Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2024 Mar 21;2(4):818–828. doi: 10.1021/acsaenm.3c00564

Functional Grading Between Soft-Magnetic Fe–Co/Fe–Ni Alloys and the Effect on Magnetic and Microstructural Properties

Jesse Min-Tze Adamczyk †, Erin J Barrick †, Charles J Pearce †, Robert E Delaney †, Nicolas Ury †,‡, Robert Peter Dillon †,‡, Todd C Monson †, Jay D Carroll †, Donald F Susan †, Nichole R Valdez †, Eric Lang §, Khalid Hattar ∥, Ana S Love †, Hyein Choi †, Andrew B Kustas †,*, Samad A Firdosy †,‡
PMCID: PMC11080042  PMID: 38737588

Abstract

graphic file with name em3c00564_0008.jpg

Producing soft magnetic alloys by additive manufacturing has the potential to overcome cracking and brittle fracture issues associated with conventional thermomechanical processing. Fe–Co alloys exhibit high magnetic saturation but low ductility that makes them difficult to process by commercial methods. Ni–Fe alloys have good ductility and high permeability in comparison to Fe–Co, but they suffer from low magnetic saturation. Functional grading between Fe–Co and Ni–Fe alloys through blown powder directed energy deposition can produce soft magnetic materials that combine and enhance properties beyond the strengths of the individual magnetic materials. This work focuses on the microstructure, crystal structure, and magnetic properties of functionally graded Fe49Co49V2/Ni80Fe16Mo4 coupons. The grading between the two materials is found to refine the microstructure, thereby improving the mechanical hardness without the use of a nonmagnetic element. Postbuild thermal treatments are found to recrystallize the microstructure and increase the grain size, leading to improved magnetic properties. Analysis of crystal structures provides an understanding of the solubility limits and phase equilibria between the BCC (Fe–Co) and FCC (Ni–Fe) structures. Success in functional grading of soft magnets may provide a pathway toward improving energy conversion efficiency through strategic combinations of high saturation and high strength materials.

Keywords: additive manufacturing, soft magnetic materials, functional grading, metallurgy, microstructure engineering

Introduction

With the increased electrification of the national energy infrastructure, advancements in soft magnetic materials remains critical for realizing higher energy efficiency electromagnetic devices for industrial and consumer applications, such as motors, transformers, and switches.1 Efforts to improve upon the performance of soft magnetic alloys, such as those from the Fe–Co, Ni–Fe and Fe–Si systems, have focused on two classic approaches: (1) modification of alloy composition and (2) development of advanced material processing routes. Both methods have been broadly utilized toward achieving desired microstructures and combinations of electrical, magnetic and mechanical properties in soft-magnetic alloys, with a particular emphasis on increasing the tensile ductility without sacrificing the magnetic performance to achieve so-called magnetically soft but mechanically tough alloys.2−4 For the alloy composition approach, strategic modifications of the alloy chemistry are employed, for instance through ternary additions (e.g., V, Nb, Mo or rare earth elements) or composition restrictions (e.g., reduce Co or Si content) to increase ductility.2,5,6 However, such modifications often reduce device performance due to higher coercivity, lower saturation magnetization, and lower permeability relative to the unaltered chemistry, thereby increasing energy conversion losses and device size/weight. Furthermore, mechanical performance improvements afforded by this approach are often modest. The latter approach on the development of bulk advanced manufacturing routes continues to gain traction where advances in powder processing, deformation processing, melt spinning, and additive manufacturing offer opportunities to design unique microstructures toward improving mechanical properties while minimizing impacts on the magnetic performance of these alloys, beyond the capabilities of conventional ingot metallurgy methods, i.e., casting followed by classical bulk thermomechanical forming.7 Of these methods, solidification-based additive manufacturing (AM) offers exciting opportunities to impart unique control over mechanical and functional properties of materials, by virtue of the highly nonequilibrium thermal history imposed during part manufacturing. In the context of mechanical properties, recent efforts have shown that AM processes can develop metastable microstructures toward enabling unconventional combinations of strength and ductility in traditionally brittle soft magnetic alloys.8−12 Furthermore, these processes offer unusual control of part geometry and quality, relative to laminated soft-magnetic alloy sheet, that can be utilized to control eddy current losses.13,14

In addition to these examples for monolithic soft magnetic materials, AM processes also offer opportunities for multimaterial integration via functional grading between dissimilar materials, wherein composition variations are leveraged in order to explore and enhance functional or mechanical properties. One method of this approach enables a combinatorial framework for discovery of new material properties, achievable by a pseudocontinuous change in the chemical composition and process parameters across a spatial domain. Probing the local properties within single sets of graded samples therefore removes the need to synthesize a large array of independent/discrete alloys. This approach has been applied previously to magnetic materials. Chaudhary15 used directed energy deposition (DED) to create functionally graded Co100–xFex and Ni100–xFex that showed similar properties to those of traditionally produced bulk samples. Another example by Geng16 showed a DED approach for grading between Fe and Co to assess microstructure, Curie temperature, saturation magnetization and magnetocrystalline constant evolution as a function of composition. In addition, functional grading has also been demonstrated to enable additional flexibility in the design of systems that utilize functional and structural materials. Rotors and stators in electric motors must be made of high saturation soft magnetic materials to achieve high efficiency, but the present usage relies on sheet metal fabrication techniques that are not amenable to spatial composition variation. For high performance applications where weight is critical (drones and electric aircraft), new rotor assemblies may be manufactured where high saturation materials are deposited on top of lightweight, nonmagnetic shaft materials. Andreiev17 used DED to grade between mechanically brittle electrical steel and high strength shaft steel with the aim reducing the weight of rotors to improve electric motor efficiency. Firdosy18 showed the value of utilizing DED to join dissimilar Hiperco and austenitic stainless steel toward maximizing magnetic and mechanical properties response in a single component. While unrelated to magnetic materials, it is worth mentioning that functional grading has shown great success in thermal barrier coatings, ballistic armor, thermoelectric power generation, and many other applications.19−22 Fabrication of functionally graded materials can be performed by a variety of methods, but the present work focuses on AM using a blown powder DED system with multiple powder hoppers to change the composition on the fly.

The flexibility of DED enables probing of the rich Fe–Co–Ni phase space that contains order–disorder and structural transitions that influence the magnetic and mechanical properties.2,23−25 Hiperco-50A (FeCo–2V), known for its high magnetic saturation, is a near-equiatomic alloy of Fe and Co with 2% V added to improve the ductility.2 At temperatures between 730–950 °C, Hiperco-50A forms a disordered BCC (α) structure that exhibits greater ductility, enabling sheet forming methods, however, the magnetic saturation is lower26,27 in comparison to its low temperature counterpart.28,29 Below 730 °C, ordering to the B2 (α′) phase improves the magnetic saturation, but ductility drops significantly.23,30 Hymu-80 is a high magnetic permeability alloy containing 80% nickel, ≈16% iron, and ≈4% molybdenum. As a disordered solid solution, Hymu-80 forms in the FCC structure (γ) above 517 °C. Below 517 °C, Ni80Fe20 (similar composition to Hymu-80) undergoes an ordering transition to the L12 structure, however, quenching from high temperature can easily avoid atomic ordering due do sluggish diffusion.29,31 In Hymu-80 alloys, the ordered structure shows improved permeability in comparison to the high temperature disordered phase,32−34 however, the magnetic properties of Hymu-80 alloys are highly dependent on grain size.35 Simultaneous improvement of both mechanical and functional properties requires the development of new techniques that can independently control the chemistry, microstructure, and crystal structures of Hiperco-50A and Hymu-80.

