Influence of the roughness of floor tiles on the cleanability from wheat flour residues in agri-food buildings

The floors in agri-food buildings are the surfaces most at risk of microbial contamination because, during processing phases, small amounts of foodmay be accidentally spilled and become the principal source of nourishment for the metabolic activities of microorganisms and pests. The buildingindustry offers a series of solutions for floorings in agri-food buildings, characterized by various types of materials and surface finishing. However,they are often proposed for different production sectors without any distinction. Actually, each food product has peculiar physico-chemicalcharacteristics and contaminates flooring surfaces in a different manner. The objective of this paper was to establish the correlations between thegeometry of unglazed dry-pressed ceramic tiles and their dry-cleanability from wheat flour residues. Four different types of 20 cm x 20 cm unglazeddry-pressed ceramic tiles, commonly used in agri-food buildings, were analysed. Thirty surface roughness measurements were performed for eachsample tile to measure roughness parameters in three different directions (horizontal, vertical, diagonal). Cleanability was measured by utilizing aprocedure proposed and described in a previous paper published by the same authors. An other essential characteristic of floors in agri-food buildingsis their slip resistance. Ten measurements were taken on each tile, which enabled to calculate the mean value of the dynamic friction. The relationbetween the roughness parameters of the tile surface and the flour residues was studied through a correlation analysis. A further linear correlation wasobtained between the values of friction coefficients and roughness parameters. In both relations, the roughness parameters with the highest values ofcorrelation were Rz Rt and Mr2. This study, showed that the cleanability of unglazed dry-pressed ceramic tiles from flour residues did not dependdirectly on the height of the profile peaks but rather on the linear combination of the above-mentioned parameters. Therefore, it is necessary toconsider the complete geometry of the profile in relation to the size of the residues. Similarly, the study also showed that the coefficient of dynamicfriction on the surface did not depend directly on the average height of the profile, yet it depended on the linear combination of roughness parametersRz Rt and Mr2.