Assessment of intra-build variations in tensile strength in electron beam powder-bed fusion Ti–6Al–4V part 2: Effects of powder mixing

In Part 2 of this paper series, high-throughput tensile testing and characterization of porosity, microstructure, and oxygen content was carried out on Ti–6Al–4V samples that were fabricated via electron beam powder-bed fusion (PBF-EB) to test our hypothesis that current industrial powder mixing strategies lead to stochastic chemistry content and tensile property variation throughout a given build. Fifteen rectangular blocks were built at locations across the entire build plate and machined into high-throughput tensile racks that contained 25 mini tensile coupons each (375 total specimens). There was a spread of 138.9 MPa in yield strength (YS) across all specimens. Considering the YS spread was only 74.2 MPa for a virgin build with no mixing, as detailed in Part 1, the difference between these two YS ranges is expected to be attributed to mixing. To determine the cause for this observed spread in YS, systematic investigation of porosity, microstructure, and oxygen content was performed via scanning electron microscopy (SEM), micro X-ray computed tomography (μXCT), inert gas fusion, and time-of-flight secondary ion mass spectrometry (ToF-SIMS). The large variation in YS for these specimens could not be attributed to internal porosity, fracture surface porosity, α-lath thickness, or crystallographic texture. Inert gas fusion measurements indicated large, stochastic variation in oxygen content, and measurements via ToF-SIMS near the fracture surfaces indicate localized chemistry variations that could be responsible for the large variations in tensile properties, specifically the yield strength. Specific alternative powder mixing and reuse methodologies are proposed to mitigate undesirable chemistry and tensile property variation within PBF-EB Ti–6Al–4V.