Effect of the Temperature on the Magnetic and Energetic Properties of Soft Magnetic Composite Materials

: In recent years, innovative magnetic materials have been introduced in the ﬁeld of electrical machines. In the ambit of soft magnetic materials, laminated steels guarantee good robustness and high magnetic performance but, in some high-frequency applications, can be replaced by Soft Magnetic Composite (SMC) materials. SMC materials allow us to reduce the eddy currents and to design innovative 3D magnetic circuits. In general, SMCs are characterized at room temperature, but as electrical machines operate at high temperature (around 100 ◦ C), an investigation analysis of the temperature effect has been carried out on these materials; in particular, three SMC samples with different binder percentages and process parameters have been considered for magnetic and energetic characterization.

Normally SMCs characterization data are provided at room temperature, but the operating temperature of electrical motors, generally at about 70-100 • C, can reach up to 150 • C, making necessary further activity for the materials characterization.
The work focuses on the investigation of the magnetic and energetic properties of these materials for the typical operating temperatures of the applications in which they should be adopted. Different SMC specimens have been tested and analyzed: the polymeric binder content and compacting pressure level have been considered as primary parameters of the study. Interesting and useful results emerged, with a view towards industrial applications.

Phenomenon Description and Procedure
The role of the thermal energy in helping to overcome the Barkhausen jumps is wellknown in the literature [47,48]. Such an advantage is more pronounced in the first part of the magnetic characteristic, in which Barkhausen jumps play a primary role. Near the saturation zone, the magnetization process is much more reversible than irreversible, and thus the effects of the considered temperature variations are also less pronounced.
This fact influences the hysteresis cycle area and, therefore, its energy, with an immediate impact on the global efficiency value. Still, few attempts tried to quantify its influence in the typical electrical machine temperature range.
A starting point on this topic could eventually lead in the future to an increase in the rated working temperature for an electrical machine equipped with SMCs, to reach better efficiency values.
This phenomenon is less pronounced in the laminated steel, where the Barkhausen jumps are limited.

Specimen Preparation
The SMC samples used in this research study have been produced through the mixing of very pure iron powder and a polymer binder. Such binder is an epoxy resin suitable to operate in high temperature, until 200 • C and more, showing good mechanical strength. The binder content is usually under 1 wt %: in the proposed activity, two percentages in weight have been selected: 0.5 wt %. and 0.2 wt %. [28]. After the mixing process, the SMC powder was compacted through a hydraulic press. Two pressure levels are adopted in work: 400 MPa and 700 MPa [16,49]. The obtained shape is toroidal and is wound with two windings.
The specimen identification has the following coding rules: the E letter represents the epoxy resin, while the next digits correspond to the binder content in weight; the last number is the pressure level. For instance, the SMC material with 0.5% of epoxy resin, compacted at 700 MPa is identified with E 0.5-700.

Experimental Procedure
The electrical machines usually operate around 100 • C. For this reason, the prepared SMC samples have been tested at different temperatures [50]. Mainly, it was necessary to implement a particular system able to warm the samples without dispersing the heat. In this context, a special chamber was prepared with wood walls ( Figure 1); inside the chamber, a thermocouple was positioned next to the SMC sample to control the temperature values. The chamber has been placed above a heater, and the temperature feedback loop involved the aforementioned thermocouple, through the heater's external TC input ( Figure 2). The temperature distribution inside the chamber is reported in Figure 3, for the temperature set at 125 • C.
The magnetic characterization has been performed at the following temperature values: 20 • C, 50 • C, 75 • C, 100 • C, and 125 • C. The measurements (Figure 4) began after a suitable temperature settling time of at least 15 min, in a steady-state condition.
No air flux compensation coil was used, as stated in the standard.

Experimental Results
A detailed experimental activity has been conducted on three different SMC specimens: low binder content and low compaction (E 0.2-500), low binder content and high compaction (E 0.2-700), and high binder content and high compaction (E 0.5-700). In Table 1, the main characteristics of each specimen are reported, together with the mechanical information in terms of Transverse Rupture Strength (TRS) obtained with the three points bending test. The measurement procedure followed the EN 60404-4 standard, except for the diameter ratio (external over internal), which was 1.33 in the present work, against the maximum stated of 1.25.
The magnetic and energetic characterization [51] is carried out with the so called "transformer approach", with two windings wounded around the toroidal magnetic core. The first provides the magnetic field H proportional to its magnetizing current, and the second, realized with thin copper wire wrapped very close to the magnetic core, provides the magnetic flux density B, proportional to the induced voltage.
The test bench realized in own laboratories allows to supply the toroids with compensated waveforms until 2 kHz, providing sinusoidal waveforms with very low harmonic content (Total Harmonic Distortion, THD, lower than 1%). A dedicated data acquisition system prepared with LabView code allows to acquire data calculating the magnetic and energetic properties. The scheme of the acquisition system is reported in Figure 5, while Figure 6 shows an example of the data acquisition and elaboration.
A complete data acquisition can require around 15 min.

