A Study on the Influence of the Properties of Commercial Soft Magnetic Composite Somaloy Materials on the Compaction Process

Abstract

This study aimed to determine optimal forming conditions by comparing the compaction behavior and microstructural characteristics of two Fe-based Soft Magnetic Composite (SMC) powders, Somaloy 700HR 5P and Somaloy 130i 5P. A full factorial design was employed with powder type, compaction temperature, and punch speed as variables. Finite element modeling (FEM) using experimentally derived properties predicted density and stress distributions in toroidal geometries. 700HR 5P exhibited higher stress under most conditions, while both powders showed similar axial density gradients. Experimental results validated the simulations. SEM analysis revealed that 130i 5P had fewer microvoids and clearer particle boundaries. As revealed by TEM-EDS analyses, after heat treatment, both powders exhibited a tendency for the insulation layers to become more uniform and continuous. The insulation layer of 700HR 5P was relatively thicker but retained some pores, whereas that of 130i 5P was thinner yet exhibited smoother and more continuous coverage. XRD analysis indicated that both powders retained an α-Fe solid solution. These results demonstrate that powder properties, composition, and insulation stability significantly influence compaction and microstructural evolution. This work systematically compares the formability and insulation stability of two commercial Somaloy powders and elucidates process–structure–property relationships through an application-oriented evaluation integrating experimental design, FEM, and microstructural characterization, providing practical insights for optimal process design.

1. Introduction

With the global policy shift toward carbon neutrality, the demand for high-efficiency electrification technologies has been rapidly increasing. Consequently, the proportion of electric drive systems in environmentally friendly mobility is expanding. Among these, electric motors—critical components that directly influence the performance of next-generation mobility—have garnered significant attention due to their essential role in reducing energy consumption and enhancing system efficiency. Therefore, the development of high-performance motor core materials and the optimization of forming processes have become key research challenges [1,2,3].

Recently, Axial Flux Permanent Magnet (AFPM) motors have attracted considerable interest in electric vehicle applications owing to their compact design, high torque density, and suitability for high-speed operation. These characteristics increase the demand for core materials capable of supporting complex geometries while maintaining superior magnetic and mechanical properties under dynamic conditions [4]. The performance of electric motors is largely determined by the magnetic properties of the core materials. In high-speed and high-frequency environments, minimizing core losses while maintaining high saturation flux density and structural stability is essential. Traditionally, non-oriented silicon steel sheets have been widely used due to their favorable workability, mechanical stability, and adequate magnetic properties [5]. However, silicon steels inherently exhibit magnetic anisotropy, which restricts complex flux path designs, and suffer from significant eddy current losses at high frequencies. Moreover, the laminated structure is inefficient for fabricating curved or three-dimensional geometries, and mechanical damage or chip losses during cutting and stacking limit mass production and precision [1,6].

The efficiency and reliability of electric motors depend not only on magnetic properties but also on processing factors such as formability, mechanical strength, and dimensional accuracy. In particular, mechanical stability is crucial in drive environments where high-speed rotation, cyclic loading, and thermal shocks frequently occur [5]. This stability is strongly influenced by both the intrinsic properties of the material and the manufacturing process parameters. Silicon steels are disadvantageous for precision forming, and the stress concentrations and micro-damage induced during cutting and stacking can degrade mechanical reliability. To overcome these limitations, optimization of powder-based Soft Magnetic Composites (SMCs) is essential, as their stress states and microstructures are sensitive to forming conditions. Systematic investigations into the correlations among material properties, processing conditions, and resulting structures are therefore required [5,7,8].

SMCs are magnetic materials composed of high-purity iron particles coated with an insulating layer, characterized by three-dimensional magnetic isotropy and high shape flexibility [9]. Their high electrical resistivity suppresses eddy current losses in high-frequency regions, enabling efficient magnetization and demagnetization. Furthermore, SMCs can be directly formed into complex shapes under high-temperature and high-pressure conditions without additional post-processing, offering advantages in manufacturing simplicity and production efficiency over conventional laminated steels. Unlike silicon steels, which require stacking, SMCs can be shaped in a single forming process, reducing scrap generation and minimizing magnetic performance degradation or stress concentrations. Industrial reports have demonstrated scrap rates of less than 3% [4], highlighting their benefits in material efficiency and process optimization [1,7,8].

Recent studies have reported that the magnetic and mechanical properties of SMCs are highly sensitive to manufacturing parameters [5]. For instance, Zhang et al. demonstrated that particle size distribution in Fe–Si–Al SMCs directly affects compaction density and core losses, and that mixed medium–fine particle sizes effectively improve both properties. Liu et al. examined the influence of compaction pressure on mechanical strength and magnetic performance, reporting that excessive pressures lead to insulation layer breakdown and microvoid formation, resulting in increased losses. Additionally, Wang et al. quantitatively analyzed the effects of insulation layer thickness (10–20 nm) on core losses and thermal stability, experimentally verifying that an optimal thickness range can balance magnetic performance with insulation stability [10,11,12].

Nevertheless, many previous studies have tended to focus on specific variables, and comprehensive analyses that encompass stress distribution during compaction, changes in porosity, and the microstructural behavior of the insulation layer under heat treatment are rare. To enable high-performance SMC-based core design, it is essential to quantitatively assess how processing parameters influence magnetic and mechanical properties and to understand their complex interactions [13,14,15]. To bridge this gap, the present study systematically analyzes the effects of forming conditions on the structural stability and magnetic performance of SMCs using toroidal-shaped specimens [16].

This geometry-based approach enables precise comparison of the sensitivity of magnetic and mechanical properties to experimental variables and provides an objective evaluation framework for the actual process responsiveness of commercial SMC powders. In practice, the process guidelines provided by manufacturers for commercial SMC powders are often based on limited empirical data, lacking systematic information on material responses and microstructural evolution across diverse temperatures, compaction speeds, and pressure conditions [17].

