Difference between angular and pseudoangular momentum

Simon Streib Department of Physics and Astronomy, Uppsala University, Box 516, SE-75120 Uppsala, Sweden (Dated: March 16, 2021) In condensed matter systems it is necessary to distinguish between the momentum of the con- stituents of the system and the pseudomomentum of quasiparticles. The same distinction is also valid for and pseudoangular momentum. Based on Noether’s theorem, we demonstrate that the recently discussed orbital angular momenta of and magnons are pseudoangular momenta. This conceptual difference is important for a proper understanding of the transfer of angular momentum in condensed matter systems, especially in spintronics applications.

In 1915, Einstein, de Haas, and Barnett demonstrated experimentally that magnetism is fundamentally related to angular momentum. When changing the magnetiza- tion of a magnet, Einstein and de Haas observed that the magnet starts to rotate, implying a transfer of an- (a) gular momentum from the magnetization to the global rotation of the lattice [1], while Barnett observed the in- verse effect, magnetization by rotation [2]. A few years later in 1918, Emmy Noether showed that continuous (b) imply conservation laws [3], such as the con- servation of momentum and angular momentum, which links magnetism to the most fundamental symmetries of nature. Condensed matter systems support closely related con- Figure 1. (a) Invariance under rotations of the whole system servation laws: the conservation of the pseudomomentum implies conservation of angular momentum, while (b) invari- and pseudoangular momentum of quasiparticles, such as ance under rotations of fields with a fixed background implies magnons and phonons. While the distinction between conservation of pseudoangular momentum. momentum and pseudomomentum (or quasimomentum) is well-known [4–13] and is crucial for understanding the momentum of light in dielectric materials [14], the con- tum. We demonstrate in this Letter that these orbital an- cept of pseudoangular momentum has not been widely gular momenta are actually pseudo-orbital angular mo- discussed in the literature, so far only in the context of menta, which phonons and magnons may carry. optical and acoustical vortices [15] and for chiral phonons The difference between pseudomomentum and momen- with a discrete rotational [16–19]. tum can most easily be explained by considering under The transfer of angular momentum between lattice which conditions they are conserved. Conservation of and magnetization has been investigated in several re- momentum follows from the uniformity of space and the cent works, for example the dynamics of a single spin invariance under translations of the whole system, embedded in an elastic solid [20–22] and the transfer of r → r + a, (1) angular momentum in magnetic insulators [23–27]. The spin angular momentum of phonons was first introduced where r is the coordinate and a an arbitrary by Vonsovskii and Svirskii [28], which was later rediscov- translation vector. The conservation of the pseudomo- ered [29], as well as experimentally confirmed [30]. It has mentum associated with a field f(r) follows from invari- been suggested that spin could be used to trans- ance of the Lagrangian under translations of the field, fer angular momentum over longer than would be possible with or magnons [25, 31, 32]. f(r) → f(r − a), (2) Several recent publications have discussed the orbital which is only valid in a uniform medium. For a crys- arXiv:2010.15616v2 [cond-mat.mes-hall] 16 Mar 2021 angular momentum of phonons [21, 22, 33] and magnons tal, pseudomomentum (“crystal momentum”) is only con- [34, 35]. The phonon orbital angular momentum plays served modulo ×reciprocal lattice vector [37]. The pseu- an important role in the relaxation process of a single ~ domomentum of a quasiparticle with wavevector k is sim- spin [21, 22], whereas the magnon orbital angular mo- ply given by k, where k is defined by an expansion of mentum could be used for the transfer of information ~ the field in waves, due to its topological stability [34] and for the manipu- lation of Skyrmions [35], analogous to using the orbital ik·r fk(r) ∼ e . (3) angular momentum of light [36]. However, there is a con- ceptual issue