Experimental Analysis of Eccentric Compression Bearing Capacity of Concrete-Filled Steel Tubes

Literature Overview

This paper published in the Journal of Fuzhou University (Natural Science Edition) in 2002 (Vol. 30, No. 6, pp. 838-844) by Chen Baochun, Ou Zhijing, Wang Laiyong, and Han Linhai from Fuzhou University presents experimental investigations on the eccentric compression behavior of concrete-filled steel tube (CFST) columns. The study was supported by the National Natural Science Foundation of China (Grant 50078016) and the Open Fund of the State Key Laboratory of Disaster Prevention in Civil Engineering. The experimental matrix included variations in eccentricity ratio, steel ratio, concrete compressive strength, and confining coefficient, providing a comprehensive parametric study that is still highly relevant to modern CFST design practice.

Experimental Program and Test Configuration

Specimen Parameters

The test specimens were circular concrete-filled steel tube columns subjected to eccentric compressive loading. The key experimental parameters were systematically varied to isolate the influence of each factor on the structural response. The eccentricity ratio (e/h, where e is the eccentricity distance and h is the column depth) was the primary variable, representing the degree of bending moment relative to axial force. The material parameters included the steel ratio (ρ_s, defined as the ratio of steel tube cross-sectional area to total cross-sectional area), concrete compressive strength (f_c), and confining coefficient (ξ, defined as the ratio of steel tube confining stress to concrete compressive strength).

Loading and Measurement Setup

The eccentric compression tests were conducted under displacement-controlled loading, with load cells measuring the applied axial force and eccentricity moment, while strain gauges and displacement transducers captured the deformation response. The test setup was designed to simulate realistic eccentric loading conditions encountered in building columns and bridge piers, where both axial compression and bending moments act simultaneously. The loading was applied monotonically until failure, allowing the complete load-displacement curve and failure mode to be recorded.

Core Technical Findings

Effect of Eccentricity Ratio on Bearing Capacity

The most significant finding of this study is that the eccentricity ratio has a profound effect on the bearing capacity and failure mode of CFST columns. As the eccentricity ratio increases, the confining effect of the steel tube on the core concrete progressively weakens, leading to a marked reduction in ultimate bearing capacity. This behavior is fundamentally different from plain concrete columns, where the eccentricity ratio effect is more linearly proportional. In CFST columns, the confining effect is most effective under pure axial compression, where the steel tube uniformly constrains the core concrete in all directions. Under eccentric loading, the compression zone and tension zone of the cross-section experience different levels of concrete confinement, with the tension zone experiencing reduced or even loss of confinement as the steel tube yields and opens.

Effect of Material Parameters

The steel ratio and concrete strength both positively influence the bearing capacity, but their interaction with the eccentricity ratio is non-linear. Higher steel ratios provide greater confining capacity, which partially compensates for the loss of confinement at higher eccentricity ratios. Higher concrete strengths increase the compressive capacity of the core, but also reduce the ductility of the member, leading to more brittle failure modes at high eccentricity ratios. The confining coefficient ξ, which encapsulates the combined effect of steel tube thickness, steel yield strength, and concrete strength, was found to be the most effective single parameter for predicting the bearing capacity across different eccentricity ratios.

Comparison with Design Codes

The authors compared the experimental results with six different calculation methods from international design codes and standards, including Chinese codes, Eurocode 4, Japanese AIJ recommendations, and American ACI provisions. The comparison revealed significant discrepancies between code predictions and experimental values, particularly at intermediate eccentricity ratios where the transition from axial compression behavior to flexural behavior occurs. Most codes tend to be conservative at low eccentricity ratios (pure axial compression region) and non-conservative at high eccentricity ratios (flexure-dominated region), indicating that the interaction curves in current codes need refinement for CFST members.

Parametric Analysis and Design Implications

Interaction Between Confinement and Eccentricity

The following table summarizes the key trends observed in the parametric study:

Engineering Practice Considerations

From a practical standpoint, the findings have direct implications for the design of CFST columns in multi-story buildings, bridge piers, and offshore platforms. For columns subjected to significant eccentric loading, such as those at building corners or bridge pier bases, the designer should not rely solely on the enhanced bearing capacity predicted by CFST interaction formulas developed for axial compression. Instead, a more nuanced approach that accounts for the progressive loss of confinement with increasing eccentricity is recommended. The steel ratio should be increased for eccentrically loaded CFST columns to maintain adequate confinement, even though this increases material cost.

Defect Analysis and Failure Mode Characterization

The failure modes observed in the tests provide important insights for quality control and damage assessment of CFST structures in service. At low eccentricity ratios, the failure was characterized by concrete crushing on the compression face with visible bulging of the steel tube, indicating effective confinement. At medium eccentricity ratios, the failure involved concrete crushing on the compression face combined with local yielding of the steel tube on the tension face, with the steel tube showing significant outward deformation. At high eccentricity ratios, the failure was dominated by steel tube yielding and buckling on the tension face, with concrete crushing occurring only in a limited compression zone. The transition between these failure modes is gradual, and the exact transition point depends on the combination of material parameters and geometric proportions.

Key Questions and Reflections

Several important questions emerge from this study that remain relevant to current practice. First, the study does not address the effect of shear force on the eccentric compression behavior, which is critical for seismic design where combined axial, bending, and shear demands act simultaneously. Second, the interaction between concrete confinement and steel tube local buckling under eccentric loading is not fully explored, particularly for columns with high slenderness ratios. Third, the comparison with design codes highlights the need for code updates, but the specific recommendations for code revision are not provided. The study demonstrates that CFST columns are highly efficient structural members, but their behavior under eccentric loading is more complex than current design codes assume, and engineers should exercise caution when applying axial compression-based design formulas to eccentrically loaded members.

Study Insights and Implications

This paper remains a foundational reference for the understanding of CFST column behavior under eccentric compression. The systematic parametric study and comparison with multiple design codes provide a clear picture of the limitations of current design approaches. For practicing engineers, the key takeaway is that the confinement effect of the steel tube is not constant but varies with the loading condition, and design formulas must account for this variation to ensure safety and economy. Future research should extend the investigation to include cyclic loading, shear combined with eccentric compression, and the effect of concrete curing conditions on the long-term performance of CFST members. The findings also underscore the importance of using appropriate material models in numerical simulations, as simplified linear elastic models cannot capture the complex interaction between steel tube confinement and concrete behavior under eccentric loading.