This work aims to explore the magnetic and mechanical properties of a functionally graded Hiperco-50A/Hymu-80 material. Bulk functionally graded samples were produced using the DED process by adjusting the relative powder flow rates of Hiperco-50A and Hymu-80 with respect to the build layer height. Postdeposition annealing shows how the as-built grain size and morphology evolve as a function of composition. Phase fraction and structural parameters determined by X-ray diffraction provide insights into the phase equilibria and atomic structure of Hiperco-50A and Humu-80 across the graded region. Nominally composition-homogeneous samples extracted from the graded samples show how the soft magnetic properties change as a function of composition. This work is motivated by the pursuit of materials that can improve energy conversion efficiency, thereby reducing emissions and improving energy conversion technology on a global scale.

Experimental Procedure

Hiperco-50A and Hymu-80 powders from Carpenter Technology were used to additively manufacture four ≈12.7 mm diameter by 76 mm tall graded cylinders on a 1018 steel build plate using an RPMI222 laser deposition system. Nominal compositions for the powders are found in Table 1. The gradient region within each cylinder was ≈3 mm tall and designed such that the alloy volume percent was linearly transitioned from Fe–80Ni–5Mo to FeCo–2V in 12 layers, with equal amounts of the pure alloy compositions on each respective end. The build parameters were optimized for density and build quality using similar soft magnetic materials.36−38 A laser power of 650 W with a beam diameter of 1.016 mm was used in the build. Scan speed was 16.9 mm/s while the hatch spacing and layer thickness were 0.635 and 0.254 mm, respectively. Three of the rods were machined into standard round E8 tensile specimens, with the gradient region located nominally in the center of the gauge section. The remaining rod was sectioned into smaller samples for microhardness and magnetic measurements. Specimens that underwent thermal treatment were subjected to a stress relief anneal at 788 °C for 1 h followed by an anneal at 1175 °C for 5 h, both under a dry hydrogen atmosphere. A final 865 °C anneal under vacuum was done for 4 h. The dual-step heat treatment was conducted in this order to account for alloy-specific phase equilibria. Specifically, the final 865 °C anneal was conducted last to minimize retained austenite (nonmagnetic FCC phase) in the Hiperco side that would have formed during the higher 1175 °C exposure, while still enabling a high temperature anneal for the Hymu-80 alloy region. It is likely that a slow cool to room temperature from 1175 °C was sufficient to remove all austenite, but this dual-step heat treatment was performed toward optimizing the saturation magnetization.

Table 1. Nominal Compositions of Hymu-80 and Hiperco-50A Powders Used to Build Functionally Graded Samples.

Hymu-80 wt % Hiperco-50A wt %
Fe balance Fe balance
Ni 80 Co 48.5
Mo 4.20 V 2.00
Si 0.35 Nb 0.01
C 0.02 C 0.001
Open in a new tab

The dogbone tensile specimens were tested at an initial strain rate of 0.025 mm/s using an MTS servohydraulic mechanical test system. One camera (Pointgrey, 90 fps, 4.1 MP, 50 mm lens) was used to capture images for digital image correlation. Image analysis performed by using Vic2D was used to determine the strain within the individual alloy and functionally graded sections.

Samples for microscopy and magnetic measurements were cut by wire electrical discharge machining by using brass wire. Recast from EDM was removed via ambient temperature nitric/deionized water and phosphoric/deionized water etch (20% v/v) solutions. Samples were immersed into nitric solution for 30–60 s, followed by phosphoric solution for 30–60 s; rinsed in tap water, followed by a deionized water rinse; blown dry with house N2, and further dried in laminar fume hood. Rod samples for EBSD were cut along the graded build direction, while samples for magnetic measurement were cut into 1 mm thick 3 mm × 3 mm squares. Metallographic preparation consisted of initial polishing via SiC paper and then with a 1 μm diamond suspension. For electron backscatter diffraction (EBSD) analysis, two vibratory polishing steps were also employed with a 0.3 μm Al2O3 slurry followed by a 0.04 μm SiO2 slurry. Vickers microhardness measurements were taken on a Struers Durascan 70 with a load of 500 g load. Microstructure analysis was performed on a Zeiss Supra 55-VP field emission scanning electron microscope equipped with a symmetry EBSD detector. The SEM working distance was 10 mm and the accelerating voltage was 20 kV for EBSD/general imaging. Electron probe microanalysis was performed using a JEOL Hyperprobe JXA 8530-F, field emission electron probe microanalyzer equipped with five wavelength dispersive spectrometers. An accelerating voltage of 15 keV with a current of 20 nA was used to create a beam diameter of 5 μm. Magnetometry was performed on a 14 T DynaCool PPMS instrument with the VSM option (Quantum Design, San Diego, CA). Samples were measured at 293 K and with a 5 quadrant field sweep starting at zero field to ±5 T (±50,000 Oe) at a field sweep rate of 10 mT/s (100 Oe/s). X-ray diffraction was performed on the graded samples using a Bruker D2 Phaser X-ray diffractometer with a θ – 2θ geometry and a copper-Kα source. Rietveld refinement to obtain the structural parameters was performed using JadePro/PDF4+ software. X-ray fluorescence measurements used to determine the elemental composition were taken at 4 locations per sample using an Oxford Maxxi-6.

Results and Discussion

Grain Structure

A thorough analysis of functionally graded materials is imperative to gaining a comprehensive understanding of their complex behavior and properties. In this regard, Figure 1 provides a visual presentation of EBSD of the as-built and annealed samples, spatial variation of composition (WDS), microhardness, and grain size as a function of sample location. Alignment of the data sets in Figure 1 enables a holistic understanding of the interplay between the AM process, thermal treatments, and compositional variation within the samples. EBSD inverse pole figure (IPF) maps in Figure 1 provide a macroscopic view of the grain structure and size evolution for both pure alloy ends and across the functionally graded region under as-built and annealed conditions. The as-built and annealed samples were built starting with Hymu-80 on the left (build plate) and finish with the Hiperco-50A region on the right (top of AM build). Notably, in both the as-built and annealed samples, a relatively discrete boundary is observed that separates the two distinct microstructures of FCC and BCC, despite imposing a continuous compositional gradation during specimen manufacturing. For the as-built sample, Figure 1a shows a single-phase FCC structure with a predominately coarse columnar grain morphology for the Hymu-rich region with an orientation that is parallel to the build direction, likely the result of epitaxial solidification during layerwise deposition. The columnar microstructure morphology is slightly serrated at the grain boundaries with steps correlated to the approximate layer height (250 μm) used during processing. The scan path of the laser on the DED system likely caused subtle deviations of the local thermal gradient to be slightly offset from the previous layer, resulting in the serrated columnar grain appearance. Large grains form in the Hymu-80 rich section of the sample form due to a lack of microstructure refining events, such as partitioning or structural transitions, during solidification.39 An ordering transition for Hymu-80 occurs at 520 °C; however, this transition is sluggish relative to the layerwise solidification rates of DED, suggesting that the structure is likely only partially ordered at most. EBSD is unable to detect the order/disorder in Hymu-80; a full characterization of the extent and size of atomically ordered phases was beyond the scope of the present study. Furthermore, it is expected that the disorder–order transition would minimally impact the grain structure’s morphology and size. EBSD also reveals evidence of thermally induced residual stress accumulation, evident by the color gradation within individual grains.

Figure 1.

