Research articles
Enhancement of Brillouin light scattering signal with anti-reflection layers on magnetic thin films

https://doi.org/10.1016/j.jmmm.2020.166565Get rights and content

Highlights

  • Brillouin light scattering signal is significantly enhanced over 450% by introducing an additional anti-reflection layer.

  • Due to the large signal enhancements, the accuracy of the interfacial Dzyaloshinskii-Moriya interaction energy density measurement is increased as well.

Abstract

The significant enhancement of Brillouin light scattering (BLS) spectroscopy intensity in a ferromagnetic thin film with an additional dielectric anti-reflection layer is experimentally investigated. The anti-reflection layer thickness dependent BLS measurements on ferromagnetic layers are performed systematically. Consequently, we observe that BLS signals are dramatically enhanced by more than 450% at a specific dielectric layer thickness due to the pure optical effect. Because of the large signal enhancements, the errors of the spin wave resonance peak frequencies are noticeably reduced as well. Since many magnetic properties such as the saturation magnetization, the surface anisotropy, and the exchange stiffness constant are determined by the spin wave resonance frequencies from the BLS spectra, the additional anti-reflection layer can help to improve the reliability of BLS experiments. Especially, the BLS signal improvement plays a crucial role in the precise determination of the interfacial Dzyaloshinskii-Moriya interaction (iDMI) energy density, since the iDMI energy density is calculated from the difference of Stokes and anti-Stokes resonance frequencies, which is typically order of 1 GHz.

Introduction

The Brillouin light scattering (BLS) spectroscopy is a versatile tool to directly investigate thermally excited propagating spin waves (SWs) with a range of wave vector from 103 to 105 cm−1 [1]. Past few decades, BLS has been widely used for the determination of various magnetic properties such as the saturation magnetization (MS), the perpendicular magnetic anisotropy (PMA) [2], [3], [4], the exchange stiffness constant (Aex) [5], [6], [7], and the magnetic damping constant (αdamping) [8], [9] in the magnetic thin films. This useful spectroscopy is based on an inelastic light scattering mechanism between propagating SWs in a ferromagnetic thin film and photons of visible lights. There are two considerable advantages of BLS measurements: First, many magnetometries such vibrating sample magnetometer (VSM) and torque magnetometer measure the energies, which should have certain errors from the evaluations of area or volume. However, BLS measurements are free from those errors, because the units of all measurable magnetic properties are the energies per area or volume in BLS measurement [10]. Second, many magnetic properties at a surface or an interface of magnetic thin film are precisely measured by the surface mode as well-known “Damon-Eshbach (DE) mode” [11]. Furthermore, the monolayer sensitivity of the BLS spectroscopy is apt to measure for ultrathin magnetic films. Recently, BLS spectroscopy has thrown new lights on the quantitative determination of the interfacial Dzyaloshinskii-Moriya interaction (iDMI) energy density at the interface between ferromagnets and heavy metals via non-reciprocal SW dispersion relations [12], [13], [14], [15], [16], [17], [18], [19], [20], [21].

Contrary to numerous advantages, the impoverished BLS intensity, which has several origins, is one of the serious disadvantages of BLS measurements. For the detail of BLS mechanism is that, the total amplitude of a SW mode is connected with the number of nm of excited magnons, which is given by nm=1expωkBT-1 [22]. Thermal fluctuations excite the magnetic normal mode of the film, causing the magnetization to precess around its equilibrium direction. The refraction index of the magnetic film is modulated by the precessing magnetization direction by the magneto-optical effect. In the point of magnon, the modulated refractive index forms a grating. The incident light scattered by the magnon grating when the diffraction conditions are satisfied. Therefore, in the experiment geometry, back-scattering geometry of BLS spectroscopy with an inelastic light scattering process, the number of back-scattered photons is extremely small compared the reflected photons. Therefore, the BLS intensity increase is mainly due to an increase of the thermal SWs amplitudes within scattering volume close to the surface of the film. And it is needed to enhanced interaction of the light scattering of the light with the magnetic film can lead to a remarkable increase of the BLS intensity. Besides, BLS measurement is based on the magneto-optical Kerr effect (MOKE), which is directly proportional to the off-diagonal elements of the dielectric tensor of ferromagnetic layer [23], [24], [25]. Here, the off-diagonal elements are generally much smaller than the diagonal elements. Therefore, BLS signals are much smaller than the Raman scattering signal, where the scattering lights are coming from the diagonal elements of the dielectric tensor. Secondly, the photon counting system of BLS setup, which is consist of photomultiplier tubes can have a certain noise level [10], [26]. Furthermore, the various defects of samples such as polycrystalline grain boundaries and surface roughness are sources of the random diffractions of lights. The disadvantages give us the uncertainties of experiment data. Therefore, the BLS has a relatively poor signal-to-noise ratio (SNR), which should be significantly improved to determine various physical properties quantitatively. In this study, we experimentally demonstrate significantly improve BLS intensities by introducing an additional anti-reflection (AR) layer. In Ref. [27], there is an experimental attempt to improve BLS signal with changing the thickness of SiO2 dielectric buffer layer [27]. They have shown that the BLS signal is strongly dependent on the thickness of SiO2 substrates and enhance the BLS intensity. To break the inversion symmetry of the system and improve the BLS signal simultaneously, we use a dielectric capping layer in this study. Consequently, we experimentally observe the intensity of the propagating spin wave signals from BLS spectra are increased over 450% by changing the thickness of the dielectric layer due to an AR coating effect. This fact indicates that the AR coating layer gives us precisely determined material parameters.

