ABSTRACT
Structurally deficient bridges are commonly retrofitted using conventional methodologies, including reinforced concrete, steel jackets, and fiber-reinforced polymers. Although these retrofit methods aim to improve structural performance, exposure to aggressive environments may undermine the durability performance of the retrofit material. More recently, ultra-high-performance concrete (UHPC) has provided an alternative to conventional construction methods, with its superior material characteristics favoring its use in retrofit applications. In this study, a large-scale reinforced concrete (RC) T-beam is constructed and artificially damaged. The T-beam is then retrofitted with an external envelope of UHPC on all faces. Sandblasting is introduced to the surface, providing partially exposed reinforcement in the T-beam to simulate material deterioration. Additional reinforcement is placed in the web and flange, followed by casting the enveloping layer of UHPC around the specimen. The feasibility of this method is discussed, and the structural performance of the beam is assessed by subjecting the beam to cyclic and ultimate flexural loading. This paper presents the results of cyclic and ultimate testing on the RC-UHPC composite T-beam regarding load-displacement, failure mode, and strain responses. The retrofitted T-beam specimen is subjected to a cyclic loading range of 131 kN for 1.576 million cycles. Despite no visible cracks in the cyclic testing, the specimen experiences a 12.22% degradation in stiffness. During the ultimate flexural testing, the specimen shows no relative slip between the two concretes, and the typical flexural failure mode is observed. By increasing the longitudinal reinforcement ratio in the web, the failure mode can shift from localized cracking, predominantly observed in the UHPC shell, toward a more distributed cracking pattern along the length of the beam, which is similar to conventional reinforced concrete beams.
ABSTRACT
Ultra-High Performance Concrete (UHPC) has been a material of interest for retrofitting reinforced concrete elements because of its pioneer mechanical and material properties. Numerous experimental studies for retrofitting concrete structures have shown an improvement in durability performance and structural behaviour. However, conservative and sometimes erroneous estimates for bond strength are used for numerically calculating the strength of the composite members. In addition, different roughening methods have been used to improve the bond mechanism; however, there is a lack of numerical simulation for the force transfer mechanism between the concrete substrate and UHPC as a repair material. This paper presents an experimental and numerical programme designed to characterize the interfacial properties of concrete substrate and its effect on the bond strength between the two materials. The experimental programme evaluates the bond strength between the concrete substrates and UHPC with two different surface preparations while using bi-surface test and additional material tests, including cylinder and cube tests for compression property, direct tension test, and flexural test to complement UHPC tensile properties. Non-linear finite element analysis was conducted, which uses a numerical zero thickness volume model to define the interface bond instead of a traditional fixed contact model. The numerical results from the zero thickness volume model show good agreement with the experimental results with a reduction in error by 181% and 24% for smooth and rough interface surfaces if compared to the results from the model with a fixed contact.
