Lecture Notes of Mechanics of Solids, Chapter 9 1

Chapter 9 Buckling of Columns

9.0 INTRODUCTION TO BCUKLING (SI&4
th
:649-652)
In discussing the analysis and design of various structures in the previous chapters, we had
two primary concerns: (1) the strength of the structure, i.e. its ability to support a specified
load without experiencing excessive stresses; (2) the ability of the structure to support a
specified load without undergoing unacceptable deformations. In this chapter, we shall be
concerned with stability of the structure, i.e. with its ability to support a given load without
experiencing a sudden change in its configuration. Our discussion will relate mainly to
column, i.e. to the analysis and design of vertical prismatic members supporting axial loads.

If a beam element is under a compressive load and its length if the orders of magnitude are
larger than either of its other dimensions such a beam is called a columns. Due to its size its
axial displacement is going to be very small compared to its lateral deflection called buckling.
Quite often the buckling of column can lead to sudden and dramatic failure. And as a result,
special attention must be given to design of column so that they can safely support the loads.

In looking at columns under this type of loading we are only going to look at three different
types of supports: pin-ended, doubly built-in and cantilever.


9.1 SLENDER PIN-ENDED COLUMN (SI 649-657; 4
th
:652-657; 3
rd
Ed p.653-
661)
Due to imperfections no column is really straight. At some critical compressive load it will
buckle. To determine the maximum compressive load (Buckling Load) we assume that
buckling has occurred as shown in Fig. 9.1,
x
y,v
P P
L

Fig. 9.1 Deflection column due to applied compressive load P

Look closely at the FBD of the left hand end of the beam as in Fig. 9.2:
y,v
P
x
v
P
V(x)
M(x)

Fig. 9.2 FBD of section of length x of deflected column

Equating moments at the cut end:
( ) 0 0 = + = =

x M Pv M ( ) Pv x M = (9.1)
But since the deflection of a beam is related with its bending moment distribution, then:
Pv
dx
v d
EI =
2
2
(9.2)
which simplifies to: 0
2
2
=

+ v
EI
P
dx
v d
(9.3)
where P/EI is a constant. This expression is in the form of a second order differential
equation of the following type:
Lecture Notes of Mechanics of Solids, Chapter 9 2

0
2
2
2
= + v
dx
v d
(9.4)
where:
EI
P
=
2
(9.5)
The solution of this equation is:
( ) ( ) x sin B x cos A v + = (9.6)
where A and B are constants, which can be determined using the columns kinematic
boundary conditions.

Kinematic Boundary Conditions
at x = 0, v = 0: 0 = A + 0, giving that A = 0
at x = L , v = 0, then: ( ) L sin B = 0
If B = 0, No bending moment exists, so the only logical solution is for: ( ) 0 = L sin and the
only way that this can happen is if :
= n L , (9.7)
where L , , , n 3 2 1 = . But since:
2
2

= =
L
n
EI
P
(9.8)
then we get that buckling load as:
2
2
2
L
EI
n P

= (9.9)
The values of 'n' define the buckling mode shapes, as in Fig. 9.3:

First mode of buckling
Second mode of buckling
Third mode of buckling
2
2
1
L
EI
P

=
2
2
2
4
L
EI
P

=
2
2
3
9
L
EI
P

=
P
1
P
1
P
2
P
2
P
3
P
3


Fig. 9.3 First three modes of buckling loads

Critical Buckling Load
However, since P
1
< P
2
< P
3
, the column buckles at P
1
and never gets to

P
2
or P
3
unless
bracing is placed at the points where v = 0 to prevent buckling at lower loads.
The critical load for a pin ended column is therefore:
E Crit
P
L
EI
P =

=
2
2
(9.10)
which is also called Euler Buckling Load,
P
Crit
Critical or maximum axial load on the column just before it begins to buckle
E Youngs modulus of elasticity
I least second moment of area for the columns cross sectional area
L unsupported length of the column, whose ends are pinned
Lecture Notes of Mechanics of Solids, Chapter 9 3



9.2 BUILT-IN COLUMN (SI&4
th
: 658-668; 3
rd
Ed p.662-672)
The critical load for other columns can be expressed in terms of the critical buckling load for
a pin- ended column P
E
. A built-in column looks like Fig. 9.4:

