Electrical determination of spin mixing conductance at metal/insulator interface using inverse spin Hall effect

R. Takahashi; R. Iguchi; K. Ando; H. Nakayama; T. Yoshino; E. Saitoh

doi:10.1063/1.3673429

Electrical determination of spin mixing conductance at metal/insulator interface using inverse spin Hall effect

R. Takahashi, R. Iguchi, K. Ando, H. Nakayama, T. Yoshino, E. Saitoh

Research output: Contribution to journal › Article › peer-review

16 Citations (Scopus)

Abstract

The spin mixing conductance at Au/BiY ₂Fe ₅O ₁₂ and Pt/BiY ₂Fe ₅O ₁₂ interfaces was electrically determined using the inverse spin Hall effect (ISHE) and the spin pumping. By measuring the electromotive force due to the ISHE and the ferromagnetic resonance spectra, we evaluated the magnitude of the generated spin currents and the magnetization-precession trajectory. The spin mixing conductance was estimated as 1.82 × 10 ¹⁸m ^-2 for the Pt/BiY ₂Fe ₅O ₁₂ film, and as 2.21 × 10 ¹⁸m ^-2 for the Au/BiY ₂Fe ₅O ₁₂ film, demonstrating efficient spin exchange at these metal/insulator interfaces.

Original language	English
Article number	07C307
Journal	Journal of Applied Physics
Volume	111
Issue number	7
DOIs	https://doi.org/10.1063/1.3673429
Publication status	Published - 2012 Apr 1

ASJC Scopus subject areas

Physics and Astronomy(all)

Access to Document

10.1063/1.3673429

Cite this

@article{5e759f3d05d54d7e802965a3a71bebb9,

title = "Electrical determination of spin mixing conductance at metal/insulator interface using inverse spin Hall effect",

abstract = "The spin mixing conductance at Au/BiY 2Fe 5O 12 and Pt/BiY 2Fe 5O 12 interfaces was electrically determined using the inverse spin Hall effect (ISHE) and the spin pumping. By measuring the electromotive force due to the ISHE and the ferromagnetic resonance spectra, we evaluated the magnitude of the generated spin currents and the magnetization-precession trajectory. The spin mixing conductance was estimated as 1.82 × 10 18m -2 for the Pt/BiY 2Fe 5O 12 film, and as 2.21 × 10 18m -2 for the Au/BiY 2Fe 5O 12 film, demonstrating efficient spin exchange at these metal/insulator interfaces.",

author = "R. Takahashi and R. Iguchi and K. Ando and H. Nakayama and T. Yoshino and E. Saitoh",

