The foundry industry generates large amounts of spent foundry sands, which are stored, available for recovery in other industrial sectors but unfortunately poorly exploited. Different authors have studied the possibility of recovering them in concretes, which would also allow production of more sustainable cementitious materials. The variability of their results highlights the importance of a better understanding of the potential influential parameters of the by-products. Unfortunately, exhaustive characterizations of the materials are rarely performed, especially for chemically bound foundry sands. This article presents a case study for the recovery of a spent chemical foundry sand with an exhaustive physicochemical characterization of the by-product and an analysis of its influence on the workability and mechanical strengths of cementitious materials. The tests recommended by the European standard for aggregates for concrete confirmed the suitability of the by-product. Associated with additional chemical tests (scanning electron microscopy, X-ray fluorescence, X-ray diffraction, etc.) as well as metallic particles characterization, they highlighted possible influential parameters. The workability and mechanical resistance tests carried out on mortars and concretes confirmed the influence of the fineness of the by-product associated with other parameters. Its use at a substitution rate of 30% results in a strength class C 30/37 concrete.

In this paper, geometrical roughness numbers named RNx, RNy and RNS are formulated as a possible analytical tool for cement-based material surface characterization using coherence scanning interferometry (CSI) and STIL confocal microscopy technique (SCM). Recently, cement paste surface maps have been used to establish a link between surface statistical roughness parameters and the measuring scale (Apedo et al., 2015). The objectives of the present paper are to study how geometrical roughness numbers can be used as a tool to analyze the colonization of cement-based material surfaces by microorganisms as well as to perform other subsequent studies. Observations from a series of images acquired using both techniques (CSI and SCM) on both polished and unpolished samples are described. Using a new method named “window resizing”, the results from CSI are compared with those from SCM. It appears that the surface available for colonization (convolution) is smaller than the surface developed by the measuring tool (sampling). The new method also allows the identification of the fractal regions and the associated fractal dimensions of both polished and unpolished cement pastes.