Open Submitted: 16 February 2021 Accepted: 09 April 2021 Published Online: 22 April 2021
Appl. Phys. Lett. 118, 162405 (2021); https://doi.org/10.1063/5.0047737
more...View Affiliations
View Contributors
• G. Gubbiotti
• E. Beginin
• S. Sheshukova
• S. Nikitov
• G. Talmelli
• I. Asselberghs
• I. P. Radu
• F. Ciubotaru

In this work, we investigate the spin-wave propagation in three-dimensional nanoscale CoFeB/Ta/NiFe meander structures fabricated on a structured SiO2/Si substrate. The magnonic band structure has been experimentally determined by wavevector-resolved Brillouin light scattering spectroscopy and a set of stationary modes interposed by two dispersive modes of Bloch type have been identified. The results could be understood by micromagnetic and finite element simulations of the mode distributions in both real space and the frequency domain. The dispersive modes periodically oscillate in frequency over the Brillouin zones and correspond to modes, whose spatial distributions extend over the entire sample and are either localized exclusively in the CoFeB layer or the entire CoFeB/Ta/NiFe magnetic bilayer, with in-phase precession of the dynamic magnetization in the two layers. Low-frequency stationary modes are concentrated in horizontal segments of the topmost NiFe layer with sizeable amplitudes in the vertical CoFeB and NiFe segments and out-of-phase precession. The findings are compared with those of single-layer CoFeB meander structures with the same geometry parameters, which reveals the influence of the dipolar coupling between the two ferromagnetic layers on the magnonic band structure.
In this Letter, we investigate by BLS spectroscopy the dispersion of collective SWs in 3D MCs based on meander structures including Co40Fe40B20/Ta/Ni80Fe20 (CoFeB/Ta/NiFe) magnetic bilayers. The unit cell of the meander structures had a height of h = 50 nm and periodicity of a = 600 nm, leading to a lattice periodicity in reciprocal space (π/a =0.52 × 107 rad/m) and a wavevector range that was accessible by our experimental BLS setup. The SW dispersion relation was mapped up to the fourth Brillouin zone (BZ) by sweeping the wavevector along the periodicity direction. The measured SW spectra contained a discrete set of dispersionless modes, i.e., with frequencies independent of the wavevectors $k$, at low and high frequencies, whereas two dispersive SWs modes were found in the intermediate frequency range. The theoretical SW dispersions and magnonic band structure were determined by micromagnetic (MuMax3) simulations while COMSOL calculations were carried out to reconstruct the spatial profiles of the main SW modes. The results indicated that propagating modes correspond to SWs with a spatial profile extending throughout the structures. Finally, the magnonic band structure and mode profiles were compared to those of identical meander structures comprising a single CoFeB layer.
The samples were processed as follows: in a first step, 300 nm thick thermal SiO2 was grown on a 300 mm Si (100) wafer and patterned into a periodic grating (height h =50 nm, line and trench widths of 300 nm each) by conventional deep-UV photolithography and reactive ion etching. Next, a Ta(2 nm)/Co40Fe40B20(23 nm)/Ta(2 nm) stack was deposited by physical vapor deposition (PVD) onto the grating, as shown in Fig. 1. The length of the horizontal segments was L1 = 300 nm and L2 =150 nm, so that the unit cell had a periodicity of a = L1 + 2 L2 = 600 nm, resulting in BZedge of π/a =0.52 × 107 rad/m. The limited conformality of the PVD process led to approximately half the film thickness on the vertical segments (sidewalls) with respect to the horizontal ones.1717. E. N. Beginin, A. V. Sadovnikov, A. Yu. Sharaevskaya, A. I. Stognij, and S. A. Nikitov, Appl. Phys. Lett. 112, 122404 (2018). https://doi.org/10.1063/1.5023138 After air exposure, a Ta layer of 3 nm was deposited onto the structure, leading to a total Ta spacer of t1 = 5 nm (t2 ≈ 2.5 nm), followed by the in situ deposition of a 23 nm thick Ni80Fe20 magnetic layer. An otherwise identical sample without the NiFe layer was also fabricated used as a reference. Figures 1(b) and 1(c) show scanning electron microscopy (SEM) images of the CoFeB/Ta/NiFe sample. It can be seen clearly that all layers coat the patterned SiO2 structures and are connected by sidewall segments, despite the limited conformality of the PVD process.
