Received: 03 Mar 2020
Revised: 31 Mar 2020
Accepted: 06 Apr 2020
Published online: 01 May 2020
Influence of the Annealing Process on Magnetic Performance of Iron based Soft Magnetic Composites
Yan An,1 Shuai Feng,1 Guoqing Shao,2 Wanli Yuan,1 Kai Sun,1,* Xiaofeng Li1 and Runhua Fan1,*
1 College of Ocean Science and Engineering, Shanghai Maritime University, Shanghai 201306, China
2 Shandong Jingchuang Magnetoelectric Industry Technology Research Institute Co., Ltd. Shandong 276000, China
*: kais@shmtu.edu.cn (K. Sun) and rhfan@shmtu.edu.cn (R. Fan)
Abstract
In this work, the reduced iron powders were phosphated to form ferrous phosphate insulating layer on their surface. Then the iron based soft magnetic composites (SMCs) were prepared successfully by powder metallurgy and annealing process. The presence of insulating layer decreased the eddy current loss of iron powders. Furthermore, the effect of the annealing process on magnetic performance was investigated. The appropriate annealing temperature and holding time were carried out in this work. The samples were annealed in a N2 atmosphere from 550 to 650 oC and kept for 10 to 30 min. According to the results, the phosphated iron based SMCs that were annealed in a N2 atmosphere at 600 oC for holding 15 min of holding time had the satisfying magnetic properties. It has a high resistivity (ρ) of 0.0173 Ω·cm, a high saturated magnetic flux density (Bs) of 1.408 T, a high maximum permeability (μm) of 263, a low coercivity (Hc) of 197 A/m and a low core loss (Ps) of 119.70 W/kg at 1000 Hz (Bm=1 T).
Keywords: soft magnetic composites metallurgy powders annealing process magnetic performance
1. Introduction
The iron based soft magnetic composites (SMCs) have a potential value and broad developing prospects in the high-frequency electronic components field for their easily flexible design and low eddy current loss.[1-3] They can be pressed into desirable shapes using the powder metallurgy technology. Moreover, the insulating layer on the surface of iron powders effectively decreases the internal and external eddy current of raw iron powders.[4] Compared with conventional coating materials such as SiO2, Al2O3 and epoxy resin,[5-12] the phosphated iron powders have the advantage in avoiding the external impurity during the in-situ generating ferrous phosphate process. Meantime, the thickness of ferrous phosphate insulating layer is easy to be controlled by adjusting the concentration of phosphate acid and reacting conditions. Therefore, it has a great potential value in an industrial batch production with high economic benefits.
On the other hand, it is reported that the introduced non-ferromagnetic insulating layer can lead to a magnetic dilution.[13-15] For this reason, the annealing process was applied in order to eliminate this adverse effect.[16] The aim of annealing process is to eliminate the residual stress of the iron based SMCs during the molding process and then to improve the permeability and saturated magnetic flux density.[17] In the annealing process, the annealing temperature and holding time are two important factors in order to release the residual stress of internal iron powders. Moreover, the defects and holes in the iron powders can be eliminated in the annealing process, which makes a dense internal structure of the molded SMCs and then improves the magnetic performance. Based on this fact, the effect of annealing process on the magnetic performance of iron based SMCs was investigated by adjusting the annealing temperature and holding time in this paper.
2. Experimentals
2.1. Materials
The reduced iron powders with an average particle size of 50 μm were supplied by Jingchuang Co., Ltd (China). The purity of Fe was above 99 %. The phosphoric acid, alcohol and zinc stearate were supplied by Sinopharm Chemical Reagent Co., Ltd (China).
2.2. Preparation of iron based SMCs
Firstly, the phosphoric acid was diluted by the alcohol, then the raw iron powders were added to coat the surface of iron powders by ferrous phosphate. After that, pure iron powders were separated and dried for 30 min. The zinc stearate powders (0.2 wt% of iron powders) as the lubricant were mixed with the prepared powders. Subsequently, the phosphated iron powders were cold-compacted into annular samples (40 mm outer diameter, 32 mm inner diameter and 9 mm height) under a 1000 MPa pressure. Eventually, the prepared samples were annealed in the N2 atmosphere. In order to investigate the effect of annealing process, the annealing temperature were designed from 550 to 650 oC and the holding time from 10 to 30 min, respectively.
