# Fileset [Manuscript_R1_wo highlight.docx](https://mdr.nims.go.jp/filesets/12e84862-422d-4b89-a18a-fd16b566498a/download) ## Creator [Nitipriya Tripathi](https://orcid.org/0009-0001-8178-709X), [Shrawan K. Mishra](https://orcid.org/0000-0001-6140-7443), [Shinji Isogami](https://orcid.org/0000-0001-7230-6090) ## Rights ©2026 American Physical Society[In Copyright](http://rightsstatements.org/vocab/InC/1.0/) ## Other metadata [Influence of spacer layers on spin transport efficiency in Mn 3 PtN / CoFeB heterostructures](https://mdr.nims.go.jp/datasets/39550b8e-1991-449d-b494-ee6e72bf5726) ## Fulltext Influence of Spacer Layers on Spin Transport Efficiency in Mn3PtN/CoFeB Heterostructures Nitipriya Tripathi1*, Shrawan K. Mishra1 ** 1School of Materials Science and Technology, Indian Institute of Technology (BHU), Varanasi-221 005, India. Shinji Isogami22Center for Magnetic and Spintronic Materials, National Institute for Materials Science (NIMS), Tsukuba 305-0047, Japan.Abstract The epitaxially grown multilayers, Mn₃PtN/Cu/Co₂₀Fe₆₀B₂₀, with the thickness of Cu spacer layers from 0 to 5 nm were prepared on the 100-oriented MgO substrates. The multilayer exhibited a pronounced exchange bias field in the absence of a Cu spacer, which became minimal with the introduction of a finite Cu spacer. We measured spin-torque ferromagnetic resonance to estimate the spin Hall angle (θSH) in the multilayers, emphasizing the role of the Cu spacer layer in the spin current transport across the interface. We found that both θSH and effective Gilbert damping parameter decrease significantly upon the insertion of Cu, suggesting that the spin-orbit torque (SOT) primarily originates from interfaces. A correlation between SOT and exchange bias with Cu spacer layer indicates the significance of interfacial contributions. These experimental results demonstrate the critical importance of interfacial engineering in antiferromagnetic-based spintronics, revealing a new mechanism for generating spin torque in magnetic heterostructures that is essential for controlling magnetisation dynamics in energy-efficient spintronics applications.Keywords: Spin-orbit coupling, magnetic thin films, exchange bias, spin current, spin-orbit torque, and spintronics.Contact: *nitipriyatripathi.rs.mst20@itbhu.ac.in **shrawan.mst@iitbhu.ac.in1. IntroductionNoncollinear antiferromagnets (NC-AFM) have recently gained considerable attention in spintronics owing to their negligible stray fields, robustness against magnetic perturbations, and ultrafast spin dynamics [1,2]. Beyond their conventional use in exchange biasing, NC-AFM can actively participate in spin-current generation, detection, and manipulation through mechanisms such as the spin Hall effect, spin pumping, and spin–orbit torque (SOT) [3-5]. Among them, Mn based compounds with metallic Mn₃X and antiperovskite nitrides Mn₃XN (X = Sn, Ga, Ni, Pt, etc.) have emerged as promising spin-current sources due to their magnetic structures and strong spin-orbit coupling [6-8]. However, despite increasing interest in their spin-transport properties, a detailed understanding of interfacial spin dissipation and magnetic damping at NC-AFM/FM interfaces remains limited. Previous studies on NC-AFM/FeGa bilayers have shown the unusual dependence of effective Gilbert damping constant (αeff) on MnIr thickness, i.e., a MnIr layer with the thickness of ~2 nm exhibits strong intrinsic magnetic anisotropy and enhanced αeff was observed due to spin pumping phenomenon involved [9]. In the noncollinear antiperovskite Mn₃PtN (MPN), the reported spin Hall angle (θSH ≈ 0.033) exceeds that of Mn₃Pt (MP) (θSH ≈ 0.025), indicating efficient spin–orbit coupling [10]. Nevertheless, the microscopic mechanism of SOT in Mn₃XN compounds remains unclear, particularly when metallic spacers modify interlayer exchange coupling. Inserting Cu or Al between NC-AFM/ferromagnet (FM) interfaces (e.g., γ-Mn₃Ir/Py) has been shown to suppress SOT, underscoring the critical role of interfacial coupling in spin transport [11-13]. These studies have attracted much attention in terms of tailoring SOT caused by interfacial material designs. In this study, the effects of interlayer magnetic exchange interaction and αeff on the θSH are investigated using NC-AFM/FM bilayers with and without insertion of Cu spacer layers. Magnetic exchange interaction was modulated significantly by changing the Cu spacer thickness (tCu), and a pronounced exchange bias field (Hex) was observed in the absence of Cu spacer layer. The αeff decreased with increasing tCu, suggesting less spin-pumping contribution by spacer layer. The θSH decreased with increasing tCu. These results show that the interfacial exchange interaction and spin injection efficiency can be precisely engineered through spacer-layer control, offering practical guidelines for designing low-damping, NC-AFM based spintronic heterostructures.2. Experimental DetailsThe trilayer structures were deposited on (100)-oriented MgO substrates, following a stacking sequence from bottom to top: MPN(5) / Cu(tCu) / CoFeB(3) / MgO(3) (thickness unit in nm), utilizing both DC and RF sputtering techniques. The base pressure was less than 1×10⁻⁶ Pa, and tCu was varied from 0 to 5.0 nm. Reactive nitride sputtering, using N₂ gas, was employed to deposit the MPN layer, with a gas flow ratio of approximately 10% (N₂ / (Ar + N₂)). The deposition temperature for the MPN layer was 723 K [10]. The CoFeB (CFB), Cu, and MgO layers were deposited at room temperature. A control sample, which replaced the MPN layer with an MP layer, was prepared for comparative analysis. Structural characterization was conducted using X-ray diffractometry (XRD) with a Cu Kα radiation. The growth rate for Cu spacer was 0.70 Å/s, in contrast to 0.103 Å/s for the CFB layer. Note that the deposition rates for Cu and CFB allow for sharp interface with negligible intermixing and/or alloying (see Fig. S1 in the Supplementary Materials). The surface roughness (Ra) and morphology of the thin films were assessed using atomic force microscopy (AFM). The magnetic properties were measured using a vibrating sample magnetometer (VSM) under magnetic fields along in-plane directions. The exchange bias field (m0Hex) was estimated from magnetic hysteresis loops recorded at different temperatures using a SQUID magnetometer. Spin-torque ferromagnetic resonance (ST-FMR) measurements were performed using microfabricated devices to determine the θSH. This dynamic process is captured as a rectified voltage (Vmix) due to anisotropic magnetoresistance. Notably, the RF power was +15 dBm throughout the measurement. A broadband FMR was utilized to estimate the αeff over a frequency range of 2 to 20 GHz, employing a coplanar waveguide transmission line. In-plane magnetic field was applied between ±150 mT at an angle of φ = 45° with respect to the signal line. FMR spectra were examined to derive the resonance field (m0Hr) and the line width (m0DH).3. Results and DiscussionA. Structural characterization Figure 1(a) shows the schematic illustration of stacking structure. Figure 1(b) shows the out-of-plane XRD profiles for the samples with tCu = 0, 1.5, 2.5, and 5.0 nm. The diffraction peaks at 2θ/w ≈ 23° and 47° correspond to the (100) and (200) planes of MPN, respectively, and 2θ/w ≈ 51° does to the (200) plane of FCC-Cu. Notably, no diffraction peak was detected originating from the 3.0 nm-thick CFB layer, indicating its amorphous and/or nanocrystalline nature. Therefore, it is inferred that the strain on the CFB layer induced by the Cu layer can be ruled out, because of no epitaxial growth of the CFB layer. Figure 1(c) shows the Ra values on tCu, suggesting the pronounced grain growth in the Cu spacer layer with increasing tCu, as shown in Fig. 1(d), which might be due