Effect of Build Parameters and Build Geometries on Residual Microstructures and Mechanical Properties of Ti-6Al-4V Components Built by Electron Beam Melting (EBM)

In this study, involving additive manufacturing (AM) using electron beam melting (EBM), we have examined build defects which result from beam tripping, porosities (including unmelted or unsintered zones) due to excursions from optimal build parameters (especially variations in melt scan beam current and scan speed), and gas bubbles trapped in atomized Ti-6Al-4V starting powder as well as recycled powder, and retained in the build. At optimized build conditions we have also examined microstructure-mechanical property (hardness, tensile strength, and elongation) variations for multiple rake building and multiple melt scans using optical metallography and scanning and transmission electron microscopy (SEM and TEM). These build variances cause cooling rate variances which promote α-phase growth and variations in dislocation density, as well as α-to-α' (martensite) phase changes, all of which produce some degree of mechanical property variations. These features (especially α-to-α' phase changes) are notable on comparing solid builds in comparison with a variety of mesh arrays where strut dimension and build-element complexities alter the cooling rates in a significant way. We illustrate these microstructure variations with corresponding variations in microindentation hardness measurements made directly on fine mesh (strut) structures. Finally, we have examined Ti-6Al-4V powder chemistries and solid build chemistries which for single-pass melt scans at optimized build conditions are shown to be relatively constant up to 40 cycles of powder reuse with the exception of Al content which was reduced by 10 to 15% in solid builds at optimized conditions. However, Al loss in solid builds approached 25% for multiple (2 and 3) melt scans, while no changes in Ar gas-bubble density were observed with changes in α-phase (grain) width which increased from 3 µm for a single melt scan to 4.5 and 6 µm for 2 and 3 melt scans, respectively. Corresponding Rockwell C-scale (HRC) hardness varied from 37, 36, and 35, respectively; with ultimate tensile strengths exceeding 1.2 GPa at elongations of 12% or higher for this melt scan sequence.