This paper was converted on www.awesomepapers.org from LaTeX by an anonymous user.
Want to know more? Visit the Converter page.

Voltage-controlled significant spin wave Doppler shift in FM/FE heterojunction

Shaojie Hu Department of Physics, Kyushu University, 744 Motooka, Fukuoka, 819-0395, Japan    Kang Wang Center for Spintronics and Quantum Systems, State Key Laboratory for Mechanical Behavior of Materials, School of Materials Science and Engineering, Xi’an Jiaotong University, Xi’an, Shaanxi, 710049, China    Tai Min Center for Spintronics and Quantum Systems, State Key Laboratory for Mechanical Behavior of Materials, School of Materials Science and Engineering, Xi’an Jiaotong University, Xi’an, Shaanxi, 710049, China    Takashi Kimura Department of Physics, Kyushu University, 744 Motooka, Fukuoka, 819-0395, Japan
Abstract

Efficient manipulation of spin waves (SWs) is considered as one of the promising means for encoding information with low power consumption in next generation spintronic devices. The SW Doppler shift is one important phenomenon by manipulating SWs propagation. Here, we predict an efficient way to control the SW Doppler shift by voltage control magnetic anisotropy boundary (MAB) movement in FM/FE heterojunction. From the micromagnetic simulation, we verified that the SW Doppler shift aligns well with our theoretical predictions in Fe/BaTiO3\rm BaTiO_{3} heterostructure. The SW Doppler shift also shows ultra wide-band shift (over 5 GHz) property with voltage. Such efficient SW Doppler shift may provide one possible way to measure the Hawking radiation of an analogue black hole in the SW systems.

Magnonics, a burgeoning field rooted in the exploration of quantum magnetic dynamic phenomena, is primarily dedicated to the in-depth research and practical application of spin waves (SWs). These SWs hold promise for advanced information processing at the nanoscale, potentially revolutionizing how we store and compute data with ultra-low power consumption.neusser2009magnonics ; 2010Kruglyak ; 2010Serga ; 2011LENK ; demokritov2012magnonics ; barman20212021 ; pirro2021advances Efficiently controlling the spin wave propagation offers a potential alternative for encoding information beyond CMOS computing due to their absence of Joule heating during propagation, scalability to atomic dimensions, and diverse nonlinear and nonreciprocal phenomena.2005Kostylev ; 2008Lee ; 2014Klingler ; 2020Dieny Much like sound or light waves, SWs inherently exhibit wave properties, including the Doppler effect that denotes frequency shifts due to relative motion between the source and observer. Recent studies have highlighted numerous innovative SW Doppler shift occurrences. Stancil, et al. reported the experimental observation of an inverse Doppler shift from left-handed dipolar SWs with a moving pick-up antenna, which is the intrinsic characteristic of left-handed, or backward SWs.2006Stancil Vlaminck and Bailleul found the current-induced SW Doppler shift that related to the adiabatic spin-transfer torque, which can be used as a probe of spin-polarized transport in various metallic ferromagnets.2008Vlaminck Meanwhile, Yu, et al. predicted the SW Doppler shift by magnon drag in magnetic insulators.2021Yu In parallel, Nakane and Kohno studied the current-induced SW Doppler shift in antiferromagnets.2021Jotaro So, researchers have hypothesized that it’s possible to create magnonic black-hole and white-hole horizons using an efficient method involving spin current-driven SW Doppler shifts. This means that the SW Doppler shift might not only provide a mechanism for simulating gravitational phenomena, but also offer a way to study analogue gravity and even measure the Hawking radiation emitted by an analogous black hole.jannes2011hawking ; roldan2017magnonic The intriguing nature of these phenomena has significantly heightened research interest in the SW Doppler shift, pushing scientists and scholars to delve deeper into its intricacies and potential applications.

So far, the preponderance of investigations into the SW Doppler shift has been concentrated on individual ferromagnetic materials that are modulated through electric current control. Such a constrained approach inadvertently poses limitations on realizing the aspirational goal of ultra-low power consumption pivotal for the nuanced manipulation of SWs. Compounding this challenge, it has been observed that the SW Doppler shift frequencies are consistently far from the GHz benchmark. In the broader academic discourse, the magnetoelectric effect emerges as a promising avenue, widely regarded for its efficacy in directing the propagation of SWs. 2018Balinskiy ; 2018Sadovnikov ; 2018Rana ; 2019Rana ; 2021Qin ; wang2022dual This is achieved with a commendable reduction in power consumption, especially when deployed within multiferroic materials. In the pantheon of multiferroic systems, the artificial ferromagnetic/ferroelectric (FM/FE) heterojunction stands as an archetypal representative.sahoo2007ferroelectric ; venkataiah2012strain ; Lahtinen2012 ; venkataiah2013strain ; gorige2017magnetization ; Yamada2021Electric ; Usami2021 ; Fujii2022Giant ; hu2023efficient Within this context, we present a pronounced SW Doppler shift attributed to the motion of the magnetic anisotropy boundary (MAB) inherent in the FM/FE heterojunction, a process which is deftly modulated by voltage.

