2020 Abstracts YRLGW

Jakub Zelezny

Spin currents are one of the key concepts of spintronics. In the past, two types of spin currents have been predominantly discussed and utilized: the spin-polarized current in ferromagnetic materials and the spin Hall effect. The spin-polarized current only exist in magnetic materials, has a non-relativistic origin and flows in the same direction as the charge current. In contrast, the spin Hall effect exists also in non-magnetic materials, has typically a relativistic origin and is transverse to the charge current. In recent years it has been discovered, however, that the phenomenology of spin currents is much richer. We have shown that the spin-polarized current can also exist in some antiferromagnetic materials and that a new type of spin Hall effect exists, which has origin in the magnetic order, and occurs in ferromagnetic and some antiferromagnetic materials [1]. This effect is now referred to as the magnetic spin Hall effect and has been recently experimentally demonstrated in non-collinear antiferromagnet Mn3Sn [2]. We have also shown that the conventional spin Hall effect can exist in some non-collinear magnetic systems even in absence of the relativistic spin-orbit interaction [3].
Furthermore, we have found that a non-relativistic magnetic spin Hall effect can exist in a collinear antiferromagnet [4]. Such system is distinct from systems where the transverse spin currents have been previously discussed because in the non-relativistic limit it conserves spin and thus it allows for spin-charge conversion in a spin-conserving system. Here we review the various types of spin currents that can occur in magnetic systems and give general conditions for their existence as well as a symmetry classification. In addition, we present calculations of these novel spin currents in various collinear and non-collinear antiferromagnets.

[1] J. Železný et al., Phys. Rev. Lett. 119, 187204 (2017)
[2] M. Kimata et al., Nature 565, 627–630 (2019)
[3] Y. Zhang et al: New J. Phys. 20 ( 2018 )
[4] R. G. Hernández el., arXiv:2002.07073 (2020)

Yuki Shiomi

Magnetopiezoelectric effect [1,2], which refers to a linear strain response to electric currents and its inverse response in low-symmetric magnetic metals, is a generalization of magnetoelectric effects in insulators to metals. In metallic materials with high conduction-electron densities, static (dc) piezoelectric responses are not allowed, even if the metals have a symmetry group low enough to support a static polarization. This is because the static surface charge density is screened out by bulk conduction electrons. However, it was recently proposed [1] that dynamic distortion can arise in response to electric currents without screening effects in antiferromagnetic metals that simultaneously break time-reversal and spatial-inversion symmetries. Note that another magnetopiezoelectric effect of a topological origin has also been proposed recently [2].
Here, we have experimentally studied the magnetopiezoelectric effect [3-5] in antiferromagnetic conductors with low crystal symmetries: EuMnBi2 [3,5] (TN = 315 K) and CaMn2Bi2 [4] (TN = 150 K). Using laser Doppler vibrometry at low temperatures, we found that dynamic displacements emerge along the [110] direction upon application of ac electric currents in the c direction in EuMnBi2 below TN [3,5]. The displacement signals showing up in response to the electric current increase in proportion to the applied electric currents. We confirmed that such displacements are not observed along the c direction of EuMnBi2 or EuZnBi2 with nonmagnetic Zn ions, consistent with the symmetry requirement of the magnetopiezoelectric effect [1]. As temperature increases from the lowest temperature, the displacement signals decrease monotonically, showing that magnetopiezoelectric signals are larger for higher conductivity states as opposed to the conventional piezoelectric effect.

[1] H. Watanabe and Y. Yanase, Phys. Rev. B 96, 064432 (2017).
[2] D. Varjas, A. G. Grushin, R. Ilan, and J. E. Moore, Phys. Rev. Lett. 117, 257601 (2016).
[3] Y. Shiomi, H. Watanabe, H. Masuda, H. Takahashi, Y. Yanase, and S. Ishiwata, Phys. Rev. Lett. 122, 127207 (2019).
[4] Y. Shiomi, Y. Koike, N. Abe, H. Watanabe, and T. Arima, Phys. Rev. B 100, 054424 (2019).
[5] Y. Shiomi, H. Masuda, H. Takahashi, and S. Ishiwata, Sci. Rep. 10, 7574 (2020).

James M. Taylor

The field of antiferromagnetic spintronics is based on recent developments in the manipulation and detection of antiferromagnetic properties using electrical methods, opening up the possibility of these materials evolving from passive to active components of spintronic devices. Doing so offers a number of advantages, such as improved stability, reduction in stray fields and increased speed of dynamics. However, changes in the orientation of typical antiferromagnets’ Néel vectors do not produce read-out signals of the size required for applications. Topological antiferromagnets may offer the solution.