The SW dispersion was measured by BLS in the backscattering configuration using a Sandercock (3 + 3)-pass tandem Fabry–Pérot interferometer.1919. J. R. Sandercock, in Light Scattering in Solids III, Springer Series in Topics in Applied Physics Vol. 51, edited by M. Cardona and G. Guntherodt ( Springer-Verlag, Berlin, 1982), p. 173. An in-plane magnetic field of μ0H =50 mT was applied along the groove length (z-direction) and perpendicular to the incidence plane of light (xy plane), in the so-called magnetostatic surface wave (MSSW) configuration. BLS spectra were recorded by changing the incidence light angle from 0 to 60° in steps of 2°. This corresponds to a sweep of in-plane SW wavevector k = (4π/λ)×sin(θ) along the x-direction from 0 to 2.05 × 107 rad/m, with λ = 532 nm the wavelength of the laser.2020. G. Carlotti and G. Gubbiotti, J. Phys.: Condens. Matter 14, 8199 (2002). https://doi.org/10.1088/0953-8984/14/35/303 The hysteresis loop (not shown), measured by vibrating sample magnetometry with the external magnetic field applied along the z-direction, was square with a coercive field of about 5 mT. No evidence of separate switching fields of the individual CoFeB and NiFe layers was found. The hysteresis loop further indicated that the sample magnetization was saturated at μ0H =50 mT, the field used for the BLS measurements.
Figure 2(a) represents a sequence of BLS spectra for the CoFeB/Ta/NiFe sample measured at different incidence angles, i.e., at different wavevectors k. The spectra are characterized by a significant asymmetry of the peak intensity when comparing negative (Stokes) and positive (anti-Stokes) frequency shifts. Consequently, some peaks are only clearly visible at the Stokes side or vice versa. Three frequency regions could be identified, as marked by the blue (7.5–10.5 GHz), yellow (16.1–22.1 GHz), and green (25.3–28.0 GHz) areas in Fig. 2(a). The regions are characterized by peaks whose frequency position does not changeover the range of explored wavevectors. A notable feature was the variation of the peak intensity as a function of k, which is a well-known effect for stationary modes and depends on the BLS cross section dependence on the mode spatial profiles and their symmetry with respect to the incident plane of light.2121. J. Jorzick, S. O. Demokritov, C. Mathieu, B. Hillebrands, B. Bartenlian, C. Chappert, F. Rousseaux, and A. N. Slavin, Phys. Rev. B 60, 15194 (1999). https://doi.org/10.1103/PhysRevB.60.15194 For example, in the lowest frequency range (blue region), a triplet of modes was observed whose intensities change as a function of k. The peak at 7.52 GHz had the largest intensity at small wavevectors and decreased monotonically with k. By contrast, the peak at 8.78 GHz had maximum intensity around k ≈ 1.2 × 107 rad/m. In the intermediate frequency range between the blue and yellow regions, two peaks were detected that are marked by the red (blue) arrows in the Stokes (anti-Stokes) spectra. These peaks exhibited significant frequency variation with k, which indicates that the corresponding modes are dispersive. Furthermore, these two peaks showed opposite frequency dependences: the Stokes (anti-Stokes) peak frequency increased (decreased) up to k =1.18 × 107 rad/m; however, the trend was subsequently reversed for at higher wavevectors.