Refer to caption
Figure 1: The schematic of SW Doppler shift induced by the voltage-controlled MAB movement in FM/FE heterojunction. The out-of-plane and in-plane polarization domains in the ferroelectric layer transfer different strains to the ferromagnetic layer. The ferromagnetic layer exhibits different anisotropy regions owing to inverse magnetostriction. (a) A case of SW Doppler red shift caused by a positive voltage. (b) A case of SW Doppler blue shift caused by a negative voltage.

Figure 1 provides a schematic representation of SW Doppler shift precipitated by the voltage-controlled MAB movement in FM/FE heterojunction. Within this configuration, the ferromagnetic layer, positioned atop a ferroelectric substrate with both in-plane and out-of-plane polarized domains, manifests two distinct magnetic anisotropy regions: A1 and A2. Varied magnetic anisotropies result from the strain-induced inverse magnetostriction effect across distinct polarized domains.2012Lahtinen The MAB separating the two magnetic anisotropy regions is pinned on the ferroelectric domain wall and parallel to y axis. A positive (negative) voltage-controlled perpendicular electric field can drive the motion of the ferroelectric domain wall along x (-x) axis as well as the MAB.2015Franke The frequencies of reflected and transmitted SWs will show the red or blue shift by moving MAB. This letter presents an analytical model of the SW Doppler shift, which is further validated through magnetic simulation.

For simplicity, we assume that the MAB moves at a constant velocity vMABv_{\mathrm{MAB}}. The incident SW propagates along x axis and perpendicular to the MAB. In the observer’s system S which is stationary with respect to the lab frame, the general dispersion relations of SWs in A1 and A2 are described as different functions f1=F1(𝐤)\rm f_{1}=F_{1}(\mathbf{k}) and f2=F2(𝐤)\rm f_{2}=F_{2}(\mathbf{k}) with wave vector 𝐤\mathbf{k}. In the moving system S’ which is stationary with respect to the MAB, the dispersion functions f1=F1(𝐤)\rm f_{1}^{\prime}=F_{1}^{\prime}(\mathbf{k}) and f2=F2(𝐤\rm f_{2}^{\prime}=F_{2}^{\prime}(\mathbf{k}) with vector 𝐤\mathbf{k^{\prime}} in A1 and A2 are given as follows by Lorentz transformation:

fi=1,2\displaystyle f^{\prime}_{i=1,2} =Fi=1,2(𝐤)=Γ(Fi=1,2(𝐤)𝐤𝐯MAB2π),𝐤i=1,2\displaystyle=F^{\prime}_{i=1,2}(\mathbf{k^{\prime}})=\Gamma\left(F_{i=1,2}(\mathbf{k^{\prime}})-\frac{\mathbf{k^{\prime}}\cdot\mathbf{v}_{\mathrm{MAB}}}{2\pi}\right),\ \mathbf{k^{\prime}}_{i=1,2} =Γ(𝐤i=1,22πfi=1,2𝐯MABc2)\displaystyle=\Gamma\left(\mathbf{k}_{i=1,2}-\frac{2\pi f_{i=1,2}\mathbf{v}_{\mathrm{MAB}}}{c^{2}}\right) (1)

where Γ=(1vMAB2/s2)12\Gamma=\left(1-v_{\mathrm{MAB}}^{2}/s^{2}\right)^{-\frac{1}{2}}, ss is the maximal group velocity of spin waves.kim2014propulsion Under the limitation of vMABv_{\mathrm{MAB}} \ll  ss, the ff^{\prime} and 𝐤\mathbf{k}^{\prime} can be written as the following:

fi=1,2\displaystyle f^{\prime}_{i=1,2} =Fi=1,2(𝐤)\displaystyle=F^{\prime}_{i=1,2}(\mathbf{k^{\prime}}) Fi=1,2(𝐤)𝐤𝐯MAB2π,𝐤i=1,2\displaystyle\approx F_{i=1,2}(\mathbf{k^{\prime}})-\frac{\mathbf{k^{\prime}}\cdot\mathbf{v}_{\mathrm{MAB}}}{2\pi},\ \mathbf{k}^{\prime}_{i=1,2} 𝐤i=1,2\displaystyle\approx\mathbf{k}_{i=1,2} (2)

