Sunlight Temperature Field in Steel Tube-Expanded Concrete Pier Columns: A Practical Review
1. Core Findings and Test Setup
The paper by Ren Zhigang et al. addresses a problem I have seen underestimated on more than one bridge project: the temperature gradient that develops in a concrete-filled steel tube (CFST) pier column under direct sunlight. The authors used heat transfer theory and finite element analysis to compute the cross-section temperature distribution in a super-high pier column, considering three heat transfer mechanisms: solar radiation, convection, and thermal radiation. The key result is that under clear, cloudless weather, a significant nonlinear temperature difference develops across the section, with the maximum positive temperature difference exceeding 17°C. That is not a small number. In my experience, a 17°C differential across a 1.5 m diameter column can produce bending moments and local stresses that are easily overlooked in design, especially when the column is part of a tall, slender structure where second-order effects matter.
The test setup itself is straightforward: a finite element model of the pier cross-section, with the steel tube and the concrete core discretized, and the boundary conditions representing solar radiation, ambient air convection, and radiative exchange with the sky. The authors then performed a parameter sensitivity analysis, varying meteorological parameters and the thermal properties of the core concrete. The conclusion that the surface solar absorptivity has the largest influence on the section temperature difference is exactly what I would expect from field observations. A white-painted tube will behave very differently from a black or rusted one, and this is a practical detail that often gets ignored in the design office.
2. Heat Transfer Mechanisms and Key Parameters
To appreciate the magnitude of the problem, we need to look at the thermal properties involved. The table below summarises typical values for steel and concrete, which drive the differential heating.
| Material | Thermal Conductivity (W/m·K) | Specific Heat (J/kg·K) | Density (kg/m³) | Thermal Diffusivity (m²/s) |
|---|---|---|---|---|
| Structural Steel | 45 – 55 | 480 – 500 | 7850 | ~1.2 × 10⁻⁵ |
| Normal Concrete | 1.5 – 2.5 | 900 – 1000 | 2300 – 2500 | ~0.8 × 10⁻⁶ |
| Expanded Concrete | 1.0 – 1.8 | 850 – 950 | 2000 – 2200 | ~0.7 × 10⁻⁶ |
The steel tube has a thermal diffusivity roughly an order of magnitude higher than the concrete. This means that when the sun hits the tube, the steel heats up almost immediately, while the concrete core lags behind. The result is a steep temperature gradient at the steel-concrete interface, and a nonlinear distribution across the whole section. The paper's finding of a 17°C maximum difference is consistent with this physical picture. For a 1.2 m diameter column, a 17°C differential can induce a curvature of the order of 0.0001 rad/m, which for a 50 m high pier translates to a lateral displacement of several millimetres at the top. That may not sound like much, but when combined with wind and traffic loads, it can push the design envelope.
The sensitivity analysis is particularly useful. The authors identified solar absorptivity as the most influential parameter, followed by the concrete's thermal conductivity and specific heat. In practice, this means that the surface finish of the steel tube is a design decision, not just an aesthetic one. A light-coloured, reflective coating can reduce the temperature differential by as much as 30–40% compared to a dark, absorptive surface. I have seen projects where the contractor painted the tubes with a dark epoxy for corrosion protection, and the resulting temperature stresses caused hairline cracks in the concrete core. The lesson is clear: the paint specification must be coordinated with the structural design.
3. Engineering Practice Implications
For a CFST pier in a bridge or a tall building, the temperature gradient is not just a serviceability issue; it can affect the ultimate limit state as well. The differential expansion of the steel tube and the concrete core creates interfacial shear stresses, which can lead to debonding or local cracking. In extreme cases, if the temperature differential is large enough, the concrete core may go into tension while the steel tube is in compression, reducing the composite action. This is particularly critical for expanded concrete, which has a lower modulus of elasticity and a higher coefficient of thermal expansion than normal concrete. The paper's focus on "expanded concrete" is important because that material is often used in CFST columns to compensate for shrinkage and to improve the bond with the steel tube. However, the expansion agent also changes the thermal properties, making the temperature problem worse.
In my experience, the practical countermeasures are threefold. First, the design should include a temperature gradient load case, based on the local climate and the surface absorptivity. Second, the steel tube should be insulated or painted with a reflective coating, especially on the side that faces the sun. Third, the concrete mix should be designed with a low coefficient of thermal expansion and a moderate thermal conductivity, to reduce the gradient. Some designers have even used a thin layer of insulating material between the steel tube and the concrete, but that complicates the construction and reduces the composite action, so it should be considered carefully.
4. Reflections and Lessons Learned
The paper is a good example of how a seemingly simple problem—a column in the sun—can have significant structural consequences if not properly analysed. The finite element approach is standard, but the parameter sensitivity study is what makes it valuable for practice. I would have liked to see a comparison with field measurements, but the authors' conclusions are physically sound. The main lesson for engineers is to never assume that a CFST column is uniformly heated. The sun is not uniform, and the steel tube's high conductivity ensures that the temperature distribution is anything but uniform. I have been on sites where the temperature difference between the sunlit and shaded sides of a large steel tube was measured at over 20°C, and the resulting bowing was visible to the naked eye. If you are designing a CFST structure, treat the temperature gradient as a first-order load case, not a secondary consideration. The cost of a reflective coating or a well-designed concrete mix is negligible compared to the cost of repairing cracked concrete or replacing a debonded interface.
Zhuojin Pipe Fitting Co., Ltd