Abstract
Concrete-filled steel tube columns are widely used in civil engineering structures due to their excellent ductility, energy absorption capacity, ultimate load-bearing capacity, and seismic behavior. In this paper, a numerical study modeling of eight lightweight concrete and conventional concrete filled steel tubes was carried out using ABAQUS software, and the lateral load-carrying capacity of square and circular steel tubes under cyclic load was compared. The quarter and one-third height of the tubes was filled with concrete with respect to the pier's height, to improve the base performance of the piers. The results show that the capacity of steel tubes filled with lightweight concrete increased by 40%–70% regarding energy absorption. The square tubes showed better performance than the circular tubes in terms of yielding load, yielding displacement, and energy dissipation.
1 Introduction
A concrete filled steel tube Fig. 1 is a composite system that ideally combines the advantages of steel and concrete materials, providing a higher load-bearing capacity due to the superposition of axial strength of the concrete core [1], and the encased steel tube. In addition to high load-bearing capacity, Concrete-Filled Steel Tube (CFST) columns also have high strength, high ductility, and good energy absorption capacities. For these reasons these column systems are widely utilized in modern engineering works for example high-rise buildings and bridges [2]. Because concrete prevents the local buckling of hollow steel sections and increases ductility significantly, steel tubes act as longitudinal and lateral reinforcement for concrete cores making them preferable in high seismic regions [3]. No additional reinforcement is needed for CFST piers apart from the steel tube, which encases the concrete core, and enhances the core's strength, ductility and prevents the concrete from crashing.

Cross-section of the concrete-filled steel tube of a bridge pier
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618

Cross-section of the concrete-filled steel tube of a bridge pier
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
Cross-section of the concrete-filled steel tube of a bridge pier
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
2 Research significance
Many researchers of CFST, were considered filling the whole steel tubes with concrete to get higher resistance, and capacity due to the applied loads. Hence this method shows a new attempt aims to increase the capacity of the steel tubes, by filling the critical parts only with Conventional Concrete (CC) or LightWeight Concrete (LWC) where plastic hinges formed, which is the lower part of the piers due to the huge base shear, and bending moment, so by this will not cause an extra vertical loads on the Pier's foundation or the bridge itself as a whole in case of filling the entire steel tube besides, using LWC, which considered less weight than CC, in brief this method leads to higher capacity with less amount of concrete and less vertical weight.
3 Materials and methods
3.1 Properties of the materials used
C50 and C50.4 grade concrete was used for the modeling of the CC, and LWC respectively, the Poisson ratio is considered as 0.2 for both grades of concrete, and the density of the conventional concrete used was 2,400 kg m−3 and 1,400 kg m−3 for lightweight concrete, Young's modulus was 33,400 N mm−2 and 21,400 N mm−2, respectively. The properties of conventional concrete were derived according to [4] and for the lightweight concrete according to [5].
S355 was the steel grade used in all the modeling, with 18 mm walls, according to Eurocode [6], the yield and ultimate stress of the steel are 355 and 470 MPa, respectively; the Young's modulus of steel is 210,000 MPa. No reinforcement was used in the models because the concrete acted as the filling material.
3.2 Model's specification and detailing
As stated previously there were two types of concrete used and, in addition, there were two heights of concrete inside the tube. The height of the piers was 5 m for all models, quarter height = 0.25 H =1.25 m and one-third height = 0.333 H = 1.65 m, were the two heights of concrete inside the steel tubes. In order to simplify the model's results, Table 1 shows the model Identification (ID), category, and abbreviation.
State of all types with ID and abbreviation
Model ID | Category | Abbreviation |
SST | Steel (NC) | Square Steel Tube |
CST | Circular Steel Tube | |
CCL4S | Conventional concrete (CC) | Conventional Concrete Filled Steel Tube -(Length/4 = quarter) - Square Steel Tube |
CCL4C | Conventional Concrete Filled Steel Tube -(Length/4 = quarter) - Circular Steel Tube | |
CCL3S | Conventional Concrete Filled Steel Tube - (Length/3 = one-third) - Square Steel Tube | |
CCL3C | Conventional Concrete Filled Steel Tube -(Length/3 = one-third) - Circular Steel Tube | |
LWCL4S | Lightweight concrete (LWC) | Lightweight Concrete Filled Steel Tube -(Length/4 = quarter) - Square Steel Tube |
LWCL4C | Lightweight Concrete Filled Steel Tube -(Length/4 = quarter) - Circular Steel Tube | |
LWCL3S | Lightweight Concrete Filled Steel Tube -(Length/3 = one-third) - Square Steel Tube | |
LWCL3C | Lightweight Concrete Filled Steel Tube -(Length/3 = one-third) - Circular Steel Tube |
The concrete was modeled using Finite Element Analysis (FEM) software ABAQUS and to simulate the plastic behavior of the concrete considered in FEM software concrete damage plasticity is used. Hence the stress-strain behavior for both CC and LWC were included in the analysis as well as described in [7].
As it is shown in Table 1, the two types of steel profile are shown in Fig. 2 (all dimensions are in millimeters).