The frequencies and wave vectors of incident, reflected and transmitted SWs in system S and S’ are defined as: f0f_{0}, 𝐤𝟎\mathbf{k_{0}}, frf_{\mathrm{r}}, 𝐤r\mathbf{k}_{\mathrm{r}}, ftf_{\mathrm{t}}, 𝐤t\mathbf{k}_{\mathrm{t}}, f0f^{\prime}_{0}, 𝐤0\mathbf{k}^{\prime}_{0}, frf^{\prime}_{\mathrm{r}}, 𝐤r\mathbf{k}^{\prime}_{\mathrm{r}}, ftf^{\prime}_{\mathrm{t}} and 𝐤t\mathbf{k}^{\prime}_{\mathrm{t}}, respectively. If the incident SW from the A1 region propagates to A2 shown in Fig.1, the frequencies of reflected and transmitted SWs in system S can be derived as follows:

fr\displaystyle f_{\mathrm{r}} =Γ(fr+𝐤r𝐯MAB2π)\displaystyle=\Gamma\left(f^{\prime}_{\mathrm{r}}+\frac{\mathbf{k}^{\prime}_{\mathrm{r}}\cdot\mathbf{v}_{\mathrm{MAB}}}{2\pi}\right) (3)
f0(𝐤0𝐤r)𝐯MAB2π\displaystyle\approx f_{0}-(\mathbf{k}_{0}-\mathbf{k}^{\prime}_{\mathrm{r}})\cdot\frac{\mathbf{v}_{\mathrm{MAB}}}{2\pi}
ft\displaystyle f_{\mathrm{t}} =Γ(ft+𝐤t𝐯MAB2π)\displaystyle=\Gamma\left(f^{\prime}_{\mathrm{t}}+\frac{\mathbf{k}^{\prime}_{\mathrm{t}}\cdot\mathbf{v}_{\mathrm{MAB}}}{2\pi}\right) (4)
f0(𝐤0𝐤t)𝐯MAB2π\displaystyle\approx f_{0}-(\mathbf{k}_{0}-\mathbf{k}^{\prime}_{\mathrm{t}})\cdot\frac{\mathbf{v}_{\mathrm{MAB}}}{2\pi}

where, incident frequency f0f_{0} is known, 𝐤0\mathbf{k}_{0} can be obtained from the dispersion relation of SWs in the incident region. 𝐤r\mathbf{k}^{\prime}_{\mathrm{r}} and 𝐤t\mathbf{k}^{\prime}_{\mathrm{t}} can be obtained from Eq. (2), respectively. So the frequency shift of SW Doppler effect is dominated by the second terms of Eq. (3) and Eq. (4), which not only relate to the velocity of MAB but also the deviation of the incident and reflected (or transmitted) wave vectors determined by different magnetic anisotropy.

Refer to caption
Figure 2: (a) The cubic anisotropy is induced in Fe\mathrm{Fe} layer on ferroelectric BaTiO3\mathrm{BaTiO_{3}} c-domain. The two white double-head arrows represent easy axes of cubic anisotropy along with 4545^{\circ} to y axis. (b) The uniaxial anisotropy is induced in Fe\mathrm{Fe} layer on ferroelectric BaTiO3\mathrm{BaTiO_{3}} a-domain. The white double-head arrows represent the easy axis of uniaxial anisotropy along y axis. The image plots of the dispersion relation of SWs in cubic anisotropy region (c) and uniaxial anisotropy region (d) by micromagnetic simulation. The dashed lines correspond to the theory calculation data.

To validate such SW Doppler shift, we conducted the micromagnetic simulation in a well-build artificial multiferroic system Fe/BaTiO3\mathrm{BaTiO_{3}} 2006Duan ; 2014Radaelli by using the GPU-accelerated micromagnetic software Mumax32014Vansteenkiste . Figure 2(a) and (b) show the Fe film on ferroelectric BaTiO3\mathrm{BaTiO_{3}} substrate with in-plane and out-of-plane polarization domains exhibits uniaxial and cubic magnetic anisotropy.2012Lahtinen In the simulation, a 6000×400×2nm36000\times 400\times 2\ \mathrm{nm}^{3} Fe film is discretized using 3000×200×13000\times 200\times 1 finite difference cells. The periodic boundary condition is used along y axis to avoid the boundary effect. The simulation parameters used are as follows2015Franke ; 2021Qin : saturation magnetization Ms=1.7×106M_{\mathrm{s}}=1.7\times 10^{6} A/m, exchange constant Aex=2.1×1011A_{\mathrm{ex}}=2.1\times 10^{-11} J/m, Gilbert damping α\alpha = 0.003, cubic anisotropy constant Kc=4.4×104J/m3K_{\mathrm{c}}=4.4\times 10^{4}\ \mathrm{J/m^{3}} and uniaxial anisotropy constant Ku=1.5×104J/m3K_{\mathrm{u}}=1.5\times 10^{4}\ \mathrm{J/m^{3}}. To prevent SW reflection at both ends, α\alpha is gradually increased from 0.003 to 0.033 at both end regions of the Fe film (3000-3000 nm < x2200-2200 nm, 22002200 nm < x30003000 nm). An external magnetic field μ0Hext=100mT\mu_{0}H_{\mathrm{ext}}=100\ \mathrm{mT} is applied to magnetize the Fe film along y axis. To obtain the dispersion relation of SWs, we perform two dimensional Fourier transform2012Kumar ; 2013Venkat on mx(x,t)m_{x}(x,t) in response to a sinc-based exciting field h0sinc(2πfc(tt0))e^x\mathrm{h}_{0}\mathrm{sinc}(2\pi f_{\mathrm{c}}(t-t_{0}))\hat{e}_{x} with μ0h0=10mT\mu_{0}h_{0}=10\ \mathrm{mT}, cutoff frequency fc=50GHzf_{c}=50\ \mathrm{GHz} and t0=5t_{0}=5 ns, at the center section (6×400×2nm36\times 400\times 2\ \mathrm{nm}^{3}) of Fe film. The dispersion relations are shown in Fig. 2(c) and (d) for cubic and uniaxial regions, respectively. The red dashed lines represent the theoretical calculations from the dispersion relation equations of Damon–Eshbach (DE) SWs in cubic and uniaxial anisotropy regions as follow1986Kalinikos ; wang2022dual :