In Fig. 2(b), the measured BLS spectrum at k =0.41 × 107 rad/m of a reference meander structure with a single CoFeB film is shown for comparison. Significant differences were observed between the spectra of the two samples in terms of mode frequencies and intensities. For example, there were no peaks below 12.5 GHz for the single-layer CoFeB sample, whereas the CoFeB/Ta/NiFe sample presented a triplet of peaks, as discussed above. This indicates that the triplet corresponds to either SW modes localized in the NiFe layer or to hybridized modes in the CoFeB/Ta/NiFe magnetic bilayer. Both samples display weakly dispersive peaks near 27 GHz, as seen in the green area of Figs. 2(a) and 2(b). The modes can thus be attributed to first-order perpendicular standing SW modes (PSSW) resonating across the thickness of the CoFeB film.
To model the magnonic band structure of the 3D MC, micromagnetic simulations were performed using the open-source GPU accelerated MuMax3 software.2222. A. Vansteenkiste, J. Leliaert, M. Dvornik, M. Helsen, F. Garcia-Sanchez, and B. Van Waeyenberge, AIP Adv. 4, 107133 (2014). https://doi.org/10.1063/1.4899186 The single-layer CoFeB and magnetic bilayer CoFeB/Ta/NiFe meander-shaped unit cell structure was discretized into cubic micromagnetic cells of dimensions $Δx×Δy×Δz$ = 1 × 1 × 1 nm3. To calculate the SW dispersion relation, Np = 31 periods of the meander structure were simulated. Hence, the total length of the simulated structure in x-direction was L = Np × a =18.6 μm. In the area $L×s$, where s = h +2d1+t1, a rectangular grid with the size of $Nx×Ny$ was specified, where $Nx=L/Δx,Ny=L/Δy$ are the number of grid nodes along the x and y axes, respectively. An out-of-plane sinc-shaped magnetic field with an amplitude of 1 mT and a cutoff frequency of fc = 27 GHz was applied in a 50 nm wide region in the center of the unit cell to excite SWs. The sample was magnetized by a bias field of μ0H =50 mT along the z-axis [see Fig. 1(a)]. This procedure was carried out in the time domain within a simulation time of T = 300 ns. The sampling time step was 9.26 ps. The SW dispersion was then obtained by performing a 2D Fast Discrete Fourier Transform.2323. A. V. Sadovnikov, S. A. Odintsov, E. N. Beginin, A. A. Grachev, V. A. Gubanov, S. E. Sheshukova, Y. P. Sharaevskii, and S. A. Nikitov, JETP Lett. 107, 25–29 (2018). https://doi.org/10.1134/S0021364018010113 The magnetic materials were represented in the MuMax3 simulations by the following parameters: saturation magnetization Ms (CoFeB) = 1275 kA/m, Ms (NiFe) = 795 kA/m, exchange constant Aex (CoFeB) = 1.43 × 10−11J/m, Aex (NiFe) = 1.11 × 10−11J/m.21,2421. J. Jorzick, S. O. Demokritov, C. Mathieu, B. Hillebrands, B. Bartenlian, C. Chappert, F. Rousseaux, and A. N. Slavin, Phys. Rev. B 60, 15194 (1999). https://doi.org/10.1103/PhysRevB.60.1519424. A. Vansteenkiste and B. Van De Wiele, J. Magn. Magn. Mater. 323, 2585 (2011). https://doi.org/10.1016/j.jmmm.2011.05.037
Figure 3 presents a comparison between measured and calculated dispersion relations for the CoFeB/Ta/NiFe and CoFeB meander structures, as well as the dispersion of an equivalent planar CoFeB/Ta/NiFe magnetic bilayer. In general, excellent agreement was observed between measured and calculated dispersion relations. For the CoFeB/Ta/NiFe meander structure [Fig. 3(a)], the observed three modes below 11 GHz (labeled I, II, and III) and the two modes above 16 GHz (modes VI and VII) were dispersionless, i.e., their frequency was independent of the wavevector. By contrast, two modes (IV and V) exhibited an antiphase frequency oscillation with identical amplitude in the frequency range between 12 and 15.3 GHz. These two modes correspond to Bloch-type SWs that propagate over the entire sample. Mode crossing and the absence of a bandgap were observed at k = /a (n is an odd number), similar to what already seen in CoFeB meander-shaped film and explained in terms of gliding-plane symmetry of the sample.1717. E. N. Beginin, A. V. Sadovnikov, A. Yu. Sharaevskaya, A. I. Stognij, and S. A. Nikitov, Appl. Phys. Lett. 112, 122404 (2018). https://doi.org/10.1063/1.5023138 Furthermore, when compared to the behavior of the single-layer CoFeB MC,1717. E. N. Beginin, A. V. Sadovnikov, A. Yu. Sharaevskaya, A. I. Stognij, and S. A. Nikitov, Appl. Phys. Lett. 112, 122404 (2018). https://doi.org/10.1063/1.5023138 the dispersive modes in the CoFeB/Ta/NiFe sample had smaller group velocity and a smaller magnonic bandwidth (3.3 GHz vs 5.5 GHz).