related to these recent advances: neither The difference between pseudomomentum and momen- phonons nor magnons carry an orbital angular momen- tum arises only for fields in a medium and not in vacuum 2 where both momenta are the same since the two transfor- much higher scales. For the elementary particles mations (1) and (2) are then equivalent [6, 10]. The pseu- in vacuum their momentum and pseudomomentum and domomentum of quasiparticles in condensed matter sys- their angular momentum and pseudoangular momentum tems is an important and the qualifier “pseudo” are identical, respectively. is usually omitted in the literature. A well-known exam- The total angular momentum of a material can be de- ple is the momentum of phonons [6, 10, 38]. The momen- composed into tum of a phonon with pseudomomentum ~k is exactly zero since the sum of the momenta of the atoms cancels. J = Le + Se + Ln + Sn, (5) We note that it is possible to define phonons in liquids where L and L are the orbital angular momenta of the in such a way that they carry momentum [6, 7, 11]. e n electrons and nuclei and S and S their spin angular The definition of pseudoangular momentum is com- e n momenta. Both the orbital and spin angular momenta pletely analogous to that of pseudomomentum. While the conservation of angular momentum follows from in- of the corresponding elementary fields are related to a variance under rotations of all the constituents of a solid, circulating momentum flow [48–50]. In the following, we will denote the pseudomomentum and pseudoangu- the conservation of pseudoangular momentum follows ˜ from invariance under rotations of fields while keeping lar momentum by a tilde, e.g., P for the pseudomomen- the solid fixed, as illustrated in Fig. 1. The corresponding tum, to distinguish them from momentum and angular conserved can then be derived from Noether’s momentum. theorem [3]. Scalar fields only have an orbital angu- As a first example, we consider the inelastic neutron lar momentum, while vector and spinor fields also have scattering process where a phonon is created [6]. The in- a spin angular momentum, which can be identified as coming neutron carries a momentum and pseudomomen- the intrinsic contribution to the angular momentum that tum ~k. In the scattering process the pseudomomentum is independent of the origin of the coordinate system is transferred to the phonon with pseudomomentum ~k. [39, 40]. Similarly to pseudomomentum, pseudoangular Since the phonon does not carry momentum, the momen- momentum cannot be exactly conserved in a real ma- tum ~k of the neutron has to be absorbed by the center of motion of the lattice. As the associated kinetic terial due to the discrete nature of atoms. However, for 2 long-wavelength excitations the discrete atomic structure energy of a body of mass M is given by (~k) /(2M), of a material is not important and the pseudoangular mo- the contribution of this kinetic energy term can be dis- mentum can be approximately conserved if the material regarded in the conservation of energy for macroscopic is very isotropic. A prime example is the magnetic in- bodies and the energy lost by the neutron is completely sulator yttrium iron garnet (YIG), for which the energy transferred to the phonon. For a complete understand- dispersion of long-wavelength magnons and phonons is ing of this process we therefore have to consider both the isotropic [41–44]. YIG is widely used in magnon spin- conservation of momentum and pseudomomentum. When considering symmetries and their corresponding tronics applications due to its record-low magnetization conservation laws, it is important to take the boundary damping [32, 45–47]. conditions into account. For lattice vibrations, periodic While in the case of pseudomomentum we can identify boundary conditions imply that the lattice cannot ro- the pseudomomentum with the wavevector k of a plane tate and angular momentum is not conserved, as it can wave, in the case of pseudoangular momentum we can leave the system via the boundary conditions. Lenstra identify the pseudoangular momentum projection along and Mandel showed that the angular momentum of the the z axis with the angular momentum m of an expansion of the field in terms of spherical har- quantized electromagnetic field is