fcu=Fcu(k)=γμ02π(Hext2Kcμ0Ms+2Aexμ0Msk2)(Hext+Kcμ0Ms+Ms+2Aexμ0Msk2)+Ms2Ff_{cu}=F_{cu}(k)=\frac{\gamma\mu_{0}}{2\pi}\sqrt{\left(H_{\mathrm{ext}}-\frac{2K_{\mathrm{c}}}{\mu_{0}M_{\mathrm{s}}}+\frac{2A_{\mathrm{ex}}}{\mu_{0}M_{\mathrm{s}}}k^{2}\right)\left(H_{\mathrm{ext}}+\frac{K_{\mathrm{c}}}{\mu_{0}M_{\mathrm{s}}}+M_{\mathrm{s}}+\frac{2A_{\mathrm{ex}}}{\mu_{0}M_{\mathrm{s}}}k^{2}\right)+M^{2}_{\mathrm{s}}F} (5)
fku=Fku(k)=γμ02π(Hext+2Kuμ0Ms+2Aexμ0Msk2)(Hext+2Kuμ0Ms+Ms+2Aexμ0Msk2)+Ms2Ff_{ku}=F_{ku}(k)=\frac{\gamma\mu_{0}}{2\pi}\sqrt{\left(H_{\mathrm{ext}}+\frac{2K_{\mathrm{u}}}{\mu_{0}M_{\mathrm{s}}}+\frac{2A_{\mathrm{ex}}}{\mu_{0}M_{\mathrm{s}}}k^{2}\right)\left(H_{\mathrm{ext}}+\frac{2K_{\mathrm{u}}}{\mu_{0}M_{\mathrm{s}}}+M_{\mathrm{s}}+\frac{2A_{\mathrm{ex}}}{\mu_{0}M_{\mathrm{s}}}k^{2}\right)+M^{2}_{\mathrm{s}}F} (6)

where, γ\gamma is gyromagnetic ratio, μ0\mu_{0} is vacuum permeability, F=1e|k|d|k|d(11e|k|d|k|d)F=\frac{1-\mathrm{e}^{-\lvert k\rvert d}}{\lvert k\rvert d}(1-\frac{1-\mathrm{e}^{-\lvert k\rvert d}}{\lvert k\rvert d}), dd is the thickness of Fe film. Obviously, the variance in anisotropy constants between uniaxial and cubic regions yields distinct dispersion relations.

Refer to caption
Figure 3: (a) The schematic of reflected and transmitted SWs by moving MAB with voltage. The MAB moves along x axis between 1100-1100 nm and 1900 nm marked by the blue dash lines. The black square (A) represents the cell at 1202-1202 nm < x1200-1200 nm, 2-2 nm < y0 nm, where mx(t)m_{x}(t) is recorded as the incident and reflected SWs. The green square (B) represents the cell at 19981998 nm < x20002000 nm, 2-2 nm < y0 nm, where mx(t)m_{x}(t) is recorded as the transmitted SW. (b) The snapshot of mxm_{x} taken at y=1y=-1 nm with the static MAB at x = 400 nm. The red line is the position of exciting source and the black dashed line is the position of MAB. (c) The spectra of reflected and incident SWs when vMAB=200v_{\mathrm{MAB}}=-200, 0 and 200 m/s. (d) The spectra of transmitted SW when vMAB=200v_{\mathrm{MAB}}=-200, 0 and 200 m/s.

