Symposium CP08 : Additive Manufacturing of Metals

Additive manufacturing has the potential to disrupt traditional fabrication methods for high add value applications because the flexibility of design, fast production speed for complex parts, and shorter lead times. However, there is currently a reduced range of metallic alloys that can be industrially processed using this technique. This is particularly problematic in the case of high strength Al alloys processed by additive manufacturing which present high porosity and an undesired microstructure formed by large columnar grains surrounded by longitudinal cracks (hot-cracking).
It has been recently shown [1] that these problems can be overcome through the addition of ZrH₂nanoparticles to the Al powders. In particular, hot cracking was significantly reduced and the microstructure was refined due to the stimulation of grain nucleation by the nanoparticles. In this investigation, a systematic study was carried out to analyze the effect of processing parameters (laser power and speed hatch distance) on the microstructure of 7075 Al alloy. Commercial powders of the 7075 Al alloy were mixed with zirconium dihydride particles by planetary milling without any grinding medium. Samples were additive manufactured by selective laser melting with and without zirconium hydride particles, using a wide range of process parameters. The microstructure of the processed samples was characterized by electron backscattered diffraction (EBSD), electron microscopy and transmission electron microscopy (TEM) in order to evaluate the influence of the addition of refining particles along the building direction.

The addition of ZrH₂led to an equiaxed grain structure (which was attributed to enhanced nucleation promoted by the Al-Zr precipitates) and to the elimination of hot cracking. A hardness increase of 30HV was observed in the Zr containing samples probably due to the hardening effect of an enhanced precipitate volume fraction. The effect of processing parameters on the microstructural features and on the hardness was analyzed.

[1] J. H. Martin et al. Nature, 549, 365–369, 2017.

The microstructure variation and heterogeneity in mechanical properties of 316L stainless steel fabricated using selective laser melting (SLM) have been investigated. The crack free, dense samples were made using SS316L alloy powder and melt pool morphology was analysed using optical and scanning electron microscopy. Extremely fast cooling rates during laser melting/solidification process, accompanied by slow diffusion of alloying elements, produced distinct microstructure with colonies of fine-grain size as well as elongated grains grown along the direction of thermocapillary convection in the melt pool. Precipitates were mostly observed at the grain boundaries while some times also appeared inside grains. The micro hardness measurements along the build direction revealed slight but gradual increase in hardness along the sample height. The tensile tests of as build samples showed that the mechanical properties of samples produced using SLM were observed to be quite heterogeneous. The ultimate tensile strength (UTS) varied from 460MPa to 510MPa along with elongation variation from 3% to 15%. The corrosion behaviour of SLM samples were also investigated by exposing the tensile samples to 5% NaCl solution for 240 hrs, at temperature of 46C. The UTS of a sample exposed to the corrosion environment revealed similar strength as of the original sample, indicating good corrosion resistance of SLM samples.

We demonstrate how Sn3Ag4Ti alloy can robustly bond to silicon via selective laser melting (SLM). By employing this technology, thermal management devices (e.g., vapor chamber evaporators, heat pipes, micro-channels) can be directly printed onto the electronic package without using thermal interface materials. Reduction of the chip operating temperature up to 10 °C, consumed power and electronic-waste are some of the advantages of this technology which lies through the elimination of thermal interface materials in the electronic package. Bonding of common metal alloys used in additive manufacturing and silicon is relatively weak and generally possesses high contact angles (poor wetting and interfacial strength). However, reactivity and wettability of the silicon substrate increases drastically by using proper interlayer material. Furthermore, conventional dissimilar material bonding can take tens of minutes to form a strong bond, this study demonstrates how this kinetic limitation can be overcome to form a good bond in microseconds via intense laser heating. Rapid bonding of Sn3Ag4Ti to silicon occurs through formation of a thin (∼µm) titanium-silicide interfacial layer that makes the silicon wettable to the Sn3Ag4Ti. This rapid bonding occurs due to formation of strong intermetallic phases at low temperature, and by exposing of the sample by laser multiple times which provides sufficient diffusion time. The printed parts are mechanically robust with no visible failures after over a week of continuous thermal cycling (−40 °C and 130 °C).

Powder bed fusion systems such as direct metal laser sintering (DMLS) provide unprecedented abilities to manufacture complex 3-D parts and structures; however, the process produces unused metal powder that can undergo significant change in chemistry and morphology. The ability to recycle the used powder that is typically found within the build volume and in the overflow compartments requires a thorough chemical and structural analysis in order to determine the extent of reusability. In this context, characterization of virgin powder and used powder for nickel-based and stainless steel based alloys was performed in this work in a rigorous fashion to compare and contrast properties such as composition, particle size distribution and morphology. This information was then used in conjunction with other process parameters (e.g. dose factor, spreader velocity, layer thickness, and laser energy density) to determine conditions that can enable the ready integration of used powder along with virgin powder. Experimental analysis was supported by ANSYS based process modeling in the optimization process, providing a path forward for cost effective additive manufacturing methods.

