betaFeedback

WorksPhysics and Astronomy

Symmetry-Selective Topological Magnon Engineering by Phonon Angular Momentum

Abstract

Dynamical control of Berry curvature remains an outstanding challenge in the engineering of topological phases. Here, we demonstrate control of magnon band structures via coherently driven phonons, based on ab initio spin-lattice coupling and Floquet theory. We show that this control is symmetry selective: linearly polarized phonons leave the spectrum unchanged, whereas circular and elliptical phonons carrying finite phonon angular momentum (PAM) induce chiral interactions that open and tune gaps at Dirac points, generating and reversing topological magnon phases. The gap magnitude and Chern numbers are directly governed by the PAM, enabling handedness-selective topology control. Applied to monolayer CrI3_3, and supported by symmetry analysis, our results establish driven lattice dynamics as a general route to engineering topological bosonic excitations and a versatile platform for Floquet control of magnetism.

Introduction

Berry curvature has emerged as a central concept in modern condensed matter physics, underpinning a wide range of topological phenomena [1, 2, 3, 4]. It acts as an effective magnetic field in momentum space, giving rise to anomalous transport responses and quantized topological invariants such as Chern numbers. In bosonic systems, and particularly for magnons, Berry curvature plays an equally important role: it governs topological magnon bands, chiral edge modes, and thermal Hall transport [5, 6, 7, 8, 9, 10, 11].

From a symmetry perspective, Berry curvature vanishes identically in the presence of both inversion (P\mathcal{P}) and time-reversal (T\mathcal{T}) symmetry. Breaking either symmetry allows for a finite Berry curvature; however, only broken time-reversal symmetry permits a nonzero Chern number, since otherwise the Berry curvature remains antisymmetric in momentum and integrates to zero over the Brillouin zone.

Phonons provide a natural route to dynamically control these symmetries. Linearly polarized phonons, characterized by real polarization vectors and atomic motion along a fixed direction, can break inversion symmetry. In contrast, circularly or elliptically polarized phonons — where atoms undergo rotational motion around their equilibrium positions — break time-reversal symmetry [12, 13]. These modes carry a finite phonon angular momentum (PAM) defined as L=∑iui×pi\mathbf{L} = \sum_i \mathbf{u}_i \times\mathbf{p}_i, with ui\mathbf{u}_i and pi\mathbf{p}_i being atomic displacements and momenta [14].

Phonons with finite PAM have recently attracted considerable attention due to their interplay with magnetism. They enable ultrafast manipulation of magnetic order, including transient induction of magnetism in paramagnets and multiferroics [15, 16], and reversal of magnetic order in ferrimagnets [17]. Conversely, ultrafast magnetic dynamics can excite phonons carrying finite PAM, highlighting the reciprocal coupling between spin and lattice degrees of freedom [18, 19, 20]. These developments point to chiral lattice dynamics as a powerful handle for controlling magnetic properties far from equilibrium [21].

Motivated by these considerations, we show in this Letter how driven phonon modes can be used to control magnon band structures. We demonstrate that this control is highly symmetry selective: while linearly polarized phonons leave the magnon spectrum in CrI3_3 unchanged, phonons with finite PAM strongly renormalize the band structure. In particular, they can open or close gaps at Dirac points, thereby inducing or reversing topological magnon phases.

Our approach combines a recently developed ab initio framework for spin-lattice coupling [22, 23] with Floquet theory [24, 25, 26, 27, 28], enabling a quantitative description of phonon-driven magnetic excitations. A complementary symmetry analysis based on spin-group theory provides a rigorous understanding of the observed effects and identifies the key role of phonon angular momentum in breaking the relevant symmetries. This establishes a general mechanism for engineering magnon topology via lattice dynamics, which is not limited to CrI3_3 but is expected to apply broadly to magnetic materials with finite spin-lattice coupling.

Spin-lattice coupling and Floquet theory

To describe spin excitations, we adopt an atomistic picture based on the Heisenberg Hamiltonian, which in the absence of the relativistic spin-orbit coupling (SOC) is given by [29]

H^=∑ij,ss′Jijsis′jS^is⋅S^js′.(1)\hat{\mathcal{H}} = \sum_{ij,ss'} \mathcal{J}_{ijs_is'j} \hat{\mathbf{S}}_{is} \cdot\hat{\mathbf{S}}_{js'} . \tag*{(1)}

Here, S^is(js′)\hat{\mathbf{S}}_{is(js')} are spin operators for spins of length Ss(s′)S_{s(s')} located at sublattice s(s′)s(s') at unit cell (i,j)(i,j), and Jis,js′\mathcal{J}_{is,js'} are the exchange interactions. Note that on-site interactions Jisis\mathcal{J}_{isis} are zero. The exchange interactions Jisjs′\mathcal{J}_{isjs'} are highly sensitive to interatomic distances and local bonding geometries. Therefore, they depend implicitly on the atomic displacements uks′′\boldsymbol{u}_{ks''} of atoms in the surrounding environment, such that Jisjs′=Jisjs′({uks′′})\mathcal{J}_{isjs'}=\mathcal{J}_{isjs'}(\{\boldsymbol{u}_{ks''}\}). Importantly, modifications of the exchange interactions are not restricted to displacements of the two atoms directly involved in the exchange path, (i,s)(i,s) and (j,s′)(j,s'). In general, displacements of neighboring atoms (k,s′′)(k,s'') in the vicinity can also alter the exchange coupling by changing bond lengths, bond angles, and hybridization pathways [23].

