Interface Normal Stress Analysis of Square Steel Tube Concrete Axially Compressed Short Columns

Overview and Context

The paper by Cheng Gao, Zhang Zhengqi, and Liu Yongjian, published in the Journal of Hefei University of Technology (2019, Vol. 42, No. 7, pp. 947-951), presents a nonlinear finite element analysis of the interface normal stress between the steel tube and concrete in square steel tube concrete (STC) axially compressed short columns. The study is supported by the National Natural Science Foundation of China (Grants 51778058, 51508027) and the Shaanxi Provincial Natural Science Foundation (Grant 2018JQ5219), reflecting the significant research interest in composite structural systems.

This work is particularly relevant to the steel pipe and steel tube industry because the behavior of the steel-concrete interface governs the overall performance of composite structural members. Understanding the confining effect (套箍作用) and the distribution of interface normal stress is essential for the rational design of steel tube concrete columns used in high-rise buildings, bridges, and infrastructure projects.

Core Technical Content

Load Modes and Finite Element Modeling

The authors establish nonlinear finite element models for three distinct loading modes:

Combined loading: Both the steel tube and concrete are loaded simultaneously (the typical design condition).
Concrete-only loading: Only the concrete core is loaded, with the steel tube acting as a confining shell.
Steel-tube-only loading: Only the steel tube is loaded, with the concrete acting as a passive core.

The finite element models employ a nonlinear material model for concrete (accounting for cracking and crushing) and an elastic-plastic model for the steel tube (accounting for yielding and post-yield hardening). The interface between the steel tube and concrete is modeled using contact elements that allow for the transfer of normal stress (compressive confining force) while preventing tensile separation.

Interface Normal Stress Distribution

The study reveals several important findings regarding the distribution of interface normal stress:

Load Mode	Stress Distribution Pattern	Relative Magnitude	Confining Effect Onset
Steel tube only	Trough-shaped (U-shaped) along width	Lowest	N/A (no confining action)
Combined loading	Trough-shaped along width	Medium	After 0.5 × ultimate load
Concrete only	Trough-shaped along width	Highest	Immediately upon loading

The "trough-shaped" (槽形) distribution pattern along the section width indicates that the interface normal stress is highest near the corners of the square section and lowest at the mid-span of each face. This pattern is consistent with the Poisson effect: as the concrete is compressed axially, it expands laterally, pressing against the steel tube. The corners of the square section experience greater lateral expansion due to the geometric constraint, resulting in higher interface normal stress.

Confining Effect and Effective Confinement Length

One of the most significant findings is that the confining effect in combined loading mode only becomes significant after the axial load reaches 0.5 times the ultimate load. Before this threshold, the steel tube and concrete behave largely independently, with minimal interaction through the interface. This finding has important implications for the design of STC columns: the confining effect, which is often relied upon for ductility and post-yield behavior, is not available during the elastic stage.

The effective confinement length is found to be approximately 0.55 times the section width. Beyond this length from the loaded end, the interface normal stress drops to near zero, indicating that the confining action is localized near the load application points. This phenomenon is related to the stress diffusion characteristic of composite columns and has implications for the design of connection details and the distribution of reinforcement.

Engineering Implications for Steel Tube Design

Design Parameter	Influence on Interface Stress	Design Recommendation
Section width	Larger width reduces stress concentration at corners	Consider rounded corners for improved stress distribution
Steel tube thickness	Thicker tube reduces local buckling and increases confinement capacity	Minimum thickness per relevant code (e.g., GB 50936)
Concrete strength	Higher strength increases lateral expansion pressure	Match concrete strength to steel grade for balanced behavior
Steel grade	Higher strength steel provides greater confinement resistance	Use Q345 or Q355 for high-strength applications
Loading mode	Combined loading is the most critical for design	Design for combined loading with appropriate safety factors

Integration with Engineering Practice

In my experience with steel tube concrete structures, the findings of this paper align with practical observations from construction and testing. The delayed onset of the confining effect (after 50% of ultimate load) explains why STC columns can appear to behave linearly during the initial loading stages, only to exhibit significant nonlinearity and ductility near failure. This behavior is critical for seismic design, where the ability of a column to sustain large deformations without catastrophic failure is paramount.

The effective confinement length of 0.55 times the section width has practical implications for the design of column connections. If the connection detail extends beyond this length from the loaded end, the confining effect will not contribute to the connection's load-carrying capacity. Engineers should therefore ensure that connection details are designed to transfer loads independently of the confining action, or that the connection geometry is optimized to take advantage of the confined region.

From a fabrication standpoint, the interface stress distribution highlights the importance of achieving a tight fit between the steel tube and the concrete core. Any gaps or voids at the interface will reduce the effectiveness of the confining action and may lead to premature concrete crushing. The steel tube must therefore be fabricated with tight dimensional tolerances, and the concrete must be placed with adequate compaction to ensure full contact with the steel tube walls.

Key Questions and Reflections

Several questions arise from this study that warrant further investigation:

How does the interface normal stress distribution change under cyclic loading conditions, as would be experienced in seismic events? The monotonic loading models used in this study may not capture the degradation of interface contact under repeated loading.
What is the effect of surface roughness and coating on the steel tube on the interface bond and stress transfer? Real steel tubes may have surface treatments that affect the contact mechanics.
How do the findings for square sections compare with those for circular sections? The trough-shaped distribution is specific to the square geometry, and circular sections would exhibit a uniform distribution around the circumference.

Study Insights and Implications

This paper provides valuable insight into the mechanics of steel tube concrete columns, particularly regarding the interface behavior that governs the composite action between the steel tube and concrete. The findings underscore the importance of understanding the load-dependent nature of the confining effect and the localized nature of the confinement zone. For engineers designing and fabricating steel tube concrete structures, the key takeaway is that the interface is not a passive boundary but an active load-transfer mechanism whose behavior evolves with the loading state. Rational design must account for this evolution, ensuring that the steel tube is adequately thick to resist the confining pressure without local buckling, and that the concrete is properly confined to achieve the desired ductility and energy dissipation capacity. The finite element approach employed in this study provides a powerful tool for investigating these complex interactions, and its results should be validated against experimental data to ensure reliability in design applications.