The material proposed in this paper is Manganese (IV) oxide (MnO₂) powder activated, ∼85% (Merck Life Science United Kingdom Limited), with particle size <10 μm (as analysed by the supplier). Similarly, this study made use of Silicone oil supplied by Merck Life Science United Kingdom Limited, which was utilised as the adhesive and diluted by way of using Aston as a pure solution. The adhesives physical characteristics are listed in Table 1.

Single Sheet Testers (SST) or Epstein frames are regularly used to determine the characteristics of electrical steel, as recommended by IEC-Standards (IEC 404-2) (S. Standard, 1996) and (IEC 404-3) (S. Standard, 1992). The significance of this type of measurement system is specific to the permeability and power loss (W/kg) at various magnetising frequencies. The “Wolfson Centre for Magnetics” at Cardiff University developed the Single Strip Tester (SST) system, which delivers highly accurate and automatic measurements. With the exception of the frame itself, the Epstein frame system is comparable to the (SST) (Bertotti, 1987). The system is shown in Figure 2 and comprises of a PC with LabVIEW version 2019 already installed, a DAQ (data acquisition Card) from NI PCI-6120, a power of amplifier, a 1 Ω shunt resistor (Rshunt), as well as an air flux adjusted SST (Somkun, 2010; Instruments, National, 2002). According to IEC 404-3, twin vertical yokes of GO silicon steel or nickel-iron alloy are employed. Furthermore, twin vertical yokes are made of GO silicon steel or nickel-iron alloy. The main coil of 865 turns (N1) is wrapped around the secondary winding, whilst a secondary coil of 250 turns (N2) is wrapped around the plastic frame. Between the yokes is a typical Epstein strip size sheet (305) mm long and 30 mm wide. This setup provides a low reluctance path. AC tests were performed under comparable settings, with frequencies ranging from 50 to 1000 Hz.

The samples were sliced and prepared (roughly 15–20 mm square) for examination under a scanning electron microscope to determine diffusion effects.

The final distribution relating to the effectiveness of the diffusion of the Manganese to both surfaces was investigated for strips of non-oriented electrical steel were tested as follows: concentration profiles measured Energy Dispersive X-ray Spectrometer (EDS) by using Scanning Electron Microscope (SEM).

Scanning Electron Microscope (SEM) was used to characterisation the sample’s composite-coated steel. Inclusions were observed by Scanning Electron Microscope (SEM) and energy-dispersive spectroscopy (EDS) for elemental analysis.

The overall loss contains both hysteresis and eddy current loss components. The per-cycle hysteresis loss is presently determined using a different method. This approach measures core loss at various frequencies. When the magnetisation frequency curves at different flux densities are extrapolated to zero frequency, the hysteresis loss per cycle, which is the hysteresis energy loss per cycle, can be determined (Mthombeni and Pillay, 2006). P c f = C h f B p k n + C e f B p k 2 (1)

The linear equation is the reason for the magnetising frequency using the maximum flux density B. Subsequently, the core loss data are applied with the aim of plotting the curves of Pc/f versus f. The curves are straight lines. P c f = A + B f (2) Where A = Kℎ B p k n means hysteresis loss per cycle and B = B p k 2 .

More information for determining these coefficients is available in (Chen and Pillay, 2002; Ionel et al., 2006; Ibrahim and Pillay, 2012).

To obtain the overall hysteresis loss, multiply the static hysteresis power by the magnetising frequency. The hysteresis power loss per cycle is frequency independent. Having this knowledge is essential when applying the extrapolation method. Nevertheless, only at low frequencies is this supposition is valid. It would result in a non-uniform magnetic field throughout the lamination at higher frequencies, which would complicate the calculation.

Consequently, at every point within the lamination, each cycle’s hysteresis loop and hysteresis power losses are distinct. On account of this, it is better to use the extrapolation method to determine the core loss separation at a low frequency; otherwise, the skin effect should indeed be considered (Boon and Robey, 1958).