The dependence of the exchange interactions on the displacements can be made explicit by performing a Taylor expansion around the equilibrium positions rks′′0\boldsymbol{r}^{0}_{ks''}. To first order, this expansion reads [22]

H^=∑ij,ss′[Jisjs′+∑k,s′′Aisjs′ks′′⋅uks′′]S^is⋅S^js′(2)\hat{\mathcal{H}}=\sum_{ij,ss'}\left[\mathcal{J}_{isjs'}+\sum_{k,s''}\boldsymbol{A}_{isjs'ks''}\cdot\boldsymbol{u}_{ks''}\right]\hat{\boldsymbol{S}}_{is}\cdot\hat{\boldsymbol{S}}_{js'} \tag*{(2)}

with the equilibrium exchange interactions Jisjs′=Jisjs′({uks′′≡0})\mathcal{J}_{isjs'}=\mathcal{J}_{isjs'}(\{\boldsymbol{u}_{ks''}\equiv0\}) and the spin-lattice coupling parameters Aisjs′ks′′=∂Jisjs′∂uks′′({uks′′≡0})\boldsymbol{A}_{isjs'ks''}=\frac{\partial\mathcal{J}_{isjs'}}{\partial\boldsymbol{u}_{ks''}}(\{\boldsymbol{u}_{ks''}\equiv0\}). Recently, efficient methods to calculate spin-lattice coupling parameters from first principles have been developed [23, 30, 31].

Now we consider a time-periodic displacement of all atoms by a single phonon mode,

uks′′(t)=A2ℏ2ms′′ωqnϕqn,s′′eiq⋅Rke−iωqnt+c.c.(3)\boldsymbol{u}_{ks''}(t)=\frac{A}{2}\sqrt{\frac{\hbar}{2m_{s''}\omega_{\boldsymbol{q}n}}}\boldsymbol{\phi}_{\boldsymbol{q}n,s''}e^{i\boldsymbol{q}\cdot\boldsymbol{R}_{k}}e^{-i\omega_{\boldsymbol{q}n}t}+\mathrm{c.c.} \tag*{(3)}

Here, q\boldsymbol{q} is the phonon wave vector, nn is the phonon band index, AA is the (unitless) amplitude, ϕqn,s′′\boldsymbol{\phi}_{\boldsymbol{q}n,s''} is the (complex) polarization vector, ωqn\omega_{\boldsymbol{q}n} is the frequency, and Rk\boldsymbol{R}_{k} is the position of the kk-th unit cell. Inserting uks′′(t)\boldsymbol{u}_{ks''}(t) into Eq. (2), we get

H^(t)=∑ij,ss′[Jisjs′+Ae−iωqntmisjs′+c.c.]S^is⋅S^js′(4)\hat{\mathcal{H}}(t)=\sum_{ij,ss'}\left[\mathcal{J}_{isjs'}+Ae^{-i\omega_{\boldsymbol{q}n}t}m_{isjs'}+\mathrm{c.c.}\right]\hat{\boldsymbol{S}}_{is}\cdot\hat{\boldsymbol{S}}_{js'} \tag*{(4)}

with misjs′=12∑k,s′′ℏ2ms′′ωqnAisjs′ks′′⋅ϕqn,s′′eiq⋅Rkm_{isjs'}=\frac{1}{2}\sum_{k,s''}\sqrt{\frac{\hbar}{2m_{s''}\omega_{\boldsymbol{q}n}}}\boldsymbol{A}_{isjs'ks''}\cdot\boldsymbol{\phi}_{\boldsymbol{q}n,s''}e^{i\boldsymbol{q}\cdot\boldsymbol{R}_{k}} being the coefficients of the time-dependent corrections to the exchange interactions. They obey the symmetry misjs′=mjs′ism_{isjs'}=m_{js'is}. Note that these corrections do not modify the magnetic ground-state energy, since their time average vanishes for any static spin configuration. This Hamiltonian is time-periodic, H^(t)=H^(t+T)\hat{\mathcal{H}}(t)=\hat{\mathcal{H}}(t+T), with period T=2πωqnT=\frac{2\pi}{\omega_{\boldsymbol{q}n}}.

Floquet theory [27] provides a convenient way to analyze time-periodic systems. If the phonon frequency is larger than those of magnons in the system, we can use the van Vleck expansion to obtain an effective, time-independent Hamiltonian [25],

H^eff=H^0+∑m≠0[H^−m,H^m]2mℏωqn+O(ωqn−2)(5)\hat{\mathcal{H}}_{\mathrm{eff}}=\hat{\mathcal{H}}_{0}+\sum_{m\ne0}\frac{[\hat{\mathcal{H}}_{-m},\hat{\mathcal{H}}_{m}]}{2m\hbar\omega_{\boldsymbol{q}n}}+O(\omega_{\boldsymbol{q}n}^{-2}) \tag*{(5)}

with H^m=1T∫0Tdt e−imωqntH^(t)\hat{\mathcal{H}}_{m}=\frac{1}{T}\int_{0}^{T}dt\,e^{-im\omega_{\boldsymbol{q}n}t}\hat{\mathcal{H}}(t) being the Fourier components of the Hamiltonian. Here, only Fourier components with m∈{−1,0,1}m\in\{-1,0,1\} are finite, with H^0\hat{\mathcal{H}}_{0} being the Heisenberg Hamiltonian without displacements. A short calculation (details in the SM [32]) gives the lowest order correction term,

H^eff(1)=4i∣A∣2∑ijk,ss′s′′misjs′(mis′′ks′)∗ℏωqnS^is⋅(S^js′×S^ks′′)(6)\hat{\mathcal{H}}_{\mathrm{eff}}^{(1)}=4i|A|^{2}\sum_{ijk,ss's''}\frac{m_{isjs'}(m_{is''ks'})^{*}}{\hbar\omega_{\boldsymbol{q}n}}\hat{\boldsymbol{S}}_{is}\cdot\left(\hat{\boldsymbol{S}}_{js'}\times\hat{\boldsymbol{S}}_{ks''}\right) \tag*{(6)}

which describes a three-spin interaction proportional to the scalar spin chirality S^is⋅(S^js′×S^ks′′)\hat{\boldsymbol{S}}_{is}\cdot(\hat{\boldsymbol{S}}_{js'}\times\hat{\boldsymbol{S}}_{ks''}). Such a term has been found to emerge for light-induced magnetic interactions [33], but so far has not been discussed in the context of phonon-driven manipulation of magnetic interactions. The next term in the expansion H^eff(2)\hat{\mathcal{H}}_{\mathrm{eff}}^{(2)} is proportional to ωqn−2\omega_{\boldsymbol{q}n}^{-2} and describes a four-spin interaction (see SM [32], for details).