Therefore, this study focuses on two widely used commercial SMC powders from Höganäs—Somaloy 700HR 5P and Somaloy 130i 5P—and quantitatively compares their compaction behavior, internal stress distribution, microstructure formation, and magnetic properties under identical processing conditions using toroidal specimens [18,19]. The objective is to elucidate how each powder responds to processing parameters and how these responses affect resulting structures and performance. Through this analysis, the study establishes fundamental data applicable to the evaluation of formability in commercial SMC powders, the determination of optimal forming conditions, and the development of motor cores under practical processing environments [10,20,21].

The specific objectives of this research are as follows:

• To evaluate the material properties of commercial SMC powders under various thermal and mechanical processing conditions.
• To determine the optimal forming parameters using a full factorial experimental design and to quantitatively analyze compaction behavior through finite element analysis (FEA).
• To fabricate toroidal core specimens under optimal conditions and compare experimental results with simulations.
• To validate the analysis through density measurements and high-resolution microstructural characterization of the fabricated specimens.
• To assess the process suitability of commercial SMC powders and propose a practical process design guide for the development of high-reliability motor cores.

2. Materials and Methods

2.1. Evaluation of Basic Properties of Somaloy Powders for Compaction Analysis

In this study, two commercially available SMC powders, Somaloy 700HR 5P and Somaloy 130i 5P, supplied by Höganäs AB in Skåne, Sweden, were employed. These powders were specifically developed for electric motor core applications, featuring properties that are well-suited for high-density compaction and the reduction in magnetic losses at high frequencies. Both materials consist of iron-based particles individually coated with an insulating layer, which effectively suppresses eddy current losses due to their high electrical resistivity [18,22].

Despite their widespread use in prototype manufacturing and electric motor applications, comprehensive comparative studies that evaluate their formability under practical processing conditions remain limited. Therefore, this study aims to quantitatively compare the compaction response and structural characteristics of these two powders under identical forming conditions, thereby establishing a consistent basis for understanding the process–material interactions of commercial SMC powders.

Somaloy 700HR 5P is designed to exhibit excellent thermal stability, making it suitable for high-temperature compaction processes, whereas Somaloy 130i 5P offers superior formability even at relatively low temperatures. The two powders differ in terms of particle size distribution, specific surface area, and apparent density, all of which are critical factors influencing their compaction behavior and microstructural development under identical processing conditions [22].

The basic material properties of each powder were obtained primarily from the manufacturer’s datasheets, with selected parameters verified through preliminary experiments [18]. Based on these properties, this study combines finite element simulations and experimental investigations to compare the compaction response and structural evolution of the two powders. The objective is to provide fundamental data that can support the optimization of forming processes for SMCs used in electric motor cores.

2.1.1. Evaluation of the Compaction Behavior of Somaloy Powders

The elevated-temperature compression test method has also been employed in previous studies to evaluate the plastic deformation characteristics and processing stability of materials under varying temperatures during forming [8]. To establish a precise correlation between powder compaction simulations and actual processing conditions, it is essential to obtain stress–strain curves through elevated-temperature compression tests prior to the evaluation of toroidal specimens. This process allows the investigation of the temperature-dependent plastic behavior of the material.

To assess the compaction behavior of the Somaloy powders, compression tests were conducted in accordance with ASTM E209 using a Gleeble 3800-GTC thermomechanical simulator (Gleeble, Poestenkill, NY, USA) [23]. To reflect actual forming conditions, the tests were performed at three temperature levels: 25 °C, 100 °C, and 200 °C. According to the Höganäs technical guide, the recommended compaction temperature for both powders is around 100 °C. Therefore, these temperature conditions were selected to analyze the temperature sensitivity of the powders, with 100 °C serving as the reference [24,25]. The strain rate was fixed at 0.1 mm/s.

The results indicated that both powders exhibited the lowest stress at 100 °C, as shown in Figure 1, and thus 100 °C was adopted as the reference temperature for evaluating formability. For both powders, the stress did not exhibit a linear dependence on temperature across the tested range. This non-linear behavior can be explained by the stable activation of the binder at 100 °C, which reduces interparticle friction and enhances formability, whereas at 200 °C, microstructural changes and the onset of thermal instability in the insulating layer contribute to non-linear stress responses.

To further investigate the cause, X-ray fluorescence (XRF) analysis was conducted using an SEA 1200 VX instrument (Seiko Instruments Inc., Chiba, Japan). The results showed that Somaloy 700HR 5P contained Fe 94.58 wt.%, Si 2.74 wt.%, and P 1.58 wt.%, while Somaloy 130i 5P contained Fe 96.51 wt.%, Si 1.38 wt.%, and P 1.18 wt.%. Compared with 130i 5P, 700HR 5P exhibited higher Si content by 1.36 wt.% and higher P content by 0.4 wt.%. Since higher Si content tends to increase brittleness, compaction under identical conditions may require higher temperatures or applied loads [26,27]. Additionally, the P content influences the properties of the binder and the thermal activation behavior of the coating. These compositional differences likely affect the thermal activation of lubricants and binders, the insulating characteristics, and the temperature-dependent compaction response of the powders.

2.1.2. Analysis of Thermal Properties Applicable to Compaction Simulation

The thermal properties of the commercial Somaloy powders supplied by Höganäs were evaluated based on the manufacturer’s technical datasheets and standard testing guidelines. The thermal conductivity was referenced from measurements according to ISO 22007-2 [28], while the coefficient of thermal expansion (CTE) was obtained following ASTM E228 [29], as specified in the Somaloy 5P material datasheet. These thermal property values were incorporated into the boundary conditions of the FEA for simulating compaction behavior and predicting heat transfer. The relevant data are summarized in

Table 1. Thermal conductivity of Somaloy 700HR 5P, Somaloy 130i 5P.

