Numerical and experimental evaluation of temperature field and melt flow in keyhole laser welding of dissimilar duplex stainless steel and nickel base alloy

To achieve high quality joint in keyhole laser welding of two dissimilar metals, phase transition behavior, the temperature and velocity field according to the variation of the process parameters were evaluated by utilizing both experimental and numerical approach. Due to the existing complex phenomena, the comprehensive analysis of the weld geometry and temperature field dependency in keyhole formation was performed either numerically or experimentally. An accurate numerical simulation of temperature and velocity fields, as well as material phase change at circular geometry path of laser beam movement were analyzed on dissimilar metals of duplex 2205 stainless steel and AISI 685 alloy metals to estimate such mentioned phenomena that could not be merely evaluated via experiments. A multi-physics numerical model that employed the finite volume method (FVM) and volume of fluid method (VOF) was utilized. The major novelty of dissimilar circular weld joint was simultaneous estimation the effect of different size and thereby volume of AISI 685 alloy and duplex 2205 alloy on the parts heat sink capacity, temperature gradient, melting ratio, fusion zone microstructure and fusion zone melt volume. The main reason for this is the asymmetric temperature distribution, resulting from the combined effects of material properties and the differing geometries and material volumes of the welded parts. To distinguish the laser process parameters, impact on the weld characterization according to the numerical simulation, the findings demonstrated that increasing the speed of the laser beam leads to the formation of bulge on the part's surface and around the keyhole while simultaneously diminishing the vapor volume. Furthermore, the laser beam's deviation from −0.25 mm at the AISI 685 alloy sheet to +0.25 at duplex 2205 led to the temperature reduction up to 300 °C at 1 mm distance from the joint centerline. Comparing the weld bead geometrical changes according to the variation of laser power and welding speed implies that the predicted temperature field of numerical simulation results is in good agreement with experimental results of weld bead geometry. The maximum error for experimental temperature measurement according to the variation of welding speed and laser power was less than 3 percent. By increasing laser power from 300 to 400 W, not only has the weld bead width become twofold, but also it penetrated toward the thickness completely, and the amount of weld bead overlap evidently increased more than 40 percent. The dissimilar joint fusion zone is mainly composed of cellular and columnar dendrite microstructure mainly created from nickel base alloy solidification according to the rapid heating followed by fast cooling induced by laser heating during welding.