also not conserved un- der periodic boundary conditions [51]. Experimentally, monics Ylm for spherical [22] or in terms of Bessel functions for cylindrical geometries [34]. In both a sample is usually fixed to a sample holder, which pre- cases the fields have the typical azimuthal depen- vents the sample from moving or rotating. Therefore, dence momentum and angular momentum are not conserved within the sample, whereas pseudomomentum and pseu- imφ fm(r) ∼ e , (4) doangular momentum can still be conserved. This has two implications: the conservation of pseudomomentum which is the angular analogue to the plane wave (3). and pseudoangular momentum is immensely useful for Since we have these two different types of symmetries describing experimentally relevant situations and, if we in condensed matter, invariance under translations or ro- find that the total momentum or angular momentum is tations of the whole system or only of some fields, we not conserved, it most likely has been lost to the bound- have to consider two different types of conservation laws: ary conditions via the lattice. the conservation of momentum and angular momentum Nakane and Kohno have derived from Noether’s theo- and the conservation of pseudomomentum and pseudoan- rem two different expression for the angular momentum gular momentum. The elementary particles that consti- of phonons [21], which they call angular momentum in tute condensed matter systems are photons, electrons, field theory and angular momentum in Newtonian me- and nuclei, disregarding the decomposition of nuclei into chanics. Their angular momentum in field theory follows proton and neutrons or quarks, which is only relevant at from invariance under global rotations of the phonon (or 3 ) field u(r), pseudo-orbital angular momentum are not. Both the mo- mentum Pph and orbital angular momentum Lph are lin- −1  u(r) → Ru R r , (6) ear in the phonon field and vanish for phonon excitations, e.g., for any plane wave with k 6= 0, where the rotation matrix R appears twice because both Z the positions and directions of the fields are rotated. The P ∼ eik·r d3r = 0. (16) phonon field u(r) is a vector field that describes the dis- ph placement of the atoms at position r. From our discus- A finite value of L corresponds to a global rotation sion of pseudoangular momentum, it is clear that this an- ph ˜ ˜ gular momentum corresponds to the pseudoangular mo- of the whole lattice. Like Pph and Pph, Lph and Lph mentum of the phonon field, since it is based on the in- do not represent the same , contrary to variance under rotations of an emergent field. It takes the suggestion in Ref. [21]. The phonon spin is finite the form [21] for circularly polarized phonons, analogous to the spin angular momentum of light [52]. The importance of distinguishing between the angular J˜ph = L˜ ph + S˜ph, (7) Z momentum and pseudoangular momentum of phonons X 3 L˜ ph = ρu˙ αr × (−∇) uαd r, (8) can be demonstrated by considering two recent publi- α=x,y,z cations. Ref. [22] considers a single spin embedded in Z an elastic medium, where angular momentum is trans- ˜ 3 Sph = ρ (u × u˙ ) d r, (9) ferred from the spin to the orbital angular momentum of phonons. Since we have established that phonons do not where ρ is the mass density of the solid and both the or- carry orbital angular momentum, the phonon orbital an- bital contribution L˜ ph and the spin contribution S˜ph are gular momentum in Ref. [22] has to be the pseudo-orbital ˜ bilinear in the phonon field. The corresponding phonon angular momentum Lph, which phonons may carry. This pseudomomentum reads [21] can be confirmed by considering that the orbital angu- lar momentum of the phonon modes in Ref. [22] is based X Z on an expansion in spherical harmonics Y , which is P˜ = ρu˙ (−∇) u d3r. (10) lm ph α α analogous to an expansion in plane waves with pseudo- α=x,y,z momentum k. The angular momentum of the embedded The “Newtonian” angular momentum derives from in- spin enters the conversation of pseudoangular momen- variance under rotations of the whole lattice [21], tum because the system is only under rotations of both the phonon fields and the spin direction as both u(r) + r → R [u(r) + r] . (11) are coupled [21]. Therefore a change in spin has to be balanced by a change