NEF4101 Research Project: CFDST Column Nonlinear Analysis

This project presents a numerical model for analyzing the nonlinear behavior of double square steel tubular short columns filled with concrete (CFDST) under axial compression. The study focuses on the effects of concrete confinement and local buckling of the outer steel tube. The project includes an introduction covering background information, the importance of the research, and objectives. A literature review examines experimental and numerical studies of square CFDST short columns, including discussions on strength, bearing capacity, and finite element modeling. The methodology details steel and concrete stress-strain relationships and the analysis procedure. Results and discussions compare the numerical model with experimental results. The research aims to develop a predictive model for future designs, addressing gaps in existing studies related to local buckling and high-strength materials. The project also considers pre-loads, the confinement of infill concrete, and high-strength materials.

1
Nonlinear Analysis of Double Square Steel Tubular Short Columns Filled with Concrete
Under Axial Compression
Student Name:
Student ID:
Supervisor:
Date of Submission:

2
Contents
Chapter 1. Introduction...........................................................................................................................2
Abstract......................................................................................................................................................2
1.1 Background Information................................................................................................................3
1.2 Importance of the Research study..................................................................................................5
1.3 Objectives of the Research topic.....................................................................................................6
1.4 Layout of This Report.....................................................................................................................6
Chapter 2. Literature Review...................................................................................................................7
2.1 Introduction.....................................................................................................................................7
2.2 Experimental work on square CFDST short columns..................................................................8
2.3 Numerical studies of Square CFDST short columns...................................................................10
2.3.1 Strength Bearing capacity of the composite..........................................................................10
2.3.2 Modelling using Finite Element Method...............................................................................10
2.4 Buckling of CFDST short columns...............................................................................................11
2.4 Conclusion......................................................................................................................................13
Chapter 3. Nonlinear Analysis of square CFDST short columns.........................................................14
3.1 Introduction.............................................................................................................................14
3.2 Methodology.............................................................................................................................16
3.3 Steel stress-strain relationship................................................................................................16
3.4 Stress-strain Relationship for concrete..................................................................................17
3.5 Analysis procedure..................................................................................................................18
3.6 Conclusion................................................................................................................................20
Chapter 4. Results and Discussions........................................................................................................21
4.1 Introduction...................................................................................................................................21
4.2 Comparisons with experimental results.......................................................................................23
4.3 Project Management.....................................................................................................................23
4.4 Resources........................................................................................................................................24
4.5 Risk analysis...................................................................................................................................24
References.................................................................................................................................................25

3
Chapter 1. Introduction
Abstract
The construction of thin-walled square CFDST columns is done by having the space between
double sheet steel tubes filled with concrete. Use of CFDST columns has been widely applied
especially when designing performance-based structures in building projects globally. The
desirable properties of the composite that makes it to be widely adopted are the excellent
ductility and strength when high strength steel is combined with concrete. This research study
therefore seeks to come up with numerical models that can accurately predict the strength and
ductility, and also utilized in the process of designing and analyzing nonlinear square CFDST
columns which are loaded axially. It will contribute a wide body of knowledge since certain gaps
have been identified in the previous studies.
1.1 Background Information
A number of studies have been conducted in the recent past to show that when steel tubes are
filled with concrete to make square CFDST short columns, their performance is significantly
increased to levels exceeding that of conventional columns made of concrete (Mizan, 2018,
p.12). The square CFDST column is a high-performance composite material that is made up of
square tube on the outer side, while the inner tube is double layer and has concrete infill,
resulting into higher resistance to axial loading in short columns. When making a square CFDST
short column, both normal strength or high strength hollow steel tube can be used and filled with
normal strength concrete or a high strength concrete. (VipulKumar, 2013). A high strength
concrete has compressive strengths exceeding 50 Mpa while high strength steels have yield
strengths exceeding 460 Mpa.
It is important to point out that the design concepts acquired in this field of study are not only
elemental for construction purposes but also in the development of International Building Codes
of Practice to enhance structural integrity as well as performance of the structures in the built-up
environment.
In their research work, Qing and Brian (2005) alluded to the fact that square CFDST short
columns are primarily axial-load carrying members in offshore structures, bridges and high-rise
buildings because of their exceptionally higher strengths, ductility and large energy absorption