In this talk, we focus on the particular example of noncollinear antiferromagnets of the type Mn3X, whose chiral spin textures break time- and inversion-symmetries, leading to novel magnetotransport properties driven by momentum-space Berry curvature. These include a large intrinsic anomalous Hall effect [1], anomalous Nernst effect [2], and both intrinsic- [3] and magnetic-spin Hall effects [4].

However, for the utilization of these Berry curvature generated phenomena in topological antiferromagnetic spintronic devices, further exploration of the behavior of Mn3X materials in thin film form is required. We therefore present results from our recent work, where we grow high-quality thin films of such noncollinear antiferromagnets with different crystal structures by exploiting epitaxial engineering. Specifically, we demonstrate the thin film deposition of two distinct varieties of noncollinear antiferromagnet: Mn3Ir, with a cubic structure [5], and Mn3Sn, with a hexagonal structure [6].

The crystal structure of the films is characterized using a combination of x-ray diffraction and transmission electron microscopy, whilst their magnetic properties are studied using vibrating sample magnetometry and x-ray magnetic circular dichroism experiments. In doing so, we illuminate the important role played by uncompensated moments in both materials, exploring how these are affected by sample microstructure and how, in turn, they affect antiferromagnetic domain distribution.
Such chiral domains play a key role in governing topological magnetotransport in these compounds. We elucidate this by measuring the Hall effect in lithographically patterned samples of both Mn3Ir and Mn3Sn, and find very different behavior in both cases. Whilst Mn3Ir shows a small conventional anomalous Hall effect [7], we observe a large anomalous Hall effect in Mn3Sn. Films down to 30 nm in thickness demonstrate an anomalous Hall conductivity of σxy (µ0H = 0 T) = 21 Ω-1cm-1, which we attribute to a Berry curvature mechanism [8]. Following cooling of Mn3Sn below its transition temperature into a glassy ferromagnetic state, we identify a change in transport behavior from anomalous to topological Hall effects.
Finally, we bring noncollinear antiferromagnets closer to functionality by moving to investigate the generation and interaction of spin currents in our thin films. Specifically, we use spin-torque ferromagnetic resonance measurements in Mn3X / NiFe bilayers to quantify their spin Hall angle. Significant charge-to-spin current conversion is identified, which depends intimately on epitaxial growth conditions, thin film magnetic state, and chiral domain structure. We conclude by discussing the origin of these different phenomena, and the potential for Mn3X materials to be used in chiralitronic devices.

[1] S. Nakatsuji, N. Kiyohara, and T. Higo, Nature 527, 212 (2015)
[2] H. Reichlová et al., Nature Communications 10, 5459 (2019)
[3] W. Zhang, W. Han, S. H. Yang, Y. Sun, Y. Zhang, B. Yan, and S. S. P. Parkin, Science Advances 2, e1600759 (2016)
[4] M. Kimata et al., Nature 565, 627 (2019)
[5] J. M. Taylor et al., Physical Review Materials 3, 074409 (2019)
[6] A. Markou, J. M. Taylor, A. Kalache, P. Werner, S. S. P. Parkin, and C. Felser, Physical Review Materials 2, 051001(R) (2018)
[7] J. M. Taylor et al., Applied Physics Letters 115, 062403 (2019)
[8] J. M. Taylor, A. Markou, E. Lesne, P. K. Sivakumar, C. Luo, F. Radu, P. Werner, C. Felser, and S. S. P. Parkin, Physical Review B 101, 094404 (2020)

Andrew Ross

With spin dynamics in the THz regime, stability in the presence of external magnetic fields, and a lack of stray fields, antiferromagnetic materials are positioned to become key in future low power spintronic devices [1]. Here, we grow and investigate high quality thin films of hematite (α-Fe2O3) (< 500 nm) of different orientations. Through measurements of the spin Hall magnetoresistance in hematite/Pt bilayers, the magnetic anisotropies of the thin films can be extracted, and the critical temperature of the Morin transition from the easy plane to the easy axis antiferromagnetic phase is electrically observed [2, 3]. Whilst a key part of antiferromagnetic spintronics is to encode and read information in the Néel vector, the efficient transfer of information is crucial for integration of antiferromagnets into devices. Recently, we demonstrated that a diffusive magnon current can be carried over micrometres in antiferromagnetic single crystals, but such crystals are not suitable for spintronic devices [4,5]. Despite theoretical works investigating the mechanisms for the long-distance propagation of pure spin currents carried by the antiferromagnetic order [6, 7], studies on thin film antiferromagnets making use of single-frequency or broadband excitations have failed to achieve efficient transport of angular momentum by magnons [8]. Making use of hematite thin films, a robust magnon current can propagate with intrinsic diffusion lengths of hundreds of nanometres. The efficiency of the transport mechanisms can be tuned by field cycling of the domain structure, the growth orientation, and the relative orientations of the magnetic field and magnetic anisotropies. The manner by which the stabilisation of the antiferromagnetic domain structure (see Fig. 1) results in frequency dependent length scales and proves to be critical in the magnon transport will be discussed [9].