Performing the Holstein-Primakoff transformation [34] of the terms in Heff\mathcal{H}_{\mathrm{eff}} up to linear order in ωqn−1\omega_{\boldsymbol{q}n}^{-1}, and keeping terms of quadratic order in the magnon operators a^ks(†)\hat{a}^{(\dagger)}_{ks}, we get [32]

H^=∑kss′(J~kss′+∣A∣2(m~kss′−m~ks′s))a^ks†a^ks′−J~0ss′a^ks†a^ks(7)\hat{\mathcal{H}}=\sum_{kss'}\left(\tilde{J}_{kss'}+|A|^{2}(\tilde{m}_{kss'}-\tilde{m}_{ks's})\right)\hat{a}^{\dagger}_{ks}\hat{a}_{ks'}-\tilde{J}_{0ss'}\hat{a}^{\dagger}_{ks}\hat{a}_{ks} \tag*{(7)}

Note that we have chosen an expansion around a ferromagnetic state with equal magnetic moments at each sublattice (since below we will apply the framework to CrI3_3) and the summation over s,s′s,s' is only over magnetic sublattices. We want to emphasize that magnon Hamiltonians for any other magnetic state (e.g., antiferromagnets, or non-collinear states) can also be derived from H^eff\hat{\mathcal{H}}_{\mathrm{eff}} in a similar manner.

The coefficients to the magnon Hamiltonian are

J~kss′=2S∑jJis,js′eik⋅(rj−ri)(8)\tilde{J}_{kss'}=2S\sum_{j}J_{is,js'}e^{i\boldsymbol{k}\cdot(\boldsymbol{r}_{j}-\boldsymbol{r}_{i})} \tag*{(8)}
m~kss′=4S2ℏωqn∑j,k,s′′mis′′ks′′(mis′js′)∗eik⋅(rj−ri)e−ik⋅(rk−ri)+(misjs′(misks′)∗−mis′′ks′′(misjs′)∗)eik⋅(rj−ri)(9)\tilde{m}_{kss'}=\frac{4S^{2}}{\hbar\omega_{\boldsymbol{q}n}}\sum_{j,k,s''}m_{is''ks''}(m_{is'js'})^{*}e^{i\boldsymbol{k}\cdot(\boldsymbol{r}_{j}-\boldsymbol{r}_{i})}e^{-i\boldsymbol{k}\cdot(\boldsymbol{r}_{k}-\boldsymbol{r}_{i})} +\left(m_{isjs'}(m_{isks'})^{*}-m_{is''ks''}(m_{isjs'})^{*}\right)e^{i\boldsymbol{k}\cdot(\boldsymbol{r}_{j}-\boldsymbol{r}_{i})} \tag*{(9)}

and can be directly computed from the exchange interactions and spin-lattice coupling parameters.

Results for monolayer CrI3_3. For our study we consider CrI3_3, which is a layered magnetic van der Waals material exhibiting ferromagnetic order with a Curie temperature of 61 K in bulk crystals and 45 K in the monolayer limit [35]. The bulk (monolayer) crystal structure belongs to the symmetry group R3ˉR\bar{3} (P3ˉ1mP\bar{3}1m), and the magnetic lattice consists of two distinct chromium sublattices forming a honeycomb network, with localized moments characterized by spin SS. Inelastic neutron scattering revealed that the energies of magnons in

CrI3_3 go up to approximately 20 meV [36]. We compute the magnetic exchange interactions Jisjs′J_{isjs'} and the spin-lattice coupling parameters Aisjs′ks′′A_{isjs'ks''} using the relativistic formulation of the so-called LKAG (Liechtenstein-Katsnelson-Antropov-Gubanov) formalism [37, 38] and its extension for spin-lattice coupling [23], implemented in the SPRKKR package [32, 39]. Among the dominant interactions, we obtain nearest-neighbor, next-nearest-neighbor, and third-nearest-neighbor exchange couplings of magnitude J1/S2=3.027J_1/S^2 = 3.027 meV, J2/S2=0.538J_2/S^2 = 0.538 meV to J3/S2=−0.048J_3/S^2 = -0.048 meV, respectively (see Figure 1), with S=3.2/2S = 3.2/2.

Top view of the atomic lattice of monolayer CrI3 and schematic representations of degenerate phonon modes

Figure 1. (a) Top view of the atomic lattice of monolayer CrI3_3. J1J_1 to J3J_3 denote the nearest- to third-nearest-neighbor exchange interactions between magnetic Cr atoms. (b) and (c) Schematic representations of one representative eigenvector from each pair of analyzed degenerate phonon modes with energies and group representations as labeled.

The phonon properties are obtained from density functional theory calculations performed using VASP [40, 41] and phonopy [42]. We focus on degenerate optical phonon modes at the Γ\Gamma point. To avoid resonant magnon-phonon processes, only phonons with energies well above those of the magnons are considered. In particular, we analyze two pairs of symmetry-protected degenerate modes, illustrated in Figure 1(b) and (c), belonging to the EuE_u and EgE_g irreducible representations. Direct offresonant excitation of magnons by a THz laser resonant with any of the considered phonon modes is negligible [32].

Circularly polarized phonon modes are constructed from the degenerate linear polarization vectors according to ϕqn,s±=(ϕqn,s(1)±iϕqn,s(2))/2\phi_{qn,s}^{\pm} = (\phi_{qn,s}^{(1)} \pm i\phi_{qn,s}^{(2)})/\sqrt{2} which correspond to clockwise (+) and counter-clockwise (−) rotational lattice vibrations.

