Assessment methods for longitudinal shear bond strength in Recycled Aggregate Concrete Composite Slabs (RACCS)

Concrete holds a central role in contemporary construction, yet its production and lifecycle raise major environmental concerns due to extensive natural aggregate consumption, high energy demands, and substantial demolition waste generation. In particular, the growing demand for sand and gravel has led to resource depletion, riverbed erosion, and increased greenhouse gas emissions. In response, the reuse of concrete waste in the form of Recycled Concrete Aggregate (RCA) represents a viable pathway toward sustainability, aligning with circular economy principles by reducing landfill volumes and conserving non-renewable resources. However, the adoption of RCA in structural applications raises technical challenges due to its heterogeneous composition, increased porosity, and altered interfacial bond performance. This study investigates the impact of RCA incorporation on the longitudinal shear behaviour of Recycled Aggregate Concrete Composite Slabs (RACCS), supporting the evolution of design methodologies to better represent the behaviour of recycled aggregate systems. Existing formulations, such as the m–k method in Eurocode 4, were calibrated on traditional materials and require adaptation for sustainable alternatives. A comprehensive dataset from 48 full-scale four-point bending tests was analysed, covering six RCA replacement ratios rRCA (0–100%) and three span lengths. An Exploratory Data Analysis (EDA) approach was implemented to ensure dataset robustness, followed by regression analysis to extract m and k parameters representing mechanical interlock and frictional resistance, respectively. Results revealed a non-monotonic behaviour: interlock efficiency increased up to 30% RCA but declined significantly beyond this threshold due to ITZ degradation and premature microcracking. Frictional contribution decreased initially but rose again at high RCA levels due to increased surface roughness from residual mortar. Correction coefficients α(rRCA) and β(rRCA) were introduced to scale m and k relative to NAC reference values, interpolated through a combined parabolic–exponential function. The resulting modified design equation achieved prediction errors generally below ±10%, enabling accurate, code-compatible, and sustainability-driven assessment of RACCS performance.