With the objective of combining advantages of both additive and subtractive manufacturing, several manufacturers are developing hybrid machine tools. Directed energy deposition (DED) method was the additive manufacturing method adopted in these hybrid machines because of the adaptability of the equipment to the milling platform. In order to take advantage of a hybrid platform and obtain a work piece with a good final quality, it is necessary to understand the influence of the additive process parameters. In this work, laser power and scan speed were systematically varied to fabricate 316L test samples to determine the optimal parameters in a recent developed hybrid machining center (Romi DCM 620 5x Hybrid). Optical stereoscopy indicates that the optimal energy input is 50 J/mm². Lower energy inputs are not sufficient to melt the powder and higher energy inputs than 50 J/mm² (with a shield gas flow of 5 L/min) result in porous samples due to an oxidation between deposited layers. Chemical composition was evaluated by Scanning Electron Microscopy coupled to an Energy Dispersive X-ray detector (SEM-EDS) along the height of the samples. Even though an oxidation was observed, EDS results do not reveal any significant Cr and Ni content variation from one measured point to another. Findings of this study will be used as a basis to develop the process with other alloys in hybrid machine-tools.

Microtextured and mesoporous metal alloys, due to its high surface area, have been widely used as fast-charging anodes for Li-ion batteries [1], as metal electrodes for electrolytic methane production [2], and for electrocatalysis applications [3]. In recent years, there has been a growing interest in designing architected and hierarchical metal electrodes by integrating wet-chemical methods [4] with 3D printing techniques in order to design and scale topologically complex catalyst systems [5] with increased surface area [6]. In this study, we examine a specific corrosion-based wet chemistry to induce the microtexturization of gas atomized metal alloy powder as a strategy to introduce surface roughness at microscales to AM feedstock intended for use in selective laser sintering. Microstructurization of NiCu powders was done by electroless chemical etching in 1M HNO3 with the formation of roughened surfaces with RMS at length scales from 20 nm to 5 μm. Additionally, post-processing of the etched microtextured powders in diluted HNO3 reduced the oxygen content from 4 % to 1 % by mass compared to the initial content present in the gas atomized metal powder. Thereby, development of the new controllable chemistry for modification and microtexturing of oxygen-free metal powders may be a promising way for fabrication of the new high surface area metal feedstock for 3D printing of high-surface area electrodes.
References
[1]
J. Li, J. Pu, Z. Liu, J. Wang, W. Wu, H. Zhang and H. Ma, "Porous-nickel-scaffolded Tin-Antimony anodes with enhances electrochemical properties for Li/Na-ion batteries," Applied Materials and Interfaces, vol. 9, pp. 25250-25256, 2017.
[2]
Y. Shen, M. Ge and A. Lua, "Deactivation of bimetallic nickel-copper alloy catalysts in thermocatalytic decomposition of methane," Catalysis Science & Technology, vol. 8, no. 15, pp. 3853-3862, 2018.
[3]
V. Vij, S. Sultan, A. Harzandi, A. Meena, J. Tiwari, W. Lee, T. Yoon and K. Kim, "Nickel-based electrocatalysts for energy-related applications: oxygen reduction, oxygen evolution, and hydrogen evolution reactions," ACS Catalysis, vol. 7, no. 10, pp. 7196-7225, 2017.
[4]
E. Seker, M. Reed and M. Begley, "Nanoporous gold: fabrication, characterization, and applications," Materials, vol. 2, pp. 2188-2215 , 2009.
[5]
C. Zhu, Z. Qi, V. Beck, M. Luneau, J. Lattimer, W. Chen, M. Worsley, J. Ye, E. Duoss, C. Spadaccini, C. Friend and J. Biener, "Toward digitally controlled catalyst achitectures: hierarchial nanoporous gold via 3D printing," Science Advances, vol. 4, no. 8, 2018.
[6]
T. Song, M. Yan and M. Qian, "The enabling role of dealloying in the creation of specific hierarchial porous metal structures - a reivew," Corrosion Science, vol. 134, no. 15, pp. 78-98, 2018.