note = "Funding Information: Injection of spin currents into solids is a key ingredient for developing spintronic devices. Recent progress in the field of spintronics 1–3 revealed that the spin pumping, 4–12 generation of spin currents from magnetization precession, in a metal/ferromagnetic (or ferrimagnetic) bilayer film enables efficient spin injection into metals and semiconductors free from the impedance mismatch problem. 4 Thus, the spin pumping offers an easy and versatile route for generating spin currents, making it a powerful technique for exploring spin currents in a wide range of materials. The spin current density generated by the spin pumping at the metal/ferromagnetic (or ferrimagnetic) interface j s 0 is proportional to the product of the spin mixing conductance g r ↑ ↓ and an elliptical area of a magnetization-precession trajectory S : j s 0 ∝ g r ↑ ↓ S. 7,8 Since S is proportional to 1/ α 2 , a material with a small Gilbert damping constant α is expected to be an efficient spin current generator. Thus, a promising material for generating large spin currents is yttrium iron garnet (Y 3 Fe 5 O 12 ; YIG) because of its very small Gilbert damping constant. 13 Therefore, quantitative estimation of g r ↑ ↓ at metal/YIG interfaces is essential for developing spintronic devices, which utilizes the spin pumping. Recently, Heinrich et al. 12 reported the large g r ↑ ↓ at a Au/YIG interface. They estimated g r ↑ ↓ ≃ 1.2 × 10 18 m −2 at the interface from the damping enhancement due to the spin pumping by monitoring the ferromagnetic resonance (FMR) spectra in various frequency ranges. In this paper, we report electrical determination of g r ↑ ↓ in Au/BiY 2 Fe 5 O 12 and Pt/BiY 2 Fe 5 O 12 films using the inverse spin Hall effect (ISHE). 5,7,9,13–15 g r ↑ ↓ for the Pt/BiY 2 Fe 5 O 12 film obtained from the ISHE signal is comparable to that for the Au/BiY 2 Fe 5 O 12 film. This demonstrates efficient spin exchange not only at the Au/BiY 2 Fe 5 O 12 interface, but also at the Pt/BiY 2 Fe 5 O 12 interface. Figure 1(a) is a schematic illustration of the sample used in this study. The samples are Au/BiY 2 Fe 5 O 12 and Pt/BiY 2 Fe 5 O 12 films, where the thickness of the BiY 2 Fe 5 O 12 layer is 50 nm. The BiY 2 Fe 5 O 12 film was fabricated on a substituted Gd 3 Ga 5 O 12 (SGGG) (111) substrate by the metal-organic decomposition method. 16 The 50-nm-thick Au and 10-nm-thick Pt layers were sputtered on the top of the BiY 2 Fe 5 O 12 films in Ar atmosphere. Two electrodes are attached to both ends of the metal layer. For the measurement, the sample is placed near the center of a TE 011 microwave cavity. During the measurements, a microwave mode with frequency f = 9.44 GHz exists in the cavity, and an external magnetic field H is applied perpendicular to the direction across the electrodes [see Fig. 1(a) ]. When H and f fulfill the FMR condition, 7 magnetization precession is induced. When this precession pumps spin currents into the metal layer, an electromotive force V is generated via the ISHE. We measured V. All the measurements were performed at room temperature. Figures 1(b) and 1(c) show H dependence of the FMR spectrum dI ( H )/ dH at θ = 0 and the electromotive force V ( H ) for the Au/BiY 2 Fe 5 O 12 film at θ = 0 and 180° under the 200 mW microwave excitation, respectively. Here, I and θ denote the microwave absorption intensity and an out-of-plane magnetic field angle [see Fig. 1(a) ], respectively. The microwave absorption spectrum shows that the magnetization in the BiY 2 Fe 5 O 12 layer resonates with the microwave at H FMR = 307.2 mT. The spectral shape is well reproduced by a first derivative of a Lorentz function as shown in Fig. 1(b) , showing that inhomogeneous-magnetic-field effects on the spectrum are negligibly small in the present system. At the FMR field, a clear V appears as shown in Fig. 1(c) . The sign of V is reversed by reversing the H direction (compare the signal at θ = 0 and 180°), as expected for the ISHE induced by the spin pumping. 6,7,9 The spectral shape also supports that the V signal is attributed to the spin pumping; the shape of the V signals is well reproduced using a Lorentz function, consistent with the prediction of the dc spin pumping model. 