PSI - Issue 68

Carlos Fernandes da Silva et al. / Procedia Structural Integrity 68 (2025) 1252–1258 C. F. das Silva et al. / Structural Integrity Procedia 00 (2024) 000–000

1256

5

The specific work of fracture ( w f ) was obtained from the total work of fracture ( W f ) by numerically integrating the load ( F ) - displacement ( ∆ ℓ ) curve, normalized by the original ligament area: W f = F d ℓ ⇒ w f = W f b . t = σ d ℓ (1) Table 2. Summary of the properties of the di ff erent films groups investigated in the present work, uncertainty is expressed in the form of the standard deviation of the measurements, and the last columns present the number of samples tested for that given group. Sample group σ y [MPa] σ d [MPa] w f [10 3 Jm − 2 ] # samples PET 44.3 ± 4.7 31.8 ± 6.6 95.8 ± 23.3 24 A1 43.5 ± 5.7 31.4 ± 3.3 73.6 ± 35.9 28 A2 42.9 ± 3.9 32.4 ± 2.3 96.9 ± 15.6 27 A3 41.9 ± 5.3 31.6 ± 7.7 81.0 ± 14.1 21 A4 41.2 ± 3.2 28.2 ± 4.1 26.4 ± 26.2 15 B1 46.9 ± 7.7 32.9 ± 4.8 53.8 ± 41.1 19 B2 46.8 ± 3.9 30.7 ± 6.1 76.9 ± 36.6 14 B3 48.2 ± 4.7 36.3 ± 10.3 60.0 ± 34.2 24 B4 47.7 ± 4.6 36.1 ± 5.2 47.6 ± 28.8 14 Figure 3(a) shows the dependence of the yield stress on the PA content for the two PA resins used in the present work. The uncertainty, represented in error bars, correspond to the 95% confidence limit for the average value, assum ing the validity of the central limit theorem Box (1978); Walpole et al. (1998). This allows a graphic visualization of the data dispersion.

50

50

40

40

30

30

σ d [MPa]

σ y [MPa]

20

20

10

10

PET PA 66 PA 6T6I

PET PA 66 PA 6T6I

0

0

0

5

10

15

20

0

5

10

15

20

% PA (b) Fig. 3. Dependence of the yield stress ( σ y ) (a) and drawing stress ( σ d ) (b) on the concentration of PA. Error bars represent the 95% confidence limit for the average, lines are just guides to the eye. The results show a slight tendency for the variation of the yield stress, with the PA 6T6I resin leading to an increase of the yield stress and the PA 66 leading to a maintenance of the yield stress (with a small tendency of softening, though not statistically significant). This result can be rationalized by the e ff ect of the cyclohexane ring in the PA 6T6I resin, which likely interacts better with the terephtalate group in PET. The changes, however are very subtle, and from a technological perspective these results show that the addition of PA to PET has little e ff ect on the strength. The drawing stress (Fig. 3(b)) follows a similar trend, there is a slight tendency for increase in the case of the PA 6T6I resin (statistically significant only for the 20% PA sample, when compared with the corresponding PA 66 sample), and a slight tendency for decrease in the case of the PA 66 resin although the values are to be considered equal to the one observed for the pure PET resins, in the statistical analysis. Both yield stress and drawing stress are defined by the ability of the chain segments to slip against each other in the macromolecular mixture, and their numerical values depend on how the di ff erent molecules interact Meijer (a) % PA