In recent years, ultra-lightweight lattices with high stiffness, strength and resilience have been fabricated by coating polymeric microlattices with a thin layer of metal (e.g. Ni, Au) or ceramic (e.g. Al₂O₃). However, there is limited scalability in the production techniques, which typically involve time consuming processes such as 2-photon polymerization, atomic layer deposition and plasma etching. In this study, we aim to circumvent these limitations by making use of macroscopic stereolithography to 3D print lattices before applying a thin coat of metal through dip coating. The polymeric core of the lattices and binders in the coatings are then removed in a subsequent sintering step at ~ 1000 °C. The mechanical properties, morphology and chemical composition of the resulting hollow metal lattice are characterized with respect to different sintering duration and temperature. The low cost and easy processing of different shapes allow this technique to be commercially feasible for the mass production of lightweight hollow metallic structures.

The ability to 3D print aluminum alloys is an area of high interest due to the wide use of aluminum in the aerospace, automotive, and engineering industries. This range of applications requires an equally broad array of material properties that can be achieved through the various series of aluminum alloys. Recently, nanofunctionalization has been shown to be a viable technique for printing crack-free, high-strength aluminum 7075. Furthermore, this technique is fundamentally applicable to a variety of metal alloys and has now been successfully applied to 2xxx, 6xxx, and 7xxx series aluminum alloys. These alloy systems are historically highly crack susceptible due to their solute contents, freezing ranges, and microstructure evolution during welding, and by extension additive manufacturing. This work demonstrates the success of nanofunctionalization in printing crack-free aluminum alloys belonging to different alloy series, with a specific focus on microstructure evolution and mechanical properties. Attention is also paid to the expansion of the design space around aluminum alloy 3D printing made possible by this technology.

Metal matrix composites (MMC) combine conventional materials to provide highly specialized and otherwise unattainable mechanical performance and material properties such as improved stiffness, strength, hardness, and thermal conductivity. These benefits have had restricted applications, however, because the traditional methods of MMC production, melt casting or powder metallurgy, are limited by high cost and the difficulty of forming desirable geometries. To overcome these limitations we integrated ceramic particles into a metal laser powder bed fusion (L-PBF) additive manufacturing process to enable production of highly complex geometries including topology optimized structures and lattice configurations. In this work, ceramic particles including SiC and TiB2 were introduced into aluminum and steel matrix materials to study the effects on processability in L-PBF and resulting material properties. In addition, reactive techniques were also explored to form ceramic particles in-situ. Ceramic particles were found to enhance manufacturability during the L-PBF process by altering laser energy absorption and dispersion into the metal powder bed. Relationships between process parameters, microstructure and mechanical performance were elucidated using a combination of X-ray computed tomography, microscopy and hardness testing on samples processed under varying energy conditions. By leveraging the improved manufacturability of metal matrices due to ceramic addition, processing was tailored to result in a greater than 50% reduction in defects and reduced processing energy requirements by up to 56% allowing faster build speeds compared to the conventional metal AM processing. MMCs in complex geometries were demonstrated, and a method for producing metal-to-ceramic gradient materials was developed as another approach to maximize the benefits of both metal and ceramic phases.

The variation in thermal conductivity of 316L stainless steel samples produced with selective laser melting with varying process parameters is investigated in the bulk and in the microscale. A critical scan rate was observed, while holding all other process parameters constant, above which the porosity started to rapidly increase. For the lowest-porosity sample, a local thermal conductivity map was produced using frequency-domain thermoreflectance. The local stainless steel thermal conductivity varied between 10.4 and 19.8 W/m-K. The average thermal conductivity of the thermal conductivity map agrees within measurement uncertainty with flash diffusivity measurements. The reduction in thermal conductivity with increasing scan rate is not fully explained by the porosity. The average measured values are less than conventionally produced bulk 316L due to the unique processing conditions of laser powder bed fusion, which modifies the crystallographic texture and microstructure. Similar trends are expected for all metal laser printed alloys. Complementary SEM and TEM characterizations will be presented.

With the objective of combining advantages of both additive and subtractive manufacturing, several manufacturers are developing hybrid machine tools. Directed energy deposition (DED) method was the additive manufacturing method adopted in these hybrid machines because of the adaptability of the equipment to the milling platform. In order to take advantage of a hybrid platform and obtain a work piece with a good final quality, it is necessary to understand the influence of the additive process parameters. In this work, laser power and scan speed were systematically varied to fabricate 316L test samples to determine the optimal parameters in a recent developed hybrid machining center (Romi DCM 620 5x Hybrid). Optical stereoscopy indicates that the optimal energy input is 50 J/mm². Lower energy inputs are not sufficient to melt the powder and higher energy inputs than 50 J/mm² (with a shield gas flow of 5 L/min) result in porous samples due to an oxidation between deposited layers. Chemical composition was evaluated by Scanning Electron Microscopy coupled to an Energy Dispersive X-ray detector (SEM-EDS) along the height of the samples. Even though an oxidation was observed, EDS results do not reveal any significant Cr and Ni content variation from one measured point to another. Findings of this study will be used as a basis to develop the process with other alloys in hybrid machine-tools.

