Nonlinear Magnonic Frequency Shift

• Nonlinear Magnonic Frequency Shift is an amplitude-, population-, or modulation-dependent change in magnon frequencies, arising from intrinsic interactions or extrinsic boundary dynamics.
• It encompasses mechanisms from multi-magnon scattering and resonance renormalization to a Doppler-type effect caused by time-dependent magnetic-energy boundaries.
• Controlling parameters like drive amplitude, boundary velocity, and dispersion mismatch enables tunable spectral synthesis for advanced magnonic devices and signal processing.

Searching arXiv for recent and directly relevant papers on nonlinear magnonic frequency shifts and related mechanisms. Nonlinear magnonic frequency shift denotes an amplitude-, population-, or modulation-dependent change in magnon or spin-wave frequency that departs from the fixed-eigenfrequency description of linear spin dynamics. Across contemporary magnonics, the term encompasses several distinct mechanisms: amplitude-dependent resonance renormalization in confined modes and auto-oscillators, frequency conversion by three- and four-magnon interactions, cross-phase modulation between co-propagating spin waves, and, in a more recent formulation, a kinematic Doppler-type shift generated by time-dependent motion of a magnetic energy boundary rather than by intrinsic magnon–magnon nonlinearity (Hou et al., 5 Jan 2026). In the 2026 proposal of a nonlinear spin-wave Doppler effect, a moving magnetic energy or anisotropy boundary converts boundary motion directly into spin-wave phase accumulation and hence into instantaneous frequency modulation, producing harmonics, equidistant combs, and chirped sidebands while remaining in the small-angle regime and avoiding multi-magnon scattering (Hou et al., 5 Jan 2026). This mechanism broadens the concept of nonlinear magnonic frequency shift from intrinsic interaction physics to nonlinear spectral synthesis controlled by boundary kinematics.

1. Concept and scope

In linear magnonics, the eigenfrequency of a mode is fixed by geometry, bias field, saturation magnetization, anisotropy, exchange, and dipolar interactions. A nonlinear magnonic frequency shift appears when that frequency acquires dependence on oscillation amplitude, magnon population, drive strength, or an externally imposed modulation. In phenomenological form, several works describe this as $\omega = \omega_0 + N p$ or $\omega(|\varphi|) = \omega_0 + \lambda |\varphi|^2$, where the nonlinear coefficient may be positive or negative depending on mode character and geometry (0906.5224, Fripp et al., 28 Nov 2025, Ge et al., 20 May 2025, Matveev et al., 2023).

The phrase also applies to nonlinear spectral conversion. In skyrmion- and twisted-crystal-based systems, a monochromatic or two-tone input generates sidebands at shifted frequencies such as $\omega_0 \pm n\omega_r$ or $\omega_1 \pm m\omega_l$ through three- or four-magnon scattering, yielding magnonic frequency combs rather than a single displaced resonance (Wang et al., 2021, Li et al., 15 Jul 2025, Li et al., 29 Aug 2025). In magnon-polaron and magnetoelastic settings, the shift may be mediated by hybridization and parametric modulation of ferromagnetic resonance by acoustic waves (Chang et al., 2016, Künstle et al., 21 May 2026).

A conceptually distinct class is introduced by the nonlinear spin-wave Doppler effect: the frequency shift remains linear in the instantaneous boundary velocity, but the spectrum becomes nonlinear because the boundary velocity itself is time dependent, so the accumulated phase contains quadratic and sinusoidal terms (Hou et al., 5 Jan 2026). This suggests that “nonlinear magnonic frequency shift” now spans both intrinsic nonlinear dynamics and extrinsic kinematic transduction.

2. Kinematic nonlinear shift from moving magnetic-energy boundaries

The 2026 work “Nonlinear spin-Wave Doppler effect for flexible tuning of magnonic frequencies” formulates a nonlinear magnonic frequency shift arising purely from the time-dependent motion of a magnetic energy/anistropy boundary (MEB/MAB), not from magnon–magnon interactions (Hou et al., 5 Jan 2026). A spin wave of incident frequency $\omega_0$ encounters a spatial boundary between regions of different anisotropy and thus different dispersion. If the boundary moves non-uniformly, the reflected or transmitted wave experiences a time-dependent Doppler phase, and the observable instantaneous frequency becomes time dependent.