to grain growth of Cu.Figure 1 (a) Schematic illustration of the representative stacking structure in this study. (b) Out-of-plane XRD profiles as a function of Cu layer thickness (tCu), where (200)* corresponds to the diffraction from FCC-Cu. (c) Dependence of surface roughness (Ra) on tCu. (d) Atomic force microscopy (AFM) images of the MgO top surfaces for all deposited samples with various tCu. B. Interlayer magnetic exchange interactionFigure 2(a) shows the magnetic hysteresis loops at 5 K and 300 K for the MPN/CFB without Cu spacer layer (tCu = 0 nm). While the saturation magnetization (m0Ms) values were identical regardless of the measurement temperature, the loop at 5 K shifted toward negative direction (indicated by a red arrow), suggesting the presence of unidirectional exchange anisotropy originating from interfacial exchange magnetic coupling between the CFB and MPN layers. The m0Hex was estimated to be 28 mT. Figure 2(b) shows the same results as Fig. 1(a), but with Cu spacer (tCu = 1.5 nm). The m0Hex ≈ 0 mT was observed for both measurement temperatures, suggesting that unidirectional exchange magnetic anisotropy vanished due to Cu spacer. Figure 2(c) summarizes the m0Hex values as a function of tCu with 20 K and 300 K for the same MPN/Cu/CFB. While the m0Hex was independent of tCu for 300 K, the m0Hex drastically increased without Cu spacer for 20 K. These results show that interlayer magnetic exchange interaction is present at the direct heterojunction of MPN/CFB, which vanishes by the Cu spacer as thin as 1.5 nm, suggesting the magnetic decoupling between MPN and CFB layers. For comparison, the temperature dependent magnetic property was investigated for the MP/Cu(1.5 nm ~ 5 nm)/CFB. It was found that the magnetic hysteresis loops are consistent with the MPN/Cu(1.5 nm ~ 5 nm)/CFB for the temperature range of 20 K ~ 300 K, suggesting that the CFB layer on the Cu spacer might be magnetically decoupled with the MP and MPN layers (see Fig. S2 in the Supplementary Materials).Figure 2 (a,b) Magnetic hysteresis loops recorded at 300 K and 5 K for the MPN/CFB (a) and MPN/Cu(1.5 nm)/CFB (b). (c) Cu thickness (tCu) dependence of exchange bias filed (m0Hex).C. Effective Gilbert damping constant (aeff)Figures 3(a-d) show the representative FMR spectra for the MPN/Cu/CFB with various tCu with a frequency of 10 GHz. The spectrum shape suggests the superposition of both symmetric (Vs) and asymmetric (Va) Lorentzian functions regardless of tCu. To extract the m0Hr and m0DH, the spectra are fitted using a line shape equation as described in the previous reports [14], . (1)Figure 3(e) shows the m0Hr dependences of f for the MPN/Cu/CFB with various tCu. To analyse the interfacial magnetic anisotropy in the CFB layers, the plots were fitted using the Kittel’s formula [15], , (2)where g, f and Meff represent the gyromagnetic ratio, the applied frequency and the effective demagnetization field defined by μ0Meff = μ0Ms – (2Ki/Ms)(1/tCFB), respectively. Ki, Ms, and tCFB represent the interfacial magnetic anisotropy energy density originating from the MPN/CFB and MP/CFB interfaces, the saturation magnetization of the CFB layer, and the thickness of CFB layer, respectively. Figure 3(f) shows the f dependences of m0DH for the symmetric component of FMR spectra in the MPN/Cu/CFB. To estimate αeff, the plots were fitted using the Kittel’s formula [15], , (3)where μ0DH0 represents the f independent and extrinsic line broadening caused by the magnetic inhomogeneity such as magnetic anisotropy distribution [16]. Figure 3(g) summarizes the μ0Meff as a function of tCu for the MPN/Cu/CFB together with the MP/Cu/CFB (see Fig. S3 in the Supplementary Materials). The μ0Meff was found to decrease with increasing tCu for both samples. This is caused dominantly by the Ki reduction by Cu insertion (see Fig. S4 in the Supplementary Materials), which led us to consider that