METALS AND MATERIALS INTERNATIONAL, cilt.30, sa.7, ss.1977-1996, 2024 (SCI-Expanded)
This study investigates the influence of the shielding gas composition on the microstructural and mechanical properties of wire arc additive manufactured duplex stainless steel. The choice of shielding gas influences the quality and performance of these steels in various industries. Three shielding gases, namely Ar, Ar + 2wt%O2 (Ar-O), and Ar + 2wt%N2 (Ar-N) were used for the deposition of walls with 2209 duplex stainless steel wire. Extensive microstructural analysis and mechanical property evaluations were performed. Based on microscopic observations, the walls produced under different gases exhibited varying inclusion levels, primarily composed of silicon-rich oxide inclusions, with the Ar-O sample exhibiting the highest inclusion concentration at 1.5 vol%. Moreover, the microstructure of all samples consisted of ferrite and austenite phases with different austenite morphologies, with intergranular austenite dominating in the Ar-O sample. In three samples, different degrees of preferred orientation were found in both the ferrite and austenite phases. In this regard, the Ar and Ar-O samples had the strongest and weakest texture, respectively, owing to the influence of inclusions on the formation of a random texture. The microhardness mapping revealed the highest hardness of 250 HV in the Ar-N sample and the lowest hardness of 220 Hv in the Ar-O sample. Nitrogen's strengthening effect on austenite in the Ar-N sample and the higher austenite content in the Ar-O sample were believed to be responsible for these variations. Nanohardness indentations confirmed the lower hardness of austenite compared to ferrite, corroborating the overall lower hardness observed in the Ar-O sample. The Ar-N sample exhibited the highest tensile strength and yield strength of 826 and 539 MPa, respectively in the horizontal direction, which is consistent with the observed hardness results. On the other hand, the Ar sample exhibited the highest elongation of all samples at 39%, which is consistent with its lower inclusion content.