Strength, fatigue strength and toughness of dissimilar Ti17–Ti64 linear friction welded joints: Effect of soft surface contamination and depletion of α precipitates

Abstract

The effect of microstructure of three dissimilar Ti17-Ti64 linear friction welded joints on the strength, fatigue strength and fracture toughness was studied. A special attention was paid to role of soft contaminants and α precipitate depletion. Three joints were produced: one in the as machined state after electrical discharge machining, one ground prior to welding, and one welded in the as machined state and post-weld heat treated. The microstructure of the weld centre zone (WCZ) exhibited an acicular entangled α martensite with a Widmanstätten morphology on the Ti64 side and fine equiaxed distorted β grains on the Ti17 side. A depletion of the strengthening α precipitates was observed in the WCZ and the thermo-mechanically affected zone (TMAZ) on the Ti17 side. For the studied welding parameters, the LFW selfcleaning mechanism failed to extrude the contaminants in the weld interface of the joint welded in the as machined state. These contaminants were related to prior to welding machining and were denominated soft contaminants due to their relatively low melting point. A defect layer was formed that led to a pseudo-brittle fracture in at the WCZ during tensile testing, cyclic loading and fracture toughness testing. During tensile testing of the joint that was ground prior to welding, failure occurred sometimes at the Ti64 parent material (PM) and sometimes at the Ti17 TMAZ. Fatigue cracks initiated at the latter weak zone with a reduced fatigue strength. Compared to the PM, fracture toughness of samples even without fatigue pre-crack was also reduced. Failure at the weak zone highlighted unexpected colonies of elongated dimples that nucleated on transgranular deformation bands.

Fatigue Behavior of Linear Friction Welded Ti-6Al-4V and Ti-6Al-2Sn-4Zr-2Mo-0.1Si Dissimilar Welds

“…Previous studies on LFW of titanium alloy joints (similar and dissimilar) have also applied different microscopic techniques (e.g., optical, SEM) to link the microstructural transformation to the hardness evolution [ 10 , 11 , 15 , 22 , 26 , 27 , 40 , 48 , 49 , 50 ]. A particularly favored method for linear friction welds is the use of electron backscatter diffraction (EBSD) to map the orientation of the α-phase grains [ 13 , 18 , 23 , 26 , 27 , 28 , 51 , 52 ]; this method allows visual differentiation of the WC (with its recrystallized fine grain structure of α′ martensite) and the TMAZs (with their plastically deformed and elongated α-grain structure) from the bimodal BM microstructure.…”

“…This explains the observations of García and Morgeneyer [ 33 ] in their study of the fatigue performance of Ti6242 linear friction welds that exhibited low fatigue strength values, as well as premature/early failures due to remnant porosity at the joint interface. Recently, Garcia et al [ 11 ] also reported on the detrimental impact of defects—such as oxides, pores and contaminants that remain in the joint and originate from inadequate preparation of the faying surfaces prior to LFW—on the fatigue behavior of dissimilar Ti17–Ti64 linear friction welds that failed prematurely at a very low number of cycles or within the first fatigue cycle. Moreover, their work indicated that in the presence of defects, the application of PWHT to relieve residual stresses or modify the microstructure was ineffective in overcoming the compromised fatigue resistance from poor weld integrity, which aligns precisely with the expected outcome from the structural hierarchy.…”

“…Though titanium alloys generally exhibit good weldability, even the application of high-energy-density fusion welding in a reducing vacuum environment would present performance losses—due to, at best, only grain coarsening in the fusion and heat-affected zones (HAZs), if contamination and solidification defects (such as porosity and underfill) could be prevented [ 2 , 3 , 4 , 5 , 6 , 7 , 8 ]—that would compromise the operability/reliability of such critically stressed structures. A key niche technology for manufacturing titanium alloy blisks has been linear friction welding (LFW), a solid-state joining process that can produce sound welds (defect-free) with a joint efficiency—strength of a welded joint with respect to the strength of the base metal (BM)—above 100% and failure in the BM [ 9 , 10 , 11 ]. This technology is suitable for the assembly of complex part geometries and titanium alloys can be reliably joined without the need for any controlled or shielding atmosphere due to the low process temperatures (considerably below the melting point) and rapid welding times (approximately 2 s).…”

“…For alloy Ti–6Al–2Sn–4Zr–2Mo–0.1Si (Ti6242), García and Morgeneyer [ 34 ] also studied the fatigue performance in the AWed condition and reported lower fatigue strength values (with a couple of early failures), as compared to the BM, which they attributed to remnant pores in the weld originating from contamination on the faying surfaces of the joint prior to LFW. Garcia et al [ 11 ] also recently examined the fatigue strength of dissimilar joints between near-β alloy Ti17 and α + β Ti64 and further highlighted the important role of surface preparation prior to LFW on the static and cyclic performance of the joints; their work also indicated that PWHT was ineffective in reversing weld integrity—and concomitant performance—losses that occurred due to bonding defects from surface contaminants.…”

On the use of stereo‐digital image correlation for the alignment of a fatigue testing machine in accordance with international standards: A feasibility study

“…Furthermore, this assumes that Young's modulus of the studied material is known and constant, which is not the case for cross‐weld specimens. [ 8–10 ] An analytical expression of the stress field induced in fatigue specimens as function of the displacement field imposed by the misaligned load frame, in particular the bending deflection, is required.…”