10,17 Figure 2(a) shows H dependence of V for the Au/BiY 2 Fe 5 O 12 film at various microwave power P MW when θ are 0, 90°, and 180°, which shows that V decreases with decreasing P MW . V ISHE is proportional to P MW as shown in Fig. 2(b) . Here, V ISHE denotes the peak height of the V spectra [see Fig. 1(c) ]. This result is consistent with the phenomenological model of the dc spin pumping: 10,17 j s = g r ↑ ↓ γ 2 h 2 ℏ [ 4 π M eff γ sin 2 θ M + ( 4 π M eff ) 2 γ 2 sin 4 θ M + 4 ω 2 ] 8 π α 2 [ ( 4 π M eff ) 2 γ 2 sin 4 θ M + 4 ω 2 ] × λ N d N tanh ( d N 2 λ N ) , (1) where j s , h, θ M , γ, α , ω , λ N , d N , and 4 πM eff are the spacial averaged spin current density, the amplitude of the microwave magnetic field, the magnetization angle to the normal vector of the film plane, 5,7 the gyromagnetic ratio, the Gilbert damping constant, the microwave angular frequency, the spin-diffusion length of the metal layer, the thickness of the metal layer, and the effective demagnetizing field 11 composed of the shape-anisotropic and the perpendicular-crystal-anisotropic magnetic field, respectively. g r ↑ ↓ at the Au/BiY 2 Fe 5 O 12 interface can be estimated using Eq. (1) and the relation V = w σ e ( 2 e ℏ θ SH j s ) , (2) where θ SH , σ e , and w denote the spin Hall angle, the electric conductivity of the metal layer, and the width of the metal layer [see Fig. 1(a) ], respectively. 7,18 Using the parameters h = 0.16 mT, γ = 1.77 × 10 11 T −1 s −1 , α = 0.004, ω = 5.94 × 10 10 s −1 , 4 π M eff = 0.0584 T, λ N = 60 nm, 19 θ SH Au = 0.0035, 15 d N = 50 nm, σ e = 1.5 × 10 7 (Ωm) −1 , w = 2.5 mm, and V ISHE = 2.9 μ V, g r ↑ ↓ at the Au/BiY 2 Fe 5 O 12 interface is estimated as 2.21 × 10 18 m −2 . This result is comparable to the value obtained from the FMR measurement; 12 the combination of the spin pumping and ISHE allows the electrical determination of g r ↑ ↓ at metal/insulator interfaces. Figure 3(a) shows H dependence of V measured for the Pt/BiY 2 Fe 5 O 12 film under the 200 mW microwave excitation. The change of sign by reversing H and the shape of the V spectra are consistent with the prediction of the spin pumping and ISHE at the Pt/BiY 2 Fe 5 O 12 interface as discussed above. P MW dependence of V as shown in Fig. 3(b) also supports the V signal is due entirely to the spin pumping. Figure 3(c) shows P MW dependence of j s for the Pt/BiY 2 Fe 5 O 12 and Au/BiY 2 Fe 5 O 12 films estimated from Eq. (2) . Here, we used the spin Hall angle of the Pt layer as θ SH Pt = 0.04 (Ref. 7 ) and that of the Au layer as above, respectively. Figure 3(c) shows that j s obtained in the Pt/BiY 2 Fe 5 O 12 film is as large as that in the Au/BiY 2 Fe 5 O 12 film. Using Eq. (1) with the parameters h = 0.16 mT, γ = 1.78 × 10 11 T −1 s −1 , α = 0.004, ω = 5.94 × 10 10 s −1 , 4 πM eff = 0.0593 T, λ N = 7 nm, 19 d N = 10 nm, σ e = 2.3 × 10 6 (Ωm) −1 , w = 4.0 mm, and V ISHE = 340 μ V, g r ↑ ↓ at the Pt/BiY 2 Fe 5 O 12 interface is estimated as 1.82 × 10 18 m −2 . This is comparable to the value for the Au/BiY 2 Fe 5 O 12 film, demonstrating that the Pt/BiY 2 Fe 5 O 12 film also enables the efficient spin pumping. In summary, we demonstrated electrical determination of g r ↑ ↓ at metal/insulator interfaces using the ISHE. Considering that g r ↑ ↓ at various metal/metal interfaces are of the same order, 8,20 the order of g r ↑ ↓ at metal/insulator interfaces could be comparable to 10 18 m −2 . We found the large g r ↑ ↓ at a Pt/BiY 2 Fe 5 O 12 interface. This result will be essential for developing spintronic devices based on insulators. This work was supported by a Grant-in-Aid for Scientific Research (A) (21244058) from MEXT, Japan, the NEXT program from the Cabinet Office, Government of Japan, the Sumitomo Foundation, and Fundamental Research Grant from CREST-JST “Creation of Nanosystems with Novel Functions through Process Integration” and TRF, Japan. FIG. 1. (Color online) (a) A schematic illustration of the Au or Pt/BiY 2 Fe 5 O 12 film used in the present study (a 2.0 × 2.5 mm 2 and 2.0 × 4.0 mm 2 rectangular shape, respectively). H is the external magnetic field. θ is the external-magnetic-field angle from the normal vector of the film plane. w is the width of the metal layer. (b) H dependence of the FMR signals dI ( H )/ dH for the Au/BiY 2 Fe 5 O 12 film at θ = 0. Here, I denotes the microwave absorption intensity. The open circles are the experimental data. The green solid line shows the fitting result using a first derivative of a Lorentz function. (c) H dependence of the electromotive force V for the Au/BiY 2 Fe 5 O 12 film under the 200 mW microwave excitation, where θ are 0 and 180°. The solid lines show the fitting results using Lorentz function. Here, V ISHE is estimated as the peak height of the resonance shape in the V spectra. FIG. 2. (Color online) (a) H dependence of V at various microwave power P MW for the Au/BiY 2 Fe 5 O 12 film when θ are 0, 90°, and 180°. H FMR denotes the resonance field. (b) P MW dependence of V ISHE for the Au/BiY 2 Fe 5 O 12 at θ = 0, 90°, and 180°. FIG. 3. (Color online) (a) H dependence of V for the Pt/BiY 2 Fe 5 O 12 film under the 200 mW microwave excitation, where θ are 0 and 180°. The solid lines show the fitting results using Lorentz function for the V data. (b) P MW dependence of V ISHE for the Pt/BiY 2 Fe 5 O 12 film at θ = 0, 90°, and 180°. (c) P MW dependence of the spacial averaged spin current density j s for the Au/BiY 2 Fe 5 O 12 film and the Pt/BiY 2 Fe 5 O 12 film at θ = 0. Here, j s is obtained from Eq. (2) as a function of V ISHE , which depends on P MW shown in Fig. 2(b) and Fig. 3(b) . ",