the Cu layer breaks the interlayer exchange interaction between MPN and CFB layers. Figure 3(h) summarizes the aeff as a function of tCu for both samples. Distinct behaviours of aeff were observed. The obtained values at tCu = 0 were 0.018 for the MP/Cu/CFB and 0.015 for the MPN/Cu/CFB, which decreased with increasing tCu, reaching a saturation value. The experimental findings indicate that aeff is significantly higher in trilayers with magnetically coupled layers compared to those where the layers are separated by Cu. This suggests that the increased aeff in coupled trilayers cannot be solely attributed to spin-pumping effects. Instead, the study proposes that the m0DH broadening in these trilayers may be influenced by interlayer exchange coupling as discussed below. Figure 3 (a-d) FMR spectra for the MPN(5 nm)/Cu(tCu)/CFB(3 nm)/MgO(3 nm) at a frequency of 10 GHz. (e) Resonance field (m0Hr) dependence of applied frequency (f). (f) f dependence of linewidth (m0DH). (g,h) Cu thickness (tCu) dependences of effective demagnetization field (m0Meff) (g) and effective Gilbert damping constant (aeff) (h).D. Spin-Hall angles (qSH)Figures 4(a-c) show the representative ST-FMR spectra for the MPN/Cu/CFB. The measured Vmix can be analysed by fitting with Eq. (1), and the total component of Vs (orange line) and Va (blue line) reproduced the Vmix as shown by the red line). Figure 4(d) shows the f as a function of m0Hr to determine the m0Meff. Following the same method as Fig. 3(e), the plots were fitted using Eq. (2). The resultant m0Meff ≈ 1.12 T was obtained for the MPN/CFB, which decreased with increasing tCu. Figure 4(e) shows the m0DH as a function of f determined by the spectra fitting shown in Figs. 4(a-c). The solid lines reproduce the plots using Eq. (3) with aeff 0.0138 for the MPN/CFB, which reduced as tCu increases. While both and aeff via the ST-FMR were typically smaller than those via the broadband FMR by ~30 % (Figs. 3(g) and 3(h)), tCu dependences of which were found to be identical. Figure 4(f) summarizes the θSH as a function of tCu that relies on the ratio of the Vs and Va components recorded through the ST-FMR spectra as [17-20], , (4)where e, d, and represent the elementary charge, the layer thickness, and the Dirac’s constant, respectively. Note that the thickness of Cu layer is included in the dMP/MPN. The θSH of MPN (MP) was estimated to be 0.0320.004 (0.0240.003) for tCu = 0, while θSH suppressed with increasing tCu (see Fig. S5 in the Supplementary Materials). These results suggest that θSH is associated with aeff as well as the exchange interaction at MPN/CFB interfaces as discussed below.Figure 4 (a-c) Measured ST-FMR signals of MPN(5 nm)/Cu(tCu)/CFB (3 nm) at room temperature with f = 9 GHz, where orange, blue, and red lines represent the fitting results by symmetric (Vs), asymmetric (Va), and total (Vmix) component of Eq. (1). (d) Resonance field (m0Hr) dependence of frequency (f), where the solid lines represent the fitting result using Eq. (2). (e) f dependence of linewidth (m0DH), where the solid lines represent the fitting results using Eq. (3). (f) Spin-Hall angles (qSH) as a function Cu thickness (tCu).The aeff was enhanced for tCu = 0, while drastically suppressed by tCu even with the thickness of 1.5 nm, in which interlayer exchange interaction vanishes as shown in Fig. 1(c). We discuss the possible mechanisms by taking two parts into account: (i) the bulk part of CFB layer; (ii) the MPN/Cu/CFB stacking. (i) As for the bulk part, the aeff may be contaminated by the contribution from a two-magnon scattering in the present CFB layer, which is generally regarded as an extrinsic mechanism that arises from inhomogeneities, defects, or surface roughness of a magnetic film. To achieve lower aeff, it is necessary to reduce the two-magnon scattering. This can be accomplished through minimizing defects and improving film quality [21]. Our results on AFM shown in Fig. 1, Ra increased with tCu, so that aeff is expected to enhance in terms of the two-magnon scattering in principle. However, aeff decreased with tCu, as shown in Fig. 3(h), which is opposite to the expectation. Therefore, the contribution from two-magnon scattering in the present bulk CFB layer is minor, if any, comparing to the interfacial contribution. (ii) As for the entire MPN/Cu/CFB stacking, the interlayer exchange interaction could be one of the possible mechanisms for aeff that is decreased by Cu spacer. For example, an impact of the interlayer exchange interaction has been investigated using IrMn/FeCo bilayer system [22]. An in-plane angular dependence of aeff is explained by the presence of in-plane exchange bias with a simple cosine function, that is, large (small) aeff is obtained in the anti-parallel (parallel) configuration between the exchange bias and applied field. The previous report led us to consider that aeff reduction with tCu can be explained by the reduction of interlayer exchange interaction. Another possible mechanism for the interfacial contribution is the spin-pumping effect depending on tCu [23], which has been also observed in the AFM/FM bilayers [24]. The spin-angular momentum transfers from the CFB layer to the MPN layer under the magnetization precession state in principle. In the case of tCu = 0, the spin-pumping effect can be dominantly detected as an additional line broadening phenomena of FMR spectra. As a result, the aeff of CFB layer can be enhanced as shown in Fig. 3(h). In the case of tCu > 1.5 nm, on the other hand, it is speculated that the spin angular momentum returns to the CFB layer as a back-flow phenomena and/or electron scattering due to the interfacial electronic structures [12,25], resulting in the reduction of aeff. To sum up, the dependence of aeff on tCu is explained by the interlayer exchange interaction as well as the spin-pumping effect.Next, we discuss the relationship between θSH and tCu shown in Fig. 4(f). The possible mechanisms are split into two origins: (iii) the bulk part of MPN layer; (iv) the interface of MPN/Cu/CFB stacking. (iii) As for the bulk contribution to the θSH, which typically saturates at thicker thickness of the nonmagnetic layer up to the spin diffusion length [26]. Thus, while bulk spin-Hall effect provides a baseline efficiency, it alone cannot explain the large θSH values observed in ultrathin heterostructures. (iv) As for the interfacial contribution, which arises at the boundary between the ferromagnetic and nonmagnetic layers, where broken inversion symmetry and strong spin–orbit coupling induce Rashba-type spin–momentum locking. This generates an interfacial spin polarization transverse to the applied current, producing an additional spin current via the Rashba–Edelstein effect. The interfacial θSH is largely thickness-independent in the ultrathin regime and primarily contributes to the field-like torque component in SOT measurements. Intermixing, nitrogen incorporation, or crystal orientation can further enhance this effect, yielding interfacial θSH values up to an order of magnitude larger than the bulk [26]. Furthermore, based on the spin circuit theory in spin accumulation of FM/NM interface [27], the total electron density current in NM region (j) is qualitatively governed by the magnetization direction (sinq) of FM layer. Therefore, enhanced j is expected when the magnetization direction of FM layer tilts with respect to the in-plane direction, which corresponds to the efficient spin current injection at the FM/NM interface. Considering that the magnetization direction of CFB layer significantly tilts due to the interlayer exchange interaction, spin injection efficiency might be enhanced, resulting in the high qSH at tCu = 0 nm. In the case of tCu > 1.5 nm, on the other hand, the tilt angle might be suppressed due to