Figure 1

Open in a new tab

(a) EBSD IPF map of the as-printed coupon showing the as-printed microstructure. (b) EBSD IPF of the annealed coupon showing grain growth and transformation to equiaxed grains in the Hymu-80 material. Note that coloring of the IPF maps is aligned with the build direction. (c) Quantitative compositions across the samples showing a gradual change from the pure Hymu-80 composition to the Hiperco-50A composition. Diffusion of Ni in the build direction causes a small concentration of nickel in what is nominally nickel-free Hiperco-50A. (d) Hardness across the as-built and annealed samples showing constant hardness until the rapid microstructural change at the FCC-BCC transition. (e) Grain size as a function of distance across the graded samples.

Composition measurements in Figure 1c complement the EBSD microstructure analysis and are used to quantify the spatial composition evolution from nickel-rich Hymu-80 to equiatomic Fe–Co near the end of the build (right side of Figure 1c). The beginning of the graded region, demarcated in Figure 1a,c by the left most dashed vertical lines, confirms the Ni–Fe-based composition of the single-phase FCC Hymu-80 region. With the continued addition of Hiperco-50A (Fe–Co–2V) in the graded region, the Fe and Co contents increase with a corresponding decrease in the Ni-content, ultimately terminating with a near-binary equiatomic Fe–Co composition. Notably, despite the continual decrease in Ni-content in the graded region with increased additions of Hiperco-50A, the columnar grain morphology, characteristic of the Hymu-80 rich region, was retained throughout a majority of the graded section. These findings were observed in both as-built and annealed conditions, suggesting that a high solubility of Fe and Co in the FCC Hymu-80 lattice structure, in agreement with prior literature.40,41 The single-phase FCC structure is observed until the composition of the alloy in the graded region is approximately 40Co–40Fe–20Ni, see Figure 1c, after which the microstructure switches to refined BCC grains. It should be noted that this grain refinement is likely a result of a martensitic transformation unique to this region of the graded structure, which is highlighted in Figure 2. Measurement of grain sizes in Figure 1e shows how there is an order of magnitude reduction in the average grain size within these graded samples. In the as-built sample, an unstable mixed FCC/BCC region exists within the melt pool marks region, constrained to roughly a single additive layer (<0.5 mm) at the boundary between the large and fine grained sections. Annealing causes a disappearance of this mixed FCC/BCC region indicating lack of 2-phase thermodynamic stability. Just beyond the sharp boundary, the microstructure appears highly refined, with the EBSD in Figure 1a showing little detail at the given magnification. Higher magnification EBSD is shown in Supporting Information Figures S5 and S6 further confirms a refined, equiaxed microstructure with grain sizes between 10 and 50 μm according to the grain size measurements in Figure 1e. This refined microstructure is likely to increase electrical resistivity by charge carrier scattering caused by an increased concentration of grain boundaries and residual stress-induced point defects. Greater resistivity in this region of the sample has the potential to improve magnetic performance through reduction of eddy currents. Beyond the graded section in the as-built sample, the grain structure appears to be equal to an approximate 25 μm size. Repetitive thermal cycling during the deposition process causes rapid FCC to BCC structural transitions with each pass, causing the formation of a refined microstructure.25 Knowledge of such process-microstructure relationships has the potential to be incorporated into the upstream design process, which could be used to change the local mechanical properties within a single part.

Figure 2.

Figure 2

Open in a new tab

BCC region of the graded section showing formation of a martensitic microstructure after thermal treatment. Additional high magnification EBSD images are available in the Supporting Information.

The two-stage thermal treatment, consisting of a higher temperature anneal at 1175 °C for 5 h and a subsequent lower temperature anneal at 865 °C for 4 h, results in significant recrystallization and grain growth of the functionally graded material microstructure (Figure 1b). The columnar grains in the as-built condition of the Hymu-80 region, as well as throughout a majority of the graded region, are recrystallized into lower aspect ratio (not necessarily equiaxed) grains with a more random crystallographic texture, evident by the increase in the number of total grains and respective grain orientations. In contrast, the microstructure of the Hiperco-50A region, which is equiaxed in the as-built state, coarsened during the heat treatment compared, but retained a nearly identical equiaxed grain morphology and similarly uniform crystallographic texture. The extent of grain coarsening on the Hiperco-50A side is significantly less than the Hymu-80, likely due to limited opportunities for grain growth in the Hiperco-50A alloy as a result of the austenite-FCC to ferrite-BCC phase transition at approximately 985 °C. Specifically, this phase boundary was traversed during the first-stage anneal at 1175 °C, promoting the nucleation and growth of a predominately austenitic-FCC microstructure for the Hiperco-50A region.42 Upon cooling to 865 °C for the second heat treatment stage, the ferritic-BCC structure would again nucleate and grow. The combination of BCC-nucleation, from prior austenite-FCC grain boundaries, and the (necessarily) lower temperature of the second heat treatment results in a finer microstructure for the Hiperco-50A. Notably, the microstructure in the graded region remained exceptionally fine due to the formation of martensite, suggesting significant resistance to grain growth.

Closer inspection of the graded section in the annealed condition (Figure 2) shows evidence of a martensitic-type lath microstructure.43 EBSD cannot distinguish between the martensitic and BCC phases, because, the tetragonal distortion is too small to cause meaningful peak splitting in the diffraction patterns. Due to the small distortion, Figure 3 shows only FCC and BCC phases, even though martensite is present near the phase boundary. However, the morphology of the microstructure paired with the high degree of microstructural refinement, suggests that a martensitic phase transition occurs within this section of the graded sample. Absence of martensitic phases in AM processed Fe–Co alloys suggests that addition of Ni, in this case from the Hymu alloy, aids in the formation of martensite.43,44 The return to a martensite-free microstructure in the Hiperco-50A rich layers supports the proposed role of Ni in stabilization of martensite. Novotny43 found that additions of Ni greater than 0.3 wt % decreases the austenite to ferrite transformation temperature and decreases the critical cooling rate required to form martensite, both of which enable martensite to form in this alloy.2,45 The martensitic microstructure persists past the boundaries of the graded section in Figure 1b, into what should be pure Hiperco-50A. As shown in Figure 1c, there is residual Ni located in the pure Hiperco-50A region just beyond the boundary of the graded region. Melting and remelting during the AM process transports the residual Ni along the build direction thereby causing alloy dilution.46 As further successive melt tracks are built upon one another, the concentration of Ni decreases to a level that can no longer be detected. Diffusion of Ni during the postbuild heat treatment process is also likely occurring, but at much slower rates due to solid state, rather than liquid, diffusion.

Figure 3.

Figure 3

Open in a new tab

(a) As-built EBSD phase map showing an unstable 2-phase region that highlights the melt pool region. (b) Postannealing, the melt pool regions have evolved into a coarser structure.

In addition to the observed grain structure evolution, the heat treatment led to other diffusion-driven structural changes. Most notably, microscopic porosity was observed at the graded interface (see Supporting Information Figure S2 porosity), indicating Kirkendall-type diffusion along the build direction.47 Microporosity formation at microstructural interfaces is common in AM materials due to the stochastic nature of the thermal history and solidification conditions that can cause compositional micro/macrosegregation.48 For these particular alloy combinations, Ni atoms from the Hymu-80 alloy are likely diffusing via vacancy substitution across the graded boundary, leading to the accumulation of the microvoids. Furthermore, heat treatment also caused the disappearance of the weld pool marks, which are observed in the as-built condition in Figure 1a, as further evidence of diffusion during annealing. Two likely explanations exist for the disappearance of the weld pool marks: (a) the BCC weld pool marks (Figure 1a) dissolved into the Hymu-80 lattice and (b) annealing promoted diffusion near the weld pool marks, resulting in a physical shift of the Hymu-80/Hiperco-50A microstructure interface.