2.3. Materials characterization
The microstructure analysis and phase identification for the prepared SMCs were characterized by scanning electron microscopy (SEM, FEI Quanta 650) and X-ray diffraction (XRD, Bruker D8), respectively. The mass loss after the annealing process was measured by a thermal gravimetric analyzer (HS, TGA-101). The resistivity was measured by a four-point probe resistance analyzer (Kunde, KDY-1) at room temperature. The magnetic properties, including saturated magnetic flux density (Bs), maximum permeability (μm), core loss (Ps) and coercivity (Hc) were measured by a silicon steel material testing device (Lianzhong, Co., Ltd., MATS-3000M). The measurement frequency was below 1000 Hz.
3. Results and discussion
Fig. 1 SEM images of raw iron powders (a), phosphated iron powders (b), and elemental analysis (c).
Fig.1 shows the SEM images of the raw iron powders (Fig.1a) and phosphated iron powders (Fig.1b). It can be seen that pure iron powders had a rough and irregular surface, as shown in Fig.1a. After processed by the phosphate acid, the surfaces of iron powders were coated by a uniform and smooth insulating layer (Fig. 1b). The composition and elements analysis are shown in Fig. 1c. The iron, phosphate and oxide elements were detected on the surface of phosphated iron powders by the mapping scan. According to the acquired table of the element contents, the mass fraction (wt%) of P element was 7.45. However, the wt% of P element in ferrous phosphate was 18.9. This result indicates that the surface of iron powders is not completely covered by the ferrous phosphate.
The compositions of raw powders and phosphated iron powders with different annealing temperatures (from 550 to 650 oC) were detected by the X-ray diffraction measurement, as shown in Fig. 2. It can be seen that the characteristic peaks (110), (200), (211) and (220) of the Fe phase were observed. Moreover, the intensity of the peaks from the Fe phase became stronger with the increase of the annealing temperature, which resulted from the improved recrystallization degree of iron powders in the annealing process. The peaks from ferrous phosphate compounds on the surface of iron powders were not distinctly observed. The reason is that the phosphate compounds exist in the form of amorphous. Meanwhile, the amorphous ferrous phosphate insulating layer could be too thin to be detected by the X-ray.
The FT-IR spectra of the iron based SMCs are depicted in Fig. 3. The chemical bond of POfrom the ferrous phosphate can be verified by the presence of characteristic peaks, which appear at around 570 and 1040 cm-1, respectively. The characteristic peak at around 3400, 1630 and 1380 cm-1 represent the vibration absorption peak of -OH. The vibration peaks at 2850 and 2924 cm-1 correspond to the -CH2 and -CH3. The peak at around 2360 cm-1 can be attributed to the asymmetric stretching vibration of -NH, which originates from the impurities in the iron powders. According to the FT-IR spectra, it can be judged that ferrous phosphate was formed on the surface of iron powders.
Fig. 2 The X-ray diffraction pattern of phosphated iron powders with different annealing temperatures.
Fig. 3 The FT-IR spectra of raw iron powders and phosphated iron powders.
Fig. 4 shows the thermal gravimetric analysis (TGA) curve of the iron based SMCs. For this measurement process, it is different from conventional measurement that the heating rate is variable to match the actual annealing process. The first weight mass can be observed at around 180 °C, which originates from the evaporation of crystal water. The process can be described as follows:
(1)
It results from the weight loss of crystal water in the ferrous phosphate. Besides, it can be seen that the mass loss is about 0.3 % when the temperature reaches 280 oC, which corresponds to the evaporation of zinc stearate and the lubricant. The weight mass at 370 °C corresponds to the oxidation process of ferrous phosphate. When the temperature attaches at around 485 oC, it has a small weight loss. According to the previous investigation,[18-19] the ferrous phosphate that was coated on the surface of iron powders begins to decompose at 475 °C. Thus, it can be considered that the mass loss originates from the decomposition of iron phosphate insulting layer.