Factor	700HR 5P	130i 5P
Thermal expansion	0.000011	0.000011
Thermal conductivity	26	26

[25,28,29].

According to the datasheet, both powders exhibit identical thermal conductivity under the same compaction pressure, reflecting the intrinsic property of the base material. However, in actual compacts, process variables and manufacturing conditions can alter the uniformity and thickness of the insulating coating. These variations significantly influence the heat flow pathways within SMCs, potentially leading to differences in heat transfer behavior between the two powders [13].

2.1.3. Identification of the Crystalline Phases of the Powders

X-ray diffraction (XRD) analysis is a standard technique for identifying the lattice structure and crystalline phases of materials. In this study, XRD was performed to verify whether the two Somaloy powders maintained stable crystalline structures before and after compaction and heat treatment. The diffraction patterns obtained from the analysis are presented in Figure 2 [30].

The results revealed distinct diffraction peaks near 44.63°, 64.97°, and 82.3°, corresponding to the [1, 1, 0], [2, 0, 0], and [2, 1, 1] crystal planes, respectively. These peaks are characteristic of the α-Fe structure and match the solid solution phase Fe_0.96Si_0.04. This finding indicates that both powders retain a cubic crystal structure and do not undergo structural instability associated with phase transformations during mechanical compaction. Therefore, the two powders can be compared on the same crystallographic basis, allowing the differences in compaction response and microstructural development to be attributed solely to material properties and processing parameters [31,32].

The phosphate-based insulating coating applied to the particle surfaces did not produce any distinct diffraction peaks, suggesting that the coating is either amorphous or below the detection limit of the XRD analysis. Overall, both powders possess a crystallographic foundation that supports isotropic magnetic properties, confirming their suitability as stable magnetic core materials.

2.1.4. Analysis of Particle Morphology and Size Distribution

To quantitatively and qualitatively compare the particle characteristics and surface morphologies of the Somaloy powders, microstructural observations were performed using scanning electron microscopy (SEM, JEOL Ltd., Tokyo, Japan). Particle size distributions were measured using a Mastersizer 3000 laser diffraction particle size analyzer (Malvern Panalytical Ltd., Worcestershire, UK) [32]. The results indicated that the average particle size of Somaloy 700HR 5P was approximately 425 µm, while that of Somaloy 130i 5P was around 150 µm. The representative morphologies of both powders are shown in Figure 3.

Particle size and shape significantly influence powder packing behavior and densification during compaction. In general, larger particles are more prone to pore formation and agglomeration, while excessively fine particles may exhibit poor flowability, negatively affecting achievable compaction density.

Both powders were manufactured using the same water atomization process. Somaloy 700HR 5P predominantly consisted of large, irregularly shaped particles with frequent interparticle bonding and agglomerates observed. In contrast, Somaloy 130i 5P featured smaller particles with higher sphericity and more uniform dispersion, resulting in greater interparticle independence and improved flow characteristics [22,33].

2.2. Experiment Methods

2.2.1. Insulation Properties of the Powder

The Somaloy 700HR 5P and 130i 5P used in this study are coated with an inorganic phosphate-based insulation layer. These coatings are designed to ensure interparticle electrical insulation, suppress eddy current losses, and enhance magnetic performance under high-frequency operating conditions.

According to the material specifications provided by the manufacturer, the electrical resistivity of 700HR 5P is approximately 600 µΩm, while that of 130i 5P is around 17,000 µΩm. These values were measured using the four-point method on toroidal specimens with an outer diameter of 55 mm, inner diameter of 45 mm, and height of 5 mm. As this geometry differs from the specimens used in the present study, the data are cited for reference purposes [25].

Both powders were used in their as-received condition without additional surface coating or post-treatment. Differences in the composition and thickness of the insulation layers are expected to affect the uniformity of stress distribution and magnetic properties during compaction and subsequent testing [34].

2.2.2. Yield Criterion and Model Selection of Porous SMCs

In this study, the Shima–Oyane yield criterion was employed to model the yielding behavior of porous SMC materials, and the findings of previous research were cited [24]. This model is an extension of the conventional von Mises yield criterion, introducing a porosity factor f to account for the stress concentration effects caused by internal pores. In general, the von Mises criterion calculates the equivalent stress based on the differences between principal stresses; however, for porous materials, this approach alone is insufficient to accurately describe the actual yielding behavior.

The yield condition in the Shima–Oyane model is defined as follows:

$${\sigma }_{eq}= {\sigma }_y(1-D \mathit{f})$$

(1)

Here, ${\sigma }_y$ represents the yield strength of the material, D is a material constant, and f denotes the porosity.

Equation (1) reflects the tendency for the yield stress to decrease as porosity increases and serves as a physically based model suitable for the forming analysis of porous metals. Equation (1) was used as the governing equation for the FEM analysis conducted in this study. The reliability of the analysis results was ensured by refining the number of mesh nodes according to the geometry, and for this reason, the analysis model was selected as a toroidal core. The toroidal geometry eliminates edge effects and provides a uniform magnetic flux distribution, making it suitable for the precise evaluation of internal stresses and magnetic performance after compaction [35,36].

2.2.3. Full Factorial Experiment Design

To quantitatively evaluate the effects of key process variables on the sintering characteristics of SMCs, a full factorial experimental design was employed. The full factorial experimental design is a structured methodology that includes all possible combinations of the levels of each factor involved, allowing the outcome of every combination to be observed and analyzed. In the case of SMC materials, it is often difficult to predict the quantitative impact of each variable on the final properties. Particularly in powder compaction and sintering processes—where multiple variables act simultaneously—this method enhances experimental reliability and serves as an effective tool for process optimization and property enhancement [37,38].