Microtextured and mesoporous metal alloys, due to its high surface area, have been widely used as fast-charging anodes for Li-ion batteries [1], as metal electrodes for electrolytic methane production [2], and for electrocatalysis applications [3]. In recent years, there has been a growing interest in designing architected and hierarchical metal electrodes by integrating wet-chemical methods [4] with 3D printing techniques in order to design and scale topologically complex catalyst systems [5] with increased surface area [6]. In this study, we examine a specific corrosion-based wet chemistry to induce the microtexturization of gas atomized metal alloy powder as a strategy to introduce surface roughness at microscales to AM feedstock intended for use in selective laser sintering. Microstructurization of NiCu powders was done by electroless chemical etching in 1M HNO3 with the formation of roughened surfaces with RMS at length scales from 20 nm to 5 μm. Additionally, post-processing of the etched microtextured powders in diluted HNO3 reduced the oxygen content from 4 % to 1 % by mass compared to the initial content present in the gas atomized metal powder. Thereby, development of the new controllable chemistry for modification and microtexturing of oxygen-free metal powders may be a promising way for fabrication of the new high surface area metal feedstock for 3D printing of high-surface area electrodes.
References
[1]
J. Li, J. Pu, Z. Liu, J. Wang, W. Wu, H. Zhang and H. Ma, "Porous-nickel-scaffolded Tin-Antimony anodes with enhances electrochemical properties for Li/Na-ion batteries," Applied Materials and Interfaces, vol. 9, pp. 25250-25256, 2017.
[2]
Y. Shen, M. Ge and A. Lua, "Deactivation of bimetallic nickel-copper alloy catalysts in thermocatalytic decomposition of methane," Catalysis Science & Technology, vol. 8, no. 15, pp. 3853-3862, 2018.
[3]
V. Vij, S. Sultan, A. Harzandi, A. Meena, J. Tiwari, W. Lee, T. Yoon and K. Kim, "Nickel-based electrocatalysts for energy-related applications: oxygen reduction, oxygen evolution, and hydrogen evolution reactions," ACS Catalysis, vol. 7, no. 10, pp. 7196-7225, 2017.
[4]
E. Seker, M. Reed and M. Begley, "Nanoporous gold: fabrication, characterization, and applications," Materials, vol. 2, pp. 2188-2215 , 2009.
[5]
C. Zhu, Z. Qi, V. Beck, M. Luneau, J. Lattimer, W. Chen, M. Worsley, J. Ye, E. Duoss, C. Spadaccini, C. Friend and J. Biener, "Toward digitally controlled catalyst achitectures: hierarchial nanoporous gold via 3D printing," Science Advances, vol. 4, no. 8, 2018.
[6]
T. Song, M. Yan and M. Qian, "The enabling role of dealloying in the creation of specific hierarchial porous metal structures - a reivew," Corrosion Science, vol. 134, no. 15, pp. 78-98, 2018.

Powder bed fusion systems such as direct metal laser sintering (DMLS) provide unprecedented abilities to manufacture complex 3-D parts and structures; however, the process produces unused metal powder that can undergo significant change in chemistry and morphology. The ability to recycle the used powder that is typically found within the build volume and in the overflow compartments requires a thorough chemical and structural analysis in order to determine the extent of reusability. In this context, characterization of virgin powder and used powder for nickel-based and stainless steel based alloys was performed in this work in a rigorous fashion to compare and contrast properties such as composition, particle size distribution and morphology. This information was then used in conjunction with other process parameters (e.g. dose factor, spreader velocity, layer thickness, and laser energy density) to determine conditions that can enable the ready integration of used powder along with virgin powder. Experimental analysis was supported by ANSYS based process modeling in the optimization process, providing a path forward for cost effective additive manufacturing methods.

The integration of nanofibers into the lattice of metals has been a long-standing goal, mainly due to the gained superior properties. Although intensive research has been undertaken to incorporate carbon based materials into metals, such as carbon fibers of carbon nanotubes (CNTs), the results have shown that these materials tend to dissolve and form undesired carbides, which are detrimental to the composite.
A viable alternative is presented in the form of boron nitride nanotubes (BNNTs), a structural analog to CNTs. Thus formed BNNT/metal-matrix-composite (BNNT/MMC) material is expected to be light weight with an improved elastic modulus, strength, toughness and thermal conductivity, while offering additional benefits in the form of a high oxidation temperature, corrosion-free and radiation shielding capabilities.
We will present the results of quantum chemical calculations, of the binding strength and geometry of a variety of metals with carbon and boron nitride nanotubes with special focus on the integration of BNNTs into an Al, Ti and Cu metal matrix.

