Structural integrity assessment of Inconel 617/P92 steel dissimilar welds for different groove geometry

Abstract

The work is focused on examining the effect of the weld groove geometry on microstructure, mechanical behaviour, residual stresses and distortion of Alloy 617/P92 steel dissimilar metal weld (DMW) joints. Manual multi-pass tungsten inert gas welding with ERNiCrCoMo-1 filler was employed to fabricate the DMW for two different groove designs: Narrow V groove (NVG) and Double V groove (NVG). The microstructural examination suggested a heterogeneous microstructure evolution at the interface of the P92 steel and ERNiCrCoMo-1 weld, including the macrosegregation and element diffusion near the interface. The interface structure included the beach parallel to the fusion boundary at the P92 steel side, the peninsula connected to the fusion boundary and the island within the weld metal and partially melted zone along Alloy 617 fusion boundary. An uneven distribution of beach, peninsula and island structures along the fusion boundary of P92 steel was confirmed from optical and SEM images of interfaces. The major diffusion of the Fe from P92 steel to ERNiCrCoMo-1 weld and Cr, Co, Mo, and Ni from ERNiCrCoMo-1 weld to P92 steel were witnessed from SEM/EDS and EMPA map. The Mo-rich M6C and Cr-rich M23C6 phases were detected in inter-dendritic areas of the weld metal using the weld’s SEM/EDS, XRD and EPMA study, which formed due to the rejection of Mo from the core to inter-dendritic locations during solidification. The other phases detected in the ERNiCrCoMo-1 weld were Ni3(Al, Ti), Ti(C, N), Cr7C3 and Mo2C. A variation in the microstructure of weld metal from top to root and also along the transverse direction in terms of composition and dendritic structure and also due to the composition gradient between dendrite core and inter-dendritic areas, a significant variation in hardness of weld metal was observed from both top to root and also in the transverse direction. The peak hardness was measured in CGHAZ of P92 while the minimum was in ICHAZ of P92 steel. Tensile test studies of both NVG and DVG welds joint demonstrated that failure occurred at P92 steel in both, room-temperature and high-temperature tensile tests and ensured the welded joint’s applicability for advanced ultra-supercritical applications. However, the strength of the welded joint for both types of joints was measured as lower than the strength of the base metals. In Charpy impact testing of NVG and DVG welded joints, specimens failed in two parts with a small amount of plastic deformation and impact energy of 99 ± 4 J for the NVG welds joint and 91 ± 3 J for the DVG welded joint. The welded joint met the criteria for boiler applications in terms of impact energy (minimum 42 J as per European Standard EN ISO15614-1:2017 and 80 J as per fast breeder reactor application). In terms of microstructural and mechanical properties, both welded joints are acceptable. However, the DVG welded joint showed minimum distortion and residual stresses compared to the NVG welded joint.