EBSD Characterization of Microstructural Variations in Solid-State Welds as a Function of Distance from the Weld Interface in Ti-17

Abstract

Recent advances in welding technology have enabled the aerospace industry to reduce further the weight of aircraft, by welding titanium alloys reliably. Solid-state welding, in which two metal pieces are welded without either one melting, has recently been proven to produce reliable and consistent welds, and is rapidly becoming the preferred method for joining titanium. Rapid heating rates, short hold-times, plastic deformation, thermal gradients, and fast cooling rates all contribute to the non-equilibrium nature of the process. All of these factors vary with respect to their proximity to the weld interface, resulting in various regions of the welded piece. Thus, the resulting microstructure varies with distance from the weld plane, and can typically be divided into four zones: the dynamically recrystallized zone (DRX), the thermo-mechanically affected zone (TMAZ), the heat-affected zone (HAZ) and the base material (BM) [1]. Commercial titanium alloys have two equilibrium phases: the low-temperature hexagonal closepacked (hcp) phase, α, and the high-temperature body-centered cubic (bcc) phase, β. Ti-17 (Ti-5Al-4Cr4Mo-2Sn-2Zr) is a metastable- β alloy, meaning that it can retain essentially 100% β phase, upon fast cooling [2]. Since the mechanical properties of titanium alloys are particularly dependent upon microstructure, it is important to understand the phase transformations occurring and the microstructures produced during the solid-state welding process. Once the various regions of the weld can be characterized, the data can help better inform integrated computational materials engineering (ICME) models. In this study, samples of Ti-17-to-Ti-17 solid-state welds were sectioned via electric-discharge machining (EDM), normal to the weld interface. The cut face was then polished according to metallographic polishing procedures for titanium alloys outlined in [3]. Once polished, the surface was indented with a Buhler microhardness indenter, using 500 g load force, and a 10 s dwell time. The first indent was placed at the weld interface and subsequent intents were placed every 100 µm, extending to 10 mm from the weld interface, on each side. To ensure accuracy of the values, each sample received multiple, parallel, identical rows of indentations. The Vicker’s Microhardness values were recorded and plotted in Figure 1. These microhardness indents could then be used as fiducial markers for recording of EBSD data, enabling precise correlation between microhardness and microstructure. Inverse pole figure maps were used to determine the amount of recrystallization in the beta phase, as well as the degree of dissolution of the alpha phase and whether any re-precipitation occurred. The microhardness profiles displayed symmetry with respect to the weld interface. The lowest hardness levels occurred at the weld interface, in the DRX zone, and generally increased as distance increased. Both sides of the interface showed local peaks and local minima in hardness at distances of approximately 1100 µm, and 1600 µm from the interface, respectively. The hardness returned to the baseline level, at a distance of approximately 2500 µm from the weld interface, which is in agreement with microstructural observations. EBSD data were collected at ~100 µm intervals for the first 2 mm,