[1] T. Jungwirth et al., Nat. Phys. 14, 200-203 (2018)
[2] R. Lebrun et al., Comm. Phys. 2 50 (2019)
[3] A. Ross et al., arXiv:2001.03117 (2020)
[4] R. Lebrun, A. Ross et al., Nature 561, 222-225 (2018)
[5] R. Lebrun et al., arxiv: 2005.14414 (2020)
[6] S. Takei et al., Phys. Rev. B, 90, 94408 (2014)
[7] S. Bender et al., Phys. Rev. Lett. 19, 056804 (2017)
[8] H. Wang et al., Phys. Rev. B 91, 220410 (2015)
[9] A. Ross, R. Lebrun et al., Nano Lett. 20 1, 306-313 (2020)

Kai Litzius

Magnetic skyrmions are topologically stabilized spin configurations that, like domain walls (DWs), can react to external stimuli by collective displacement, which is both physically intriguing and bears promises to realize next generation non-volatile data storage technologies. [1] However, skyrmions in ferromagnets move at an angle with respect to the current direction, which complicates the use of skyrmions in wire devices because the motion component perpendicular to the current can move the skyrmion to a wire edge and thereby annihilate it. [2] Antiferromagnetically coupled systems with compensated angular momentum (such as compensated ferrimagnets and natural antiferromagnets) can reduce this skyrmion Hall effect to zero and could additionally provide high speed dynamics to move spin structures at unprecedented speeds. [3,4] Skyrmions are predicted to move at even higher speeds in these materials, thus making these materials challenging but promising candidates for future spintronic devices.
Besides the compensation of perpendicular motion of skyrmions with respect to the drive, the predictability of their trajectories is also of major importance. Analytical equations of motion describe straight 180° DWs in the one-dimensional (1D) model while rigid, circular bubble domains and skyrmions are predicted to move according to the Thiele equation. [5] However, DWs and skyrmions are often not perfectly straight or circular. Here, we study how strongly deformed DWs and bubble skyrmions move in uncompensated ferrimagnetic Pt/CoGd/W in response to current pulses. We find that all 1D spin textures as well as all fully enclosed spin textures, reach speeds >500 m/s and display identical dynamics. While high speeds are indeed reached, the predicted differences between skyrmion and DW dynamics could not be observed. We attribute this deviation from the commonly used model to significant deformations of the skyrmions during their motion.

[1] K. Everschor-Sitte et al., Journal of Applied Physics 124, 240901 (2018)
[2] W. Jiang et al., Physics Reports 704, 1-49 (2017)
[3] S. Woo et al., Nature Communications 9, 959 (2018)
[4] L. Caretta et al., Nature Nanotechnology 13, 1154 (2018)
[5] F. Büttner, I. Lemesh & G. S. D. Beach, Scientific Reports 8, 4464 (2018)

Kouta Kondou

Spin Hall effect (SHE) provides the spin-charge interconversion in non-magnetic materials, which has drawn much attention because of its potential application for efficient magnetization switching via the spin torque [1].
Here we focus on the topological antiferromagnet Mn3Sn to realize the new functionality in spin-charge conversion. Mn3Sn exhibits the large anomalous Hall effect comparable with ferromagnet at room temperature [2]. Figure 1 shows a devise structure for spin accumulation detection. By applying the charge current on a Mn3Sn strip, spin accumulation can be detected electrically by the ferromagnetic electrode. This technique enables us to observe the SHE in Mn3Sn, exhibiting a sign change when its small magnetic moment switches orientation. Additionally, we succeeded in observation of the sign change in the inverse effect by means of spin pumping method [3]. These new SHEs were named Magnetic spin Hall effect (MSHE) and Magnetic inverse spin Hall effect (MISHE). Recently we fabricated the ferromagnetic layer/Mn3Sn bilayer to investigate the spin torque due to MSHE in Mn3Sn, which can be expected to realize the unconventional spin torque generation.