We have demonstrated a reactive additive manufacturing process that results in the formation of single-crystal semiconducting GaN. Printed GaN regions have been fabricated through the presented method with a single GaN(002) peak present in 2theta-omega XRD scans, indicating a single orientation crystal film. X-ray diffraction and energy dispersive spectroscopy have been used to confirm full conversion of reactive precursors to form GaN with no remaining precursors present in the crystal structure. Through this method, single-crystal GaN films up to 1µm in thickness have been fabricated with crystal growth rates that are orders of magnitude higher than standard semiconductor synthesis methods. We will outline the semiconductor printing method as well as the challenges associated with this process including print resolution, crystal orientation and impurity control.

The application of additive manufacturing (AM) techniques to problems in fusion power provides an exciting opportunity to advance both technologies. Small plasma-facing components such as divertors and armor tiles for tokamak power plants are required to withstand extremely high heat and neutron fluxes, and must therefore be manufactured from difficult-to-machine materials such as tungsten. Additive manufacturing allows fabrication of components in new geometries out of new materials, with a high degree of control over component microstructure. Microstructure is particularly important for small plasma-facing tungsten components, where intergranular fracture is a common failure mode and grain boundary placement must therefore be carefully considered. To be discussed is AM of tungsten components for fusion applications via electron beam melting (EBM). EBM offers a high degree of control over heat input, including the possibility of site-specific microstructure control by means of a point melt scan strategy. Spot melting strategies were developed through computational modelling and validated by experiments. The geometry-specific texture and microstructure of the resulting parts are evaluated using optical and electron microscopy and electron backscatter diffraction to develop a process-structure-properties relationship for the scan pattern.

4D printing can be defined as Additive Manufacturing (AM) of complex structures that can change shape on exposure to a stimulus. By leveraging the unique processing capabilities of AM in combination with shape memory capabilities of alloys like NiTi, there is potential to enable new structures and geometries with adaptive capabilities for space, undersea, special operations and biomedical applications. This work has been focused on solving the fundamental research challenges associated with designing and additively manufacturing functional, stimulus-responsive components from NiTi shape memory alloy. Initially, a processing parameter sweep was used to fabricate single tracks, which were subsequently analyzed for composition and microstructure. A new analytical model relating hatch spacing to energy parameters was developed and used in conjunctions with the detailed compositional and microstructural analyses to identify ideal processing parameters for achieving desired stoichiometry in 3-dimensioanl parts. Build plate composition and geometry were also found to be critical to NiTi stoichiometry and microstructure in built parts due to reactivity with the build plate or adhesion challenges. Several build plate options and processing solutions were identified to ensure optimal material formation. Finally, thermally-activated shape change was demonstrated in a complex part. Future work will focus on additional refinement of processing to achieve tailorable actuation in a complex geometry.

Metal additive manufacturing and in particular selective laser melting (SLM) is a very promising production method. SLM alloy AlSi10Mg, as a common lightweight structural alloy, has already drawn the attention of the aerospace industry due to its good strength to weight ratio. However, when aiming at reaching the requirements of highly demanding structural applications, the fatigue life of SLM AlSi10Mg does not reach the performances met by wrought aluminium alloys. The low fatigue resistance is attributed to residual stresses, the surface roughness and remaining large porosities. Improvements can be carried out in many steps of the processing chain such as actual SLM parameter optimisation and/or post-treatment.
The optimisation of process parameters has already been extensively studied for bulk AlSi10Mg and is provided by the machine manufacturers. Now, post-treatments may sometimes be unavoidable to improve the fatigue life. Hot isostatic pressing (HIP) yields very satisfactory results for Ti based AM alloys and is routinely applied in this industry, but the typical 500°C/100MPa/2h cycle used for AlSi10Mg cast parts increases ductility, but also ruins the SLM fine microstructure, depleting mechanical strength to an inadmissibly low level. In this sense, friction stir processing (FSP), a post-treatment method derived from friction stir welding that has been proven useful in Al cast material, was implemented. This post-treatment could be applied locally in regions of stress concentration more prone to fatigue failure. A significant improvement in fatigue life was demonstrated on polished bulk samples. This is attributed to the positive impact of FSP allowing to eliminate large porosities. These porosities were studied by 3D synchrotron X-ray tomography (voxel size of 0.75 µm) also involving ex-situ fatigue crack propagation. When excessive roughness is the cause of fatigue failure, other post-treatments may be envisioned. Vibro-finishing was found to be an efficient post-processing method for fatigue life improvement.
Contrarily to the bulk structures mentioned above, for thin structures (typically a couple of 100^th of µm) the contouring strategy plays a dominant role on the roughness and sub-surface porosities. The surface roughness of a large variety of strategies was compared and their origin discussed. The sub-surface porosities were studied by laboratory X-ray tomography. The post-treatments mentioned previously may cause excessive deformation of very thin samples. Thus, a contour strategy optimisation is a first step towards their fatigue resistance improvement.

