The cracking of the pre-/post-casting UHPC joint in the steel-UHPC composite bridge deck system can lead to continuous tensile damage in the UHPC layer, reducing its ability to stiffen the steel deck. This study aims to clarify the cracking mechanism of the segmented-casting UHPC joint, and to provide design recommendations for cracking control. Axial tension tests on full-scale composite deck specimens were conducted, in which the influence of with and without joint, and varying reinforcement ratios on cracking response were identified. Moreover, the nonlinear numerical model for the composite deck, where the UHPC-UHPC interface was simulated in three methods, i.e. unbonded case, cohesive zone model (CZM), and perfectly bonded case, was developed and validated to simulate the crack initiation and propagation of the UHPC layer. Based on the validated numerical model incorporating CZM, improving the bond strength of CZM is more effective in controlling crack opening at the interface than increasing the failure displacement. Finally, optimized reinforcement arrangement and joint shape recommendations were provided to enhance construction convenience, minimize stress concentration, and limit crack opening.

The short-headed stud connectors play a critical role on the interaction of the orthotropic steel deck (OSD) and the ultra-high performance concrete (UHPC) layer in orthotropic steel-UHPC composite bridge deck. In this paper, the fatigue behavior of these short-headed stud connectors was experimentally investigated in a beam test. The failure modes of the short-headed stud connectors were identified and classified into 5 types. The fatigue test results were analyzed by linear regression analysis neglecting run-outs and treating run-outs as failure respectively. On the other hand, the maximum likelihood estimation (MLE) approach was used to shape the S-N curve by considering the influence of run-outs. Additionally, the push-out and beam fatigue test data were compared, and the push-out test presented a relatively conservative result. Last, the applicability of existing specifications on design guidelines regarding the short-headed stud connectors design in orthotropic steel–UHPC composite bridge deck is discussed, and a design S-N curve with 95% survival probability is proposed.

In order to investigate load-carrying capacity, ductility and plastic development of orthotropic steel-concrete composite deck used in composite girder cable-stayed bridge, six single-point-loaded simply-supported orthotropic steel-concrete composite decks with varying axial compressive forces and different concrete grades were designed and tested. The test results show that the failure modes of all specimens are similar and exhibit cracks of the lower concrete section combined with crushing of the upper concrete section at midspan. The axial compressive force imposed in this test has a neglected effect on elastic stiffness and vertical load-carrying capacity of tested specimens. While the axial compressive force considerably lowers the ductility and plastic development of the orthotropic steel-concrete composite deck under vertical load. Besides, improving concrete grade from C60 to C80 could not help to enhance elastic stiffness, vertical load-carrying capacity, ductility and plastic development of the composite deck.