Axial Compression Behaviour of CFRP-Steel Tube RPC Short Columns
Overview of the Literature and Its Engineering Context
This paper by Cui Lishi, Jiao Chujie, Li Song and their colleagues from Guangzhou University presents an experimental study on 12 short columns consisting of a reactive powder concrete (RPC) core, a steel tube with wall thicknesses of 2 mm or 4 mm, and two layers of carbon fibre reinforced polymer (CFRP) as external confinement. The RPC core had a compressive strength of 100 to 160 MPa. The columns were tested under axial compression, and the authors proposed a simplified formula for the ultimate bearing capacity based on a limit equilibrium analysis. For a veteran of the steel pipe industry, this paper is particularly interesting because it addresses the interaction between the steel tube and the high-performance concrete core, and it highlights the role of the steel tube as a confining element rather than just a structural member.
The paper's main contribution is the introduction of an influence factor IF to quantify the enhancement of the ultimate bearing capacity due to the combined confinement of the CFRP and the steel tube. The authors also developed a linear regression expression for IF and a simplified formula for the ultimate load. The experimental results showed that the RPC core exhibited more brittle behaviour than ordinary concrete, with a larger strain at peak load, and that the failure modes were either shear failure or end compression bulging.
Technical Points and Engineering Interpretation
Material Behaviour and Confinement Mechanism
The RPC core in this study had a compressive strength of 100 to 160 MPa, which is significantly higher than that of ordinary concrete. The paper notes that RPC is more brittle in an unconfined state, but when confined by the steel tube and CFRP, it exhibits a more ductile response. The confining pressure provided by the steel tube is limited by the yield strength of the steel, while the CFRP provides additional confinement that is elastic until failure. The combined effect is a higher ultimate capacity and a more gradual post-peak softening.
From a steel pipe perspective, the key parameter is the diameter-to-thickness ratio of the steel tube. For a 2 mm wall thickness, the D/t ratio is quite high, which means the steel tube is prone to local buckling before it reaches its yield strength. The paper observed that the columns with 2 mm tubes failed by local buckling of the steel tube, while those with 4 mm tubes failed by concrete crushing or by a combination of concrete crushing and steel yielding. This is consistent with my experience in the field: a thin-walled steel tube cannot effectively confine a high-strength concrete core because it buckles locally, and the confinement is lost.
The CFRP wrapping provides an external confinement that is independent of the steel tube. The two layers of CFRP in this study were applied with the fibres oriented in the circumferential direction, which is the most effective orientation for axial compression. The CFRP has a high tensile strength but a low modulus of elasticity, so it provides confinement that increases with the lateral expansion of the core. The paper's results show that the CFRP significantly increased the ultimate capacity, especially for the columns with thin steel tubes, because the CFRP prevented the local buckling of the steel tube.
Failure Modes and Load-Displacement Curves
The paper describes three failure modes: shear failure, end compression bulging, and a combination of both. The shear failure occurs when the concrete core fails along a diagonal plane, and the steel tube and CFRP are unable to prevent it. The end compression bulging occurs when the concrete at the loaded end expands laterally, causing the steel tube to bulge outward. This is a common failure mode in CFST columns with high-strength concrete, and it is often caused by the high stress concentration at the loading plate.
The load-displacement curves were divided into four stages: linear elastic, elasto-plastic, failure, and plateau. In the linear elastic stage, the steel tube and the concrete core deform elastically, and the load is shared between them. In the elasto-plastic stage, the concrete core begins to micro-crack, and the lateral expansion increases, which activates the confinement from the steel tube and CFRP. The failure stage is characterised by a rapid increase in displacement with a small increase in load, and the plateau stage represents the post-peak residual capacity.
For the pipe industry, the elasto-plastic stage is the most important because it determines the ductility of the column. The paper's results show that the columns with CFRP wrapping had a more gradual post-peak softening, which is beneficial for seismic applications. This is because the CFRP provides a continuous confinement that prevents the sudden loss of capacity that occurs when the steel tube buckles.
Simplified Formula and Its Practical Limitations
The authors proposed a simplified formula for the ultimate bearing capacity based on the assumption that the steel tube reaches its yield strength at the ultimate load. They used a limit equilibrium analysis, which is a common approach for CFST columns. The formula includes an influence factor IF that accounts for the enhancement due to the CFRP and steel tube confinement. The IF was obtained by linear regression of the experimental data.
The simplified formula is useful for preliminary design, but it has limitations. First, it does not account for the local buckling of the steel tube, which can reduce the confinement effectiveness. Second, it assumes that the steel tube reaches its yield strength, which may not be true if the tube is thin-walled and buckles locally. Third, the formula was derived from a limited number of specimens, and the regression may not be accurate for all geometries and material combinations.