We then simulate the reflection and transmission of SWs by moving MAB, as depicted in Fig. 3(a). The exciting source of SWs is 1.478 μm\mathrm{\mu m} from the left end with a size of 42×400×2nm342\times 400\times 2\ \mathrm{nm}^{3}. The SW is excited by a consistent Sine function field μ0h0sin(2πf0t)e^x\mu_{0}\mathrm{h}_{0}\mathrm{sin}(2\pi f_{\mathrm{0}}t)\hat{e}_{x} (μ0h0=10mT\rm\mu_{0}h_{0}=10\ \mathrm{mT}, f0=20GHzf_{0}=20\ \mathrm{GHz}) under the external magnetic field μ0Hext=100mT\mu_{0}H_{\mathrm{ext}}=100\ \mathrm{mT}. The excited SW propagates in Fe film along x axis perpendicular to the MAB. The width of MAB same as the pinned ferroelectric domain wall (2 - 5 nm 1992Zhang ; 2006Hlinka ; 2006ZhangQingsong ) is far smaller than the wavelength of SWs and we neglect it in simulation. The MAB moves along x axis between x = 1100-1100 nm and x = 1900 nm marked by the blue dash lines in Fig. 3(a). It has been confirmed the velocity of ferroelectric domain wall driven by electric field can be up to 1000 m/s in BaTiO3\rm BaTiO_{3}.1963Stadler ; 2017Boddu And, the velocity of MAB (vMABv_{\mathrm{MAB}}) could be calculated by Miller-Weinreich theory as following equation1963Stadler ; tagantsev2010domains :

vMAB(U)=vn=1ne(δ/(U/D))n32v_{\mathrm{MAB}}(U)=v_{\infty}\sum_{n=1}^{\infty}n\mathrm{e}^{-(\delta/(U/D))n^{\frac{3}{2}}} (7)

where UU is the voltage to provide perpendicular electric field in BaTiO3\rm BaTiO_{3} , vv_{\infty} = 5 m/s, δ\delta = 4 kV/cm, the thickness of BaTiO3\rm BaTiO_{3} substrate DD = 100 nm. So, a significant velocity of MAB could be realized by a much lower voltage.

Refer to caption
Figure 4: The frequencies of reflected and transmitted SWs as a function vMABv_{\mathrm{MAB}} (voltage) in Fe/BiTiO3\rm BiTiO_{3} at the exciting frequency of 20 GHz. (a) The SWs incident from cubic to uniaxial region. The positive voltage provides the positive velocity of MAB. (b) The SWs incident from uniaxial to the cubic regions. The negative voltage provides the positive velocity of MAB. The symbols are corresponding to the simulation results. The solid lines correspond to the theoretical calculation.

In the simulation, we implement MAB moving at velocity vMABv_{\mathrm{MAB}} by shifting it over one discretization cell (δx=2\delta x=2 nm) every time interval δt=δx/vMAB\delta t=\delta x/v_{\mathrm{MAB}}.2015Franke ; 2016Wiele The mxm_{x} in the cell of 1202-1202 nm < x1200-1200 nm, 2-2 nm < y0 nm is recorded as incident and reflected SWs, while the mxm_{x} in the cell of 19981998 nm < x20002000 nm, 2-2 nm < y0 nm is recorded as transmitted SW, marked by the black and green squares A, B in Fig. 3(a), respectively. Figure 3(b) shows the snapshot of mxm_{x} taken at y = 1-1 nm with static MAB at x = 400 nm. The black dashed line indicates the position of MAB. Clearly, the SWs have different wavelengths in cubic and uniaxial regions. To accurately determine all the related frequencies of the SWs, we perform a fast Fourier transform on mx(t)m_{x}(t) at A and B to obtain the spectra of reflected and transmitted SWs. Figure 3(c) shows the spectra of the incident and reflected SWs at various vMABv_{\mathrm{MAB}}. At vMABv_{\mathrm{MAB}} = 0 m/s, a single 20 GHz peak in the spectrum (black solid curve) indicates the reflected SW frequency matches exciting frequency. However, for vMABv_{\mathrm{MAB}} = 200 m/s, the spectrum (red dashed curve) displays peaks at 20 GHz and 17.07 GHz. The latter peak, the reflected SW, shows a significant red shift of the Doppler effect. With a negative voltage setting vMABv_{\mathrm{MAB}} = 200-200 m/s, reflected SW frequency blue shifts to 24 GHz. Figure 3(d) shows the spectra of transmitted SW for various MAB velocities. They all have a single peak due to the solitary transmitted spin wave in the B. The transmitted SW frequencies display a Doppler red shift to 19.4 GHz and blue shift to 20.4 GHz at vMABv_{\mathrm{MAB}} = 200 and 200-200 m/s respectively. Notably, these Doppler shifts are much lower than those of the reflected SWs, which hold opposing wave vectors to the incident SW.

To comprehensively analyze the Doppler shift, we simulate the SW Doppler shift across a broader range of velocities (voltages). Figure 4(a) plots both simulated and theoretical Doppler shifts of reflected and transmitted SWs’ frequencies at the exciting frequency of 20 GHz in the cubic region. The simulation aligns closely with theoretical predictions. The frequencies of reflected and transmitted SWs show a monotonic shift with the velocity (voltage). This also means all the reflected and transmitted SWs show a red shift when the MAB is far away from the source of the SW. Meanwhile, the SWs show a blue shift when the MAB is close to the SW source. In addition, the SW Doppler shift can exceed several GHz, nearly two orders greater than the electric current control methods. It’s also interesting the transmission SW shows a different velocity-dependent tendency when the incident SW from uniaxial region to cubic region shown in Fig. 4(b). Here, the transmitted SWs show a blue shift when the MAB is far away from the SW source, and a red shift when the MAB is close to the source. Such unusual Doppler shift comes from the negative value of 𝐤0𝐤𝐭\mathbf{k}_{0}-\mathbf{k_{t}}^{\prime} based on the Eq.4. These results also indicate that the type of transmitted SW Doppler shift is influenced not only by the MAB’s movement direction but also by the deviation between the two different regions.