The printability of CoCrFeMnNi high-entropy alloy in laser powder bed fusion is evaluated in this study. A single-track of the alloy was examined to study the solidification microstructure in rapid cooling. Subsequently, multi-layer builds were fabricated with different scanning strategies to study the consolidation and how the microstructure evolves in the subsequently repeated deposition. The present study shows that CoCrFeMnNi has excellent printability with high consolidation and excellent strength. Both monotonic tension and cyclic loading were carried out to study the plasticity and fatigue performance of the alloy additively manufactured with varying scanning strategies.

Additive Manufacturing (AM) literature has shown that the final properties of AM fabricated components are influenced by the complex interactions between geometry, process parameters and alloy chemistry. These interactions can be traced back to spatial and temporal variations of thermal and mechanical signatures (i.e., T {x, y, z, time}). These variations affect all phenomena starting from defect formation, solidification grain structure, solid state transformation and accumulated plastic deformations. Although the above changes are similar to welding, we do observe large reversals of these signatures in AM because of the spatial and temporal transients brought about by the rapid changes in energy delivery modes. The evolution of residual stress distribution due to the non-uniform accumulation of plastic strains within the build is a common concern among all different methods of AM of metallic parts. This plastic strain evolution during repeated thermal cycling is brought about by macro scale changes in the thermal gradients, changes in thermal expansion co-efficient and crystallographic changes between phases and constitutive stress-strain properties. The fundamental response of metals and alloys that undergo solid state phase transformations while being subjected to alternating temperature (above and below the instability temperature) and stress (elastic and plastic) have not been measured before and forms the basis for the current research. The model alloy chosen to understand this complex behavior is Ti – 6Al – 4V. Thermomechanical processing of α/β titanium alloys are largely based on the β -> α + β transformation on cooling. It is well known in titanium literature that α plates usually maintain a specific orientation relationship with the β matrix, referred to as the Burgers orientation relationship which is (101)_β || (0001)_α and [-111]_β || [-2110]_α. There are 12 crystallographically equivalent orientation variants of the α phase within a single β grain. Each of the variants has a different transformation strain, different degrees of elastic self-accommodation when in contact with each other and different interfacial energies when in contact with grain boundaries. Phase field simulations have shown that this leads to variant selection (i.e., some variants appear more frequently than the other) accompanying α precipitation which then leads to the formation of a transformation texture or microtexture. Since the α phase is highly anisotropic in its physical and mechanical properties, this evolving microtexture will determine to a large extent the response of the alloy towards thermo-mechanical cycling. Thus, it would be detrimental if only a single or few variants of alpha percolate the whole matrix. For our research, three different scan strategies (a line scan and 2 types of spot scan) were incorporated to build 15 x 15 x 25 mm cubes, in order to get a variation of thermal gradients across the build. Initial characterization of the as fabricated Ti – 6Al – 4V cube of the line scan sample, using back scattered electrons with SEM, has shown that different α morphology has precipitated along either side of the prior β grain boundary. On one side we see a basket-weave type α while on the other side we see a fine lath type α. Further EBSD analysis has shown that the texture of α is different at the top of the cube and the bottom of the cube. Our work here illustrates the response of this α/β Ti-alloy to thermo-mechanical cycling with proper registration of boundary conditions to simulate typical thermal cycling conditions during AM of metals. Based on the literature, we hypothesize that the response will be dictated by redistribution of global stress/strain in between the rapidly changing phase fractions (i.e., HCP - α and BCC - β). A semi-analytical heat conduction model was used to predict the transient heat transfer behavior during powder bed metal AM. Using this model we can register proper boundary conditions.