4
capacities resulting into excellent performance. Their popularity is also partly due to their
relatively lower costs as compared to conventional concrete columns.
When a thin-walled steel column is filled with concrete and subjected to axil loading, the
concrete prevents local inward buckling causing the steel to experience outward local buckling.
This will increase the buckling strength of steel walls and consequently that of the composite
column, ultimately resulting into a high-performance structure.
According to VipulKumar (2013), the use of square CFDST short columns offers a number of
advantages, making them preferred to conventional columns. The steel tubes create a
confinement for the concrete infill. The confinement effects results into increased strength of
concrete and higher ductility of the core within this square cross-sectional area of steel tubes.
The increased strength and ductility results into higher performance of the structure. Another
added advantage of CFDST column is the delay of local bucking of the steel tube caused by the
concrete infill as explained in the previous paragraph. The final advantage which is also essential
is that the steel tubes act as reinforcements in the longitudinal direction as well as formwork for
the concrete infill.
This saves a great deal in financial budgeting of materials and increases the speed of
construction. The formwork created already supports construction loads even prior to pouring of
concrete between the double sheet steel tubes.
As discussed previously, the concrete infill plays a vital role in prevention of inward local
buckling, thus the steel tube can only experience outward local buckling.
Liang and Uy (2000) observed that when examining load-deflection curves of real plates that are
subjected to compressive stresses, there is no bifurcation point. This is explained by the presence
of imperfections in the initial geometry of the real thin steel plates. As a result, it is difficult to
determine the local loads initially subjected to these thin steel plates with imperfections. It is
therefore important to consider the effects of local buckling in the analysis of nonlinear square
concrete-filled double steel tubular short columns.
During construction, steel tubes are erected to several storeys up in a Highrise building before
they are filled with concrete. This means that they are able to support a considerable number of
loads before and after construction supported by the framework established.

5
The effects of these preloads must be taken into consideration when designing square concrete-
filled double steel tubular short columns (VipulKumar, 2013).
Parametric characteristics of square concrete-filled double steel tubular short columns is largely
based on parameters like the ratio of depth to thickness, the eccentricity ratio of loading, column
slenderness ratio, the compressive strength of concrete used as well as yield strength of steel.
Studying all these parameters in this research work would not be cost effective, and would
consume a lot of time.
Conducting a full-scale behavioral analysis of square concrete-filled double steel tubular short
columns would require usage of verified nonlinear inelastic techniques of analysis.
1.2 Importance of the Research study
The use of square concrete-filled double steel tubular short columns has gained popularity in
Highrise structures and bridges in the recent past.
However, there is still limited understanding of their fundamental behavior as a result of
inadequate experimental and numerical research studies covering this composite material.
Thus, efficient design and analysis methods have not been fully developed for square CFDST not
only columns but also other structural members like beams which are also subjected to eccentric
axial loading. The absence of this vital numerical data constitutes a research gap that is one area
to be addressed by this research paper.
High strength concrete is brittle, and its usage in CFDST columns results to reduced ductility of
the column even though the composite material has a relatively higher ultimate strength.
Conversely, usage of high strength thin-walled steel provides economical designs even though
local buckling is possible. Local buckling in steel tubes is associated with a reduction in loads
that the square concrete-filled double steel tubular short columns can support safely.
Even after the initiation of local buckling, loads can still be carried effectively with the thin steel
plates without incidences of failure. As the load is increased gradually, the steel plates undergo
local buckling, a phenomenon referred to as post-local buckling by Liang et al. (2007).
The current nonlinear inelastic methods of analyzing the CFDST composites inadequately
addresses this progressive local buckling, which is yet another research gap to be filled by this
study. These nonlinear inelastic methods do not account for local buckling hence overestimating
the ultimate loads that can be safely carried by the CFDST columns without failure.