tiny or no interlayer exchange interaction, spin injection efficiency might be also suppressed, resulting in the reduction of qSH.In magnetic heterostructures central to spintronics, in general, exchange bias, θSH, and Dzyaloshinskii-Moriya interaction (DMI) exhibit interconnected behaviours driven by interfacial spin-orbit coupling (SOC). The exchange bias provides an effective in-plane field to break symmetry, allowing field-free magnetization switching in the SOT-devices with high θSH, and DMI enhances this effect by inducing chiral exchange bias. The DMI and θSH correlate via interface effects: stronger SOC boosts both, as seen in IrMn/CoFeB, where DMI strength scales with spin mixing conductance (Geff↑↓), which modulates effective SOT efficiency (proportional to θSH). In AFM/FM systems, DMI can drive exchange bias by causing spin canting at compensated interfaces, enhancing coercivity via spin-flop mechanisms. This interplay is tuneable; e.g., varying AFM thickness increases DMI and Geff↑↓, correlating with higher effective θSH for SOT [28-30]. 4. ConclusionIn conclusion, our systematic investigation of MP/Cu/CFB and MPN/Cu/CFB trilayers using ST-FMR and FMR techniques reveals that the magnetization and in-plane uniaxial anisotropy are largely affected by the presence and thickness of the Cu spacer layers. The absence of a Cu spacer leads to a significant enhancement of the θSH and the αeff, which we attribute to the combined effects of spin pumping and interfacial exchange interaction. These findings highlight the crucial role of the Cu spacer, that is, θSH and the αeff decrease and saturate by the Cu thickness of only 5 nm. The results indicate a promising pathway for engineering spintronic devices through careful control of interfacial coupling and spacer layer thickness.AcknowledgmentsN.T. expresses thanks to DST-INSPIRE for the fellowship and to the NIMS-ICGP graduate fellowship program. IIT (BHU), Varanasi, is acknowledged for its partial financial support.Reference: [1] V. Baltz, A. Manchon, M. Tsoi, T. Moriyama, T. Ono, and Y. Tserkovnyak, Antiferromagnetic spintronics, Rev. Mod. Phys. 90, 015005 (2018). [2] T. Jungwirth, X. Marti, P. Wadley, and J. Wunderlich, Antiferromagnetic spintronics, Nature Nanotechnology 11, 231 (2016).[3] K. O’Grady, L. E. Fernandez-Outon, and G. Vallejo-Fernandez, A new paradigm for exchange bias in polycrystalline thin films, J. Magn. Magn. Mater. 322, 883 (2010).[4] W. Zhang, M. B. Jungfleisch, W. Jiang, J. E. Pearson, and A. Hoffmann, F. Freimuth and Y. Mokrousov, Spin Hall effects in metallic antiferromagnets, Phys. Rev. Lett. 113, 196602 (2014).[5] H. Saglam, W. Zhang, M. B. Jungfleisch, J. Sklenar, J. E. Pearson, J. B. Ketterson, and A. Hoffmann, Spin transport through the metallic antiferromagnet FeMn, Phys. Rev. B 94, 140412(R) (2016).[6] J. Sinova, S. O. Valenzuela, J. Wunderlich, C. H. Back, T. Jungwirth, Spin Hall effects, Rev. Mod. Phys. 87, 1213 (2015).[7] I. Martin and C. D. Batista, Itinerant electron-driven chiral magnetic ordering and spontaneous quantum Hall effect in triangular lattice models, Phys. Rev. Lett. 101, 156402 (2008).[8] S. Nakatsuji, N. Kiyohara, and T. Higo, Large anomalous Hall effect in a non-collinear antiferromagnet at room temperature, Nature 527, 212 (2015).[9] S. M. Rezende, R. L. Rodríguez-Suárez, and A. Azevedo, Diffusive magnonic spin transport in antiferromagnetic insulators, Phys. Rev. B 93, 054412 (2016).[10] N. Tripathi, S. K. Mishra, Y. Miura, S. Isogami, Impact of nitrogen on the charge-to-spin conversion efficiency in antiferromagnetic Mn3PtN compared to Mn3Pt thin films, Phys. Rev. B 109, 224406 (2024).[11] A. Kumar, P. Gupta, N. Chowdhury, K. A. Khan, U. Shashank, S. Gupta, Y. Fukuma, S. Chaudhary, and P. K. Muduli, Interfacial origin of unconventional spin-orbit torque in Py/IrMn3, Adv. Quantum Techno. 6, 2300092 (2023).