A
L
L/4 L/2 L/4
L
E
B P P
P P
Zero Bending Moment


Fig. 9.4 Built-in column at both ends showing the effective pin-ended length

From symmetry conditions, at the points of inflection
( ) x M
dx
v d
= = 0
2
2

which occurs at 1/4L points. Thus the middle half of the column can be taken out and treated
as a pin-ended column of length L
E
= L/2 as shown in Fig. 9.4. The critical load for this half
length is then :
E
E
Crit
P
L
EI
L
EI
P 4
4
2
2
2
2
=

= (9.11)


9.3 CANTILEVER COLUMN

A
L=L
E
/2
B
P
P
L
E


Fig. 9.5 Cantilever column and its effective length

This is similar to previous case. However, this span is equivalent to 1/2 of the Euler span L
E
,
as illustrated in Fig. 9.5, thus:
4
4
2
2
2
2
E
E
Crit
P
L
EI
L
EI
P =

= (9.12)

Note: Since P
Crit
is proportional to I, the column will buckle in the direction corresponding to
the minimum value of I, as shown in Fig. 9.6:
Lecture Notes of Mechanics of Solids, Chapter 9 4

x
y
z
y
z
I
y
> I
z
Cross-section
P
P
Buckling Direction
A
b
h

Fig. 9.6 Column cross section showing the direction of buckling (here:
12 12
3 3
hb
I
bh
I
y z
= < = )

9.4 CRITICAL COLUMN STRESS
A column can either fail due to the material yielding, or because the column buckles, it is of
interest to the engineer to determine when this point of transition occurs.
Consider the Euler buckling equation 9.10
2
2
L
EI
P
E

=
Because of the large deflection caused by buckling, the least second moment of area term I
can be expressed as follows:
2
Ar I = (9.13)
where: A is the cross sectional area and r is called radius of gyration of the cross sectional
area, i.e. A / I r = . Note that the smallest radius of gyration of the column, i.e. the least
second moment of area I should be taken in order to find the critical stress.
Dividing the buckling equation by A, gives:
( )
2
2
r / L
E
A
P
E
E

= = (9.14)
where:
E
is the compressive stress in the column and must not exceed the yield stress
Y
of the
material, i.e.
E
<
Y
, L / r is called the slenderness ratio, it is a measure of the column's flexibility.
If this equation is plotted for steel it gives:
L/r

x
240MPa
89
( )
2
2
r / L
E
Crit

=

Y

Fig. 9.7 Critical stress vs slenderness ratio for steel

For a column not to fail by either yielding or buckling, its stress must remain underneath this
diagram in Fig. 9.7.

Example 9.1 A 2m long pin ended column of square cross section. Assuming E=12.5GPa,

allow
=12MPa for compression parallel to the grain, and using a factor of safety of 2.5 in
computing Eulers critical load for buckling, determining the size of the cross section if the
column is to safely support (a) a P = 100kN load and (b) a P = 200kN load.
Lecture Notes of Mechanics of Solids, Chapter 9 5

P
P
A
B
I
z
y
a
a
Section a-a
s
s

Part (a)
Second moment of area
12 12
1
4
3
s
ss I I
y z
= = =
Buckling criterion
Using given Factor of Safety FS=2.5

=
allow
fail
F
F
FS , we make the required critical load as
N kN . P FS P
Crit
3
10 250 100 5 2 = =
Based on Eulers formula, Eq. (9.10), we have
N
L
EI
P
Crit
3
2
2
10 250

=
E
L
I
2
2 3
10 250

or: mm . m .
. E
L
s
B
3 99 0993 0 12
10 5 12
2 10 250
12
10 250
4
9 2
2 3
4
2
2 3
1
= =


=

Stress criterion
allow
A
P
= =
allow
P
s A

=
2

i.e. mm . m .
P
s
allow
3 91 0913 0
10 12
10 100
6
3
1
= =

Comparing the results from these two criteria, we have { } mm . s , s max s
B
3 99
2 2
=

. In this
case, the design is taken against the buckling criterion. Finally, one may select a round-up
amount, e.g. s = 100mm, as the design of the size of cross section.

