Issue 78

A. Sancak et alii, Fracture and Structural Integrity, 78 (2026) 186-206; DOI: 10.3221/IGF-ESIS.78.11

The uncorroded reference specimens were tested at 23, 0, − 20 and − 40 °C to establish the temperature-dependent baseline impact behaviour of the chain steel. These reference tests were used to evaluate the change in absorbed energy with decreasing temperature and to provide a comparison for the corrosion-exposed specimens. For the corrosion-exposed specimens, the main comparison was performed at 0 °C after 10 and 30 days of exposure in seawater and substitute seawater. Since the specimens exposed to seawater showed a stronger reduction tendency in absorbed impact energy than those exposed to substitute seawater, additional testing was focused on seawater exposure. Therefore, specimens exposed to seawater for 30 days were also tested at 23 °C. In addition, pre-notched specimens exposed to seawater for 30 days were tested at both 0 and 23 °C to evaluate the effect of direct corrosion exposure at the notch root.

Figure 4: a) Corroded Charpy V-notch specimens after removal from seawater and before impact testing, b) V-notch machined on uncorroded Charpy specimens, c) pre-notched specimen removed from seawater before impact testing, d) representative corroded specimen. Before testing at reduced temperatures, the specimens were conditioned to the target test temperature using a LAUDA Edition X temperature conditioning unit. After removal from the conditioning medium, each specimen was transferred to the impact testing machine and tested within 5 s, in accordance with the time requirement of ISO 148-1 [32]. The same testing procedure was applied to both reference and corrosion-exposed specimens. For each condition, three specimens were tested, and both the mean absorbed energy and the scatter between specimens were considered in the evaluation. Statistical analysis The absorbed impact energy values of the eleven experimental groups were evaluated using one-way analysis of variance (ANOVA), with the experimental group treated as a single categorical factor. The analysis was performed to determine whether an overall statistically significant difference existed among the group means. Statistical significance was accepted at p < 0.05. Because only three specimens were tested for each condition and considerable within-group scatter was observed in several groups, the individual measurements and mean ± standard deviation values were also reported, and differences between specific experimental conditions were evaluated based on the observed numerical trends. Pearson’s correlation coefficient (r) was used to evaluate the linear relationships of hardness and corrosion mass loss with respect to absorbed impact energy. The hardness–energy analysis included all individual specimens from the eleven

192