Applied Mechanics and Materials Vols.

256-259 (2013) pp 2709-2712 Online: 2012-12-13

© (2013) Trans Tech Publications, Switzerland
doi:10.4028/www.scientific.net/AMM.256-259.2709

Analytical Behavior of Square Reinforced Concrete Stub Columns

under Overall Fire and Axial Compression

Yongqian Zheng 1,2,a, Jinping Zhuang 1,2,b

1
Department of Civil Engineering, Fujian University of Technology, No.3 Xue Yuan Road,
University Town, Shangjie Minhou, Fuzhou, Fujian, P.R. China, 350108
2
College of Civil Engineering, Fuzhou University, No.2 Xue Yuan Road, University Town,
Shangjie Minhou, Fuzhou, Fujian, P.R. China, 350108
a
[email protected], [email protected]

Keywords: Reinforced concrete (RC), Column, Fire, Axial compression

Abstract. A finite element model for square reinforced concrete stub column under overall fire and
axial compression is developed. The temperature of columns subjected to simulated heating and
cooling standard fire and the maximal temperature field were determined. The creep of steel bars,
bond slip between steel bars and concrete, transformation of material properties were considered in
the force analysis. The deformation under overall fire and the load bearing capacity of the column
with initial load after fire are provided.

Introduction
Fire resistant behavior is one of the most important components of structural performance. The fire
research has caused public concern to limit the damage of structures in fire and develop the retrofit
method after fire. In the past, most studies focused on the behavior of members, joints and frames
under fire condition especially on the heating phase not considering the cooling phase. As for most
post-fire researches, the effect of full range fire and the initial load were not taken into account.
Actually the overall fire includes the heating and cooling phases with an initial load, which is more
practical and valuable in the fire research.
In the past few years, some experiments were carried out to study the performance of RC columns,
concrete filled steel tubular (CFST) columns and joints under overall fire condition by Song et al. [1]
and Zhang [2]. Finite element models were presented for CFST stub column and steel reinforced
concrete (SRC) columns in fire including the cooling phase by Song et al. [3], Du [4] etc..
RC columns have been widely used in structural engineering, which is very important to know
their behavior in overall fire. This paper developed a finite element model for square RC stub column
under overall fire and axial compression using software ABAQUS. The analysis consists of two
primary steps. Firstly, the temperature fields of columns subjected to ISO-834 heating and cooling
standard fire were determined. Secondly, the mechanical behaviors including the ambient, heating,
cooling and post-fire phases were analyzed based on the temperature results. The programs made by
the authors were incorporated into ABAQUS to consider the creep of steel and transform the material
properties in deferent phases.

Temperature analysis
The thermal properties for steel bars and concrete (e.g. conductivity, specific heat) presented by Lie
and Denham [5] are adopted. The water vapor has influence on temperature movements of concrete at
about 100 °C in the heating phase not in the cooling phase. The modified expression of concrete
thermal properties proposed by Hamarthy [6] is used at elevated temperature.
The node numbers in the thermal analysis mesh should be the same as these in the structural
analysis mesh. The heat transfer 8-node brick elements (DCC3D8D) were used for the concrete and
the 2-node link elements (DCC1D2D) were used for the steel bars. The output file (*.FIL) was

All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans
Tech Publications, www.ttp.net. (ID: 141.211.4.224, University of Michigan Library, Media Union Library, Ann Arbor, USA-13/07/15,17:16:31)
2710 Advances in Civil Engineering II