[12] C. Swindells and D. Atkinson, Interface enhanced precessional damping in spintronic multilayers: A perspective, J. Appl. Phys. 131, 170902 (2022).[13] V. P. Amin, P. M. Haney, and M. D. Stiles, Interfacial spin-orbit torques, J. Appl. Phys. 128, 151101 (2020).[14] N. Mecking, Y. S. Gui, and C.-M. Hu, Microwave photovoltage and photo-resistance effects in ferromagnetic microstrips, Phys. Rev. B 76, 224430 (2007).[15] C. Kittel, On the theory of ferromagnetic resonance absorption, Phys. Rev. 73, 155 (1948).[16] W. Platow, A. N. Anisimov, G. L. Dunifer, M. Farle, and K. Baberschke, Correlations between ferromagnetic-resonance linewidths and sample quality in the study of metallic ultrathin films, Phys. Rev. B 58, 5611 (1998). [17] S. Karimeddiny, J. A. Mittelstaedt, R. A. Buhrman, and D. C. Ralph, Transverse and Longitudinal Spin-Torque Ferromagnetic Resonance for Improved Measurement of Spin-Orbit Torque, Phys. Rev. Applied 14, 024024 (2020).[18] A. R. Mellnik, J. S. Lee, A. Richardella, J. L. Grab, P. J. Mintun, M. H. Fischer, A. Vaezi, A. Manchon, E.-A. Kim, N. Samarth, and D. C. Ralph, Spin-transfer torque generated by a topological insulator, Nature 511, 449 (2014).[19] M. Harder, Z. X. Cao, Y. S. Gui, X. L. Fan, and C.-M. Hu, Analysis of the line shape of electrically detected ferromagnetic resonance, Phys. Rev. B 84, 054423 (2011).[20] K. Kondou, H. Sukegawa, S. Kasai, S. Mitani, Y. Niimi, and Y. Otani, Influence of inverse spin Hall effect in spin-torque ferromagnetic resonance measurements, Appl. Phys. Express 9, 023002 (2016).[21] L. Zhu, L. Zhu, D.C. Ralph, and R.A. Buhrman, Origin of Strong Two-Magnon Scattering in Heavy-Metal/Ferromagnet/Oxide Heterostructures, Phys. Rev. Appl. 13, 034038 (2020).[22] J. B. Mohammadi, J. M. Jones, S. Paul, B. Khodadadi, C. K. A. Mewes, T. Mewes, and C. Kaiser, Broadband ferromagnetic resonance characterization of anisotropies and relaxation in exchange-biased IrMn/CoFe bilayers, Phys. Rev. B 95, 064414 (2017).[23] M. Caminale, A. Ghosh, S. Auffret, U. Ebels, K. Ollefs, F. Wilhelm, A. Rogalev, and W. E. Bailey, Spin pumping damping and magnetic proximity effect in Pd and Pt spin-sink layers, Phys. Rev. B 94, 014414 (2016).[24] K. Roy, A. Mishra, P. Gupta, S. Mohanty, B. B. Singh, and S. Bedanta, Spin pumping and inverse spin Hall effect in CoFeB/IrMn heterostructures, J. Phys. D: Appl. Phys. 54, 425001 (2021).[25] E. Barati, M. Cinal, D. M. Edwards, and A. Umerski, Gilbert damping in magnetic layered systems, Phys. Rev. B 90, 014420 (2014).[26] T. Fache, J. C. Rojas-Sanchez, L. Badie, S. Mangin, and S. Petit-Watelot, Determination of spin Hall angle, spin mixing conductance, and spin diffusion length in CoFeB/Ir for spin-orbitronic devices, Phys. Rev. B 102, 064425 (2020).[27] M. Zwierzycki, Y. Tserkovnyak, P. J. Kelly, A. Brataas, and G. E. W. Bauer, First-principles study of magnetization relaxation enhancement and spin transfer in thin magnetic, Phys. Rev. B 71, 064420 (2005).[28] R. Yanes, J. Jackson, L. Udvardi, L. Szunyogh, and U. Nowak, Exchange Bias Driven by Dzyaloshinskii-Moriya Interactions, Phys. Rev. Lett. 111, 217202 (2013).[29] X. Ma, G. Yu, S. A. Razavi, L. Chang, L. Deng, Z. Chu, C. He, K. L. Wang, and X. Li, Correlation between the Dzyaloshinskii-Moriya interaction and spin-mixing conductance at an antiferromagnet/ferromagnet interface, Phys. Rev. B 98, 104428 (2018).[30] Y. Guo, J. Zhang, Q. Cui, R. Liu, Y. Ga, X. Zhan, H. Lyu, C. Hu, J. Li, J. Zhou, H. Wei, T. Zhu, H. Yang, and S. Wang, Effect of interlayer Dzyaloshinskii-Moriya interaction on spin structure in synthetic antiferromagnetic multilayers, Phys. Rev. B 105, 184405 (2022). 2image1.emfimage2.emf0 1 2 3 4 5051015202530 0Hex (mT)tCu (nm) 300K 20K-300 -200 -100 0 100 200 300-1.0-0.50.00.51.00Hinp (mT)0M (T) Cu (1.5nm) 300K 20K-300 -200 -100 0 100 200 300-1.0-0.50.00.51.0 0M (T)0Hinp (mT) 300K 5Kwithout Cu Hex(a) (b) (c) image3.emfimage4.emfimage5.pngimage6.png