[1] L. Liu, C-F. Pai, Y. Li, H. W. Tseng, D. C. Ralph and R. A. Buhrman, Science 336, 555 (2012).
[2] S. Nakatsuji, N. Kiyohara and T. Higo, Nature 527, 212 (2015).
[3] M. Kimata, H. Chen, K. Kondou et al., Nature 565, 627 (2019).

Annika Johansson

The Edelstein effect, also known as inverse spin-galvanic effect, is a magnetoelectric phenomenon providing charge-current-to-spin conversion in systems with broken inversion symmetry. In pristine nonmagnetic materials, a finite macroscopic spin polarization can be induced purely electrically by the application of an electric field [1,2]. Originally, the Edelstein effect has been discussed for two-dimensional Rashba systems at surfaces or interfaces. Whereas for isotropic Rashba systems a large spin-orbit splitting is crucial for efficient charge-to-spin conversion, current research aims at finding novel materials beyond conventional Rashba systems providing a large direct or inverse Edelstein effect, for example three-dimensional topological insulators and oxide interfaces.
In this talk the Edelstein effect in Rashba systems and topological materials is discussed within the semiclassical Boltzmann transport theory [3,4]. Here, one focus is on finding materials in which a large Edelstein effect can be realized. Considering geometrical and topological properties, Weyl semimetals are identified as candidates for a highly efficient charge-to-spin conversion [4].
Further, SrTiO_3-based two-dimensional electron gases (2DEGs) have been found to provide a large inverse Edelstein effect [5], in particular the 2DEG emerging at the interface between SrTiO_3 and AlO [6]. The application of a gate voltage leads to a strong variation and even sign changes of the spin-to-charge conversion efficiency. This unconventional gate dependence is explained by a band-resolved analysis of the Edelstein signal. The experimentally observed spin-to-charge conversion is related to the band structure as well as the topological character and the spin texture of the 2DEG [6]. In addition the orbital Edelstein effect, originating from the orbital moments, is analyzed, which can exceed the conventionally discussed spin Edelstein effect by one order of magnitude.

[1] A. G. Aronov, Y. B. Lyanda-Geller, JETP Lett. 50, 431 (1989)
[2] V. M. Edelstein, Solid State Commun. 73, 233 (1990)
[3] A. Johansson et al., Phys. Rev.B 93, 195440 (2016)
[4] A. Johansson et al., Phys. Rev.B 97, 085417 (2018)
[5] E. Lesne et al., Nat. Mater. 15, 1261 (2016)
[6] D. Vaz et al., Nat. Mater. 18, 1187 (2019)

Max Hirschberger

Recently, we have explored a variety of metallic materials where non-coplanar magnetic order occurs on characteristic length scales λ comparable to the size of a single crystallographic unit cell. In this limit, the standard (real-space) model of the ‘topological’ Hall effect, the key transport signature of Berry curvature due to canted magnetism, is expected to fail. In metallic magnets where the carrier mean free path exceeds λ, a momentum-space picture of the THE is expected to be more appropriate for an adequate description.
We aim to approach the momentum space regime using four model compounds, listed here in order of decreasing λ: (1) In rare-earth magnets with inversion center such as Gd2PdSi3 and Gd3Ru4Al12, we observed nanometer-sized skyrmion textures in highly symmetric lattices and in absence of Dzyaloshinskii-Moriya interactions (λ~2-3 nm). (2) The (breathing) Kagome system Dy3Ru4Al12 realizes a peculiar arrangement of antiferromagnetically stacked canted spin-trimers (λ~1-2 nm). (3) The metallic pyrochlore oxide Nd2Mo2O7 is a canted ferromagnet where non-coplanarity occurs within a single unit cell (λ~1 nm).
The topological Hall and Nernst effects of these materials were studied while tuning a variety of ‘external knobs’. We changed the Fermi energy via substitutional doping and also modified the lattice spacing via hydrostatic pressure. Thus, we aim to develop new phenomenology of transport signatures characteristic in the limit of entangled real-space canted magnetism and momentum space Berry curvature.