specified to record the nodal temperatures at each time step, which was used as the input in structural
analysis. The ISO-834 standard fire curve including cooling phase was used for the outer temperature.
Heat conduction and radiation were applied to the surface of column. In the typical example, the fire
duration time is 120 min and the temperature drops to ambient temperature at 367 min. The column
with square section of 400×400 mm and was exposed to four-face fire. Fig. 1 shows the predicted
temperature (T) versus time (t) curves under overall fire, where d is the distance from external surface
to section point. It can be seen that the temperature of external surface falls accordingly as the outer
temperature drops. However, the internal temperature still keeps rising after 120 min and the deeper
section point reaches maximal temperature later. It is evident that the temperature lags in different
degrees at temperature close to 100 °C.
As mentioned above, the most sectional points reach maximal temperature at different time. It is
very important to get the maximal temperature of each section point, which should be used in force
analysis due to the material constitutive models in cooling phase and post fire phase. The
programming made by the authors is applied to read the ABAQUS file (*.odb) to get the maximal
temperature and write to the output file, as is shown in Fig. 2. The maximal temperatures of external
corner and center are 1039 °C and 365.8 °C respectively.
1200
Outer temperature
1000 d=200mm
d=121mm
800 d=42.5mm
T (°C)

d=0mm
600

400

200

0
0 120 240 360 480 600 720
t (min)

Fig. 1 Predicted temperature-time curves Fig. 2 Maximal temperature of the section

Force analysis
The total strains of steel and concrete under temperature and loading consist of three and four
components respectively. The free thermal strain of steel and concrete presented by Lie and Denham
[5] are adopted. The stress related strain of steel in heating phase proposed by Lie and Denham [5] and
the double broken-line model are adopted. In the cooling phase, it can be interpolated. As for concrete,
the stress related strain described by Lie and Denham [5] is used for the heating phase and the model
provided by Lu [7] can be used for the cooling and post fire phases. The thermal strains of steel and
concrete can be calculated according to the expressions proposed by Lie and Denham [5]. The creep
strain at high temperature is more visible than that at ambient temperature and the strain value of steel
is more obvious. The AIJ model [8] for creep strain of steel at high temperature is used for the analysis
and the user subroutine is incorporated into ABAQUS. The creep strain and transient strain of
concrete can be ignored for the analysis.
Typical RC stub column was design according to the relevant code. The geometries are square
cross section of 400×400 mm, height of 1200 mm, 8 longitudinal steel bars with a diameter of 25 mm,
stirrup diameter of 10 mm, stirrup spacing of 150 mm. The concrete cover thickness to longitudinal
steel bar is 30mm. The yield stresses of steel bars and the cubic strength of concrete are 400 MPa and
50 MPa respectively. In order to avoid local damage of end parts, each end range about 100 mm is
kept ambient temperature.
Fig. 3 shows the finite element analysis model of ABAQUS. Two rigid plates were used for the
bearing load plates and the load was applied to the center of the upper plate. The upper surface of rigid
plate was restrained to have only the displacement in Z direction, and the displacements in all
 Applied Mechanics and Materials Vols. 256-259 2711

directions at the bottom of the lower rigid plate were constrained, shown as in Fig. 3. The concrete and
rigid plate were modeled using 8-node brick elements (C3D8R), with three translation degrees of
freedom at each node. The 2-node truss elements (T3D2) were used to model all the steel bars. The
stirrups were embedded to the concrete while the longitudinal steel bars had the relative
displacements in axial direction. The 2-node spring elements (SPRING2) were used to simulate the
interaction of the longitudinal steel bars and concrete to get the bond slip mechanism. The nonlinear
behavior was defined by nodal force-relative displacement relationship and the bond stress versus slip
relationship model provided by Zhu et al. [9] was adopted in the analysis.
At ambient temperature, the load was apply to the column and kept constant in the heating and
cooling phases. After that, the temperature of column reverted to the ambient temperature and the
displacement in axial direction was applied to fail finally. The field variables were set to transform the
different material force properties in different phases and the user subroutine was carried out to
achieve it. Some material properties related to the current temperature especially at elevated
temperature while some related to the maximal temperature in cooling and post fire phases
particularly. During the calculation, the current temperature and the maximal temperature were read
in time from the output file of ABAQUS.

