betaFeedback

WorksPhysics and Astronomy

Optomagnetic non-thermal modification of the ferromagnetic resonance

Abstract

We investigate the photoinduced shift of the ferromagnetic resonance (FMR) frequency in magnets caused by the inverse Cotton-Mouton effect (ICME) under linearly polarized light. Using a Lagrangian description of magnetization dynamics, we derive the equations of motion, and obtain analytical expressions for the resonance frequency in both in-plane and out-of-plane equilibrium configurations. The theory shows that the FMR frequency depends on the polarization angle and propagation direction of light, with ICME producing a frequency shift that can dominate over thermal effects. The analytical results agree well with numerical simulations and with available experimental data for bismuth-substituted yttrium iron garnet, enabling estimation of the ICME contribution. These findings demonstrate that linearly polarized light can be used to control ferromagnetic resonance through magneto-optical effects.

Introduction

Over the past several decades, the interaction between light and spins in magnetically ordered materials has emerged as a focal point of condensed matter research [1, 2, 3, 4, 5, 6]. Initial investigations primarily utilized femtosecond laser pulses to induce both thermal [7, 8, 9] and non-thermal [10] modifications of spin states. Non-thermal mechanisms are driven by photoinduced magnetic anisotropy [7, 11] or optomagnetic phenomena, including the inverse Faraday and Cotton-Mouton effects (IFE and ICME, respectively). These interactions facilitate advanced functionalities such as ultrafast all-optical switching [7, 9, 11, 12, 13] and the coherent excitation of spin waves [14, 15].

While transient pulse excitation is prominent, continuous-wave laser irradiation also offers a viable pathway for spin manipulation. Early research focused on photomagnetic data storage [16, 17], largely relying on localized laser heating to modulate magnetization and anisotropy—a principle now industrialized via Heat-Assisted Magnetic Recording [18, 19]. However, such thermal effects necessitate significant optical absorption to drive electronic transitions. In transparent dielectrics, specifically rare-earth iron garnets, absorption-mediated heating is minimized, allowing non-thermal, inverse magneto-optical effects to dominate. These include the Inverse Faraday Effect, driven by circularly polarized light [1, 20], and the Inverse Cotton–Mouton Effect, triggered by linearly polarized light [21, 22, 23].

Ferromagnetic resonance lies in roots of many magnetic experiments and applications and is governed by the effective field entering the Kittel [24, 25] relation and therefore depends on external bias. The FMR in garnet films is modulated by external field magnitude and orientation [26], film thickness [27, 28, 29, 30], chemical composition [31, 32, 33], and temperature-dependent magnetic properties [34, 35, 36, 37]. Furthermore, tuning is achieved through elastic strain [29, 30, 38], electric-field coupling [39, 40, 41], and light-induced anisotropy changes ranging from steady-state shifts to ultrafast precessional triggering [8, 42, 43].

Dynamic control of ferromagnetic resonance frequency via external stimuli is a critical objective, with optical modulation offering high precision. The Inverse Cotton-Mouton Effect facilitates non-thermal magnetic anisotropy modification within a material’s transparency range. In ferrimagnetic garnets like Y3_3Fe5_5O12_{12} (YIG) and bismuth-substituted variants (BiYIG), ICME enables efficient magnetization precession triggering via linearly polarized light, often surpassing the efficiency of the inverse Faraday effect or photo-induced anisotropy [3, 44]. Furthermore, in antiferromagnets such as orthoferrites and NiO, ICME-driven coherent magnon excitation by linearly polarized pulses is orders of magnitude stronger than circular polarization-driven effects [2, 45]. These findings indicate that ICME can be used for fundamentally shifting FMR frequency through optical modulation of the effective magnetic field.

As the experimental foundation for our research, we employ the results obtained by [46], where a key experiment relevant to the present problem was carried out. While the experimental data are of significant value, the theoretical description is not complete. In this work, a consistent and rigorous theory is formulated, providing full agreement with the experimental observations from [46] and given additional insights into light mediated non-thermal control of FMR.

In this work, we investigate the dependence of the FMR frequency on the in-plane orientation of the polarization of light at room temperature. Experimental data indicate that ICME provides the main contribution to the polarization-dependent frequency shift, that dominates photothermal contribution to shift of FMR frequency. We analyze the observed frequency shifts in terms of anisotropy fields generated by ICME and their relation to static anisotropies.

Theory

Let’s consider an iron garnet thin film in an external in-plane magnetic field H=(H,0,0)\mathbf{H}=(H,0,0). The Cartesian coordinate system is chosen to get zz-axis out-of-plane and xx-axis along the external magnetic field (Figure 1). The film exhibits uniaxial magnetic anisotropy in the cases of easy-axis, while the effects associated with cubic crystallographic anisotropy are not taken into account. The sample is illuminated by linearly polarized light characterized by the electric field E=E0(sin⁡βcos⁡α,sin⁡βsin⁡α,cos⁡β)\mathbf{E}=E_0(\sin\beta\cos\alpha,\sin\beta\sin\alpha,\cos\beta), which is parameterized by two angles α\alpha and β\beta. Here, β\beta denotes the angle of deviation of the polarization of incident light from the normal of the film, while α\alpha represents the rotation of the in-plane polarization projection relative to the xx-axis.

Figure 1. Ferromagnetic film with uniaxial magnetic anisotropy in the external magnetic field HextH_{\mathrm{ext}} along xx axis.

To describe the dynamics of magnetization in this system we employ the Lagrangian formalism. The spherical coordinates are chosen as the polar and azimuthal angles (θ,φ)(\theta,\varphi) of the magnetization M=M(sin⁡θcos⁡φ,sin⁡θsin⁡φ,cos⁡θ)\mathbf{M}=M(\sin\theta\cos\varphi,\sin\theta\sin\varphi,\cos\theta) in Cartesian system. The Lagrangian of the system is constructed from the kinetic term and the potential energy.

L=−Mγφ˙cos⁡θ−Ua−UZ−Ud−Ucm(1)\mathcal{L}=-\frac{M}{\gamma}\dot{\varphi}\cos\theta-U_a-U_Z-U_d-U_{cm} \tag*{(1)}

The total potential energy UU includes several contributions: Zeeman energy UZU_Z due to the static magnetic field, uniaxial magnetocrystalline anisotropy UaU_a, demagnetization energy UdU_d, spin-photon interaction energy described by the inverse Cotton–Mouton effect UcmU_{cm}. The exchange energy was omitted, since we consider the film to be in a monodomain state. The above energy contributions can be written in spherical coordinates as

Ua=−Kucos⁡2θ,(2)U_a=-K_u\cos^2\theta, \tag*{(2)}
UZ=−MHsin⁡θcos⁡φ,(3)U_Z=-MH\sin\theta\cos\varphi, \tag*{(3)}
Ud=2πM2cos⁡2θ,(4)U_d=2\pi M^2\cos^2\theta, \tag*{(4)}
Ucm=Kcm[cos⁡2β+cos⁡2θ+3cos⁡2βcos⁡2θ+sin⁡2βsin⁡2θ(4cos⁡2αcos⁡2φ+sin⁡2αsin⁡2φ)+cos⁡(α−φ)sin⁡2βsin⁡2θ],(5)\begin{aligned} U_{cm}=K_{cm}\bigl[&\cos2\beta+\cos2\theta+3\cos2\beta\cos2\theta\\ &+\sin^2\beta\sin^2\theta(4\cos2\alpha\cos2\varphi+\sin2\alpha\sin2\varphi)\\ &+\cos(\alpha-\varphi)\sin2\beta\sin2\theta\bigr], \tag*{(5)} \end{aligned}

where derivation of Cotton-Mouton energy is given in detail in Appendix B.