The dispersion relation of the planar CoFeB/Ta/NiFe magnetic bilayer contains two modes, which due to dynamic dipolar coupling, correspond to acoustic (at a higher frequency) and optic modes (at a lower frequency) and are associated with the in-phase and out-of-phase precession of the dynamic magnetization in the two magnetic layers, respectively.2525. G. Gubbiotti, M. Kostylev, N. Sergeeva, M. Conti, G. Carlotti, T. Ono, A. N. Slavin, and A. Stashkevich, Phys. Rev. B 70, 224422 (2004). https://doi.org/10.1103/PhysRevB.70.224422 The acoustic mode is mainly localized in the bottom CoFeB layer whereas the optic branch possesses higher intensity in the NiFe layer. It is worth noting that the frequency of the dispersive modes with positive group velocity (i.e., modes IV) in the two meander structures depended on $k$ in a similar manner than the acoustic mode in planar CoFeB/Ta/NiFe films [see Fig. 3(c)].
To calculate the mode spatial profiles, we have performed the eigenmode analysis for the meander bilayer using COMSOL and the mode separation technique described in Refs. 2323. A. V. Sadovnikov, S. A. Odintsov, E. N. Beginin, A. A. Grachev, V. A. Gubanov, S. E. Sheshukova, Y. P. Sharaevskii, and S. A. Nikitov, JETP Lett. 107, 25–29 (2018). https://doi.org/10.1134/S0021364018010113 and 2626. A. V. Sadovnikov, K. V. Bublikov, and E. N. Beginin, J. Commun. Technol. Electron. 59, 914 (2014). https://doi.org/10.1134/S106422691408018X. Here we take into account one period of the bilayer (primitive cell), which is shown in Fig. 1. The periodic boundary conditions are used in the x-direction. The fields are assumed to be in the form of Bloch wave with wavevector component kx, which was perpendicular to the external magnetic field.