Parts made of the titanium alloy Ti-6Al-4V make up for about 50% of the titanium market. The material is today widely used for lightweight applications, e.g. in the aerospace industry. Additive Manufacturing can significantly contribute to lightweight design due to its high design freedom and only very few manufacturing restrictions. Therefore, lightweight Ti-6Al-4V parts represent some of the most promising industrial business cases for Additive Manufacturing these days, such as the brake caliper of Bugatti Chiron and brackets holding the flight crew rest compartment in the Airbus A350 produced by laser beam melting. Part integration, where larger assemblies of parts are integrated to one additively manufactured part, has shown to save additional costs in industrial applications. This leads to the desire to print ever larger parts through Additive Manufacturing technologies such as laser beam melting.
Large scale parts, with dimensions exceeding 0.3 m in at least one orientation, however feature increasing problems from residual stresses during the process. As a result, a high amount of support structures is needed to handle the stresses, which in turn causes high amount of post processing. Depending on the exact geometry, a first-time right production cannot be guaranteed or the parts are simply not manufacturable. Taking into account the comparatively low productivity of today’s laser beam melting machines it is also less economical to produce large scale parts.
In this context, a new approach for laser beam melting is proposed. Instead of optimising process parameters with regard to the highest possible density, the process parameters are optimised to yield the highest possible speed and the lowest possible residual stresses while maintaining a density above 95%. It can be shown that parts fulfilling these criteria may be fully densified through hot-isostatic pressing in a following step. The presentation will look at the differences in residual stresses in test specimens as well as in a large scale part, on the resulting microstructures, as well as on the economics of the process, which profits from a >60% speed gain and thus reduced machine time.

Additive manufacturing (AM) enables design and production of structural metallic components with complex geometries. Recent work has shown that, in addition to complex geometries, site-specific microstructures can be achieved through careful control of processing condition at every layer. These advances are made by leveraging the similarity of physics between welding and AM processes. The interactions between boundary conditions imposed by component geometry, wide variations of thermal signatures brought about by mode of energy delivery, and composition of the alloys can be used for qualification of AM components. This overview presentation will discuss pragmatic integration of recent innovations related to modeling, measurements and manufacturing methodologies with case studies from titanium and nickel alloys.
The first part of the paper will focus on the modeling tools for topology optimization of component geometry, process modeling of heat and mass transfer, solidification and solid-state transformations, plastic deformations, residual stress, distortion and performance of the components. The process flow involved in linking all these discrete modeling tools, challenges and future directions will be outlined with published examples from both welding and additive manufacturing. The second part of the paper will focus on the in-situ characterization of process and also correlation of the same to evolution of defects and microstructure in titanium and nickel alloys. The third part of the paper will focus on case studies involving both computational modeling and in-situ characterization to arrive at strategies for qualification of the AM components.
Finally, this paper will also focus on some of the unresolved questions related to stability of interfaces between liquid and solid, solid and solid, as well as, deformation conditions within the solid during additive manufacturing and approaches to address the same.

Material property variance in powder bed fusion processes is heavily influenced by both pore like defects and microstructure, complicating extension of traditional microstructural models to optimization and qualification of AM materials. In addition while traditional A/B basis testing approaches capture variability in an ideal process, they do not allow prediction of properties for local variations in laser power, melt pool size, or powder layer thickness. As in-situ monitoring becomes more wide spread the ability to predict the impacts of these common deviations will be vital to process certification. In this work we demonstrate how simple nonlinear regression techniques including neural networks and random forests can be used to quickly model the influence of lack of fusion defects on mechanical properties of Ti64. These models can then be leveraged to predict experimental regions of interest to minimize the number of high cost experiments needed for more in depth analysis. 128 vertical Ti64 ASTM E8 tensile bars were produced using different laser processing parameters on a Renishaw AM400. Four parameters were varied using a Latin hypercube sampling approach: laser power, hatch distance, point distance, and exposure time. Test outputs were modulus, ductile vs. brittle fracture, yield strength, and ultimate tensile strength. Given a set of processing conditions, predictive models were developed that could successfully predict ductile vs. brittle fracture at 84% accuracy, and yield strength within 90%. Five samples selected to span the entire mechanical output range were further analyzed using X-Ray computed tomography to evaluate the degree of lack of fusion cracking and explain underlying microstructural basis for varying mechanical behavior. Going forward, these models can be used both to estimate process variability and selectively tailor AM parts for desired mechanical performance.