6
It is worthwhile mentioning that even the available design codes for instance the Eurocode
4(2004), ACI 318 (2002) and LRFD (1999) have no provisions specifying design of high-
strength CFDST structural members.
Having looked at the highlighted inefficiencies of the previous studies, this research work seeks
to fill the gaps by developing efficient numerical-based models to be adopted in simulation of
non-linear behavior of CFDST composites under axial loading.
These developed simulation models would therefore provide essential information to be used in
the design and analysis of structures which are safer, economical and meets the needs of clients
at the same time. The models to be developed by the study will enable efficiency when analyzing
and designing square concrete-filled double steel tubular short columns using any high strength
concrete or any other grade of concrete as may be appropriate.
1.3 Objectives of the Research topic
The major objective of the study involves coming up with efficient numerical-based models that
can accurately predict the strength and ductility, and also utilized in designing and analyzing
nonlinear square CFDST columns subjected to axial loads. These numerical models are to have
capabilities to account for; (a) pre-loads created by local buckling, (b) confinement of infill
concrete and (c) high strength materials. In order to achieve this main objective task, the
following specific aims needs to be achieved:
 To come up with a numerical model that can be used in the design and nonlinear analysis of
high strength square CFDST short columns with square cross-sectional areas subjected to
axial loading.
 To come up with a numerical model that can be used in the design and nonlinear analysis of
square CFDST short columns subjected to both axial loading and local buckling.
 To come up with a model that can be used in nonlinear analysis of high-strength CFDST
columns under axial loading when they are subjected to both local buckling and pre-load
effects.
 To carry out verification of the developed models by cross-checking them with experimental
data that is already available.
 To conduct in-depth parametric studies so as to examine the effects of various parameters on
the behavior of square concrete-filled double steel tubular short columns when subjected to
axial loads and bending.

7
1.4 Layout of This Report
For convenience and easy understanding of this report, it has been divided into 4 chapters.
Having completed chapter one that deals with the background information on CFDST columns,
chapter 2 presents an in-depth literature review on the nonlinear analysis of square CFDST short
columns subjected to axial loading. A number of journal articles are reviewed that contain
previous similar research work to form the basis of argument. Local and post-local buckling is
then reviewed for both CFDST columns and steel plates. A logical conclusion is then drawn at
the end of the chapter to summarize the literature review in relation to the research paper.
In chapter 3, the stress and strain relationships for both steel and concrete are examined in this
section. Finally, an analysis procedure is developed for the nonlinear square CFDST columns
and relevant conclusions drawn.
Chapter 4 has only two sections, 4.1 and 4.2. It describes the results obtained in the previous
sections and provide well explained discussions on the models developed and interpretations are
made. The second section conducts the comparison of experimental data for verification of the
models developed, and conclusions are made based on these results obtained.

8
Chapter 2. Literature Review
2.1 Introduction
A substantial amount of work in terms of research studies has been done in the area of nonlinear
analysis of square CFDST columns. This section highlights most of the research studies
conducted in this field, and has been used to form the background of this project. It involves a
series of experimental studies and the conclusions that were made in those studies.
Nonlinear techniques of analysis are fundamental in predicting cross sectional ductility
performance and ultimate strengths needed in the design of thin-walled CFDST columns (Qing
Quan, 2008). The current codes {Eurocode 4(2004), ACI 318 (2002) and LRFD (1999)} have
shown discrepancies in their strength predictions since the designs have used different
approaches in designing square concrete-filled double steel tubular short columns.
As much as adoption of this composite is gaining popularity in axil load bearing due to their
high strength, ductility and greater energy absorptions. This is because the steel casing confines
the concrete core increasing its ductility remarkably. The CFDST columns experience outward
local lateral buckling of the outer square steel tube (Klöppel, 1957). However, the effects of local
buckling have not been accounted for in the fiber-based analysis models in the design codes.
Since the analysis of thin-walled square CFDST columns involve a complex interaction of local
and global buckling, its analysis becomes quite challenging.
The most basic characteristics of square concrete-filled double steel tubular short columns is
dictated by a number of parameters like the d/t ratio, slenderness ration, compressive strength of
concrete and the yield strength of steel plates. Even though it is possible to conduct experiments
to investigate the behavior of CFDST columns under different loading conditions, it proves to be
expensive as all the parametric studies need to be conducted in order to conclusively reach a
conclusion, which is also time consuming. Besides, the limited data obtained would not be
sufficient to make a rational conclusion by deriving a theory from the tests that can be utilized
when designing the CFDST composite (Furlong, 1967). A well-developed numerical modelling
can be used in order to prevent the shortcomings of experiments, and used to investigate the
elemental characteristics of square CFDST short columns.