In my experience, a more robust approach is to use a fibre model or a finite element model that accounts for the actual stress-strain behaviour of the steel, the concrete, and the CFRP. The paper's formula is a good starting point, but it should be validated against a larger database of test results before it is used in design.
Integration with Manufacturing and Delivery Practice
Steel Tube Quality and Dimensional Tolerances
The steel tubes used in this study had wall thicknesses of 2 mm and 4 mm, which are relatively thin for structural applications. In practice, such thin tubes are typically used for small diameter columns or for temporary structures. The manufacturing of thin-walled tubes requires strict control of the wall thickness tolerance, because a variation of 0.5 mm can change the D/t ratio by 25% and significantly affect the local buckling capacity.
For the pipe industry, the key quality parameters are the wall thickness, the diameter, and the surface condition. The wall thickness should be measured at four points around the circumference, and the average should be within the specified tolerance. The diameter should be within the tolerance specified in the standard, and the ovality should be less than 1% of the diameter. The surface should be free from scratches, dents, and corrosion, because these can initiate local buckling.
CFRP Application and Bonding
The application of CFRP to the steel tube requires careful surface preparation. The steel surface must be cleaned, degreased, and roughened to achieve a good bond. The CFRP is typically applied using a wet lay-up process, where the fibres are saturated with an epoxy resin and wrapped around the tube. The resin must be allowed to cure for the specified time before the column is loaded.
In my experience, the most common problem with CFRP wrapping is the formation of air bubbles, which reduce the bond strength and can lead to premature debonding. The application should be done in a controlled environment with a temperature above 10°C and a relative humidity below 75%. The CFRP should be applied with sufficient tension to ensure that it is in intimate contact with the steel surface, and the overlapping joint should be at least 150 mm long.
Welding and Fabrication of the Steel Tube
If the steel tube is welded, the weld quality is critical. For a 2 mm wall thickness, the weld is typically a butt weld, and it should be full penetration. The weld should be visually inspected and, if required, tested with dye penetrant or magnetic particle inspection. The weld should be ground flush to avoid stress concentrations that could initiate local buckling.
For a 4 mm wall thickness, the weld can be a fillet weld or a butt weld, depending on the application. The weld should be sized to develop the full strength of the tube. In the paper's study, the tubes were likely seamless or welded, but the paper does not specify. In practice, I would recommend seamless tubes for small diameter columns, because they have no weld and therefore no weld-induced residual stress or imperfections.
Key Questions and Reflections
One question that the paper does not address is the effect of the loading rate on the behaviour of the columns. The tests were conducted at a quasi-static loading rate, but in a seismic event, the loading rate is much higher, and the concrete core may exhibit a higher strength but a lower ductility. The CFRP may also behave differently under high strain rates, with a higher tensile strength but a lower ultimate strain.
Another question is the effect of the length-to-diameter ratio on the behaviour. The paper tested short columns with a length-to-diameter ratio of about 3, which is typical for a stub column. However, in practice, columns can have a much higher slenderness ratio, and the failure mode changes from material failure to global buckling. The simplified formula does not account for the slenderness effect, so it should only be used for short columns.
From a practical perspective, the most important question is whether the CFRP wrapping is cost-effective. The CFRP material and the application labour are expensive, and the increase in capacity may not justify the cost. In many cases, it is more economical to increase the wall thickness of the steel tube or to use a higher strength steel. The paper does not provide a cost analysis, but I would recommend that engineers consider the cost-benefit before specifying CFRP wrapping.
Study Insights and Implications for the Industry
This paper provides valuable experimental data on the behaviour of CFRP-steel tube RPC columns, and it demonstrates the effectiveness of combined confinement. The proposed formula is a useful tool for preliminary design, but it should be used with caution. For the steel pipe industry, the paper highlights the importance of the steel tube as a confining element, and it shows that the wall thickness and the D/t ratio are the key parameters that determine the confinement effectiveness.
In my practice, I would recommend that engineers specify a minimum wall thickness for the steel tube to ensure that it can provide effective confinement without local buckling. The minimum wall thickness should be based on the diameter and the yield strength of the steel, and it should be checked against the local buckling limits in the design code. The CFRP wrapping should be considered as a supplementary confinement, not as a substitute for an adequate steel tube.
Conclusion
The paper by Cui et al. is a well-executed experimental study that contributes to our understanding of the behaviour of CFRP-steel tube RPC short columns under axial compression. The proposed influence factor and simplified formula are practical, but they have limitations that should be recognised. For the pipe industry, the key takeaway is that the steel tube must be designed to provide effective confinement, and the wall thickness should be sufficient to prevent local buckling. The CFRP wrapping can enhance the capacity and ductility, but it should be used judiciously, considering the cost and the application quality. I would recommend this paper to engineers working on high-strength concrete columns, and I would advise them to supplement the simplified formula with a more detailed analysis for critical applications.
Zhuojin Pipe Fitting Co., Ltd