In summary, we predict an efficient way to control the SW Doppler shift by voltage control MAB movement in FM/FE heterojunction. The SW Doppler shift could be modulated in an ultra-wide-band range with voltage. The micromagnetic simulation results align closely with theoretical predictions. Such a significant SW Doppler shift may provide one possible system to do analogue gravity and measure the Hawking radiation of an analogue black hole. On the other hand, the SW Doppler shift could be employed as an accurate probe of the ferroelectric domain wall motion controlled by electric field.

Acknowledgements.
This work is partially supported by National JSPS Program for Grant-in-Aid for Scientific Research (S)(21H05021), and Challenging Exploratory Research (17H06227) and JST CREST (JPMJCR18J1). Natural Science Foundation of Shaanxi Province (2021JM-022).

References

  • [1] Sebastian Neusser and Dirk Grundler. Magnonics: Spin waves on the nanoscale. Advanced materials, 21(28):2927–2932, 2009.
  • [2] V V Kruglyak, S O Demokritov, and D Grundler. Magnonics. Journal of Physics D: Applied Physics, 43(26):260301, 2010.
  • [3] A A Serga, A V Chumak, and B Hillebrands. YIG magnonics. Journal of Physics D: Applied Physics, 43(26):264002, 2010.
  • [4] B. Lenk, H. Ulrichs, F. Garbs, and M. Münzenberg. The building blocks of magnonics. Physics Reports, 507(4):107–136, 2011.
  • [5] Sergej O Demokritov and Andrei N Slavin. Magnonics: From fundamentals to applications, volume 125. Springer Science & Business Media, 2012.
  • [6] Anjan Barman, Gianluca Gubbiotti, Sam Ladak, Adekunle Olusola Adeyeye, Maciej Krawczyk, Joachim Gräfe, Christoph Adelmann, Sorin Cotofana, Azad Naeemi, Vitaliy I Vasyuchka, et al. The 2021 magnonics roadmap. Journal of Physics: Condensed Matter, 33(41):413001, 2021.
  • [7] Philipp Pirro, Vitaliy I Vasyuchka, Alexander A Serga, and Burkard Hillebrands. Advances in coherent magnonics. Nature Reviews Materials, 6(12):1114–1135, 2021.
  • [8] M. P. Kostylev, A. A. Serga, T. Schneider, B. Leven, and B. Hillebrands. Spin-wave logical gates. Applied Physics Letters, 87(15):153501, 2005.
  • [9] Ki-Suk Lee and Sang-Koog Kim. Conceptual design of spin wave logic gates based on a Mach–Zehnder-type spin wave interferometer for universal logic functions. Journal of Applied Physics, 104(5):053909, 2008.
  • [10] S. Klingler, P. Pirro, T. Brächer, B. Leven, B. Hillebrands, and A. V. Chumak. Design of a spin-wave majority gate employing mode selection. Applied Physics Letters, 105(15):152410, 2014.
  • [11] B. Dieny, I. L. Prejbeanu, K. Garello, P. Gambardella, P. Freitas, R. Lehndorff, W. Raberg, U. Ebels, S. O. Demokritov, J. Akerman, A. Deac, P. Pirro, C. Adelmann, A. Anane, A. V. Chumak, A. Hirohata, S. Mangin, Sergio O. Valenzuela, M. Cengiz Onbaşlı, M. d’Aquino, G. Prenat, G. Finocchio, L. Lopez-Diaz, R. Chantrell, O. Chubykalo-Fesenko, and P. Bortolotti. Opportunities and challenges for spintronics in the microelectronics industry. Nature Electronics, 3(8):446–459, 2020.
  • [12] Daniel D. Stancil, Benjamin E. Henty, Ahmet G. Cepni, and J. P. Van’t Hof. Observation of an inverse Doppler shift from left-handed dipolar spin waves. Phys. Rev. B, 74:060404, 2006.
  • [13] Vincent Vlaminck and Matthieu Bailleul. Current-Induced Spin-Wave Doppler Shift. Science, 322(5900):410–413, 2008.
  • [14] Tao Yu, Chen Wang, Michael A. Sentef, and Gerrit E. W. Bauer. Spin-Wave Doppler Shift by Magnon Drag in Magnetic Insulators. Phys. Rev. Lett., 126:137202, 2021.
  • [15] Jotaro J. Nakane and Hiroshi Kohno. Current-Induced Spin-Wave Doppler Shift in Antiferromagnets. Journal of the Physical Society of Japan, 90(10):103705, 2021.