Constructing the Euler–Lagrange equations based on the Lagrangian with the energy terms presented in Eqs. (2)-(5), one obtains the equations of magnetization motion for θ(t)\theta(t) and φ(t)\varphi(t).

θ˙=−γHsin⁡φ−ωcm[sin⁡θsin⁡2β(sin⁡2αcos⁡2φ−4cos⁡2αsin⁡2φ)+cos⁡θsin⁡2βsin⁡(α−φ)](6)\dot{\theta}=-\gamma H\sin\varphi-\omega_{cm}\left[\sin\theta\sin^2\beta(\sin2\alpha\cos2\varphi-4\cos2\alpha\sin2\varphi)+\cos\theta\sin2\beta\sin(\alpha-\varphi)\right] \tag*{(6)}
φ˙=cos⁡θ(ωu−2ωcm(1+3cos⁡2β)−γsin⁡θHcos⁡φ)+ωcm[cos⁡θsin⁡2β(4cos⁡2αcos⁡2φ+sin⁡2αsin⁡2φ)+cos⁡2θsin⁡θsin⁡2βcos⁡(α−φ)](7)\begin{aligned} \dot{\varphi}={}&\cos\theta\left(\omega_u-2\omega_{cm}(1+3\cos2\beta)-\frac{\gamma}{\sin\theta}H\cos\varphi\right)\\ &+\omega_{cm}\left[\cos\theta\sin^2\beta(4\cos2\alpha\cos2\varphi+\sin2\alpha\sin2\varphi)+\frac{\cos2\theta}{\sin\theta}\sin2\beta\cos(\alpha-\varphi)\right] \tag*{(7)} \end{aligned}

where γ\gamma is a gyromagnetic ratio, ωcm=2γKcmM\omega_{cm}=\frac{2\gamma K_{cm}}{M}, ωu=2γ(Ku−2πM2)M\omega_u=\frac{2\gamma(K_u-2\pi M^2)}{M} and Kcm=(a1−a2)M2E02/8K_{cm}=(a_1-a_2)M^2E_0^2/8 with constants a1,a2a_1,a_2, representing two possible ICME energy contributions admitted by the symmetry group of the system (Appendix B). These equations describe the precessional dynamics of magnetization determined by uniaxial anisotropy and external fields of a magnetic field and linearly polarized light, when the system is perturbed from an equilibrium position.

The equilibrium state (∂U∂θ=0,∂U∂φ=0 with ωcm=0)\left(\frac{\partial U}{\partial\theta}=0,\frac{\partial U}{\partial\varphi}=0\text{ with }\omega_{cm}=0\right) is determined by the value of the uniaxial anisotropy and the external magnetic field. Here we will consider in detail the case when initially the magnetization is alongside the external magnetic field: θ0=π/2\theta_0=\pi/2, φ0=0\varphi_0=0. For the magnetic film with the "easy-axis" anisotropy (Ku>2πM2K_u>2\pi M^2) it is realized if H>2KuM−4πMH>\frac{2K_u}{M}-4\pi M. If the uniaxial anisotropy of the magnetic film is "easy-plane" type (Ku<2πM2K_u<2\pi M^2) then any external magnetic field along xx-axis provides this equilibrium state. The case of the out-of-plane magnetization equilibrium position is described in the Appendix C.

Therefore, to linearize Eqs. (6)-(7) it is convenient to pass small angles θ1,φ1≪1\theta_1,\varphi_1\ll1 as φ=φ0+φ1\varphi=\varphi_0+\varphi_1 and θ=θ0+θ1\theta=\theta_0+\theta_1. Taking into account terms up to the first order in θ1,φ1\theta_1,\varphi_1, as numerical modeling indicates that this is sufficient to accurately describe the system, we get the following linearized equation for the free magnetization precession

θ˙1=−ω1φ1+ω3θ1+ω5,(8)\dot{\theta}_1=-\omega_1\varphi_1+\omega_3\theta_1+\omega_5, \tag*{(8)}
φ˙1=ω2θ1+ω4φ1+ω6,(9)\dot{\varphi}_1=\omega_2\theta_1+\omega_4\varphi_1+\omega_6, \tag*{(9)}

where ω1=γH−8ωcmcos⁡2αsin⁡2β\omega_1=\gamma H-8\omega_{cm}\cos2\alpha\sin^2\beta, ω2=γH−ωu+2ωcm(1+3cos⁡2β−2cos⁡2αsin⁡2β)\omega_2=\gamma H-\omega_u+2\omega_{cm}(1+3\cos2\beta-2\cos2\alpha\sin^2\beta), ω3=−ω4=ωcmsin⁡αsin⁡2β\omega_3=-\omega_4=\omega_{cm}\sin\alpha\sin2\beta, ω5=−ωcmsin⁡2αsin⁡2β\omega_5=-\omega_{cm}\sin2\alpha\sin^2\beta and ω6=−ωcmsin⁡2βcos⁡α\omega_6=-\omega_{cm}\sin2\beta\cos\alpha. These linearized equations 8, 9 can be solved exactly under the initial conditions θ1(0)\theta_1(0) and φ1(0)\varphi_1(0), assuming ωcmγHx≪1\frac{\omega_{cm}}{\gamma H_x}\ll1 we have

φ1(t)=1ωr2[Ω12+Ω22cos⁡ωrt+ωrΩ3sin⁡ωrt],(10)\varphi_1(t)=\frac{1}{\omega_r^2}\left[\Omega_1^2+\Omega_2^2\cos\omega_rt+\omega_r\Omega_3\sin\omega_rt\right], \tag*{(10)}
θ1(t)=1ωr2[Ω42+Ω52cos⁡ωrt+ωrΩ6sin⁡ωrt],(11)\theta_1(t)=\frac{1}{\omega_r^2}\left[\Omega_4^2+\Omega_5^2\cos\omega_rt+\omega_r\Omega_6\sin\omega_rt\right], \tag*{(11)}
ωr=ω1ω2−ω32,(12)\omega_r=\sqrt{\omega_1\omega_2-\omega_3^2}, \tag*{(12)}

where ωr\omega_r is the resonant frequency, Ω12=ω2ω5−ω3ω6\Omega_1^2=\omega_2\omega_5-\omega_3\omega_6 and Ω42=ω3ω5−ω1ω6\Omega_4^2=\omega_3\omega_5-\omega_1\omega_6 are determined only by the parameters of the system, Ω2=φ1(0)ωr2−Ω12\Omega_2=\varphi_1(0)\omega_r^2-\Omega_1^2, Ω3=θ1(0)ω2−φ1(0)ω3+ω6\Omega_3=\theta_1(0)\omega_2-\varphi_1(0)\omega_3+\omega_6, Ω52=θ1(0)ωr2−Ω42\Omega_5^2=\theta_1(0)\omega_r^2-\Omega_4^2 and θ1(0)ω3−φ1(0)ω1+ω5\theta_1(0)\omega_3-\varphi_1(0)\omega_1+\omega_5 are determined by the system and the initial conditions of magnetization precession θ1(0)\theta_1(0) and φ1(0)\varphi_1(0). The magnitude and polarization of a linearly polarized electric field affects not only the magnitude of the Cotton–Mouton effect but also induces a frequency shift. The Eq. (12) is in good agreement with Kittel formula (Appendix § 1.7)

In the case of normal incidence of light (β=π/2\beta=\pi/2) the Eq. (12) is simplified (ω3=ω4=ω6=0\omega_3=\omega_4=\omega_6=0) and resonant frequency becomes ωr=ω1ω2\omega_r=\sqrt{\omega_1\omega_2}.