Square and circle steel dimensions
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618

Square and circle steel dimensions
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
Square and circle steel dimensions
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
4 Analytical study
4.1 Finite element modeling
All models either filled or not with concrete, are shown in Figs 3 –5 with a longitudinal cross-section for each. The steel part was taken as a shell element with rectangular mesh, and the concrete part as a solid element with hexahedral mesh, with considering matching the size and joints between shells and solids elements.



CCL4C, LWCL4C and CCL4S, LWCL4S models cross-sections
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618

CCL4C, LWCL4C and CCL4S, LWCL4S models cross-sections
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
CCL4C, LWCL4C and CCL4S, LWCL4S models cross-sections
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618

CCL3C, LWCL3C and CCL3S, LWCL3S models cross-sections
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618

CCL3C, LWCL3C and CCL3S, LWCL3S models cross-sections
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
CCL3C, LWCL3C and CCL3S, LWCL3S models cross-sections
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
The interaction between the concrete and steel was taken in normal direction as a hard contact, and in the tangential direction as a friction contact with 0.3 friction factor according to Ding et al. [8] and Li [9]. The same cyclic load as Yadav [10] shown in Fig. 6 was applied on all models at the top of the bridge piers. The bottom side had fixed support.


5 Results and discussion
Each variation, shown in Table 1, was modeled and a stress-strain graph computed for the steel and concrete section.
Figure 7 shows the stress of the steel tubes for both square and circular tubes respectively.


For the concrete-filled steel tube using lightweight concrete, the stress of the steel and concrete elements are shown in terms of Von-Mises principles in Figs 8 and 9. Generally, the maximum deformed part near the bottom side as Vulcu et al. [11] and Danku et al. [12], which has red color in the figures referred to the part of steel tube and concrete, which reached their ultimate stress limit of failure, 470 MPa for steel and 50 MPa for concrete.

LWCL4S and LWCL4C steel stress contour
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618

LWCL4S and LWCL4C steel stress contour
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
LWCL4S and LWCL4C steel stress contour
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618

LWCL3S and LWCL3C steel stress contour
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618

LWCL3S and LWCL3C steel stress contour
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
LWCL3S and LWCL3C steel stress contour
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
The square and circular steel tubes deformed at the bottom part of the piers where they were fixed but had different characteristics. The square concrete profile inside the square steel tubes shows the stresses under compression were concentrated at the corners on the upper face of the concrete as it can be seen in Fig. 10, which shows the concentration of stresses at the corners only in red color then the stresses decrease inward, so these corners has reached to the ultimate limit. In contrast, Fig. 11 shows the circular concrete profile inside the circular steel tube and here the stress of compression from the cyclic load was spread over a wider area on the concrete bottom face, and completely different compare to the cubic profile.