For a slowly moving boundary satisfying the quasistatic condition $|\mathbf{v}|\ll v_g$, the instantaneous reflected or transmitted frequency is written as

$$\omega_{r,t}(t) = \omega_0 - \Delta\mathbf{k}(\omega_{r,t})\cdot\mathbf{v}(t),$$

with $$\Delta\mathbf{k}(\omega)=\mathbf{k}_0(\omega)-\mathbf{k}'_{r,t}(\omega)$$ the wavevector mismatch across the boundary (Hou et al., 5 Jan 2026). In first order, $\Delta\mathbf{k}(\omega)\approx \Delta\mathbf{k}_0$, giving

$$\omega_{r,t}(t)\approx \omega_0-\Delta\mathbf{k}_0\cdot\mathbf{v}(t).$$

The spin-wave field is expressed as

$\omega(|\varphi|) = \omega_0 + \lambda |\varphi|^2$0

with $\omega(|\varphi|) = \omega_0 + \lambda |\varphi|^2$1 (Hou et al., 5 Jan 2026).

The boundary velocity is modeled as

$\omega(|\varphi|) = \omega_0 + \lambda |\varphi|^2$2

so that integrating the Doppler shift yields

$\omega(|\varphi|) = \omega_0 + \lambda |\varphi|^2$3

The three resulting phase terms are central: uniform motion gives the standard linear Doppler shift, acceleration gives quadratic phase and therefore a linear chirp, and periodic motion gives sinusoidal phase and thus frequency modulation (Hou et al., 5 Jan 2026).

This mechanism is explicitly contrasted with conventional nonlinear magnonics. The magnetization remains in the small-angle regime; there is no need for multi-magnon scattering, self-phase modulation due to large precession angles, or nonlinear dispersion at high amplitudes (Hou et al., 5 Jan 2026). The nonlinearity lies in the kinematics of the boundary rather than in the intrinsic spin-wave Hamiltonian.

3. Spectral consequences: harmonics, combs, chirps, and thresholdless synthesis

When the acceleration term is negligible, the boundary-induced phase modulation reduces to a frequency-modulated wave,

$\omega(|\varphi|) = \omega_0 + \lambda |\varphi|^2$4

with carrier frequency

$\omega(|\varphi|) = \omega_0 + \lambda |\varphi|^2$5

and modulation index

$\omega(|\varphi|) = \omega_0 + \lambda |\varphi|^2$6

Using the Jacobi–Anger expansion, the spectrum acquires equidistant lines at

$\omega(|\varphi|) = \omega_0 + \lambda |\varphi|^2$7

with intensities

$\omega(|\varphi|) = \omega_0 + \lambda |\varphi|^2$8

Accordingly, the comb spacing is fixed entirely by the boundary oscillation frequency,

$\omega(|\varphi|) = \omega_0 + \lambda |\varphi|^2$9

In the special case $\omega_0 \pm n\omega_r$0, the boundary acts as an emitter generating harmonics at $\omega_0 \pm n\omega_r$1 (Hou et al., 5 Jan 2026).

When acceleration dominates and $\omega_0 \pm n\omega_r$2, the instantaneous frequency becomes

$\omega_0 \pm n\omega_r$3

so the signal is linearly chirped. Over an observation time $\omega_0 \pm n\omega_r$4, the chirp bandwidth is estimated as

$\omega_0 \pm n\omega_r$5

This is explicitly identified as a nonlinear magnonic frequency shift because the effective frequency sweeps across a designed band rather than remaining a constant offset (Hou et al., 5 Jan 2026).

The same spectral objects—combs, harmonics, and chirped sidebands—also arise in other nonlinear magnonic settings, but usually through intrinsic interactions. Exceptional-point-enhanced magnonic frequency combs exploit a time-periodic nonlinear coupling between a pump-induced magnon mode and a Kittel mode, with lines at $\omega_0 \pm n\omega_r$6 and spacing $\omega_0 \pm n\omega_r$7 MHz in a demonstrated example (Wang et al., 2023). Strongly bistable nonlinear resonators have generated more than 350 comb lines spanning a 450 MHz bandwidth with spacing continuously tunable by a two-tone drive (Jiang et al., 28 Nov 2025). Skyrmion-based combs and topological edge-state combs similarly use discrete internal-mode frequencies or drive detuning to set the comb spacing (Wang et al., 2021, Li et al., 29 Aug 2025). The moving-boundary mechanism differs in that the comb spacing and topology are determined solely by boundary kinematics rather than by resonator free spectral range or intrinsic scattering thresholds (Hou et al., 5 Jan 2026).

4. Relation to intrinsic nonlinear mechanisms

A major point of classification concerns whether the shift is intrinsic to magnetization dynamics or extrinsic to a moving boundary or external modulation. In intrinsic cases, the effective field depends on the dynamic magnetization itself, so large precession amplitudes renormalize the mode frequency. This is explicit in nanoscale magnonic neurons, where the edge mode of a Permalloy nano-disk obeys

$\omega_0 \pm n\omega_r$8

and the nonlinear fit yields $\omega_0 \pm n\omega_r$9, corresponding to a positive blue shift of the physical edge-mode resonance (Fripp et al., 28 Nov 2025). In a nonlinear nano-ring resonator, the ring dispersion is written as

$\omega_1 \pm m\omega_l$0

with $\omega_1 \pm m\omega_l$1, producing a red shift of the resonance and foldover-like activation and limiter behavior (Wang et al., 2020).