Properties of the optomagnetic shift of FMR

Figure 2 demonstrates dependence on α\alpha, β\beta and KcmK_{cm}. The parameters of system are selected for bismuth substituted iron garnet film BiY2_2Fe4.4_{4.4}Sc0.6_{0.6}O12_{12} [46]: H=8H=8 Oe, 4πM=18304\pi M=1830 Oe, γ=1.76×10−5\gamma=1.76\times10^{-5} ps−1^{-1}Oe−1^{-1} and Ku=61.6×103K_u=61.6\times10^3 erg cm−3^{-3}. For this set of parameters Ku−2πM2=−70.6×103K_u-2\pi M^2=-70.6\times10^3 erg cm−3^{-3} the in plane equilibrium condition is satisfied.

Figure 2. The resonant frequencies dependence on the parameters α\alpha, β\beta and KcmK_{cm} with equilibrium position in-plane with Ku−2πM2<0K_u-2\pi M^2<0. (a) The colormap with fixed parameter Kcm=−5K_{cm}=-5 erg/cm3^3 illustrates analytical expression for Δωr/ωr,0\Delta\omega_r/\omega_{r,0} derived from the Eq. (12), where ωr,0=ωr(Kcm=0)\omega_{r,0}=\omega_r(K_{cm}=0) and Δωr=ωr−ωr,0\Delta\omega_r=\omega_r-\omega_{r,0}. (b) Dependence of ωr\omega_r at normal light incidence (β=π/2\beta=\pi/2) on the polarization angle α\alpha for several values of ICME parameter KcmK_{cm}. (c) Dependence of ωr\omega_r at normal light incidence on KcmK_{cm} for α=0,π/4,π/2\alpha=0,\pi/4,\pi/2.

Figure 2(a) presents the relative resonant frequency change Δωr/ωr,0\Delta\omega_r/\omega_{r,0} calculated by Eq. (12), where ωr,0=ωr(Kcm=0)\omega_{r,0}=\omega_r(K_{cm}=0) and Δωr=ωr−ωr,0\Delta\omega_r=\omega_r-\omega_{r,0}. The largest influence of light on FMR appears for normal incidence (β=π/2\beta=\pi/2).

The dependence of ωr\omega_r on the polarization angle α\alpha at normal light incidence is plotted in Figure 2(b) for several values of ICME parameter KcmK_{cm}. For selected parameters, the peak values of ωr\omega_r correspond to the angles around α=0,π/2,π,3π/2\alpha=0,\pi/2,\pi,3\pi/2, where polarization of light is either parallel or perpendicular to the equilibrium position of magnetization. On the contrary, if α\alpha is nearly π/4,3π/4,5π/4,7π/4\pi/4,3\pi/4,5\pi/4,7\pi/4, then there is no change in FMR frequency and ωr=ωr,0\omega_r=\omega_{r,0}.

Larger KcmK_{cm} produces a stronger shift of FMR. Figure 2(c) demonstrates that the effect gets stronger for increasing KcmK_{cm} and the frequency shift is almost linear in KcmK_{cm}, i.e. in the incident light intensity.

In this section we apply the developed theory to describe the experimental results obtained in [46] where dependence of ωr\omega_r on light polarization was measured. Optical pumping was performed using a continuous-wave laser at a wavelength of λ=680\lambda=680 nm with a beam spot size of approximately 66 mm2^2, which illuminated roughly 24%24\% of the total surface of the 5×55 \times5 mm2^2 sample. The static and alternating microwave magnetic fields were aligned orthogonal to each other within the sample plane. To optimize the signal-to-noise ratio, each spectral trace was obtained by averaging 50 consecutive scans. The experimental investigations were conducted on a bismuth-substituted yttrium iron garnet (BiY2_2Fe4.4_{4.4}Sc0.6_{0.6}O12_{12}) film with a thickness of d=12d = 12 μm, synthesized via liquid-phase epitaxy on a (111)-oriented gadolinium gallium garnet (GGG) substrate [46].

Figure 3. Dependence of the FMR frequency on the polarization direction α\alpha of the light wave at T=300T = 300 K. A constant magnetic field Hx=8H_x = 8 Oe is applied. The light beam with power P=25P = 25 mW makes an angle of ≈5∘\approx5^\circ with the normal to the sample plane. Experimental data points (black dots) and error bars are taken from article [46]. The solid theoretical curve is calculated according to equation 12 using the experimental parameters.

Fig. 3 presents a theoretical description of the experimental data (black dots) by Eq. (12) (solid curve). The magnetic parameters are taken the same as in the beginning of section. The best correspondence is achieved for Kcm=−1.25K_{cm} = -1.25 erg cm−3^{-3}. Since in the experiment light beam power was 25 mW one can find a1−a2=−3.1×10−7a_1 - a_2 = -3.1 \times10^{-7}/Oe2^2, which value is the same order as presented in [47].

Field-Swept Resonance Field Shift Induced by ICME

To analyze the impact of the inverse Cotton-Mouton effect under standard field-swept ferromagnetic resonance conditions, the microwave driving frequency ω\omega is treated as a strict constant. In the unperturbed regime, defined by the absence of optical pumping (ωcm=0\omega_{cm} = 0), the resonance condition is satisfied precisely at the baseline magnetic field H=HrH = H_r. Linearizing the unpumped equations of motion around the saturated in-plane equilibrium configuration (θ0,φ0)=(π/2,0)(\theta_0, \varphi_0) = (\pi/2, 0) yields the unperturbed

dispersion relation:

ω2=γHr(γHr−ωu).(13)\omega^2 = \gamma H_r(\gamma H_r - \omega_u). \tag*{(13)}

When the system is optically pumped, the spin-photon coupling (ωcm≠0\omega_{cm} \ne0) acts as an effective opto-magnetic anisotropy that modifies the free energy curvature. Introducing small-signal harmonic perturbations δθ,δφ∝eiωt\delta\theta, \delta\varphi\propto e^{i\omega t} for the magnetization variations, the dynamics are governed by the coupled linear system: iωδθ=Aδθ+Bδφi\omega\delta\theta= A\delta\theta+ B\delta\varphi and iωδφ=Cδθ+Dδφi\omega\delta\varphi= C\delta\theta+ D\delta\varphi, where A=∂θ˙/∂θA = \partial\dot{\theta}/\partial\theta, B=∂θ˙/∂φB = \partial\dot{\theta}/\partial\varphi, C=∂φ˙/∂θC = \partial\dot{\varphi}/\partial\theta and D=∂φ˙/∂φD = \partial\dot{\varphi}/\partial\varphi are as follows

A=ωcmsin⁡2βsin⁡α=−D,(14)A = \omega_{cm}\sin2\beta\sin\alpha= -D, \tag*{(14)}
B=−γ(Hr+ΔHr)+8ωcmsin⁡2βcos⁡2α,(15)B = -\gamma(H_r + \Delta H_r) + 8\omega_{cm}\sin^2\beta\cos2\alpha, \tag*{(15)}
C=γ(Hr+ΔHr)−ωu+2ωcm(1+3cos⁡2β−2sin⁡2βcos⁡2α).(16)C = \gamma(H_r + \Delta H_r) - \omega_u + 2\omega_{cm}(1 + 3\cos2\beta- 2\sin^2\beta\cos2\alpha). \tag*{(16)}

To maintain the fixed-frequency resonance condition (δω2=0\delta\omega^2 = 0) the resonant external magnetic field is taking the form H=Hr+ΔHrH = H_r + \Delta H_r, where ΔHr\Delta H_r represents the exact field displacement required to compensate for the optomagnetic ICME ωcm≠0\omega_{cm} \ne0.

