# Du, Chen, Liew and Xiong 2017

## Experimental Study, Results, and Discussion

This paper studies the effects of high-strength steel in rectangular CFT beam-columns with parameters of in-fill concrete, eccentricity ratio, and width-to-thickness ratio, and slenderness. Chinese code DB 29-57 states that the concrete bears the axial load, whereas the steel withstands the bending moment and a fraction of the axial load. The concrete increases the critical local buckling stress of the steel, and the steel also provides confinement to the concrete. This paper studies rectangular cross sections as opposed to square or circular because they have a different flexural resistance in each direction, enabling further optimization and convenience for connections. Design approaches were developed based on methods from EC4 and AISC 360 based on the interaction equations of steel columns. DB 29-57 ignores the steel-concrete interaction, and only the steel tubes are verified in formulas for bearing capacity when the tube is subjected to axial load and bending moment. Many codes have limitations for width-to-thickness ratio, and AISC 360 prevents the use of high-strength steel in rectangular CFT columns, however this paper serves to modify the current standards. Three different cross sections were utilized with varying eccentricity, and the thickness was maintained. The axial and flexural loads were increased proportionally to reach the maximum interaction strength. Linear variable displacement transducers were placed along the specimen to monitor the declection, as well as the axial shortening. Strain gages were also utilized to measure longitudinal and lateral strain.

The behavior of the beam-columns can be divided into seven stages: A: Yielding of compression flange, B: Concrete goes into tension, C: Concrete reaches failure strain, D: Yielding of tension flange, E: local buckling of compression flange, F: Local buckling of web, and G: Cracking of steel tube. It was observed that the stiffness was reduced after yielding of the steel compression flange, and the concrete went into tension after the steel tube cracked. For the rectangular CFT specimens, buckling and cracking of the steel occurred at the same location, and the concrete was crushed near mid height, corresponding to the initial imperfection. It was seen that the in-fill concrete had little influence on the stiffness until yielding of the steel. The yielding of compression flange was delayed due to the in-fill concrete, and the load level corresponding to the local buckling of steel increased due to the in-fill concrete. An increase in strength of concrete (C40 to C50) increases the ultimate load by 4.7%. The increase in eccentricity ratio (from .2 to .6) causes a 23% decrease in ultimate load, and a 65% increase in ultimate bending moment. In conclusion, DB 29-57 is conservative for high-strength steel rectangular CFT beam-columns as the ratios of ultimate load to design load are greater than one.

A finite element analysis was also performed using ABAQUS to simulate the experiments. The models simulated a pinned-pinned condition and the nodes in the middle of the column were limited to XOZ displacement. A simplified bilinear model without strain-hardening was used to simulate the Q460 steel, and the stress-strain curves of concrete were used to simulate the in-fill concrete. The load-axial deformation and load-deflection curves match that of the experimental test results proving the accuracy of the finite element analysis.

## References

Du, Y., Chen, Z., Liew, J., Xiong, M.-X. (2017). “Rectangular concrete-filled steel tubular beam-columns using high-strength steel: Experiments and design.” Journal of Construction Steel Research 131, April pp. 1-18 doi:10.1016/j.jcsr.2016.12.016