Part (b)
Buckling criterion
N kN . P FS P
Crit
3
10 500 200 5 2 = =
Step 2: Eulers formula N
L
EI
P
Crit
3
2
2
10 500

=
E
L
I
2
2 3
10 500

or: mm . m .
. E
L
s
B
1 118 1181 0 12
10 5 12
2 10 500
12
10 500
4
9 2
2 3
4
2
2 3
2
= =


=

Stress criterion
allow
A
P
= =
allow
P
s A

=
2

i.e. mm . m .
P
s
allow
1 129 1291 0
10 12
10 200
6
3
2
= =

Comparing the results from these two criteria, we have { } mm . s , s max s
B
1 129
2 2
=

. In this
case, the design is taken against the stress criterion. One may select s = 130mm as the design
of the size of cross section.
Lecture Notes of Mechanics of Solids, Chapter 9 6

Example 9.2 Determine the largest load P which may be applied to the structure as shown.
Assume that E=200GPa, allowable vertical deflection at point A
allow
=0.5mm and allowable
compressive and tensile stress
allow
=50MPa.
y
z
h=50mm
b=100mm
P
A
B
C
F
AB
F
AC
P
30
Pin A
A
8m
3m
30
Cross section for AB & AC

Step 1: Determine the members internal forces
P sin F F
AB y
= = +

30 0 P F
AB
2 = (+ tensile force)
AC AB x
F cos F F + = = +

30 0 P F
AC
3 = (- compressive force)

Step 2: Buckling criterion F
AB
is in tension, we do not considered its buckling. But bar AC is
a strut and we need to check for buckling. I about y and z is computed respectively

= =
4 6
3 3
10 04267 1
12
5 0 1 0
12
m .
. . bh
I
z
<

= =
4 6
3 3
10 667 41
12
1 0 5 0
12
m .
. . hb
I
y

( ) ( )
kN .
.
L
I E
P
AC
AC AC
AC , Crit
128 32
8
10 04267 1 10 200
2
6 9 2
2
2
=

=

=


But
B AC , Crit AC
P P F 3 = = , kN . / P P
Crit B
55 18 3 = =
Step 3: Strength criterion Consider tensile and compressive stresses in AB and AC respectively.
6
10 50
1 0 05 0
2
=

= =
allow
AB
AB
AB
. .
P
A
F
kN P 125
400
10 50
6
=

=
6
10 50
1 0 05 0
3
=

= =
allow
AC
AC
AC
. .
P
A
F
kN . P 3 144
400
10 50
6
=

=
From stress criterion, the maximum allowable load should be the smallest one i.e. P

=125kN

Step 4: Stiffness criterion Consider vertical deflection at point A using Castiglianos method.
Total strain energy due to axial forces:
AC AC
AC AC
AB AB
AB AB
i i i
i i
A E
L F
A E
L F
A E
L F
U
2 2 2
2 2 2
+ = =

The displacement can be then computed as: ( )

=

i i
i
i
i
i
i i i
i i
P
A E
L
P
F
F
A E
L F
P 2
2

Member F
i
(N) P F
i
L
i
(m) A
i
(m
2
) ( )( ) P F A E L F
i i i i i


AB 2P 2 6 0.05

2.410
-9
P
AC
- 3 P - 3
8 0.05 2.410
-9
P
Thus we have: ( ) ( )
allow P
P . P . P . = + =
9 9 9
10 8 4 10 4 2 10 4 2
kN .
.
.
P 17 104
10 8 4
0005 0
9
=

Step 5: Determine the maximum allowable load P from the above three criteria
Clearly, for the safety reason, we should pick the lowest level as the allowable load
{ } kN . P , P , P min P
B
55 18 = =


Proc. of the 10
th
Intl. Conf. on Advances in Steel Concrete Composite and Hybrid Structures
The deflection at the mid-height of the hollow steel tube caused by the preload (u
mo
) is treated
as initial geometric imperfection at the mid-height of the CFST beam-column. The external
moment at the mid-height of the CFST beam-columns can be determined as
( )
m mo o me
u u u e P M + + + = (9)
where P is the applied axial load, e is the eccentricity of the applied load and u
o
is the initial
geometric imperfection at the mid-height of the hollow steel tube.
Figure 2: Strain distribution in column section
The analysis procedure is given as follows:
(1) Calculate the deflection u
mo
of the hollow steel tube under the preload;
(2) Set u
o
= u
o
+ u
mo
;
(3) Initialize the mid-height deflection of the beam-column:
m m
u u A = ;
(4) Calculate the curvature
m
| at the mid-height of the beam-column;
(5) Adjust the neutral axis depth
n
d using the Mllers method;
(6) Calculate the force Pand moment M ;
(7) Repeat Steps (5) to (6) until
k me m
M M r c < = , where
k
c is set to
4
10