First, we calculate mis,js′m_{is,js'}, the time-dependent corrections to the exchange interactions, induced by circularly polarized phonons. In general, we find that their absolute value is the same for all neighbors in a certain shell; they only differ by a phase. We find the phase difference

phonon energysymmetrym1/S2m_1/S^2m2/S2m_2/S^2m3/S2m_3/S^2
25.46EuE_u00.0190
27.32EgE_g0.1540.0120.011

Table 1. Absolute values of time-dependent corrections to the exchange interactions mis,js′m_{is,js'} in the first three neighbor shells induced by circular phonon modes constructed from two pairs of degenerate phonon modes with energies and symmetries as labeled. All values are in meV.

between the induced exchange of two pairs of neighbors to be twice the angle enclosed by the respective bond vectors. That is why we can write mis,js′′=mle∓i2ϕisjs′+iαlm_{is,js'}' = m_l e^{\mp i2\phi_{isjs'}+i\alpha_l}, with l=1,2,3l = 1,2,3, indexing the neighbor shells (analogous to the JlJ_l; see Figure 1), ϕisjs′\phi_{isjs'} being the bond angle and αl\alpha_l some constant phase for each shell. We find “−” in the exponential for clockwise circular phonons and “+” for counter-clockwise circular phonons. The absolute values for the first three neighbor shells are listed in Table I.

Interestingly, the EuE_u circular modes only modify the intrasublattice exchange (via m2m_2) and do not affect the intersublattice exchange. For EuE_u phonons, the displacements are odd under inversion and Cr sublattice exchange. Since we consider only terms up to linear order in displacements in Eq. (2), the mis,js′m_{is,js'} are also odd under this transformation. Thus, it follows that mi1,j2=−mj2,i1m_{i1,j2} = -m_{j2,i1}, which, together with mis,js′=mjs′,is′m_{is,js'} = m_{js',is'}, implies that the intersublattice exchange mi1,j2m_{i1,j2} is zero. From the mis,js′m_{is,js'}, we obtain the coefficient m~kseff\tilde{m}_{ks}^{\mathrm{eff}} for the magnon Hamiltonian (7). The calculation of its eigenenergies and eigenmodes is performed numerically using Colpa’s method [32, 43].

Fig. 2 illustrates the modification of the magnon spectrum induced by clockwise circular EgE_g phonon modes. In the absence of phonon driving, the two magnon bands exhibit Dirac-like crossings at the KK and K′K' points. For a finite circular phonon amplitude, these degeneracies are lifted and energy gaps open at both high-symmetry points. The magnitude of the induced gaps scales linearly with ∣A∣2|A|^2, and thus quadratically with the displacement amplitude of the Cr atoms from their equilibrium positions (see inset of Fig. 2).

The gap opening signals a topological phase transition. For any finite drive amplitude, the magnon bands acquire nontrivial Chern numbers Cn=12π∫Ωknd2kC_n = \frac{1}{2\pi}\int\Omega_{kn}d^2k where the Berry curvature is defined as Ωkn=i(⟨∂kxn∣∂kyn⟩−⟨∂kyn∣∂kxn⟩)\Omega_{kn} = i(\langle\partial_{k_x}n|\partial_{k_y}n\rangle- \langle\partial_{k_y}n|\partial_{k_x}n\rangle) [1]. We obtain Cn=±1C_n = \pm1 for the two bands [32], demonstrating that the circular-phonon-induced gap is topological in nature. Reversing the sense of atomic rotation to counter-clockwise inverts the sign of the Berry curvature throughout the Brillouin zone and consequently flips the Chern numbers.

The dependence of the topological gap opening on the phonon properties, including the PAM, can be made explicit by considering a superposition of the two degenerate linear polarization vectors, $\phi_{qn,s}(\psi) = (\phi_{qn,s}^{(1)} +

Energy bands of CrI$_3$ along the $M$–$K'$–$\Gamma$–$K$–$M$ path, with Berry curvature color coding and inset showing the magnon energy gap versus Cr displacementVector (PDF)

caption: Engineering of magnon bandstructure of CrI3_3. Arrows indicate the opening of a gap at the Dirac points KK and K′K' upon driving by clockwise circular EgE_g phonon mode and the color coding shows the Berry curvature of the magnon bands. The inset shows the magnon energy gap versus Cr displacement.

:::

eiψϕqn,s(2)/2e^{i\psi}\phi_{qn,s}^{(2)}/\sqrt{2}, where ψ\psi captures the polarization: for ψ=0,±π\psi= 0, \pm\pi the atomic motion is linearly polarized, whereas ψ=π/2\psi= \pi/2 corresponds to clockwise circular motion and ψ=−π/2\psi= -\pi/2 to counter-clockwise circular motion.

Refs. [14, 44] showed that the PAM of a phonon mode is related to its polarization vectors through Lkn=iℏ∑sϕqn,s×(ϕqn,s)∗\mathbf{L}_{kn} = i\hbar\sum_s\phi_{qn,s}\times(\phi_{qn,s})^*. For the parametrization above, this expression simplifies to Lkn=ℏsin⁡ψ∑sϕkn,s′(1)×ϕkn,s′(2)\mathbf{L}_{kn}=\hbar\sin\psi\sum_s\phi_{kn,s'}^{(1)}\times\phi_{kn,s'}^{(2)}.

This sine-like behavior is confirmed in Fig. 3(a), which shows the PAM for ψ∈(−π,π)\psi\in(-\pi,\pi) for both pairs of degenerate phonon modes considered here. The maximum PAM, given by ℏ∑sϕkn,s′(1)×ϕkn,s′(2)\hbar\sum_s\phi_{kn,s'}^{(1)}\times\phi_{kn,s'}^{(2)}, is slightly smaller for the EgE_g modes.

Figures 3(b) and 3(c) further reveal that the magnon gaps ΔE\Delta E at KK and K′K' induced by EgE_g phonons scale as ∣sin⁡ψ∣|\sin\psi|, while the corresponding Chern numbers follow ±sgn⁡(ψ)\pm\operatorname{sgn}(\psi). This establishes the relations

ΔE∝∣PAM∣,Cn∝±sgn⁡(PAM),(10)\Delta E \propto|\mathrm{PAM}|,\qquad C_n \propto\pm\operatorname{sgn}(\mathrm{PAM}), \tag*{(10)}

which constitute a central finding of this work: the gap and topology of magnon bands can be controlled by the PAM of lattice vibrations.