Non-trivial solutions for the coefficients exist only if the determinant of the coefficients vanishes, generating the secular equation ω2=−A2−BC\omega^2 = -A^2 - BC. The parameter AA is proportional to the spin-photon coupling ωcm\omega_{cm}. Consequently, the term A2∝ωcm2A^2 \propto\omega_{cm}^2 constitutes a second-order perturbation and is neglected in the weak optomagnetic coupling regime (ωcm≪γH,ωu\omega_{cm} \ll\gamma H, \omega_u). The resonance condition simplifies to

ω2≈γHr(γHr−ωu)+γΔHr(2γHr−ωu)+2ωcmγHr(1+3cos⁡2β−2sin⁡2βcos⁡2α)−8ωcmsin⁡2βcos⁡2α(γHr−ωu).(17)\begin{aligned} \omega^2 \approx{}& \gamma H_r(\gamma H_r - \omega_u) + \gamma\Delta H_r(2\gamma H_r - \omega_u) \\ &+ 2\omega_{cm}\gamma H_r(1 + 3\cos2\beta- 2\sin^2\beta\cos2\alpha) \\ &- 8\omega_{cm}\sin^2\beta\cos2\alpha(\gamma H_r - \omega_u). \tag*{(17)} \end{aligned}

Subtracting the unperturbed system ω2=γHr(γHr−ωu)\omega^2 = \gamma H_r(\gamma H_r - \omega_u) requires the sum of all first-order perturbation terms to strictly vanish. Isolating ΔHr\Delta H_r provides the explicit analytical expression for the optomagnetic resonance field shift:

ΔHr(α,β)=2ωcmγ(2γHr−ωu)(−γHr(1+3cos⁡2β)+2sin⁡2βcos⁡2α(3γHr−2ωu)).(18)\Delta H_r(\alpha,\beta) = \frac{2\omega_{cm}}{\gamma(2\gamma H_r - \omega_u)}\left(-\gamma H_r(1 + 3\cos2\beta) + 2\sin^2\beta\cos2\alpha(3\gamma H_r - 2\omega_u)\right). \tag*{(18)}

In the configuration corresponding to the normal incidence of light relative to the sample plane, the polar polarization angle simplifies to β=π/2\beta= \pi/2. Under this geometric condition, one can reduce the general expression for the optomagnetic resonance field shift to the following relation:

ΔHr(α)=4ωcmγ[γHr+cos⁡2α(3γHr−2ωu)2γHr−ωu](19)\Delta H_r(\alpha) = \frac{4\omega_{cm}}{\gamma}\left[\frac{\gamma H_r + \cos2\alpha(3\gamma H_r - 2\omega_u)}{2\gamma H_r - \omega_u}\right] \tag*{(19)}

To analyze the evolution of this opto-magnetic response across distinct magnetic operating regimes, the dependence of ΔHr\Delta H_r on the azimuthal polarization angle α\alpha is evaluated for a wide range of unperturbed baseline fields: Hr=8H_r = 8, 100, 500, and 2000 Oe. The simplified expression demonstrates that the field shift retains a strict π\pi-periodicity dictated by the cos⁡2α\cos2\alpha symmetry of ICME (Figure 4). Furthermore, the expression reveals that both the vertical offset and the harmonic amplitude of the modulation are fundamentally governed by the ratio of the external Zeeman energy γHr\gamma H_r to the effective uniaxial anisotropy frequency ωu\omega_u. As HrH_r increases far beyond the anisotropy field limit (Hr≫ωu/γH_r \gg\omega_u/\gamma), the fraction within the brackets asymptotically approaches (1+3cos⁡2α)/2(1 + 3\cos2\alpha)/2, causing the shape of the polarization dependence to transition towards a high-field invariant profile.

Figure 4. Azimuthal polarization dependence of the photoinduced resonance field shift ΔHr\Delta H_r evaluated at a fixed polar angle β=π/2\beta= \pi/2. The continuous curves track the field shift profile across α/π∈[0,2]\alpha/\pi\in[0, 2] for three discrete unperturbed resonance fields: Hr=8H_r = 8, 100, 500, 2000 Oe. Fixed material constants utilized in the calculation are M=1830/(4π)M = 1830/(4\pi) G, Ku=(62.619×103−2πM2)K_u = (62.619 \times10^3 - 2\pi M^2) erg/cm3^3, γ=2π×2.8×106\gamma= 2\pi\times2.8 \times10^6 s−1^{-1}·Oe−1^{-1}, and Kcm=−5K_{cm} = -5 erg/cm3^3.

Discussion

To unambiguously distinguish the macroscopic ICME from the localized dynamics of photoactive impurity centers, the system must be analyzed in the time domain via ultra-fast subpicosecond laser excitation. The fundamental distinction between these two competing mechanisms lies in their characteristic temporal response functions. The ICME is a non-resonant, purely coherent opto-magnetic phenomenon mediated by a virtual electronic excitation wrapper. Consequently, the effective photoinduced magnetic field generated by the ICME tracks the optical pulse profile instantaneously, turning on and off within the subpicosecond duration of the laser pulse (τ∼100\tau\sim100 fs). This impulsive stimulus acts as a δ\delta-functional torque that initiates coherent magnetization precession (ferromagnetic resonance) without transferring thermal energy or modifying the ground-state populations.

In sharp contrast to ICME, the mechanism involving photoactive impurity centers (such as Pb3+^{3+}/Pb4+^{4+} [3] or structural defect complexes) relies on real, dissipative electronic transitions. The absorption of light drives charge transfer or reorientation among these localized centers, altering the local single-ion anisotropy energy surfaces. Because this process involves real population transfers and subsequent lattice or spin-lattice relaxation, it exhibits a characteristic finite activation rise time and a long-lived metastable decay profile that typically spans from tens of picoseconds to macroscopic timescales. By deploying subpicosecond optical pulses, the instantaneous ICME torque can be temporally separated from the delayed, cumulative changes in anisotropy induced by impurity center refills, establishing a clear pathway for isolating purely opto-magnetic phenomena.

Conclusion

The study demonstrated that linearly polarized CW light leads to the shift of FMR frequency by the inverse Cotton-Mouton effect. The theoretical consideration was based on the Lagrangian formalism considering magnetization dynamics in the magnetic potential taking into account a contribution from the ICME. The latter was derived from the group theory analysis.

The FMR frequency shift is almost linear with light intensity. It depends on the polarization of light with respect to the equilibrium magnetization state. If initial magnetization is along the in-plane external magnetic field then the maximum influence of light takes place for the light polarization either parallel or perpendicular to the equilibrium magnetization. The case when initial magnetization is out-of-plane was considered in Appendix § 1.9.

The dependence of the resonance frequency shift on varying system parameters was further analyzed. Two equilibrium configurations were examined: one with the magnetization lying in the film plane and another with the magnetization oriented out of the plane. For both cases, the frequencies obtained from nonlinear numerical simulations were systematically compared with those predicted by the linearized model, showing quantitative agreement.