;
(8) Increase the deflection at mid-height of the column by
m m m
u u u A + = ;
(9) Repeat Steps (4) to (8) until ultimate load P
u
is obtained or deflection limit is reached.
2.4 Mllers Method Algorithms
The Mllers method algorithms are implemented in the fiber element analysis program to
adjust the depth of the neutral axis (d
n
) in the section to obtain equilibrium conditions. The
depth of the neutral axis is adjusted by
m m m m
m
, n , n
c a b b
c
d d
4
2
2
3 4

= (10)
( )( ) ( )( )
( )( )( )
3 1 3 2 2 1
3 2 3 1 3 1 3 2
, n , n , n , n , n , n
, m , m , n , n , m , m , n , n
m
d d d d d d
r r d d r r d d
a


= (11)
( ) ( ) ( ) ( )
( )( )( )
3 1 3 2 2 1
3 1
2
3 2 3 2
2
3 1
, n , n , n , n , n , n
, m , m , n , n , m , m , n , n
m
d d d d d d
r r d d r r d d
b


= (12)
3 , m m
r c = (13)
|
D
d
c
t
c
e,i
de,i
Steel fibers
Concrete fibers
y
i
y
n,i
x
n
y
N.A.
398
Proc. of the 10
th
Intl. Conf. on Advances in Steel Concrete Composite and Hybrid Structures
The sign of the square root term in the denominator of Eq. (10) is taken to be the same as that
of b
m
. The values of d
n,1
, d
n,2
and d
n,3
and corresponding residual moments r
m1
, r
m2
and r
m3
need to be switched as discussed by Patel et al. (2012).
3. Comparisons with Experimental Results
The predicted and experimental ultimate axial strengths of circular CFST columns with
preload effects are given in Table 1. Specimens A122, A124, B122, A202 and A204 were
tested by Zhang et al. (1997) and the remaining specimens shown in Table 1 were conducted
by Liew and Xiong (2009). It can be seen from Table 1 that there is a good agreement
between predicted and experimental results. The mean value of the predicted to the
experimental ultimate axial strength is 0.93. The predicted and experimental load-deflection
curves for Specimen A204 are given in Fig. 3. The figure shows that the model predicts well
the load-deflection curves of the specimen.
Table 1: Ultimate strengths of circular CFST columns with preload effects
Specimens
D
(mm)
t
(mm)
L
(mm)
o
u
(mm)
'
c
f
(mm)
sy
f
(MPa)
su
f
(MPa)
a
| .exp u
P
.num u
P
exp . u
. u
p
P
num
A122 133 4.3 1670 L/1000 35.9 325 430 0.22 430 394 0.92
A124 133 4.3 1670 L/1000 35.9 325 430 0.42 416 369 0.89
B122 133 4.3 1670 L/1000 35.9 325 430 0.23 347 346 1.00
A202 133 4.3 2730 L/1000 35.9 325 430 0.22 293 262 0.90
A204 133 4.3 2730 L/1000 35.9 325 430 0.41 282 280 0.99
CFT-I-40-30P 219 6.3 1728 L/1200 44 405 518 0.299 3648 3235 0.89
CFT-I-100-30P 219 6.3 1728 L/1570 113 405 518 0.305 5278 5121 0.97
CFT-I-130-40P 219 6.3 1728 L/4300 139 405 518 0.380 5437 5860 1.08
CFT-L-40-30P 219 6.3 3078 L/2800 49 393 506 0.306 3160 2845 0.90
CFT-L-100-30P 219 6.3 3078 L/2800 111 393 506 0.310 4580 3896 0.85
CFT-L-130-40P 219 6.3 3078 L/1700 125 393 506 0.399 4827 4078 0.84
Mean 0.93
Standard deviation 0.07
Coefficient of variation 0.08
Figure 3: Comparison of predicted and experimental load-deflection curves
4. Behavior
The numerical model developed was used to investigate the behavior of circular CFST
slender beam-columns with preload effects. The initial geometric imperfection at the mid-
0
50
100
150
200
250
300
0 20 40 60
A
x
i
a
l