Section snippets

Physical mechanism and experimental details

This method, an anti-reflection coating, has been widely used to enhance the intensity of MOKE signals by employing a dielectric layer [28], [29], [30], [31], [32]. There are many common and different points in the physical mechanisms of the two individual measurements (MOKE and BLS). For the case of MOKE, the polarization of the reflected lights from ferromagnetic materials is changed and it is based on Fresnel reflection optics. On the other hand, BLS spectroscopy measures inelastically

Results

In order to verify our hypothesis that the signals of BLS measurements can be modulated by introducing additional MgO layer, a series of Ta(5 nm, buffer layer)/Pt(5 nm, heavy metal layer)/Co(2 nm, ferromagnetic layer)/MgO(tMgO)/Ta(2 nm, capping layer) samples are prepared as shown in Fig. 1(b). The multilayer systems are prepared on thermally oxidized Silicon wafers. All layers are deposited by using a magnetron sputtering system with the base pressure of 3×10−8 Torr. As a heavy metal layer, a

Conclusion

In conclusion, to determine the accurate magnon intensity, an inelastic light scattering measurement as a function of AR layer thickness is performed systematically. Since the underlying physics of BLS is the same with the MOKE, the BLS signals can be significantly improved at a certain thickness of the AR layer, when the MOKE signal is enhanced as well. From our systematic measurement, we reveal that the SW intensities of the Stokes and the anti-Stoke regions are significantly changed by an AR

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was supported by the National Research Foundation of Korea in the Ministry of Science and Education, and ICT (NRF-2017R1A2B3002621, NRF-2015M3D1A1070465, and NRF-2018R1D1A1B07051237).

References (44)

  • J. Cho et al.

    J. Magn. Magn. Mater.

    (2015)
  • S.O. Demokritov et al.

    Phys. Rep.

    (2001)
  • R.W. Damon et al.

    J. Phys. Chem. Solids

    (1961)
  • Spin dynamics in Confined Magnetic Structures I, edited by B. Hillebrands and K. Ounadjela (Springer, Berlin,...
  • J.R. Dutcher et al.

    J. Appl. Phys.

    (1988)
  • S. Park et al.

    J. Appl. Phys.

    (2002)
  • G. Gubbiotti et al.

    J. Appl. Phys.

    (2003)
  • S.P. Vernon et al.

    Phys. Rev. B

    (1984)
  • J. Cho et al.

    J. Magn. Magn. Mater.

    (2013)
  • J. Sinha et al.

    RSC Adv.

    (2015)
  • N.-H. Kim et al.

    Appl. Phys. Lett.

    (2016)
  • A. Fert et al.

    Nat. Nanotechnol.

    (2013)
  • J. Cho et al.

    Nat. Commun.

    (2015)
  • N.-H. Kim et al.

    Appl. Phys. Lett.

    (2015)
  • M. Belmeguenai et al.

    Phys. Rev. B

    (2015)
  • H.T. Nembach et al.

    Nat. Phys.

    (2015)
  • J. Cho et al.

    J. Phys. D Appl. Phys.

    (2017)
  • W.-Y. Kim et al.

    Appl. Phys. Expr.

    (2019)
  • K. Di et al.

    Phys. Rev. Lett.

    (2015)
  • A.V. Sadovnikov et al.

    Phys. Rev. Lett.

    (2018)
  • Avinash Kumar Chaurasiya et al.

    Phys. Rev. B

    (2019)
  • P. Grunberg, Ch. 8. Light scattering from Spin waves in Thin Films and Layered Magnetic Structures in Light Scattering...
  • 1

    This authors contributed equally to this work.

    View full text