In this study, three primary factors were selected: powder type, compaction temperature, and press forming speed. The full factorial experimental design was structured based on all possible level combinations of these factors. The resulting experimental matrix was used to analyze the influence of each factor on key post-compaction and sintered properties, including density, porosity, and mechanical strength. Special emphasis was placed on identifying the interactions between material properties and forming conditions, with the goal of determining optimal processing conditions for achieving high quality and improved performance in SMC-based components.

To quantitatively evaluate the material-related factors affecting the compaction behavior of SMC powders, this study designated powder type as a primary experimental variable based on mechanical and thermal properties [39]. The compaction temperature was selected with reference to application conditions of commercially used powders. In particular, the recommended processing temperatures for 700HR 5P and 130i 5P were considered, and the experiments were conducted at 60 °C, 80 °C, and 100 °C to reflect the applicable range. The forming speed was determined based on the stroke speed range of the single-axis press mechanism, which governs the vertical motion of the upper punch. The control factors and their respective levels are summarized in

Table 2. Design of factor and level.

Factor	Description	Level
		1	2	3
A	Powder Type	700HR 5P	130i 5P	-
B	Forming Speed [mm/s]	2	5	8
C	Forming Temperature [°C]	60	80	100

and

Table 3. Full factorial experiment design.

Simulation No.	A	B	C
1	1	1	1
2	1	2	2
3	1	3	3
4	1	1	2
5	1	2	3
6	1	3	1
7	1	1	3
8	1	2	1
9	1	3	2
10	2	1	1
11	2	2	2
12	2	3	3
13	2	1	2
14	2	2	3
15	2	3	1
16	2	1	3
17	2	2	1
18	2	3	2

.

3. Powder Compaction Simulation

3.1. Simulation Setup

To quantitatively analyze the effects of process variables, such as powder type, compaction temperature, and punch speed, on the density and stress distribution during the compaction of SMC toroidal cores, the commercial FEA software DEFORM-3D (SFTC, version 14.2) was employed. The simulation model was constructed based on the geometry of the toroidal core used in the experiments. The initial shape of the porous green body, designed using CATIA V5, along with the rigidly assumed die and punch, is shown in Figure 4. The flow characteristics of the powder material during compaction were defined based on the stress–density relationship obtained from compression tests on fabricated specimens, as previously described, and these material properties were used as input parameters for the simulation [17,40,41,42,43].

The boundary conditions were set as follows: the lower punch was fixed, while the upper punch applied a load at the specified speed in the axial direction to carry out compaction. A heat transfer condition was applied at the interface between the die and the powder to account for temperature effects during compaction [44]. To compare the compaction pressures of the two powders, the stopping condition for the punch was set to a final forming height of 5 mm.

The initial powder loading mass was set to 8 g, identical to that of the actual toroidal bulk specimens. After filling the die with the powder, the initial powder loading height was set to 11.2 mm for Somaloy 700HR 5P and 10.3 mm for 130i 5P. These values were calculated by back-calculation based on the apparent density provided in the Höganäs material data, assuming the target density of 7.4 g/cm³ as the reference relative density [25]. Prior to compaction, the mesh of the model was generated with a minimum element size of 0.18 mm, resulting in 237,735 elements for 700HR 5P and 208,655 elements for 130i 5P, thereby ensuring high accuracy in evaluating the density and stress distributions during compaction [17].

3.2. Simulation Result

Based on the simulation results, a comparative analysis was conducted to evaluate the compaction characteristics of the two powders. Relative density, effective stress, and mean stress were analyzed for each case, and the simulation outcomes are presented in Figure 5, Figure 6 and Figure 7 and

Table 4. Results for each case according to simulation No.

No.	Average Relative Density	Max-Min Relative Density	Average Effective Stress [MPa]	Max-Min Effective Stress [MPa]	Average Mean Stress [MPa]	Max-Min Mean Stress [MPa]	Compaction Pressure [MPa]
1	0.948	0.241	454.191	294	−733.793	789.3	886.4
2	0.948	0.250	452.507	291	−735.668	889.5	851.8
3	0.948	0.235	469.555	278	−764.448	995.1	921.4
4	0.947	0.238	451.574	268	−724.896	826.5	879.4
5	0.947	0.243	453.001	297	−730.856	873.5	905.5
6	0.947	0.234	462.702	269	−730.856	873.5	891.1
7	0.948	0.240	451.908	290	−730.468	788.0	877.3
8	0.948	0.235	469.555	278	−764.448	995.1	872.2
9	0.947	0.241	466.648	295	−752.274	829.4	877.3
10	0.945	0.175	407.656	216	−649.772	821.7	780.8
11	0.945	0.176	402.984	214	−641.903	745.6	782.3
12	0.945	0.182	428.198	227	−679.811	770.0	826.6
13	0.945	0.178	405.701	217	−644.882	757.8	765.9
14	0.946	0.167	405.239	205	−646.356	784.5	833.5
15	0.945	0.177	419.569	218	−668.699	836.5	785.6
16	0.945	0.174	404.850	209	−643.860	746.7	826.8
17	0.945	0.181	401.339	218	−641.267	764.5	791.2
18	0.946	0.174	426.868	221	−682.785	843.5	786.5

. Somaloy 700HR 5P generally requires a higher compaction load to reach the final core height of 5 mm, which corresponds to its tendency to exhibit higher relative density under the same processing conditions. In fact, across all temperature and forming speed conditions, 700HR 5P consistently showed higher relative density values than 130i 5P.

Analysis of the post-compaction relative density distribution revealed that, under all conditions, the maximum density was located near the top or bottom boundaries, while the minimum density was commonly observed in the lower central region. This pattern reflects the pressure attenuation during downward compaction by the upper punch, where pressure decreases toward the bottom, and limited powder flow leads to density loss in the constrained region.

