PSI - Issue 82

Goran Vukelić et al. / Procedia Structural Integrity 82 (2026) 3– 8 Vukelić et al./ Structural Integrity Procedia 00 (2026) 000–000

5

3

peened to achieve the final surface finish. Rectangular welding blanks were subsequently joined using the tungsten inert gas (TIG) welding technique, ensuring complete joint penetration (Khedr et al., 2025). Three types of welded blanks were prepared: (i) CM-CM plate combinations, (ii) AM-AM plate combinations, and (iii) hybrid CM-AM plate combinations, Fig. 1a. Butt-welded specimens were then sectioned from the welded blanks using laser jet cutting technology, Fig. 1b. Each category comprised twenty specimens, grouped into sets of five. Specimen groups were exposed to the natural marine environment for one, three, and six months, respectively, while one control group was stored under laboratory atmospheric conditions. The specimens were exposed to natural seawater at a depth of 10 m below the surface in the northern Adriatic Sea, off the coast of Rijeka, Croatia. Following immersion, the specimens were retrieved, Fig. 1c, and cleaned of corrosion products using a soft brush under running water. The cleaned samples were subsequently rinsed with distilled water and dried at ambient room temperature. The cleaning procedure followed established protocols reported in previous studies. Specimens were weighed before and after exposure using an Ohaus 3000 digital balance to determine mass loss. Uniaxial tensile testing was carried out on a Hegewald & Peschke Inspekt Table 20 kN universal testing machine. The tensile specimens were prepared in accordance with the relevant standard (Vukelic et al., 2025). All tensile tests were performed following standardized procedures to obtain engineering stress-strain curves for subsequent mechanical property evaluation.

(a)

(b)

(c)

Fig. 1. (a) Welded blanks with waterjet cut specimens; (b) Three types of specimens tested; (c) CM-AM after 6 months exposure.

3. Results Based on the measured values obtained before and after marine exposure, the average mass loss percentage for each specimen group was calculated and correlated with the corresponding exposure duration, as shown in Fig. 2. In addition to the experimental data, approximation curves were generated to illustrate the relationship between average mass loss and exposure time, providing a visual representation of the observed trend. The coefficient of determination (R²) is included to indicate the quality of fit (Vukelic et al., 2022).