Fig. 4 The TGA images of phosphated iron based SMCs.
Fig. 5 The static hysteresis loops of the prepared phosphated iron based SMCs; the inset shows the enlarged part of the hysteresis loops.
Fig. 5 shows the static hysteresis loop and residual magnetic induction (shown in the enlarged zone) of the prepared phosphated SMCs with different annealing temperatures from 550 to 650 oC. In Fig.5, it can be observed that all lines show a typical soft magnetic character with a long and narrow surrounded area. Generally, the area presents the value of hysteresis loss. The magnetic induction (B) had not reached the saturated degree when the magnetic field strength (H) was attached at 4000 A/m. With the increase of annealing temperature, the SMCs samples were easier to be magnetized because a higher temperature makes a lower internal stress and then it is convenient to the motion of domain wall. As a result, the sample annealed at 650 oC has the highest magnetic induction and residual magnetic induction (inset of Fig. 5).
The magnetic performance including maximum permeability (μm), saturated magnetic induction (Bs) and coercive force (Hc) are shown in Fig. 6. It suggested that the permeability and saturated magnetic induction increased when the annealing temperature changed from 550 to 650 °C.
Fig. 6 The magnetic performance of iron based SMCs with different annealing temperatures.
This can be illustrated by the following reason. Mainly, the invalidation of the ferrous phosphate insulating layer at higher temperatures resulted in a lower non-magnetic content. Besides, according to the magnetic chip theory,[20] a higher temperature is conducive to relieving the internal stress of molding iron powders and to acquiring a better crystal structure. For the same reason, the coercive force dropped with the increase of annealing temperature.
In Table 1, the density, resistivity and core loss of phosphated SMCs with different annealing temperatures from 550 to 650 °C are listed. It can be seen that the density of SMCs hardly changed with increasing the annealing temperature. Moreover, as shown in Table 1, the resistivity and core loss of phosphated iron based SMCs with different annealing temperatures from 550 to 650 °C are listed. It can be obviously seen that the resistivity dropped sharply when the temperature increased from 550 to 575 °C. It confirmed that the ferrous phosphate insulating layer began invalid at this temperature. Furthermore, the sample had the lowest core loss (Ps), i.e., 113.50 W/kg at 575 oC. The reason is that parts of insulating layers still existed. This conclusion can also be verified by the magnetic performance in Fig. 7. The saturated magnetic induction had a rapid rise when the annealing temperature increased from 575 to 600 °C. Combining with the data of resistivity, it can be suggested that the phosphated insulating layer got into valid totally when the annealing temperature was higher than 600 °C.
According to the above data and investigation, the prepared SMCs annealed at 600 °C had a satisfying loss and a better magnetic performance. Thus, the annealing temperature was set at 600 °C to investigate the influence of holding time on magnetic performance. Fig.7 shows the magnetic performance of the samples processed at 600 oC with different annealing holding time from 10 to 30 min. The tendency of maximum permeability, saturated magnetic induction and coercive force are different from those on the condition of increasing the annealed temperature (as shown in Fig. 6), which changed in a linear manner. It is clear that the maximum value of maximum permeability, 286, saturated magnetic induction, 1.428 T and coercive force, 213 A/m appeared when the annealing time attached at 20 min. Moreover, the saturated magnetic induction was below 1.41 T when the holding time was maintained at 15 and 25 min. The reason for this non-linear variety can be suggested that the ferrous phosphate insulating layer was completely destroyed during the annealing process when the holding time was over than 15 min.
In Table 2, the density, resistivity and core loss (Ps) of phosphated SMCs with different annealing time from 10 to 30 min at 600 °C were listed. It shows that the density increases with the extension of holding time in the annealing process. The reason is that the internal stress can be released more and more fully with increasing the holding time at a certain temperature.[21] Furthermore, the internal defects and holes are also diminished.
Fig. 7 The magnetic performance of iron based SMCs in 600 oC with different holding time.
