PSI - Issue 68

V. Milani et al. / Procedia Structural Integrity 68 (2025) 1181–1187 V. Milani, G. Angella, G. Timelli/ Structural Integrity Procedia 00 (2025) 000–000

1185

5

“young” bifilms to generate a second distribution of defects, which is confirmed by the obtained positive threshold (Tiryakioglu & Campbell, 2010), as displayed in Table 3. The threshold values from the Weibull analysis also provide insights into the value of elongation to fracture below which no specimen is expected to fail. Regardless of the processing temperature, the threshold values of the samples treated with flux are higher compared to the non-treated AlSi9Cu3(Fe) alloy, which indicates that the use of flux increases the value of elongation below which fracture will occur. Furthermore, Flux A performs slightly better than Flux B at both processing temperatures. According to Zahedi et al. (Zahedi et al., 2007), who investigated the effect of Fe-rich intermetallics on the Weibull distribution of tensile properties in a cast AlSi5Cu3Fe1Mg0.3 alloy, found that Sr and Mn additions to the alloy affect the threshold stress by modifying the distribution of features that create the highest stress concentration. They concluded that an increase in the threshold parameter should be addressed as an increase in safety of the castings, rather than an increase in reliability as used in the two-parameter Weibull statistics. Thus, the present study highlights how using fluxes can also increase the safety of castings not only by leading to narrower Weibull distributions, but also by moving the distribution to higher values of ductility, as shown in Figure 2. The three-parameter Weibull distribution fits the data well, as indicated by high correlation coefficients ρ in the range between 0.93 to 0.98 (see Table 3). Higher p -values, especially 0.77, indicate excellent fits. Table 3. Estimated Weibull moduli, β , scale parameters, η , and threshold parameter, λ , for the elongation to fracture values obtained from the tensile tests at different processing temperatures and fluxing conditions. The correlation coefficient ρ and p -values are also reported. Processing Temperature (°C) Fluxing Condition β η (%) λ (%) ρ p -value

Flux A Flux B No Flux Flux A Flux B No Flux

1.70 1.52 4.07 1.85 1.73 2.06

0.33 0.64 1.66 0.67 0.95 1.57

1.88 1.69 0.61 2.31 2.05 1.55

0.93 0.93 0.98 0.95 0.96 0.97

0.12 0.12 0.77 0.19 0.35 0.41

720

760

Figures 3a and 3b display the higher porosity levels in the samples treated without flux. The porosity is more significant in the alloy treated at 760 °C (Figure 3b) due to increased oxidation rates and hydrogen absorption in the melt occurring at higher processing temperatures. Using flux reduced the amount of porosity in the specimens, as shown in Figures 3c and 3d, thus leading to improved casting reliability and safety, in accordance with the 2- and 3 parameter Weibull analysis from the present study. Conversely, when fluxes are not used, the higher porosity levels are associated with wider Weibull distributions, as shown in Figure 2.

Fig. 3 Macrographs showing the porosity distribution of (a,b) untreated and (c,d) flux treated with Flux A AlSi9Cu3(Fe) alloy processed at (a,c) 720 and (b,d) 760 °C casting temperature.