AMP_06_September_2021

A D V A N C E D M A T E R I A L S & P R O C E S S E S | S E P T E M B E R 2 0 2 1 1 6 content along the slip band, thereby di- minishing the strength at the head of the pileup [30,31] . This is why the deforma- tion pathway B to B’ in Fig. 4 is not the weakest link despite having a long slip length. Pathways C to C’ and D to D’ have the primary α both crystallographical- ly and morphologically aligned, but the differently oriented secondary α colo- ny in C to C’ and the basketweave mor- phology in D to D’ are sufficient to block the slip band. Pathway E to E’ consists of interconnected primary α with simi- lar orientations, but the spatial arrange- ment of the particles does not permit extended basal slip and hence only sin- gle-grain-sized slip bands can devel- op. A continuous network of primary α grains with similar orientations well ori- ented for basal slip and spatially orient- ed for slip transfer is required to create the long-range slip bands and disloca- tion pileups for early crack nucleation. the same qualities that produced facet- ed regions on a dwell fracture surface. This work highlights the need for specif- ic MTR segmentation strategies for dif- ferent purposes. Harr et al. proposed that MTR segmentation strategies for long-range slip activity need to include a requirement for grain connectivity and a higher degree of c-axis alignment. This differs from the behavior of MTRs in crack growth studies, where similarly oriented grains do not need to be inter- connected to behave cooperatively and permit rapid crack growth. When considering MTRs and their influence on increased slip length and strain localization, it is important to keep in mind both slip transfer within MTR grains as well as the longer wave- length neighborhood effects of the MTRs. These examples demonstrate that the presence of MTRs facilitates for- mation of long-range plastic slip with- in MTRs below the macroscopic yield stress in titanium alloys. This is further supported by computational work by Kasemer et al. [27] . Additionally, the defi- nition of an MTR should be considered from the mechanistic basis of the pro- posed data application. Long-range slip at stresses below macroscopic yield in titanium alloys requires a stricter set of MTR definition parameters than when considering cooperative grains under fatigue crack growth. MICROSTRUCTURAL ‘WEAK LINK’ Because titanium alloys lack hard second phases, inclusions, and pores when properly melted and convert- ed to billet, the microstructural weak links are special aggregates of micro- structure that permit strain localization, crack nucleation, and easy small crack extension. Based on the authors’ ex- perimental observations and survey of the literature cited here, as well as the entire body of relevant literature, this section hypothesizes the microstruc- tural weak link for dwell fatigue in tita- nium alloys and describes why other attributes of the microstructure cannot be the weakest link based on what is known about the underlying deforma- tion behavior. Fig. 4 shows a schematic bimodal microstructure. It is constructed from underlying β phase, which is similarly oriented (red background hues) along with similarly oriented primary α, as ev- idenced by the dotted basal plane slip traces. The secondary α phase features either colony or basketweave morphol- ogy. Most cases show colony morpholo- gy in the secondary α phase and further, the underlying β phase was suitably ori- ented such that the secondary α adopt- ed a similar orientation to the adjacent primary α. There are a total of five deforma- tion pathways illustrated in Fig. 4. The first, connecting point A to A’, is the critical arrangement that would al- low development of a large slip band that creates a pileup of appreciable strength. It is bounded on both ends by obstacles to further dislocation motion, i.e., basketweave secondary α on top and a highly misoriented “hard grain” on the bottom. The large slip length is possible due to the sim- ilar orientation of the primary α grains as well as the spatial arrange- ment that aligns the basal slip traces of the interconnected α. Moreover, if one considers the effects of Al and O on deformation behavior [28,29] and the al- loying element parti- tioning effect [1] , one can conclude that slip in the primary α phase will be the most planar, with the most strain local- ized into a small num- ber of slip bands in primary α compared to the secondary α phase, which is depleted in Al and O. Further, the work of Suri et al. and Sav- age et al. indicates that slip in the secondary α phase would result in residual dislocations being left at α/β inter- faces; this would dis- tribute the dislocation Fig. 4 — Schematic bimodal microstructure of an α/β titanium alloy with representative (0001) slip traces. Long continuous (0001) traces that could contribute to dislocation pileups occur through regions of similarly oriented primary α. Long-range slip is inhibited by primary α of a different orientation, β phase, and basketweave microstructures. ’