Furthermore, the internal defects and holes are also diminished. Compared with the data from Table 1, it can be easily seen that the density of iron based SMCs dropped obviously from 7.46 to 7.43 g/cm3 with the decrease of annealing time, which confirmed that the internal stress can be reduced during the annealing process. Furthermore, when the holding time attached at 15 min in 600 oC, the core loss (Ps) was reduced to the lowest value of 119.70 W/kg. As mentioned above, when the annealing time was over than 575 °C, the insulating layer has been destructed totally. At this condition, the extended holding time in the annealing process has no influence on the structure of iron powders. Therefore, the resistivity shows a nonlinear variation with increasing the holding time. Consequently, the Ps was mainly affected by the hysteresis loss.
Fig. 8 The core loss (Ps) below 1000 Hz of the iron based SMCs with (a) different annealing temperatures and (b) holding time.
Table 1. The resistivity and core loss (Ps) for the prepared SMCs with different annealing temperatures.
|
Annealing temperature (°C) |
Density (g/cm3) |
Resistivity (Ω·cm) |
Ps @ Bm = 1 T (W/kg) |
||||
|
Frequency (Hz) |
|||||||
|
|
100 |
200 |
400 |
500 |
1000 |
||
|
550 |
7.46 |
40.0741 |
10.44 |
21.62 |
45.88 |
57.68 |
118.00 |
|
575 |
7.47 |
0.2004 |
9.93 |
20.23 |
42.08 |
53.21 |
113.50 |
|
600 |
7.46 |
0.0908 |
10.20 |
20.53 |
41.70 |
53.29 |
122.33 |
|
625 |
7.47 |
0.0099 |
9.48 |
19.66 |
42.85 |
55.88 |
138.73 |
|
650 |
7.47 |
0.0042 |
9.44 |
20.69 |
48.19 |
64.19 |
168.83 |
Fig. 8 directly shows the core losses with different annealing temperatures and holding time for the prepared iron based SMCs. The measurement frequency points are 100, 200, 400, 500 and 1000 Hz, respectively. As shown in Fig. 8a, the Ps grew slowly with the increase of frequency from 100 to 400 Hz, then had a rapid growth when the frequency was over than 600 Hz. It has the same trend for the Ps with different holding time, as shown in Fig. 8b. For the iron based SMC, the Ps is usually composed of hysteresis loss (Ph), eddy current loss (Pe) and residual loss (Pr). Usually, the Pr can be ignored at a low frequency. Thus, the total core loss for the SMCs can be expressed by Equations (2) and (3):[22]
(2)
(3)
where the Ch represents the coefficient for hysteresis loss, Bm represents the maximum magnetic flux density and f is the test frequency. In this work, the Bm was a fixed value in the measurement process. According to the previous investigation,[23] the Pe can be classified into two parts: the inter eddy current loss (Pein) and intra eddy current loss (Petra) of particles, which can be described by the following formula:[24]
(4)
(5)
In this formula, d is related to the size of iron particle and deff is related to the effective size, which means the average size of integral iron powders; β1 represents the rectangular geometrical factor and β2 represents the granular geometrical factor, which depends on the geometry of iron particle; ρs and ρp can be regarded as the bulk resistivity of molded samples and intra-particle resistivity of iron powders.[25,26]
For the prepared phosphated iron based SMCs, raw materials and the molding process were controlled in the same condition, so the d, deff, β1 and β2 had nearly the same values. Based on this fact, the core loss was determined by the Ch , 
and 
at the same frequency. The Ch represents the coefficient for the hysteresis loss and the , 
represent the coefficients for the eddy current losses, which are related to the changes of external magnetic field and the size, resistivity of measured samples. At this condition, the value of core loss only depends on the annealing process. Appropriate temperature and holding time can release enough internal stress inside the iron powders and keep the ferrous phosphate insulating layer from being damaged. Under this condition, the iron based SMCs had lower values of Ch, , and higher values of ρs and ρp.
