ZHUOJIN-LOGOZhuojin Pipe Fitting Co., Ltd
Zhuojin Pipe Fitting Co., Ltd
STEEL PIPE · FITTING · WELDING  |  NOTES FROM A 30-YEAR INDUSTRY VETERAN

Steel Tube Columns Filled with Coarse-Aggregate Ultra-High Performance Concrete: Axial Compression Behaviour

1. What the Paper Is About

Zeng Yanqin, Xu Lihua, Wu Fanghong, Yu Min and Chi Yin from Wuhan University published a study in Engineering Mechanics, Vol. 39, No. 10, 2022, on the axial compression behaviour of short columns consisting of a circular steel tube filled with coarse-aggregate ultra-high performance concrete (CA-UHPC). The work was supported by a key project of the National Natural Science Foundation of China. The study considered three parameters — steel fibre content, coarse aggregate replacement ratio, and steel tube thickness — and tested 14 short column specimens under full-section axial compression.

The motivation is clear: UHPC has very high compressive strength but is brittle and expensive. Coarse aggregate can improve the ductility and reduce the cost, but it may also affect the bond with the steel tube and the confinement effectiveness. The paper investigates whether the benefits of coarse aggregate outweigh the drawbacks.

2. Core Technical Points and Test Observations

The test results showed two distinct failure modes depending on the confinement coefficient ξ, which is defined as the ratio of the steel tube's confinement force to the concrete's axial capacity:

The transition between the two modes occurs at ξ ≈ 0.883, which is an important design boundary. Below this value, the column is under-confined and fails in a brittle shear mode. Above this value, the column is adequately confined and fails in a ductile barrel mode.

The paper also found that, for the same steel tube thickness, the addition of coarse aggregate increased the bearing capacity. This is because the coarse aggregate improves the concrete's internal friction and interlock, which enhances the confinement effect. The coarse aggregate also reduces the shrinkage of the concrete, which improves the bond with the steel tube.

Confinement coefficient ξ Failure mode Behaviour Design implication
0.217 ≤ ξ < 0.883 Shear failure Brittle, sudden loss of capacity Avoid this range; increase steel ratio or concrete strength
0.883 ≤ ξ ≤ 1.431 Barrel (waist drum) failure Ductile, gradual post-peak Preferred range; ensure adequate confinement
ξ > 1.431 Not tested in this study Likely very ductile Extrapolate with caution

3. A Veteran's Perspective on Material Selection and Fabrication

I have specified concrete-filled steel tubes for columns in high-rise buildings, bridge piers and industrial structures, and the choice of concrete is always a balance between strength, workability, cost and compatibility with the steel tube. The paper's finding that coarse aggregate improves the capacity is consistent with my experience, but there are practical considerations that the paper does not fully address.

First, the maximum aggregate size. In a steel tube, the aggregate size must be small enough to allow the concrete to flow around the tube wall and to fill the space between the tube and any internal stiffeners. For a typical circular tube with a diameter of 300–600 mm, a maximum aggregate size of 10–12 mm is practical. Larger aggregate can cause honeycombing and voids.

Second, the steel fibre content. The paper recommends that the steel fibre volume fraction should not exceed 0.75% for the micro-expansion concrete studied in the related work (Topic 5), but for the CA-UHPC in this study, the fibre content is one of the parameters. Too much fibre can cause balling and poor workability. In my experience, a fibre content of 1.0–1.5% by volume is typical for UHPC, but the mix must be designed carefully to avoid fibre segregation.

Third, the steel tube thickness. The confinement coefficient ξ depends on the steel tube thickness and the concrete strength. For a given concrete strength, increasing the tube thickness increases ξ and improves the ductility. The paper's test range of ξ from 0.217 to 1.431 covers a wide practical range, but the lower end (ξ < 0.883) is a warning zone. I would never design a CFST column with ξ below 0.883 unless there is a very good reason, because the shear failure mode is brittle and unpredictable.

Fourth, the concrete placement. In a short column, the concrete can be placed by pouring from the top, but in a long column, pumping is required. The coarse aggregate can cause blockages in the pump line if the aggregate is too large or the mix is too stiff. I have seen this happen on site, and it is a costly delay. The mix must be designed with a sufficient slump flow and a maximum aggregate size that is compatible with the pump.

4. Study Insights and Implications for Design

The paper proposes a formula for the axial bearing capacity of CA-UHPCFST short columns, based on existing CFST design codes. This is a useful contribution, but the formula is empirical and should be validated against a wider range of test data. The paper's test matrix of 14 specimens is a good start, but it is not enough to cover all practical combinations of steel ratio, concrete strength and fibre content.

From a design perspective, the key insight is that the confinement coefficient ξ is the controlling parameter. The designer should:

  1. Calculate ξ using the actual steel tube dimensions and the concrete strength.
  2. Ensure ξ ≥ 0.883 to avoid the brittle shear failure mode.
  3. Use the coarse aggregate to improve the capacity and reduce the cost, but verify the workability and pumpability.
  4. Specify the steel fibre content based on the required ductility and the concrete mix design.

From a fabrication perspective, the steel tube must be manufactured to the correct dimensions and the weld must be sound. A circular tube is usually made from a plate that is rolled and welded longitudinally. The weld must be full penetration and inspected by UT or RT. The tube ends must be flat and square to ensure a uniform bearing surface.

5. Concluding Summary

This paper is a valuable contribution to the understanding of CFST columns filled with coarse-aggregate UHPC. The test results clearly show that the confinement coefficient ξ governs the failure mode, and that the coarse aggregate improves the bearing capacity. The proposed formula is a useful design tool, but it should be used with caution until it is validated against a broader database. For the practising engineer, the key message is to design for ξ ≥ 0.883 and to specify a concrete mix that is workable, pumpable and compatible with the steel tube. The paper is a good example of how material science and structural engineering can work together to produce a better composite column, and I would recommend it to anyone involved in the design or construction of CFST structures.