of the phonon pseudoangular mo- Therefore this angular momentum is the orbital angular mentum. The key question is: what happens to the an- momentum of the atoms and is given by [20, 21] gular momentum of the spin if it is not transferred to the phonons? As Ref. [22] applies periodic boundary condi- Z tions to the phonons, the total angular momentum is not J = L + S = ρ ([u + r] × u˙ ) d3r, (12) ph ph ph conserved. Physically, the angular momentum is in this Z case absorbed by the boundary conditions and would be 3 Lph = ρ (r × u˙ ) d r, (13) transferred to a global rotation of the lattice if it were Z not for the boundary conditions that prevent such a ro- 3 Sph = ρ (u × u˙ ) d r, (14) tation. Ref. [24], on the other hand, considers the trans- fer of angular momentum between magnons, phonons, and the global lattice rotation of a magnetic insulator. where the spin contribution Sph is formally identified as the intrinsic contribution which is independent of the ori- There, the orbital angular momentum of phonons is not gin of the coordinate system [20]. The momentum of the mentioned at all since the quantity under consideration lattice, is the angular momentum of the phonons, which is car- ried only by the phonon spin. The apparent contradic- Z 3 tion between Refs. [22] and [24] is resolved by taking Pph = ρu˙ d r, (15) into account that Ref. [22] is considering the pseudoan- gular momentum and Ref. [24] the angular momentum is only finite when there is a center of mass motion of the of phonons. Both approaches are valid, but it is impor- whole lattice, contrary to the pseudomomentum which tant to be aware of which kind of angular momentum is is finite for phonon excitations without a center of mass described and which conservation laws apply. motion. Similarly to the phonon case, we analyze next the an- Comparing the angular momentum and pseudoangu- gular momentum of magnons in a collinear ferromag- lar momentum associated with the phonon field, we see net, motivated by recent advances related to their or- that spin and pseudospin are identical, while orbital and bital angular momentum [34, 35]. We define the classical 4 magnon field ψ(r) in the following way. We start from The magnon field ψ(r) is a complex scalar field and there- the Holstein-Primakoff transformation of the spin opera- fore the magnon pseudoangular momentum has only an ˆ± ˆx ˆy tor Si = Si ± iSi at lattice site i [53], orbital contribution and no pseudospin. The pseudo- momentum k of magnons enters the conservation of r ~ √ nˆ pseudomomentum, which is for example important in Sˆ+ = 2S 1 − i ˆb , (17) i ~ 2S i spin transfer processes [54, 55]. The total spin angu- r lar momentum is determined by the number of excited √ nˆ Sˆ− = 2Sˆb† 1 − i , (18) magnons, i ~ i 2S X ˆz Sz = S0 − hnˆ i Si = ~ (S − nˆi) , (19) tot tot ~ i i Z where S is the dimensionless spin quantum number, 0 ∗ 3 ˆ†ˆ ˆ† ˆ = Stot − ~ ψ (r)ψ(r)d r, (29) nˆi = bi bi the magnon number operator, and bi and bi the magnon creation and annihilation operators with the 0 commutation relations where Stot = ~NS, with N the number of lattice sites, is the total spin if no magnons are excited. hˆ ˆ†i hˆ ˆ i hˆ† ˆ†i We note that it is also possible to derive the pseudoan- bi, bj = δij, bi, bj = bi , bj = 0. (20) gular momentum of the magnetization field M(r), which ˆ is a vector field with a spin contribution to its pseudoan- We can introduce the continuous magnon field ψ(r), gular momentum [56, 57], although there are complica- Z tions with respect to the application of Noether’s theorem ˆ 1 ˆ 3 bi = √ ψ(r)d r, (21) since the resulting pseudomomenta are gauge dependent V i Vi and not uniquely defined [57–60]. This is however not a fundamental issue because these quantities are pseu- where Vi is the of the unit cell associated with the spin at lattice site i. The replacement of the magnon domomenta and not proper momenta, which would have ˆ ˆ to be well-defined. Similarly, the pseudomomentum and operators bi by the field ψ(r) is valid for long-wavelength magnons (i.e., in the continuum limit). The normaliza- pseudoangular momentum of magnons are only defined tion is chosen such that the magnon field fulfills with respect to a given magnon quantization axis, which is in principle arbitrary. However, magnons