The absence of any modification of the magnon dispersion under EuE_u phonon driving originates from the lack of induced intersublattice interactions for these modes (see Tab. I). Such interactions are required for finite m~kss′−m~−ks′s\tilde{m}_{kss'}-\tilde{m}_{-ks's}, and their absence therefore prevents the opening of a gap in the magnon spectrum (details in the SM [32]).

To understand why circular or elliptical phonons can open a gap through nonrelativistic spin-lattice coupling while linearly polarized phonons cannot, we analyze the symmetry constraints on m~kss′\tilde{m}_{kss'} imposed by the magnon Hamiltonian (7). We first rewrite the magnon

PAM, magnon gap, and Chern numbers versus phonon polarization phase for $E_g$ and $E_u$ modesVector (PDF)

caption: PAM-selective control of magnon gap and topologies in the absence of SOC. (a) PAM and (b) magnon gap at KK, K′K' (for uCr=∣A∣ℏ2mCrωqn∣ϕqn,Cr∣=0.184u_{\mathrm{Cr}}=|A|\sqrt{\frac{\hbar}{2m_{\mathrm{Cr}}\omega_{qn}}}|\phi_{qn,\mathrm{Cr}}|=0.184 Å) versus phonon polarization for EgE_g and EuE_u modes. (c) Chern numbers of both magnon bands versus phonon polarization upon driving by EgE_g modes (results for ψ=±π\psi=\pm\pi are not shown).

:::

Hamiltonian in Bogoliubov–de Gennes form [45], H^=12∑kss′X^ks†Hkss′X^ks′\hat{H}=\frac{1}{2}\sum_{kss'}\hat{X}_{ks}^{\dagger}\mathcal{H}_{kss'}\hat{X}_{ks'}, where X^ks†=(a^−ks,a^ks†)\hat{X}_{ks}^{\dagger}=(\hat{a}_{-ks},\hat{a}_{ks}^{\dagger}) and Hkss′=(Hkss′0−∣A∣2ΔHkss′)\mathcal{H}_{kss'}=\left(\mathcal{H}_{kss'}^0-|A|^2\Delta\mathcal{H}_{kss'}\right). The 2×22\times2 matrices Hkss′0=(J~kss′−J~0ss′δss′)σ0\mathcal{H}_{kss'}^0=(\tilde{J}_{kss'}-\tilde{J}_{0ss'}\delta_{ss'})\sigma_0 and ΔHkss′=2m~kss′σ3\Delta\mathcal{H}_{kss'}=2\tilde{m}_{kss'}\sigma_3 describe the equilibrium interactions and the phonon-induced corrections, respectively. Here σi\sigma_i with i∈{0,1,2,3}i\in\{0,1,2,3\} denote the identity and the three Pauli matrices [46].

The Bogoliubov–de Gennes structure implies a generalized particle–hole symmetry requiring σ1Hkss′σ1=H−kss′∗\sigma_1\mathcal{H}_{kss'}\sigma_1=\mathcal{H}_{-kss'}^*. Consequently, ΔHkss′=−ΔH−kss′∗\Delta\mathcal{H}_{kss'}=-\Delta\mathcal{H}_{-kss'}^* independent of which phonon mode is excited. In the absence of phonons carrying PAM, and neglecting SOC (see below), the system is additionally invariant under combined time reversal and spin reversal, implying ΔHkss′=ΔH−kss′∗\Delta\mathcal{H}_{kss'}=\Delta\mathcal{H}_{-kss'}^* [47]. The two symmetry relations therefore enforce ΔHkss′=0\Delta\mathcal{H}_{kss'}=0 for linearly polarized modes. Circular or elliptical phonons, however, carry finite PAM and thus break this effective time- and spin-reversal symmetry for magnons, allowing for a finite m~kss′\tilde{m}_{kss'} and the opening of a topological magnon gap.

According to experiments, the magnon spectrum of CrI3_3 is already gapped in the absence of phonon driving [36]. The microscopic origin of this gap remains under debate – ranging from Kitaev-type interactions [48, 49] to Dzyaloshinskii–Moriya interactions (DMI) [36, 50, 51]. We adopt the latter scenario in the following. Nearest-neighbor DMI is forbidden by inversion symmetry, and we therefore include next-nearest-neighbor DMI with a strength taken from Ref. [36]. As a SOC effect, DMI breaks the combined time-reversal and spin-reversal symmetry discussed above, which explains the presence of a finite gap already in equilibrium (see Figure 4(a) and (Figure 4f)).

Figure 4. Engineering of magnon bandstructure of CrI3_3 in the presence of SOC, assumed to manifest in DMI. (a–e) and (f–j) show the magnon energies around KK driven by clockwise (CW) and counter-clockwise (CCW) EgE_g phonons with Cr displacements uCru_{\mathrm{Cr}} as labeled. The color coding indicates the Berry curvature. (k) and (l) depict the magnon energy gap and the anomalous thermal Hall conductivity versus Cr displacement.

Upon driving the system with circular phonons, the gap can be tuned in a handedness-selective manner: clockwise phonons enhance the gap, whereas counter-clockwise phonons reduce it, eventually closing and reopening it with inverted topology, i.e., with exchanged Chern numbers of the lower and upper magnon bands (see Figure 4(a)–(Figure 4k)). Reversing the sign of the DMI reverses the role of the two phonon handednesses.

An experimental signature of this topological gap reversal can be found in transverse thermal transport. The anomalous thermal Hall conductivity κxy\kappa_{xy} relates a temperature gradient to a transverse heat current via jx=−κxy∂yTj_x=-\kappa_{xy}\partial_yT. Within linear response theory, κxy\kappa_{xy} is directly linked to the Berry curvature [6, 52, 53, 54],

κxy=−kB2TℏA∑knc(fkn)Ωkn.(11)\kappa_{xy}=-\frac{k_{\mathrm{B}}^2T}{\hbar A}\sum_{kn}c(f_{kn})\Omega_{kn}. \tag*{(11)}

Here, AA denotes the system area, fknf_{kn} is the Bose-Einstein distribution function, and c(x)=(1+x)(ln⁡1+xx)2−2Li⁡2(−x)c(x)=(1+x)\left(\ln\frac{1+x}{x}\right)^2-2\operatorname{Li}_2(-x), with Li⁡2(x)\operatorname{Li}_2(x) being the second order polylogarithm function.