The normalized dynamic component of magnetization (My/Ms) at $k=0$ (i.e., in the center of the BZ), as shown in Fig. 4. For the CoFeB/Ta/NiFe meander structure [Fig. 4(a)], the spatial distributions of both dispersive and nondispersive modes were symmetric with respect to the center of the unit cell (x =0). Furthermore, the simulations show that the amplitude of modes I, II, and III were concentrated in the topmost NiFe layer with an increasing number of nodes in the horizontal segments and resulting in an increase in the mode frequency. A sizeable and out-of-phase oscillation amplitude is also observed in the vertical segments of the CoFeB and NiFe layers. Moreover, since this triplet of modes lies in the frequency range of the optic mode of the planar structure, as shown in Fig. 3(c), they can be interpreted as due to the spatial quantization of this mode.2525. G. Gubbiotti, M. Kostylev, N. Sergeeva, M. Conti, G. Carlotti, T. Ono, A. N. Slavin, and A. Stashkevich, Phys. Rev. B 70, 224422 (2004). https://doi.org/10.1103/PhysRevB.70.224422 We notice that the frequencies of these modes are below the ferromagnetic resonance of a continuous CoFeB film (the saturation magnetization of NiFe is smaller than that of CoFeB) so that the magnetization precession in CoFeB is not resonantly excited but undergoes a forced oscillation with reduced amplitude.27,2827. M. L. Sokolovsky and M. Krawczyk, J. Nanopart. Res. 13, 6085 (2011). https://doi.org/10.1007/s11051-011-0303-528. G. Gubbiotti, S. Tacchi, M. Madami, G. Carlotti, S. Jain, A. O. Adeyeye, and M. P. Kostylev, Appl. Phys. Lett. 100, 162407 (2012). https://doi.org/10.1063/1.4704659
By contrast, the profile of mode IV was quasi-uniform within the CoFeB layer [Fig. 4(b)] and showed negligible amplitude in NiFe. This mode had an extended spatial character and corresponded to the dispersive mode represented by the blue circles in Fig. 3(a). A similar distribution was observed also for the single-layer CoFeB MC [see Fig. 4(b)], even if the corresponding frequency of this mode was lower at around 10 GHz. Mode V of the CoFeB/Ta/NiFe MC also had an extended profile with the spin precession amplitude mainly concentrated in NiFe, although it was complemented by a sizeable in-phase amplitude in the CoFeB layer. This supports the propagating character of this mode, in agreement with the experimental results. Similar behavior was also observed for the single-layer CoFeB meander structure. Mode VI was mainly concentrated in the top NiFe and exhibited a large number of oscillations. This explains its relatively small peak intensity observed in the experimental spectra in Fig. 2(a).
In conclusion, we have studied both experimentally and numerically the SW dispersion relation of in-plane saturated CoFeB/Ta/NiFe meander-shaped MCs and compared the results to those of an equivalent sample comprising a single CoFeB layer. The magnonic band structures of the two samples are markedly different, both in terms of observed modes and in their dependences on the wavevector, i.e., whether they are stationary or dispersive in nature. A narrower width of the magnonic band has been observed for the CoFeB/Ta/NiFe structure than for the CoFeB sample. This can be related to the interlayer dipolar coupling which changes the spin-wave dispersion relation.2929. M. Grassi, M. Geilen, D. Louis, M. Mohseni, T. Brächer, M. Hehn, D. Stoeffler, M. Bailleul, P. Pirro, and Y. Henry, Phys. Rev. Appl. 14, 024047 (2020). https://doi.org/10.1103/PhysRevApplied.14.024047 The properties of the individual modes have been further characterized by the phase relation (in-phase or out-of-phase) between the magnetization oscillations in the two layers and their localization in the horizontal and vertical segments. The results show that meander structures containing layered magnetic structures can be considered as prototypes of 3D MCs that may be used as a basic element in complex multilevel magnonic waveguide networks.
Imec's contribution to this work has been supported by its industrial affiliate program on beyond-CMOS logic as well as by the European Union's Horizon 2020 research and innovation program within the FET-OPEN project CHIRON under Grant Agreement No. 801055. The numerical simulation and theoretical model for three-dimensional magnonic band structure has been supported by the Russian Science Foundation (Project No. 20-79-10191). E.B. acknowledges support from the Russian Ministry of Education and Science (Project No. FSRR-2020-0005) and Russian Foundation for Basic Research (Project No. 19-29-03034). S.N. acknowledges support by the Russian Science Foundation (Project No. 19-19-00607). The authors would like to thank Danny Wan, Anshul Gupta, Shreya Kundu, and Robert Carpenter at imec for support of the sample fabrication.
The data that support the findings of this study are available from the corresponding author upon reasonable request.