usually de- h i scribe fluctuations with respect to a given magnetic equi- ψˆ(r), ψˆ†(r0) = δ(r − r0). (22) librium configuration that physically defines a preferred quantization axis for each spin. This allows us to define the canonically conjugate field By comparing our results for the magnon pseudomo- mentum and pseudoangular momentum, Eqs. (27-28), πˆ(r) = i ψˆ†(r), (23) ~ with the magnon momentum and magnon orbital angular with the canonical commutation relation momentum of Ref. [34], we find that they are indeed the same quantities: the pseudomomentum and pseudoan- h 0 i 0 gular momentum of magnons. Ref. [35] uses instead the ψˆ(r), πˆ(r ) = i~δ(r − r ). (24) definition of the pseudo-orbital angular momentum of the The corresponding classical magnon field fulfills then the magnetization field [56, 57]. canonical Poisson bracket relations at equal , Since a magnon excitation is only associated with a change of the magnetization and therefore of the angu- {ψ(r, t), π(r0, t)} = δ(r − r0). (25) lar momentum, a magnon cannot have any momentum, only pseudomomentum. The angular momentum associ- Because a general spin Hamiltonian H[ψ, π] is a func- ated with the magnon field is determined by the number tional of the fields and their gradients but not of their of magnons alone, Eq. (29), and does not depend on the derivatives, we obtain the classical Lagrangian [40] spatial structure of the magnon field, whereas the pseudo- orbital angular momentum L˜ shows such a dependence Z m L = πψ˙ d3r − H [ψ, π] . (26) and does not correspond to any contribution to the an- gular momentum. If the magnetization is not due to the spin alone but does also have an orbital contri- It is now straight-forward to apply Noether’s theorem to bution, then there is a correlation between the electron this Lagrangian [40] and to derive the classical magnon orbital angular momentum and the number of magnons. pseudomomentum and pseudoangular momentum, We out that the recently proposed orbital mag- Z netic moments of phonons [61] and magnons [62] are ˜ ∗ 3 Pm = −i~ ψ ∇ψ d r, (27) not related to pseudoangular momentum. Magnetic Z moments are always due to the angular momentum of ˜ ∗ 3 charged particles. In the case of Ref. [61] the underlying Lm = −i~ r × (ψ ∇ψ) d r. (28) angular momentum is the orbital angular momentum of 5 ions, and in the case of Ref. [62] the orbital angular mo- closed system, pseudoangular momentum is not exactly mentum of electrons. conserved in real materials, which should be kept in In summary, we have shown that the conservation mind for potential experimental applications. The dis- of pseudoangular momentum arises when a condensed tinction between angular momentum and pseudoangular matter system is invariant under rotations of the momentum could also be crucial for the of the emergent fields that are associated with quasiparticles. angular momentum associated with the pseudospin in We have considered the pseudoangular momentum of graphene, which has been argued to represent a real phonons and magnons, which are derived from Noether’s angular momentum [63, 64]. theorem, and demonstrated that the recently discussed orbital angular momenta of phonons [21, 22, 33] and I thank Charles Sebens for discussions on angular magnons [34, 35] are in fact pseudoangular momenta. momentum and spin, Mikhael Katsnelson and Johan In general, we conclude that when Noether’s theo- Mentink for discussing the concept of pseudomomentum, rem is applied to emergent quasiparticle fields, the and Jorge E. Hirsch for discussions on the transfer of resulting conserved momenta are pseudomomenta. The angular momentum in superconductors and ferromag- examples discussed here show that it is important to nets and his hospitality after the cancellation of the APS consider the difference between angular momentum March Meeting 2020. Special thanks go to Anna Delin, and pseudoangular momentum, which is especially Danny Thonig, and Olle Eriksson for feedback on this re- important in spintronics applications where angular search project. Financial support from the Knut and Al- momentum is used as an information carrier. While ice Wallenberg Foundation through Grant No. 2018.0060 angular momentum is in principle always conserved in a is gratefully acknowledged.