Concrete stress in case of quarter's and one-third height of the square piers
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618

Concrete stress in case of quarter's and one-third height of the square piers
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
Concrete stress in case of quarter's and one-third height of the square piers
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618

Concrete stress in case of quarter's and one-third height of the circular piers
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618

Concrete stress in case of quarter's and one-third height of the circular piers
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
Concrete stress in case of quarter's and one-third height of the circular piers
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
Table 2 shows the proportions and mass of materials used in the study. Square profiled steel sections were slightly heavier than the circular profiles, but the biggest difference was whether lightweight concrete or conventional concrete was used. As it is shown in Table 2 LWC was 41.67% lighter than CC.
Weight of steel, concrete and the reduction in weight
Model ID | Weight of Steel (kg) | Weight of Concrete (kg) | Reduction in Weight % |
SST | 128.6 | - | 0 |
CST | 101.3 | - | 0 |
CCL4S | 128.6 | 2017.2 | 0 |
CCL4C | 101.3 | 1584 | 0 |
CCL3S | 128.6 | 2664 | 0 |
CCL3C | 101.3 | 2088 | 0 |
LWCL4S | 128.6 | 1176.7 | 41.67 |
LWCL4C | 101.3 | 924 | 41.67 |
LWCL3S | 128.6 | 1554 | 41.67 |
LWCL3C | 101.3 | 1218 | 41.67 |
Along with the cyclic load applied on each model, Table 3 shows the results in terms of the yielding load, ultimate load, yielding displacement, ultimate displacement and energy dissipation, where dissipated energy is calculated by measuring the area of the loops for each model using AutoCAD software.
Yielding and ultimate load, displacement and energy dissipation
Model ID | Fy (kN) | Fu (kN) | Dy (mm) | Du (mm) | Energy dissipation |
SST | 1609.9 | 2003.3 | 30.5 | 49.01 | 1.67 |
CST | 1140.1 | 1301.5 | 37.5 | 56.25 | 1.52 |
CCL4S | 1878.7 | 1893.6 | 37.2 | 40.6 | 2.39 |
CCL4C | 1322.7 | 1361.8 | 50.0 | 59.3 | 2.28 |
CCL3S | 1968.4 | 1961.01 | 40.4 | 47.82 | 2.86 |
CCL3C | 1239.1 | 1412.5 | 39.1 | 68.3 | 2.26 |
LWCL4S | 1875.7 | 1895.4 | 33.6 | 38.4 | 2.35 |
LWCL4C | 1198.4 | 1377.9 | 39.1 | 68.3 | 2.27 |
LWCL3S | 1924.2 | 1925.1 | 40.6 | 49.4 | 2.85 |
LWCL3C | 1230.8 | 1406.9 | 39.6 | 68.5 | 2.25 |
Table 3 shows that LWC gives similar results to the CC models with the same cross-section, which means that using lightweight concrete has the same capacity for absorbing the cyclic applied load but has 41.67% less weight in concrete.
Comparing the circular cross-section with the square cross-section results in Fig. 12, the circular piers give less capacity in terms of yielding load compared to the square profile.


Figures 13 and 14 compare the square and circular profiles for both quarter and one-third heights of tubes filled with concrete. The square profile gives a higher-yielding capacity and also higher energy absorption, the circular section shows 97% of the energy absorption compared with square tubes in the case of quarter height of tubes filled with concrete and 80% for one-third height of tubes filled with concrete.

Cyclic loops of LWCL4S and LWCL4C
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618

Cyclic loops of LWCL4S and LWCL4C
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
Cyclic loops of LWCL4S and LWCL4C
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618

Cyclic loops of LWCL3S and LWCL3C
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618

Cyclic loops of LWCL3S and LWCL3C
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
Cyclic loops of LWCL3S and LWCL3C
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
Figure 15 shows the effectiveness of filling the steel tube with LWC, which means the bottom part of the tube has higher friction between the concrete and the steel, which means the cyclic loop of model LWCL4S is 40% higher than the square steel tube without it, and 70% higher for LWCL3S model.

Cyclic loops of SST, LWCL4S and LWCL3S
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618

Cyclic loops of SST, LWCL4S and LWCL3S
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
Cyclic loops of SST, LWCL4S and LWCL3S
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
Similarly, Fig. 16 shows the effectiveness of filling the circular tube with LWC, both LWCL4C and LWCL3C models give 49% higher capacity than the ordinary circular steel tube.