A CoCrFeNiTiMo-based high entropy alloy (HEA) powder was used to successfully print near-net-shape hardware and test-blocks for examination of selective laser melting (SLM) materials properties. Additive manufacturing (AM) has enabled the manufacture of this difficult-to-process material. By simultaneously exploiting the fast cooling rate of the AM process with the uniformity of compositionally controlled powder, it is possible to reduce the chemical segregation seen in cast-wrought alloys. An anticipated benefit of reduced chemical segregation is improved corrosion resistance.
All specimens received a post-processing heat treatment which precipitated the strengthening phase gamma prime (γ’) and the intermetallic phases eta (η acicular morphology and sigma (σ globular morphology). The resultant microstructures and mechanical properties were evaluated as a function of build orientation. Results so far suggest the build orientation and location on the build plate have little effect on properties and microstructure.
A bimodal grain size distribution exists in the non-HIP and heat-treated condition (C1). Mechanical properties for C1 at room temperature (20C) yield the following: impact energy measured by instrumented Charpy testing is 39.7-40.9 J, yield strength is 890-892 MPa, and the UTS is 1420 – 1459 MPa. A K_Q was measured using J-R fracture toughness testing with no build orientation effects observed. Additional work is ongoing to evaluate the aqueous corrosion resistance of the SLM specimens compared to industry standard cast and wrought alloys.

Additively manufactured metallic materials are known to exhibit high levels of micro-structural inhomogeneity, defects and residual stress. These undesirable features originate from the non-equilibrium manufacturing process involving very high spatial and temporal gradients in temperature. As a result, these materials have defects spanning multiple length-scales and suffers from drastic loss in ductility and toughness. The conventional approach to mitigate this issue is to perform thermal treatment, such as annealing or hot isostatic pressing.

In this study, we explore an alternate route that is non-thermal in nature. The selected material was Ti-6Al-4V alloy manufacture using laser-based, directed energy- deposition system. Thin section (50-150 microns) specimens were prepared to be used in an experimental setup that allows us to study the synergy of electrical current and tensile strain instead of temperature. Our approach involves passing electrical current in the specimens at much lower current density than the electro-migration damage limit. This directly causes Joule heating, which is eliminated by both passive and active heat removal techniques to keep the specimen temperature at or near ambient conditions. We then study the role of mechanical strain on the grain growth at room temperature. The experiments were performed inside a SEM with Electron back scattered diffraction (EBSD) for in-situ characterization. For electrical tuning below 150 °C, the basket-weave a lath structure showed increase in average grain sizes from 144 to 193 microns. The average width of the a laths increased from 0.46 to 0.68 microns. Inverse pole figure maps. Show that the microtextures are random in nature. Our study show decrease in residual strain through EBSD measurement of local orientation spread or Kernel Average Misorientation (KAM). The higher the residual strain, the higher the KAM value, and vice versa for low residual strain. In the as-build condition, about 64% of the kernels had misorientation less than 2°, whereas ~12% had misorientation greater than 4°. Upon electrical tuning below 150 °C, the fraction of kernels with misorientation less than 2° increased to ~70% and 99%, respectively; whereas the kernels with misorientation greater than 4° dropped to ~9% and 0%, respectively. This changes are with respect to the as-deposited specimen materials, which is a clear evidence of the reduction/elimination of residual strain via electric field annealing.

This study describes a multi-channel in-situ monitoring system developed to better understand defect formation signatures in metal additive manufacturing. Three high-speed imaging modes coupled with an image computer capable of processing and storing these data streams allowed an examination of defect formations signatures and mechanisms. It was found that defects later detected in x-ray computed tomography (CT) scans were related to regions with anomalous heat signatures and powder bed morphology. Automated defect detection algorithms based on these defect signatures captured 80% of defects greater than 300µm.

Advanced alloy materials are frequently utilized in applications comprising extreme conditions such as high temperatures or corrosive materials. Nickel chromium based alloys are one such group which is commonly used in industries including jet engine manufacturing, thermal and nuclear power generation, and petroleum engineering, and thus it is crucial to learn how reliably they operate at various scales and in different conditions. There is broad interest in further understanding the mechanical behaviors and reliability of these advanced alloys in settings which more closely mimic those in which they are applied, particularly pertaining to how they are affected at the nano and microscales. This work focuses on the experimental investigation of one such alloy, Inconel 718, which has been fabricated by means of additive manufacturing, and the dependence of its mechanical properties on temperatures ranging from – 200 °C up to 800 °C. Advanced nanoindentation methods are used to characterize mechanical and tribological properties, including tribological responses, creep behaviors, and the variation of cracking thresholds for the determination of fracture toughness. Previous work has shown that mechanical properties in such alloys can be altered by variation of their environment, and it is reported that for large changes in temperature there are noticeable effects in the basic mechanical properties such as hardness and elastic modulus, which was seen to decrease by nearly 30 % at increased temperature, as well as more vital qualities like fracturing behavior. It was concluded that the fracture toughness of additively manufactured Inconel 718 at room temperature is approximately 70 MPa-m^1/2, and the cracking threshold load is variable and is dependent on the material temperature.