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

Literature Overview

The study by Cheng Gao, Zhang Zhengqi, and Liu Yongjian, published in the "Journal of Hefei University of Technology" (Vol. 42, No. 7, 2019, pp. 947-951), investigates the interaction between steel tubes and confined concrete in square steel tube concrete (STC) short columns under various loading conditions. Funded by the National Natural Science Foundation of China (Grants 51778058, 51508027) and Shaanxi Provincial Natural Science Foundation (Grant 2018JQ5219), this research employs nonlinear finite element analysis to characterize the interface normal stress distribution and the confinement effect (hooping action) between the steel tube and the concrete core.

Core Technical Content and Key Points

Research Motivation and Methodology

Steel tube concrete (STC) structures have gained widespread adoption in bridge piers, building columns, and offshore platforms due to their superior load-bearing capacity, ductility, and construction efficiency compared to conventional reinforced concrete. The confinement effect—the lateral restraint provided by the steel tube on the concrete core—is the fundamental mechanism that enhances the compressive strength and ductility of STC members. However, the precise characterization of this confinement effect, particularly its spatial distribution and activation threshold, has remained an area of uncertainty in design codes.

The authors established nonlinear finite element models in ABAQUS, incorporating:

• Material nonlinearity: Concrete modeled with the Concrete Damage Plasticity model; steel modeled with bilinear isotropic hardening
• Interface behavior: Frictional contact with penalty method, coefficient of friction μ = 0.4
• Loading scenarios: (1) Combined loading of steel and concrete, (2) Concrete-only loading, (3) Steel-only loading

Key Findings on Interface Normal Stress Distribution

The study reveals several critical observations regarding the interface normal stress (radial pressure between steel tube wall and concrete):

The "trough-shaped" distribution indicates that the interface normal stress is highest at the center of each wall face and decreases toward the corners. This distribution pattern is consistent with the Poisson effect: as concrete is compressed axially, it expands laterally, pressing against the steel tube wall. The stress concentration at the wall center occurs because corner regions are constrained by adjacent walls, reducing the lateral expansion demand.

Confinement Effect Activation and Effective Length

One of the most significant findings is that the confinement effect (hooping action) activates only after the axial load reaches approximately 0.5 times the ultimate load capacity. Before this threshold, the concrete and steel tube behave largely independently, with minimal interaction at the interface. This has direct implications for design:

• The effective confinement length is approximately 0.55 times the cross-section width
• Outside this effective zone, the interface normal stress is nearly zero
• This "end effect" is analogous to the stress concentration phenomenon observed in bolted joints and welds

Implications for Design Codes

Current design codes (GB 50017, Eurocode 4, AISC 360) typically assume uniform confinement across the entire cross-section, which is a simplification. The findings suggest:

1. Load-dependent confinement: Design equations that assume constant confinement pressure throughout the loading history may overestimate the effective confinement at low load levels and underestimate it near ultimate capacity.
2. Geometric limitations: For square sections with width-to-thickness ratios exceeding certain limits, the effective confinement zone may not cover the full cross-section, reducing the overall confinement benefit.
3. Corner regions: The reduced interface stress at corners suggests that the confinement effect is less effective in preventing corner concrete spalling, which may be relevant for seismic design.

Engineering Practice Integration

From a fabrication and welding perspective, the findings have several practical implications for steel tube concrete column production:

• Steel tube geometry: The "trough-shaped" stress distribution implies that the wall thickness distribution in square tubes should ideally be optimized, with potentially thicker walls at the center of each face. However, manufacturing constraints typically require uniform wall thickness, making this a design consideration rather than a fabrication one.
• Welding of built-up square tubes: When square tubes are fabricated from flat plates welded at the corners, the corner welds experience complex stress states due to the reduced interface pressure in those regions. Weld procedure qualification should account for the multi-axial stress state.
• Concrete pouring and compaction: Ensuring complete filling of the steel tube is critical, as voids would eliminate the confinement effect entirely. The effective confinement length of 0.55 times the section width suggests that concrete placement from both ends of a column may result in inadequate compaction in the middle region.

Study Insights and Reflections

This research contributes valuable quantitative data to the understanding of STC behavior, particularly the load-dependent nature of the confinement effect. The threshold of 0.5 times ultimate load for confinement activation is a practically important parameter that should inform both design and testing protocols. Engineers designing STC columns should recognize that the confinement benefit is not instantaneous but develops progressively as the axial load increases, which has implications for serviceability limit state checks at lower load levels.

The finding that the effective confinement length is limited to approximately 0.55 times the section width raises important questions about the applicability of STC technology for very wide sections. For sections exceeding 1000 mm in width, the confinement effect may not effectively constrain the central concrete region, potentially limiting the strength enhancement achievable through confinement. Future research should investigate whether internal steel reinforcement or additional confinement elements can extend the effective confinement zone for large sections.