Cyclic loops of CST, LWCL4C and LWCL3C
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618

Cyclic loops of CST, LWCL4C and LWCL3C
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
Cyclic loops of CST, LWCL4C and LWCL3C
Citation: Pollack Periodica 17, 3; 10.1556/606.2022.00618
6 Conclusion
This study involves the analytical investigation on CFST using ABAQUS software, The following conclusions are drawn based on finite element analysis:
-
Lightweight concrete gives almost similar performance to conventional concrete models for both square and circular tubes;
-
Square tubes show higher capacity in terms of yielding load compared to circular tubes;
-
Circular tubes show higher capacity in yielding displacement compared to square tubes;
-
CCL3S shows the highest capacity in terms of energy dissipation compared to others;
-
Absorbed energy of CCL4C is equal to 97% from the CCL4S capacity and 80% in the case of CCL3C compare to CCL3S;
-
Absorbed energy of LWCL4C is equal to 96% from the LWCL4S capacity and 78% in the case of LWCL3C compare to LWCL3S;
-
Introducing lightweight concrete in construction as a filler material reduce the vertical loads of the bridge piers by 41.7% in addition getting similar performance compare to conventional concrete which leads to lower costs for a project as whole.
References
- [1]↑
K. Sakino and H. Ishibashi , “Experimental studies on concrete-filled square steel tubular short columns subjected to cyclic shear force and constant axial force,” J. Struct. Construction Eng., no. 353, pp. 82–83, 1985.
- [2]↑
B. Evirgen , A. Tuncan , and K. Taskin , “Structural behavior of concrete-filled steel tubular sections (CFS/CFST) under axial compression,” Thin-Walled Struct., vol. 80, pp. 46–56, 2014.
- [3]↑
J. G. Nie , Y. H. Wang , and J. S. Fan , “Experimental study on seismic behavior of concrete-filled steel tube columns under pure torsion and compression–torsion cyclic load,” J. Constructional Steel Res., vol. 79, pp. 115–126, 2012.
- [4]↑
M. Hafezolghorani , F. Hejazi , R. Vaghei , M. S. B. Jaafar , and K. Karimzade , “Simplified damage plasticity model for concrete,” Struct. Eng. Int., vol. 27, no. 1, pp. 73–74, 2017.
- [5]↑
C. Koh , M. Teng , and T. H. Wee , “A plastic-damage model for lightweight concrete and normal weight concrete,” Int. J. Concrete Structures Mater., vol. 2, no. 2, pp. 123–136, 2008.
- [6]↑
EN 1993-1-8, 2005 , “Eurocode 3: Design of steel structures, Part 1-8: Design of joints,” BSI, London, 2005.
- [7]↑
X. Dai and D. Lam , “Numerical modeling of the axial compressive behaviour of short concrete-filled elliptical steel columns,” J. Constr. Steel Res., vol. 66, no. 7, pp. 931–942, 2010.
- [8]↑
F. X. Ding , Y. F. Chen , Y. J. Yu , L. P. Wang , and Z. W. Yu , “Composite action of rectangular concrete-filled steel tube columns under lateral shear force,” Struct. Concrete, vol. 22, no. 2, pp. 726–740, 2020.
- [9]↑
P. Li , T. Zhang , and C. Wang , “Behavior of concrete-filled steel tube columns subjected to axial compression,” Adv. Mater. Sci. Eng., vol. 2018, 2018, Paper no. 4059675.
- [10]↑
R. Yadav , B. Chen , H. Yuan , and Z. Lian , “Analytical behavior of CFST bridge piers under cyclic loading,” Proced. Eng., vol. 173, pp. 1731–1738, 2017.
- [11]↑
C. Vulcu , A. Stratan , and D. Dubina , “Numerical simulation of the cyclic loading for welded beam-to-CFT column joints of dual-steel frames,” Pollack Period., vol. 7, no. 2, pp. 35–46, 2012.
- [12]↑
G. Danku and D. Dubina , “Extensive study of plastic hinges in composite steel-concrete members subjected to shear and/or bending,” Pollack Period., vol. 1, no. 1, pp. 37–46, 2011.