Crucially, the theoretical model was extended to describe traditional field-swept FMR spectroscopy by implementing a fixed-frequency constraint. By expanding the perturbed resonance condition to first order, an explicit analytical expression was derived for the optomagnetic resonance field shift. This formulation directly maps the microscopic spin-photon coupling parameter ωcm\omega_{cm} onto the displacement of the external resonance field, establishing its parametric dependence on the incident light polarization angles.

Finally, the theoretical predictions were confronted with available experimental results. The comparison confirmed the applicability of the developed theory and enabled the determination of the ICME constants for the studied sample.

ACKNOWLEDGMENTS

This work was financially supported by Russian Science Foundation (Project No. 23-62-10024) in part related to electromagnetic modeling of the spin dynamics. N.I.G. and V.I.B. also acknowledge support from the Foundation for the Advancement of Theoretical Physics and Mathematics “BASIS” (Project No. 25-1-1-49-4) for analytical theoretical studies.

Kittel formula for FMR

Kittel formula for the ferromagnetic resonance frequency

ωfmr2=γ2M2sin⁡2θ0(∂2U∂θ2∂2U∂φ2−(∂2U∂θ∂φ)2)θ=θ0,φ=φ0(20)\omega_{\mathrm{fmr}}^{2} = \frac{\gamma^{2}}{M^{2}\sin^{2}\theta_{0}}\left(\frac{\partial^{2}U}{\partial\theta^{2}}\frac{\partial^{2}U}{\partial\varphi^{2}}-\left(\frac{\partial^{2}U}{\partial\theta\partial\varphi}\right)^{2}\right)_{\theta=\theta_{0},\varphi=\varphi_{0}} \tag*{(20)}

is obtained from the second derivatives of the total energy UU [24, 25, 48]. The ferromagnetic resonance frequency calculated ωfmr\omega_{\mathrm{fmr}} using the Kittel formula A1 coincides with the previously obtained value ωr\omega_{r} (Eq. (12)). This confirms the validity of the derivation and the consistency of theoretical description of the system’s dynamics.

Derivation of Cotton-Mouton energy

A system of iron garnet thin film is considered, where a Cartesian system is introduced: the film lies in the x,yx,y plane and its normal coincides with the zz-axis. The sample is illuminated by linearly polarized light characterized by the electric field E=E0(sin⁡βcos⁡α,sin⁡βsin⁡α,cos⁡β)\mathbf{E}=E_{0}(\sin\beta\cos\alpha,\sin\beta\sin\alpha,\cos\beta), which is parameterized by two angles α\alpha and β\beta. Here, β\beta denotes the angle of deviation of the incident light from the normal to the film, while α\alpha represents the rotation of the polarization plane relative to the xx-axis. The spherical coordinates are chosen as the polar and azimuthal angles (M,θ,φ)(M,\theta,\varphi) of the magnetization vector M\mathbf{M} (in Cartesian coordinate system M=M(sin⁡θcos⁡φ,sin⁡θsin⁡φ,cos⁡θ)\mathbf{M}=M(\sin\theta\cos\varphi,\sin\theta\sin\varphi,\cos\theta)).

When analyzing a physical system, the first step is to identify the symmetry group of its Hamiltonian, that is, the set of transformations that leave the Hamiltonian invariant [49]. In our work, we consider a system with uniaxial anisotropy, meaning that there exists a single distinguished direction along the zz-axis.

The crystallographic class of type ∞∞\infty\infty corresponds to the symmetry class of infinitely extended cylindrical objects [50]. The first ∞\infty denotes an infinite rotational axis along a certain direction, while the second ∞\infty indicates the existence of an infinite set of equivalent directions perpendicular to this axis. In other words, the object possesses cylindrical symmetry. From the tensor of the Cotton–Mouton coefficients [50], one can derive the expression [49] for the Cotton–Mouton energy as

UCM=a1(Mx2Ex2+My2Ey2+Mz2Ez2)+a2(Mx2(Ez2+Ey2)+My2(Ez2+Ex2)+Mz2(Ex2+Ey2))+a1−a22(MxMyExEy+MxMzExEz+MzMyEzEy),(21)U_{\mathrm{CM}}=a_{1}(M_{x}^{2}E_{x}^{2}+M_{y}^{2}E_{y}^{2}+M_{z}^{2}E_{z}^{2})+a_{2}\left(M_{x}^{2}(E_{z}^{2}+E_{y}^{2})+M_{y}^{2}(E_{z}^{2}+E_{x}^{2})+M_{z}^{2}(E_{x}^{2}+E_{y}^{2})\right)+\frac{a_{1}-a_{2}}{2}(M_{x}M_{y}E_{x}E_{y}+M_{x}M_{z}E_{x}E_{z}+M_{z}M_{y}E_{z}E_{y}), \tag*{(21)}

where zz axis is directed along the easy axis of the uniaxial anisotropy. Writing equations in spherical coordinates with M=M(sin⁡θcos⁡φ,sin⁡θsin⁡φ,cos⁡θ)\mathbf{M}=M(\sin\theta\cos\varphi,\sin\theta\sin\varphi,\cos\theta) and E=E0(sin⁡βcos⁡α,sin⁡βsin⁡α,cos⁡β)\mathbf{E}=E_{0}(\sin\beta\cos\alpha,\sin\beta\sin\alpha,\cos\beta), omitting constants we get

UCM=a1−a28M2E02[cos⁡2β+cos⁡2θ+3cos⁡2βcos⁡2θ+sin⁡2βsin⁡2θ(4cos⁡2αcos⁡2φ+sin⁡2αsin⁡2φ)+cos⁡(α−φ)sin⁡2βsin⁡2θ].(22)U_{\mathrm{CM}}=\frac{a_{1}-a_{2}}{8}M^{2}E_{0}^{2}\left[\cos2\beta+\cos2\theta+3\cos2\beta\cos2\theta+\sin^{2}\beta\sin^{2}\theta(4\cos2\alpha\cos2\varphi+\sin2\alpha\sin2\varphi)+\cos(\alpha-\varphi)\sin2\beta\sin2\theta\right]. \tag*{(22)}

In the case β=π/2\beta=\pi/2, which is equivalent to the normal incidence of light (along the zz axis), we have

UCM=Kcm[−2cos⁡2θ+sin⁡2θ(4cos⁡2αcos⁡2φ+sin⁡2αsin⁡2φ)],(23)U_{\mathrm{CM}}=K_{\mathrm{cm}}\left[-2\cos2\theta+\sin^{2}\theta(4\cos2\alpha\cos2\varphi+\sin2\alpha\sin2\varphi)\right], \tag*{(23)}

where Kcm=a1−a28M2E02K_{\mathrm{cm}}=\frac{a_{1}-a_{2}}{8}M^{2}E_{0}^{2}

The case of out-of-plane equilibrium magnetization state

In this section, we consider the case when the equilibrium position with ωcm=0\omega_{\mathrm{cm}}=0 of magnetization lies out of the (xy)(xy) plane. This situation is realized in the case of an easy axis, when Ku>2πM2K_u>2\pi M^2, and the external magnetic field ∣H∣<2KuM−4πM\lvert\mathbf{H}\rvert< \frac{2K_u}{M}-4\pi M. The equilibrium position of magnetization is determined by θ0=arcsin⁡(∣H∣2KuM−4πM)\theta_0=\arcsin\left(\frac{\lvert\mathbf{H}\rvert}{\frac{2K_u}{M}-4\pi M}\right) and φ0=0\varphi_0=0, since we consider without limitation generality H\mathbf{H} along the xx axis.