[1] A. Einstein and W. J. de Haas, Experimental proof of the (2003). existence of Ampère’s molecular currents, Proc. KNAW [16] L. Zhang and Q. Niu, Chiral Phonons at High-Symmetry 18, 696 (1915). Points in Monolayer Hexagonal Lattices, Phys. Rev. Lett. [2] S. J. Barnett, Magnetization by rotation, Phys. Rev. 6, 115, 115502 (2015). 239 (1915). [17] H. Zhu, J. Yi, M.-Y. Li, J. Xiao, L. Zhang, C.-W. Yang, [3] E. Noether, Invariante Variationsprobleme, Nachr. Ges. R. A. Kaindl, L.-J. Li, Y. Wang, and X. Zhang, Obser- Wiss. Göttingen, Math.-Phys. Kl. 2, 235 (1918). vation of chiral phonons, Science 359, 579 (2018). [4] J. P. Gordon, Radiation and momenta in dielectric [18] Y. Tatsumi, T. Kaneko, and R. Saito, Conservation law media, Phys. Rev. A 8, 14 (1973). of angular momentum in helicity-dependent Raman and [5] R. Peierls, The Momentum of Light in a Refracting Rayleigh scattering, Phys. Rev. B 97, 195444 (2018). Medium, Proc. R. Soc. A 347, 475 (1976). [19] W. Zhang, A. Srivastava, X. Li, and L. Zhang, Chiral [6] R. Peierls, Surprises in Theoretical Physics (Princeton phonons in the indirect optical transition of a MoS2/WS2 University Press, 1979). heterostructure, Phys. Rev. B 102, 174301 (2020). [7] A. Thellung, Two-fluid equations for phonons without [20] D. A. Garanin and E. M. Chudnovsky, Angular momen- momentum, Ann. Phys. (N. Y.) 127, 289 (1980). tum in spin-phonon processes, Phys. Rev. B 92, 024421 [8] M. E. Mcintyre, On the ‘wave momentum’ myth, J. Fluid (2015). Mech. 106, 331 (1981). [21] J. J. Nakane and H. Kohno, Angular momentum of [9] D. F. Nelson, Momentum, pseudomomentum, and wave phonons and its application to single-spin relaxation, momentum: Toward resolving the Minkowski-Abraham Phys. Rev. B 97, 174403 (2018). controversy, Phys. Rev. A 44, 3985 (1991). [22] J. H. Mentink, M. I. Katsnelson, and M. Lemeshko, [10] R. Peierls, More Surprises in Theoretical Physics Quantum many-body dynamics of the Einstein–de Haas (Princeton University Press, 1991). effect, Phys. Rev. B 99, 064428 (2019). [11] A. Thellung, Momentum and Quasimomentum in the [23] S. Streib, H. Keshtgar, and G. E. W. Bauer, Damping of Physics of Condensed Matter, in Die Kunst of Phonons: magnetization dynamics by phonon pumping, Phys. Rev. Lectures from the Winter School of Theoretical Physics Lett. 121, 027202 (2018). (Springer US, Boston, MA, 1994). [24] A. Rückriegel, S. Streib, G. E. W. Bauer, and R. A. [12] M. Stone, Phonons and forces: Momentum versus pseu- Duine, Angular momentum conservation and phonon domomentum in moving fluids, in Artificial Black Holes spin in magnetic insulators, Phys. Rev. B 101, 104402 (World Scientific, Singapore, 2002). (2020). [13] E. J. Heller and D. Kim, Schrödinger Correspondence [25] A. Rückriegel and R. A. Duine, Long-range phonon spin Applied to Crystals, J. Phys. Chem. A 123, 4379 (2019). transport in ferromagnet–nonmagnetic insulator het- [14] R. N. C. Pfeifer, T. A. Nieminen, N. R. Heckenberg, erostructures, Phys. Rev. Lett. 124, 117201 (2020). and H. Rubinsztein-Dunlop, Colloquium: Momentum of [26] X. Zhang, G. E. W. Bauer, and T. Yu, Unidirectional an electromagnetic wave in dielectric media, Rev. Mod. pumping of phonons by magnetization dynamics, Phys. Phys. 79, 1197 (2007). Rev. Lett. 125, 077203 (2020). [15] J.