year = "2012",

month = apr,

day = "1",

doi = "10.1063/1.3673429",

language = "English",

volume = "111",

journal = "Journal of Applied Physics",

issn = "0021-8979",

publisher = "American Institute of Physics Publising LLC",

number = "7",

}

TY - JOUR

T1 - Electrical determination of spin mixing conductance at metal/insulator interface using inverse spin Hall effect

AU - Takahashi, R.

AU - Iguchi, R.

AU - Ando, K.

AU - Nakayama, H.

AU - Yoshino, T.

AU - Saitoh, E.

N1 - Funding Information: Injection of spin currents into solids is a key ingredient for developing spintronic devices. Recent progress in the field of spintronics 1–3 revealed that the spin pumping, 4–12 generation of spin currents from magnetization precession, in a metal/ferromagnetic (or ferrimagnetic) bilayer film enables efficient spin injection into metals and semiconductors free from the impedance mismatch problem. 4 Thus, the spin pumping offers an easy and versatile route for generating spin currents, making it a powerful technique for exploring spin currents in a wide range of materials. The spin current density generated by the spin pumping at the metal/ferromagnetic (or ferrimagnetic) interface j s 0 is proportional to the product of the spin mixing conductance g r ↑ ↓ and an elliptical area of a magnetization-precession trajectory S : j s 0 ∝ g r ↑ ↓ S. 7,8 Since S is proportional to 1/ α 2 , a material with a small Gilbert damping constant α is expected to be an efficient spin current generator. Thus, a promising material for generating large spin currents is yttrium iron garnet (Y 3 Fe 5 O 12 ; YIG) because of its very small Gilbert damping constant. 13 Therefore, quantitative estimation of g r ↑ ↓ at metal/YIG interfaces is essential for developing spintronic devices, which utilizes the spin pumping. Recently, Heinrich et al. 12 reported the large g r ↑ ↓ at a Au/YIG interface. They estimated g r ↑ ↓ ≃ 1.2 × 10 18 m −2 at the interface from the damping enhancement due to the spin pumping by monitoring the ferromagnetic resonance (FMR) spectra in various frequency ranges. In this paper, we report electrical determination of g r ↑ ↓ in Au/BiY 2 Fe 5 O 12 and Pt/BiY 2 Fe 5 O 12 films using the inverse spin Hall effect (ISHE). 5,7,9,13–15 g r ↑ ↓ for the Pt/BiY 2 Fe 5 O 12 film obtained from the ISHE signal is comparable to that for the Au/BiY 2 Fe 5 O 12 film. This demonstrates efficient spin exchange not only at the Au/BiY 2 Fe 5 O 12 interface, but also at the Pt/BiY 2 Fe 5 O 12 interface. Figure 1(a) is a schematic illustration of the sample used in this study. The samples are Au/BiY 2 Fe 5 O 12 and Pt/BiY 2 Fe 5 O 12 films, where the thickness of the BiY 2 Fe 5 O 12 layer is 50 nm. The BiY 2 Fe 5 O 12 film was fabricated on a substituted Gd 3 Ga 5 O 12 (SGGG) (111) substrate by the metal-organic decomposition method. 16 The 50-nm-thick Au and 10-nm-thick Pt layers were sputtered on the top of the BiY 2 Fe 5 O 12 films in Ar atmosphere. Two electrodes are attached to both ends of the metal layer. For the measurement, the sample is placed near the center of a TE 011 microwave cavity. During the measurements, a microwave mode with frequency f = 9.44 GHz exists in the cavity, and an external magnetic field H is applied perpendicular to the direction across the electrodes [see Fig. 1(a) ]. When H and f fulfill the FMR condition, 7 magnetization precession is induced. When this precession pumps spin currents