Crystal Structure Evolution

Crystal structure evolution is characterized as a function of the alloy composition within the graded specimen, complementing grain structure characterization. Crystal structure analysis via X-ray diffraction is presented in Figure 4 and shows behavior that is consistent with known phase equilibria for binary Fe–Co49 and Ni–Fe39 alloys, as well as the Fe–Ni–Co ternary.50 The lower diffraction patterns in Figure 4a are representative of the FCC structure that is typical of Fe–Ni alloys, especially following rapid cooling/solidification. There is the potential that these peaks may also represent an ordered Ni–Fe structure (L12), but low peak intensities make this determination difficult. Future work is recommended using differential scanning calorimetry (DSC), neutron diffraction, or transmission electron microscopy (TEM) to aid in determining the presence and size of the ordered phase in this material. With an increasing concentration of Hiperco-50A, a transition is observed from FCC to the BCC structure that is characteristic of equiatomic Fe–Co alloys. None of the diffraction patterns in Figure 4a show more than one set of diffraction peaks, however, the room temperature Fe–Ni–Co phase diagram50 shows a 2-phase region between the Fe50Co50 and Ni80Fe20 alloys. This is also discussed in the Phase Equilibria Modeling section. There is potential for some texturing or strain as indicated by changes in relative peak heights across the compositions, but this is not expected to influence refinement of the lattice parameters in a meaningful way. A sharp transition in crystal structure can be seen in 3 where there is a nearly discrete boundary between the FCC and BCC structures. Notably, no major shifts or evolution of the diffraction peaks were observed following the two-stage thermal treatment, suggesting that the as-built samples are relatively stable in terms of crystal structure. It is important to note here that while the parent lattice structures, i.e., FCC-Hymu and BCC-Hiperco, were seemingly retained in both as-built and annealed conditions based on the diffraction analysis, the extent of atomic ordering remains unknown, that is, the extent of FeNi3 (L12) and FeCo (B2). In the case of FeCo alloys, the equilibrium B2 structure can be, at least partially, suppressed when manufactured by fusion-based AM methods, such the method utilized in this study due to rapid layerwise cooling/solidification rates.8 Thus, some degree of metastable disordered BCC Hiperco-50A and FCC Hymu-80 lattice may be expected for the as-built conditions in this study. Additional characterization is needed to quantify the full extent of atomic ordering in bulk material.

Figure 4.

Figure 4

Open in a new tab

(a) XRD patterns on coupons sectioned from the functionally graded region of the sample. A change in crystal structure from the FCC to BCC structure is seen in the build direction. No significant difference is found between the annealed and as-built samples. (b) Volume per atom increases linearly as a function of Hiperco-50A fraction, indicating well-behaved substitutional alloying.

Volume per atom plotted as a function of Hiperco-50A composition, shown in (Figure 4b), is also obtained by refinement of the X-ray diffraction patterns. This metric provides a structure-agnostic view of lattice evolution in the graded material without the discontinuity that would be present in a plot of the FCC and BCC lattice parameters. Overall, the measurements show a linear behavior in agreement with Vegard’s law, indicating a stable atomic substitution as a function of composition. For an FCC Ni80Fe20 alloy, the reported volume per atom33 is ≈11.16 Å3; very close to the measured ≈11.20 Å (Figure 4b) even with the presence of additional molybdenum. Similarly, the Hiperco-50A rich side of Figure 4b aligns well with expectations from literature.51 The linearly increasing volume per atom as a function of composition is a result of substitutional alloying between the FCC Fe–Ni and BCC Fe–Co alloys. Ordering in Fe–Ni and Fe–Co alloys causes small changes in lattice parameter on the order of 10–3 Å, which is far below the expected level of inhomogeneity in our functionally graded samples.33,51,52 The slight outlying Hymu-80-rich point for the annealed condition is likely the result of local heterogeneity within the sample; this outlier is within a range that does not significantly deviate from the overall linear trend. This consistent and predictable structural evolution for the graded specimens aids in decoupling the effects of structure and composition on magnetic property measurements, which is discussed in section.

Microhardness

Mechanical properties, as determined by microhardness measurements, are also provided toward elucidating effects of microstructure (as-built vs annealed) and composition. Hardness is expressed as a function of distance across the graded specimen, including the single alloy ends in Figure 1d. As-built Hymu-80 exhibits a hardness close to 200 HV, which is slightly lower than wrought cold drawn/rolled bar/strip Hymu (approximately 230–260 HV).53 Annealing reduces the hardness to ≈160 HV, which is similar to the so-called process-annealed condition at 871 °C for wrought bar/strip (approximately 170 HV).53 While annealing reduces the grain size in Hymu-80 (Figure 1b), a reduction in residual (thermal) stresses and dislocation density is the likely cause for the thermal softening. Furthermore, Hall–Petch strengthening effects were likely not captured from the microhardness indentation experiments, because, indentations were an order of magnitude smaller than the grain sizes in the Hymu-80 section. Throughout a majority of the graded section, measured hardness is similar to the Hymu-80, with essentially no discernible change despite the near monotonic increase of Fe and Co constituents in the Hymu FCC lattice. This observation suggests that solid solution strengthening in the ternary Fe–Ni–Co system is, perhaps, a negligible effect. Remarkably, a rapid increase in hardness is observed in the graded region near the Hiperco-50A section, as evident by the spike in hardness in Figure 1c. High hardness of 500–600 HV in the graded region, representing a near 300% increase relative to the Hymu-80 region, is observed for the as-built condition. The sharp increase in hardness is likely driven by extensive microstructure refinement and martensite formation in this region. As the fraction of Hiperco-50A increases, the hardness decreases to approximately 400 HV and plateaus (≈6.5 mm) due to the consistent grain size (Figure 1d). An in-depth analysis of the Hall–Petch relationship within this refined Hiperco microstructure is forgone, as there are prior studies showing how mechanical properties change as the grain size changes from 10 to 100 μm.54,55 The effect of annealing on the specimen hardness was relatively minor, with a negligible change for the Hymu-80 region and the graded section. On the Hiperco-50A side of the sample, annealing shows the largest reduction in the microhardness by ≈100 HV, likely driven by grain growth effects. Overall, the range of hardness and the effect of annealing on Hiperco-50A is consistent with our prior work and literature findings.18,54−56 While micromechanical properties can be explained at a high level by examining the microstructure, an understanding of the atomic structure and phase equilibria will provide a more complete picture of this functionally graded system.

Phase Equilibria Modeling

To inform the experimental microstructure evolution results, phase equilibria modeling of the Hiperco–Hymu system was performed using Thermo-Calc software and the TCFE8 Steels/Fe-alloys database.57Figure 5a shows the calculated pseudobinary phase diagram for the Hiperco-50A and Hymu-80 alloys, expressed in terms of temperature as a function of mass fraction Hiperco-50A. The pseudobinary phase diagram effectively represents all equilibrium alloy compositions predicted to appear in this study, bound by the single Hymu-80 and Hiperco-50A alloys. Superimposed on Figure 5a are isothermal temperatures for the two-stage heat treatment, along with the initial stress relief anneal. Complementing the pseudobinary phase diagram are the corresponding (calculated) relative weight fractions of Ni, Fe, Co, and V constituents, again expressed as a function of Hiperco-50A fraction and shown in Figure 5b. Molybdenum, as a minor constituent in the Hymu-80 material, is left out of the calculated phase equilibria analysis, as the addition of Mo led to carbide formation predictions, which are not observed experimentally. At high temperatures or high nickel concentrations, a single FCC phase field is predicted to form (Figure 5a). At lower temperatures and/or with increasing fractions of the Hiperco-50A alloy, that is, increased concentrations of Fe, Co, and V, a 2-phase BCC + FCC region is expected to form. This 2-phase region spans from a lower bound 40% Hiperco-50A fraction until approximately 90%, after which a single-phase BCC phase field is predicted below the austenite transformation temperature. The solvus lines that bound the 2-phase region generally increase in temperature, while the width of the 2-phase region itself decreases, with increasing Hiperco-50A fraction.