-L. Thomas and R. Marchiano, Pseudo Angular Mo- [27] D. M. Juraschek, P. Narang, and N. A. Spaldin, Phono- mentum and Topological Charge Conservation for Non- magnetic analogs to opto-magnetic effects, Phys. Rev. linear Acoustical Vortices, Phys. Rev. Lett. 91, 244302 Research 2, 043035 (2020). 6

[28] S. V. Vonsovskii and M. S. Svirskii, Phonon Spin, Sov. [46] A. V. Chumak, V. I. Vasyuchka, A. A. Serga, and Phys. Solid State 3, 1568 (1962). B. Hillebrands, Magnon spintronics, Nat. Phys. 11, 453 [29] L. Zhang and Q. Niu, Angular Momentum of Phonons (2015). and the Einstein–de Haas Effect, Phys. Rev. Lett. 112, [47] S. A. Nikitov, D. V. Kalyabin, I. V. Lisenkov, A. Slavin, 085503 (2014). Y. N. Barabanenkov, S. A. Osokin, A. V. Sadovnikov, [30] J. Holanda, D. S. Maior, A. Azevedo, and S. M. Rezende, E. N. Beginin, M. A. Morozova, Y. A. Filimonov, Y. V. Detecting the phonon spin in magnon-phonon conversion Khivintsev, S. L. Vysotsky, V. K. Sakharov, and E. S. experiments, Nat. Phys. 14, 500 (2018). Pavlov, Magnonics: a new research area in spintronics [31] K. An, A. N. Litvinenko, R. Kohno, A. A. Fuad, V. V. and spin wave electronics, Phys. Usp. 58, 1002 (2015). Naletov, L. Vila, U. Ebels, G. de Loubens, H. Hurd- [48] H. C. Ohanian, What is spin?, Am. J. Phys. 54, 500 equint, N. Beaulieu, J. Ben Youssef, N. Vukadinovic, (1986). G. E. W. Bauer, A. N. Slavin, V. S. Tiberkevich, and [49] C. T. Sebens, How electrons spin, Stud. Hist. Philos. Sci. O. Klein, Coherent long-range transfer of angular mo- B 68, 40 (2019). mentum between magnon kittel modes by phonons, Phys. [50] K. Y. Bliokh, H. Punzmann, H. Xia, F. Nori, and Rev. B 101, 060407(R) (2020). M. Shats, Relativistic field-theory spin and momentum [32] A. Brataas, B. van Wees, O. Klein, G. de Loubens, and in water waves (2020), arXiv:2009.03245. M. Viret, Spin insulatronics, Phys. Rep. 885, 1 (2020). [51] D. Lenstra and L. Mandel, Angular momentum of the [33] M. K. Ayub, S. Ali, and J. T. Mendonca, Phonons with quantized electromagnetic field with periodic boundary orbital angular momentum, Phys. Plasmas 18, 102117 conditions, Phys. Rev. A 26, 3428 (1982). (2011). [52] S. J. van Enk and G. Nienhuis, Spin and Orbital Angular [34] C. Jia, D. Ma, A. F. Schäffer, and J. Berakdar, Twisted Momentum of Photons, EPL 25, 497 (1994). magnon beams carrying orbital angular momentum, Nat. [53] T. Holstein and H. Primakoff, Field Dependence of the Commun. 10, 2077 (2019). Intrinsic Domain Magnetization of a Ferromagnet, Phys. [35] Y. Jiang, H. Y. Yuan, Z.