PSI - Issue 78

Riccardo Liberotti et al. / Procedia Structural Integrity 78 (2026) 1919–1926

1922

• C. Lime+pozzolana. Proportions: 1.5 parts black pozzolana+1.5 parts red pozzolana+1 part lime (volume ratio 3:1). Mineralogical type: inorganic natural hydraulic lime mortar (NHL) aged 60 months, handmade according to traditional methods and fired in limestone kilns fuelled with bundles; black pozzolana; red pozzolana. Grain size: black and red pozzolana 0–6 mm. • D. Lime+earthenware powder. Proportions: 3 parts crock powder to 1 part lime (volume ratio 3:1). Mineralogical type: artisanal hydraulic slaked lime with crushed ceramic shards from recycled elements (e.g. tiles). Grain size: earthenware powder 0–10 mm. • E. Lime+pozzolanic cement+pozzolana. Proportions: 1.5 parts black pozzolana+1.5 parts red pozzolana+0.5 part lime+0.5 part pozzolanic cement. Mineralogical type: artisanal hydraulic slaked lime; black pozzolana; red pozzolana; pozzolanic cement. Grain size: black and red pozzolana 0–6 mm. • F. Lime+pozzolanic cement+earthenware powder. Proportions: 3 parts earthenware pow der+0.5 part lime+0.5 part pozzolanic cement. Mineralogical type: artisanal hydraulic slaked lime; crushed ceramic shards from recycled ancient elements (e.g. tiles); pozzolanic cement. Grain size: black and red pozzolana 0–6 mm. • G. Lime+sand. Proportions: 3 parts sand to 1 part lime (volume ratio 3:1). Mineralogical type: artisanal hydraulic slaked lime; river/siliceous sand. Grain size: sand 0–3 mm. • H. Lime+sand+pozzolanic cement. Proportions: 3 parts sand+0.5 part lime+0.5 part pozzolanic cement. Mineralogical type: artisanal hydraulic slaked lime; river/siliceous sand; pozzolanic cement. Grain size: sand 0–3 mm. • I. Alabastrine gypsum. Water/gypsum ratio: 100/140. Mineralogical type: calcium sulphate hemi hydrate powder, obtained by grinding gypsum stones with specific characteristics, calcined in methane furnaces at constant temperatures, and refined to fine grains. • L. Alabastrine gypsum+carded sisal fibre. Water/gypsum ratio: 100/140. Mineralogical type: calcium sulphate hemihydrate powder, obtained by grinding gypsum stones with specific character istics, calcined in methane furnaces at constant temperatures, and refined to fine grains. Natural additive: sisal fibre composed of cellulose (65.80%), hemicellulose (12%), pectin (0.80%), water (10%), lignins (9.90%), water-soluble substances (1.20%), and fats (0.30%). • M. Red pozzolana (unsifted)+pozzolanic cement (not tested yet). Proportions: 3 parts red pozzolana+1.5 parts pozzolanic cement. Mineralogical type: stabilised raw red pozzolanic soil, workable upon water addition and air-dried. • N. Red pozzolana (sifted)+pozzolanic cement (not tested yet). Proportions: 3 parts red pozzolana+1.5 parts pozzolanic cement. Mineralogical type: stabilised raw red pozzolanic soil, workable upon water addition and air-dried. 3. Experimental results The outcomes of the three-point bending tests conducted on the artificial stone specimens are pre sented in Fig. 2. Each subplot compares two nominally identical samples of a given mix design, with the force–displacement curves represented by dashed and dotted lines, respectively. At this stage, specimens of types M and N have not yet been tested; therefore, the graphs refer exclusively to specimens ranging from type A to type L. For clearer visualisation, due to the significant differences observed between the various mix designs, the scales of the x-axis (displacement) and y-axis (force) differ across the graphs. It is therefore essential, when analysing the results, to refer to the specific scale of each plot rather than focusing just on the shape of the curves. In terms of peak resistance, best samples are B, E, F probably due to the presence of pozzolana or pozzolanic cement and I and L which are made of compact alabastrine gypsum. Intermediate values are reached by samples A and H, the first one is weaker than B as it has sand instead of black poz zolana while the second one is weaker than F as it has sand instead of earthenware powder. Finally, samples C, D and G are the weakest ones, indicating that lime based components without pozzolanic cement do not exhibit satisfactory mechanical performance. For example, sample G which is basically H without pozzolanic cement has a huge drop in resistance with respect to H: 0.05 kN vs 0.7 kN. Moreover, the two tested G specimens provided a very different outcome which shows the unreliability of this mix-design. With regard