into the metal layer, an electromotive force V is generated via the ISHE. We measured V. All the measurements were performed at room temperature. Figures 1(b) and 1(c) show H dependence of the FMR spectrum dI ( H )/ dH at θ = 0 and the electromotive force V ( H ) for the Au/BiY 2 Fe 5 O 12 film at θ = 0 and 180° under the 200 mW microwave excitation, respectively. Here, I and θ denote the microwave absorption intensity and an out-of-plane magnetic field angle [see Fig. 1(a) ], respectively. The microwave absorption spectrum shows that the magnetization in the BiY 2 Fe 5 O 12 layer resonates with the microwave at H FMR = 307.2 mT. The spectral shape is well reproduced by a first derivative of a Lorentz function as shown in Fig. 1(b) , showing that inhomogeneous-magnetic-field effects on the spectrum are negligibly small in the present system. At the FMR field, a clear V appears as shown in Fig. 1(c) . The sign of V is reversed by reversing the H direction (compare the signal at θ = 0 and 180°), as expected for the ISHE induced by the spin pumping. 6,7,9 The spectral shape also supports that the V signal is attributed to the spin pumping; the shape of the V signals is well reproduced using a Lorentz function, consistent with the prediction of the dc spin pumping model. 10,17 Figure 2(a) shows H dependence of V for the Au/BiY 2 Fe 5 O 12 film at various microwave power P MW when θ are 0, 90°, and 180°, which shows that V decreases with decreasing P MW . V ISHE is proportional to P MW as shown in Fig. 2(b) . Here, V ISHE denotes the peak height of the V spectra [see Fig. 1(c) ]. This result is consistent with the phenomenological model of the dc spin pumping: 10,17 j s = g r ↑ ↓ γ 2 h 2 ℏ [ 4 π M eff γ sin 2 θ M + ( 4 π M eff ) 2 γ 2 sin 4 θ M + 4 ω 2 ] 8 π α 2 [ ( 4 π M eff ) 2 γ 2 sin 4 θ M + 4 ω 2 ] × λ N d N tanh ( d N 2 λ N ) , (1) where j s , h, θ M , γ, α , ω , λ N , d N , and 4 πM eff are the spacial averaged spin current density, the amplitude of the microwave magnetic field, the magnetization angle to the normal vector of the film plane, 5,7 the gyromagnetic ratio, the Gilbert damping constant, the microwave angular frequency, the spin-diffusion length of the metal layer, the thickness of the metal layer, and the effective demagnetizing field 11 composed of the shape-anisotropic and the perpendicular-crystal-anisotropic magnetic field, respectively. g r ↑ ↓ at the Au/BiY 2 Fe 5 O 12 interface can be estimated using Eq. (1) and the relation V = w σ e ( 2 e ℏ θ SH j s ) , (2) where θ SH , σ e , and w denote the spin Hall angle, the electric conductivity of the metal layer, and the width of the metal layer [see Fig. 1(a) ], respectively. 7,18 Using the parameters h = 0.16 mT, γ = 1.77 × 10 11 T −1 s −1 , α = 0.004, ω = 5.94 × 10 10 s −1 , 4 π M eff = 0.0584 T, λ N = 60 nm, 19 θ SH Au = 0.0035, 15 d N = 50 nm, σ e = 1.5 × 10 7 (Ωm) −1 , w = 2.5 mm, and V ISHE = 2.9 μ V, g r ↑ ↓ at the Au/BiY 2 Fe 5 O 12 interface is estimated as 2.21 × 10 18 m −2 . This result is comparable to the value obtained from the FMR measurement; 12 the combination of the spin pumping and ISHE allows the electrical determination of g r ↑ ↓ at metal/insulator interfaces. Figure 3(a) shows H dependence of V measured for the Pt/BiY 2 Fe 5 O 12 film under the 200 mW microwave excitation. The change of sign by reversing H and the shape of the V spectra are consistent with the prediction of the spin pumping and ISHE at the Pt/BiY 