France



In this study, we report room temperature negative magneto capacitance due to 60% Fe doping and positive magneto capacitance due to 60% Ti doping at Mn site in La&lt;sub&gt;0.7&lt;/sub&gt;Pb&lt;sub&gt;0.3&lt;/sub&gt;MnO&lt;sub&gt;3&lt;/sub&gt;. We have studied the structural, mixed valence studies using synchrotron X-ray absorption spectroscopy (XAS), magnetic and magneto capacitance properties of La&lt;sub&gt;0.7&lt;/sub&gt;Pb&lt;sub&gt;0.3&lt;/sub&gt;Mn&lt;sub&gt;0.4&lt;/sub&gt;Fe&lt;sub&gt;0.6&lt;/sub&gt;O&lt;sub&gt;3-δ &lt;/sub&gt;and La&lt;sub&gt;0.7&lt;/sub&gt;Pb&lt;sub&gt;0.3&lt;/sub&gt;Mn&lt;sub&gt;0.4&lt;/sub&gt;Ti&lt;sub&gt;0.6&lt;/sub&gt;O&lt;sub&gt;3-δ&lt;/sub&gt;. La&lt;sub&gt;0.7&lt;/sub&gt;Pb&lt;sub&gt;0.3&lt;/sub&gt;Mn&lt;sub&gt;0.4&lt;/sub&gt;Fe&lt;sub&gt;0.6&lt;/sub&gt;O&lt;sub&gt;3-δ &lt;/sub&gt;and La&lt;sub&gt;0.7&lt;/sub&gt;Pb&lt;sub&gt;0.3&lt;/sub&gt;Mn&lt;sub&gt;0.4&lt;/sub&gt;Ti&lt;sub&gt;0.6&lt;/sub&gt;O&lt;sub&gt;3-δ &lt;/sub&gt;were synthesized by solid state method. The x-ray diffraction (XRD) patterns show (i) successful substitution of Fe and Ti at Mn site and (ii) change in lattice parameters of La&lt;sub&gt;0.7&lt;/sub&gt;Pb&lt;sub&gt;0.3&lt;/sub&gt;Mn&lt;sub&gt;0.4&lt;/sub&gt;Fe&lt;sub&gt;0.6&lt;/sub&gt;O&lt;sub&gt;3-δ&lt;/sub&gt; and La&lt;sub&gt;0.7&lt;/sub&gt;Pb&lt;sub&gt;0.3&lt;/sub&gt;Mn&lt;sub&gt;0.4&lt;/sub&gt;Ti&lt;sub&gt;0.6&lt;/sub&gt;O&lt;sub&gt;3-δ&lt;/sub&gt;.&lt;sub&gt; &lt;/sub&gt; L edge absorption studies using synchrotron XAS show the presence of mixed valence states in these compositions. La&lt;sub&gt;0.7&lt;/sub&gt;Pb&lt;sub&gt;0.3&lt;/sub&gt;Mn&lt;sub&gt;0.4&lt;/sub&gt;Fe&lt;sub&gt;0.6&lt;/sub&gt;O&lt;sub&gt;3-δ &lt;/sub&gt;show maximum&lt;sub&gt; &lt;/sub&gt;~64 % negative magneto capacitance at 500 gauss(G) magnetic field, ~60 % at 5kG magnetic field and -56 % at 10 kG magnetic field at 300 K. La&lt;sub&gt;0.7&lt;/sub&gt;Pb&lt;sub&gt;0.3&lt;/sub&gt;Mn&lt;sub&gt;0.4&lt;/sub&gt;Fe&lt;sub&gt;0.6&lt;/sub&gt;O&lt;sub&gt;3-δ &lt;/sub&gt;show ~51% negative magneto capacitance at 15 kG magnetic field while La&lt;sub&gt;0.7&lt;/sub&gt;Pb&lt;sub&gt;0.3&lt;/sub&gt;Mn&lt;sub&gt;0.4&lt;/sub&gt;Ti&lt;sub&gt;0.6&lt;/sub&gt;O&lt;sub&gt;3-δ &lt;/sub&gt;show ~ 12 % positive magneto capacitance at 15 kG magnetic field. Magneto capacitance values in these compositions at room temperature in accordance with the Maxwell Wagner effect, which was explained by Catalan et al. Interaction between Mn&lt;sup&gt;+3&lt;/sup&gt; and Mn&lt;sup&gt;+4&lt;/sup&gt; via oxygen atom in the lattice originates the local magnetic region in the middle of non magnetic regions which originated negative magneto capacitance while Ti doped samples show positive magneto capacitance due to the presence quantum interference and coulombic interaction effects. Presence of unpaired electrons in the d orbitals of Fe, Mn and Ti dictates the magneto capacitance properties of these materials.

**References:**

1. G. Catalan Appl Phys Lett 88, 102902 (2006)&lt;br/&gt;2. P. Chen , D. Y. Xing and Y. W. Du &lt;i&gt;Phys. Rev. &lt;/i&gt;B 64, 104402 (2001)&lt;br/&gt;3. S. S. Manoharan, B. Singh, R.K. Sahu, et al. J. Phys.: Condens. Matter 20, 235205 (2008)&lt;br/&gt;4. B. Singh Mater. Let. 156, 76 (2015)

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/d8756a16e22f94e3a35f045e93a84cc9.png)
Fig. 1 Magneto capacitance vs frequency plots at different applied Magnetic field for La&lt;sub&gt;0.7&lt;/sub&gt;Pb&lt;sub&gt;0.3&lt;/sub&gt;Mn&lt;sub&gt;0.4&lt;/sub&gt;Fe&lt;sub&gt;0.6&lt;/sub&gt;O&lt;sub&gt;3-δ &lt;/sub&gt;polycrystalline sample.

![](https://assets.underline.io/uploads/markdown_image/1/image/60a94b74ab5d40cb913da542d67977db.png)
Fig. 2 Magneto capacitance vs frequency plots at different applied Magnetic field for La&lt;sub&gt;0.7&lt;/sub&gt;Pb&lt;sub&gt;0.3&lt;/sub&gt;Mn&lt;sub&gt;0.4&lt;/sub&gt;Ti&lt;sub&gt;0.6&lt;/sub&gt;O&lt;sub&gt;3-δ &lt;/sub&gt;polycrystalline sample.

INTERMAG 2021

 Huge room temperature negative magneto capacitance in La0.7Pb0.3Mn0.4Fe0.6O3-δ and positive magneto capacitance in La0.7Pb0.3Mn0.4Ti0.6O3-δ

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.