  • [16] Gil Jannes, Philippe Maïssa, Thomas G Philbin, and Germain Rousseaux. Hawking radiation and the boomerang behavior of massive modes near a horizon. Physical Review D, 83(10):104028, 2011.
  • [17] A Roldán-Molina, Alvaro S Nunez, and RA Duine. Magnonic black holes. Physical review letters, 118(6):061301, 2017.
  • [18] M. Balinskiy, A. C. Chavez, A. Barra, H. Chiang, G. P. Carman, and A. Khitun. Magnetoelectric Spin Wave Modulator Based On Synthetic Multiferroic Structure. Sci Rep, 8(1):10867, 2018.
  • [19] A. V. Sadovnikov, A. A. Grachev, S. E. Sheshukova, Y. P. Sharaevskii, A. A. Serdobintsev, D. M. Mitin, and S. A. Nikitov. Magnon Straintronics: Reconfigurable Spin-Wave Routing in Strain-Controlled Bilateral Magnetic Stripes. Phys Rev Lett, 120(25):257203, 2018.
  • [20] Bivas Rana and YoshiChika Otani. Voltage-Controlled Reconfigurable Spin-Wave Nanochannels and Logic Devices. Phys. Rev. Applied, 9:014033, 2018.
  • [21] Bivas Rana and YoshiChika Otani. Towards magnonic devices based on voltage-controlled magnetic anisotropy. Communications Physics, 2(1):90, 2019.
  • [22] Huajun Qin, Rouven Dreyer, Georg Woltersdorf, Tomoyasu Taniyama, and Sebastiaan van Dijken. Electric-Field Control of Propagating Spin Waves by Ferroelectric Domain-Wall Motion in a Multiferroic Heterostructure. Advanced Materials, 33(27):2100646, 2021.
  • [23] Kang Wang, Shaojie Hu, Fupeng Gao, Miaoxin Wang, and Dawei Wang. Dual function spin-wave logic gates based on electric field control magnetic anisotropy boundary. Applied Physics Letters, 120(14):142405, 2022.
  • [24] Sarbeswar Sahoo, Srinivas Polisetty, Chun-Gang Duan, Sitaram S Jaswal, Evgeny Y Tsymbal, and Christian Binek. Ferroelectric control of magnetism in BaTiO3/ Fe heterostructures via interface strain coupling. Physical Review B, 76(9):092108, 2007.
  • [25] G Venkataiah, Y Shirahata, I Suzuki, M Itoh, and T Taniyama. Strain-induced reversible and irreversible magnetization switching in Fe/BaTiO3 heterostructures. Journal of Applied Physics, 111(3), 2012.
  • [26] Tuomas HE Lahtinen, Yasuhiro Shirahata, Lide Yao, Kévin JA Franke, Gorige Venkataiah, Tomoyasu Taniyama, and Sebastiaan van Dijken. Alternating domains with uniaxial and biaxial magnetic anisotropy in epitaxial Fe films on BaTiO3. Applied Physics Letters, 101(26), 2012.
  • [27] G Venkataiah, E Wada, H Taniguchi, M Itoh, and T Taniyama. Electric-voltage control of magnetism in Fe/BaTiO3 heterostructured multiferroics. Journal of Applied Physics, 113(17), 2013.
  • [28] Venkataiah Gorige, Anupama Swain, Katsuyoshi Komatsu, Mitsuru Itoh, and Tomoyasu Taniyama. Magnetization reversal in Fe/BaTiO3 (110) heterostructured multiferroics. physica status solidi (RRL)–Rapid Research Letters, 11(11):1700294, 2017.
  • [29] S Yamada, Y Teramoto, D Matsumi, D Kepaptsoglou, I Azaceta, T Murata, K Kudo, VK Lazarov, T Taniyama, and K Hamaya. Electric field tunable anisotropic magnetoresistance effect in an epitaxial Co2FeSi/BaTiO3 interfacial multiferroic system. Physical Review Materials, 5(1):014412, 2021.
  • [30] T Usami, S Fujii, S Yamada, Y Shiratsuchi, R Nakatani, and K Hamaya. Giant magnetoelectric effect in an L21-ordered Co2FeSi/Pb(Mg1/3Nb2/3)O3PbTiO3Co_{2}FeSi/Pb(Mg_{1/3}Nb_{2/3})O_{3}-PbTiO_{3} multiferroic heterostructure. Applied Physics Letters, 118(14), 2021.
  • [31] Shumpei Fujii, Takamasa Usami, Yu Shiratsuchi, Adam M Kerrigan, Amran Mahfudh Yatmeidhy, Shinya Yamada, Takeshi Kanashima, Ryoichi Nakatani, Vlado K Lazarov, Tamio Oguchi, et al. Giant converse magnetoelectric effect in a multiferroic heterostructure with polycrystalline Co2FeSi. NPG Asia Materials, 14(1):43, 2022.
