Local bond behavior of partially encased composite steel and recycled concrete structural members

Bond failure at the steel–concrete interface is a leading cause of performance degradation in partially encased composite (PEC) structural members. Compared with other composite structures, PEC structural members are more susceptible to bond failure owing to the unique characteristics of their construction, highlighting the need to study and improve the interfacial bond behavior. In this study, the bond behavior of partially encased composite steel and recycled concrete (PERC) structural members was explored via push-out testing, and the influences of diverse parameters (the strength of recycled concrete (RC), anchorage length, and the number of layers) on the bond behavior were assessed based on the gray correlation theory. The results suggest that the average ultimate bond strength has a positive relationship with both the number of stud layers and RC strength. Moreover, the number of stud layers had a more significant effect on the ultimate bond strength than on the RC strength. The average ultimate bond strength increased by 665 % when studs were arranged in a double layer. Conversely, the average ultimate bond strength decreased by 14.3 % and 23.4 % as the anchorage length increased from 300 mm to 400 mm and then to 500 mm, respectively. Furthermore, equations for determining the characteristic points essential for constructing a predictive model of the average bond stress–slip behavior at the PERC bond interface were established through statistical regression of the characteristic points from the average bond–slip curve. The bond stress distribution along the anchorage region was further analyzed based on the strain data from the RC and main steel component. A position function was introduced to describe the spatial distribution of the bond stress along the anchorage depth, enabling a more accurate prediction of the bond–slip behavior at different bonding locations of the main steel component. These findings contribute to a better prediction model for the bond performance in PERC structures, providing valuable insights into the finite element analysis of PERC structure members.