In this study, we report room temperature negative magneto capacitance due to 60% Fe doping and positive magneto capacitance due to 60% Ti doping at Mn site in La0.7Pb0.3MnO3. We have studied the structural, mixed valence studies using synchrotron X-ray absorption spectroscopy (XAS), magnetic and magneto capacitance properties of La0.7Pb0.3Mn0.4Fe0.6O3-δ and La0.7Pb0.3Mn0.4Ti0.6O3-δ. La0.7Pb0.3Mn0.4Fe0.6O3-δ and La0.7Pb0.3Mn0.4Ti0.6O3-δ were synthesized by solid state method. The x-ray diffraction (XRD) patterns show (i) successful substitution of Fe and Ti at Mn site and (ii) change in lattice parameters of La0.7Pb0.3Mn0.4Fe0.6O3-δ and La0.7Pb0.3Mn0.4Ti0.6O3-δ. L edge absorption studies using synchrotron XAS show the presence of mixed valence states in these compositions. La0.7Pb0.3Mn0.4Fe0.6O3-δ show maximum ~64 % negative magneto capacitance at 500 gauss(G) magnetic field, ~60 % at 5kG magnetic field and -56 % at 10 kG magnetic field at 300 K. La0.7Pb0.3Mn0.4Fe0.6O3-δ show ~51% negative magneto capacitance at 15 kG magnetic field while La0.7Pb0.3Mn0.4Ti0.6O3-δ show ~ 12 % positive magneto capacitance at 15 kG magnetic field. Magneto capacitance values in these compositions at room temperature in accordance with the Maxwell Wagner effect, which was explained by Catalan et al. Interaction between Mn+3 and Mn+4 via oxygen atom in the lattice originates the local magnetic region in the middle of non magnetic regions which originated negative magneto capacitance while Ti doped samples show positive magneto capacitance due to the presence quantum interference and coulombic interaction effects. Presence of unpaired electrons in the d orbitals of Fe, Mn and Ti dictates the magneto capacitance properties of these materials.

**References:**

1. G. Catalan Appl Phys Lett 88, 102902 (2006) 2. P. Chen , D. Y. Xing and Y. W. Du Phys. Rev. B 64, 104402 (2001) 3. S. S. Manoharan, B. Singh, R.K. Sahu, et al. J. Phys.: Condens. Matter 20, 235205 (2008) 4. B. Singh Mater. Let. 156, 76 (2015)

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/d8756a16e22f94e3a35f045e93a84cc9.png)
Fig. 1 Magneto capacitance vs frequency plots at different applied Magnetic field for La0.7Pb0.3Mn0.4Fe0.6O3-δ polycrystalline sample.

![](https://assets.underline.io/uploads/markdown_image/1/image/60a94b74ab5d40cb913da542d67977db.png)
Fig. 2 Magneto capacitance vs frequency plots at different applied Magnetic field for La0.7Pb0.3Mn0.4Ti0.6O3-δ polycrystalline sample.

technical paper

In spintronics, one of the long standing questions is why the MgO barrier is almost the only option to achieve a large tunnelling magnetoresistance (TMR) ratio at room temperature (RT) but not as large as the theoretical prediction [1]. This can be due to the spin fluctuation at the ferromagnet/MgO tunnel barrier interfaces and/or spin-independent hopping within the barrier in a magnetic tunnel junction (MTJ) [2]. One of the approaches to overcome these issues is to employ the lattice softening of the ferromagnetic layers in MTJ. Recently a metastable body-centred cubic (bcc) CoxMn100-x (CoMn) ferromagnetic films have been reported to exhibit the enhancement of their magnetic moments with an ideal lattice matching with MhO [3], which can be an ideal candidate for the TMR enhancement in MTJ. MTJ stacks were grown using conventional magnetron sputtering on MgO(001) substrates with the structure of Cr (40)/CoxMn100-x (10)/MgO (2.4)/CoxMn100-x (4)/Co3Fe(1.5)/IrMn (10)/Ru (5) (thickness in nm) [4]. For CoMn, four different compositions of x = 66, 75, 83 and 86 were used, allowing to control the CoMn lattice constants. To promote the crystallisation of the seed and CoMn layers, in-situ annealing was performed after the deposition of the Cr seed layer and CoxMn100-x layers at 700oC and 200oC, respectively. MTJs were then post-annealed at 325oC for their crystallisation after patterned into pillars with Ti/Au electrodes by photolithography. TMR measurements were carried out using the conventional four-terminal setup with elevating temperature, showing two distinctive TMR ratios of 229% and 142% at room temperature for x = 75 and 86, respectively. The corresponding atomic structures were imaged by cross-sectional transmission electron microscopy (TEM, JEOL JEM-2100 Plus) at 600k magnification. Figures 1(a) shows a cross-sectional high-resolution (HR) TEM image of the MTJ with x = 75 showing the larger TMR ratio. Fully epitaxial growth of the entire MTJ is confirmed with some dislocations, of which period is calculated to be (11.4 ± 0.3) nm. This indicates that the Co75Mn25 layer may induce plastic deformation rather than elastic deformation with inducing dislocations at the boundaries between crystals. The x = 86 sample, on the other hand, shows dark black regions at the CoMn/MgO interfaces indicating CoMn/MgO is not fully crystallised within the CoMn layer [see Fig. 1(b)]. These features are found to be induced by dislocations formed at the CoMn/MgO/CoMn interface with the period of (8.9 ± 0.3) nm. These dislocations may be the origin of the interfacial spin fluctuation. In the MgO layer for x = 75, two distinctive crystallographic phases are observed as shown as the regions (i) and (ii), which are crystallised along MgO[001] and [100] directions, respectively, as similarly reported in previous reports [5]. For x = 86, an additional partially crystallised MgO grains are found, which induces spin-independent hopping to reduce the corresponding TMR ratio. By measuring the lattice constants of CoMn and MgO in the HRTEM images, the top CoMn lattice constants are almost constant across MTJs apart from x = 86, confirming the lattice softening of these layers with forming almost strain-free MTJ. These results were compared with ab initio calculations on the lattice stability of the CoMn alloys. Calculations confirm the crystalline deformation stability across a broad compositional range in CoMn, proving a strain-free interface for larger TMR ratios. Further optimisation of the CoMn-based MTJs can achieve &gt; 1,000% TMR ratio at RT.

**References:**

[1] A. Hirohata et al., J. Magn. Magn. Mater., vol. 509, p. 166711 (2020). [2] C. H. Shang et al., Phys. Rev. B, vol. 58, p. R2917 (1998). [3] R.J. Snow et al., J. Magn. Magn. Mater., vol. 419, p. 490 (2016). [4] K. Kunimatsu et al., Appl. Phys. Exp., vol. 13, p. 083007 (2020). [5] G. X. Miao et al., Appl. Phys. Lett., vol. 93, p. 142511 (2008).

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/cd295ebb917f5d44588a534b3c4a60c8.jpg)
Fig. 1 HRTEM images taken on the x = (a) 75 and (b) 86 sample.

