PSI - Issue 53

8

Luca Marchini et al. / Procedia Structural Integrity 53 (2024) 203–211 Author name / Structural Integrity Procedia 00 (2019) 000–000

210

Moreover, the thickness of the intermetallic layers growing on the AM surface remains relatively constant, at approximately 10 μ m, in the period from 30 minutes to 2 hours. In contrast, the η phase layer on the F sample exhibited a notable increase of thickness (about 50 μ m) within the same testing period. The relation existing between thin intermetallic layers and reduced interdiffusion rates is confirmed by the data obtained for H11 samples, where the layer thickness remained very small also after 2 hours of immersion test, in the range of 5 - 10 μ m (Fig. 5f). As for the AM maraging steel, the intermetallic layer formed on H11 comprises two distinct layers. However, in this instance, Al 3 Fe is the inner layer, while a ternary Al 4.5 FeSi - τ 3 phase is the outer one facing the molten Al alloy. Notably, chromium carbide particles from the steel matrix, are embedded in the intermetallic layer as already observed in the H11 sample after 30 minutes. In agreement with the existing literature, such carbides serve as a physical barrier to the interdiffusion of Al and Fe atoms, significantly increasing the corrosion resistance of H11 steel (Xu et al. , 2021). Al 3 Fe represents the second most prevalent Al-Fe intermetallic compound after Al 5 Fe 2 , possibly due to its lower negative enthalpy of mixing and slower growth kinetics (Dangi). It is established that Al 3 Fe intermetallic primarily forms through precipitation within aluminum melt regions that are supersaturated with Fe atoms during cooling from relatively high temperatures, or alternatively, it forms at the steel interface via diffusion mechanisms (Chen et al. , 2016). Nevertheless, the presence of this intermetallic compound, contrary to Al 5 Fe 2 , may slow down the diffusion process. In fact, the growth of Al 5 Fe 2 intermetallic layer proceeds rapidly when contact is maintained between Al/ Al 5 Fe 2 and Al 5 Al 5 /Fe interfaces. Furthermore, another reason of the superior corrosion resistance of the H11 steel could be linked to a higher C content. In fact, as Si, the enrichment of this element at the reaction interface with the melt can inhibit the interdiffusion of Fe and Al atoms. 4. Conclusion In this study, a comparative analysis of the hot-corrosion resistance in a molten low-iron AlSi7Mg alloy have been carried out by comparing the 1.2709 maraging steel produced via additive manufacturing with the conventional forging process. Additionally, a well-established H11 tool steel was examined as a reference material. The results showed that the primary mechanism governing the hot corrosion of steel samples is the diffusion of elements across the intermetallic layers formed at the surface due to the prolonged static immersion. The higher material loss was observed in case of forged maraging samples, particularly in comparison with the parent AM sample. Using the best performing H11 tool steel as a reference point, the forged samples exhibited a 2.5 times higher thickness loss compared to the 1.7 times greater loss observed in the AM samples (after 2 h of immersion time). Furthermore, both maraging samples displayed non-uniform corrosion, especially in the case of the forged samples. This non-uniformity could be attributed to the segregation of alloy elements within the forged components, characterized by the typical microstructural banding. However, further analyses are necessary to gain a more comprehensive understanding of this behavior. SEM-EDS analysis revealed that the AM samples were able to form a more stable intermetallic layer rich in silicon compared to the forged counterparts, which dissolves completely for testing times above 2 hours. Considering that literature states that the Al-Fe-Si phases hinder the growth of intermetallic compounds, this may be one of the primary reasons for the slower diffusion-related material loss observed in the AM sample. On the other hand, the analysis of H11 samples showed the formation of slightly different intermetallic compounds, with embedded chromium carbides. The extremely thin interaction layer formed over the H11 steel surface is indicative of a low diffusion rate and can be attributed to the inhibitory effects of chromium and carbon, which impede the diffusion of Fe and Al atoms. Chemical analysis of the AlSi7Mg alloy post-testing revealed the selective diffusion and dissolution of Ni from the maraging steels into the molten alloy. This observation may suggest a greater hindrance to Ni diffusion imposed by the intermetallic Al-Fe-Si layer formed over the AM sample. The presence of slightly differences in the alloying elements of these maraging steels complicates the interpretation of their corrosion behavior. Indeed, as evidenced, small differences in the quantities of alloying elements within the intermetallic layers can have some influence on diffusion coefficients and, consequently, on the dissolution rate of steel into the molten Al alloy. Consequently, it becomes challenging to determine the role of the differential microstructures in the corrosion behavior in liquid Al. Therefore, further analyzes are needed to