Sergey Grytsiuk

Three-dimensional magnetic textures with particle-like properties [1-2] have been recently growing in popularity due to a great potential for innovative spintronic applications [3-5] and brain-inspired computing. However, only little is known about solids where such 3D magnetic solitons may exist and a complete theoretical model for the underlying magnetic interactions is remarkably elusive until now. While, for instance, the basic magnetic properties of the 2D skyrmions are determined by an intricate competition involving the Heisenberg exchange and the chiral relativistic Dzyaloshinskii–Moriya interaction (DMI), such models fail to explain the 3D-magnetic texture of few-nm size observed in MnGe.
Recently, we have discovered a conceptually new class of the magnetic interactions rooted in the so-called topological orbital moment, a Berry-phase effect that results from the orbital motion of the electron in a complex magnetic background [6]. We refer to these interactions as topological–chiral interactions, favouring the emergence of non-coplanar magnetic structures with scalar spin chirality of specific sign even without an external magnetic field. The novel interactions offer fundamentally different opportunities for imprinting chiral magnetism, as they manifest in the scalar chirality of spin arrangements on triangular plaquettes, as opposed to the vector chirality between pairs of spins in the case of DMI. By means of density functional calculations we demonstrate that these interactions are not small but can dominate over the celebrated DMI in selecting the chiral ground state, providing possibly a key for solving the open question of the recently observed complex 3D magnetic structures in B20-type chiral magnet MnGe. In addition, we show that in the continuum limit, the spin-chirality relates to the curvature of the magnetization field and one flavour of the topological–chiral interaction reverts to the Faddeev model [1] with solutions for the 3D magnetic solitons, known as hopfions.
In this talk, I will give brief introduction to the topological–chiral interactions focusing on rigorous derivation based on Multiple scattering theory. Also, I will discuss the role of the novel interactions in stabilising the nontrivial magnetic orders in bulk and thin films of different materials.

[1] Faddeev, L. D. Preprint IAS Print-75-QS70 (Inst. Advanced Study, Princeton, NJ, 1975), 32 pp. 31.
[2] N. Manton and P. Sutcliffe, Topological Solitons (Cambridge University Press, Cambridge, England, 2004).
[3] A. M. Kosevich et al., Phys. Rep. 194, 117 (1990).
[4] X. S. Wang et al., PRL 123, 147203 (2019)
[5] Yizhou Liu et al., 124, 127204 (2020)
[6] S. Grytsiuk et al., Nature Comm., 11, 511 (2020)

Dongwook Go

Spin current is one of the central concepts in spintronics. While early studies of giant magnetoresistance and spin-transfer torque have shown good agreement between the theory and experiment, recent experiments of current-induced torques in spin-orbit coupled systems imply that we need a theory which goes beyond “spin current picture”. In general, angular momentum can be carried by other degrees of freedom as well as the spin. For electrons, the angular momentum is encoded in not only the spin but also orbital part of the wave function, thus one can think of transport of orbital angular momentum carried by electrons in analogy to the spin transport.
In this talk, I will explain how to electrically generate orbital current and utilize it to exert a torque on the magnetization. As a way to generate the orbital current, I introduce a mechanism of orbital Hall effect, which is defined as orbital current response along transverse directions to an external electric field [1]. Then I show that injection of the orbital current to a ferromagnet can excite magnetization dynamics, which we call orbital torque [2]. One advantage of utilizing the orbital current is that it does not require spin-orbit coupling for electrical generation, which is in contrast to spin current generation, e.g., by spin Hall effect. Thus, the orbital torque mechanism predicts sizable current-induced torque even for weakly spin-orbit coupled materials. However, since the spin and orbital angular momenta transform in the same way upon symmetry operations, it is challenging to disentangle the orbital transport effect from the spin transport effect in experiments. For this purpose, we recently developed a general theory which can track angular momentum transfer between different angular-momentum-carrying degrees of freedom, which include not only the spin and orbital of the electron but also crystal lattice and local magnetic moment [3]. From a first-principles implementation of the formalism, we show that the orbital torque mechanism behaves qualitatively different from the “conventional” contribution caused by the spin Hall effect. This provides microscopic understanding of the orbital torque in terms of the electronic structure. Finally, I discuss further experimental implications and conceptual difference between the orbital transport and spin transport.
We acknowledge funding under SPP 2137 “Skyrmionics” (project MO 1731/7-1) and TRR 173 − 268565370 (project A11) of the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation).

[1] D. Go, D. Jo, C. Kim, and H.-W. Lee, Intrinsic Spin and Orbital Hall Effects from Orbital Texture, Phys. Rev. Lett. 121, 086602 (2018).
[2] D. Go and H.-W. Lee, Orbital Torque: Torque Generation by Orbital Current Injection, Phys. Rev. Res. 2, 013177 (2020).
[3] D. Go, F. Freimuth, J.-P. Hanke, F. Xue, O. Gomonay, K.-J. Lee, S. Blügel, P. M. Haney, H.-W. Lee, and Y. Mokrousov, Theory of Current-Induced Angular Momentum Transfer Dynamics in Spin-Orbit Coupled Systems, arXiv:2004.05945.