  • [32] Shaojie Hu, Shinya Yamada, Po-Chun Chang, Wen-Chin Lin, Kohei Hamaya, and Takashi Kimura. Efficient Electrical Manipulation of the Magnetization Process in an Epitaxially Controlled Co2FeSi/BaTiO3Co_{2}FeSi/BaTiO_{3} Multiferroic Interface. Physical Review Applied, 20(3):034029, 2023.
  • [33] Tuomas H. E. Lahtinen, Yasuhiro Shirahata, Lide Yao, Kévin J. A. Franke, Gorige Venkataiah, Tomoyasu Taniyama, and Sebastiaan van Dijken. Alternating domains with uniaxial and biaxial magnetic anisotropy in epitaxial Fe films on BaTiO3\mathrm{BaTiO_{3}}. Applied Physics Letters, 101(26):262405, 2012.
  • [34] Kévin J. A. Franke, Ben Van de Wiele, Yasuhiro Shirahata, Sampo J. Hämäläinen, Tomoyasu Taniyama, and Sebastiaan van Dijken. Reversible Electric-Field-Driven Magnetic Domain-Wall Motion. Phys. Rev. X, 5:011010, 2015.
  • [35] Se Kwon Kim, Yaroslav Tserkovnyak, and Oleg Tchernyshyov. Propulsion of a domain wall in an antiferromagnet by magnons. Physical Review B, 90(10):104406, 2014.
  • [36] Chun-Gang Duan, S. S. Jaswal, and E. Y. Tsymbal. Predicted Magnetoelectric Effect in Fe/BaTiO3\mathrm{Fe}/{\mathrm{BaTiO}}_{3} Multilayers: Ferroelectric Control of Magnetism. Phys. Rev. Lett., 97:047201, 2006.
  • [37] G. Radaelli, D. Petti, E. Plekhanov, I. Fina, P. Torelli, B. R. Salles, M. Cantoni, C. Rinaldi, D. Gutiérrez, G. Panaccione, M. Varela, S. Picozzi, J. Fontcuberta, and R. Bertacco. Electric control of magnetism at the Fe/BaTiO3\mathrm{Fe}/\mathrm{BaTiO_{3}} interface. Nature Communications, 5(1):3404, 2014.
  • [38] Arne Vansteenkiste, Jonathan Leliaert, Mykola Dvornik, Mathias Helsen, Felipe Garcia-Sanchez, and Bartel Van Waeyenberge. The design and verification of MuMax3. AIP Advances, 4(10):107133, 2014.
  • [39] Dheeraj Kumar, Oleksandr Dmytriiev, Sabareesan Ponraj, and Anjan Barman. Numerical calculation of spin wave dispersions in magnetic nanostructures. Journal of Physics D: Applied Physics, 45(1):015001, 2011.
  • [40] G. Venkat, D. Kumar, M. Franchin, O. Dmytriiev, M. Mruczkiewicz, H. Fangohr, A. Barman, M. Krawczyk, and A. Prabhakar. Proposal for a Standard Micromagnetic Problem: Spin Wave Dispersion in a Magnonic Waveguide. IEEE Transactions on Magnetics, 49(1):524–529, 2013.
  • [41] B. A. Kalinikos and A. N. Slavin. Theory of dipole-exchange spin wave spectrum for ferromagnetic films with mixed exchange boundary conditions. Journal of Physics C: Solid State Physics, 19(35):7013–7033, 1986.
  • [42] X. Zhang, T. Hashimoto, and D. C. Joy. Electron holographic study of ferroelectric domain walls. Applied Physics Letters, 60(6):784–786, 1992.
  • [43] J. Hlinka and P. Márton. Phenomenological model of a 90° domain wall in BaTiO3\mathrm{Ba}\mathrm{Ti}{\mathrm{O}}_{3}-type ferroelectrics. Phys. Rev. B, 74:104104, 2006.
  • [44] Qingsong Zhang and William A. Goddard. Charge and polarization distributions at the 9090^{\circ} domain wall in barium titanate ferroelectric. Applied Physics Letters, 89(18):182903, 2006.
  • [45] H. L. Stadler and P. J. Zachmanidis. Nucleation and Growth of Ferroelectric Domains in BaTiO3\rm BaTiO_{3} at Fields from 2 to 450 kV/cm\mathrm{kV/cm}. Journal of Applied Physics, 34(11):3255–3260, 1963.
  • [46] V. Boddu, F. Endres, and P. Steinmann. Molecular dynamics study of ferroelectric domain nucleation and domain switching dynamics. Sci Rep, 7(1):806, 2017.
  • [47] Alexander Kirillovich Tagantsev, L Eric Cross, and Jan Fousek. Domains in ferroic crystals and thin films, volume 13. Springer, 2010.
  • [48] Ben Van de Wiele, Jonathan Leliaert, Kévin J A Franke, and Sebastiaan van Dijken. Electric-field-driven dynamics of magnetic domain walls in magnetic nanowires patterned on ferroelectric domains. New Journal of Physics, 18(3):033027, mar 2016.