Nonlinear Finite Element Analysis of Square CFST Frames Under Low-Cycle Reversed Loading

Literature Overview

This paper by Wang Tiecheng, Lu Mingqi, and Rong Xian from Tianjin University (2004) presents a three-dimensional nonlinear finite element analysis of square concrete-filled steel tube (CFST) frames subjected to low-cycle reversed loading. The study was supported by the Tianjin Municipal Commission of Housing and Urban-Rural Development major science and technology project (01-2-073), and was published in the journal Earthquake Engineering and Engineering Dynamics, Volume 24, Issue 4, pages 82–85. The core objective was to develop and validate a numerical model that accounts for the bond-slip interaction between the steel tube and the infill concrete, and to assess the seismic performance of such frames through comparison with experimental hysteresis and stiffness degradation curves.

Core Technical Content and Model Description

The numerical model established in this study is a three-dimensional nonlinear finite element model that explicitly considers the bond-slip mechanism between the steel tube and the enclosed concrete. This is a critical technical decision, as many simplified models treat the steel and concrete as perfectly bonded, which can significantly overestimate the frame's lateral stiffness and underestimate the energy dissipation capacity. The bond-slip interface is modeled using contact elements or interface elements that capture the progressive separation and re-engagement of the two materials under cyclic loading.

The model incorporates the following key material constitutive relationships:

• Steel tube: bilinear or multilinear isotropic hardening model with kinematic hardening to capture the Bauschinger effect under reversed loading
• Core concrete: confined concrete model (e.g., Mander or Kent-Park) that accounts for the lateral confinement pressure exerted by the steel tube
• Bond-slip interface: a nonlinear traction-separation law that transitions from elastic bonding through partial slip to full debonding

The frame geometry, boundary conditions, and loading protocol are calibrated to match the experimental test setup. Displacement-controlled cyclic loading is applied at the beam ends, with drift ratios typically ranging from 0.5% to 4.0%, representing serviceability through collapse-level seismic demands.

Validation and Key Findings

The model's accuracy is validated by comparing the computed hysteresis loops and stiffness degradation curves against experimental data obtained from physical low-cycle reversed loading tests on square CFST frames. The comparison shows that the numerical predictions are in reasonably good agreement with the test results, particularly in the following aspects:

The study concludes that the square CFST frame possesses favorable seismic performance, characterized by good ductility, stable energy dissipation, and progressive degradation of lateral stiffness rather than sudden brittle failure. The inclusion of bond-slip in the model improves prediction accuracy compared to perfectly bonded models, particularly at drift ratios beyond 2.0%.

Integration with Engineering Practice

From a steel pipe manufacturing and welding perspective, the findings of this study have direct implications for the design and fabrication of structural steel tubes used in seismic regions. The following practical considerations emerge:

1. Steel tube quality control: The local buckling of the steel tube is a critical failure mode. Manufacturing tolerances for wall thickness, out-of-straightness, and ovality must be tightly controlled. ERW or HFW welded tubes should meet API 5L or GB/T 9711 requirements for dimensional accuracy and mechanical properties.
2. Welding quality at joints: The beam-column connections in CFST frames are typically welded or bolted. The weld quality at these connections directly affects the frame's seismic performance. GTAW or FCAW processes are commonly used, and the welds must be inspected by RT or UT to detect lack of fusion, porosity, and undercut defects.
3. Concrete infill quality: The concrete placed inside the steel tube must achieve adequate compaction despite the confined geometry. Pumping pressure, slump, and admixture selection are critical. For high-rise CFST columns, the concrete strength grade typically ranges from C50 to C80.
4. Bond-slip mitigation: The study highlights that bond-slip reduces the effective stiffness of CFST members. In practice, this can be mitigated through the use of shear keys, helical stirrups inside the tube, or the use of self-compacting concrete with enhanced adhesion properties.

Key Questions and Reflections

A significant question that arises from this study is whether the bond-slip model used adequately captures the three-dimensional nature of the interface behavior. In reality, the bond-slip at the steel-concrete interface is not uniform around the perimeter of the square tube; it varies with the loading direction, the position within the member, and the stage of deformation. A more refined model with spatially varying interface properties could improve prediction accuracy, particularly in the post-peak region.

Another reflection is that the study focuses on square CFST frames, but circular CFST frames are more common in practice due to their superior confinement efficiency and simpler fabrication. The bond-slip behavior in circular tubes is inherently more uniform, and the confinement effect is more predictable. Future research should extend the model to circular and other cross-sectional shapes to broaden its applicability.

Study Insights and Implications

The most valuable contribution of this paper is the demonstration that a properly formulated bond-slip interface in a nonlinear FEA model can reproduce the experimentally observed hysteresis and stiffness degradation behavior of CFST frames with acceptable accuracy. This validates the use of numerical simulation as a complementary tool to physical testing, particularly for parametric studies and design optimization where full-scale experiments are impractical.

For engineers involved in steel pipe supply and structural welding, the practical implication is that the seismic performance of CFST structures depends not only on the material properties of the steel tube and concrete but also on the quality of the interface between them. This means that surface treatment of the steel tube interior, concrete mix design, and placement procedures all play a role in the ultimate structural performance. Quality assurance programs should incorporate interface bond testing in addition to standard mechanical property verification of the steel tube and concrete.

In summary, this study provides a validated numerical framework for the seismic assessment of square CFST frames, with the key technical advancement being the explicit modeling of bond-slip behavior, which significantly improves prediction accuracy over simplified perfectly bonded models.