1. 1. J. Topp, D. Heitmann, M. P. Kostylev, and D. Grundler, Phys. Rev. Lett. 104, 207205 (2010). https://doi.org/10.1103/PhysRevLett.104.207205, Google ScholarCrossref
2. 2. G. Gubbiotti, X. Zhou, Z. Haghshenasfard, M. G. Cottam, and A. O. Adeyeye, Phys. Rev. B 97, 134428 (2018). https://doi.org/10.1103/PhysRevB.97.134428, Google ScholarCrossref
3. 3. D. Kumar, J. W. Klos, M. Krawczyk, and A. Barman, J. Appl. Phys. 115, 043917 (2014). https://doi.org/10.1063/1.4862911, Google ScholarScitation, ISI
4. 4. P. Frey, A. A. Nikitin, D. A. Bozhko, S. A. Bunyaev, G. N. Kakazei, A. B. Ustinov, B. A. Kalinikos, F. Ciubotaru, A. V. Chumak, Q. Wang, V. S. Tiberkevich, B. Hillebrands, and A. A. Serga, Commun. Phys. 3, 17 (2020). https://doi.org/10.1038/s42005-020-0281-y, Google ScholarCrossref
5. 5. A. Barman, S. Mondal, S. Sahoo, and A. De, J. Appl. Phys. 128, 170901 (2020). https://doi.org/10.1063/5.0023993, Google ScholarScitation, ISI
6. 6. M. Krawczyk and D. Grundler, J. Phys.: Condens. Matter 26, 123202 (2014). https://doi.org/10.1088/0953-8984/26/12/123202, Google ScholarCrossref
7. 7. S. Tacchi, G. Gubbiotti, M. Madami, and G. Carlotti, J. Phys.: Condens. Matter 29, 073001 (2017). https://doi.org/10.1088/1361-648X/29/7/073001, Google ScholarCrossref
8. 8. B. Obry, P. Pirro, T. Brächer, A. V. Chumak, J. Osten, F. Ciubotaru, A. A. Serga, J. Fassbender, and B. Hillebrands, Appl. Phys. Lett. 102, 202403 (2013). https://doi.org/10.1063/1.4807721, Google ScholarScitation, ISI
9. 9. M. Inoue, A. Baryshev, H. Takagi, P. B. Lim, K. Hatafuku, J. Noda, and K. Togo, Appl. Phys. Lett. 98, 132511 (2011). https://doi.org/10.1063/1.3567940, Google ScholarScitation, ISI
10. 10. H. Merbouche, M. Collet, M. Evelt, V. E. Demidov, J. L. Prieto, M. Munoz, J. B. Youssef, G. de Loubens, O. Klein, S. Xavier, O. D'Allivy Kelly, P. Bortolotti, V. Cros, A. Anane, and S. O. Demokritov, ACS Appl. Nano Mater. 4, 1 121 (2021). https://doi.org/10.1021/acsanm.0c02382, Google ScholarCrossref
11. 11. A. V. Chumak, A. A. Serga, and B. Hillebrands, Nat. Commun. 5, 4700 (2014). https://doi.org/10.1038/ncomms5700, Google ScholarCrossref, ISI
12. 12. L. Brunet, P. Batude, C. Fenouillet-Beranger et al., in IEEE Symposium on VLSI Technology, Honolulu, HI, USA (2016), p. 1. Google Scholar
13. 13. A. Vandooren, L. Witters, J. Franco et al., in IEEE Symposium on VLSI Technology, Honolulu, HI, USA, 69 (2018), p. 69. Google Scholar
14. 14. G. Gubbiotti, Three-Dimensional Magnonics ( Jenny Stanford, Singapore, 2019). Google ScholarCrossref
15. 15. V. K. Sakharov, E. N. Beginin, Y. V. Khivintsev, A. V. Sadovnikov, A. I. Stognij, Y. A. Filimonov, and S. A. Nikitov, Appl. Phys. Lett. 117, 022403 (2020). https://doi.org/10.1063/5.0013150, Google ScholarScitation, ISI