In directional couplers, the nonlinear frequency shift can become so large that it suppresses inter-waveguide energy transfer. The effective frequency becomes

$\omega_1 \pm m\omega_l$2

with $\omega_1 \pm m\omega_l$3 in the perpendicularly magnetized YIG nano-waveguide considered, and the transition to negligible transfer occurs when $\omega_1 \pm m\omega_l$4 (Ge et al., 20 May 2025). In spin-transfer nano-oscillators, the oscillation frequency follows

$\omega_1 \pm m\omega_l$5

and the dimensionless nonlinear frequency shift coefficient $\omega_1 \pm m\omega_l$6 governs amplitude-to-phase noise conversion; reported values $\omega_1 \pm m\omega_l$7 and $\omega_1 \pm m\omega_l$8 were associated with relatively narrow linewidths (0906.5224).

Three- and four-magnon scattering provide another intrinsic route. In skyrmion scattering, discrete lines at $\omega_1 \pm m\omega_l$9 arise above a threshold because the three-magnon processes involving the skyrmion breathing mode prevail (Wang et al., 2021). In twisted magnonic crystals, finite twist activates the cubic Hamiltonian term $\omega_0$0 through a non-collinear ground state, enabling combs at $\omega_0$1 and $\omega_0$2 with spacing fixed by the Kittel mode (Li et al., 15 Jul 2025). In topological skyrmion lattices, nonlinear four-magnon scattering among chiral edge modes generates comb lines without an amplitude threshold (Li et al., 29 Aug 2025). By contrast, the moving-boundary Doppler shift in (Hou et al., 5 Jan 2026) yields analogous spectra without multi-magnon coupling.

This distinction is often a source of confusion. A nonlinear magnonic frequency shift need not imply large-angle precession or magnon-number-dependent dispersion. It may instead reflect nonlinear phase accumulation under a time-dependent boundary motion (Hou et al., 5 Jan 2026), linear synthetic-dimension hopping between fixed eigenfrequencies (Xu et al., 2024), or parametric modulation of the ferromagnetic resonance frequency by elastic strain (Chang et al., 2016). The common thread is nonlinear spectral transformation of magnonic excitations, but the microscopic origin differs sharply across platforms.

5. Micromagnetic and device realizations

The moving-boundary realization in (Hou et al., 5 Jan 2026) uses a BaTiO$\omega_0$3/Fe ferroelectric/ferromagnetic heterostructure simulated in MuMax3. The Fe parameters are $\omega_0$4, $\omega_0$5, $\omega_0$6, $\omega_0$7, $\omega_0$8, and $\omega_0$9 (Hou et al., 5 Jan 2026). The anisotropy boundary is described by

$|\mathbf{v}|\ll v_g$0

with time-dependent boundary position $|\mathbf{v}|\ll v_g$1 controlled by RF voltage (Hou et al., 5 Jan 2026).

Two operating regimes are identified. In the unbalanced-torque case, with $|\mathbf{v}|\ll v_g$2, $|\mathbf{v}|\ll v_g$3, and RF voltage amplitude $|\mathbf{v}|\ll v_g$4 giving boundary velocity $|\mathbf{v}|\ll v_g$5, the oscillating boundary acts as a source for spin waves and produces harmonic sequences with spacing $|\mathbf{v}|\ll v_g$6 (Hou et al., 5 Jan 2026). The simulated harmonic intensities follow the predicted Bessel-law redistribution $|\mathbf{v}|\ll v_g$7, supporting a Doppler/FM interpretation rather than a multi-magnon one (Hou et al., 5 Jan 2026).

In the balanced-torque case, for example at $|\mathbf{v}|\ll v_g$8 and $|\mathbf{v}|\ll v_g$9, the boundary does not emit new waves but modulates the phase of an incident one. With $$\omega_{r,t}(t) = \omega_0 - \Delta\mathbf{k}(\omega_{r,t})\cdot\mathbf{v}(t),$$0 and $$\omega_{r,t}(t) = \omega_0 - \Delta\mathbf{k}(\omega_{r,t})\cdot\mathbf{v}(t),$$1, the Fourier spectrum at $$\omega_{r,t}(t) = \omega_0 - \Delta\mathbf{k}(\omega_{r,t})\cdot\mathbf{v}(t),$$2 shows a comb centered near $$\omega_{r,t}(t) = \omega_0 - \Delta\mathbf{k}(\omega_{r,t})\cdot\mathbf{v}(t),$$3 with spacing $$\omega_{r,t}(t) = \omega_0 - \Delta\mathbf{k}(\omega_{r,t})\cdot\mathbf{v}(t),$$4 GHz, and varying $$\omega_{r,t}(t) = \omega_0 - \Delta\mathbf{k}(\omega_{r,t})\cdot\mathbf{v}(t),$$5 changes the comb spacing accordingly (Hou et al., 5 Jan 2026). Under accelerated boundary motion, with $$\omega_{r,t}(t) = \omega_0 - \Delta\mathbf{k}(\omega_{r,t})\cdot\mathbf{v}(t),$$6 and $$\omega_{r,t}(t) = \omega_0 - \Delta\mathbf{k}(\omega_{r,t})\cdot\mathbf{v}(t),$$7, the simulation yields a right-shifted side peak with $$\omega_{r,t}(t) = \omega_0 - \Delta\mathbf{k}(\omega_{r,t})\cdot\mathbf{v}(t),$$8, in agreement with the chirp-bandwidth estimate (Hou et al., 5 Jan 2026).