Figure 3 explains the SEM and EDS analysis of typical inclusions in the final annealed samples. The samples were coated and polished using acetone. Scanning electron microscope and EDS was performed to analyse the Manganese inclusions. The EDS analysis revealed that a significant part of the sample that contains Manganese. The inclusions containing Manganese were predominantly (<250 µm) in size. Inclusions with different Manganese contents can be seen in Figures 3–5. The results of the EDS analysis are shown in the tables in relation to each figure.

In the table pertaining to Figure 5, it can be noted that the inclusions that contain small amounts of Mn are exceedingly complex. The inclusions contain even more zirconium than Manganese. The high oxygen and carbon contents of the inclusions reveal the presence of Manganese oxides.

After coating the electrical steel with MnO2, the analysis of the deep elements was performed using EDS. During the EDS measurement, different areas were focused on. Figure 3 illustrates the identical peak. All of the MnO₂ can be noticed in the synthesised, nanostructures in the EDS spectrum. Where it was revealed that the major elements present were manganese, oxygen and carbon, in spectrum 55, the quantity of Mn, O₂ and C were 59.2, 25.79, and 10.18 weight %, respectively. Subsequently, after heat treatment, a small amount of impurity in conjunction with the presence of reduced titanium, was observed. However, while observed in spectrum 24, the values were noted to be 20.2, 28.95, and 2.68 measured in weight % for Mn, O₂ and C, respectively.

It was also observed that after mixing the Mn compound with Si oil particles followed by a heat treatment at 525°C and 700°C, the carbon amount was ascertained to increase, Similarly, the EDS graph in the figures indicates the presence of Fe that confirms the successful coating of the surface of the Mn with Si oil.

The elemental composition of the material was analysed using EDS to determine the nature of the composite, revealing that true functionalization occurred. Details of the three EDS spectra of the concentration (Mn) values measured in atomic and weight % are listed.

Figure 6 describes the impact of temperature on the characteristics of the sample in terms of power losses at 400 Hz and 1.5 T. The power losses increase linearly when the temperature of the sample increases (uncoated or coated). The effect of temperature on the samples’ properties confirms that the highest increase in power loss at 52 W/kg occurs at 700°C when the firing time was 45 min.

At the melting point of Mn at 525°C, the following values were obtained: uncoated at 39 W/kg and coated at 35 W/kg when the firing time was 60 min. Also, from Figure 6, as can be seen, the maximum temperature of 800°C has a slight reduction in power losses to 49 W/kg when compared to 700°C. Thus, its dissolving power is maximised.

Figure 6 explains the power losses with different temperatures for the Manganese oxide sample both with and without coating whilst heating from 525°C to 800°C, respectively. In both cases, the power losses increased as the temperature increased from 525°C to 700°C. However, when the temperature reached 800°C, there was an insignificant decrease in power losses. Furthermore, the peak of power losses appeared at 700°C, indicating a significant increase in power losses, as shown in Figure 6 in both cases. Based on the results of the melting performance for manganese oxide, the annealing temperatures implemented in this study were 525°C, 700°C, and 800°C. Therefore, the fact that the melting point of manganese oxide is 525°C has contributed to improving the performance of samples and reducing power losses, is considered.

Figure 7 demonstrates permeability at 400 Hz, 1.5 T for coated and uncoated samples with manganese oxide contents at different temperatures during sintering times of 1 hour, 45 min, and 30 min. It was observed that the permeability rose with the temperature and some of samples used (coated and uncoated). It was established that permeability decreased when the sample reached 525°C but was high when used at the same temperature without being coated. It was observed that coated specimens had less permeability than the uncoated samples at 700°C. Magnetic permeability decreases by increasing the temperature rate to 800°C. This is caused by conditions brought on by an excess of lubricant that reduced permeability. Holding the sintering time had a visible impact on magnetic permeability, which will help minimise specimen flaws. It is apparent that specimens with an hour-long holding period have greater magnetic permeability than those with a 30-min holding period.

Consequently, the permeability of composites increases with increasing annealing temperature, whereas power loss diminishes when the annealing temperature is increased.

From the results obtained by applying the following method to calculate power loss separation, the power loss of the non-oriented electrical steel samples was measured using the SST system explained in (Chen and Pillay, 2002) and (Ibrahim and Pillay, 2012), in which all power loss measurements were made.

