Development and statistical database analysis of hardened concrete properties made with recycled concrete aggregates

Abstract

The use of recycled concrete aggregates (RCA) as a replacement in new concrete has gained popularity worldwide as a method of reducing natural aggregate consumption. Many engineers, owners, and agencies are reticent to use RCA due to a lack of understanding on how the material will impact overall concrete properties. This lack of acceptance has been hindered by the difficulty of doing large-scale studies because of the time and cost of casting different concrete mixtures. Many individual studies have been completed but little work has been done to analyze that work to make broad conclusions on RCA concrete mechanical properties. This paper presents the development and analysis of a database of mechanical properties of concrete containing coarse RCA generated from 115 peer-reviewed journal articles. A sensitivity study of RCA systems was conducted through a full-factorial analysis to explore how the mixture design proportions influence the RCA concrete hardened properties. On average, the compressive strengths of concrete systems with 20% or less RCA replacement were 2% lower than that of the natural aggregate systems. Increased levels of RCA replacements decreased the average concrete compressive strengths with the highest average strength loss, in which a 10% reduction compared to natural aggregate systems is observed at 100% RCA replacement level. Results indicate that, in general, adequate concrete strengths for a variety of applications can be achieved at all replacement ratios of coarse RCA. Impacts of mixture design properties on particular mechanical properties are presented along with bivariate contour plots that can be used by concrete mixture designers to better estimate RCA concrete properties. This work supports an increased adoption of RCA by providing broad conclusions supported by statistical analysis that show typical performance of RCA concrete across a variety of design inputs.