The same linearization procedure around the equilibrium position has been performed for the case θ0≠π/2\theta_0\ne\pi/2. where κ1\kappa_1, κ2\kappa_2, κ3\kappa_3 and κ4\kappa_4 are determined under the assumption 4κ1κ2>(κ3−κ4)24\kappa_1\kappa_2>(\kappa_3-\kappa_4)^2. This result coincides with the formula 12 in the case of θ0=π/2\theta_0=\pi/2.

In this case, the general form of the equations remains the same, but the coefficients have a more form. The angles that determine the orientation of the magnetization can be represented as ϕ=ϕ0+ϕ1\phi=\phi_0+\phi_1 and θ=θ0+θ1\theta=\theta_0+\theta_1 with small deviations from the ground state θ1,ϕ1≪1\theta_1,\phi_1\ll1. Taking into account terms up to the first order in θ1,ϕ1\theta_1,\phi_1, as numerical modeling indicates that this is sufficient to accurately describe the system, we derive the Euler-Lagrange linearized equations from equations (6) and (7)

θ˙1(t)=−κ1φ1+κ3θ1+κ5,(24)\dot{\theta}_1(t)=-\kappa_1\varphi_1+\kappa_3\theta_1+\kappa_5, \tag*{(24)}
φ˙1(t)=κ2θ1+κ4φ1+κ6,(25)\dot{\varphi}_1(t)=\kappa_2\theta_1+\kappa_4\varphi_1+\kappa_6, \tag*{(25)}

where κ1=γHx−ωcm(cos⁡θ0cos⁡αsin⁡2β+8sin⁡θ0cos⁡2αsin⁡2β)\kappa_1=\gamma H_x-\omega_{\mathrm{cm}}(\cos\theta_0\cos\alpha\sin2\beta+8\sin\theta_0\cos2\alpha\sin^2\beta), κ3=ωcm(sin⁡θ0sin⁡αsin⁡2β−cos⁡θ0sin⁡2αsin⁡2β)\kappa_3=\omega_{\mathrm{cm}}(\sin\theta_0\sin\alpha\sin2\beta-\cos\theta_0\sin2\alpha\sin^2\beta), κ5=−ωcm(cos⁡θ0sin⁡αsin⁡2β+sin⁡θ0sin⁡2αsin⁡2β)\kappa_5=-\omega_{\mathrm{cm}}(\cos\theta_0\sin\alpha\sin2\beta+\sin\theta_0\sin2\alpha\sin^2\beta), κ2=ωcm[(−2+cos⁡2θ0)cos⁡θ0sin⁡2θ0cos⁡αsin⁡2β+2sin⁡θ0(1+3cos⁡2β−2cos⁡2αsin⁡2β)]−ωusin⁡θ0+γHxsin⁡2θ0\kappa_2=\omega_{\mathrm{cm}}\left[\left(-2+\cos2\theta_0\right)\frac{\cos\theta_0}{\sin^2\theta_0}\cos\alpha\sin2\beta+2\sin\theta_0(1+3\cos2\beta-2\cos2\alpha\sin^2\beta)\right]-\omega_u\sin\theta_0+\frac{\gamma H_x}{\sin^2\theta_0}, κ4=ωcm(2cos⁡θ0sin⁡2αsin⁡2β+cos⁡2θ0sin⁡θ0sin⁡αsin⁡2β)\kappa_4=\omega_{\mathrm{cm}}\left(2\cos\theta_0\sin2\alpha\sin^2\beta+\frac{\cos2\theta_0}{\sin\theta_0}\sin\alpha\sin2\beta\right) and κ6=ωcm[cos⁡2θ0sin⁡θ0cos⁡αsin⁡2β−2cos⁡θ0(1+3cos⁡2β−2cos⁡2αsin⁡2β)]+ωucos⁡θ0−cos⁡θ0sin⁡θ0γHx\kappa_6=\omega_{\mathrm{cm}}\left[\frac{\cos2\theta_0}{\sin\theta_0}\cos\alpha\sin2\beta-2\cos\theta_0(1+3\cos2\beta-2\cos2\alpha\sin^2\beta)\right]+\omega_u\cos\theta_0-\frac{\cos\theta_0}{\sin\theta_0}\gamma H_x. These linearized equations can be solved under the initial conditions θ1(0)\theta_1(0) and φ1(0)\varphi_1(0), that have the form

φ1(t)ξ12=ξ22+eω3+ω42t[(φ1(0)ξ12−ξ22)cos⁡ωrt+sin⁡ωrt2ωr(2θ1(0)κ2ξ12−φ1(0)(κ3−κ4)ξ12+κ2κ5(κ3+κ4)+κ6(2κ1κ2−κ32+κ3κ4))],(26)\varphi_1(t)\xi_1^2=\xi_2^2+e^{\frac{\omega_3+\omega_4}{2}t}\left[(\varphi_1(0)\xi_1^2-\xi_2^2)\cos\omega_rt+\frac{\sin\omega_rt}{2\omega_r}\left(2\theta_1(0)\kappa_2\xi_1^2-\varphi_1(0)(\kappa_3-\kappa_4)\xi_1^2+\kappa_2\kappa_5(\kappa_3+\kappa_4)+\kappa_6(2\kappa_1\kappa_2-\kappa_3^2+\kappa_3\kappa_4)\right)\right], \tag*{(26)}
θ1(t)ξ12=−ξ32+eω3+ω42t[(θ1(0)ξ12+ξ32)cos⁡ωrt−sin⁡ωrt2ωr(2φ1(0)κ1ξ12−θ1(0)(κ3−κ4)ξ12+κ5(−2κ1κ2−κ3κ4+κ42)+κ1κ6(κ3+κ4))],(27)\theta_1(t)\xi_1^2=-\xi_3^2+e^{\frac{\omega_3+\omega_4}{2}t}\left[(\theta_1(0)\xi_1^2+\xi_3^2)\cos\omega_rt-\frac{\sin\omega_rt}{2\omega_r}\left(2\varphi_1(0)\kappa_1\xi_1^2-\theta_1(0)(\kappa_3-\kappa_4)\xi_1^2+\kappa_5(-2\kappa_1\kappa_2-\kappa_3\kappa_4+\kappa_4^2)+\kappa_1\kappa_6(\kappa_3+\kappa_4)\right)\right], \tag*{(27)}
ωr2=κ1κ2−(κ3−κ4)24,(28)\omega_r^2=\kappa_1\kappa_2-\frac{(\kappa_3-\kappa_4)^2}{4}, \tag*{(28)}

where ξ12=κ1κ2+κ3κ4\xi_1^2=\kappa_1\kappa_2+\kappa_3\kappa_4, ξ22=κ2κ5−κ3κ6\xi_2^2=\kappa_2\kappa_5-\kappa_3\kappa_6 and ξ32=κ4κ5+κ1κ6\xi_3^2=\kappa_4\kappa_5+\kappa_1\kappa_6.