700HR 5P, characterized by a larger average particle size and agglomerated particle structure, exhibited overall higher peak density values. However, its inhomogeneous particle size distribution hindered particle rearrangement during compaction, leading to localized high-density regions near the bottom. In contrast, 130i 5P, with its finer and better-dispersed particle characteristics, maintained a relative density above 0.78 even in the lower region, demonstrating superior compaction uniformity.

Comparison of the effective stress distributions revealed that, for 700HR 5P, conditions in which the maximum relative density was concentrated near the lower region also exhibited localized stress concentration at the bottom interface. In contrast, 130i 5P—due to its finer and more dispersed particle characteristics—showed a more uniform stress transmission, with the effective stress distributed moderately around the upper central region. This spatial consistency between effective stress and relative density distribution in 130i 5P highlights the improved internal compaction uniformity. Furthermore, the well-formed interparticle insulation enhances the overall electrical resistivity of the specimen, which is consistent with the previously discussed density and stress distribution results [45,46].

130i 5P also exhibited a non-linear response in both effective and mean stress values with respect to compaction temperature. Specifically, under a forming speed of 2 mm/s, the effective stress decreased from 407.656 MPa at 60 °C to 401.339 MPa at 80 °C, and then increased to 419.569 MPa at 100 °C. The mean stress reached its minimum at 80 °C, with a value of −641.267 MPa. This U-shaped trend was also consistently observed at a forming speed of 5 mm/s, indicating that 80 °C represents the optimal compaction temperature for 130i 5P.

In contrast, when the forming speed was increased to 8 mm/s, both the effective and mean stresses exhibited a monotonic increase with rising temperature. However, the mean stress variation reached 784.5 MPa, indicating a reduction in stress uniformity during compaction. At 5 mm/s, the mean stress variation was comparatively lower at 745.6 MPa, and the compaction behavior showed a well-balanced performance across multiple indicators: effective stress of 402.984 MPa, mean stress of −641.903 MPa, density deviation of 0.176, and compaction pressure of 782.3 MPa. Based on these results, the combination of 80 °C and 5 mm/s is considered the optimal condition for achieving both mechanical stability and uniform density distribution during SMC powder compaction.

For the 700HR 5P powder, the effective stress exhibited a gradual increase with rising temperature across all forming speeds. Specifically, the effective stress increased from 454.191 MPa at 60 °C to 455.296 MPa at 80 °C and 462.702 MPa at 100 °C. Under the condition of 100 °C and 2 mm/s, the effective stress variation was relatively low at 269 MPa, which was comparable to the 268 MPa observed at 60 °C and 5 mm/s. This indicates that stable and uniform stress distribution can be maintained at elevated temperatures when forming at a low speed.

However, when the forming speed was increased to 8 mm/s at 100 °C, the effective stress variation rose to 278 MPa, and the mean stress variation reached as high as 995.1 MPa, indicating a significant intensification of stress concentration. These results reflect increased process instability and a higher risk of stress localization at high forming speeds. Based on these observations, the combination of 100 °C and 2 mm/s was selected as the optimal compaction condition for 700HR 5P.

The difference in optimal forming speed between the two powders originates from variations in compaction behavior caused by differences in particle size and microstructural characteristics. 700HR 5P, with an average particle size of approximately 425 µm—three times larger than that of 130i 5P—has an agglomerated particle structure. As the forming speed increases, the limited particle mobility and uneven pressure transmission lead to intensified stress concentration and localized density non-uniformity. In contrast, 130i 5P, with an average particle size of approximately 150 µm and a finely dispersed structure, allows smooth particle rearrangement even at a forming speed of 5 mm/s, thereby maintaining uniform density and stress distributions. These findings indicate that selecting forming conditions tailored to the microstructural characteristics of each powder is crucial for optimizing the process and ensuring the quality and reliability of SMC core manufacturing [47,48].

4. Experimental Validation

4.1. Toroidal Core Fabrication

Toroidal core specimens were fabricated based on the optimal compaction conditions derived from FEM simulations. The fabricated samples are shown in Figure 8 and Figure 9. A hydraulic powder compaction press with a maximum capacity of 50 tons was used for the forming process, and the powder charge was set to 8 g. This corresponds to a compaction load of approximately 8 tons per unit cross-sectional area of the specimen, which was determined to ensure consistency with the compaction force levels obtained from the simulation. The setup was designed to replicate comparable stress conditions during actual compaction [49].

Since the magnetic performance of SMCs is highly sensitive to internal stress and microstructural conditions, heat treatment is a critical step in achieving the desired final properties. In this study, a mixed atmosphere of hydrogen and argon was used during heat treatment to ensure both uniform thermal distribution and surface reduction effects. Hydrogen, with its high thermal conductivity and strong reducing properties, contributes to temperature uniformity and facilitates the reduction in metal oxides, thereby stabilizing the microstructure. Simultaneously, argon was used as the primary component of the atmosphere to suppress the excessive reactivity of hydrogen and to maintain process stability.

Heat treatment was carried out at 700 °C, which was selected as an appropriate temperature for relieving residual stress within the SMC structure and minimizing internal defects to optimize magnetic properties. Excessively high temperatures can cause thermal instability and phase transformations in the insulation layer, leading to the deterioration of magnetic properties. Therefore, the heat treatment temperature and atmosphere were carefully controlled to achieve a balance between magnetic properties and structural stability.

4.1.1. Comparison of Press and Simulation Load–Stroke Behavior

Figure 10 shows the load–stroke curves obtained from simulations conducted under the optimal forming conditions for the two powders. As previously determined, Somaloy 700HR 5P exhibited an optimal forming load of 86,408.3 N and a forming pressure of 877.3 MPa under a punch speed of 2 mm/s and a compaction temperature of 100 °C, while Somaloy 130i 5P recorded a forming load of 77,049 N and a forming pressure of 782.3 MPa at a punch speed of 5 mm/s and a compaction temperature of 80 °C.