Figure 5.

Figure 5

Open in a new tab

(a) Pseudobinary phase diagram between Hymu-80 and Hiperco-50A. Arrows on the left side indicate the order of the heat treatment steps. At high Ni concentrations and high temperatures, the FCC phase forms. Low temperature and low Ni concentrations cause partitioning into the BCC + FCC 2-phase region. Very high fractions of Fe–Co form the BCC structure. (b) Relative compositions of the alloy in the pseudobinary phase diagram. Hymu-80 has a high Ni concentration while Hiperco is rich in Fe and Co.

The degree to which the experimental microstructure results and computational predictions agree is mixed. The strongest alignment with the experimental results is found in the single-phase FCC and BCC regions of the pseudobinary phase diagram. Indeed, in the Hymu-rich region of the graded specimen, for both as-built and annealed conditions, a single-phase FCC structure is observed, in agreement with the expectations for the higher-Ni-content region of the pseudobinary phase diagram. Similarly, at high Hiperco-50A concentrations, a single BCC phase is predicted by modeling and observed experimentally. The largest discrepancies between the modeling and experimental results is found for compositions that consist of appreciable mixtures of Hymu-80 and Hiperco-50A. Specifically, the single-phase FCC structure is observed for a majority of the graded region experimentally, until approximately 20 wt % Ni, see Figure 1. By comparison, phase equilibria modeling predicts the formation of a two-phase region when the Ni-content is approximately 40 wt %. Furthermore, the 2-phase BCC + FCC region was not observed experimentally in the graded specimen. Reasons for the discrepancies between the experimental measurements and phase equilibria predictions remain speculative but are perhaps partially the result of the unique thermal history imparted by AM, namely, rapid cooling/solidification, which can extend solubility limits and promote microstructural metastability (i.e., martensite formation). Additionally, the two-phase region was not experimentally observed, likely due to the formation of a nonequilibrium martensitic structure, which is not captured in the phase equilibria modeling. It is thus hypothesized that the predicted BCC + FCC region in Figure 5a is actually the martensitic phase region, identified as a BCC crystal structure during EBSD analysis.

The phase equilibria modeling also provides insights into origins of the microstructure size and morphological evolution, especially in the context of the refined martensite containing microstructure that forms in the Hiperco-50A rich section of the graded region of the specimen (see fine grain structures in Figure 1a,b). While the phase diagram in Figure 5a is calculated assuming thermal equilibrium, the presence of martensite in our graded samples indicates that our material is not at thermal equilibrium. For the as-built condition, continuous thermal cycling imposed during AM processing would likely cause repetitive martensite BCC–FCC phase transitions, characterized by nucleation of the FCC structure at high temperatures and subsequent reforming of martensite upon rapid cooling. It is conceivable that this repetitive heating and cooling, in combination with the phase transitions, leads to overall refinement of the microstructure. Indeed, a similar mechanism has been proposed previously for AM Hiperco-50A, as well as for wrought Hiperco following cyclic (postprocessing) thermal treatments, wherein repetitive FCC–BCC transitions in the solid state lead to microstructure refinement.8,9,25,44,58 Mechanisms for the subtle gradation in the as-built and annealed grain size in this graded region are also elucidated by the phase equilibrium modeling predictions. Specifically, the grain size is finest in the graded region just after the switch from FCC to BCC, and continues to increase in size with decreasing Ni fraction. It is hypothesized that this microstructure gradation is the result of thermal cycling through the FCC–BCC martensite transition, approximated in Figure 5 by the two-phase FCC + BCC region. The solvus temperatures for this two-phase region are also predicted to increase monotonically with an increasing Hiperco-50A phase fraction. As such, microstructure coarsening locally, with increasing Hiperco-50A fraction, may perhaps be expected due to opportunities for more time at higher temperatures in the BCC phase field. Compositions with the lower transformation temperature (0.4–0.5 wt % Hiperco-50A) would be expected to cycle into the single phase FCC region more frequently during the thermal cycling and thus would experience new FCC grain formation, thereby refining the grain size.

Tensile Mechanical Properties

Three tensile tests were performed on functionally graded samples to provide a measure of the mechanical properties for the functionally graded samples, including measurements to quantify strain/flow localization through noncontact digital image correlation. Samples had a machined external surface to avoid effects of process-induced defects. Interestingly, all three tensile samples broke in the grips during testing on the Hiperco-50A side, indicating limited workability and high notch sensitivity of the alloy. Figure 6a shows that a functionally graded tensile bar at the moment before failure of the Hiperco-50A material in the grips. Nonetheless, the stress and strain values (≈500 MPa, 5% strain) are within the expected range for nominally tested Hiperco-50A samples.2,23 Spatially dependent strain accumulation can be seen in Figure 6a where the Hymu-80 material has a much greater strain in comparison with the Hiperco-50A material. The Hymu-80 side of the sample has also necked down in comparison to the lower strain sections of the sample. In agreement with the microstructure and microhardness analysis, the lowest strain is identified at the narrow band in the lower half of the sample within the graded region, where the finest microstructure and highest hardness/strength was found. Stress strain curves in Figure 6b show how the strain to failure and elastic moduli differ across the sample. Hiperco-50A exhibits a Lüders band followed by a nearly linear increase in stress common to hard, brittle materials. On the other hand, Hymu-80 and the graded material exhibit a higher degree of strain that further highlights the relative differences in ductility between each region of the sample. In the context of controlling mechanical properties, these results suggest opportunities to tailor gradients in plastic strain accommodation through the local control of alloy composition. In this case, low strain and high strength are observed in the graded region, providing opportunities to increase the mechanical strength in otherwise low strength and low ductility functional alloys.

Figure 6.

Figure 6

Open in a new tab

(a) Strain map of the tensile sample the moment before failure. Dashed lines on the tensile sample indicate the locations of DIC extensometers on the specimen. (b) Stress–strain curves of the three materials, using the DIC extensometers. Because, failure occurred in the Hiperco-50A grip, strains should be interpreted as final strains at failure rather than inherent ductility measurements of the individual materials.

Magnetic Properties

Magnetic properties are characterized, emphasizing the full-field induction, to illustrate the effects of alloy chemistry and composition gradation enabled through AM. Figure 7 shows that the full-field induction for both as-built and annealed conditions plotted as a function of local composition (fraction Hiperco-50A). A near monotonic increase in saturation induction from ≈0.8 to ≈2.2 T is observed with increasing fraction of Hiperco-50A. These lower and upper limits of saturation induction are found to match the literature values in Table 2. Annealing slightly increases the saturation induction of the Hiperco-50A rich samples in comparison to the as-built samples due to ordering of the B2 phase.27 Sluggish diffusion in Fe–Ni alloys prevents ordering that would increase the saturation induction of the annealed Hymu-80 rich samples.33 The orders of magnitude differences in grain sizes within these graded samples indicate that microstructure has a limited impact on the saturation induction. The lack of dependence on thermal treatment and microstructure provides a positive outlook for soft-magnets produced by AM: near net shapes can be produced without the need for thermal treatment, enabling faster turnaround times and increased agility.

