This paper deals with the experimental determination of ductility of ultra-high-performance fiber-reinforced concrete (UHPFRC) columns with longitudinal passive reinforcement, tested under combined axial and alternate flexural load. The influence of axial load has been studied, as well as the presence or not of a reduced shear reinforcement. One companion column made of ordinary concrete and with a high shear reinforcement ratio has also been tested for comparison. The ductility has been assessed with energy dissipation and ductility coefficient calculations. In addition, the maximum bending moment has been compared with theoretical calculation, using the material characteristics determined from characterization tests. The paper concludes that UHPFRC columns with a longitudinal reinforcement which have not been specifically designed for seismic resistance do display a ductility coefficient higher than 2 and a noticeable ability to protect the rebars from buckling, but have a rather high decrease of bending moment under the succession of series of alternate flexural load cycles.

Over the last 30 years, structural reinforcement and retrofitting with externally bonded composite materials have proven to be efficient and cost-effective solutions to increase both the safety and the lifespan of civil engineering structures, including nuclear power plants. The effectiveness of the strengthening system highly depends on the level of adhesion between the fiber-reinforced polymer (FRP) composite material and the concrete surface. Therefore, on-site evaluation of the bond quality is critical to assess the performance and predict the durability of the system in place. The direct tension pull-off test is most commonly used to quantify the adhesion level, but this standardized method has many drawbacks. In the present study, it is proposed to evaluate the bond properties by using a nondestructive test (NDT) derived from the standard pull-off test. This innovative test enables the measurement of an interfacial “stiffness” which may be used as a bond quality criterion. This paper gives an insight into the performance of the proposed NDT method, when applied in laboratory conditions to concrete slabs reinforced with bonded pultruded carbon FRP plates (CFRP). Three different epoxy adhesive systems with a broad range of Young’s moduli were used for the specimens’ preparation, in order to vary the stiffness of the concrete/CFRP interface. The purpose was to simulate different levels of interfacial adhesion that could be observed for a single adhesive system. It was shown that the test method was able to detect differences in the interface stiffness beyond experimental uncertainties, and it should therefore enable the detection of differences in the bond quality for a given adhesive system as well. The sensitivity of the NDT was then discussed, and its detection capabilities were predicted for standard field conditions. In the last part, strain measurements were collected during the NDT, thanks to distributed optical fiber sensors (DOFS) embedded in the adhesive joints of the strengthened specimens. An analysis of the strain profiles was found to provide complementary information on the quality of the adhesive bond.