 Strain-Free Magnetic Tunnel Junction with Metastable bcc CoxMn100-x(001) Ferromagnetic Layers

Magnetic tunnel junctions (MTJs) are a key element for various spintronic devices such as magnetic random-access memories including novel ones, i.e., brain-morphic devices and magnetic logic circuits. For future applications, room temperature (RT) TMR ratios of 1000% are targeted [1]. However, fabricating MTJs with a TMR ratio much larger than 200% is difficult because the spin-polarized tunneling current strongly depends on the ferromagnet (FM)/oxide interfaces quality. MTJs suffer from imperfect interfaces due to dislocations, formation of unfavorable oxide layers at the electrode/barrier interfaces or oxygen vacancies [2,3]. Especially, the fabrication of high-quality FM electrodes on the oxide barrier is difficult due to their different chemical properties which is challenging for fabrication of sharp interfaces. Such imperfections disturb highly spin polarized electron tunneling and lead to reduced TMR ratios experimentally. In fact, the TMR ratio of well-studied Fe/MgO/Fe merely exceed 200% at RT even though theoretical calculations predicted &gt;1000% TMR ratios [2]. Here, we report giant enhancement in RT-TMR ratios of Fe/MgO/Fe by tuning the growth condition of each layer and optimization of barrier interfaces: maximum 417% at RT and 914% at 5 K in a single-crystal, spin-valve-type Fe/MgO/Fe MTJ [4]. Exchange-spin-valve MTJ structures were fabricated in an ultrahigh vacuum magnetron sputtering system (base pressure: 4×10-7 Pa) with the structure: MgO(001) substrate//Cr (60)/Fe (30 or 50)/CoFe (0 or 2.24)/Mg (0.5)/wedge-shaped MgO (dMgO = 0.5 – 2.5)/Fe (5)/IrMn (10)/Ru (20) (number in brackets in nm). The MgO barrier was deposited using electron-beam evaporation method and a linear shutter for the wedge-shaped barrier. In-situ post-annealing was performed for each layer. After deposition, the wafers were ex-situ annealed in a 0.7 T magnetic field at 200°C and subsequently, the TMR ratio and resistance area product (RA) was measured using current in-plane tunneling (CIPT) method. The TMR ratio is defined as TMR = (RAP – RP)/RP × 100%, where RP (RAP) are the resistance for the parallel (antiparallel) state. Thereafter, the wafers were pattered into micrometer scale MTJs using photo- and EB-lithography. The transport properties were investigated using a standard dc 4-probe method. In Fig. 1 (a) and (b), RA and TMR ratio vs. dMgO for the MTJ with a 30 nm and 50 nm bottom-Fe is shown, respectively, for the patterned MTJs (dc 4-probe). For the MTJ with 30 nm bottom-Fe the unpatterned wafer (CIPT) results are also shown. As expected, a linear relationship between log(RA) and dMgO in a wide dMgO range was observed, suggesting good MgO barrier formation. For dMgO &lt; 1 nm, the MgO barrier may be discontinuous. For dMgO &gt; 1 nm, the TMR ratio increases with dMgO and it finally reaches &gt;400%. The inset of Fig. 1 (b) shows the MR loop of the Fe (50 nm)/MgO/Fe MTJ with a maximum 417% TMR ratio which almost doubles previously reported values of Fe/MgO/Fe. Improved Fe/MgO interfaces and well-controlled (001)-orientation of our Fe/MgO/Fe resulted in this significant enhancement of the TMR ratio. In addition, it shows significant oscillatory behavior with dMgO. The oscillation period is 0.31 nm, which agrees with the previous Fe/MgO/Fe report [2]. The oscillation is clearly observed in both the patterned and unpatterned cases, which excludes possible errors of our measurements. The peak-to-valley difference of the oscillation reaches ~80%, which is ~7 times larger than the previous report [2]. The RA curves also possess an oscillatory component as seen in Fig.1 (a). Furthermore, the TMR ratio was increased to 497% at RT by using a Fe/CoFe/MgO/Fe structure (red square in Fig. 1 and MR loop in inset of Fig. 1(b)). This value is larger than other ones of exchange-spin-valve MTJs such as Co/MgO/Co (410%) [5] and Co2(FeMn)Si/MgO/Co2(FeMn)Si (429%) [6]. In Fig. 2, the temperature dependence of the TMR ratio for an Fe (30 nm)/MgO/Fe MTJ is shown. The TMR ratio is 386% at RT and increases to 914% at 3 K. The respective MR loops are displayed in the inset of Fig. 2. The low temperature (LT) value is around three times larger as compared to other reported values of Fe/MgO/Fe MTJs (250-370%) [2,7]. Our Fe/MgO/Fe MTJ exhibits a giant TMR ratio close to 1,000% at LT, which is often referred as a theoretical value of Fe/MgO/Fe. The study demonstrated that there is more room for improving TMR ratios of Fe/MgO/Fe MTJs by the optimization of each layer and interface. Further improvement in TMR ratios using the technology in the present study can be expected as demonstrated by the CoFe insertion result. This work was partly supported by the ImPACT Program of the Council for Science, Technology and innovation (Cabinet Office, Government of Japan), JSPS KAKENHI Grant No. 16H06332, TIA collaborative research program “Kakehashi”, and JPNP16007, commissioned by the New Energy and Industrial Technology Development Organization (NEDO).

**References:**

[1] D. Ielmini and S. Ambrogio, Nanotechnology, Vol. 31, p. 092001 (2020). [2] S. Yuasa et al., Nat. Mater., Vol. 3, p. 868 (2004). [3] X.-G. Zhang, et al., Phys. Rev. B, Vol. 68, p. 092402 (2003). [4] T. Scheike et al., Appl. Phys. Lett., in press; arXiv: 2011.08739. [5] S. Yuasa et al., Appl. Phys. Lett., Vol. 89, p. 042505 (2006). [6] H.-x. Liu, et al., J. Phys. D: Appl. Phys., Vol. 48, p. 164001 (2015). [7] S. G. Wang, et al., Phys. Rev. B, Vol. 78, p. 180411(R) (2008).

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/fcd71b3caf74c7f35790944728c9ec6d.jpg)
Fig. 1 RA (a) and TMR ratio (b) vs. dMgO for Fe/MgO/Fe MTJs with 30 nm (circle and cross) and 50 nm (diamond) bottom-Fe and Fe(50 nm)/CoFe(2.24 nm)/MgO/Fe (square) MTJ. Inset: MR loops of both MTJs at their maximum TMR ratios.

![](https://assets.underline.io/uploads/markdown_image/1/image/d4a9e77a8c2d894c8d8c4e9022dcd09d.jpg)
Fig. 2 TMR ratio vs. temperature of a Fe(30 nm)/MgO/Fe MTJ. Inset: MR loops at 300 K and 5 K.