Figure 4(l) shows the magnon anomalous thermal Hall conductivity for clockwise and counter-clockwise phonon driving. Its behavior closely follows that of the gap ΔE\Delta E: clockwise phonons enhance κxy\kappa_{xy}, while counter-clockwise phonons suppress it and ultimately induce a sign reversal at sufficiently large driving amplitudes.

We note that the lattice displacements required to completely close and reopen the topological magnon gap in CrI3_3 are comparatively large, reflecting the already sizable equilibrium gap (4 meV). At such amplitudes, anharmonic effects and structural instabilities may become relevant. The results presented here should therefore be viewed as a proof of principle. We expect that materials with substantially smaller intrinsic magnon gaps, such as CrCl3_3 [55], could exhibit the same phonon-induced topological engineering at considerably smaller lattice displacements.

Conclusion. We have demonstrated that magnon band structures can be systematically engineered by driving lattice vibrations, establishing THz-driven phonons as an effective and symmetry-selective control knob for magnon topology. Using an ab initio-based spin-lattice coupling framework combined with Floquet theory, we showed that circular and elliptical phonons carrying finite phonon angular momentum (PAM) break time-reversal symmetry for magnons and thereby induce topological gaps in otherwise gapless spectra. In contrast, linearly polarized phonons do not modify the magnon dispersion due to symmetry constraints within the nonrelativistic spin-lattice coupling considered here.

A central result of this work is that both the magnitude and topology of magnon bands are directly governed by the PAM: the induced gap scales with ∣PAM∣\lvert\mathrm{PAM}\rvert,

while the associated Chern numbers are determined by its sign. This enables handedness-selective control, where reversing the sense of lattice rotation switches the Berry curvature and topological invariants. We further showed that only specific phonon symmetries (such as EgE_g modes in CrI3_3) modulate intersublattice exchange interactions and can thus modify the magnon spectrum, whereas others (e.g., EuE_u modes) remain ineffective.

Including spin-orbit coupling (SOC) via Dzyaloshinskii-Moriya interactions (DMI), we find that phonon driving allows continuous tuning of an already existing magnon gap, including its closure and reopening with inverted topology, depending on PAM. This behavior is directly reflected in experimentally accessible observables such as the anomalous thermal Hall conductivity. We note that relativistic spin-lattice coupling, describing phonon-induced or -modulated DMI and anisotropies, provides a subleading correction to this mechanism and enables linearly polarized phonons to modify the gap through an effect that is quadratic in SOC [32].

Importantly, the theoretical framework is robust: we explicitly benchmarked the lowest-order van Vleck expansion against the numerically converged Floquet spectrum obtained in an extended Hilbert space framework [32], finding excellent agreement and confirming the reliability of the effective Hamiltonian. While demonstrated for CrI3_3, the underlying mechanism relies only on general symmetry principles, and spin-lattice coupling, and is therefore expected to be broadly applicable. Promising candidate systems include other two-dimensional van der Waals magnets (such as CrBr3_3 [56], CrCl3_3 [55], or Fe3_3GeTe2_2 [57]) as well as magnetic insulators with strong spin-lattice coupling and accessible optical phonon modes. These findings open a route toward dynamical control of magnon topology using tailored phonon excitations.

Acknowledgments

M.W. and P.M.O. acknowledge funding from the German Research Foundation (Deutsche Forschungsgemeinschaft) through CRC/TRR 227 “Ultrafast Spin Dynamics” (Project MF, Project ID No. 328545488). This work was further supported by the Swedish Research Council (VR) and the Knut and Alice Wallenberg Foundation (Grants No. 2022.0079 and No. 2023.0336). We acknowledge funding from the European Union’s HORIZON EUROPE, under Grant Agreement No. 101129641, “OBELIX”. The calculations were partially supported by resources provided by the National Academic Infrastructure for Supercomputing in Sweden (NAISS) at NSC Linköping, partially funded by VR through Grant No. 2022-06725.