2 Fe 5 O 12 interface as discussed above. P MW dependence of V as shown in Fig. 3(b) also supports the V signal is due entirely to the spin pumping. Figure 3(c) shows P MW dependence of j s for the Pt/BiY 2 Fe 5 O 12 and Au/BiY 2 Fe 5 O 12 films estimated from Eq. (2) . Here, we used the spin Hall angle of the Pt layer as θ SH Pt = 0.04 (Ref. 7 ) and that of the Au layer as above, respectively. Figure 3(c) shows that j s obtained in the Pt/BiY 2 Fe 5 O 12 film is as large as that in the Au/BiY 2 Fe 5 O 12 film. Using Eq. (1) with the parameters h = 0.16 mT, γ = 1.78 × 10 11 T −1 s −1 , α = 0.004, ω = 5.94 × 10 10 s −1 , 4 πM eff = 0.0593 T, λ N = 7 nm, 19 d N = 10 nm, σ e = 2.3 × 10 6 (Ωm) −1 , w = 4.0 mm, and V ISHE = 340 μ V, g r ↑ ↓ at the Pt/BiY 2 Fe 5 O 12 interface is estimated as 1.82 × 10 18 m −2 . This is comparable to the value for the Au/BiY 2 Fe 5 O 12 film, demonstrating that the Pt/BiY 2 Fe 5 O 12 film also enables the efficient spin pumping. In summary, we demonstrated electrical determination of g r ↑ ↓ at metal/insulator interfaces using the ISHE. Considering that g r ↑ ↓ at various metal/metal interfaces are of the same order, 8,20 the order of g r ↑ ↓ at metal/insulator interfaces could be comparable to 10 18 m −2 . We found the large g r ↑ ↓ at a Pt/BiY 2 Fe 5 O 12 interface. This result will be essential for developing spintronic devices based on insulators. This work was supported by a Grant-in-Aid for Scientific Research (A) (21244058) from MEXT, Japan, the NEXT program from the Cabinet Office, Government of Japan, the Sumitomo Foundation, and Fundamental Research Grant from CREST-JST “Creation of Nanosystems with Novel Functions through Process Integration” and TRF, Japan. FIG. 1. (Color online) (a) A schematic illustration of the Au or Pt/BiY 2 Fe 5 O 12 film used in the present study (a 2.0 × 2.5 mm 2 and 2.0 × 4.0 mm 2 rectangular shape, respectively). H is the external magnetic field. θ is the external-magnetic-field angle from the normal vector of the film plane. w is the width of the metal layer. (b) H dependence of the FMR signals dI ( H )/ dH for the Au/BiY 2 Fe 5 O 12 film at θ = 0. Here, I denotes the microwave absorption intensity. The open circles are the experimental data. The green solid line shows the fitting result using a first derivative of a Lorentz function. (c) H dependence of the electromotive force V for the Au/BiY 2 Fe 5 O 12 film under the 200 mW microwave excitation, where θ are 0 and 180°. The solid lines show the fitting results using Lorentz function. Here, V ISHE is estimated as the peak height of the resonance shape in the V spectra. FIG. 2. (Color online) (a) H dependence of V at various microwave power P MW for the Au/BiY 2 Fe 5 O 12 film when θ are 0, 90°, and 180°. H FMR denotes the resonance field. (b) P MW dependence of V ISHE for the Au/BiY 2 Fe 5 O 12 at θ = 0, 90°, and 180°. FIG. 3. (Color online) (a) H dependence of V for the Pt/BiY 2 Fe 5 O 12 film under the 200 mW microwave excitation, where θ are 0 and 180°. The solid lines show the fitting results using Lorentz function for the V data. (b) P MW dependence of V ISHE for the Pt/BiY 2 Fe 5 O 12 film at θ = 0, 90°, and 180°. (c) P MW dependence of the spacial averaged spin current density j s for the Au/BiY 2 Fe 5 O 12 film and the Pt/BiY 2 Fe 5 O 12 film at θ = 0. Here, j s is obtained from Eq. (2) as a function of V ISHE , which depends on P MW shown in Fig. 2(b) and Fig. 3(b) .