9
2.2 Experimental work on square CFDST short columns
This section highlights a series of experimental work and other research studies that have been
conducted and presents the outcomes of such experiments in a way that relates to this research
project.
It would be interesting to know that the very first experiment to determine the ultimate strength
of square CFDST short columns was done by (Knowles, R. B. and Park, R., 1969). Both steel
and concrete resist axial loading subjected to a square CFDST column (Furlong, 1967). The
experiment he conducted involved testing a number of circular CFDST columns having a ratio of
d/t as 29:98. One of the parameters examined in the study was the effect of thickness of the steel
tube on the ultimate strength of the CFDST columns. A general conclusion in the experiments
was made that when concrete is confined in steel sheets, its ductility is increased and not the
strength. However, a lesser confinement is created when a square steel tube is used, resulting into
lower strength compared to that of circular CFDST column. A recommendation was made that a
CFDST column cross-sectional area can be used as a reinforced concrete section for
determination of axial load-moment interaction diagrams. The circular cross-sectional area of a
CFDST column meant that it has more strength and performs better than a square CFDST short
column in practice.
(Knowles, R. B. and Park, R., 1969) conducted experimental tests using different slenderness ratio to
determine the effect of confinement of infill concrete on the behavior of CFDST columns of
different cross-sectional areas. For the square cross-sectional area of the CFDST column, only
ductility was increased by the confinement and not the ultimate strength. However, for the
circular CFDST column, both the ultimate strength and ductility of the concrete were achieved
by the confinement.
The study also found that increased slenderness decreases the effect created by the confinement
on the concrete core. This results into premature failure of the column by local buckling even
before the compressive strength is increased by strain. During the study, all the columns that
were tested after being subjected to axial loading failed considerably due to column buckling.
However, some cases of the experiment showed that not all the columns failed by local buckling.
The concrete failed after the attainment of the maximum loading. The studies concluded that a
tangent modulus method made reliable and accurate estimations of compressive strength of

10
concrete where the ratios of L/D exceeded 11, and being conservative for those with the same
ratio being lower than 11.
An experimental study conducted by (Hu, H. T., Huang, C. S., Wu, M. H. and Wu, Y. M., 2003)
utilized the analysis method of finite element model (FEM) in the nonlinear analysis of CFDST
short columns to examine the effect of concrete confinement. The research adopted the general-
purpose model of finite element program ABAQUS to examine the inelastic properties of
CFDST short columns. In the nonlinear analysis, both concrete-filled circular and square cross-
sectional areas reinforced with ties were used. A simulation was carried out with the double steel
tubes using model of “elastic-perfectly-plastic” accompanied by the theory of flow whereas the
modelling of concrete was done using “elastic-plastic” theory with associated flow with the rule
isotopic hardening. Since the double steel tubes were subjected to a variety of multiple stresses,
defining the elastic limit was appropriately done by the use of Von Mises yield criterion.
A series of experiments were carried out and the end support orthogonal knife edge were
designed to counteract the effect of friction at the two ends of the column, so as to ensure free
rotational mechanism at the two orthogonal bending planes. The two ends of the composite
CFDST column were pinned and subjected to a single bending curvature. The results obtained
from the experimental study indicated a delay in local buckling of the thin steel walls caused by
the infill concrete core. For a slender column, outward local buckling on all faces of the steel
tube took place at maximum load. Even though a number of experimental studies have been
conducted on the CFDST columns subjected to axial loading, not much is covered on the square
concrete-filled double steel tubular short columns with similar load conditions especially with
steels having higher yield strengths and also high strength concrete. A combination of the two
materials would result to a high strength composite material whose properties would be useful in
the application of civil and structural engineering, thus there is need to come up with numerical
models in their analysis.
2.3 Numerical studies of Square CFDST short columns
2.3.1 Strength Bearing capacity of the composite
The most effective way in solving the axial compressive bearing capacity of a square CFDST
short column is by the use of limit equilibrium (Qing Quang and Brian Uy, 2006). This is
because the method does not give consideration to the way of loading and deformation process.