Figure 7.

Figure 7

Open in a new tab

Magnetic flux density, B, as a function of composition. Saturation induction increases as the fraction of Hiperco-50A increases.

Table 2. Literature and DED Produced Hiperco-50A and Hymu-80 Showing Comparable Saturation Induction Values.

sample B (T)
Hiperco-50A59 2.2 T (H = 1600 A/m)
DED Hiperco-50A 2.1 T (H = 1,672,000 A/m)
Hymu-8053 0.7 T (H = 16 A/m)
DED Hymu-80 0.8 T (H = 621,400 A/m)
Open in a new tab

Electrical properties are often characterized as complementary measures to the magnetic performance. While beyond the scope of the present study, the use of grading and AM processing is anticipated to have effects on the electrical resistivity. For the pure alloy ends, the as-built condition would likely have higher electrical resistivity compared to wrought versions of the same alloys due to a higher defect density (i.e., dislocation and vacancy content) from thermal residual stresses imposed during processing. As-built microstructures may also increase resistivity due to the presence of a solidification substructure or grain refinement (in the case for Hiperco-50A). These as-built features promote charge carrier scattering. The annealed conditions are expected to have electrical resistivity values closer to those of the wrought material. In contrast, the graded region may exhibit higher resistivity in both as-built and annealed conditions, relative to the pure alloy forms, due to finer grain size and the presence of a dual phase microstructure, both of which would increase scattering. Future work is recommended to characterize the electrical properties of these alloys.

Conclusions

Functional grading has the potential to create new magnetic materials that can overcome the limitations of conventional sheet metal processing. The martensitic transformation in the material, paired with repeated thermal cycling in the AM process, causes the formation of a highly refined microstructure with significantly greater hardness in comparison to the pure alloy endmembers. Typical microstructure refinement techniques, that rely on mechanical deformation or rapid solidification are prone to causing grain coarsening during thermal treatment. Functional grading, on the other hand, is able to rapidly identify regions where nonequilibrium structural behavior can simultaneously benefit soft magnetic and mechanical properties. Functional grading between two soft-magnetic materials maintains a high saturation magnetization unlike other materials that may utilize a nonmagnetic material to refine the microstructure.

Magnetic properties vary as a function of both composition and microstructure, allowing for tunability to a specific application depending on mechanical or magnetic requirements. This paves the way for the discovery of new high-strength functional materials that can be produced in near-net-shape geometries. Further work in this space should leverage the automated nature of the powder DED process to generate process/property maps that will enable the discovery of revolutionary soft magnetic materials.

Acknowledgments

The authors acknowledge Shaun R. Whetten for valuable discussions and consultations regarding this manuscript. A portion of this research was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with NASA and supported by JPL Research and Technology Development funds, under a contract with the National Aeronautics and Space Administration (80NM0018D0004). Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International, Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA-0003525. This paper describes objective technical results and analysis. Any subjective views or opinions that might be expressed in this paper do not necessarily represent the views of the U.S. Department of Energy or the United States Government.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsaenm.3c00564.

  • Additional microstructural data and magnetic properties (ZIP)

  • Functional grading between soft-magnetic Fe-Co/Fe-Ni alloys; effect on magnetic and microstructural properties (PDF)

Author Contributions

Jesse M. Adamczyk: Writing—Original draft, formal analysis, investigation, data curation, visualization, supervision. Erin J. Barrick: Writing—review and editing, formal analysis, investigation, data curation. Charles J. Pearce: Methodology, investigation, formal analysis. Robert E. Delaney: Methodology, investigation, formal analysis. Nicolas Ury: Methodology, investigation, resources. R. Peter Dillon: Methodology, investigation, resources. Todd C. Monson: Writing—review and editing, formal analysis, resources. Jay D. Carroll: Resources, Writing—review and editing. Donald F. Susan: Writing—review and editing. Nichole Valdez: Resources, methodology, investigation. Eric Lang: Investigation, validationKhalid Hattar: Conceptualization, funding acquisition. Ana S. Love: Writing—Review and editing, formal analysis, visualization. Hyein Choi: Writing—Review and editing, formal analysis, visualization. Andrew B. Kustas: Conceptualization, methodology, project administration, funding acquisition. Samad Firdosy: Conceptualization, methodology, investigation, project administration.

The authors declare no competing financial interest.

Supplementary Material

em3c00564_si_001.zip (11.3MB, zip)
em3c00564_si_002.pdf (2.8MB, pdf)