l
o
a
d

(
k
N
)
Mid-height deflection u
m
(mm)
Experiment (A204)
Numerical analysis
399
Proc. of the 10
th
Intl. Conf. on Advances in Steel Concrete Composite and Hybrid Structures
height of the beam-column was taken as 1000 L . The Youngs modulus of steel was 200 GPa.
The preload ratios were taken as 0.0, 0.4 and 0.8 in the analysis.
4.1 Effects of Preloads on Load-Deflection Curves
The effects of preloads on the load-deflection curves for a high strength slender steel tube
with yield stress of 690 MPa filled with normal strength concrete of 40 MPa were examined.
The diameter of the column was 600 mm. The following parameters were considered: the
diameter-to-thickness ( ) t D ratio of 60, the column slenderness ( ) r L ratio of 80 and the
loading eccentricity ( ) D e ratio of 0.2. Fig. 4(a) shows that increasing the preload ratio
significantly reduces the stiffness and ultimate axial strength of the CFST beam-column. The
mid-height deflection at the ultimate axial load is found to increase with an increase in the
preload ratio. When increasing the preload ratio from 0.0 to 0.4 and 0.8, the ultimate axial
strength of the slender beam-column is reduced by 7.7 % and 17.4 % respectively.
(a) (b)
(c) (d)
Figure 4: Behavior of circular CFST columns with preload effects
4.2 Effects of Preloads on Column Strength Curves
The analyses of high strength circular CFST beam-columns with a diameter of 500 mm, t D
ratio of 50, r L ratio varying from 0 to 100, D e ratio of 0.2, f
sv
= 690 MPa, f
su
= 790 MPa
and f
c
= 60 MPa were undertaken to study the effects of preloads on the column strength
curves. As presented in Fig. 4(b), increasing the r L ratio significantly reduces the ultimate
axial strength of CFST beam-columns with the same preload ratio. The strength ratio tends to
increase when increasing the L/r ratio. The preload with a ratio of 0.8 might reduce the
ultimate axial strength of the CFST slender beam-column with a r L ratio of 100 by 21.9 %.
b
a
=0
b
a
=0.4
b
a
=0.8
0
2000
4000
6000
8000
10000
0 200 400 600 800
A
x
i
a
l

l
o
a
d

(
k
N
)
Mid-height deflection u
m
(mm)
b
a
=0
b
a
=0.4
b
a
=0.8
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 30 60 90 120 150
U
l
t
i
m
a
t
e

a
x
i
a
l

l
o
a
d

P
n
/
P
o
Slenderness ratio L/r
b
a
=0
b
a
=0.4
b
a
=0.8
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
0 20 40 60 80 100 120 140
U
l
t
i
m
a
t
e

a
x
i
a
l

l
o
a
d

(
k
N
)
Diameter-to-thickness ratio D/t
b
a
=0
b
a
=0.4
b
a
=0.8
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.5 1 1.5 2 2.5
U
l
t
i
m
a
t
e

