France



Grain-oriented (GO) steel sheets increasingly face novel applications, like the ones enforced by the emerging &lt;i&gt;smart grid&lt;/i&gt; technologies and the related medium-to-high frequencies &lt;i&gt;conversion devices&lt;/i&gt; and those connected with the development of highly efficient low-noise rotating machines [1-3]. Comprehensive knowledge of the material behavior, going beyond the standard assessment of its properties at power frequencies along the rolling direction (RD), is therefore appreciated in the practice of machine design. This requires both experiments made upon broader than usual range of frequencies [4] and the investigation of the complex anisotropic response of the GO sheets and the related 2D magnetization process [5]. Such a process is shown to consists in the balanced contributions by moving [001] directed 180° domain walls (dws) and 90° dw transitions towards the [100] and [010] phases. It is then observed that, with the applied field leaving the RD, not only the magnetic losses suffer a sharp increase, but also a strong geometry dependent effect, under both static and dynamic regimes. This descends from the interference of the demagnetizing field, invariably arising in Fe-Si single crystals, but for closed samples cut along the &amp;lt;100&amp;gt;, &amp;lt;110&amp;gt;, and &amp;lt;111&amp;gt; axes. Consequently, the popular approach to GO characterization using Epstein strips cut at different angles θ to RD can provide the intrinsic properties of the material only for θ = 0° (RD) and θ = 90° (TD). This is ill-suited to straightforward use in those cases, like the rotating machine cores, where directions other than RD and TD are involved. However, the experimental knowledge of the geometry-independent RD and TD magnetization curves can be usefully exploited, following the lines of the Néel’s phase theory, to reconstruct the same curves measured on Epstein strips cut along a generic angle θ [6]. The same applies to the DC hysteresis loops and their area, once the fundamental role of the demagnetizing field is recognized. One can also emulate an infinitely extended sheet, the ideal configuration leading to the intrinsic material response, by adopting an X-stacked sheet configuration, where flux continuity across the sheet edges can be achieved, resulting in a net magnetization variation both along the applied field (i.e., the strip edges) and orthogonal to it. Also in this case the DC properties become predictable by starting from the RD and TD magnetic properties [6]. Any other sample geometry will then be fully represented by or will be intermediate between the Epstein and the X-stack configurations.&lt;br/&gt;In this work, we provide a general approach to DC and AC (1 Hz - 200 Hz) magnetic hysteresis loop and losses in Epstein and X-stacked HGO sheets. The experiments are made using either the standard Epstein frame or a single sheet tester, where 120 mm wide 300 mm long X-stacked sheet samples are inserted in a laminated flux-closing yoke. A phenomenological predicting method is devised, which relies on the pre-emptive knowledge of the behavior of hysteresis loop and energy loss independently measured under sinusoidal induction in the RD- and TD-cut Epstein strips. We treat, in accordance with magneto-optical observations, the magnetization process unfolding for generic angle θ as the result, dictated by applied, demagnetizing, and eddy current fields, of two mechanisms: 1) Displacement of the 180° dws, the process pertaining to the RD cut sheets and strips; 2) Motion of 90° dws, converting the antiparallel slab-like [001] domains into [010] and [100] domains, and vice versa. The frames in Fig. 1 qualitatively show an example of domain structure in an X-stacked sheet at two instants of time along a major hysteresis loop (peak polarization &lt;i&gt;J&lt;/i&gt;&lt;sub&gt;p&lt;/sub&gt; = 1.45 T). The antiparallel [001] domain structure observed at&lt;i&gt; J&lt;/i&gt; = 0, largely disappears upon attaining &lt;i&gt;J&lt;/i&gt;&lt;sub&gt; &lt;/sub&gt;= 0.87 T. With this configuration, only a faint demagnetizing field is expected to arise at the sample edges and a net 2D magnetization variation is observed. By taking then, for a given &lt;i&gt;J&lt;/i&gt;&lt;sub&gt;p&lt;/sub&gt;, the experimental energy loss and the hysteresis loops pertaining to the RD and TD samples, obtained at &lt;i&gt;J&lt;/i&gt;&lt;sub&gt;pRD&lt;/sub&gt; = &lt;i&gt;J&lt;/i&gt;&lt;sub&gt;p&lt;/sub&gt; cos(θ) and &lt;i&gt;J&lt;/i&gt;&lt;sub&gt;pTD&lt;/sub&gt; = &lt;i&gt;J&lt;/i&gt;&lt;sub&gt;p&lt;/sub&gt; sin(θ), and realizing that the fractional sample volume &lt;i&gt;v&lt;/i&gt;&lt;sub&gt;180&lt;/sub&gt;(&lt;i&gt;t&lt;/i&gt;) occupied by the [001] phase is related to the same quantity &lt;i&gt;v&lt;/i&gt;&lt;sub&gt;90&lt;/sub&gt;(&lt;i&gt;t&lt;/i&gt;) = √2 &lt;i&gt;J&lt;/i&gt;&lt;sub&gt;TD &lt;/sub&gt;/&lt;i&gt;J&lt;/i&gt;&lt;sub&gt;s&lt;/sub&gt; associated with the [010] and [100] domains as &lt;i&gt;v&lt;/i&gt;&lt;sub&gt;180&lt;/sub&gt;(&lt;i&gt;t &lt;/i&gt;) = 1 – &lt;i&gt;v&lt;/i&gt;&lt;sub&gt;90&lt;/sub&gt;(&lt;i&gt;t&lt;/i&gt;), we calculate loss and loop for any given θ value. An example of so-obtained energy loss &lt;i&gt;W&lt;/i&gt;(&lt;i&gt;f&lt;/i&gt;) at 50 Hz across a range of&lt;i&gt; J&lt;/i&gt;&lt;sub&gt;p &lt;/sub&gt;values on 0.288 mm thick high-permeability GO Epstein strips cut at different angles θ to RD is provided in Fig. 2. The experiments show that &lt;i&gt;W&lt;/i&gt;(θ) peaks around θ = 75°. The predicting approach to &lt;i&gt;W&lt;/i&gt;(θ) is based on the experimental knowledge of the energy loss measured at the same frequency in the RD and TD strips for &lt;i&gt;J&lt;/i&gt;&lt;sub&gt;pRD&lt;/sub&gt; = &lt;i&gt;J&lt;/i&gt;&lt;sub&gt;p&lt;/sub&gt; cos(θ) and &lt;i&gt;J&lt;/i&gt;&lt;sub&gt;pTD&lt;/sub&gt; = &lt;i&gt;J&lt;/i&gt;&lt;sub&gt;p&lt;/sub&gt; sin(θ). This associates with the calculation of the hysteresis loop. A more involved approach is required by the &lt;i&gt;W&lt;/i&gt;(θ) prediction in the X-stacked samples, where the simplifying assumption of zero magnetization across the sheet width does not hold.

**References:**

[1] T. Belgrand, R. Lemaître, A. Benabou, J. Blaszkowski et C. Wang, &lt;i&gt;AIP Advan&lt;/i&gt;ces 8, 047611 (2018).&lt;br/&gt;[2] G. Parent, R. Penin, J. P. Lecointe, J. F. Brudny, and T. Belgrand, &lt;i&gt;IEEE Trans. Magn&lt;/i&gt;. 49, 1977 (2013).&lt;br/&gt;[3] S. Lopez, B. Cassoret, J. Brudny, L. Lefebvre, and J. Vincent, &lt;i&gt;IEEE Trans. Magn. &lt;/i&gt;45, 4161(2009).&lt;br/&gt;[4] M. A. Shamsuddin, F. Rojas, R. Cardenas, J. Pereda, M. Diaz, and R. Kennel, &lt;i&gt;Energies&lt;/i&gt; 13, 2319 (2020).&lt;br/&gt;[5] W. Pluta, &lt;i&gt;IEEE Trans. Magn. &lt;/i&gt;52, 6300912 (2016).&lt;br/&gt;[6] F. Fiorillo, L. R. Dupré&lt;i&gt;, &lt;/i&gt;C. Appino, and A. M. Rietto, &lt;i&gt;IEEE Trans. &lt;/i&gt;&lt;i&gt;Magn. &lt;/i&gt;38, 1467 (2002).

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/6b5cb05f0e84b50643df6d3ceb985c44.jpg)
Fig. 1. X-stacked GO sheet, cut at 75° with respect to the RD. The [001] antiparallel slab-like domain structure found in the demagnetized state (&lt;i&gt;J&lt;/i&gt; = 0) is converted by 90° dw motion into symmetric [100] and [010] structures under increasing applied field (&lt;i&gt;J&lt;/i&gt;&lt;sub&gt;p&lt;/sub&gt; = 1.45 T).

![](https://assets.underline.io/uploads/markdown_image/1/image/853d1d39b26e5ad4a2c3c620a36c162b.jpg)
Fig. 2. Energy loss at 50 Hz measured versus &lt;i&gt;J&lt;/i&gt;&lt;sub&gt;p&lt;/sub&gt; and θ in Epstein strips. Solid lines and symbols: experiments. Dashed lines: prediction by a weighted combination of the 50 Hz RD and TD loss figures.

INTERMAG 2021

 The 2D Magnetization Process in HGO Fe-Si Sheets

**Please view our Welcome to MH21 video.**

<center>
<figure class="video_container">
	 <iframe src="https://screencast-o-matic.com/watch/cr1UYHV1pcz?v=6&ff=1&title=0&controls=1" width=826 height=526 frameborder="0" allowfullscreen="true"> </iframe>
</figure>
	</center>


**Walkthrough Video**

Here is a Walkthrough showing how to navigate the conference and network with others 

<center>
<figure class="video_container">
	 <iframe src="https://screencast-o-matic.com/watch/cr1vnrV1paE?v=6&ff=1&title=0&controls=1" width=826 height=526 frameborder="0" allowfullscreen="true"> </iframe>
</figure>
	</center>




	 Organized by
 
<a href="https://mrs.org.sg/">
<img src="https://assets.underline.io/uploads/markdown_image/1/image/95d5789689d5f900b90e9fe54ea3cf0d.jpg"</a>

 
	
	 In association with
 
<a href="http://www.ntu.edu.sg/">
<img src="https://assets.underline.io/uploads/markdown_image/1/image/334cd171ad2fdee033c635b368e5e160.png"</a>
	<a href="https://www.nus.edu.sg/">
<img src="https://assets.underline.io/uploads/markdown_image/1/image/32e6f82c0440f7bb24c590f9f67821bd.png"</a>
		<a href="https://www.a-star.edu.sg/">
<img src="https://assets.underline.io/uploads/markdown_image/1/image/72557865b589cfa3d2a0243ffce4a270.png"</a>
	
 


	 Supported by
 
		 <a href="https://www.visitsingapore.com/mice/en/about-us/about-secb/">
<img src="https://assets.underline.io/uploads/markdown_image/1/image/6e679842ead62be522ba6a630b82d5c9.png"</a>

To access the MH 21 Conference, [please register](https://mh21.mrs.org.sg/event/da3f9651-429f-4763-81b2-ebd7311b6bc8/websitePage:206f7605-1731-41f9-8421-56d3890f398f)

MATERIALS FOR HUMANITY (MH 21)

MH 21 is arranged around 6 themes that focus attention on pressing needs that should be addressed to enhance the well-being of communities and provide a forum for presenting breakthrough discoveries.

Grain-oriented (GO) steel sheets increasingly face novel applications, like the ones enforced by the emerging smart grid technologies and the related medium-to-high frequencies conversion devices and those connected with the development of highly efficient low-noise rotating machines [1-3]. Comprehensive knowledge of the material behavior, going beyond the standard assessment of its properties at power frequencies along the rolling direction (RD), is therefore appreciated in the practice of machine design. This requires both experiments made upon broader than usual range of frequencies [4] and the investigation of the complex anisotropic response of the GO sheets and the related 2D magnetization process [5]. Such a process is shown to consists in the balanced contributions by moving [001] directed 180° domain walls (dws) and 90° dw transitions towards the [100] and [010] phases. It is then observed that, with the applied field leaving the RD, not only the magnetic losses suffer a sharp increase, but also a strong geometry dependent effect, under both static and dynamic regimes. This descends from the interference of the demagnetizing field, invariably arising in Fe-Si single crystals, but for closed samples cut along the &lt;100&gt;, &lt;110&gt;, and &lt;111&gt; axes. Consequently, the popular approach to GO characterization using Epstein strips cut at different angles θ to RD can provide the intrinsic properties of the material only for θ = 0° (RD) and θ = 90° (TD). This is ill-suited to straightforward use in those cases, like the rotating machine cores, where directions other than RD and TD are involved. However, the experimental knowledge of the geometry-independent RD and TD magnetization curves can be usefully exploited, following the lines of the Néel’s phase theory, to reconstruct the same curves measured on Epstein strips cut along a generic angle θ [6]. The same applies to the DC hysteresis loops and their area, once the fundamental role of the demagnetizing field is recognized. One can also emulate an infinitely extended sheet, the ideal configuration leading to the intrinsic material response, by adopting an X-stacked sheet configuration, where flux continuity across the sheet edges can be achieved, resulting in a net magnetization variation both along the applied field (i.e., the strip edges) and orthogonal to it. Also in this case the DC properties become predictable by starting from the RD and TD magnetic properties [6]. Any other sample geometry will then be fully represented by or will be intermediate between the Epstein and the X-stack configurations. In this work, we provide a general approach to DC and AC (1 Hz - 200 Hz) magnetic hysteresis loop and losses in Epstein and X-stacked HGO sheets. The experiments are made using either the standard Epstein frame or a single sheet tester, where 120 mm wide 300 mm long X-stacked sheet samples are inserted in a laminated flux-closing yoke. A phenomenological predicting method is devised, which relies on the pre-emptive knowledge of the behavior of hysteresis loop and energy loss independently measured under sinusoidal induction in the RD- and TD-cut Epstein strips. We treat, in accordance with magneto-optical observations, the magnetization process unfolding for generic angle θ as the result, dictated by applied, demagnetizing, and eddy current fields, of two mechanisms: 1) Displacement of the 180° dws, the process pertaining to the RD cut sheets and strips; 2) Motion of 90° dws, converting the antiparallel slab-like [001] domains into [010] and [100] domains, and vice versa. The frames in Fig. 1 qualitatively show an example of domain structure in an X-stacked sheet at two instants of time along a major hysteresis loop (peak polarization Jp = 1.45 T). The antiparallel [001] domain structure observed at J = 0, largely disappears upon attaining J = 0.87 T. With this configuration, only a faint demagnetizing field is expected to arise at the sample edges and a net 2D magnetization variation is observed. By taking then, for a given Jp, the experimental energy loss and the hysteresis loops pertaining to the RD and TD samples, obtained at JpRD = Jp cos(θ) and JpTD = Jp sin(θ), and realizing that the fractional sample volume v180(t) occupied by the [001] phase is related to the same quantity v90(t) = √2 JTD /Js associated with the [010] and [100] domains as v180(t ) = 1 – v90(t), we calculate loss and loop for any given θ value. An example of so-obtained energy loss W(f) at 50 Hz across a range of Jp values on 0.288 mm thick high-permeability GO Epstein strips cut at different angles θ to RD is provided in Fig. 2. The experiments show that W(θ) peaks around θ = 75°. The predicting approach to W(θ) is based on the experimental knowledge of the energy loss measured at the same frequency in the RD and TD strips for JpRD = Jp cos(θ) and JpTD = Jp sin(θ). This associates with the calculation of the hysteresis loop. A more involved approach is required by the W(θ) prediction in the X-stacked samples, where the simplifying assumption of zero magnetization across the sheet width does not hold.

**References:**

[1] T. Belgrand, R. Lemaître, A. Benabou, J. Blaszkowski et C. Wang, AIP Advances 8, 047611 (2018). [2] G. Parent, R. Penin, J. P. Lecointe, J. F. Brudny, and T. Belgrand, IEEE Trans. Magn. 49, 1977 (2013). [3] S. Lopez, B. Cassoret, J. Brudny, L. Lefebvre, and J. Vincent, IEEE Trans. Magn. 45, 4161(2009). [4] M. A. Shamsuddin, F. Rojas, R. Cardenas, J. Pereda, M. Diaz, and R. Kennel, Energies 13, 2319 (2020). [5] W. Pluta, IEEE Trans. Magn. 52, 6300912 (2016). [6] F. Fiorillo, L. R. Dupré, C. Appino, and A. M. Rietto, IEEE Trans. Magn. 38, 1467 (2002).

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/6b5cb05f0e84b50643df6d3ceb985c44.jpg)
Fig. 1. X-stacked GO sheet, cut at 75° with respect to the RD. The [001] antiparallel slab-like domain structure found in the demagnetized state (J = 0) is converted by 90° dw motion into symmetric [100] and [010] structures under increasing applied field (Jp = 1.45 T).

![](https://assets.underline.io/uploads/markdown_image/1/image/853d1d39b26e5ad4a2c3c620a36c162b.jpg)
Fig. 2. Energy loss at 50 Hz measured versus Jp and θ in Epstein strips. Solid lines and symbols: experiments. Dashed lines: prediction by a weighted combination of the 50 Hz RD and TD loss figures.

technical paper

INTERMAG 2021 Welcome Video



	
<figure class="video_container">
 <iframe src="https://screencast-o-matic.com/player/crflqFVnUpZ?v=6&ff=1&title=0&controls=1" width=640 height=360 frameborder="0" allowfullscreen="true"> </iframe>
</figure>

Want to buy a ticket? 

Register for INTERMAG 2021 here

[![](https://assets.underline.io/uploads/markdown_image/1/image/17ce76e7960db633258024eef8d6da7f.png)](https://intermag.org/register)

Got your ticket, but still can’t access the event?

Contact us and our team will look into it.

Owing to the augmentation of wireless communication, the development of electromagnetic interference (EMI) shielding structures is an open area for research. The devices working at microwave frequencies are hazardous to human health and can also result in the malfunctioning of various sensitive equipments. Various techniques are reported by the researchers for the development of an efficient EMI shielding structure, some of them are composite materials, nanocomposites, metamaterials, and frequency selective surfaces (FSSs) [1]-[3]. FSSs are an array of periodic slots or apertures having excellent electromagnetic (EM) wave filtering properties. Major challenges across the research community are to create a miniaturized, angular stable, wideband, polarization-insensitive, and flexible EMI shielding structure. Multiband structures are often used by researchers to get a wide bandwidth. Whereas, miniaturization is attained either by techniques like 3-D, convoluted geometries, and lumped capacitances. All these methods of bandwidth improvement and miniaturization result in complex design and therefore fabrication becomes tedious. Almost all reported EMI shielding structures uses pure dielectric substrates like FR4 and Rogers. However, bandwidth and miniaturization can be improved by increasing the overall inductance of the shielding structure. The introduction of lumped inductors can be a solution but it not only increases the fabrication complexities but also may result in unwanted parasitic capacitance. Also introducing active or reactive components limits the conformal applications of shielding structures. One possible solution is to use magnetic substrates, which will increase the overall inductance of the shielding structure. Therefore, in this work electromagnetic properties of two magnetic substrates (Ferrite and magnetic silicon rubber) are measured and investigated to improve the bandwidth and miniaturization of an EMI shielding structure. Further, a polarization-insensitive, angular stable FSS is designed and the equivalent circuit model (ECM) is developed to have a physical insight into the structure. Moreover, a competent deep neural network (DNN) approach is employed for fast and accurate optimization of the shielding structure. At last, a prototype is developed and measured for validation of the proposed technique. The schematic of the proposed shielding structure is depicted in Figure 1(a). It comprises an array of periodic unit cells, a magnetic ferrite substrate, and a silicon rubber sheet. Figure 1(b) demonstrates an FSS unit cell with corresponding geometrical design variables. Here, ro is the radius of the outer circle and ri corresponds to the radius of the inner circle. d is the length of the patch, w is the width, and periodicity of the unit cell is represented by p. The electromagnetic properties of magnetic substrates are measured using a waveguide-based microwave measurement technique. Figure 1(c) depicts the measured relative permittivity (εr) and permeability (μr) of both the substrates. The real part of μr varies between 2.10 and 1.49 and the imaginary part varies between 0.34 and 0.0025 for magnetic silicon rubber when the frequency changes from 8.2 to 12.4 GHz. The real part of εr is 5.50 and the imaginary part varies between 0.20 and 0.005. On the other hand, the ferrite substrate real part of μr changes varies between 1.17 and 1.13 and the imaginary part varies between 0.04 and 0.001. Moving a step further, ECM of the FSS backed substrates is developed and the overall impedance is calculated. A huge database is created using the ECM and a deep neural network is trained using the Levenberg-Marquardt algorithm. The entire dataset is divided into three sections randomly for training, validation, and testing in the proportion of 70%, 15%, and 15%. Once the training, testing, and validation of the network is done, the shielding structure is optimized for resonant frequency and bandwidth. A prototype is then fabricated from the optimized parameters using the printed circuit board (PCB) technique. Figure 2(a) shows the image of the fabricated prototype. Meanwhile, the effect of the magnetic properties of the substrates on the transmission performance of the shielding structure is investigated. When the permittivity of the substrates dominates, a bandwidth of 2.75 GHz is obtained with resonant frequency at 10.42 GHz as shown in Figure 2(b). The bandwidth of surfaces with quite high impedance is directly proportional to the square root of overall inductance. To increase the bandwidth and to achieve miniaturization substrates with dominant magnetic properties are used. The overall inductance of the structure is increased by using substrates with dominant μr. Figure 2(b) depicts when substrates with dominant permeability are used in place of permittivity dominant substrates resonance shifts to lower value as the overall inductance of the structure increased. Also, the bandwidth obtained from such magnetic substrate increased to 3.37 GHz from 2.75 GHz. Therefore, it can be concluded that using magnetic substrates bandwidth enhancement and miniaturization of the shielding structure can be achieved.

**References:**

[1] M. Sonehara, T. Sato, M. Takasaki, H. Konishi, K. Yamasawa and Y. Miura, "Preparation and Characterization of Nanofiber Nonwoven Textile for Electromagnetic Wave Shielding," IEEE Transactions on Magnetics, vol. 44, no. 11, pp. 3107-3110, Nov. 2008. [2] J. Lee, B. M. Jung, S. B. Lee, S. Lee and K. H. Kim, "FeCoNi-Coated Glass Fabric/Polycarbonate Composite Sheets for Electromagnetic Absorption and Shielding," IEEE Transactions on Magnetics, vol. 53, no. 11, pp. 1-4, Nov. 2017. [3] R. Panwar, S. Puthucheri, D. Singh and V. Agarwala, "Design of Ferrite–Graphene-Based Thin Broadband Radar Wave Absorber for Stealth Application," IEEE Transactions on Magnetics, vol. 51, no. 11, pp. 1-4, Nov. 2015.

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/cfe4cd73e62b6f4782155911aa08359e.jpg)
 
Figure 1. Proposed shielding structure (a) Schematic and FSS unit cell, and (b) Fabricated prototype

![](https://assets.underline.io/uploads/markdown_image/1/image/2627666b73e7db1a89ce3751c93f1ac6.jpg)
Figure 2. EMI shielding characteristics (a) Frequency-dependent electromagnetic characteristics of silicon rubber and ferrite substrates, and (b) Transmission performance for different configurations of FSS and substrates.

 Magnetic Substrate Coupled Broadband and Miniaturized Electromagnetic Interference Shielding Structure Using Deep Neural Network

Yttrium Iron Garnet (YIG) is the benchmark material for ferromagnetic resonance; it is up to today the only magnetic material that is being integrated in on-shell electronic devices for its radiofrequency properties that combine high resonance quality factor (104) and frequency tunability. It is the goal of YIG magnonics to utilize this potential for a large variety of applications ranging from beyond complementary metal-oxide semiconductor (CMOS) computation to radiofrequency front-end and back-end analogue signal processing. YIG devices have however a severe drawback that is that they should include thermal stabilization. This is a necessity since YIG has a relatively low Curie temperature (∼550K) which results in a high sensitivity of its magnetization to temperature (∼4 Gauss/Kelvin). Here, we present a novel approach to solve this long standing issue by engineering the magnetic anisotropy in Bi substituted YIG. We evidence for those films that a vanishing effective magnetization yields high thermal stability over very wide temperature range: from 260K to 400K, where the frequency thermal drift is 50 times smaller than that of YIG. These types of properties can be leveraged for new applications where the temperature dependence of YIG is detrimental, while keeping the extremely low damping and insulating character of YIG[1]. For instance, this was key for the excitation of coherent spin-waves using SOT[2]. We anticipate that fabricating a material whose FMR frequency is temperature independent open new opportunities for the field of spintronics. In his seminal article[3] of 1948 Charles Kittel established the Ferromagnetic resonance (FMR) conditions of magnetic samples with arbitrary shapes stressing the importance of correcting the effective magnetic field that sets the Larmor frequency by taking into account the demagnetization field and the effect of uniaxial anisotropy. In today notations the FMR frequency can hence be written for a thin film in the in-plane magnetization configuration as: f=γ*(Ha*(Ha+(MS-Hu)))1/2. Where γ is the gyromagnetic factor, Ha is the applied magnetic field, MS is the saturation magnetization and Hu is the out-of-plane magnetization and the effective magnetization is defined as Meff = MS - Hu. It arises from this equation that the only temperature dependent terms are MS(T) and Hu(T). For most ferromagnetic materials Hu depends on MS with power law dependence defined by Callen and Callen theory[4]. Nonetheless, by engineering the magnetic anisotropy in Bi-YIG films, we were able to balance the saturation magnetization with the uniaxial anisotropy between 260K and 400K. These measurements were done on thin Bi1Y2Fe5O12 films of 21nm grown on (111)-oriented sGGG substrates with Pulsed Laser Deposition. We varied the substrate growth temperature within a range of 100°C and probed the films effective magnetization using FMR: we evidenced a magnetization state transition from in-plane to out-of-plane as the substrate growth temperature decreases from 500°C to 400°C. To ensure the nominal Bi content as well as the magneto-elastic anisotropy were maintained over all the growth temperature range, X-Ray Diffraction measurements were done and evidenced a perfect structural match between thin Bi1Y2IG films. Magneto-Optical Kerr Effect (MOKE) measurements were performed complementary to visualize not only the magnetic domains but also the magnetic anisotropy transition. With room temperature FMR measurements, we obtained the optimal substrate growth temperature range to be 420°C to 430°C: temperatures for which the effective magnetization Meff ≈ 0G. Interestingly, for thin films closest to the compensation, FMR measurements performed at lower temperatures evidenced that effective magnetization stay constant in a range of 140K: suggesting Hu(T)=Ms(T) contradictory with Callen and Callen theory[4]. Finally we measured the frequency thermal drift was about 1mT between 260K and 400K for films close to the compensation, FMR frequency is temperature independent.

**References:**

[1] L. Soumah, N. Beaulieu, L. Qassym, Nature Communication 9, 3355 (2018) [2] M. Evelt, L. Soumah, A.B. Rinkevich, Physical Review Applied 10, 041002 (2018) [3] C. Kittel, Physical Review 73, 155 (1948) [4] H. B. Callen and E. Callen, Journal of Physics and Chemistry of Solids 27, 1271 (1966).

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/e5e1e60b717ac1a92c59f4a6cb88abd2.jpg)
Fig. 1 Polar Magneto-Optical Kerr Effect (MOKE) measurement for a Bi-YIG thin film close to the compensation evidences the presence of magnetic domains.

![](https://assets.underline.io/uploads/markdown_image/1/image/7befc984861421507db377f3d886061b.jpg)
Fig. 2 Variation of the effective magnetization Meff(T) for Bi1YIG thin films within a range of 140K probed by Ferromagnetic Resonance (FMR) measurement.

 Fine tuning of the Magnetic Anisotropy results in Temperature Independent Ferromagnetic Resonance Frequency for Bi-YIG thin films grown by Pulsed Laser Deposition

The typical timescale for material discovery is remarkably long, taking between 10 – 20 years to propagate through idea conception, experimentation, optimisation, and deployment1. Existing global challenges such as the climate emergency require rapid development of new materials on timescales that are incompatible with the traditional linear material discovery approach. High-throughput approaches can be applied as a parallel material discovery methodology, for which a single sample contains a library of different material compositions that can be rapidly screened for the realisation of new magnetic materials. We apply the high-throughput material discovery approach to the new class of materials known as High Entropy Alloys (HEAs). HEAs often consist of five or more constituent elements mixed in near equal quantities. The name is owed to the resulting high configurational entropy of the alloy, which contributes to suppressing the formation of intermetallic phases and stabilises simple FCC and BCC solid solutions. HEAs occupy the unexplored central regions of phase diagrams, and as such, have potential to provide disruptive materials. Already HEAs have exhibited promising soft magnetic properties such as high saturation magnetisation alongside high electrical resistivity, as well as obtaining nanocrystalline textures2,3. However, the number of constituent elements and the various chemical compositions mean HEAs can be found in a very broad parameter space making it difficult to localise the search for useful compositions. This work has used combinatorial sputtering, to fabricate sample libraries containing composition variations of up to 35 at.% on a single Silicon wafer (Figure 1a). We investigate the (CoFeNi)1-x-yAxBy series, where A and B are 3d/4d transition metals (with the addition of Si and Al). Systems studied include CoFeNiMnAl, CoFeNiMnSi, and CoFeNiMnV. We implement a program to model the composition variation across a 3” Silicon wafer with three sputter sources at different incidence angles with concurrent deposition (Figure 1b). The resulting sample library is characterised using a high-throughput magneto-optical Kerr effect (MOKE) magnetometer which has been developed to automatically map the Kerr voltage (which is proportional to the magnetic moment) and coercive field across this sample library as function of HEA composition (Figure 1c). These can then be combined to achieve magnetic property mapping within the chemical composition space (Figure 1d). Following this, XRD, SQUID magnetometry, FMR and magneto-transport are used to generate a complete structural and functional property description of selected compositions. The resulting methodology not only provides a scalable process for magnetic HEA fabrication and characterisation, but can also be paired with data driven machine learning material discovery strategies such as Materials 4.04. The large datasets necessary for machine learning are costly to procure and so far lacking in functional magnetic properties (e.g. Ms, Hc) for which these high-throughput methods could provide.

**References:**

1. Y. Liu, T. Zhao, W. Ju and S. Shi, J. Materiomics 3, 159–177 (2017). 2. Y. Zhang, T. Zuo, Y. Cheng and P. K. Liaw, Sci. Rep. 3, 1455 (2013). 3. T. Zuo et al., Acta Mater. 130, 10–18 (2017). 4. R. Jose and S. Ramakrishna, Appl. Mater. Today 10, 127–132 (2018).

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/646e3021d243cac1b425ee3cb6de2f04.jpg)
Schematic representation of the high-throughput thin film methodology applied to magnetic material discovery, including (a) combinatorial sputtering, (b) modelling the composition across the wafer, (c) a measured Kerr Voltage map, and (d) mapping of magnetic properties as a function of chemical composition.

 High-throughput studies of magnetic High Entropy Alloys

The accelerated development of new soft magnetic materials is urgently required to address the challenges of next generation high frequency rotating electrical machines. We used high-throughput experimental and characterization techniques as well as thermodynamic modeling and machine learning (ML) to develop novel soft magnetic materials based on the Fe-Co-Ni alloy system. The experimental methods include (i) Direct energy deposition based additive manufacturing of compositionally graded bulk materials, (ii) Magnetic thin films with a range of compositions, (iii) Chemical flow synthesis of magnetic nanoparticles, (iv) Spark plasma sintered compositionally graded magnetic alloys. The structural, magnetic, mechanical, and electrical properties were determined. The feasibility of our strategy to develop new soft magnetic alloys has been demonstrated. Introduction: Soft magnetic materials are the key component for various applications such as electric motors and generators, magnetic shielding, robotics, electromagnets, sensors, etc[1, 2]. Such systems utilize a significant fraction of the total energy consumption of the world, hence increasing their efficiency will have a significant impact on the urgent need for energy conservation. High frequency rotating electrical machines require magnetic materials with a balance of magnetic, mechanical, and electrical properties. The traditional methods of developing such materials are costly and time consuming, limiting the opportunity to discover new excellent materials. Hence, high throughput methods are of high demand for accelerated discovery of novel materials and to improve the properties of existing materials. We present experimental and computational high throughput approaches to design new magnetic materials. Methods: Compositionally graded Fe-xNi[3], Fe-xCo, Fe-xNi-yMo, Fe-xCo-yNi samples were prepared by a DED process. High purity spherical Fe, Co, Ni, Mo powders with sizes ranging from 50 to 150 mm were employed. Thin film libraries of Fe-Co-Ni were prepared by co-sputtering of Ni-21Fe and Co targets using magnetron sputtering at a base pressure of 10-8 torr. Thermodynamic modeling was found to be very helpful to identify the heat treatment conditions [4]. The following characterization methods were employed: XRD for crystal structure and phase fraction, EPMA-WBS and EDX for composition, PPMS and magnetic field assisted TGA for magnetic properties, four-point probe for electrical resistivities, and nono(micro)-indentation for mechanical properties. A neural network model was trained on the data to predict the magnetic, mechanical and electric properties. Figure 1 shows the methods and the properties of our interest. Results and discussion: Additive manufacturing was utilized to explore a range of compositions. Iron rich ternary Fe-Co-Ni (Fe&gt; 75%, Co&lt;20%, Ni&lt;25%) alloys exhibit saturation magnetization (Ms), coercivity (Hc), Curie temperature (Tc) and hardness values ranging from 167 to 218.4 emu/g, 19.6 to 10.2 Oe, 534 to 940 °C, 185 to 359 Hv, respectively. Fe-8Co-9.1Ni and Fe-7.4Co-11.3Ni were found to have optimum magnetic and mechanical properties with Ms of 218.4 and 217 emu/g, Hc of 10.2 and 13 Oe, Tc of 765 and 741 °C, hardness of 334 and 359 Hv, respectively. In spark plasma sintered compositionally graded Ni-21Fe-xCo alloys, the largest Ms, Hc, Tc and HV were observed in Co rich samples while the resistivity was found to be larger in Ni rich samples. As an example, the compositions along the diameter of 7.6 cm sample of a co-sputtered thin film is shown in figure 2. Thus, the feasibility to accelerate the development of novel soft magnets has been demonstrated in the Fe-Co-Ni system. Acknowledgement: This work is supported by the AME Programmatic Fund by the Agency for Science, Technology and Research, Singapore under Grant No. A1898b0043.

**References:**

[1] V. Chaudhary, S.A. Mantri, R.V. Ramanujan, R. Banerjee, Progress in Materials Science 114 (2020) 100688. [2] J.M. Silveyra, E. Ferrara, D.L. Huber, T.C. Monson, Science 362 (2018) eaao0195. [3] V. Chaudhary, N.M. Sai Kiran Kumar Yadav, S.A. Mantri, S. Dasari, A. Jagetia, R.V. Ramanujan, R. Banerjee, Journal of Alloys and Compounds 823 (2020) 153817. [4] S. Dasari, V. Chaudhary, B. Gwalani, A. Jagetia, V. Soni, S. Gorsse, R.V. Ramanujan, R. Banerjee, Materialia 12 (2020) 100755.

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/920fbe1f51f6d92c3fa7416eceac2a0c.jpg)
Fig. 1 Soft magnetic materials development using accelerated (a) experimental, (b) computational and (c) characterization techniques, (d) measured/predicted properties

![](https://assets.underline.io/uploads/markdown_image/1/image/6b04517d314f4b720e708815ed8b0002.jpg)
Fig. 2 Change in composition across the diameter of circular graded thin film

 Accelerated development of soft magnetic alloys

Compositionally graded materials are attractive since the properties can be optimized according to the spatial requirement. We report the use of spark plasma sintering for the processing of compositionally graded magnetic materials. Magnetic, mechanical and electrical properties were screened using high throughput characterization techniques. A significant change in all these properties, e.g., saturation magnetization (Ms) ranging from 57 to 224 emu/g, Curie temperature (Tc) ranging from 358 to 1116 °C, and Vickers hardness ranging from 91 to 195 HV across the length of a single cylindrical sample was observed. Introduction: Elemental Fe, Co, Ni exhibit ferromagnetic behavior, and their alloys such as Fe-Co-2V (hiperco) and Ni-21Fe (permalloy) are well established commercially available soft magnetic materials. Fe-Co-2V exhibits an attractive combination of high saturation magnetization, high Curie temperature, high yield strength, and is very useful for high temperature magnetic applications. On the other hand, Permalloys exhibit low coercivity, high permeability, and moderate saturation magnetizations and are of high interest for transformers, electric motors and other electromagnetic devices. Magnetic components of a power electronic convertors functioning at high frequency can be &gt;50 % of total weight of the system[1]. It has been demonstrated that graded properties of soft magnetic materials are beneficial for next generation power convertors/generators[2, 3]. Therefore, developing magnetic components with spatially optimized properties is of high interest[1]. This work demonstrates the use of spark plasma sintering (SPS) for the accelerated preparation of a compositionally graded magnetic materials. Fe/Co/Ni/NiFe/FeCoV have been prepared using gas atomized spherical powders of Fe, Co, Ni, Ni-21Fe and Fe-Co-2V. This graded SPS technique produces a dense bulk metallic sample with discrete changes in composition, resulting in tunable, spatially optimized structural and functional properties. Experiment: Gas atomized prealloyed Fe-Co-2V and Ni-21Fe, and elemental Co, Fe and Ni, from Tosoh SMD Inc. were used as starting powders. The purity of all the powders is &gt;99.9%. The 5 compositions were loaded layer by layer in a graphite die. Sintering of compositionally graded Fe/Co/Ni/NiFe/FeCoV soft magnetic materials was performed using a Spark Plasma Sintering (SPS) equipment (Fuji Electronic Industrial, SPS-211LX) at a vacuum level below 8 Pa. The sample was vertically cut in two sections, one section was annealed at 800 °C for 4 h. The crystal structure of graded samples was determined by X-ray diffraction (XRD) using a Bruker D8 discover diffractometer (CuKa radiation). Elemental mapping was performed by an energy dispersive X-ray spectrometer attached to the SEM. The magnetic properties were measured using a physical property measurement system (PPMS) (EverCool-II, Quantum Design), equipped with a vibrating sample magnetometer. The microhardness of the graded samples was measured by a Vickers hardness tester (Future-Tech). Results and conclusion: Saturation magnetization (Ms), Curie temperature (Tc) and Vickers hardness (HV) of a typical compositionally graded Fe-Co-Ni-NiFe-FeCoV sample are shown in figure 1. Both ends (Fe and FeCoV) of the sample exhibit Ms of &gt;200 emu/g, while the central region, which is Ni, exhibits the minimum Ms of ~57 emu/g. The Curie temperature of the compositionally graded sample measured by employing a permanent magnet placed near the TGA pan. The Co region exhibits the highest Tc, with the Tc following the sequence of Co &gt; FeCoV &gt; Fe &gt; NiFe &gt; Ni. These Ms and Tc values closely matched with the values reported in earlier references. A significant change in mechanical properties across the sample was observed with a maximum Vickers hardness of 195 HV in the FeCoV region. Thus, SPS processed samples with spatially varying compositions were developed, relevant to the urgent need for magnetic components with spatially optimized properties. Acknowledgement: This work is supported by the AME Programmatic Fund by the Agency for Science, Technology and Research, Singapore under Grant No. A1898b0043.

**References:**

[1] D. Rodriguez-Sotelo, M.A. Rodriguez-Licea, A.G. Soriano-Sanchez, A. Espinosa-Calderon, F.J. Perez-Pinal, IEEE Access 8 56238-56252, (2020). [2] V. Chaudhary, S.A. Mantri, R.V. Ramanujan, R. Banerjee, Progress in Materials Science 114, 100688 (2020). [3] K. Byerly, P.R. Ohodnicki, S.R. Moon, A.M. Leary, V. Keylin, M.E. McHenry, S. Simizu, R. Beddingfield, Y. Yu, G. Feichter, R. Noebe, R. Bowman, S. Bhattacharya, JOM 70, 879-891 (2018).

