Development of Diffraction Research Methodologies for Mediloy S-CO Alloy Speciments Made Using LPBF Additive Manufacturing

Although the idea of incremental component manufacturing techniques is more than 40 years old, techniques of this type have yet to be widely implemented in mass production, still being hindered by challenges that make it difficult to obtain parts with acceptable strength properties. 3D printing processes make it possible to obtain parts with complex shapes or gradient-varying properties (known as functionally graded materials, FGM – see Bhavar et al., 2017), which in the long run can improve products across various industries; however, when it comes to metal 3D printing, the most problematic issues are significant surface roughness, the presence of pores inside the printed parts and the presence of surface tensile stresses. Certain established procedures already exist by which it is possible to eliminate residual stresses with undesirable tensile values (appropriate post-printing thermal treatment, use of a multiple laser pass procedure, heating of the print platform before the printing process). Processes are also being developed to significantly change the stresses, i.e. to introduce compressive stresses in place of tensile stresses using methods such as grinding, drag finishing, shot-peening or laser shock peening (LSP) (Pathak et al., 2023; Behjat et al., 2023). Hot isostatic pressing (HIP), in turn, makes it possible to get rid of excess, unwanted porosity.

This paper examines the properties of cobalt–chromium (CoCr) alloy printed parts obtained using the selective laser melting (SLM) method. Castings made of this alloy are usually characterized by grains of considerable size, a rather heterogeneous structure and defects within the material, resulting from the solidification process. As a result, the resulting parts suffer from inadequate mechanical properties and uneven quality. The advantage of incremental manufacturing methods lies in the ability to optimize the printing parameters in such a way as to avoid such problems. They key parameters in the printing process primarily include scanning speed, laser power, and standing strategies. In the case of alloys from the CoCr family, the optimization of these parameters proves to be particularly important because it greatly affects the microstructure and properties of the final product.

In general, cobalt–chromium alloys contain two main phases: γ-fcc (face centered cubic) and ε-hcp (hexagonal close packed) phases. The γ-fcc phase is thermodynamically stable at high temperatures (above 900°C), while the ε-hcp phase is stable at room temperature. Under certain conditions, the two phases can undergo a mutual transformation. The mechanical properties of these alloys are significantly influenced by the relative amount of γ-fcc and ε-hcp phases, as well as their grain sizes. For example, the volume fraction of the ε-hcp phase strongly influences work hardening and yield strength, but has little effect on tensile strength. On the other hand, an increase in the fraction of the ε-hcp phase results in an increase in hardness and wear resistance, but has a degrading effect on the ductility of alloys (Wei, Koizumi, Chiba, et al., 2018).

The influence of heat treatment on the microstructure and mechanical properties of Co-Cr-Mo alloy obtained by conventional methods was investigated by Lee et al. (2005). Four pre-heat treatments were performed at 1170, 1200, 1230 and 1260°C for three different time conditions: 2, 6, 15h. After every heat treatment the specimens were water-cooled. Tensile tests and X-ray diffraction analyses were used to examine the microstructure and mechanical properties. The volume fraction of the γ phase was found to increase with higher heat treatment temperature. These findings indicate that the suppression of γ → ε transformation takes place with increasing heat treatment temperature. Lee et al. (2005) also noted a slight decrease in tensile strength and an increase in ductility as both the temperature and duration of heat treatment were raised. It was observed that heat treating the alloy at 1260°C for more than 6 hours completely dissolved the σ phase.