11
2.3.2 Modelling using Finite Element Method
In analyzing square CFDST composites, software ABAQUS is vital for general purpose analysis
using the Finite Element (FE) method. This requires simulation of the steel tube using the 4-node
shell elements, and using reduced integration i.e. (S4R). for the infill concrete core, modelling is
done using 8-node brick elements (C3D8R), having 3 translation degrees of freedom at each and
every node. Normally, a coefficient friction of 0.25 is used in the simulation models of carbon
steel CFST columns (Han et al., 2007; Tao et al., 2009).
Chromium oxide is chemically stable and is used to cover the surface of a stainless steel for
corrosion resistance, the coefficient of friction and even behavior of the bond between the
stainless-steel tube and concrete core is different from that of conventional carbon CFDST steel
column.
2.4 Buckling of CFDST short columns
When a square CFDST column is subjected to axial compressive forces, the double steel tube
constraints the laterally expanding concrete, creating a compressional effect of the infill concrete
into multidirectional state of compression. This increases the compressive strength and
deformation capacity of the infill concrete, while the infill concrete restrains the local buckling
of the steel tubes. It is for this reason that (Chung, 2010)argues that CFDST columns exhibit
exceptionally high performance.
Unlike the conventional reinforced concrete (RC), the CFDST columns have reinforcement
positioned at the perimeter of the cross section of the column. This maximizes the flexural
capacity of the steel section.
The use of thin-walled steel in CFDST columns to bear compressive strength is often associated
by failure through local buckling. This has an effect of reducing the ductility and compressive
strength of the infill concrete. For a concrete filled square CFDST column, the confinement
effect creates restraints hence the steel tube can only buckle outwards. This is called local
outward buckling.
When a CFDST column composite experiences outward buckling, there is a significant increase
of the critical local buckling strength of the steel and also the load carrying capacity of the
column. Of importance is that the steel plates encasing the concrete also serve as permanent
formwork since the wet concrete is poured into it, thereby increasing the speed of construction
by avoiding prefabricated formworks which take time to erect and remove when one phase of
construction is complete.

12
This formwork also acts as longitudinal reinforcement of the composite hence economical
designs are possible. Because of these properties, the composite is known to have exceptionally
high performance because of its high strength, large amounts of energy absorption and high
ductility as compared to other composite materials. The presence of high strength concrete and
steel plates of high yield strength encourages the adoption of CDFST columns in construction
projects. However, their use is associated with the local instability problem of these thin-walled
steel plates especially when they are subjected to compressive forces and in-plane bending as
explained by (El-Heweity, 2012). When these thin walled steel plates undergo local buckling,
there is a notable reduction in strength and ductility performance, especially when the steel plates
have initial geometric imperfections and residual stresses caused by processes such as welding.
A number of studies have been conducted to explore the local stability of thin-walled steel plates
under compressive forces and in-plane bending mode. A research study done by (Ellobody,
2013) employed the Galerkin’s method to provide a solution to the governing nonlinear
simultaneous differential equation. A similar study done by Rhodes and Harvey (1971) used
steel plates that are simply supported at the loaded edges to examine the effects of load
eccentricity on local and post local buckling of the CFSDT columns, while varying the type of
support conditions at the unloaded edges. A report of load carrying capacity of thin steel plates
with initial imperfections under linearly varying displacement was given by Rhodes et al. (1975).
These studies are consistent in the results observed when local buckling occurs on thin steel
plates, describing the basic behavior of local and post-local buckling of thin-walled steel plates
for square CFDST short columns.
Further studies done by Usami (1993) incorporated energy methods and other analysis methods.
Steel plates that are simply supported were used and subjected to axial compression and bending
moments about major and minor axes. This enabled the development of effective and reliable
numerical width formulas that can be used in the prediction of ultimate strength of the steel
plates. Both analytical and experimental research works have been conducted by Narayanan and
Chan (1985) to investigate the behavior of elastic critical local and post-local buckling strengths
of steel plates. The unique feature of the study was that the steel plates had holes in them and
were then subjected to a varying range of edge displacements. An investigation into the ultimate
loads of a box column with this steel walls was done by (Shanmugam, N. E. and Lakshmi, B. ,
2001). The study also incorporated elements of local buckling in steel plates that were simply
supported and subjected to compressive forces as well as in-plane bending. In all the mentioned
studies above, it is clear that free buckling was allowed in the two lateral directions.