a
x
i
a
l

l
o
a
d

P
n
/
P
o
Eccentricity ratio e/D
400
Proc. of the 10
th
Intl. Conf. on Advances in Steel Concrete Composite and Hybrid Structures
The results indicate that when the r L ratio is less than 22, the preload effect becomes
insignificant. This means that for short CFST beam-columns with a r L ratio less than 22,
the preload effect can be ignored in the design.
4.3 Effects of Preloads and Diameter-To-Thickness Ratio
Investigations on the effects of preloads and D/t ratios on the normal strength steel slender
tubes with yield stress of 300 MPa filled with high strength concrete of 70 MPa were
performed using the numerical model. The diameter of the column section was 700 mm with
D/t ratios ranging from 20 to 100. The r L ratio of 80 and D e ratio of 0.2 were considered.
It can be seen from Fig. 4(c) that the ultimate axial strength decreases with increasing the
t D ratio regardless of the preload value. When the preload ratio increases from 0.0 to 0.4
and 0.8 for the t D ratio of 100, the ultimate axial strength is decreased by 6.9% and 14.7%
respectively.
4.4 Effects of Preloads and Loading Eccentricity Ratio
The numerical studies were carried out to examine the effects of preloads and loading
eccentricity ratios on the ultimate strength of normal strength steel slender tube with a
diameter of 550 mm filled with normal strength concrete of 40 MPa. The yield stress of the
steel tube was 300 MPa. Other parameters used were: D/t = 55, L/r = 80 and e/D ratio ranging
from 0 to 2. It can be observed from Fig. 4(d) that increasing the e/D ratio reduces the
ultimate axial strength. The reduction in the ultimate axial strength of CFST columns
increases with increasing the e/D ratio from 0.0 to 0.4. However, the strength reduction tends
to decrease with an increase in e/D ratio over 0.4. The preloads cause a maximum reduction
in the strength of the column with the e/D ratio of 0.4.
5. Conclusions
This paper has presented a numerical model for simulating the behavior of circular CFST
slender beam-columns with preload and concrete confinement effects. Computation
algorithms have been developed for predicating the load-deflection responses of circular
CFST slender beam-columns including preload effects. Verification studies demonstrate that
the numerical model is accurate and efficient for the inelastic analysis of circular CFST
slender beam-columns subjected to preloads on the steel tubes. The parametric study shows
that increasing the preload ratio reduces the ultimate axial strengths of CFST slender beam-
columns. The preload effect can be ignored in the design of CFST short beam-columns with
the column slenderness ratio less than 22. Increasing the diameter-to-thickness ratio reduces
the ultimate axial strength for the same preload ratio. The preload effect is most pronounced
when the eccentricity ratio is 0.4.
6. References
[1] Fujimoto T, Mukai A, Nishiyama I and Sakino K (2004), Behavior of eccentrically loaded concrete-filled
steel tubular columns. Journal of Structural Engineering, ASCE, Vol. 130 No. 2, pp. 203-12.
[2] Giakoumelis G and Lam D (2004), Axial capacity of circular concrete-filled tube columns. Journal of
Constructional Steel Research, Vol. 60 No. 7, pp. 1049-68.
[3] Han LH and Yao GH (2003), Behaviour of concrete-filled hollow structural steel (HSS) columns with pre-
load on the steel tubes, Journal of Constructional Steel Research, Vol. 59 No. 12, pp. 1455-75.
[4] Hu HT, Huang CS, Wu MH and Wu YM (2003), Nonlinear analysis of axially loaded concrete-filled tube
columns with confinement effect, Journal of Structural Engineering, ASCE, Vol. 129 No. 10, pp. 1322-29.
[5] Liang QQ (2009), Performance-based analysis of concrete-filled steel tubular beam-columns, Part I:
Theory and algorithms, Journal of Constructional Steel Research, Vol. 65 No. 2, pp. 363-72.
401
Proc. of the 10
th
Intl. Conf. on Advances in Steel Concrete Composite and Hybrid Structures
[6] Liang QQ (2011), High strength circular concrete-filled steel tubular slender beam-columns, Part I:
Numerical analysis, Journal of Constructional Steel Research, Vol. 67 No. 2, pp. 164-71.
[7] Liang QQ and Fragomeni S (2009), Nonlinear analysis of circular concrete-filled steel tubular short
columns under axial loading, Journal of Constructional Steel Research, Vol. 65 No 12, pp. 2186-96.
[8] Liang QQ, Uy B and Liew JYR (2006), Nonlinear analysis of concrete-filled thin-walled steel box columns
with local buckling effects, Journal of Constructional Steel Research, Vol. 62 No. 6, pp. 581-91.
[9] Liew JYR and Xiong DX (2009), Effect of preload on the axial capacity of concrete-filled composite
columns, Journal of Constructional Steel Research, Vol. 65 No.3, pp. 709-22.
[10] Mander JB, Priestley MJN and Park R (1988), Theoretical stress-strain model for confined concrete,
Journal of Structural Engineering, ASCE, Vol. 114 No. 8, pp. 1804-26.
[11] Neogi PK, Sen HK and Chapman JC (1969), Concrete-filled tubular steel columns under eccentric loading,
The Structural Engineer, Vol. 47 No. 5, pp. 187-95.
[12] Patel VI, Liang QQ and Hadi MNS (2012), High strength thin-walled rectangular concrete-filled steel
tubular slender beam-columns, Part I: Modeling, Journal of Constructional Steel Research, Vol. 70, pp.
377-84.
[13] Tang J, Hino S, Kuroda I and Ohta T (1996), Modeling of stress-strain relationship for steel and concrete
in concrete filled circular steel tubular columns, Steel Construction Engineering, JSSC, Vol. 3 No. 11, pp.
35-46.
[14] Xiong DX and Zha XX (2007), A numerical investigation on the behaviour of concrete-filled steel tubular
columns under initial stresses, Journal of Constructional Steel Research, Vol. 63 No. 5, pp. 599-611.
[15] Zhang XQ, Zhong ST, Yan SZ, Lin W and Cao HL (1997), Experimental study about the effect of initial
stress on bearing capacity of concrete-filled steel tubular members under eccentric compression, Journal of
Harbin University of Civil Engineering and Architecture, Vol. 30 No. 1, pp. 50-6 (in Chinese).
402