The resonant frequency for this case can be represented as an analytical function:

ωr2=κ1κ2−(κ3−κ4)24,(29)\omega_r^2=\kappa_1\kappa_2-\frac{(\kappa_3-\kappa_4)^2}{4}, \tag*{(29)}

References

  1. [1]AV Kimel, A Kirilyuk, PA Usachev, RV Pisarev, AM Balbashov, and Th Rasing. Ultrafast non-thermal control of magnetization by instantaneous photomagnetic pulses. Nature, 435(7042):655–657, 2005.pubmed.ncbi.nlm.nih.gov/15917826
  2. [2]AM Kalashnikova, AV Kimel, RV Pisarev, VN Gridnev, Andrei Kirilyuk, and Th Rasing. Impulsive generation of coherent magnons by linearly polarized light in the easy-plane antiferromagnet febo 3. Physical review letters, 99(16):167205, 2007.pubmed.ncbi.nlm.nih.gov/17995288
  3. [3]Isao Yoshimine, Takuya Satoh, Ryugo Iida, Andrzej Stupakiewicz, Andrzej Maziewski, and Tsutomu Shimura. Phase-controllable spin wave generation in iron garnet by linearly polarized light pulses. Journal of applied physics, 116(4), 2014.
  4. [4]A Stupakiewicz, Krzysztof Szerencos, MD Davydova, KA Zvezdin, AK Zvezdin, Andrei Kirilyuk, and AV Kimel. Selection rules for all-optical magnetic recording in iron garnet. Nature communications, 10(1):612, 2019.DOI
  5. [5]Denis M Kriechvsky, Vladislav A Ozerov, Alexandra V Bel’kova, Daria A Sylgacheva, Andrey N Kalish, Svetlana A Evstigneeva, Aleksander S Pakhomov, Tatiana V Mikhailova, Sergey D Lyashko, Alexander L Kudryashov, et al. Spatially inhomogeneous inverse Faraday effect pro- vides tunable nonthermal excitation of exchange dominated spin waves. Nanophotonics, 13(3):299–306, 2024.
  6. [6]NI Gribova, AK Zvezdin, and VI Belotelov. Unified theoretical framework for spin dynamics excitation by the magnetic field of a high-frequency electromagnetic pulse. Physical Review B, 113(13):134428, 2026.DOI
  7. [7]TA Ostler, J Barker, RFL Evans, RW Chantrell, U Atxitia, O Chubykalo-Fesenko, S El Moussaoui, LBPJ Le Guyader, E Mengotti, LJ Heyderman, et al. Ultrafast heating as a sufficient stimulus for magnetization reversal in a ferrimagnet. Nature communications, 3(1):666, 2012.DOI
  8. [8]Lucile Soumah, Davide Bossini, Abdelmadjid Anane, and Stefano Bonetti. Optical frequency up-conversion of the ferromagnetic resonance in an ultrathin garnet mediated by magnetoelastic coupling. Physical Review Letters, 127(7):077203, 2021.DOI
  9. [9]A. Frej, I. Rzadolski, A. Maziewski, and A. Stupakiewicz. Nonlinear subswitching regime of magnetization dynamics in photomagnetic garnets. Phys. Rev. B, 107:134405, Apr 2023.DOI
  10. [10]Saveliì V Lutsenko, Anastasia E Khramova, Daria O Ignatyeva, Daniil V Konkov, Natalia S Kaurova, Anatoly A Syrov, Alexander L Kudryashov, Grigory N Goltsman, Vladimir N Berzhansky, and Vladimir I Belotelov. Magnetophotonic waveguide nanostructure for selective ultrafast optical excitation of high-k spin dynamics. arXiv preprint arXiv:2408.08710, 2024.arxiv.org/abs/2408.08710
  11. [11]C. D. Stanciu, F. Hansteen, A. V. Kimel, A. Kirilyuk, A. Tsukamoto, A. Itoh, and Th. Rasing. All-optical magnetic recording with circularly polarized light. Phys. Rev. Lett., 99:047601, Jul 2007.pubmed.ncbi.nlm.nih.gov/17678404
  12. [12]K Vahaplar, AM Kalashnikova, AV Kimel, Denise Hinzke, Ulrich Nowak, R Chantrell, A Tsukamoto, A Itoh, A Kirilyuk, and Th Rasing. Ultrafast path for optical magnetization reversal via a strongly nonequilibrium state. Physical review letters, 103(11):117201, 2009.pubmed.ncbi.nlm.nih.gov/19792396
  13. [13]NI Gribova, DO Ignatyeva, NA Gusev, AK Zvezdin, and VI Belotelov. Optomagnonic logic based on optical non-thermal magnetization switching in near-compensated iron-garnets. arXiv preprint arXiv:2602.06844, 2026.
  14. [14]Takuya Satoh, Yuki Terui, Rai Moriya, Boris Ivanov, Kazuya Ando, Eiji Saitoh, Tsutomu Shimura, and Kazuo Kuroda. Directional control of spin-wave emission by spatially shaped light. Nature Photonics, 6:662–666, 10 2012.
  15. [15]IV Savochkin, M Jäckl, VI Belotelov, IA Akimov, MA Kozhaev, DA Sylgacheva, AI Chernov, AN Shaposhnikov, AR Prokopov, VN Berzhansky, et al. Generation of spin waves by a train of fs-laser pulses: a novel approach for tuning magnon wavelength. Scientific reports, 7(1):5668, 2017.DOI
  16. [16]Yipeng Jiao and R. H. Victora. Dependence of predicted areal density on common optimization strategies for heat-assisted magnetic recording. IEEE Magnetics Letters, 8:1–4, 2017.DOI
  17. [17]H. J. Richter, Gregory Parker, Matteo Staffaroni, Michael Grobis, and Barry C. Stipe. Heat assisted magnetic recording with laser pulsing. IEEE Transactions on Magnetics, 50(11):1–7, 2014.DOI
  18. [18]Christoph Vogler, Claas Abert, Florian Bruckner, Dieter Suess, and Dirk Praetorius. Heat-assisted magnetic recording of bit-patterned media beyond 10 tb/in2. Applied Physics Letters, 108(10), 2016.DOI
  19. [19]Dieter Weller, Gregory Parker, Oleksandr Mosendz, Eric Champion, BC Stipe, X Wang, T Klemmer, GJ Ju, and A Ajan. A hamr media technology roadmap to an areal density of 4 tbit/in2. IEEE Trans. Magn, 50:3100108, 2014.
  20. [20]PS Pershan. Nonlinear optical properties of solids: energy considerations. Physical Review, 130(3):919, 1963.
  21. [21]BA Zon, V Ya Kupershmidt, GV Pakhomov, and TT Urazbaev. Observation of inverse cotton-mouton effect in the magnetically ordered crystal (lu, bi) 3 (fe, ga) 5 o 12. ZhETF Pisma Redaktsiiu, 45:219, 1987.
  22. [22]Daria Popova, Andreas Bringer, and Stefan Blügel. Theory of the inverse faraday effect in view of ultrafast magnetization experiments. Physical Review B—Condensed Matter and Materials Physics, 84(21):214421, 2011.DOI
  23. [23]Anatolii Konstantinovich Zvezdin, Roman Mikhailovich Dubrovin, and Alexey Vol’demarovich Kimel. Giant parametric amplification of the inverse cotton–mouton effect in antiferromagnetic crystals. JETP Letters, 119(5):363–371, 2024.DOI
  24. [24]Charles Kittel. On the theory of ferromagnetic resonance absorption. Physical review, 73(2):155, 1948.DOI
  25. [25]Charles Kittel. Ferromagnetic resonance. Journal de Physique et le Radium, 12(3):291–302, 1951.DOI