Conversely, the study by Gaytan et al. (2010) focused on as-fabricated and fabricated and annealed femoral prototypes and reticulated mesh components, all produced with additive manufacturing, using electron beam melting (EBM). The CoCr alloy selected for this study was Co-26Cr-6Mo-0.2C. Optical metallography, scanning electron microscopy, transmission electron microscopy, energy-dispersive X-ray spectrometry, X-ray diffraction, tensile and hardness tests were involved. In this study, the fact of transformation from the hexagonal ε phase to the face-centred γ phase was confirmed also for additively manufactured elements. It was pointed out that this transformation occurs at a temperature of 430°C; with the addition of Cr this temperature increases to 970°C. The γ → ε transformation can also occur as an effect of plastic deformation or quenching from the temperature range of the stable γ phase. Quenching from the melt state results in the formation of a primarily γ phase while rapid solidification leads to the formation of a mixed phase. The EBM fabrication of Co-26Cr-6Mo-0.2C alloy powder, in the initial state characterized by a hexagonal structure (hcp) of solid elements, results in a matrix with a fcc structure (CoCr with a CrMn component), with Cr₂₃C₆ fcc orthogonal carbide arrays when viewed perpendicular to the build direction and carbide columns connected to these arrays when viewed in a plane parallel to the build direction. These findings pave the way for future development of a set of EBM printing parameters to achieve controlled microstructural architectures. Moreover, the mechanical properties slightly increased with respect to wrought or cast Co-26Cr-0.6Mo alloys.

The study by Sun et al. (2014) examined the impact of EBM printing direction on the microstructural and high-temperature strength properties of the CoCrMo alloy. The occurrence of crystallographic texture was observed with printing tilt from the axis of the samples by 0°, 45°,55° and 90°. The preferred crystalline orientations of the γ phase in specimens produced by the EBM method with angles of 0°, 45°, 55° and 90° were close to [0 0 1], [1 1 0], [1 1 1] and [1 0 0], respectively. The regular occurrence of carbide precipitates at 3 μm intervals along the building direction was observed. Notably, transformation to the hcp phase after heat treatment at 800° for 24 h was also detected. Although no significant texture was observed after aging, lamellar nitride colonies appeared. Furthermore, the study found that printing in the 55° direction relative to the specimen axis provided the highest ultimate tensile strength (806 MPa at 700 degrees). It was concluded that determining the optimal heat treatment conditions for homogenizing the microstructure over the entire height of EBM-built objects should consider the thermal history in the post-melting period of the EBM process, especially when the solid-solid transition is slow.

In the study by Barucca et al. (2015), a Co-Cr-Mo alloy produced by direct metal laser sintering (DMLS) underwent hardness tests and structural analysis. Echoing previous findings Xin et al. (2013), Mantrala et al. (2014), this study showed that the hardness value of this alloy is significantly higher than commonly obtained for the same cast or forged alloys. Seeking to explain this using X-ray diffraction (XRD), electron microscopy (SEM and TEM) and energy dispersive microanalysis (EDX), the study revealed a homogeneous microstructure consisting of an intricate network of thin ε(hcp)-lamellae distributed within the γ-phase (fcc). The authors conclude that the ε-phase lamellae that appeared on the {111} γ planes limit the dislocations movement inside the γ-phase (fcc), causing the measured increase in hardness.

The study presented in Qian et al. (2015), in turn, investigated the microstructure of CoCrMo specimens prepared with selective laser melting. The goal was to understand the influence of the structural characteristics of the alloy at different scales on the mechanical properties of the printed elements. The investigation found that, despite the presence of defects, the yield strength and ultimate tensile strength (UTS) were 870 and 430 MPa and 1300 MPa and 1160 MPa, respectively, for two types of specimens (called: ‘core’ and ‘skin’), which are higher than for the cast dental alloy. Two different sets of processing parameters developed for the internal part (‘core’) and surface (‘skin’) structures of dental protheses were tested. For ‘core’ specimens the applied laser power was 190 W, scanning speed 535 mm/s, hatched line distance was 0.14 mm and layer thickness was 40 μm, while for ‘skin’ specimens these parameters were respectively: 190W, 800 mm/s, 0.1 mm and 20 μm. The microstructures were characterized by SEM, EBSD and XRD analyses. Elemental distributions were assessed by EDS line profile analysis under TEM. Mechanical properties of the samples were measured. Through testing, the formation of grains consisting of columnar sub-grains with Mo enrichment at the sub-grain boundaries was observed. Clusters of columnar grains grew coherently along one common crystallographic direction to form much larger monocrystalline grains that intersected in different directions to form an overall dendritic microstructure. Occasionally, three types of microstructural defects were observed: small voids (<10 μm), small cracks at grain boundaries (<10 μm) and cracks at weld line boundaries (>10 μm). Using X-ray diffraction (XRD) analysis at room temperature the authors showed that the surface of the specimens consisted of the metastable, high-temperature cubic γ (fcc) phase. This seemed consistent as the outer surfaces are rapidly cooled. After grinding and polishing of the surface, additional weak peaks of the low-temperature hexagonal ε phase was detected. The partial phase transformation from γ to ε phase might occur in the bulk interior due to slower cooling. An additional annealing of the specimen at 950°C for 6 min yielded a significant increase in the γ phase at the expense of ε phase; proving the solid-state transformation of the fcc structure at medium temperature.

Lu et al. (2015) focused on the microstructure, hardness, mechanical properties, electrochemical behavior and metal release of CoCrW alloy specimens (without nickel), printed with two types of printing strategy: line-scan and island-scan. XRD revealed that γ-phase and ε-phase coexisted in the as-SLM CoCrW alloys. Optical and scanning electron microscopy observations showed that the microstructure of the CoCrW alloys appeared as a square pattern with fine dendrite cells at the boundaries. Tensile testing suggested that the difference in mechanical properties of the line-formed and island-formed samples is very small.

The effects of heat treatment on components obtained using SLM incremental methods on microstructure, mechanical properties and metal release were investigated in Lu et al. (2016) using X-ray diffraction microscopy, tensile test and ICP-AES. XRD indicated that the alloy underwent a transformation from γ-fcc to ε-hcp phase during thermal processing. The ε-hcp phase almost dominated the alloy during processing at 1200°C after water quenching. Heat treatment at 1150°C with furnace cooling contributed to the formation of larger precipitates in the grain boundary regions, while the size and number of precipitates decreased after heating above 1100°C with water quenching. The diamond-like structure was invisible at higher solution temperatures above 1150°C after water quenching. The authors showed that, compared with furnace cooling, the water-hardened alloy exhibited excellent mechanical properties and a low amount of metal release. The mechanical properties of the alloy reached an optimum combination for the alloy heated in 1200°C: UTS = 1113.6 MPa, 0.2%YS = 639.5 MPa, and E% = 20.1%, while showing a lower total amount of metal release.

The study by Mengucci et al. (2016) looked at the effect of post-production treatments (shot peening, stress relieving, firing cycle) on the mechanical and microstructural properties of Co-Cr-Mo-W alloy prints obtained using Direct Metal Laser Sintering (DMLS). Experimental results showed the existence of a complex network of ε-Co lamellae in the γ-Co matrix responsible for the high UTS and hardness values in the sintered state. Heat treatment increased the volume proportion of ε-Co martensite, but slightly changed the average size of the lamellar structure. One of the main conclusions of the study is that heat treatment is able to induce a significant increase in UTS and hardness and a strong reduction in ductility.

Further confirmation of the existence of two phases (γ-fcc and ε-hcp) in the material of Co-Cr-Mo alloy parts using incremental methods can be found in Takashima et al. (2016). The authors concluded that their specimens positioned in the central part of the matrix consisted more of the γ-fcc phase and less of the ε-hcp phase than those in the matrix’s outer part. The likely reason for this is that, although each specimen was produced under the same conditions, the specimens in the center were exposed to a higher temperature than the specimens in the outer part, and there was less heat dissipation as the neighboring specimens were also heated by the electron beam. This difference in thermal history should be taken into account when fabricating multiple objects simultaneously. The authors underline that the ε phase formed by the transformation from the γ phase gives rise to great changes in the mechanical properties, such as increased wear resistance together with a decrease in the ductility; therefore, it is important to control the phase distribution in additive-manufactured alloys.

Other sources (e.g. Wang et al., 2018), address the problem of microstructural changes and corrosion resistance after different treatments (e.g. laser polishing), applying modified printing parameters (e.g. laser power) or object distances. These results of this work indicate that laser polishing can improve the microstructure, surface structure and corrosion resistance of additively manufactured CoCr alloy by effectively controlling the hcp structures and the oxides formed. The experimental results indicated that the laser-polished samples have higher corrosion resistance and are about 30%better than some thermomechanically treated samples reported in previous studies. Furthermore, oxides on the outer layer and structures in the inner layer play an important role in the corrosion resistance of laser-polished samples, and the most important parameters of the laser polishing process are laser power and object distance. This paper concludes that the difficulty of oxide growth is influenced by the directions of the ε phase, and the transformation of the crystalline phase from the γ phase to the ε phase is accompanied by vacancies and planar defects.

Different and temperature dependent crystallographic phases were also revealed by Wei, Koizumi, Chiba, et al. (2018) for Co-Cr-Mo alloy fabricated by electron beam melting. The authors investigated the phase composition of the printed element and found that the constituent phases of the top side (built later) differ from those of the bottom side (built earlier) of the rod. The upper side consists mainly of the γ-fcc phase, while the lower side consists of the ε-hcp phase. Both phases coexist in the middle part of the bar. The average grain size in the area of the element where the γ-fcc phase dominates is approx. 3 times lower than in the area of ε-hcp phase dominance. The local strain density is much higher in the later-printed part of the element and decreases with increasing distance from this part. A large number of precipitates were observed, which consisted of the main M₂₃X₆ phase and minor, η- and π-, phases, with surface proportions ranging from 5.3% (top) to 8.7% (bottom). The hardness of the samples, as well as the surface fraction of the precipitates formed in them, increases progressively from top to bottom of the bar.

The microstructure, phase composition, strength and corrosion properties of the Co-Cr-Mo-W alloy produced by the SLM process, in turn, were investigated in the study by Wei, Zhou, et al. (2018). The presence of a diverse microstructure was demonstrated, which is homogeneous, supersaturated and composed of extremely fine columnar crystals. Higher strength properties were demonstrated compared to cast components. The high strength is attributed to their fine-grained and supersaturated matrix causing consolidation at the grain boundary and strengthening in the supersaturated state. The lower HCP phase content and the absence of intermetallic compounds or eutectics contributed to the high ductility.

In the study by Monroy, Delgado, and Ciurana (2013), image analysis showed the existence of randomly distributed pores with smooth walls resulting from gas entrapment during the SLM incremental manufacturing process. The authors demonstrated the possibility of obtaining printed CoCrMo alloy dense printed parts, i.e. with the lowest porosity of 0.9%. The lowest porosity was obtained with the thickest layer. This had to do with a better packing of the powder particles, and thus a better thermal conductivity between the particles. The average pore size was independent of layer thickness or laser power, with a strong increase when scanning speed was reduced. The authors suggest this increase in pore size, corroborated by other studies, is due to the prolonged exposure of the material to high temperatures. The most important conclusion of the paper is its identification of the optimum printing process parameters for which the lowest porosity was obtained: a scanning speed of 50 mm/s and the highest possible layer thickness, here: 50 μm. Conversely, the pores with the smallest size were obtained at a scanning speed of 83.3 mm/s and a layer thickness of 500 μm.

Seeking understand the mechanism of defect formation, accurate three-dimensional images of defects inside SLM, Zhou et al. (2015) obtained fabricated Co-Cr-Mo samples by synchrotron radiation micro-CT imaging. The porosities remaining in Co-Cr-Mo alloy parts prepared by selective laser melting were presented as a function of laser processing parameters. Accidental monolayer defects arose as gaps between adjacent laser melt paths or discontinuities in the melt paths, due to inherent fluid instability under the influence of various disturbances. The first monolayer defect formed often generates a multilayer defect involving 2-3 consecutive powder layers. By stabilizing the melt pool flow and reducing the surface roughness by adjusting the machining parameters, it is possible to reduce the concentration of defects.

The paper by Xin et al. (2013) investigates the corrosion features as well as the surface properties of a commercially available Co-Cr alloy (Co61.5, Cr26.0, W5.0, Mo6.0, Si, Fe) produced using the SLM method and compares the results of such components with those made using traditional casting methods. Hardness tests, microstructure observations, as well as surface analysis by XPS spectroscopy and an electrochemical corrosion test were carried out. The results showed that the microstructure of the printed samples was more homogeneous and their hardness was higher, averaging 458Hv compared to the cast samples, which had a hardness of 384Hv. There were no significant differences between the surface properties and corrosion properties of the printed and cast samples.