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/b66589666c74d5193a7c7b9a8854133b.jpg)
Fig. 1 Change in saturation magnetization (Ms), Curie temperature (Tc) and Vickers hardness (HV) across the length of a compositionally graded cylindrical sample

 Compositionally graded Fe/Co/Ni/NiFe/FeCoV soft magnetic materials

Alloy systems of the past tend towards single principal component alloy systems. In present research, the stoichiometry of alloy systems with large number of alloying components (&gt;4) can be adjusted to stabilise near-ideal solid solution phases (termed High-entropy alloys, HEAs due to their high configurational entropy), or to possess some mixture of ordered intermetallic phases and the near-ideal solid solution phase (termed Complex Concentrated Alloys, CCAs). HEA and CCA possess many design degree-of-freedoms. Precise control over stoichiometry and thermomechanical history lead to exciting magnetic properties; e.g. through the formation of nanocomposite microstructures [1] in CCAs can achieve good magnetic softness by allowing exchange averaging of the magnetic anisotropy. A major challenge to general application is the vast compositional design space; if ignoring stoichiometry: restricting the number of elements to 1) 3d transition metals, and 2) Si and B, there are 10C5 = 792 [RRR1] possible equimolar combinations. Millions more are possible if stoichiometry is relaxed. Appropriate strategies to filter out good candidate compositional families for study are thus essential. Alloy design can be generalised to two approaches: data-driven discovery (DDD) and functionality-driven discovery (FDD). For DDD experimental data obtained from exploratory and characterisation frameworks are extrapolated for design and prediction; in FDD search and optimisation methods are used to achieve targeted properties. Prediction models (e.g. ab-initio) are used in conjunction with FDD to determine microscopic parameters that govern material properties. FDD is useful in exploring materials with large numbers of degree-of-freedoms such as HEA/CCA systems as extrapolation from experimental data is not a good predictor of quantum mechanical behaviour (i.e. electronic structure interactions). Ab-initio models can calculate potential properties of new compositions, iterating over shedloads[RRR2] of stoichiometries using search and optimisation methods. Although computationally efficient, significant computational power is still required. Applying reasonably accurate low-fidelity pre-filters derived from quantum principles reduces the computational resources required. This work uses two such low-fidelity pre-filters using a parametric semi-empirical approach for determining the saturation magnetisation and mean Curie temperature of HEA/CCA systems. The Curie temperature can be expressed as a function alloyed system (i.e. the number of impurity atoms). At the electronic level these changes are due to 1) Variations of the local moments, and/or 2) Changes in interaction strength between moments represented by the molecular field coefficient (dependent on interatomic distances and local environment effects) [2]. Since the uniaxial pressure experienced by the valence electrons upon interatomic distance changes leads to electronic density compression [3], the rectangular band model [4] can be used to explore the onset of ferromagnetism by evaluating the bulk modulus, B with respect to the Stoner criterion. From this, the ratio B/EF (where the Fermi energy EF is approximated by the Mulliken electronegativity, XM) is used to predict the Curie temperature in HEAs and CCAs (cf. Fig. 1). The predictions are validated through synthesised as-cast HEA/CCA compositions that are characterised on a SQUID-VSM and are found to be dependent on the amount of antiferromagnetic Cr alloying in the system. The Bethe-Slater curve forms part of the early understanding of magnetism and the electronic-structure-driven magnetic ordering of 3d transition metals. Slater Rules [5] are used to evaluate the 4s:3d orbital radius ratio to reconstruct the Bethe-Slater curve (cf. Fig. 2a) using synthesised HEA/CCA compositions from literature, predictions, and our own experiments. Deviations in the 3d Bethe-Slater curve due to electronic interactions lead to curve shifts and model multiplicities, making predictions non-trivial. Orbital radius changes and valence symmetry-breaking are thus evaluated in a parametric function derived from rule-of-mixtures XM and valence electron values as a supplement to the modified Beth-Slater curve. Plotted using the same dataset, this function in Fig. 2b shows good parameterisation of MS with no model multiplicities making it suitable for design predictions. Both models show reasonable accuracies when compared to the experimental data and data from literature. One benefit of both models is that they are derived from common parameters (Bulk modulus, Mulliken electronegativity, atomic number, and d-electron count). This means that the predictive calculations can be obtained by using the rule-of-averages to determine the weighted mean of these parameters for various alloy systems. In order to further test the validity of these models, the models are used with an evolutionary algorithm to design a HEA/CCA with targeted Curie temperature and saturation magnetisation. The results of this prediction are synthesised and characterised through x-ray diffraction and SQUID-VSM and will be presented here.

**References:**

[1] N.A. Morley, B. Lim, J. Xi, A. Quintana-Nedelcos, Z. Leong, Sci. Rep., Vol. 10, 14506 (2020). [2] J.P. Gavigan, D. Givord, H.S. Li, , in: Concert. Eur. Action Magn. CEAM, p.163–173 (1989). [3] S.A. Serebrinsky, J.L. Gervasoni, J.P. Abriata, V.H. Ponce, J. Mater. Sci., Vol. 33, 167–171 (1998) [4] D.G. Pettifor, Bonding and structure of molecules and solids (1995). [5] J.C. Slater, Phys. Rev., Vol 36, p.57–64 (1930).

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/32b4c926a5a2171a4d869a6d4bfa2874.jpg)
Fig. 1. Plot of a dimensionless number, B/XM against the Curie temperature for experimental values as well as values taken from literature. The lines represent a guide to the eye for the various amounts of Cr addition to the composition.

![](https://assets.underline.io/uploads/markdown_image/1/image/94cdfef7de161c197ef79ff1235b2407.jpg)
Fig. 2. Saturation magnetisation plotted against the 4s:3d ratio calculated using Slater Rules. *New results presented here **Experimental results from literature, ***Simulated results from literature.

 Parametric Semi-Empirical Design of Magnetic Properties in Complex Concentrated Alloys: Curie Temperature and Saturation Magnetisation

Accelerated materials development has gained interest in the research community as this approach can generate large amount of data in short period of time, and models built from this data can optimize the properties of the material for specific engineering applications. One such material is soft magnetic material which are widely used in electric motors, generators, transformers, inductors, etc. To model and optimize the properties, machine learning can be deployed. We present preliminary findings of the data mining of the literature for the Fe-Co-Ni alloy system. The processing conditions, properties, and compositions were obtained from the literature. The use of a machine learning tool to predict the properties based on composition, processing parameters, and structural information is described. Introduction: Globally, in the past two decades, total energy consumption has increased exponentially and electric motors account for a significant fraction of the total power consumption[1]. Soft magnetic materials play an important role in energy conversion and are widely used in electric motors, transformer cores, electric vehicles, inductors, etc [2]. However, commercially available soft magnetic materials do not have the optimum balance of properties needed in high frequency electric motors operating in extreme environments. Hence, machine learning based magnetic property prediction, optimization and design of soft magnetic materials has gained attention [3-5]. We build a Fe-Co-Ni based metallic soft magnetic materials database from literature and use a neural network algorithm [6, 7] to optimize properties. Database and Machine Learning: A database of 891 data entries comprising magnetic, mechanical, and electrical properties of various composition of Fe-Co-Ni alloys processed at various conditions was compiled from relevant literature. To curate this database, 55 references were compiled, including standard books, handbooks, and relevant publications. The atomic % of other elements in the alloys was restricted to maximum of 10%. Further, the database was preprocessed to convert the weight % to atomic % of the elements, to convert alphanumeric type of material and processing condition entries to numerical, and to calculate grain size from coercivity and yield strength of the alloys. This preprocessed database was used to train the machine learning tool with input variables: (a) composition, (b) type of material (categorical), (c) processing parameters like treatment temperature and time, cooling rate, % reduction in area (% r.a.) either by cold rolling or hot rolling, reheating temperature and time, etc., (d) grain size, and (e) volume fraction of phases (FCC, BCC, HCP). The output properties were saturation magnetization, log(coercivity), log(maximum permeability), log(electrical resistivity), yield strength, tensile strength, log(elongation), Vickers hardness, and cost of the material. The quality of fit, coefficient of determination (R2) values was determined through cross-validation. Using the trained model, properties of various compositions were predicted. Results and Discussion: Figure 1 shows the R2 values from cross-validation and the uncertainty in prediction for the output parameters. The algorithm was able to predict the properties in good agreement with the literature data. However, the prediction for electrical resistivity and elongation were not as good as that of other properties. This can be attributed to the lack of the data and complex behavior of the properties with the input parameters. The trained model was used to predict the properties of various Fe-Co-Ni alloys with the following conditions: cold rolled sheets with 95% r.a., annealed at 1000oC for 2 hours, and furnace cooled. The predicted saturation magnetization for the above specified conditions reveals a complex interplay between the properties, hence there it could be possible to design a soft magnetic material with improved mechanical strength. Conclusion: Using this robust neural network model and data mining, preliminary results relevant to the development of a Fe-Co-Ni based soft magnetic material with optimal combination of properties could be obtained. Acknowledgement: This work is supported by the AME Programmatic Fund by the Agency for Science, Technology and Research, Singapore under Grant No. A1898b0043, and the Royal Society.

**References:**

[1] R. Saidur, Renewable and Sustainable Energy Reviews, vol. 14, no. 3, p. 877-898 (2010) [2] V. Chaudhary, S.A. Mantri, R.V. Ramanujan, et. al., Progress in Materials Science, vol. 114, p. 100688 (2020) [3] Y. Wang, Y. Tian, T. Kirk, et. al., Acta Materialia, vol. 194, p. 144-155 (2020) [4] J. Nelson, and S. Sanvito, Physical Review Materials, vol. 3, no. 10, p. 104405 (2019) [5] J. Konieczny, L. Dobrzañski, B. Tomiczek, et. al., Archives of Materials Science, vol. 106, p. 106 (2008) [6] B.D. Conduit, N.G. Jones, H.J. Stone, et. al., Scripta Materialia, vol. 146, p. 82-86 (2018) [7] B. Conduit, N.G. Jones, H.J. Stone, et. al., Materials & Design, vol. 131, p. 358-365 (2017)

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/0143b4b4607cd37cafeda7e308e49281.jpg)
Figure 1: Coefficient of determination (R2) values and the uncertainty in prediction of the properties, i.e., saturation magnetization (Ms), coercivity (Hc), Curie temperature (Tc), maximum permeability (m), tensile strength (T.S.), yield strength (Y.S.), elongation (El.), Vickers hardness (Hv), and cost (from left to right).

 Data Mining based development of Fe-Co-Ni soft magnetic material

There has been renewed interest in recent years to accelerate materials development beyond traditional approaches. A combination of both machine learning and high-throughput experiments can be used for predicting material properties and designing compositions beyond what is currently known. Herein we report the use of a computational tool, Alchemite™, to mine for important processing parameters of Fe-Si-Al alloys from a data set curated from published experimental results. The preliminary calculations suggest that this is a viable method to develop new Fe-Si based alloys. Introduction: Silicon steels (Fe-Si) are an important class of materials as they are widely used in applications like electric motors and transformers. Currently, 3.2 wt.% Si is widely used due to (i) good balance of electrical and magnetic properties, and (ii) ductility requirements in traditional processing methods.1 Recent trends have pointed to higher Si content of up to 10 wt.%, with an addition of small amounts of aluminum (Al) to improve material performance. Due to the small literature, it is imperative that we use machine learning to explore higher Si compositions and their processing methods to obtain a good balance of soft magnetic properties, electrical properties and mechanical properties. Data Set and Machine Learning Model: A typical machine learning model comprises three parts: training data, attributes for describing the material, and the machine learning algorithm to map the attributes to properties. The Fe-Si-Al data set was compiled of reported experiment values from 40 references (IEEE Ferromagnetism Handbook2 and 39 journal references, published between 2000 – 2020), resulting in a total of 572 entries. The attributes that were selected as input parameters to describe the material were (i) Composition, (ii) Type of material – like ingots, sheets or powder etc., (iii) Final thickness of the material, heat treatment conditions (iv) Treatment temperature, (v) Treatment time and (vi) Cooling rate, (vii) Frequency and (viii) Magnetic flux density. The output parameters comprised various soft magnetic, electrical and mechanical properties. Alchemite™ is a machine learning technology commercialized for materials and industrial chemicals design through Intellegens.3 The artificial neural network trains from the data set provided and predicts the selected material properties. The coefficient of determination (R2) values was obtained to determine the consistency of the data and predictions. Results and Discussion: Figure 1 shows the R2 values for the input parameters. The machine learning tool showed good predictions for most properties, although mechanical properties were not as well predicted (R2&lt;0.8). This is expected as there is a gap in the reported literature. More work will be done to address this gap, by means of improving the data set through theoretical calculations and high-throughput experimental methods. We studied the machine learning model parameters to reveal the most important variables for each output parameter: The frequency of operation affects total core loss greatly, like observations in alternating current (AC) applications – core loss is higher at increased operating frequency. To mitigate such effects, electrical resistivity must be increased – by controlling material thickness, grain size, and composition. Next, the type of material strongly affects the coercivity. The coercivity is affected by the grain size and composition, and this is in turn is affected by the method of fabrication and subsequent processing. The tensile strength is sensitive to the cooling rate because it affects the formation of different phases (A2, B2, and D03). It is reported that the ordering of phases (in B2 and D03) can interact with dislocations, resulting in a strengthening effect which decreases ductility.1 The phases formed also depend on the Si content, as reflected by the correlation between yield strength and Si content. The above highlights how the various input attributes are interrelated in affecting the output properties, and physical insights this can deliver. This underpins the main difficulty in traditional trial and error experiments and the mammoth task to undertake if it is to be done manually. Conclusions: The results show that the use of Alchemite™ for the development of Fe-Si-Al alloys are robust and promising, as it can capture property-processing relationships and highlight gaps in the data set. More work will be focused on improving the data set and high-throughput experiments performed to compare the results obtained with the predicted values. This will enable us to add to knowledge on achieving optimized and balanced properties in Fe-Si based alloys. This work is supported by the AME Programmatic Fund by the Agency for Science, Technology and Research, Singapore under Grant No. A1898b0043 and the Royal Society.

**References:**

[1] G. Ouyang, X. Chen, and Y. Liang, Journal of Magnetism and Magnetic Materials, Vol. 481, p.234–250 (2019) [2] RM Bozorth, Ferromagnetism. (IEEE Press, 1993) [3] https://intellegens.ai/ (accessed January 2021)

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/55c12021610e152d25b771cea3e2a50e.jpg)
Fig. 1: Coefficient of determination values of output properties from machine learning predictions. The properties from left to right are: W – Total core loss, μ – Permeability, Hc – Coercivity, Ms – Saturation magnetisation, Tc – Curie temperature, ρ – Electrical resistivity, T.S – Tensile strength, Y.S. – Yield strength, El. – Elongation, Hv – Hardness and Cost.

 Data Mining Approach to Development of Iron-Silicon Soft Magnetic Alloys

1Introduction Manufacturing of electrical machines, in general, involves cutting of electrical steel sheets to appropriate sizes. Different cutting techniques such as punching, electrical discharge machining (EDM) and laser cutting are popular. These methods cause damage inside the material close to the cutting edge, which results in deterioration of magnetic properties [1]. Existing studies are based on measurement result by Epstein frame, Ring sample and single-sheet tester. However, these methods do have problems, causing error when analyzing influence of manufacture on material properties[2]. For example, the such non-ideal factors as overlapping joint causes significant eddy current loss especially at high frequency, which may make effect of manufacture effect negligible. In order to obtain more accurate measurement results, this paper aims to propose a new method to clarify the cutting effect on magnetic properties of silicon steel, eliminating those non-ideal factors, and to find out mechanism behind the deterioration of magnetic properties based on macroscopic magnetic measurement and microstructure observation. 2Effect of Cutting Methods on Magnetic Properties of Electrical Steel 2.1Macroscopic Magnetic Measurement The effect of cutting on the electrical steel sheets is analyzed by magnetic measurement on steel strips of different widths. The magnetic test samples were 35WW270 electrical steel lamination (300 mm×30 mm×0.35 mm). Half of the samples were cut along rolling direction while the other half were in the transverse direction. Then the samples have been cut along their length into different width by shearing, EDM, laser cutting, and glued up with polyimide tapes. Magnetic measurements are usually performed on electrical steel samples with Epstein frame. Nevertheless, the Epstein frame has some intrinsic weaknesses. One of them is that the flux density and specific total loss is uneven near the corners. The Epstein frame can be divided as uniform area, transitional area and overlapping area. In order to eliminate the impact of overlapping area and transitional area, two Epstein frames were designed as Fig 1, E(25), 25cm (standard) and E(12.5), 12.5cm (non-standard) Epstein frames, which have identical corner areas and transitional areas. The only difference in size is that the uniform area of E(12.5) is shorter than E(25). The same steel sheet (300mm length *30mm width) were used in both Epstein frames and their total absolute loss was measured. Magnetic measurement are performed following standard IEC: 604004-2. It can be assumed the absolute loss of corner and affected area of E(25) and E(12.5) is equal because of the identical corners and affected area. Then the only difference in absolute power loss between the two Epstein frames is due to the difference in absolute loss of the different uniform areas. The magnetic properties of the samples were measured at 50-800 Hz, and the results at 50 Hz were shown in Fig 2. When B = 1 T and frequency is 50 Hz, compared with EDM, the iron loss of samples with shearing increased by 5.6% and that with laser cutting increased by 32.6%. It can be found that the difference decreased at a higher frequency.The iron loss can be divided into static loss and dynamic loss. With the frequency increasing, dynamic loss takes more share of total iron loss[3]. Therefore, it can be assumed that the effect of cutting stresses on static loss is greater than that on dynamic loss. 2.1.1Microstructure Observation The steel samples were prepared by grinding on 400, 800, and 1200 grit SiC papers and polishing using 3μm diamond paste to obtain a mirror-like surface. Specimen microstructure was examined by scanning electron microscopy (SEM). Fig 2 shows the microstructure of the samples with different cutting methods observed by SEM. There are different contrast effects due to the plastic deformation near the cutting edge. It can be seen that the width of damaged area of EDM is about 20 μm, smaller than that of shearing (83 μm). And it is obvious that there is melting or heat affect zone (108 μm) in the immediate proximity of the cut edge due to laser cutting procedure, which leads to the worst magnetic properties of samples with laser cutting. 3Conclusion This paper has compared and analyzed the effect of different cutting methods on the magnetic properties of electrical steel by double Epstein frames. It is found that EDM has the least influence on the sample, followed by shearing, and laser cutting has the greatest influence. The laser cutting effect is the worst, that is becacuse that there is no cooling in the cutting process, while EDM has coolant flow. The microstructure of the samples was observed by SEM. The results show that the macro measurement results are consistent with the micro observation results.

**References:**

&lt;!--[if supportFields]&gt;&lt;spanlang=EN-US&gt;&lt;span style='mso-element:field-begin'&gt;&lt;/span&gt;&lt;spanstyle='mso-spacerun:yes'&gt; &lt;/span&gt;ADDIN EN.REFLIST &lt;span style='mso-element:field-separator'&gt;&lt;/span&gt;&lt;/span&gt;&lt;![endif]--&gt;[1] H. Naumoski, "Investigation of the influence of different cutting procedures on the global and local magnetic properties of non-oriented electrical steel," Journal of Magnetism and Magnetic Materials, vol. 392, pp. 126-133, 2015. [2] Z. Xiaoyang, "Study on Measurement and Simulation Method of Magnetic Properties of Electrical Steel Sheets Considering Influence of Cutting Method," Shenyang University of Technology, 2016. [3] W. D. Chen J, Cheng S, "Modeling of Temperature Effects on Magnetic Property of Nonoriented Silicon Steel Lamination," IEEE Transactions on Magnetics, vol. 51, no. 11, pp. 1-4, 2015. &lt;!--[if supportFields]&gt;&lt;span lang=EN-US style='font-size:12.0pt;font-family:"Times New Roman",serif;mso-fareast-font-family:宋体;mso-bidi-theme-font:minor-bidi;mso-ansi-language:EN-US;mso-fareast-language:ZH-CN;mso-bidi-language:AR-SA'&gt;&lt;spanstyle='mso-element:field-end'&gt;&lt;/span&gt;&lt;/span&gt;&lt;![endif]--&gt;

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/4befe7ab34ce100a7257c4a4004b0ad8.jpg)

![](https://assets.underline.io/uploads/markdown_image/1/image/a40e5a0db3f0e78ca813dc19a706c8a2.jpg)

 Effect of Cutting Methods on Magnetic Properties of Electrical Steel

CoFeB is one of the most extensivity studied magnetic materials due to its atypical properties such as structurally smooth growth [1], soft magnetic properties [2], high spin polarization [3], and very low Gilbert damping [4], which makes it especially suitable for magnetic tunnel junction devices [5]. Numerous researches have proclaimed that thermal annealing is a decisive measure for a high performance of CoFeB based magnetic tunnel junction devices. Therefore, it is of utmost importance to understand the influence of thermal annealing on this alloy. In this work, the crystallisation of Co-Fe-B alloys triggered by thermal annealing was investigated by X-ray diffraction techniques and scanning electron microscopy, as well as spectroscopic ellipsometry and magneto-optical Kerr effect spectroscopy for annealing temperatures ranging from 300°C to 600°C. The transformation of ~100 nm thick CoxFe(80-x)B20 films from amorphous to polycrystalline was revealed by the sharpening of spectral features observed in optical and magneto-optical dielectric function spectra. These features are ascribed to direct inter-band transitions due to the hybridization of p- and d orbitals of Co-Fe alloys in bcc crystalline phase. Cross-sectional scanning electron microscopy of Co-Fe-B layers annealed at 600°C further confirmed that in the case of a Pt capping layer, the crystallization starts at the interface of CoFeB and Pt, and expands for a thickness, which is consistent with the vertical coherence length of the crystallites determined from XRD. The influence of B on the dielectric function was assessed both experimentally and by optical modelling. By analysing the Drude component of the optical dielectric function, a consistent trend between the charge carrier scattering time/resistivity and the annealing temperature was observed, in agreement with the electrical investigations by means of the four-point-probe method. The result obtained from this study were recently published in ref. [6].

**References:**

1. D. D. Djayaprawira, K. Tsunekawa, M. Nagai, H. Maehara, S. Yamagata, N. Watanabe, S. Yuasa, Y. Suzuki, and K. Ando, Appl. Phys. Lett. 86, 92502 (2005). 2. H. Hauser and R. Grössinger, J. Appl. Phys. 85, 5133 (1999). 3. S. X. Huang, T. Y. Chen, and C. L. Chien, Appl. Phys. Lett. 92, 242509 (2008). 4. C. Bilzer, T. Devolder, J.-V. Von Kim, G. Counil, C. Chappert, S. Cardoso, and P. P. Freitas, J. Appl. Phys. 100, 053903 (2006). 5. T. Devolder, J.-V. Kim, L. Nistor, R. Sousa, B. Rodmacq, and B. Diény, J. Appl. Phys. 120, 183902 (2016). 6. A. Sharma, M. A. Hoffmann, P. Matthes, O. Hellwig, C. Kowol, S. E. Schulz, D. R. T. Zahn, and G. Salvan, Phys. Rev. B 101, 054438 (2020).

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/c0dc6dc54b8645942697748105c792f5.jpg)
The polar Kerr effect measured polarisation rotation (θK) spectra for a Pt capped 100 nm thick Co40Fe40B20 film before and after annealing at various temperatures.

![](https://assets.underline.io/uploads/markdown_image/1/image/c008b8812afe925d41bc890b4fc5cea1.jpg)
SEM micrograph collage of substrate/ Co40Fe40B20/ Pt before and after annealing at 600°C, recorded in a FIB trench at 36° stage tilt.

The 2D Magnetization Process in HGO Fe-Si Sheets

Downloads

Next from INTERMAG 2021

Magnetic Substrate Coupled Broadband and Miniaturized Electromagnetic Interference Shielding Structure Using Deep Neural Network

Similar lecture

Helimagnetism of the Fe5GeTe2 crystal

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

The 2D Magnetization Process in HGO Fe-Si Sheets

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from INTERMAG 2021

Magnetic Substrate Coupled Broadband and Miniaturized Electromagnetic Interference Shielding Structure Using Deep Neural Network

Similar lecture

Helimagnetism of the Fe5GeTe2 crystal

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Downloads