References

  1. [1]D. Xiao, M.-C. Chang, and Q. Niu, Berry phase effects on electronic properties, Rev. Mod. Phys. 82, 1959 (2010).DOI
  2. [2]N. Nagaosa, J. Sinova, S. Onoda, A. H. MacDonald, and N. P. Ong, Anomalous Hall effect, Rev. Mod. Phys. 82, 1539 (2010).DOI
  3. [3]M. Z. Hasan and C. L. Kane, Colloquium: Topological insulators, Rev. Mod. Phys. 82, 3045 (2010).
  4. [4]X.-L. Qi and S.-C. Zhang, Topological insulators and superconductors, Rev. Mod. Phys. 83, 1057 (2011).
  5. [5]R. Matsumoto and S. Murakami, Theoretical Prediction of a Rotating Magnon Wave Packet in Ferromagnets, Phys. Rev. Lett. 106, 197202 (2011).pubmed.ncbi.nlm.nih.gov/21668195
  6. [6]R. Matsumoto, R. Shindou, and S. Murakami, Thermal Hall effect of magnons in magnets with dipolar interaction, Phys. Rev. B 89, 054420 (2014).
  7. [7]Y. Onose, T. Ideue, H. Katsura, Y. Shiomi, N. Nagaosa, and Y. Tokura, Observation of the Magnon Hall Effect, Science 329, 297 (2010).DOI
  8. [8]A. Mook, J. Henk, and I. Mertig, Magnon Hall effect and topology in kagome lattices: A theoretical investigation, Phys. Rev. B 89, 134409 (2014).DOI
  9. [9]A. Mook, J. Henk, and I. Mertig, Edge states in topological magnon insulators, Phys. Rev. B 90, 024412 (2014).DOI
  10. [10]S. A. Owerre, A first theoretical realization of honeycomb topological magnon insulator, Journal of Physics: Condensed Matter 28, 386001 (2016).DOI
  11. [11]P. A. McClarty, Topological Magnons: A Review, Annual Review of Condensed Matter Physics 13, 171 (2022).DOI
  12. [12]L. Zhang and Q. Niu, Chiral Phonons at High-Symmetry Points in Monolayer Hexagonal Lattices, Phys. Rev. Lett. 115, 115502 (2015).DOI
  13. [13]H. Zhu, J. Yi, M.-Y. Li, J. Xiao, L. Zhang, C.-W. Yang, R. A. Kaindl, L.-J. Li, Y. Wang, and X. Zhang, Observation of chiral phonons, Science 359, 579 (2018).
  14. [14]L. Zhang and Q. Niu, Angular Momentum of Phonons and the Einstein-de Haas Effect, Phys. Rev. Lett. 112, 085503 (2014).DOI
  15. [15]J. Luo, T. Lin, J. Zhang, X. Chen, E. R. Blackert, R. Xu, B. I. Yakobson, and H. Zhu, Large effective magnetic fields from chiral phonons in rare-earth halides, Science 382, 698 (2023).DOI
  16. [16]M. Basini, M. Pancaldi, B. Wehinger, M. Udina, V. Unikandanunni, T. Tadano, M. C. Hoffmann, A. V. Balatsky, and S. Bonetti, Terahertz electric-field-driven dynamical multiferroicity in SrTiO3, Nature 628, 534 (2024).
  17. [17]C. S. Davies, F. G. N. Fennema, A. Tsukamoto, I. Razdolski, A. V. Kimel, and A. Kirilyuk, Phononic switching of magnetization by the ultrafast Barnett effect, Nature 628, 540 (2024).DOI
  18. [18]C. Dornes, Y. Acremann, M. Savoini, M. Kubli, M. J. Neugebauer, E. Abreu, L. Huber, G. Lantz, C. A. F. Vaz, H. Lemke, E. M. Bothschafter, M. Porer, V. Esposito, L. Rettig, M. Buzzi, A. Alberca, Y. W. Windsor, P. Beaud, U. Staub, D. Zhu, S. Song, J. M. Glownia, and S. L. Johnson, The ultrafast Einstein–de Haas effect, Nature 565, 209 (2019).DOI
  19. [19]S. R. Tauchert, M. Volkov, D. Ehbenger, D. Kazenwadel, M. Evers, H. Lange, A. Donges, A. Book, W. Kreuzpaintner, U. Nowak, and P. Baum, Polarized phonons carry angular momentum in ultrafast demagnetization, Nature 602, 73 (2022).
  20. [20]M. S. Mrudul, M. Weißenhofer, and P. M. Oppeneer, Generation of phonons with angular momentum during ultrafast demagnetization, Phys. Rev. B 112, L180407 (2025).DOI
  21. [21]D. M. Juraschek et al., Chiral phonons, Nature Phys. 21, 1532 (2025).
  22. [22]J. Hellsvik, D. Thonig, K. Modin, D. Iuşan, A. Bergman, O. Eriksson, L. Bergqvist, and A. Delin, General method for atomistic spin-lattice dynamics with first-principles accuracy, Phys. Rev. B 99, 104302 (2019).
  23. [23]S. Mankovsky, S. Polesya, H. Lange, M. Weißenhofer, U. Nowak, and H. Ebert, Angular Momentum Transfer via Relativistic Spin-Lattice Coupling from First Principles, Phys. Rev. Lett. 129, 067202 (2022).DOI
  24. [24]J. H. Shirley, Solution of the Schrödinger Equation with a Hamiltonian Periodic in Time, Phys. Rev. 138, B979 (1965).DOI
  25. [25]M. Bukov, L. D’Alessio, and A. Polkovnikov, Universal high-frequency behavior of periodically driven systems: from dynamical stabilization to Floquet engineering, Advances in Physics 64, 139 (2015).DOI
  26. [26]A. Eckardt, Colloquium: Atomic quantum gases in periodically driven optical lattices, Rev. Mod. Phys. 89, 011004 (2017).DOI
  27. [27]T. Oka and S. Kitamura, Floquet Engineering of Quantum Materials, Annual Review of Condensed Matter Physics 10, 387 (2019).DOI
  28. [28]M. Yarmohammadi, M. Berritta, M. Bukov, L. Šmejkal, J. Linder, and P. M. Oppeneer, Spin polarization engineering in d-wave altermagnets, Phys. Rev. B 113, L060403 (2026).arxiv.org/abs/2510.02452
  29. [29]O. Eriksson et al., Atomistic Spin Dynamics (Oxford University Press, 2017).
  30. [30]S. Mankovsky, H. Lange, S. Polesya, and H. Ebert, Spin-lattice interaction parameters from first principles: Theory and implementation, Phys. Rev. B 107, 144428 (2023).
  31. [31]I. P. Miranda, M. Pankratova, M. Weißenhofer, A. B. Klautau, D. Thonig, M. Pereiro, E. Sjöqvist, A. Delin, M. I. Katsnelson, O. Eriksson, and A. Bergman, Spin-lattice couplings in 3d ferromagnets: Analysis from first principles, Phys. Rev. Mater. 9, 024409 (2025).DOI