16. 16. A. A. Martyshkin, E. N. Beginin, A. I. Stognij, S. A. Nikitov, and A. V. Sadovnikov, IEEE Magn. Lett. 10, 5511105 (2019). https://doi.org/10.1109/LMAG.2019.2957264, Google ScholarCrossref
17. 17. E. N. Beginin, A. V. Sadovnikov, A. Yu. Sharaevskaya, A. I. Stognij, and S. A. Nikitov, Appl. Phys. Lett. 112, 122404 (2018). https://doi.org/10.1063/1.5023138, Google ScholarScitation, ISI
18. 18. G. Gubbiotti, A. Sadovnikov, E. Beginin, S. Nikitov, D. Wan, A. Gupta, S. Kundu, G. Talmelli, R. Carpenter, I. Asselberghs, I. P. Radu, C. Adelmann, and F. Ciubotaru, Phys. Rev. Appl. 15, 014061 (2021). https://doi.org/10.1103/PhysRevApplied.15.014061, Google ScholarCrossref
19. 19. J. R. Sandercock, in Light Scattering in Solids III, Springer Series in Topics in Applied Physics Vol. 51, edited by M. Cardona and G. Guntherodt ( Springer-Verlag, Berlin, 1982), p. 173. Google ScholarCrossref
20. 20. G. Carlotti and G. Gubbiotti, J. Phys.: Condens. Matter 14, 8199 (2002). https://doi.org/10.1088/0953-8984/14/35/303, Google ScholarCrossref
21. 21. J. Jorzick, S. O. Demokritov, C. Mathieu, B. Hillebrands, B. Bartenlian, C. Chappert, F. Rousseaux, and A. N. Slavin, Phys. Rev. B 60, 15194 (1999). https://doi.org/10.1103/PhysRevB.60.15194, Google ScholarCrossref
22. 22. A. Vansteenkiste, J. Leliaert, M. Dvornik, M. Helsen, F. Garcia-Sanchez, and B. Van Waeyenberge, AIP Adv. 4, 107133 (2014). https://doi.org/10.1063/1.4899186, Google ScholarScitation, ISI
23. 23. A. V. Sadovnikov, S. A. Odintsov, E. N. Beginin, A. A. Grachev, V. A. Gubanov, S. E. Sheshukova, Y. P. Sharaevskii, and S. A. Nikitov, JETP Lett. 107, 25–29 (2018). https://doi.org/10.1134/S0021364018010113, Google ScholarCrossref
24. 24. A. Vansteenkiste and B. Van De Wiele, J. Magn. Magn. Mater. 323, 2585 (2011). https://doi.org/10.1016/j.jmmm.2011.05.037, Google ScholarCrossref
25. 25. G. Gubbiotti, M. Kostylev, N. Sergeeva, M. Conti, G. Carlotti, T. Ono, A. N. Slavin, and A. Stashkevich, Phys. Rev. B 70, 224422 (2004). https://doi.org/10.1103/PhysRevB.70.224422, Google ScholarCrossref
26. 26. A. V. Sadovnikov, K. V. Bublikov, and E. N. Beginin, J. Commun. Technol. Electron. 59, 914 (2014). https://doi.org/10.1134/S106422691408018X, Google ScholarCrossref
27. 27. M. L. Sokolovsky and M. Krawczyk, J. Nanopart. Res. 13, 6085 (2011). https://doi.org/10.1007/s11051-011-0303-5, Google ScholarCrossref
28. 28. G. Gubbiotti, S. Tacchi, M. Madami, G. Carlotti, S. Jain, A. O. Adeyeye, and M. P. Kostylev, Appl. Phys. Lett. 100, 162407 (2012). https://doi.org/10.1063/1.4704659, Google ScholarScitation, ISI
29. 29. M. Grassi, M. Geilen, D. Louis, M. Mohseni, T. Brächer, M. Hehn, D. Stoeffler, M. Bailleul, P. Pirro, and Y. Henry, Phys. Rev. Appl. 14, 024047 (2020). https://doi.org/10.1103/PhysRevApplied.14.024047, Google ScholarCrossref