Table 2. The resistivity and core loss (Ps) for prepared SMCs with different annealing time at 600 °C.
|
Annealing time (min) |
Density (g/cm3) |
Resistivity (Ω·cm) |
Ps @ Bm = 1 T (W/kg) |
||||
|
at frequency (Hz) |
|||||||
|
|
100 |
200 |
400 |
500 |
1000 |
||
|
30 |
7.46 |
0.0908 |
10.20 |
20.53 |
41.70 |
53.29 |
122.33 |
|
25 |
7.45 |
0.0159 |
9.28 |
19.57 |
41.65 |
53.35 |
121.77 |
|
20 |
7.45 |
0.0131 |
9.38 |
19.83 |
43.75 |
53.01 |
137.23 |
|
15 |
7.44 |
0.0173 |
9.50 |
19.53 |
40.63 |
51.89 |
119.70 |
|
10 |
7.43 |
0.0165 |
9.40 |
19.41 |
41.04 |
53.04 |
125.33 |
4. Conclusion
In this work, the phosphated iron based SMCs were produced by the powder metallurgy technology. Then the annealing process was applied to improve the magnetic performance simultaneously. In order to acquire both high permeability, saturated magnetic induction and low core loss, the effects of annealing temperature and holding time on the magnetic performance of the SMCs were investigated. According to the results, the phosphated iron based SMCs which were annealed in a N2 atmosphere at 600 °C for 15 min had the satisfying magnetic properties. It has a high resistivity (ρ) of 0.0173 Ω·cm, a high saturation magnetic flux density (Bs) of 1.408 T, a high maximum permeability (μm) of 263, a low coercivity (Hc) of 197 A/m and a low core loss (Ps) of 119.70 W/kg at 1000 Hz (Bm=1 T). This work provides a promising annealing route to produce iron based SMCs with excellent magnetic performances and low core losses.
Acknowledgements
The authors gratefully acknowledge the financial supports by the National Natural Science Foundation of China (Grant No. 51803119, 51871146 and 51771108), the Innovation Program of Shanghai Municipal Education Commission (Grant No. 2019-01-07-00-10-E00053), Chenguang Program supported by Shanghai Education Development Foundation and Shanghai Municipal Education Commission (Grant No. 18CG56) and the Science and Technology Commission of Shanghai Municipality (Grant No. 18DZ1112900) and Shanghai Engineering Technology Research Centre of Deep Offshore Material (19DZ2253100).
Conflict of Interest
There is no conflict of interest.
Supporting Information
Not Applicable.
References
[1] A. H. Taghvaei, H. Shokrollahi and K. Janghorban, J. Magn. Magn. Mater., 2009, 321, 3926-3932, doi: 10.1016/j.jmmm.2009.07.061.
[2] S. Wu, A. Sun, Z. Lu and C. Cheng, Mater. Chem. Phys., 2015, 153, 359-364, doi: 10.1016/j.matchemphys.2015.01.026.
[3] M. Strečková, Ľ. Medvecký, J. Füzer, P. Kollár, R. Bureš and M. Fáberová, Mater. Lett., 2013, 101, 37-40, doi: 10.1016/j.matlet.2013.03.067.
[4] B. Yang, Z. Wu, Z. Zou and R. Yu, J. Phys. D-Appl. Phys., 2010, 43(36), 365003, doi: 10.1088/0022-3727/43/36/365003.
[5] S. Wu, A. Sun, W. Xu, Q. Zhang, F. Zhai, P. Logan and A. A. Volinsky, J. Magn. Magn. Mater., 2012, 324, 3899-3905, doi: 10.1016/j.jmmm.2012.06.042.
[6] A. H. Taghvaei, H. Shokrollahi, K. Janghorban and H. Abiri, Mater. Design, 2009, 30, 3989-3995, doi: 10.1016/j.matdes.2009.05.026.
[7] S. Wu, J. Fan, J. Liu, H. Gao, D. Zhang and A. Sun, J. Supercond. Nov. Magn., 2018, 31, 587-595, doi: 10.1007/s10948-017-4256-5.
[8] P. Xie, K. Sun, Z. Wang, Y. Liu, R. Fan, Z. Zhang and G. Schumacher, J. Alloy. Compd., 2017, 725, 1259-1263, doi: 10.1016/j.jallcom.2017.04.248.
[9] P. Xie, Z. Zhang, Z. Wang, K. Sun and R. Fan, Research, 2019, 2019, 1021368, doi: 10.34133/2019/1021368.