Other device realizations illustrate how nonlinear shifts translate into logic-like functions. In a YIG/CoFeB Fabry–Pérot resonator, the spin-wave transmission gaps shift downward by up to about 50 MHz between –15 dBm and +5 dBm, enabling neuron-like activation and nonlinear transmission suppression (Lutsenko et al., 11 Feb 2026). In all-magnonic neurons based on Ga:YIG, the positive nonlinear shift moves the resonant wavevector into the maximum of the antenna excitation efficiency, yielding a factor of about 5 increase in BLS intensity within a narrow 0.4 dB window around a self-activation power of –2.75 dBm (Breitbach et al., 22 Sep 2025). These examples show that the functional consequence of a nonlinear frequency shift is often not the shift itself but the abrupt reconfiguration of coupling, transmission, or scattering.

6. Sign, control parameters, and broader significance

The sign of the nonlinear frequency shift is not universal. Large planar magnetic vortex dots have positive gyrotropic frequency shifts, whereas smaller or more elongated dots can have negative ones, with a zero-shift geometry defined by $$\omega_{r,t}(t) = \omega_0 - \Delta\mathbf{k}(\omega_{r,t})\cdot\mathbf{v}(t),$$9 (Metlov, 2013). Spin-transfer nano-oscillators likewise permit tuning to $$\Delta\mathbf{k}(\omega)=\mathbf{k}_0(\omega)-\mathbf{k}'_{r,t}(\omega)$$0 by selecting the bias-field orientation, with asymptotic critical angles given analytically for isotropic and anisotropic cases (Matveev et al., 2023). This suggests that zero-nonlinearity design is as important as maximizing nonlinearity, depending on whether the goal is spectral stability or nonlinear functionality.

Control parameters differ by platform. In intrinsic Kerr-like systems they include drive amplitude, precession angle, bias field, anisotropy fields, and geometry (Fripp et al., 28 Nov 2025, Ge et al., 20 May 2025, Wang et al., 2020, Matveev et al., 2023). In magnon-skyrmion and twisted-crystal combs, they include the internal-mode frequency, drive detuning, twist angle, and the strength of three-magnon coupling (Wang et al., 2021, Li et al., 15 Jul 2025). In exceptional-point-enhanced combs, pump power and polarization tune the nonlinear coupling and EP location (Wang et al., 2023). In the moving-boundary Doppler scheme, the decisive control parameters are $$\Delta\mathbf{k}(\omega)=\mathbf{k}_0(\omega)-\mathbf{k}'_{r,t}(\omega)$$1, $$\Delta\mathbf{k}(\omega)=\mathbf{k}_0(\omega)-\mathbf{k}'_{r,t}(\omega)$$2, $$\Delta\mathbf{k}(\omega)=\mathbf{k}_0(\omega)-\mathbf{k}'_{r,t}(\omega)$$3, and the dispersion mismatch $$\Delta\mathbf{k}(\omega)=\mathbf{k}_0(\omega)-\mathbf{k}'_{r,t}(\omega)$$4; the spectral topology is set by boundary kinematics rather than nonlinear susceptibility (Hou et al., 5 Jan 2026).

This diversity complicates any single definition of nonlinear magnonic frequency shift. A precise encyclopedia usage is therefore necessarily plural. The term refers, first, to amplitude- or population-dependent renormalization of a magnonic eigenfrequency; second, to nonlinear generation of shifted spectral lines through multi-magnon or hybrid-wave processes; and, in the newest kinematic formulation, to frequency modulation produced by the time dependence of a moving magnetic-energy boundary (Hou et al., 5 Jan 2026). A plausible implication is that future magnonic frequency engineering will combine these mechanisms rather than treat them as competing categories: intrinsic nonlinearity for gain and thresholding, hybridization for coupling selectivity, and moving-boundary Doppler transduction for coherent low-power spectral synthesis.