 Giant tunnel magnetoresistance up to 417% at room temperature using Fe/MgO/Fe magnetic tunnel junctions

Perpendicular magnetic tunnel junctions (p-MTJs) utilizing tetragonal Mn-based alloy films attracts attention as key devices for magnetic random access memory and THz frequency devices 1. There were many past studies demonstrating those devices, whereas their TMR ratios were much smaller than those predicted by first-principles calculations2. Interlayers between the Mn-based alloy and the barrier effectively enhance those TMR ratio, and the TMR ratio of 60% was reported in MnGa/MgO MTJs with FeCo interlayer3. It is crucial to find new interlayer materials to further enhance the TMR ratio for realizing Mn-based p-MTJs for practical applications. Recently, we have reported in-plane type MTJs using bcc-Co75Mn25 electrodes. Bcc-Co75Mn25 is known for its metastable bcc-phase and for ferromagnetism having high Curie temperature4. Those MTJs showed a high TMR ratio of over 200%, which may be due to highly spin-polarized △1 band in bcc-Co75Mn25 5. Here we demonstrate high TMR ratio in ultrathin MnGa-based p-MTJs with the bcc-Co3Mn alloy interlayer. The result indicates that bcc-CoMn alloy is one of the promising interlayer materials for obtaining high TMR ratio in Mn-based p-MTJs6,7. The samples were prepared using a magnetron sputtering system. The stacking structure of the samples was MgO(001) substrate/Cr(40)/Co55Ga45(30)/ Mn61Ga39(3)/ CoxMn100-x(0.8) /Mg(0.4)/MgO(2)/Fe60Co20B20(1.2)/Ta(3)/Ru(5) (thickness is in nanometers) Here, the CoxMn100-x interlayer was deposited using a co-sputtering technique with Co and Mn targets. All the layers were deposited at room temperature, and the deposition process for the buffer and seed layers was adopted from a previous study6. Micro-fabrication of MTJs was performed by standard ultraviolet photolithography, and Ar ion milling etching. After microfabrication, the MTJs were annealed at 250 °C in a vacuum furnace. TMR measurements were performed by a physical property measurement system (PPMS, Quantum Design) using a standard four-probe method at a bias voltage of 1 mV. Figure 1 shows the TMR curves of the MTJs with/without Co75Mn25 interlayer measured at 300 K. The Co75Mn25 interlayer shows the TMR ratio up of approximately 80%, which is much enhanced as compared with that for MnGa-MTJs, ~ 10 %. Meanwhile, the shape of the TMR curve and the increase of coercivity indicate that Co75Mn25 interlayer strongly and antiferromagnetically couples with perpendicularly-magnetized MnGa layer. Figure 2 shows the composition x dependence of the TMR ratio for the MTJs with 0.8-nm-thick CoxMn100-x interlayer. The TMR ratio for the x=0 MTJ is 70% (145%) at 300 K (10K), which is larger than the value expected from Julliere’s relation with spin-polarization of fcc-Co and CoFe(B). Moreover, the TMR ratio increases by up to 85% (209%) at 300 K(10 K) for the case of x=0.8. Those high TMR ratio would originate from coherent tunneling between highly spin-polarized Δ1 bands in bcc-Co (Co-Mn) and CoFe(B) electrodes, even though the Co-Mn interlayer thickness is very small, 0.8 nm. This may be also consistent with the temperature dependence of resistance or conductance in these MTJs, which was similar to those observed in Fe/MgO/Fe MTJs. Acknowledgement We would like to thank Y. Kondo for his technical assistance. This work was partially supported by JST CREST (No. JPMJCR17J5).

**References:**

[1] S. Mizukami et al., Scripta Materialia 118, 70 (2016). [2] J. Jeong et al., Nat. Commun., 7, 10276 (2016). [3] Q. L. Ma et al., Phys. Rev. Lett. 112, 157202 (2014). [4] R. J. Snow et al., J. Magn. Magn. Mater. 419, 490 (2016). [5] K. Kunimatsu et al., Appl. Phys. Express 13, 083007 (2020). [6] K. Z. Suzuki et al., Sci. Rep. 6 , 30249 (2016). [7] K. Z. Suzuki et al., submitted.

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/cd118f0c5f1c2b5dda5eeec663df1850.jpg)
Figure1 TMR curves measured at 300 K for the MnGa-based p-MTJs with/without Co75Mn25 interlayer.

![](https://assets.underline.io/uploads/markdown_image/1/image/ec9d8f4d534d71d7d5602117fd1c8e27.jpg)
Figure 2 The Co composition dependence of the TMR ratio in MnGa-based p-MTJs with CoxMn100-x interlayer at 300 and 10 K.

 High Tunnel magnetoresistance in ultrathin MnGa-based perpendicular magnetic tunnel junctions utilizing ferromagnetic bcc-CoMn interlayers

Spin-orbitronics, which takes advantage of spin-orbit coupling (SOC), has expanded the research objects of spintronics to nonmagnetic materials. Here, we report the emerging nonlinear spintronic phenomena in the inversion-asymmetric nonmagnetic quantum materials with SOC. For instance, the surface state of three-dimensional topological insulator (TI) owns helical spin textures with the spin and momentum perpendicularly locked. We show the observation of a nonlinear magnetoresistance (called bilinear magneto-electric resistance, BMER) [1] and nonlinear Hall effect [2] in a prototypical TI Bi2Se3, which scale linearly with both the applied electric and magnetic fields. We further reveal that these effects are originated from the conversion of a nonlinear spin current to charge current under the application of an external magnetic field. A close link between the BMER and the spin texture was established in TI surface states, which enables a novel transport probe of spin textures. We further extended the observation of BMER effect to the d-orbital two-dimensional electron gas (2DEG) at a SrTiO3 (STO) (111) surface [3]. The BMER probes a three-fold out-of-plane spin texture, in addition to an in-plane one at the STO (111) surface 2DEG. This novel spin texture is in contrast to the conventional one induced by the Rashba effect. By performing tight-binding supercell calculations, we find that this 3D spin texture is fully described by the confinement effects of the STO t2g conduction band in the (111) plane. We recently reported the observation a sign reversal of nonlinear magnetoresistance with temperature in a semimetal WTe2 [4]. Theoretical calculations revealed the critical role of Fermi surface topology and convexity on the nonlinear magneto-response. These findings open a new branch in spintronics, which discusses the nonlinear transport effects in spin-polarized quantum materials, and is therefore referred as nonlinear spintronics.