[10] K. Sun, S. Feng, Q. Jiang, X. Li, Y. Li, R. Fan, Y. An and J. Wang, J. Magn. Magn. Mater., 2020, 493, 165705, doi: 10.1016/j.jmmm.2019.165705.
[11] X. Zhu, J. Yang, D. Dastan, H. Garmestani, R. Fan and Z. Shi, Compos. Part A-Appl. S., 2019, 125, 105521, doi: 10.1016/j.compositesa.2019.105521.
[12] P. Xie, Y. Li, Q. Hou, K. Sui, C. Liu, X, Fu, J. Zhang, V. Murugadoss, J. Fan, Y. Wang, R. Fan and Z. Guo, J. Mater. Chem. C, 2020, 8, 3029-3039, doi: 10.1039/c9tc06378a.
[13] W. Li, W. Huang, Y. Kang, Y. Guo, Y. Ying, J. Yu, J. Zheng, L. Qiao and S. Che, Compos. Sci. Technol., 2019, 173, 7-14, doi: 10.1016/j.compscitech.2019.01.022.
[14] Y. Kang, W. Li, T. Ma, X. Huang, Y. Mo, Z. Chu, Z. Zhang and G. Feng, Compos. Sci. Technol., 2019, 174, 184-193, doi: 10.1016/j.compscitech.2019.02.029.
[15] W. Li, Z. Wang, Y. Ying, J. Yu, J. Zheng, L. Qiao and S. Che, Ceram. Int., 2019, 45, 3864-3870, doi: 10.1016/j.ceramint.2018.11.058.
[16] J. Lei, J. Zheng, H. Zheng, L. Qiao, Y. Ying, W. Cai, W. Li, J. Yu, M. Lin and S. Che, J. Magn. Magn. Mater., 2019, 472, 7-13, doi: 10.1016/j.jmmm.2018.09.125.
[17] Y. Liang, F. Ye, J. Lin, Y. Wang, G. Chen, J. Alloy. Compd., 2010, 491, 268-270, doi: 10.1016/j.jallcom.2009.10.118.
[18] G. Zhao, S. Wu and M. Yan, J. Alloy. Compd., 2016, 685, 231-236, doi: 10.1016/j.jallcom.2016.05.277.
[19] M. Huang, C. Wu, Y. Jiang and M. Yan, J. Alloy. Compd., 2015, 644, 124-130, doi: 10.1016/j.jallcom.2015.04.201.
[20] B. M. Tanygin, J. Magn. Magn. Mater., 2011, 323, 616-619, doi: 10.1016/j.jmmm.2010.10.028.
[21] S. Wu, A. Sun, F. Zhai, J. Wang, Q. Zhang, W. Xu, P. Logan and A. A. Volinsky, J. Magn. Magn. Mater., 2012, 324, 818-822, doi: 10.1016/j.jmmm.2011.09.026.
[22] O. de la Barrie`re, C. Appino, F. Fiorillo, C. Ragusa, H. B. Ahmed, M. Gabsi, F. Mazaleyrat and M. Lobue, J. Appl. Phys., 2011, 109, 07A317, doi: 10.1063/1.3554207.
[23] J. Li, J. Yu, W. Li, S. Che, J. Zheng, L. Qiao and Y. Ying, J. Magn. Magn. Mater., 2018, 454, 103-109, doi: 10.1016/j.jmmm.2018.01.061.
[24] A. H. Taghvaei, H. Shokrollahi and K. Janghorban, Mater. Des., 2010, 31, 142-148, doi: 10.1016/j.matdes.2009.06.044.
[25] W. Li, W. Wang, J. Lv, Y. Ying, J. Yu, J. Zheng, L. Qiao and S. Che, J. Magn. Magn. Mater., 2018, 456, 333-340, doi: 10.1016/j.jmmm.2018.02.033.
[26] W. Cai, H. Cai, Y. Kang, Y. Ying, J. Yu, J. Zheng, L. Qiao, Y. Jiang and S. Che, Acta Mater., 2019, 167, 267-274, doi: 10.1016/j.actamat.2019.01.035.
Publisher’s Note
Engineered Science Publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.