References

  1. Gutfleisch O.; Willard M.; Bruck E.; Chen C.; Sankar S. G.; Liu J. Magnetic materials and devices for the 21st century: stronger, lighter, and more energy efficient. Adv. Mater. 2011, 23, 821–842. 10.1002/adma.201002180. [DOI] [PubMed] [Google Scholar]
  2. Sourmail T. Near equiatomic FeCo alloys: constitution, mechanical and magnetic properties. Prog. Mater. Sci. 2005, 50, 816–880. 10.1016/j.pmatsci.2005.04.001. [DOI] [Google Scholar]
  3. Arnold H.; Elmen G. Permalloy, an alloy of remarkable magnetic properties. J. Franklin Inst. 1923, 195, 621–632. 10.1016/S0016-0032(23)90114-6. [DOI] [Google Scholar]
  4. George E. Magnetically soft but mechanically tough alloys. Nature 2022, 608, 270–271. 10.1038/d41586-022-02103-1. [DOI] [PubMed] [Google Scholar]
  5. Ouyang G.; Chen X.; Liang Y.; Macziewski C.; Cui J. Review of Fe-6.5 wt %Si high silicon steel—a promising soft magnetic material for sub-kHz application. J. Magn. Magn. Mater. 2019, 481, 234–250. 10.1016/j.jmmm.2019.02.089. [DOI] [Google Scholar]
  6. Couderchon G.; Tiers J. F. Some aspects of magnetic properties of Ni-Fe and Co-Fe alloys. J. Magn. Magn. Mater. 1982, 26, 196–214. 10.1016/0304-8853(82)90152-4. [DOI] [Google Scholar]
  7. Kustas A. B.; Susan D. F.; Monson T. Emerging opportunities in manufacturing bulk soft-magnetic alloys for energy applications: a review. JOM 2022, 74, 1306–1328. 10.1007/s11837-021-05019-9. [DOI] [Google Scholar]
  8. Kustas A. B.; Fancher C. M.; Whetten S. R.; Dagel D. J.; Michael J. R.; Susan D. F. Controlling the extent of atomic ordering in intermetallic alloys through additive manufacturing. Addit. Manuf. 2019, 28, 772–780. 10.1016/j.addma.2019.06.020. [DOI] [Google Scholar]
  9. Babuska T. F.; Wilson M. A.; Johnson K. L.; Whetten S. R.; Curry J. F.; Rodelas J. M.; Atkinson C.; Lu P.; Chandross M.; Krick B. A.; Michael J. R.; Argibay N.; Susan D. F.; Kustas A. B. Achieving high strength and ductility in traditionally brittle soft magnetic intermetallics via additive manufacturing. Acta Mater. 2019, 180, 149–157. 10.1016/j.actamat.2019.08.044. [DOI] [Google Scholar]
  10. Li S.; Lau K. B.; Wuu D.; Wei F.; Lin M.; Cheong A.; Wang P.; Tan C. C.; Ramamurty U. 3D printing of ductile equiatomic Fe-Co alloy for soft magnetic applications. Addit. Manuf. 2021, 47, 102291. 10.1016/j.addma.2021.102291. [DOI] [Google Scholar]
  11. Nartu M.; Dasari S.; Sharma A.; Chaudhary V.; Varahabhatla S.; Mantri S.; Ivanov E.; Ramanujan R.; Dahotre N.; Banerjee R. Reducing coercivity by chemical ordering in additively manufactured soft magnetic Fe–Co (Hiperco) alloys. J. Alloys Compd. 2021, 861, 157998. 10.1016/j.jallcom.2020.157998. [DOI] [Google Scholar]
  12. Lemke J.; Simonelli M.; Garibaldi M.; Ashcroft I.; Hague R.; Vedani M.; Wildman R.; Tuck C. Calorimetric study and microstructure analysis of the order-disorder phase transformation in silicon steel built by SLM. J. Alloys Compd. 2017, 722, 293–301. 10.1016/j.jallcom.2017.06.085. [DOI] [Google Scholar]
  13. Plotkowski A.; Pries J.; List F.; Nandwana P.; Stump B.; Carver K.; Dehoff R. Influence of scan pattern and geometry on the microstructure and soft-magnetic performance of additively manufactured Fe-Si. Addit. Manuf. 2019, 29, 100781. 10.1016/j.addma.2019.100781. [DOI] [Google Scholar]
  14. Goodall A. D.; Chechik L.; Mitchell R. L.; Jewell G. W.; Todd I. Cracking of soft magnetic FeSi to reduce eddy current losses in stator cores. Addit. Manuf. 2023, 70, 103555. 10.1016/j.addma.2023.103555. [DOI] [Google Scholar]
  15. Chaudhary V.; Sai Kiran Kumar Yadav N. M.; Mantri S. A.; Dasari S.; Jagetia A.; Ramanujan R.; Banerjee R. Additive manufacturing of functionally graded Co–Fe and Ni–Fe magnetic materials. J. Alloys Compd. 2020, 823, 153817. 10.1016/j.jallcom.2020.153817. [DOI] [Google Scholar]
  16. Geng J.; Nlebedim I. C.; Besser M. F.; Simsek E.; Ott R. T. Bulk combinatorial synthesis and high throughput characterization for rapid assessment of magnetic materials: application of Laser Engineered Net Shaping (LENSTM). JOM 2016, 68, 1972–1977. 10.1007/s11837-016-1918-x. [DOI] [Google Scholar]
  17. Andreiev A.; Hoyer K.-P.; Dula D.; Hengsbach F.; Grydin O.; Frolov Y.; Schaper M. Laser beam melting of functionally graded materials with application-adapted tailoring of magnetic and mechanical performance. Mater. Sci. Eng. A 2021, 822, 141662. 10.1016/j.msea.2021.141662. [DOI] [Google Scholar]
  18. Firdosy S.; Ury N.; Kustas A. B.; Carroll J. D.; Pathare P.; Casias Z.; Tung D.; Susan D.; Bobbitt N. S.; Chandross M.; Borgonia J.; Ravi V. A.; Dillon R. P. Compositionally graded joints between magnetically dissimilar alloys achieved through directed energy deposition. Scr. Mater. 2021, 202, 114005. 10.1016/j.scriptamat.2021.114005. [DOI] [Google Scholar]
  19. Lee W. Y.; Stinton D. P.; Berndt C. C.; Erdogan F.; Lee Y.-D.; Mutasim Z. Concept of functionally graded materials for advanced thermal barrier coating applications. J. Am. Ceram. Soc. 1996, 79, 3003–3012. 10.1111/j.1151-2916.1996.tb08070.x. [DOI] [Google Scholar]
  20. Gupta N.; Basu B.; Prasad V.; Vemuri M. Ballistic studies on TiB 2-Ti functionally graded armor ceramics. Defence Sci. J. 2012, 62, 382–389. 10.14429/dsj.62.2666. [DOI] [Google Scholar]
  21. Cramer C. L.; Wang H.; Ma K. Performance of functionally graded thermoelectric materials and devices: a review. J. Electron. Mater. 2018, 47, 5122–5132. 10.1007/s11664-018-6402-7. [DOI] [Google Scholar]
  22. Müller E.; Drašar Č.; Schilz J.; Kaysser W. Functionally graded materials for sensor and energy applications. Mater. Sci. Eng. A 2003, 362, 17–39. 10.1016/S0921-5093(03)00581-1. [DOI] [Google Scholar]
  23. Kawahara K. Structures and mechanical properties of an FeCo-2V alloy. J. Mater. Sci. 1983, 18, 3427–3436. 10.1007/BF00544171. [DOI] [Google Scholar]
  24. Leech P.; Sykes C. LXIX. The evidence for a superlattice in the nickel-iron alloy Ni3Fe. London, Edinburgh Dublin Phil. Mag. J. Sci. 1939, 27, 742–753. 10.1080/14786443908562274. [DOI] [Google Scholar]
  25. Kustas A. B.; Susan D. F.; Johnson K. L.; Whetten S. R.; Rodriguez M. A.; Dagel D. J.; Michael J. R.; Keicher D. M.; Argibay N. Characterization of the Fe-Co-1.5V soft ferromagnetic alloy processed by Laser Engineered Net Shaping (LENS). Addit. Manuf. 2018, 21, 41–52. 10.1016/j.addma.2018.02.006. [DOI] [Google Scholar]
  26. Clegg D.; Buckley R. The disorder → order transformation in iron–cobalt-based alloys. Met. Sci. J. 1973, 7, 48–54. 10.1179/030634573790445541. [DOI] [Google Scholar]
  27. Nartu M.; Dasari S.; Sharma A.; Chaudhary V.; Varahabhatla S.; Mantri S.; Ivanov E.; Ramanujan R.; Dahotre N.; Banerjee R. Reducing coercivity by chemical ordering in additively manufactured soft magnetic Fe–Co (Hiperco) alloys. J. Alloys Compd. 2021, 861, 157998. 10.1016/j.jallcom.2020.157998. [DOI] [Google Scholar]