The hydraulic press used in this study, unlike a mechanical press, applied a variable load during compaction. At the final stage, when the compact reached its maximum density, a load of approximately 8 tons (≈78,480 N) was applied, confirming that the target forming pressure was achieved. The load–stroke curve obtained from the experiment exhibited the typical behavior of a hydraulic press: a gradual load increase at the initial stage, a rapid rise during the middle stage, and a deceleration in the rate of increase as the density approached saturation. This curve was compared with the simulated load–stroke curves. At the point where maximum density was reached, the punch stroke displacement was 7.3 mm for 700HR 5P and 6.4 mm for 130i 5P.

Due to differences in press control methods and boundary conditions, it was difficult to directly calculate the relative error between the simulation and the experiment; however, the simulation results qualitatively reflected the actual behavior observed during the powder compaction process.

4.1.2. Density Analysis

The outer diameter, inner diameter, height, and mass of the fabricated specimens were measured, and the results are summarized in

Table 5. Linear dimensions of density specimens of Somaloy 700HR 5P.

Linear Dimension	Specimen 1	Specimen 2	Specimen 3
Outer Diameter [mm]	20.39	20.41	20.33
Inner Diameter [mm]	12.73	12.77	12.77
Height [mm]	5.40	5.58	5.52
Volume [mm3]	1075.43	1110.39	1084.32
Weight [g]	7.82	7.84	7.98
Calculated Density [g/cm3]	7.27	7.06	7.36
Measured Density [g/cm3]	7.32	7.24	7.37

and

Table 6. Linear dimensions of density specimens of Somaloy 130i 5P.

Linear Dimension	Specimen 4	Specimen 5	Specimen 6
Outer Diameter [mm]	20.39	20.34	20.39
Inner Diameter [mm]	12.74	12.73	12.72
Height [mm]	5.47	5.42	5.25
Volume [mm3]	1100.56	1070.75	1046.61
Weight [g]	7.67	7.53	7.52
Calculated Density [g/cm3]	6.97	7.03	7.19
Measured Density [g/cm3]	7.21	7.22	7.33

. Based on the measured dimensions, the calculated average density was 7.23 g/cm³ for the Somaloy 700HR 5P specimens and 7.06 g/cm³ for the Somaloy 130i 5P specimens. To improve the accuracy of density measurement, the specimens were coated with paraffin to minimize the influence of closed porosity, and the density was measured using the Archimedes method in accordance with ASTM B962 [50].

The results showed that the calculated average density of the Somaloy 700HR 5P specimens was 7.23 g/cm³, while the average density measured by the Archimedes method was 7.31 g/cm³, corresponding to a deviation of 1.12% between the two methods. For the Somaloy 130i 5P specimens, the calculated average density was 7.06 g/cm³, and the measured average density using the Archimedes method was 7.26 g/cm³, with a larger deviation of 2.76%. These results are summarized in

Table 7. Average calculated density, average measured density by the Archimedes method, and their error rate of Somaloy 700HR 5P.

	Calculated Density [g/cm3]	Measured Density [g/cm3]	Relative Error (%)
Somaloy 700HR 5P	7.23	7.31	1.12

and

Table 8. Average calculated density, average measured density by the Archimedes method, and their error rate of Somaloy 130i 5P.

	Calculated Density [g/cm3]	Measured Density [g/cm3]	Relative Error (%)
Somaloy 130i 5P	7.06	7.26	2.76

.

Due to the influence of press process conditions and die set design, the actual height of some specimens slightly exceeded the target value of 5 mm during compaction. Furthermore, factors such as burr formation and height variation during pressing led to an increase in the final specimen volume, resulting in calculated density values that were lower than the target density under the same mass condition. Nevertheless, the Somaloy 700HR 5P specimens exhibited both higher average and maximum density values compared to Somaloy 130i 5P, indicating that this powder promotes more effective interparticle bonding and forms a denser structure under high compaction pressure.

4.2. Microstructural Properties of Toroidal Core

The correlation between the structural characteristics observed during the forming process was analyzed by comparing the simulation results for the toroidal specimens with the properties of the actual formed samples. To investigate microstructural homogeneity, potential damage to the insulation layer, and structural consistency, fine-scale analyses were additionally conducted using SEM and FIB-TEM.

4.2.1. Bulk Microstructure Analysis of Somaloy

The preprocessed samples for SEM analysis, which underwent mounting and polishing, are shown in Figure 11. Figure 12 presents the cross-sectional view of the toroidal specimens, divided along the height into upper at 4.5 mm, middle at 3.0 mm, and lower at 0.5 mm. Differences in density and the corresponding microstructural variations across these regions were analyzed.

Somaloy 700HR 5P exhibited a relative density of 0.95 at the top and bottom regions, and 0.94 in the middle, indicating a tendency for higher compaction at both ends along the vertical axis. In contrast, Somaloy 130i 5P showed a gradual decrease in density from top to bottom, with values of 0.95 at the top, 0.94 in the middle, and 0.93 at the bottom. This trend is closely related to the nature of the uniaxial compaction process. In a configuration where the load is applied downward from the upper punch, compressive stress tends to concentrate in the upper region and attenuate as it is transmitted downward, resulting in relatively lower compaction density in the lower section.

As shown in the SEM observations in Figure 13, this density distribution is reflected in the corresponding microstructural features. For Somaloy 700HR 5P, a dense and uniform microstructure was observed in both the top and bottom regions, while the middle region exhibited relatively finer pores. This suggests that more effective particle bonding occurred at the ends, whereas insufficient compaction pressure in the middle contributed to a reduction in structural density.