-X. Li, Z. Wang, H. W. Zhang, Rev. 58, 1098 (1940). Y. Cao, and P. Yan, Twisted magnon as a magnetic [54] A. Mitrofanov and S. Urazhdin, Energy and momentum tweezer, Phys. Rev. Lett. 124, 217204 (2020). conservation in spin transfer, Phys. Rev. B 102, 184402 [36] W. Yang, H. Yang, Y. Cao, and P. Yan, Photonic orbital (2020). angular momentum transfer and magnetic skyrmion ro- [55] N. Tramsen, A. Mitrofanov, and S. Urazhdin, Effects tation, Opt. Express 26, 8778 (2018). of the dynamical magnetization state on spin transfer [37] N. Ashcroft and N. Mermin, Solid State Physics (Saun- (2021), arXiv:2101.08868. ders College Publishing, Philadelphia, 1976). [56] E. V. Gol’dshtein and V. M. Tsukernik, Angular momen- [38] C. Kittel, Introduction to Solid State Physics, 7th ed. tum of a Heisenberg ferromagnet with a magnetic dipole (Wiley, New York, 1996). interaction, Sov. Phys. JETP 60, 764 (1984). [39] D. E. Soper, Classical Field Theory (Wiley, New York, [57] P. Yan, A. Kamra, Y. Cao, and G. E. W. Bauer, Angu- 1976). lar and linear momentum of excited ferromagnets, Phys. [40] W. Greiner and J. Reinhardt, Field Quantization Rev. B 88, 144413 (2013). (Springer, Berlin, 1996). [58] N. Papanicolaou and T. Tomaras, Dynamics of magnetic [41] V. Cherepanov, I. Kolokolov, and V. L’vov, The saga vortices, Nucl. Phys. B 360, 425 (1991). of YIG: Spectra, thermodynamics, interaction and relax- [59] O. Tchernyshyov, Conserved momenta of a ferromagnetic ation of magnons in a complex magnet, Phys. Rep. 229, soliton, Ann. Phys. (N. Y.) 363, 98 (2015). 81 (1993). [60] S. Dasgupta and O. Tchernyshyov, Energy-momentum [42] A. G. Gurevich and G. A. Melkov, Magnetization Oscil- tensor of a ferromagnet, Phys. Rev. B 98, 224401 (2018). lations and Waves (CRC, Boca Raton, FL, 1996). [61] D. M. Juraschek and N. A. Spaldin, Orbital magnetic mo- [43] S. Maehrlein, Nonlinear Terahertz Phononics: A Novel ments of phonons, Phys. Rev. Materials 3, 064405 (2019). Route to Controlling Matter, Ph.D. thesis, Freie Univer- [62] R. R. Neumann, A. Mook, J. Henk, and I. Mertig, Or- sität Berlin (2017). bital magnetic moment of magnons, Phys. Rev. Lett. [44] S. F. Maehrlein, I. Radu, P. Maldonado, A. Paarmann, 125, 117209 (2020). M. Gensch, A. M. Kalashnikova, R. V. Pisarev, M. Wolf, [63] M. Mecklenburg and B. C. Regan, Spin and the honey- P. M. Oppeneer, J. Barker, and T. Kampfrath, Dissecting comb lattice: Lessons from graphene, Phys. Rev. Lett. spin-phonon equilibration in ferrimagnetic insulators by 106, 116803 (2011). ultrafast lattice excitation, Sci. Adv. 4, eaar5164 (2018). [64] D. Song, V. Paltoglou, S. Liu, Y. Zhu, D. Gallardo, [45] V. V. Kruglyak, S. O. Demokritov, and D. Grundler, L. Tang, J. Xu, M. Ablowitz, N. K. Efremidis, and Magnonics, J. Phys. D: Appl. Phys. 43, 264001 (2010). Z. Chen, Unveiling pseudospin and angular momentum in photonic graphene, Nat. Commun. 6, 6272 (2015).