**References:**

[1] Pan He, et al. Nature Physics 14, 495 (2018) [2] Pan He, et al. Phys. Rev. Lett. 123, 016801 (2019) [3] Pan He, et al. Phys. Rev. Lett. 120, 266802 (2018) [4] Pan He, et al. Nature commun., 10, 1290 (2019)

**Images:**

 Nonlinear spintronics in quantum materials with inversion symmetry breaking INVITED

Nowadays, our lives are greatly enriched by information society. Meanwhile, energy resource issue is becoming a serious problem, leading to the demand for devices with lower power consumption. When we focus on the memory, the spread of non-volatile memory, represented by the racetrack memory [1] based on current induced domain wall motion (CIDWM), has been longed for. To realize this, our groups have been investigating anti-perovskite Mn4N films as candidate materials for the racetrack memory. They show a small saturation magnetization (MS) of 80 kA m-1 and clear perpendicular magnetic anisotropy (PMA) [2], contributing to high velocity and low threshold current density, respectively. We experimentally demonstrated the domain wall velocity (vDW) of 900 ms-1 at 1.3×1012 Am-2 only by spin-transfer torque (STT) [2]. Note that we achieved this record by rare-earth free material unlikely other reports. Our recent work also pays attention to Mn4-xNixN films because they have compensation point at x ~ 0.15 where vDW takes its maximum value [3]. In Mn3.85Ni0.15N, we recorded vDW of 2000 m/s at 1.16×1012 Am-2, the fastest purely STT-driven CIDWM [4]. Despite of these amazing performances, large parts of magneto-transport properties in Mn4-xNixN films are still unknown although they are deeply linked to the relationship between conduction electrons and localized ones. Interestingly, we confirmed large anomalous Hall effect (AHE) with a relatively large anomalous Hall angle (θAHE = ρAHE/ρxx) of 2% in Mn4N films although they don’t contain any heavy metal with large spin-orbit coupling [2]. There were attempts to discover the origin of such large AHE, however, it has not been proved by now. In this work, we performed AHE and conductivity measurements at temperatures in the range of 5-300 K, and we plotted σAHE against σxx to find out the origin of AHE [6]. Also, we performed anisotropic magnetoresistance (AMR) measurements for the better understanding of spin-dependent scattering in these materials. Mn4-xNixN films (30 nm) were fabricated onto SrTiO3 (STO)(001) substrates by molecular beam epitaxy. Epitaxial growth was confirmed by X-ray diffraction, and the thickness of each layer was examined by X-ray reflectivity. AHE and AMR measurements were performed by physical properties measurement system. Before measurements, film samples were processed into Hall bars with a width of 200 mm by Ar ion milling. During AMR measurements, DC current was set to flow in the [100] azimuth and a magnetic field of 9 T was applied parallel to the plane. Figure 1 shows the σxx dependence of σAHE in Mn4-xNixN/STO in the temperature range of 5-300 K. Note that σxx became larger at lower temperatures, a typical behavior in metals. The region of σxx &lt; 104 Ω-1cm-1 is called “Bad metal regime”, where σAHE is proportional to σxxα (α~1.8) in many 3d transition metals [5]. Mn4-xNixN films (x &gt; 0) also followed this trend, while Mn4N showed rather constant σAHE relative to σxx. On the other hand, the region of σxx &gt; 104 Ω-1cm-1 is called “Good metal regime”, where σAHE did not change so much against σxx. In Mn4-xNixN at small x (x = 0 and 0.05), however, σAHE decreased with increasing σxx, thereby, with decreasing temperature. It’s also remarkable that σAHE remained almost unchanged when x = 0.15 and 0.2. This discrepancy can be explained in the following. Figure 2 shows the Fourier coefficients C2 (cos2θ) and C4 (cos4θ) of AMR curves of Mn4-xNixN/STO. θ is the angle made by the current and magnetization. C2 and C4 components derive from cubic and tetragonal crystalline fields, respectively [6]. Thus, large C4 components at temperatures lower than 100 K at x = 0 and 0.05 originate from tetragonal one although they have cubic structures. Note that C4 component was very small at x = 0.15, and AMR was barely observed at x &gt; 0.2. In 3d transition metals, intrinsic AHE comes from hopping of electrons between electronic states dyz and dzx [7]. However, these states rise under the tetragonal crystalline field due to the resolution of degeneracy, leading to smaller number of electrons related to the origin of AHE. Therefore, reflecting our experimental results, unexpected decrease in σAHE at low temperatures in Mn4-xNixN (x= 0, 0.05) can be explained by the emergence of tetragonal field. On the basis of these discussions, we conclude that large AHE in Mn4-xNixN originates from intrinsic AHE at low temperatures. Although we have not fully revealed the case at room temperature due to a limited number of reports on AHE in “Bad metal regime”, we suggest that it originates from intrinsic AHE and extrinsic AHE such as phonon-scattering. We believe this work opens the interest in the band structure of Mn4-xNixN films.

**References:**

[1] S. Parkin and S.-H. Yang, Nat. Nanotechnol. Vol.10, p.195 (2015). [2] T. Gushi, L. Vila and T. Suemasu et al., Nano. Lett. Vol.19, p.8716 (2019). [3] T. Komori, K. Amemiya and T. Suemasu et al., J. Appl. Phys. Vol.127, p.043903 (2020). [4] S. Ghosh, T. Komori and L. Vila et al., IEEE MMM Virtual Conference, R1-11, Nov. (2020). [5] N. Nagaosa, S. Onoda and N. P. Ong et al., Rev. Mod. Phys. Vol.82, p.1539 (2010). [6] S. Kokado and M. Tsunoda, Mater. Today Proceedings Vol.33, p.1864 (2020). [7] H. Kotani, T. Tanaka, K. Yamada Phys. Rev. B Vol.75, p.184416 (2007).

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/cb1bf1328229c6aeb2d2d6ad95de27fd.jpg)
Fig. 1 σxx dependence of σAHE in Mn4-xNixN at temperatures (5-300 K). A dashed line indicates the border of Good metal and Bad metal regime.

![](https://assets.underline.io/uploads/markdown_image/1/image/185d297e24af6e838eb29c195a960292.jpg)
FIg. 2 Fourier coefficients C2 and C4 of AMR curves of Mn4-xNixN.

 Magneto-transport properties in Mn4-xNixN films with large current induced domain wall mobility and investigation in their large anomalous Hall effect

Various types of Anomalous Hall effects have been widely investigated in recent years [1, 2]. In most of the cases, the predicted or measured Hall signal has been attributed to spontaneous magnetic moment arisen from the dynamics of the current flow [1]. In this work, we aim to investigate the role of anisotropic band dispersion of chiral Weyl fermions in generating a transverse current resulting in a Hall signal. For this purpose, the simplest Weyl semimetal phase consisting of two Weyl nodes with opposite chirality is adopted and the angular dependence of conductance is calculated [3, 4]. Multi-terminal devices with circular geometry as illustrated in Fig. 1 are commonly used to probe the dependence of the signal on transport direction in experiments [1, 3]. Therefore, in the proposed model, we adopt the circular channel geometry and calculate the conductance for arbitrary transport directions. The Hamiltonian of the simplest Weyl system is as follows, H = V0 + Σi hki τ (viσi+wi), where σ’s are Pauli matrixes, v’s are velocities and their sign (τ = ±) carry the chirality of Weyl nodes. We assume symmetric velocities equal to vF = 106 m/s. The dispersion of Weyl fermion can be tilted along all three directions, and the strength of the tilt is denoted by wi. In the current work, two Weyl nodes having opposite chirality are also tilted along opposite directions. The tilt direction is chosen to lie along the high anisotropy axis, which is close to one of the crystal axes in most of the Weyl semimetals reported up to date [e.g., Refs. 3, 5]. The angular dependence of the conductance [3, 4, 5] can be calculated based on transmission probability along an arbitrary direction and integrated over all transverse wave vectors based on the following equation, Gβ = G0 ∫ -(∂f/∂E)dE ∬ dSFS Tφ,γ cosφ cosγ. In the above equation, γ is the angle between Fermi wave vector k and the x-y plane, φ is the azimuthal angle with respect to the x-axis, G0 is the quantum conductance, dSFS is the infinitesimal element of the elliptical Fermi surface per unit variation of the coordinates φ and γ. The anisotropy in crystal symmetry and the bandstructure make the conductance dependent on the angle between the transport direction and the crystal axis. Based on the conductance profile, the voltage difference between the transverse electrodes can be found as follows, Vt = Vt1-Vt2 ∝ (S/ρ)[(Gt1-Gt2)/(L/√2)], where Gt1 = Gβ+(π/4) and Gt2 = Gβ-(π/4). The longitudinal potential difference can be calculated similarly by replacing the respective conductance value and distance between the voltage probes. As shown in Fig. 2 and predicted by the above equation, transverse voltage emerges at some angles which are related to the imbalance of Gt1 and Gt2. The highest Vt occurs at the maximum imbalance, where Gt1 and Gt2 coincide with the lowest and highest conductance points in the conductance profile shown in Fig. 1 (right). On the other hand, the conductance between source and drain is minimum in this direction as seen in Fig. 2. Our predicted results are quite consistent with the experimental results presented recently in literature [1] thus revealing a possible origin of the measured Hall signals in Weyl semimetals in the absence of a magnetic field.

**References:**

[1] Kang, K., Li, T., Sohn, E. et al. Nonlinear anomalous Hall effect in few-layer WTe2. Nat. Mater. 18, 324–328 (2019). [2] Ma, Q., Xu, SY., Shen, H. et al. Observation of the nonlinear Hall effect under time-reversal-symmetric conditions. Nature 565, 337–342 (2019). [3] Qisheng Wang, Can Yesilyurt, Fucai Liu, Zhuo Bin Siu, Kaiming Cai, Dushyant Kumar, Zheng Liu, Mansoor B. A. Jalil, and Hyunsoo Yang Nano Letters 2019 19 (4), 2647-2652 [4] Yesilyurt, C., Siu, Z.B., Tan, S.G. et al. Electrically tunable valley polarization in Weyl semimetals with tilted energy dispersion. Sci Rep 9, 4480 (2019). [5] Soluyanov, A., Gresch, D., Wang, Z. et al. Type-II Weyl semimetals. Nature 527, 495–498 (2015).

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/6d703c64fb6f07d5598af24065e4402f.jpg)
Illustration of the adopted model, i.e., circular multi-terminal circuit (left). Terminal A and B denote source and drain electrodes respectively, and the Hall signal is calculated as the voltage difference at two transverse points t1 and t2. Normalized anisotropic conductance profile of the Weyl system with two opposite chirality Weyl nodes (right). Blue (Red) arrow(s) indicates the conductance between the source and drain (transverse electrodes).

![](https://assets.underline.io/uploads/markdown_image/1/image/d819182570aba9d154a697f1728aa977.jpg)
Angular dependence of anomalous Hall effect, where the Hall voltage Vt is calculated as the voltage difference between the points t1 and t2 (i.e., shown in Fig. 1). The longitudinal voltage VSD is calculated as the voltage difference between points A and B (i.e., shown in Fig. 1)

 Anomalous Hall Effect in Anisotropic Weyl Semimetals

Magnetotransport properties of magnetic materials receive considerable interests, because of they make an important role in basic science and technological applications[1]. Anomalous Hall Effect (AHE), as an important topic in condensed matter, contains extensive physics for their many different mechanisms, such as intrinsic mechanism associated with the Berry curvature in entangled Bloch electronic bands, extrinsic mechanism including skew scattering and side jump scattering[2]. In this study, the magnetotransport properties in antiferromagnetic half-Heusler single crystals of TbPtBi, a magnetic-field-induced topological semimetal with simple band structure, were investigated. We found a non-monotonic dependence of the anomalous Hall resistivity on the magnetic field in a strong magnetic field (B&gt;7T), which comes from the change of band structure caused by the Zeeman-like splitting when an external magnetic field is applied. The experimental results show that credible anomalous Hall resistivity and conductivity reach up to 0.68mΩcm and 125Ω-1cm-1, respectively. A large anomalous Hall angle (AHA) up to 33% is achieved in TbPtBi, which is comparable to a typical ferromagnetic Weyl semimetal Co3Sn2S2[3]. The analysis of the results shows that this should be attributed to topological band around EF and low carrier density.

**References:**

[1] Science advances 4 : eaar7880 (2018). [2] Rev. Mod. Phys. 82 (2), 1539 (2010); Physical Review 95 (5), 1154 (1954); Physica 24, 39 (1958); Physica 21, 877 (1955); Phys. Rev. B 2 (11), 4559 (1970). [3] Nat. Phys. 14, 1125 (2018); Adv. Funct. Mater., 2000830 (2020); Nano Lett. 7476–7481, (2020).

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/72c83298ed580c527c76631a5810d453.jpg)
Figure 1. Dependence of the longitudinal resistivityρxx(a) and Hall resistivity ρxy (b) on magnetic field B at temperatures from 1.8K to 50K. (c) Separation progress of normal Hall resistivity ρxyN and anomalous Hall resistivity ρxyA at 1.8K. The inset shows the normal Hall resistivity obtained by linear fitting for low magnetic field curves at corresponding T. (d) The anomalous Hall resistivity ρxyA after subtracting the fitting normal Hall resistivity ρxyN. The inset is the maximum value of ρxyA at different temperatures.

 Large Anomalous Hall Angle in a Topological Semimetal Candidate TbPtBi

The thermoelectric generation based on the anomalous Nernst effect (ANE) has become one of the promising approaches to realize efficient energy harvesting from waste heat and heat flux sensor due to its several advantages over conventional devices based on the Seebeck effect (SE) 1,2. However, till date, the observed ANE thermopower is very small and a systematic material investigation to realize large anomalous Nernst coefficient is necessary. Recent theoretical and experimental investigations identified a Weyl semimetal as a potential material having giant ANE which can be achieved by a little tuning of Fermi level (EF) near Weyl points3,4,5. However, systematic tuning of EF and investigation of thermoelectric properties based on many films with different compositions are time-consuming. Furthermore, in this process, one losses many intermediate compositions due to the limitation of number of films that can be prepared. In this regard, we introduce an efficient approach to tune EF for Weyl semimetals and demonstrate the approach through a layer-by-layer combinatorial deposition of a Co2MnAl1-xSix (CMAS) thin film with x varied from 0 to 1 on a single substrate. The fabrication procedure is represented schematically in Fig.1 (a). The structural characterization reveals the formation of single-phase CMAS alloys throughout the composition range. Hard X-ray photoemission spectroscopy (HAXPES) directly confirmed a continuous shifting of EF from Co2MnAl to Co2MnSi with a maximum shift of 0.4 eV (see Fig.1. (b)), which is consistent with the theoretical prediction. We also measured the anomalous Ettingshausen effect (AEE), the reciprocal of ANE, for the all x range using a single strip along the composition gradient (see Fig.1. (c)) by means of the lock-in thermography technique6,7. Only one LIT image tells us that large AEE is obtained between x = 0.06 to 0.12, which well agrees with the composition dependence of the ANE signals (see Fig.1. (d)), measured by making many parallelly aligned Hall bars elongated to the orthogonal direction to the composition gradient (see Fig.1. (c)). The findings clearly demonstrate that the AEE measurement using the composition spread film is an effective approach to investigate the composition dependence of ANE and find the highest performance without fabricating many films and performing systematic measurements. These demonstrations will enable the optimization of the composition in Weyl semimetals and exploration of new materials to obtain giant ANE/AEE, which is essential to realize applications based on the transverse thermoelectric effects.

**References:**

[1] Y. Sakuraba et al., Appl. Phys. Express 6, 033003 (2013) [2] W. Zhou et al., Appl. Phys. Express 13, 043001 (2020) [3] K. Sumida et al., Commun. Mater. 1, 89 (2020) [4] J. Noky et al., NPJ Comput. Mater. 6, 77 (2020) [5] Y. Sakuraba et al., Phys. Rev. B 101, 134407 (2020) [6] S. Daimon et al., Nat. Commun. 7, 13754 (2016) [7] T. Seki et al., Appl. Phys. Lett. 112, 152403 (2018)

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/58d6615be315220821aa1a7a138fb225.jpg)
Fig.1 (a) Schematic representation of composition-spread CMAS combinatorial film deposited with wedge shaped layer by layer deposition (b) Valance band HAXPES spectra for CMAS composition-spread film at Co2MnAl (CMA) and Co2MnSi (CMS) side (c) Schematic measurement configuration used for ANE and AEE (d) AAEE/jc and SANE as a function of x. Where AAEE is the amplitude of temperature modulation due to AEE, jc is the charge current density and SANE is anomalous Nernst coefficient. The inset shows observed AAEE image in LIT.

 Efficient tuning of electronic, transport, and thermoelectric properties of Weyl semimetal Co2MnAl1-xSix composition-spread thin film

Lately, Magnetic Random-Access Memory (MRAM) has stimulated significant interests compared to traditional semiconductor memory devices due to its non-volatility combined with expeditious read and write endurance. In the preceding studies, conventional methods that had been investigated widely include spin-orbit torque (SOT)1, spin-transfer torque (STT)2 and current-driven switching. However, magnetization switching in MRAM utilizing such techniques fall short to observe the desired switching and inevitably supplemented by thermal issues, leading to inefficient functionality. Recently, considerable endeavour towards a new paradigm of straintronic devices (STR)3 for efficient magnetization switching is explored. A magnetoelectric STR constitutes of ferromagnetic and piezoelectric order parameters, in which an applied electric field across the piezoelectric order parameter instigate mechanical strain at the heterostructure interface giving rise to the magnetization switching due to the Villari effect. Though both out-of-plane and in-plane magnetization switching are examined in the aforementioned studies to explore the STR devices, magnetization switching in the in-plane direction is obscured on account of thermal instability in smaller element size and multi-domain formation in larger element size. Also, the self-biased STR-MRAM device that extinguishes the external bias magnetic field is still undiscovered. In this work, we investigated the size-dependent in-plane magnetization switching of FeGaB/PMN-PT (ferromagnetic/piezoelectric) heterostructure (Fig. 1(a)) using micromagnetic finite element method (FEM) coupled with electrodynamic and elastic analysis. Three elliptical nanodiscs of FeGaB magnetoelastic material having cross-sectional areas 100×90.9 nm2, 100×66.67 nm2, and 100×50 nm2 are investigated by varying the thickness from 2 nm to 28 nm on a 10×10×0.5 mm3 single-crystal PMN-PT (011) piezoelectric substrate. Initially, minimum thickness (dmin) at pre-stress condition for three elliptical nanodiscs are investigated (Fig. 1(b)) using in-plane induced energy well (△E) scaled by kbT (where kb is Boltzmann constant and T is the temperature in Kelvin). △E/kbT≤70 is an assumed condition for thermal instability to ensure dmin for STR-MRAM devices. To explore the maximum thickness related to self-biasing (dsb), pre-stress equilibrium exchange and demagnetization energies associated with the three elliptical nanodiscs are analyzed against thickness. The larger size magnets make an effort to minimize the energy by aligning the magnetization with the elliptical nanodisc boundary and gives rise to vortex state. On contrary, smaller size magnets minimize the energy by aligning the magnetization along the easy axis (0° or 180°) of the elliptical FeGaB nanomagnets without applying the external magnetic bias field (Fig. 2). This result is pivotal as such self-biasing would give primacy to future STR-MRAM devices by omitting the necessity for an extrinsic magnetic bias field. Although the above analysis is performed to obtain optimum size where STR-MRAM device is stable and self-biased, the maximum thickness (dmax) for in-plane 180° magnetization switching is limited by the critical stress requirement. It is observed that dsb is not necessarily the same as dmax for in-plane 180° magnetization switching. When compressive strain equal or greater than the critical compressive strain is applied from the pre-stress state, where magnetization is along the easy axis, in-plane 90° magnetization switching (along 90° or 270°) is observed. This happens as exchange and stress anisotropy energies are almost negligible and the demagnetization energy rises as the magnetization is switched from the easy to the hard axis. Magnetic precession is used to obtain in-plane 180o switching. When the magnetization component along the x-axis is negative during the precession, removal of external strain causes another 90o switching because the ferromagnet tries to reduce its total energy by gravitating to its nearest easy axis. In-plane fastest 180o magnetization switching time and associated external strain are investigated also. For fixed thickness, compressive strain requirement rises with increasing cross-sectional area. As the thickness is increased, tradeoff between fastest switching time and associated switching energy is observed for the same cross-sectional area. Whereas faster switching time and lesser switching energy are required for larger cross-sectional nanodiscs having the same thickness. This work can be potentially useful in the future STR-MRAM devices that eliminates the necessity of extrinsic magnetic field bias and completes in-plane 180o magnetization switching.