BH Curves
The main output of the magnetic characterization is the trace of the magnetic characteristic of the proposed SMC materials as a function of the temperature. The BH traces are reported hereafter, respectively, E 0.2-500 in Figure 7, E 0.2-700 in Figure 8 and E 0.5-700 in Figure 9. The results confirm that the temperature does not affect the anhysteretic magnetic characteristics. The samples with low binder content and compacted at high pressure show the best BH curves (1.33 T at 5000 A/m), but those results are independent of the temperature during the measurement.

Maximum Magnetic Permeability
The measurements of maximum magnetic permeability have been conducted and are shown in Figure 10. The impact of the operating temperature is closely related to the typologies of SMC samples. For E 0.5-700, the temperature effect is negligible, while for SMC materials with lower binder content (E 0.2-500 and E 0.2-700), the maximum magnetic permeability increases with temperature.

Hysteresis Cycles
Further information about the magnetic behavior of the SMC specimens have been obtained by measuring the hysteresis cycles, which are reported in the following Figures 11-13, as a function of the temperature, with the respective reduction of the coercive field H c , indicated in the lower quadrant of each graph. It can be pointed out that the specimen with higher density, and better hysteresis cycle, presents lower percentage reduction of the coercive field H c .

Specific Iron Losses
In the case of specific iron losses, the measurements at 1T and 50 Hz have been deduced and are shown in Figure 14. For all the considered specimens, the iron losses present uniform tendency, lowering with the temperature increase. The thermal energy adds to that of the excitation field, making it easier to overcome the pinning sites and, as a consequence, the Block walls move more freely. The temperature effect is not affected by the SMC typologies; just a slight improvement can be observed for samples with low binder content (E 0.2-500 and E 0.2-700) compared to high binder content. The tests have also been repeated for different supply frequency levels to investigate its possible impact on the losses; the consequent specific iron losses values are reported in Table 2, as a function of the operating temperature and frequency. From these results, it is possible to deduce the percentage reduction of the iron losses as reported in Table 3.  Table 3. Effect of the operating temperature condition on specific iron losses % reduction (comparison between 20 • C and 125 • C).

Specimen Typology Frequency (Hz) 50
400 500 The reduction of iron losses may be attributed to the minor hysteresis component of the losses; on the other hand, the presence of excess losses is not negligible in SMC materials [52][53][54][55][56]. Therefore, the correct evolution of different iron losses components with temperature increase requires a detailed analysis.
The magnetic and energetic properties were also checked at 50 • C after having completed the characterization as a function of the temperature, obtaining the same results previously obtained. Based on our experience of a large variety of materials, stable improvements should be obtained with heat treatments over 200-250 • C.
As the obtained results should be considered of significant interest for what concerns the behavior of the investigated SMC materials with respect to the actual operating temperature, the authors also decided to carry out a similar activity on a laminated steel. A toroidal core was realized with a M400-50A steel (0.5 mm thick) punched in the same size of the SMC specimens (external diameter 40 mm, internal diameter 30 mm, total thickness 5.07 mm), and was tested in the same temperature and frequency conditions. The test data on the above-mentioned steel are reported in Figure 15 and Table 4, providing the losses variation from 20 • C to 125 • C of operating temperature  Table 4. Effect of the operating temperature condition on specific iron losses % reduction (comparison between 20 • C and 125 • C) in a laminated steel.

Conclusions
The effect of the operating temperature on the magnetic and energetic behavior of SMC materials has been experimentally obtained. The results show that the maximum magnetic permeability improves with the temperature increase, and even better results are obtained on the specific iron losses reduction. It is evident that the binder content affects the energetic behavior of SMCs in the expected operating conditions, and the values remain constant at different frequencies. Instead, low pressure levels involve a further slight reduction at high frequencies. Probably, the dependence of the iron losses in the SMC materials at different thermal conditions lies in the hysteresis and excess losses behavior. Under this point of view, a correct separation of iron losses components must be considered mandatory, and this activity will be pursued in the immediate future, together with the mechanical performance behavior in temperature (requiring a specific climatized instrument), taking under consideration the actual electrical machine operating condition. Finally, other typologies of SMC materials will be tested in future activities.

Conflicts of Interest:
The authors declare no conflict of interest.