  32. [32]See Supplemental Material at [URL will be inserted by publisher] for (i) the derivation of Eq. (6), (ii) the derivation of Eq. (7), (iii) a summary of Colpa’s method, (iv) a proof that phonon-induced intrasublattice exchange cannot modify the magnon gap, (v) the computational details for the exchange and spin-lattice coupling parameters, (vi) a comparison between the first-order van Vleck expansion and the converged Floquet quasi-energy spectrum, (vii) the effect of phonon-modulated DMI on the first-order van Vleck term, and (viii) an estimate of the magnon and phonon excitation rates under THz excitation.
  33. [33]R. Yambe and S. Hayami, Symmetry analysis of light-induced magnetic interactions via Floquet engineering, Phys. Rev. B 108, 064420 (2023).DOI
  34. [34]T. Holstein and H. Primakoff, Field Dependence of the Intrinsic Domain Magnetization of a Ferromagnet, Phys. Rev. 58, 1098 (1940).DOI
  35. [35]B. Huang, G. Clark, E. Navarro-Moratalla, D. R. Klein, R. Cheng, K. L. Seyler, D. Zhong, E. Schmidgall, M. A. McGuire, D. H. Cobden, W. Yao, D. Xiao, P. Jarillo-Herrero, and X. Xu, Layer-dependent ferromagnetism in a van der waals crystal down to the monolayer limit, Nature 546, 270 (2017).DOI
  36. [36]L. Chen, J.-H. Chung, B. Gao, T. Chen, M. B. Stone, A. I. Kolesnikov, Q. Huang, and P. Dai, Topological Spin Excitations in Honeycomb Ferromagnet CrI3, Phys. Rev. X 8, 041028 (2018).DOI
  37. [37]A. I. Liechtenstein, M. I. Katsnelson, V. P. Antropov, and V. A. Gubanov, Local spin density functional approach to the theory of exchange interactions in ferromagnetic metals and alloys, Journal of Magnetism and Magnetic Materials 67, 65 (1987).DOI
  38. [38]H. Ebert and S. Mankovsky, Anisotropic exchange coupling in diluted magnetic semiconductors: Ab initio spin-density functional theory, Phys. Rev. B 79, 045209 (2009).DOI
  39. [39]H. Ebert, D. Ködderitzsch, and J. Minár, Calculating condensed matter properties using the KKR-Green’s function method—recent developments and applications, Reports on Progress in Physics 74, 096501 (2011).DOI
  40. [40]G. Kresse and J. Furthmüller, Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set, Computational Materials Science 6, 15 (1996).DOI
  41. [41]G. Kresse and J. Furthmüller, Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set, Phys. Rev. B 54, 11169 (1996).pubmed.ncbi.nlm.nih.gov/9984901
  42. [42]A. Togo, L. Chaput, T. Tadano, and I. Tanaka, Implementation strategies in phonopy and phono3py, Journal of Physics: Condensed Matter 35, 353001 (2023).DOI
  43. [43]J. Colpa, Diagonalization of the quadratic boson hamiltonian, Physica A: Statistical Mechanics and its Applications 93, 327 (1978).DOI
  44. [44]M. Weißenhofer, P. Rieger, M. S. Mrudul, L. Mikadze, U. Nowak, and P. M. Oppeneer, Chiral Phonons Arising from Chirality-Selective Magnon-Phonon Coupling, Phys. Rev. Lett. 135, 216701 (2025).
  45. [45]A. Mook, J. Henk, and I. Mertig, Thermal Hall effect in noncollinear coplanar insulating antiferromagnets, Phys. Rev. B 99, 014427 (2019).DOI
  46. [46]We define σ0 = (1 0 0 1), σ1 = (0 1 1 0), σ2 = (0 −i i 0), and σ3 = (1 0 0 −1).
  47. [47]M. Weißenhofer, M. S. Mrudul, S. Mankovsky, and P. M. Oppeneer, Magnon orbital nernst effect in altermagnets, npj Quantum Materials 11, 25 (2026).
  48. [48]I. Lee, F. G. Untermohlen, D. Weber, K. Hwang, C. Zhang, J. van Tol, J. E. Goldberger, N. Trivedi, and P. C. Hammel, Fundamental Spin Interactions Underlying the Magnetic Anisotropy in the Kitaev Ferromagnet CrI3, Phys. Rev. Lett. 124, 017201 (2020).DOI
  49. [49]C. Xu, J. Feng, H. Xiang, and L. Bellaiche, Interplay between Kitaev interaction and single ion anisotropy in ferromagnetic CrI3 and CrGeTe3 monolayers, npj Computational Materials 4, 57 (2018).DOI
  50. [50]A. Kartsev, M. Augustin, R. F. L. Evans, K. S. Novoselov, and E. J. G. Santos, Biquadratic exchange interactions in two-dimensional magnets, npj Computational Materials 6, 150 (2020).DOI
  51. [51]L. Chen, J.-H. Chung, M. B. Stone, A. I. Kolesnikov, B. Winn, V. O. Garlea, D. L. Abernathy, B. Gao, M. Augustin, E. J. G. Santos, and P. Dai, Magnetic Field Effect on Topological Spin Excitations in CrI3, Phys. Rev. X 11, 031047 (2021).DOI
  52. [52]L. Zhang, Berry curvature and various thermal Hall effects, New J. Phys. 18, 103039 (2016).DOI
  53. [53]S. Murakami and A. Okamoto, Thermal Hall effect of magnons, J. Phys. Soc. Jpn. 86, 011010 (2017).
  54. [54]J. N. Kløgetvedt and A. Qaiumzadeh, Tunable topological magnon-polaron states and intrinsic anomalous Hall phenomena in two-dimensional ferromagnetic insulators, Phys. Rev. B 108, 224424 (2023).
  55. [55]D. R. Klein, D. MacNeill, Q. Song, D. T. Larson, S. Fang, M. Xu, R. A. Ribeiro, P. C. Canfield, E. Kaxiras, R. Comin, and P. Jarillo-Herrero, Enhancement of interlayer exchange in an ultrathin two-dimensional magnet, Nature Physics 15, 1255 (2019).DOI
  56. [56]D. Ghazaryan, M. T. Greenaway, Z. Wang, V. H. Guarochico-Moreira, I. J. Vera-Marun, J. Yin, Y. Liao, S. V. Morozov, O. Kristanovski, A. I. Lichtenstein, M. I. Katsnelson, F. Withers, A. Mishchenko, L. Eaves, A. K. Geim, K. S. Novoselov, and A. Misra, Magnon-assisted tunnelling in van der Waals heterostructures based on CrBr3, Nature Electronics 1, 344 (2018).DOI
  57. [57]Z. Fei, B. Huang, P. Malinowski, W. Wang, T. Song, J. Sanchez, W. Yao, D. Xiao, X. Zhu, A. F. May, W. Wu, D. H. Cobden, J.-H. Chu, and X. Xu, Two-dimensional itinerant ferromagnetism in atomically thin Fe3GeTe2, Nature Materials 17, 778 (2018).DOI

Paper details

vlnc:W_v6qf687c
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