**References:**

1C. Zhang, S. Fukami, H. Sato, F. Matsukura, H. Ohno, Appl. Phys. Lett., 107, 012401 (2015). 2C.F. Pai, L. Liu, Y. Li, H. W. Tseng, D. C. Ralph, R. A. Buhrman, Appl. Phys. Lett., 101, 122404 (2012). 3K. Roy, S. Bandyopadhyay, J. Atulasimha, Appl. Phys. Lett., 99, 063108 (2011).

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/f80b0892f94f3338bf712e9f149fd36a.jpg)
FIG 1: (a) Proposed structure consists of FeGaB/PMN-PT (ferromagnetic/piezoelectric) heterostructure and corresponding magnetization components. (b) △E/kbT vs thickness curve to examine the minimum thickness (dmin) requirement for thermal stability.

![](https://assets.underline.io/uploads/markdown_image/1/image/58c037c03f736d7b2641a003318c63ad.jpg)
FIG 2: (a) Simulated in-plane 90o magnetization switching from initial random magnetization state after applying critical stress. (b) Simulated in-plane 180o magnetization switching after magnetization switches to in-plane easy axis for 100×90.9×9.5 nm3 nanodisc.

 Micromagnetic study of strain-induced magnetization switching in FeGaB nanomagnets for self-biased MRAM applications

Coupling lattice and spin degrees of freedom without the use of external magnetic fields allows for energy-efficient spintronic devices. In this context, hybrid materials composed of an oxide layer undergoing a first-order, structural phase transition (SPT), such as V2O3, and a ferromagnetic layer, such as Ni, constitute a very promising paradigm for the development of such applications. In the case of a Ni/V2O3 heterostructure, the structural domains inherent to the first-order phase transition in the V2O3 cause a differential strain pattern in the Ni layer. By virtue of the inverse magnetorestrictive effect, this strain pattern is converted into a magnetic anisotropy distribution in the Ni. This offers a promising alternative to voltage-controlled magnetism or other switching mechanisms without a magnetic field [1,2]. Here we show by direct imaging of the thermal evolution of the Ni spin structure that the magnetic domains can be tuned in both size and orientation upon crossing the structural phase transition (SPT) of the proximal V2O3 layer [3]. We find a drastic temperature-driven reorientation of the Ni magnetic domains across the SPT which is responsible for an abrupt increase in the coercive field (up to 500%). By synchrotron-based X-ray microscopy we show a reconfiguration of the ferromagnetic domain pattern in the Ni layer across the V2O3 SPT together with changes in magnitude and direction of the magnetic anisotropy. These observations are in good agreement with both angular dependent ferromagnetic resonance measurements and micromagnetic simulations. Our simulations successfully replicate the change in the coercivity as well as the reorientation of the magnetization of the Ni domains after the implementation of a simple model of strain-dependent magnetic anisotropies. Furthermore, simulations also exhibit a phase coexistence of different magnetic domains across the SPT in the V2O3. Direct observations of the lateral correlation length of the Ni domains show an increase of almost an order of magnitude at the SPT compared to room temperature as well as a broad spatial distribution of the local transition temperatures. This corroborates the phase coexistence of Ni anisotropies due to the V2O3 structural domains coexistence across the SPT. Our data reveal that the reorientation of Ni domains is controlled by the reconfiguration of the structural domains of the oxide layer across the SPT [4,5], due to strain induced proximity. Our findings reveal a novel pathway to control magnetic domains without a magnetic field through proximity to a material undergoing a first-order structural phase transition, which allows for engineering of coercive fields for novel data storage architectures and novel device concepts based on “straintronics”. Acknowledgements: This work was supported by the Spanish MINECO projects MAT2015-68772-P and PGC2018-097789-B-I00 and European Union FEDER funds. A.F.R. and M.G.M acknowledge financial support from the EU CALIPSO Transnational Access programme. J.G.R. acknowledges support from Colciencias under grant 120471250659. The work by I.V., C.W. and I.K.S. at UCSD were supported by the Office of Basic Energy Science, U.S. Department of Energy, BES-DMS funded by the Department of Energy’s Office of Basic Energy Science, DMR under grant DE FG02 87ER-45332.

**References:**

[1] de la Venta, J., Wang, S., Ramirez, J. G., et al., Appl. Phys. Lett. 102, 1 (2013); [2] de la Venta, J., Wang, S., Saerbeck, T., et al., Appl. Phys. Lett. 104, 062410 (2014). [3] Valmianski, I., Fraile Rodríguez. A., Rodriguez-Alvarez, J., et al., (in review). [4] McLeod, A. S., Van Heumen, E., Ramirez, J. G., et al. Nat. Phys. 13, 80 (2017). [5] Gilbert, D. A., Ramírez, J. G., Saerbeck, T., et al., Sci. Rep. 7 13471 (2017).

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/609df7dde4c22e1de9b043b18a9dd651.jpg)
Fig. 1. Top row panels show representative XMCD-PEEM images recorded at temperatures below (150 K), across (170 K) and above (200 K) the SPT of V2O3. The insets show their Fourier transforms. The initial state (blue contrast) at low temperature (LT) splits into a pattern of small domains (yellow) which progressively rotate and grow across the V2O3 SPT as the high-temperature (HT) phase is reached. Bottom row panels show histograms of the absolute value of the average rotation angle of the Ni magnetization, Df, at selected temperatures (calculated from the images in the top row). Sharp Df distributions are found for the LT (150 K) and HT (200 K) phases in contrast to broader distributions in the middle of the V2O3 SPT (170 K).

![](https://assets.underline.io/uploads/markdown_image/1/image/0691162ae0a39dfe6b44c065dd312054.jpg)
Fig. 2. Coercivity as a function of temperature of Ni(10 nm)/V2O3(100 nm) across the V2O3 SPT recorded while cooling the sample from 300 K to 120 K (only data from 200 K are shown). A sharp rise in the coercive field, Hc, of about a factor of four, is found as the temperature goes below the V2O3 SPT. The garnet circles represent the coercivity values obtained from the experimental loops while the green triangles correspond to those obtained from micromagnetic simulations.

Downloads

Next from INTERMAG 2021

Strain-Free Magnetic Tunnel Junction with Metastable bcc CoxMn100-x(001) Ferromagnetic Layers

Similar lecture

Magnetocaloric effect in Hf1-xTaxFe2By

.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

Strain-Free Magnetic Tunnel Junction with Metastable bcc CoxMn100-x(001) Ferromagnetic Layers

Similar lecture

Magnetocaloric effect in Hf1-xTaxFe2By

Downloads