A design strategy overcomes the strength-ductility trade-off in alloy manufacturing

Three-dimensional (3D) printing, also known as additive manufacturing, continues to reshape industries, including metals production. Among its advantages are decreasing the time and costs for creating intricate metal parts and increasing customization. However, the technology still faces challenges in achieving uniform mechanical properties in 3D-printed metallic alloys. On page 639 of this issue, Zhang et al. (1) report a design strategy for printing a robust titanium alloy. The authors show that the addition of molybdenum (Mo) to the powder metal mixture enhances phase stability and improves the strength, ductility, and uniformity of tensile properties of the 3D-printed alloy. The approach could potentially be applied to other powder mixtures and enable the customization of different alloys with enhanced properties.

In the layer-by-layer 3D printing process (typically with a high cooling rate of approximately 10³ to 10⁸ K/s), a substantial thermal gradient forms near the edge and bottom of the melt pool, where metal powder has been melted by a laser beam. The thermal gradient induces epitaxial grain growth along the interface between the freshly melted material and the underlying solid material, with grains growing toward the melt pool center. Cycles of heating and partial remelting during the printing of multiple layers ultimately results in the formation of large columnar grains and heterogeneously distributed phases, both of which are undesirable because they can lead to nonuniform (anisotropic) and compromised mechanical properties (2, 3).

Titanium alloys are among the strongest metallic materials. In engineering applications at ambient temperature, a suitable titanium alloy typically exhibits a tensile elongation (the maximum stretch or deformation that a material can withstand before breaking) ranging from ?10 to ?25%, which reflects good material reliability. Although greater elongation (ductility) facilitates easier formability and holds priority in certain applications, an increased strength within this elongation range is preferred for enduring mechanical loads. In both conventional and additive manufacturing techniques for processing metallic materials, a trade-off between strength and ductility has been prevalent (4, 5).

Three-dimensional (3D) printing or additive manufacturing (AM) of metals and alloys typically involves multiple physical and metallurgical phenomena that impart complex microstructures and varied mechanical properties in the fabricated products (1–7). For metallic alloys, which solidify with a cubic crystal structure, columnar grains often prevail in the 3D-printed part (8, 9) because grains with the easy growth directions tend to align closely to the maximum temperature gradient of the melt pool and grow epitaxially from the partially melted layers (10). Although this highly textured columnar grain structure can be beneficial for certain applications, in most cases it is undesirable because it degrades the mechanical performance and results in mechanical property anisotropy (11, 12). Accordingly, extensive effort has been devoted to transforming the coarse columnar grain structure into fine equiaxed grains to achieve superior and isotropic mechanical properties (13–17). Generally, the columnar-to-equiaxed transition (CET) and grain refinement can be promoted through process control and by adding grain-refining inoculants (12, 18). The former typically includes manipulating the 3D printing processing parameters or introducing external interferences (16, 17). However, the effectiveness and practicalities of this approach are limited to specific alloys or 3D printing technologies. Alternatively, the metallurgical approach through additives has proven highly effective but often results in an undesirable loss in ductility owing to the formation of brittle second phases (9). Therefore, simultaneously addressing the coarse columnar grains and eliminating property anisotropy without adversely affecting the ductility is highly desirable.

A further complication is that many allotropic alloy systems, including titanium alloys, are also susceptible to the heterogeneous distribution of phases associated with the solid-state thermal cycling experienced during the 3D printing process (19–22). This poses an additional challenge to achieve uniform mechanical properties of 3D-printed parts made from these alloys (23, 24). The localized heating, cooling, and reheating nature of 3D printing effectively creates dynamic in situ heat treatments that encourage the decomposition of initially formed phases and/or the precipitation of new phases through solid-state phase transformations (20, 25). Because thermal cycles are spatially variable, the associated heat treatments can produce an inhomogeneous distribution of phase along the building direction of the part, thereby resulting in the spatial variation of mechanical properties (21, 22). Postprinting heat treatments can be effective in mitigating these phase heterogeneities but introduce delays and additional costs and are not effective in refining textured columnar grains (26). The confluence of these issues has made it extremely challenging to achieve uniform and superior mechanical properties in the as-fabricated state.

We demonstrate a design strategy to address this challenge by simultaneously controlling the grain structure and constituent phases in products manufactured by laser powder bed fusion (L-PBF). We selected the Ti?5Al?5Mo?5V?3Cr (Ti-5553) metastable ? titanium alloy as a model alloy because it shows the coexistence of coarse columnar ? grains and a heterogeneous distribution of phases (Fig. 1, A to C). This results in highly nonuniform, position-dependent tensile properties from L-PBF, as we demonstrate (Fig. 1, D and E) and as has been demonstrated in other studies across multiple 3D printing technologies (27–29). We show that the single addition (up to 5.0 wt %) of elements from the ?-isomorphous group [in this case, we selected Mo; see section on alloy design in the materials and methods (30)] into Ti-5553 powder to form a composite blend achieves bifunctionality: (i) During 3D printing, some of the Mo particles partially melt, but the core survives to nucleate fine grains during solidification and prevent coarse columnar grains from forming...

Fig. 1. Microstructures and mechanical properties of Ti-5553 produced by L-PBF.
(A) The coexistence of coarse columnar ? grains and spatially dependent phases in Ti-5553 produced by L-PBF. (1) Schematic illustration of the L-PBF process. (2) EBSD IPF map showing coarse columnar ? grains along the building direction (BD). (3) SEM-BSE micrographs showing the phase distribution along the BD. The yellow arrows point out ? phases with a darker contrast in the ?-Ti matrix. (B) Schematic illustration of the microstructure heterogeneity in terms of columnar ? grains and heterogeneously distributed phases on the cross-section S?S (the yz-plane), as indicated in (A). (C) TEM micrographs of Ti-5553. (1) Dark-field TEM image showing ? phases. (2) TEM SAED pattern from the [111]? zone axis showing the presence of ? phases. The key diagram of the diffraction is shown in (3). Note that there are three ? variants with the same zone axis of (2110), which grow in different directions. (4) TEM SAED from the [113]? zone axis showing the existence of isothermal ? phases. The key diagram of the diffraction is shown in (5). Note that there are two ? variants. (D) Tensile engineering stress-strain curves of Ti-5553 horizontal tensile specimens (1, 3, 5, 7, 9, and 11). The inset shows how the horizontal tensile specimens were machined from the Ti-5553 part. (E) Tensile engineering stress-strain curves of Ti-5553 vertical tensile specimens (1, 3, 5, 7, 9, and 11). The inset shows how the vertical tensile specimens were machined from the Ti-5553 part.