In the case of Somaloy 130i 5P, a uniform and compact microstructure was observed in the upper region; however, the number and size of pores increased progressively toward the middle and bottom regions, accompanied by looser particle bonding. Particularly in the bottom region, a distinct increase in pore distribution was observed, reflecting the microstructural inhomogeneity consistent with the measured density results. These findings experimentally demonstrate that the limitations in load transmission during uniaxial compaction directly affect microstructural formation.

Overall, both powders exhibited a vertical density gradient under uniaxial forming conditions, and this gradient was more pronounced in materials with lower formability or limited flow characteristics. To improve density uniformity in future compacts, alternative forming methods such as bidirectional pressing or floating-die techniques should be considered [51,52].

4.2.2. Bulk Morphology Analysis of Somaloy

To evaluate the thickness and microstructural characteristics of the insulation coating layers of the Somaloy powders, specimens were prepared using FIB and analyzed by TEM and EDS [26,32]. The results for the heat-treated specimens are shown in Figure 14 and Figure 15, while those for the pre-heat-treated specimens are presented in Figure 16 and Figure 17. The average coating layer thicknesses measured from three specimens of Somaloy 700HR 5P were 301.3, 816.4, and 901.1 nm, yielding an inter-specimen average of 673.0 ± 323.0 nm. For Somaloy 130i 5P, the corresponding values were 140.0, 712.7, and 475.9 nm, with an inter-specimen average of 442.9 ± 286.8 nm. Detailed statistical data are summarized in

Table 9. Average thickness and statistical parameters of the insulation layer in Somaloy 700HR 5P specimens (a–c) obtained from TEM images in Figure 14.

700HR 5P	AVG [nm]	SD [nm]	Min [nm]	Max [nm]	Max-Min [nm]	CV [%]
(a)	301.3	96.0	140	340	200	31.9
(b)	816.4	52.2	719	868	149	6.4
(c)	901.1	351.9	408	1490	1082	39.0

and

Table 10. Average thickness and statistical parameters of the insulation layer in Somaloy 130i 5P specimens (a–c) obtained from TEM images in Figure 15.

130i 5P	AVG [nm]	SD [nm]	Min [nm]	Max [nm]	Max-Min [nm]	CV [%]
(a)	140.0	58.9	60	200	140	42.1
(b)	712.7	272.0	239	1060	821	38.2
(c)	475.9	302.6	68	772	704	63.6

. These results indicate that 700HR 5P forms a thicker insulation layer than 130i 5P, while the thickness variability was greater in the latter.

In 700HR 5P, pores were observed within the coating layer, predominantly distributed along the particle–coating interfaces. These pores may serve as initiation sites for stress concentration during compaction. In 130i 5P, substantial thickness variation among specimens was noted, and numerous pores and cracks were identified before heat treatment. However, after heat treatment, the number of pores decreased and the continuity of the coating layer improved [46].

Elemental distribution analysis revealed that P and O were consistently detected throughout the insulation layers of both powders, confirming the phosphate-based nature of the coatings. Bi, a minor additive introduced during manufacturing to enhance insulation properties, exhibited strongly localized signals in 700HR 5P, whereas in 130i 5P it showed weaker and more sporadic distributions [52]. Additionally, trace amounts of S were detected in 700HR 5P. Both Bi and S were mainly localized near pores and interfaces, indicating a close relationship with microstructural non-uniformities in those regions.

After heat treatment, the insulation layers of both powders exhibited improved continuity and uniformity compared with those before heat treatment, with no evidence of compositional loss or decomposition. In 700HR 5P, some pores remained even after heat treatment, which is interpreted as being related to non-uniform coating deposition during the coating process and artifacts introduced during FIB preparation. In 130i 5P, pores and cracks observed before heat treatment were markedly reduced.

Overall, 700HR 5P formed a thicker insulation layer, with an average thickness of 673.0 nm compared with 442.9 nm for 130i 5P, while the thickness variability among specimens was more pronounced in 130i 5P. Meanwhile, the average particle size of 700HR 5P was more than three times larger than that of 130i 5P, making it difficult to ensure uniform coating deposition, which could be associated with the persistence of some pores even after heat treatment.

4.2.3. Crystal Structures of Bulk Specimen

To assess the stability of the crystal structures, XRD analysis was conducted on the bulk Somaloy specimens. The measurements were performed using a SmartLab X-ray diffractometer (Rigaku Corporation, Tokyo, Japan), and the results are presented in Figure 18. In both materials, distinct diffraction peaks were observed near 44.5°, 64.8°, and 82.2°, corresponding to the [1, 1, 0], [2, 0, 0], and [2, 1, 1] planes, respectively. These patterns are characteristic of a cubic crystal structure and matched the peak positions previously identified in the powder state [30].

No evidence of phase transformation or crystallographic reorientation was observed during the forming process, indicating excellent thermal stability and minimal structural damage to the crystal lattice in both materials. Therefore, the forming conditions did not significantly affect the Fe–Si alloy crystal structure, and the crystallographic characteristics observed in the powder state were well maintained in the bulk specimens.

4.2.4. Electromagnetic Characteristics Inferred from Microstructural Differences

The quality and uniformity of the insulating layer play a critical role in suppressing eddy current losses in the high-frequency range. When the insulating layer is uniformly and continuously formed, the electrical pathways between particles are effectively blocked, resulting in reduced eddy current losses and a uniform magnetic flux distribution, which enhances the stability of the magnetic response. Conversely, a non-uniform insulating layer or the presence of internal microvoids can induce localized magnetic flux concentration and thermal imbalance, leading to increased magnetic losses [13,47].

The TEM-EDS analyses conducted in this study revealed that Somaloy 700HR 5P, despite exhibiting a generally thicker insulation layer, contained numerous micropores and showed lower uniformity, indicating structurally unstable characteristics. In contrast, Somaloy 130i 5P formed a thinner insulation layer than 700HR 5P but demonstrated a more continuous coating structure with fewer internal defects, suggesting a higher degree of structural integrity. These microstructural differences act as direct factors influencing the electromagnetic response of the two materials.