  28. George E.; Gubbi A.; Baker I.; Robertson L. Mechanical properties of soft magnetic FeCo alloys. Mater. Sci. Eng. A 2002, 329–331, 325–333. 10.1016/S0921-5093(01)01594-5. [DOI] [Google Scholar]
  29. Davies R.; Stoloff N. Order and domain hardening in Cu3Au type superlattice alloys. Acta Metall. 1963, 11, 1347–1353. 10.1016/0001-6160(63)90029-4. [DOI] [Google Scholar]
  30. Persiano A.; Rawlings R. Effect of niobium additions on the structure and magnetic properties of equiatomic iron-cobalt alloys. J. Mater. Sci. 1991, 26, 4026–4032. 10.1007/BF02402943. [DOI] [Google Scholar]
  31. Bozorth R. The permalloy problem. Rev. Mod. Phys. 1953, 25, 42–48. 10.1103/RevModPhys.25.42. [DOI] [Google Scholar]
  32. Taoka T.; Ohtsuka T. The magnetic properties and their temperature dependence of ferromagnetic alloys with an order-disorder transformation. II Ni3Mn. J. Phys. Soc. Jpn. 1954, 9, 723–729. 10.1143/JPSJ.9.723. [DOI] [Google Scholar]
  33. Wakelin R.; Yates E. A study of the order-disorder transformation in iron-nickel alloys in the region FeNi3. Proc. Phys. Soc. B 1953, 66, 221–240. 10.1088/0370-1301/66/3/310. [DOI] [Google Scholar]
  34. Herman H.Treatise on Materials Science and Technology; Academic Press, Inc.: New York, 1974; Vol. 5; p 300. [Google Scholar]
  35. Chinnasamy C.; Narayanasamy A.; Ponpandian N.; Chattopadhyay K.; Saravanakumar M. Order–disorder studies and magnetic properties of mechanically alloyed nanocrystalline Ni3Fe alloy. Mater. Sci. Eng. A 2001, 304–306, 408–412. 10.1016/S0921-5093(00)01518-5. [DOI] [Google Scholar]
  36. Firdosy S.; Ury N.; Borgonia J. P.; McEnerney B.; Conversano R.; Hofer R.; Hermann A.; Ucar H.; Ravi V. A.; Dillon R. P. Processing–microstructure–property relationships in a laser-deposited Fe-Co-V Alloy. Adv. Eng. Mater. 2022, 24, 2100931. 10.1002/adem.202100931. [DOI] [Google Scholar]
  37. Firdosy S. A.; Ury N. E.; Witt R.; Dillon R. P.; Ravi V. A. Laser-deposited soft magnetic Fe–Ni–Mo alloy: a processing—microstructure study. Metallogr. Microstruct. Anal. 2022, 11, 108–118. 10.1007/s13632-022-00819-0. [DOI] [Google Scholar]
  38. Firdosy S.; Ury N.; Dang K.; Narvaez P.; Witt R.; Berman J.; Cooperrider J.; Dillon R. P.; Ravi V. Magnetic behavior of a laser deposited Fe-Ni-Mo alloy. Adv. Eng. Mater. 2023, 25, 2201310. 10.1002/adem.202201310. [DOI] [Google Scholar]
  39. Swartzendruber L.; Itkin V.; Alcock C. The Fe-Ni (iron-nickel) system. J. Phase Equil. 1991, 12, 288–312. 10.1007/BF02649918. [DOI] [Google Scholar]
  40. Sinyova S.; Dreval L.; Starykh R.; Novoghilova O. Study of the phase equilibria of the Co-Fe-Ni ternary system in the high-temperature range. J. Phase Equilib. Diffus. 2019, 40, 583–587. 10.1007/s11669-019-00747-7. [DOI] [Google Scholar]
  41. Raynor G. V.; Rivlin V. G.. Phase Equilibria in Iron Ternary Alloys—A Critical Assessment of the Experimental Literature; The Institute of Metals: 1 Carlton House Terrace, London SW1Y 5DB, UK, 1988. [Google Scholar]
  42. Nishizawa T.; Ishida K. The Co-Fe (cobalt-iron) system. Bull. Alloy Phase Diagrams 1984, 5, 250–259. 10.1007/BF02868548. [DOI] [Google Scholar]
  43. Novotny P. The effect of nickel content on the ac magnetic properties of 49Fe-49Co-2V alloys. J. Appl. Phys. 1988, 63, 2977–2979. 10.1063/1.340920. [DOI] [Google Scholar]
  44. Babuska T. F.; Johnson K. L.; Verdonik T.; Subia S. R.; Krick B. A.; Susan D. F.; Kustas A. B. An additive manufacturing design approach to achieving high strength and ductility in traditionally brittle alloys via laser powder bed fusion. Addit. Manuf. 2020, 34, 101187. 10.1016/j.addma.2020.101187. [DOI] [Google Scholar]
  45. Ashby J.; Flower H.; Rawlings R. Gamma phase in an Fe-Co-2% V alloy. Met. Sci. 1977, 11, 91–96. 10.1179/msc.1977.11.3.91. [DOI] [Google Scholar]
  46. Unocic R.; DuPont J. Composition control in the direct laser-deposition process. Metall. Mater. Trans. B 2003, 34, 439–445. 10.1007/s11663-003-0070-5. [DOI] [Google Scholar]
  47. Strandlund H.; Larsson H. Prediction of Kirkendall shift and porosity in binary and ternary diffusion couples. Acta Mater. 2004, 52, 4695–4703. 10.1016/j.actamat.2004.06.039. [DOI] [Google Scholar]
  48. Moorehead M.; Nelaturu P.; Elbakhshwan M.; Parkin C.; Zhang C.; Sridharan K.; Thoma D. J.; Couet A. High-throughput ion irradiation of additively manufactured compositionally complex alloys. J. Nucl. Mater. 2021, 547, 152782. 10.1016/j.jnucmat.2021.152782. [DOI] [Google Scholar]
  49. Ustinovshikov Y. I.; Pushkarev B.; Sapegina I. Phase separation and ordering in the Fe-Co system. Inorg. Mater. 2006, 42, 354–359. 10.1134/S0020168506040042. [DOI] [Google Scholar]
  50. Kase T. On the equilibrium diagram of the iron-cobalt-nickel system. 164th Rep. Res. Inst. Iron Steel Other Met. 1927, 16, 491–513. [Google Scholar]
  51. Zhu Q.; Li L.; Masteller M.; Del Corso G. An increase of structural order parameter in Fe–Co–V soft magnetic alloy after thermal aging. Appl. Phys. Lett. 1996, 69, 3917–3919. 10.1063/1.117569. [DOI] [Google Scholar]
  52. Ohnuma I.; Enoki H.; Ikeda O.; Kainuma R.; Ohtani H.; Sundman B.; Ishida K. Phase equilibria in the Fe–Co binary system. Acta Mater. 2002, 50, 379–393. 10.1016/S1359-6454(01)00337-8. [DOI] [Google Scholar]
  53. Hymu 80® . Carpenter Electrification. V 10-22, 2022.
  54. Susan D. F.; Jozaghi T.; Karaman I.; Rodelas J. M. Equal channel angular extrusion for bulk processing of Fe–Co–2V soft magnetic alloys, part I: processing and mechanical properties. J. Mater. Res. 2018, 33, 2168–2175. 10.1557/jmr.2018.142. [DOI] [Google Scholar]
  55. Zhang Y.; Kang R.; Dong Z.; Wang Q.; Zhu X.; Xu R.; Guo D. Microstructural evolution of soft magnetic 49Fe-49Co-2V alloy induced by drilling. Mater. Des. 2020, 189, 108501. 10.1016/j.matdes.2020.108501. [DOI] [Google Scholar]
  56. Hasani S.; Shamanian M.; Shafyei A.; Nezakat M.; Mostaan H.; Szpunar J. A. Effect of recrystallization and phase transitions on the mechanical properties of semihard magnetic FeCo-7.15V alloy during the thermomechanical process. Metall. Mater. Trans. A 2017, 48, 1903–1909. 10.1007/s11661-017-3954-8. [DOI] [Google Scholar]
  57. Andersson J.-O.; Helander T.; Höglund L.; Shi P.; Sundman B. Thermo-Calc DICTRA, computational tools for materials science. Calphad 2002, 26, 273–312. 10.1016/S0364-5916(02)00037-8. [DOI] [Google Scholar]
  58. Robino C. V.; Michael J. R.. Novel ultrafine grain size processing of soft magnetic materials SAND2009-0536, 2009.
  59. HIPERCO® 50A . Carpenter Electrification. V 05-20, 2020.

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

em3c00564_si_001.zip (11.3MB, zip)
em3c00564_si_002.pdf (2.8MB, pdf)

Articles from ACS Applied Engineering Materials are provided here courtesy of American Chemical Society

RESOURCES