PY - 2012/4/1

Y1 - 2012/4/1

N2 - The spin mixing conductance at Au/BiY 2Fe 5O 12 and Pt/BiY 2Fe 5O 12 interfaces was electrically determined using the inverse spin Hall effect (ISHE) and the spin pumping. By measuring the electromotive force due to the ISHE and the ferromagnetic resonance spectra, we evaluated the magnitude of the generated spin currents and the magnetization-precession trajectory. The spin mixing conductance was estimated as 1.82 × 10 18m -2 for the Pt/BiY 2Fe 5O 12 film, and as 2.21 × 10 18m -2 for the Au/BiY 2Fe 5O 12 film, demonstrating efficient spin exchange at these metal/insulator interfaces.

AB - The spin mixing conductance at Au/BiY 2Fe 5O 12 and Pt/BiY 2Fe 5O 12 interfaces was electrically determined using the inverse spin Hall effect (ISHE) and the spin pumping. By measuring the electromotive force due to the ISHE and the ferromagnetic resonance spectra, we evaluated the magnitude of the generated spin currents and the magnetization-precession trajectory. The spin mixing conductance was estimated as 1.82 × 10 18m -2 for the Pt/BiY 2Fe 5O 12 film, and as 2.21 × 10 18m -2 for the Au/BiY 2Fe 5O 12 film, demonstrating efficient spin exchange at these metal/insulator interfaces.

UR - http://www.scopus.com/inward/record.url?scp=84861742506&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84861742506&partnerID=8YFLogxK

U2 - 10.1063/1.3673429

DO - 10.1063/1.3673429

M3 - Article

AN - SCOPUS:84861742506

VL - 111

JO - Journal of Applied Physics

JF - Journal of Applied Physics

SN - 0021-8979

IS - 7

M1 - 07C307

ER -

Electrical determination of spin mixing conductance at metal/insulator interface using inverse spin Hall effect

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this