Typically, as shown in Table 4, at a maximum flux density of 1.5 T and magnetising frequencies of 50, 100, 200, 400, 500, and 700 Hz to 1 kHz, the power loss of NO steel is calculated.

Figure 8 shows the measured power loss per cycle versus the frequency’s square root. By fitting a polynomial curve to this curve in Microsoft Excel, we can find the coefficients of the power loss parts.

The fitting equation’s residual values are often very near to unity, or R 2 = 1, suggesting an accurate approximation.

Table 5 shows the results of determining the power loss components at various frequencies using the equations from Figure 8. A = C h f B p k n = 0.0321 (10) B = C e f B p k 2 = 0.0002 (11) C = C a f B p k 1.5 = 0.0005 (12)

The power loss components, as illustrated in Table 5, are calculated by applying these coefficients in Eq. 6. Figure 9 demonstrates the relationship between the magnetising frequency and the hysteresis power loss and eddy current power loss at 1.5 T. The static hysteresis loops are utilised to calculate hysteresis loss. However, an equation may be used to calculate eddy current power loss per cycle as a linear function of the magnetising frequency: P e = π 6 ρ d 2 f 2 B p k 2 (13) Where B p k is the peak flux density, (T); f is frequency, (Hz); ρ is resistivity (Ω. m); D is density, (kg/ m 3 ); and d is the thickness of the lamination (m) (Bertotti, 1987).

Power loss of non-oriented steel is measured at a peak flux density of 1.5 T at different magnetizing frequencies with total power loss per cycle with coating as shown in Table 6.

The power loss separation findings for the same material with a coating are shown in Table 7.

Figure 10 shows the measured power loss per cycle versus the frequency’s square root. By fitting a polynomial curve to this curve in Microsoft Excel, we can find the coefficients of the power loss parts with coating.

Table 8 illustrates a comparison was made between the eddy current power loss obtained by the extrapolation method (experimental based) and eddy current calculated by the conventional formula (theory-based). This experiment indicated that the eddy current loss calculated in a sample at low induction and low frequency is virtually equal to that observed by the extrapolation method. Most of the extra loss changes with the frequency and strength of the field at the same eddy current power loss.

Table 9 displays the percentage reduction in power loss for uncoated and coated samples at various frequencies and verifies that the reduction in power loss occurs at a peak flux density of 1.5 T. The diffusion of manganese (IV) oxide reduces losses by approximately 8% at 500Hz and 7% at 400 Hz at the same value flux density. Moreover, as the internal effects of the internal flux distribution are greater even when AC magnetism was used, the unequal amount of Mn causes a resistivity gradient that lowers eddy current loss.

Table 10 displays the percentage reduction in permeability for coated and uncoated specimens at various frequencies. In the case of the uncoated samples, it can be observed that permeability on the samples is visibly effected by increased frequencies from 100 to 700 Hz, while reducing in the case of the coated samples, which achieve higher magnetic permeability at 1 KHz with coating and the highest percent decrease in permeability at 50 Hz with coating at a peak flux density of 1.5 T. The diffusion of manganese (IV) oxide reduces magnetic permeability at the same value of flux density. By the diffusion of MnO₂ into the Si-Fe strips, the texture of the material has been affected leading to a higher anisotropic coefficient, which was noticed when the relative permeability of the coating material has been reduced.

The hysteresis loop of coated and uncoated samples is shown in Figure 11. The diffusion and distribution of manganese are reflected by the size of the hysteresis loop, which is proportional to the amount of material stored in the samples. The hysteresis-loop area for coated is larger and has highest saturation flux density than the raw material. AC core losses were calculated at a frequency of 500 Hz, and steel strips were magnetised to saturation. At saturation, the 500 Hz flux density approaches 1.5 T based on the magnetising current frequency. At low frequencies, the lowest core loss is measured at 525°C, and the core loss increases gradually with increasing annealing temperature; at high frequencies, the minimum core loss is observed at 800°C, and it decreases gradually with increasing annealing temperature. Consequently, a 500 Hz and 1.5 T hysteresis loop was used to investigate the magnetic properties.