Microstructures and Grain Refinement of Additive-Manufactured Ti-xW Alloys

Abstract

It is necessary to better understand the composition-processing-microstructure relationships that exist for materials produced by additive manufacturing. To this end, Laser Engineered Net Shaping (LENSä), a type of additive manufacturing, was used to produce a compositionally graded titanium binary model alloy system (Ti-xW specimen (0 £ x £ 30 wt pct), so that relationships could be made between composition, processing, and the prior beta grain size. Importantly, the thermophysical properties of the Ti-xW, specifically its supercooling parameter (P) and growth restriction factor (Q), are such that grain refinement is expected and was observed. The systematic, combinatorial study of this binary system provides an opportunity to assess the mechanisms by which grain refinement occurs in Ti-based alloys in general, and for additive manufacturing in particular. The operating mechanisms that govern the relationship between composition and grain size are interpreted using a model originally developed for aluminum and magnesium alloys and subsequently applied for titanium alloys. The prior beta grain factor observed and the interpretations of their correlations indicate that tungsten is a good grain refiner and such models are valid to explain the grain-refinement process. By extension, other binary elements or higher order alloy systems with similar thermophysical properties should exhibit similar grain refinement.

High Throughput Assays for Additively Manufactured Ti-Ni Alloys Based on Compositional Gradients and Spherical Indentation

“…In alloys such as Ti-6Al-4 V, which has a very narrow freezing range, such epitaxial growth is quite common. In other alloys where it is possible to further undercool the liquid or promote nucleation, it is possible to develop equiaxed grain structures [40,41]. These results suggest that Ni may also be suitable for promoting equiaxed grains, although the precise solidification details (e.g., Marangoni convection, undercooling, growth restriction [42]) have not been established.…”

Modeling of Ti-W Solidification Microstructures Under Additive Manufacturing Conditions

“…The simulated concentration fluctuations can be compared to experimental data of the concentration variation of W collected by EDS, as described in Mendoza et al [63] and shown in Figure 11(b). The experimental concentration profile showed variations in solute concentration with periodicity that is quite close to what we predict in the simulations (about 7 µm).…”

“…[63] We employed the commercial software package COMSOL Multiphysics ® to model the macroscale process as a simplified beam melting problem, considering the effects of fluid flow, heat transport, and temperature-dependent materials properties. At the microscale, information passed from COMSOL simulations is used to determine initial conditions, thermal gradients, cooling rates, and fluid flow velocities to be passed to a LB-CA model.…”

“…The model simulates a stationary beam melting problem, with a fixed molten pool geometry (estimated based on SEM images) that is varied as a function of input power. [63] Owing to these assumptions, calculated values for temperature and fluid flow fields are used here only to estimate the thermal gradients, cooling rates, and fluid flow fields at short times near the melt pool walls after turning off the energy source.…”

Powder-based additive manufacturing - a review of types of defects, generation mechanisms, detection, property evaluation and metrology

“…Once entrained in the liquid, the melt pool dynamics are equally complex, with Marangoni convection, evaporation, wetting and capillarity playing strong roles (among many other operating physics). The liquid metal velocity is quite high, and results in features that resemble comet tails as melting particles leave molten material behind them as they move through the molten pool (Mendoza et al, 2017). These physics present challenges in understanding and modelling AM processes, but it is expected that over the next three to five years, a number of research groups will be able to contribute to the knowledgebase of the heat source/particle interactions.…”