Rigid plate Load

Constraint UX, UY

Stirrup

Outer temperature
Longitudinal
steel bar

Z
Y

Rigid plate Constraint UX, UY, UZ

Fig.3 Finite element analysis model

Fig. 4 shows the predicted axial deformation (u) versus time (t) curves at different axial load levels
(n), where the broken line represents the boundary at 120 min. The positive value represents the upper
expansion and the negative value represents the compression deformation. It is evident that the RC
columns expand in axial direction at low axial load level due to the rising temperature. Although the
outer temperature drops after 120 min, the expansion of column increases because of the increasing
temperature of internal elements. Subsequently the effect of load outweighs the expansion and the
axial compression deformation occurs and the trend becomes gently for a long time. Finally the stub
column was compressed to fail. For the axial load level of 0.3, the relative higher load leads to the
compression deformation at the initial heating stage. Fig. 5 shows the predicted load (P) versus axial
deformation (u) curves at different axial load levels (n). At ambient temperature the load increases to
regulated value. During the heating and cooling, the load keeps constant while the axial deformation
changes. After that, the typical load-axial deformation curves include the ascent and descent stages.
The calculated load bearing capacity of RC column at ambient temperature in the example is 7874.1
kN. The load bearing capacities after fire at axial load levels of 0.1, 0.2 and 0.3 are 4105.8 kN, 4111.8
kN and 4036.2 kN and the differ ratio is within 2% under the given condition. When fire duration time
is 120 min, the load bearing capacity of the column decreases about 48% compared to the one at
ambient temperature.
2712 Advances in Civil Engineering II

6 5000
4 n=0.1
n=0.2 4000
2 n=0.3
0
u (mm)
3000

P (kN)
-2
-4 2000
n=0.1
-6 n=0.2
1000
-8 n=0.3

-10 0
0 800 1600 2400 3200 -20 -15 -10 -5 0 5
t (min) u (mm)

Fig. 4 Axial deformation-time curves Fig. 5 Load-axial deformation curves

Conclusions
(1) The finite element model developed in this paper can be used to analyze the behavior of square
RC stub column under overall fire and axial compression. The user subroutines can be
incorporated into ABAQUS to consider the creep of steel and transform the material properties.
(2) The internal temperature still keeps rising when the outer temperature drops.
(3) At fire duration time of 120 min, the load bearing capacity of RC column in the typical example
has a decrease of about 48% of the one at ambient temperature and the initial load levels has little
influence on load bearing capacity after exposure to fire.

Acknowledgements
This work was financially supported by the Natural Science Foundation of Fujian Province of China
(No.2012J05083) and Scientific Research Special Project of Funded Provincial Universities
(JK2010040).

References
[1] T.Y. Song, L.H. Han and B. Uy: J. of Constr. Steel Res. Vol. 66 (2010). p. 591
[2] J.G. Zhang: Experimental research on mechanical behaviors of reinforced concrete columns
with sustained axial load after exposure to fire or high temperatures. Deg. Diss. of Hunan Univ.
Hunan. 2009. (In Chinese)
[3] T.Y. Song, L.H. Han and H.X. Yu: J. of Constr. Steel Res. Vol. 66 (2010). p. 369
[4] E.F. Du: The mechanical performance of SRC columns during the whole period of standard fire.
Deg. Diss. of Suzhou Univ. of Sci. and Technol. Suzhou. 2009. (In Chinese)
[5] T.T. Lie and E.M.A. Denham: Factors Affecting the Fire Resistance of Square Hollow Steel
Columns Filled With Bar-Reinforced Concrete Int. Rep. No. 650. NRCC. Canada. 1993
[6] T.Z Harmathy: Fire safety design and concrete (Longman Group UK Limited, UK 1993).
[7] G.Q. Li, L.H. Han, G.B. Lou and S.C. Jiang: Fire design of steel structure and steel-concrete
composite structure (China Architecture & Building Press, China 2006) (In Chinese)
[8] AIJ. Recommendations for design and construction of concrete filled steel tubular structures
(Architectural Institute of Japan. Japan 1997)
[9] B.L. Zhu, Z.D. Lu and K.X. Hu: Building Sci. Res. of Sichuan. (1990). p. 37. (in Chinese)
Advances in Civil Engineering II
10.4028/www.scientific.net/AMM.256-259

Analytical Behavior of Square Reinforced Concrete Stub Columns under Overall Fire and Axial
Compression
10.4028/www.scientific.net/AMM.256-259.2709