An accurate quantification of the bond between embedded rebar and recycled aggregate concrete (RAC) is a prerequisite for confident use of RAC. However, there is a noticeable lack of experimental data that can faithfully represent the actual bond stress state in real RAC structures. To address this limitation, a total of 37 beam specimens were tested in strict conformity with the RILEM testing protocol. The strength grade of old concrete was among several control variables studied for their impact on the bond strength. The findings reveal that the incorporation of coarse or/and fine recycled aggregate (RA) does not alter the expected mode of bond failure for a beam; however, it significantly affects the beam’s bond strength in flexure. Using RA sourced from weaker parent concrete leads to much inferior interfacial bonding, but there exists an almost linear relationship between the bond strength of RAC relative to that of natural aggregate concrete (NAC) and their relative water absorption ratio. This relationship seems to hold true regardless of the strength of RA. Based on the probabilistic analyses conducted herein, the anchorage length specified in existing codes developed for NAC is found inadequate and unsafe for RAC. A modifier is therefore proposed to remedy this unconservatism.
Read full abstract