The microstructural properties, wear susceptibility and corrosion properties of laser engineered net shaped (LENS) components have been investigated and presented in Mantrala et al. (2014). The authors of this paper investigated the properties of deposits produced using different values of the manufacturing process parameters. They found that the deposits produced were characterized by a homogeneous microstructure. The coatings produced using high laser power (350 W), low powder feed rate (5 g/min) and high scanning speed (20 mm/s) were characterised by the highest hardness (446Hv) and wear resistance (1.80 mm³/Nm). In contrast, corrosion resistance was observed at a high level for deposits produced using low laser power (200 W), low powder feed rate (5 g/min) and low scanning speed (10 mm/s). Electrochemical and metal release experiments performed in study Lu et al. (2015) indicate that the island-formed specimens made of CoCrW Ni-free alloys show slightly better corrosion resistance than the line-formed ones in PBS and Hanks solutions.

The mechanical properties and fatigue strength of SLM-produced Co-Cr alloy specimens were investigated by Kajima et al. (2016). The specimens were printed in three ways: with the print direction in line with the specimen axis, perpendicular to the specimen axis, and inclined at 45° to the specimen axis. The results were compared with the results for cast specimens. All groups of SLM alloys showed significantly higher 0.2% offset yield strength and tensile strength than the cast samples, which was probably due to the fine-grained microstructure of the SLM print-outs obtained by their rapid solidification. In contrast, SLM-prepared specimens showed significant anisotropy in fatigue strength. The fatigue strength of the specimen for which the print direction was perpendicular to the specimen axis was significantly higher than that of the cast specimens, while it was significantly lower for the other two specimen types. This phenomenon can be caused by many factors such as surface roughness, crystal orientation, residual stresses and melt pool boundaries.

The study by Wei, Koizumi, Takashima et al. (2018) investigated the effect of heat treatment and thus phase composition and microstructure induced fatigue properties of Co-Cr-Mo alloy produced by EBM. It was observed that the strengths in stress-controlled fatigue were enhanced by converting the initial dual γ + ε phase into a dominant ε phase, and were then significantly improved by inversely converting the ε phase into a γ phase with refined grains. The improvement was achieved by manipulating the fatigue deformation behavior and refining the grain structures.

The primary aim of the present study, therefore, is to investigate the influence of additive manufacturing parameters on the microstructural characteristics and residual stress profiles of Mediloy S-Co alloy specimens produced through LPBF technology. Specifically, this study seeks to elucidate the relationship between printing strategies, porosity levels, and their subsequent impact on the mechanical integrity and stress distribution within printed parts. By analyzing these relationships, we seek to contribute to optimizing printing parameters, thereby enhancing the quality, performance, and application potential of CoCr alloy components in various industrial domains.

Our study’s findings support several conclusions regarding the relationship between printing strategies, porosity, and residual stresses in printed specimens.

Firstly, specimens exhibiting the highest porosity, specifically those printed using strategies 2 and 4, displayed the lowest residual stresses. This correlation suggests that the presence of pores within the material facilitates stress relaxation. The pronounced porosity observed in specimens from strategies 2 and 4 can be attributed to the lower energy densities applied (30.2 J/mm³ and 27.8 J/mm³, respectively), and possibly to a thicker single layer, especially noted in the case of strategy 2. However, merely reducing the thickness of a single layer did not significantly lower the porosity of the prints, as demonstrated by the specimens from strategy 4.

Secondly, a surprising outcome was noted for specimens 1 and 5, where the printing direction was aligned with the specimen’s axis, and the energy density values were relatively similar. Specimen 1, printed with strategy 1, exhibited considerably higher compressive stresses (over 600 MPa), whereas specimen 5, printed using strategy 5, showed stresses that were about half as much (around 300 MPa). When printed at a 45° angle relative to the specimen axis or perpendicular to it, the stress differences between specimens from strategies 1 and 5 were marginal. These findings indicate that the choice of printing direction relative to the part’s geometry, among other parameters, plays a crucial role in determining the final properties of the printed parts.

Lastly, regarding strategy 3, although a relatively high power setting was used, the scanning speed was significantly increased compared to strategy 5, and both the single layer thickness and hatch spacing were slightly enlarged. This parameter configuration yielded average values of tensile circumferential stresses, underscoring the complex interplay between printing parameters and the mechanical stresses developed within he printed parts.