Based on the B–H curves and high-frequency core loss values provided by the manufacturer’s material data, Somaloy 700HR 5P exhibited a higher initial magnetic permeability and a higher magnetic flux density under the same field conditions [25]. Across the frequency range of 50–800 Hz, 700HR 5P recorded lower core loss than 130i 5P under all magnetic flux density conditions. At B = 1.5 T, the core losses of 700HR 5P were measured as 6.8, 29.5, 65.1, and 154.7 W/kg at 50, 200, 400, and 800 Hz, respectively, while those of 130i 5P were 8.35, 34.7, 73.1, and 160.7 W/kg under the same conditions, showing that 700HR 5P had approximately 10–18% lower values. Core loss data are summarized in

Table 11. Core loss of Somaloy 700HR 5P as a function of magnetic flux density at various frequencies.

	50 Hz	200 Hz	400 Hz	800 Hz
0.5 T	0.99	4.2	9.09	20.9
1.0 T	3.34	14.36	31.44	73.76
1.5 T	6.8	29.49	65.1	154.68

and

Table 12. Core loss of Somaloy 130i 5P as a function of magnetic flux density at various frequencies.

	50 Hz	200 Hz	400 Hz	800 Hz
0.5 T	1.22	5.02	10.44	22.48
1.0 T	4.1	17	35.61	77.65
1.5 T	8.35	34.74	73.1	160.65

. The TEM analysis further indicated that 700HR 5P formed a relatively thick insulation layer of about 0.67 µm on average, which enhances electrical insulation and contributes to reduced core loss. However, some micropores persisted in the insulation layer even after heat treatment, which could induce local electrical/magnetic non-uniformities and stress concentration under long-term cyclic loading conditions. On the other hand, although 130i 5P formed a thinner insulation layer averaging approximately 0.44 µm, its continuous and defect-scarce structure suggests that, despite higher initial core loss values, it may be advantageous for long-term suppression of eddy current losses and stable magnetic response.

In summary, while 700HR 5P exhibits superior initial performance indicators, it requires careful quality control of the insulating layer. In contrast, 130i 5P demonstrates higher structural stability, allowing for more reliable operational performance. This study primarily focused on deriving optimal forming conditions for the powders, evaluating process dependence on formability, mechanical strength, and dimensional accuracy using toroidal specimens. However, for process optimization and performance prediction in complex geometries such as AFPM motor cores, it is essential to fully assess the electromagnetic characteristics of the core materials to ensure reliability. Future work will build upon the microstructural data obtained in this study to conduct a detailed analysis of magnetic properties, enabling more comprehensive process optimization and performance prediction.

5. Conclusions

This study systematically compared and analyzed the compaction behavior and microstructural characteristics of two commercial SMC powders, Somaloy 700HR 5P and Somaloy 130i 5P, using toroidal specimens. A full factorial experimental design combined with FEM simulations was employed to comprehensively evaluate stress and density distributions as well as microstructural changes under various forming conditions, leading to the identification of the optimal compaction parameters for each material.

The toroidal specimens served as an effective model to validate the intrinsic formability of the powders and to obtain data on microstructure and stress distribution prior to extending the study to AFPM motor core geometries. This approach enabled a quantitative characterization of the inherent behavior of the materials without the complexity of design variables, thereby providing a solid basis for precise process optimization before actual motor stator design.

Somaloy 130i 5P demonstrated uniform stress transmission and homogeneous density distribution in the simulations, with the optimal compaction behavior achieved at 80 °C and 5 mm/s, where stress fluctuations and density variations were minimized. These results were supported by SEM and TEM-EDS analyses, which revealed finely dispersed particles and a thin, continuous coating layer. Somaloy 700HR 5P exhibited stable stress distribution and high average density under conditions of 100 °C and 2 mm/s in the FEM analysis. The corresponding microstructural features observed through SEM and TEM-EDS, including larger particle size, thicker coating layers, and localized stress concentration, aligned well with the simulation results. The macroscopic compaction behavior predicted by the simulations was consistent with the microstructural observations, confirming the interdependence between particle size, coating thickness, and internal porosity.

The toroidal specimens fabricated under the optimal conditions achieved average densities of 7.31 g/cm³ for 700HR 5P and 7.26 g/cm³ for 130i 5P. The deviations from the theoretical densities predicted by FEM were 1.13% and 1.57%, respectively, demonstrating high consistency between simulations and experiments and confirming the reliability of the FEM model. Minor discrepancies observed under certain conditions were attributed to limitations in particle rearrangement and deformation of the insulating layer.

SEM and TEM-EDS analyses further confirmed that differences in particle size, coating thickness, and pore distribution were consistent with the stress and density distributions predicted by simulations. Somaloy 130i 5P was characterized by a uniform microstructure and stress distribution, whereas Somaloy 700HR 5P exhibited higher density and localized stress concentration. These findings indicate that the two powders possess distinct characteristics that can meet different motor core design requirements.

Both powders exhibited density variations within the core, with 130i 5P showing a tendency for the density to decrease from the upper to the lower regions along the core height. This vertical density gradient arises from the limitations of uniaxial compaction and can potentially lead to degradation of the magnetic performance. To improve density uniformity, future work should apply bidirectional compaction, where simultaneous pressure from both upper and lower punches reduces load transmission loss and ensures more uniform density distribution in the compact [53].

Moreover, the floating-die technique, which allows the lower die to move to alleviate stress concentration and enhance powder flow, has been reported as an effective method to mitigate density gradients [54]. This forming technique can be further extended to the processing of complex AFPM motor geometries in subsequent research.