  26. [26]Seongjae Lee, Scott Grudichak, Joseph Sklenar, CC Tsai, Moongyu Jang, Qinghui Yang, Huaiwu Zhang, and John B Ketterson. Ferromagnetic resonance of a ygf film in the low frequency regime. Journal of Applied Physics, 120(3), 2016.
  27. [27]Ying Liu, Peng Zhou, Rao Bidhanapally, Jitao Zhang, Wei Zhang, Michael R Page, Tianjin Zhang, and Gopalan Srinivasan. Strain control of magnetic anisotropy in yttrium iron garnet films in a composite structure with yttrium aluminum garnet substrate. Journal of Composites Science, 6(7):203, 2022.DOI
  28. [28]Yiheng Rao, Dainan Zhang, Huaiwu Zhang, Lichuan Jin, Qinghui Yang, Zhiyong Zhong, Mingming Li, Caiyun Hong, and Bo Ma. Thickness dependence of magnetic properties in submicron yttrium iron garnet films. Journal of Physics D: Applied Physics, 51(43):435001, 2018.DOI
  29. [29]Jinjun Ding, Chuanpu Liu, Yuejie Zhang, Uppalaiah Erugu, Zhiyong Quan, Rui Yu, Ethan McCollum, Songyu Mo, Sheng Yang, Haifeng Ding, et al. Nanometer-thick yttrium iron garnet films with perpendicular anisotropy and low damping. Physical Review Applied, 14(1):014017, 2020.DOI
  30. [30]Adam Krysztofik, Sevgi Özoğlu, Robert D McMichael, and Emerson Coy. Effect of strain-induced anisotropy on magnetization dynamics in y3fe5o12 films recrystallized on a lattice-mismatched substrate. Scientific reports, 11(1):14011, 2021.DOI
  31. [31]VV Randoshkin, VY Kozlov, V Yu Mochar, NV Vasil’eva, NA Es’kov, and Yu A Durasova. Characteristic features of ferromagnetic resonance in iron-garnet films with orthorhombic magnetic anisotropy. Physics of the Solid State, 41(7):1144–1148, 1999.DOI
  32. [32]Ethan R Rosenberg, Kai Litzius, Justin M Shaw, Grant A Riley, Geoffrey SD Beach, Hans T Nembach, and Caroline A Ross. Magnetic properties and growth-induced anisotropy in yttrium thulium iron garnet thin films. Advanced Electronic Materials, 7(10):2100452, 2021.DOI
  33. [33]Sreeveni Das, Rhodri M Mansell, Lukáš Flajšman, Lide Yao, and Sebastiaan Van Dijken. Perpendicular magnetic anisotropy in bi-substituted yttrium iron garnet films.
  34. [34]vides tunable nonthermal excitation of exchange dominated spin waves. Nanophotonics, 13(3):299–306, 2024. CL Jermain, SV Aradhya, ND Reynolds, RA Buhrman, JT Brangham, MR Page, PC Hammel, FY Yang, and DC Ralph. Increased low-temperature damping in yttrium iron garnet thin films. Physical Review B, 95(17):174411, 2017.DOI
  35. [35]M Haidar, M Ranjbar, M Balinsky, RK Dumas, Sergiy Khartsev, and Johan Åkerman. Thickness-and temperature-dependent magnetodynamic properties of yttrium iron garnet thin films. Journal of Applied Physics, 117(17), 2015.DOI
  36. [36]MI Panin, NE Kupchinskaya, MV Bakhmetiev, RB Morgunov, VN Berzhansky, VI Belotelov, SN Polulyakh, and AI Chernov. Exploring magnetic anisotropy in garnet films at low temperatures using ferromagnetic resonance. Journal of Applied Physics, 137(4), 2025.
  37. [37]I Laulicht, JT Suss, and J Barak. The temperature dependence of the ferromagnetic and paramagnetic resonance spectra in thin yttrium-iron-garnet films. Journal of applied physics, 70(4):2251–2258, 1991.DOI
  38. [38]Marwan Deb, Elena Popova, Michel Hehn, Niels Keller, Stéphane Mangin, and Gregory Malinowski. Picosecond acoustic-excitation-driven ultrafast magnetization dynamics in dielectric bi-substituted yttrium iron garnet. Physical Review B, 98(17):174407, 2018.DOI
  39. [39]Rui Yu, Kang He, Qi Liu, Xucheng Gan, Bingfeng Miao, Liang Sun, Jun Du, Hongling Cai, Xiaoshan Wu, Mingzhong Wu, et al. Nonvolatile electric-field control of ferromagnetic resonance and spin pumping in pt / yig at room temperature. Advanced Electronic Materials, 5(3):1800663, 2019.
  40. [40]IV Zavislyak, MA Popov, G Sreniavaslu, and G Srinivasan. Electric field tuning of domain magnetic resonances in yttrium iron garnet films. Applied Physics Letters, 102(22), 2013.DOI
  41. [41]Xufeng Zhang, Tianyu Liu, Michael E Flatté, and Hong X Tang. Electric-field coupling to spin waves in a centrosymmetric ferrite. Physical review letters, 113(3):037202, 2014.DOI
  42. [42]A Stupakiewicz, A Maziewski, I Davidenko, and V Zablot ski. Light-induced magnetic anisotropy in co-doped garnet films. Physical Review B, 64(6):064405, 2001.DOI
  43. [43]F Atoneche, AM Kalashnikova, AV Kimel, A Stupakiewicz, A Maziewski, Andrei Kirilyuk, and Th Rasing. Large ultrafast photoinduced magnetic anisotropy in a cobalt-substituted yttrium iron garnet. Physical Review B—Condensed Matter and Materials Physics, 81(21):214440, 2010.
  44. [44]LQ Shen, LF Zhou, JY Shi, M Tang, Z Zheng, D Wu, SM Zhou, LY Chen, and HB Zhao. Dominant role of inverse cotton-mouton effect in ultrafast stimulation of magnetization precession in undoped yttrium iron garnet films by 400-nm laser pulses. Physical Review B, 97(22):224430, 2018.DOI
  45. [45]Ryugo Iida, Takuya Satoh, Tsutomu Shimura, Kazuo Kuroda, BA Ivanov, Yusuke Tokunaga, and Yoshinori Tokura. Spectral dependence of photoinduced spin precession in dyfeo 3. Physical Review B—Condensed Matter and Materials Physics, 84(6):064402, 2011.DOI
  46. [46]Sergei Nikolaevich Polulyakh, E Yu Semuk, Anatolii Konstantinovich Zvezdin, Vladimir Naumovich Berzhanskii, and Vladimir Igorevich Belotelov. Light-induced modification of the fmr spectra of a bismuth-substituted yttrium ferrite garnet film. JETP Letters, 115(4):196–201, 2022.DOI
  47. [47]RV Pisarev, IG Sinii, NN Kolpakova, and Yu M Yakovlev. Magnetic birefringence of light in iron garnets. Sov. Phys. JETP, 33:1175, 1971.
  48. [48]H Suhl. Ferromagnetic resonance in nickel ferrite between one and two kilomacycles. Physical Review, 97(2):555, 1955.DOI
  49. [49]Morton Hamermesh. Group theory and its application to physical problems. Courier Corporation, 2012.
  50. [50]Shaskolskaya M.P Sirotin YuI. Fundamentals of Crystal Physics. Mir Publishers, Moscow, 1982.

Paper details

vlnc:W_76jzqd6f
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents