Thickness dependence of magnetic properties of Ir/FeV-based systems

Ir-buffered ${\mathrm{Fe}}_{0.7}{\mathrm{V}}_{0.3}$ (FeV) thin films of variable thickness ($1.4\mathrm{nm}\le t_{\mathrm{FeV}}\le 10\mathrm{nm}$) were sputtered on thermally oxidized Si substrate and capped with Cu or MgO films. Vibrating sample magnetometery measurements allowed deducing their magnetization at saturation and the magnetic dead layer thickness, found to be capping layer independent. Microstrip line ferromagnetic resonance measurements under in-plane and perpendicular to the film plane applied magnetic fields are employed to investigate the perpendicular magnetic anisotropy (PMA) and damping. PMA is found to result from a uniaxial interfacial contribution which is slightly capping layer dependent. In contrast, the second-order PMA constant, which could result from the inhomogeneous distribution and the spatial fluctuations at the nanoscale of the uniaxial PMA at interface, is significantly stronger for Ir/FeV/Cu films. The damping dependence on the inverse of FeV effective thickness revealed a low Gilbert damping constant for FeV $({\alpha }_{\mathrm{FeV}}=1.7\times {10}^{-3})$ and allowed concluding that damping enhancement is mainly induced by Ir/FeV. The zero frequency linewidth for in-plane configuration is attributed to two magnon scattering contribution, found to be higher for Cu capped FeV. The investigation of the FeV thickness dependence of interfacial Dzyaloshinskii-Moriya interaction (iDMI) of both systems via Brillouin light scattering allowed to conclude to the zero iDMI constant of FeV/Cu. It also led to deduce iDMI constants of Ir/FeV and FeV/MgO, estimated to be $D_{\mathrm{s}}^{\mathrm{Ir}/\mathrm{FeV}}=(0.13\pm 0.02)\mathrm{pJ}/\mathrm{m}$ et $D_s^{\mathrm{FeV}/\mathrm{MgO}}=(-0.07\pm 0.02)\mathrm{pJ}/\mathrm{m}$, respectively.

Figure 1

Variation of the saturation magnetic moment per unit area versus the FeV thickness ($t_{\mathrm{FeV}}$) for Ir/FeV/Cu and Ir/FeV/MgO systems. Symbols are VSM measurements data and lines refer to the linear fits.

Figure 2

Variations of the uniform precession mode frequency versus of (a) the perpendicular to the film plane $\left(H_{\perp }\right)$ (c) the in-plane applied magnetic field ($H_{\parallel }$) for $\mathrm{Pt}/\mathrm{FeV}\left(t_{\mathrm{FeV}}\right)/\mathrm{MgO}$ system. Symbols refer to the experimental data and solid lines are fits using equations given in the paper. (b) MS-FMR spectra (measured at different microwave frequencies) representing the amplitude of the field derivative of the absorbed power as a function of the in-plane applied magnetic field for the 10-nm- and 2.8-nm-thick FeV films capped by MgO. The magnetic field was applied in the sample plane along the substrate edges. Symbols refer to experimental data and solid lines are fits using Eq. (1) of Ref. [16]. For more visibility, the amplitude of spectra at 5 GHz have been divided by 10 (6) for 10 nm (2.8 nm) FeV film.

Figure 3

(a) Eeffective magnetization $\left({\mu }_0{M}_{\mathrm{eff}}\right)$ versus the FeV reciprocal effective thickness $\left(1/\left(t_{\mathrm{FeV}}-t_d\right)\right)$ of $\mathrm{Pt}/\mathrm{FeV}\left(t_{\mathrm{FeV}}\right)/\mathrm{MgO}$ and $\mathrm{Pt}/\mathrm{FeV}\left(t_{\mathrm{FeV}}\right)/\mathrm{Cu}$ systems grown on Si substrates. ${\mu }_0{M}_{\mathrm{eff}}$ values have been extracted from the fit of MS-FMR measurements of the frequency of the uniform precession mode versus the in-plane and the perpendicular applied magnetic field using equations given in the main text. Symbols refer to experimental data while solid lines are the linear fits taking into account only the thicker films. The inset shows the second- order perpendicular magnetic anisotropy constant versus the FeV reciprocal effective thickness of the corresponding systems. Solid lines are used as guides to the eye. Variation of the resonance field versus of the out-of-plane angle measured at 9 GHz driving frequency for Ir/FeV/MgO system with (b) 2.8-nm- and (c) 10-nm-thick FeV layer. ${\theta }_H$ refers to the out-of-plane angle defining the direction of the applied field and the normal to the film plane. Symbols refer to experimental data and solid lines are fits using Eq. (1).

Figure 4

FMR half width at half maximum linewidth ($\mathrm{\Delta }H$) versus the driving frequency for (a) in-plane or (b) perpendicular to the film plane applied magnetic field of $\mathrm{Ir}/\mathrm{FeV}\left(t_{\mathrm{FeV}}\right)/\mathrm{MgO}$ of various FeV thicknesses. Symbols refer to experimental data and solid lines are linear fits. Variation of $\mathrm{\Delta }H$ versus the direction of the out-of-plane applied magnetic field with a respect to the normal to the sample plane (${\theta }_H$) measured at 9 GHz driving frequency for Ir/FeV/MgO system with (c) 2.8-nm- and (d) 10-nm-thick FeV layer. Symbols refer to experimental data and solid lines are fits using Gilbert, two magnon and mosaicity contribution.

Figure 5

(a) Gilbert damping parameter and (b) zero frequency linewidth, deduced from in-plane ($\parallel$) and perpendicular ($\perp$) applied magnetic fields measurements, as a function of the FeV reciprocal effective thickness of $\mathrm{Ir}/\mathrm{FeV}\left(t_{\mathrm{FeV}}\right)/\mathrm{MgO}$ and $\mathrm{Ir}/\mathrm{FeV}\left(t_{\mathrm{FeV}}\right)/\mathrm{Cu}$ systems. Symbols refer to experimental data and solid lines are linear fits.

Figure 6

Variation of the effective iDMI constants as a function of the FeV reciprocal effective thickness of $\mathrm{Ir}/\mathrm{FeV}\left(t_{\mathrm{FeV}}\right)/\mathrm{MgO}$ and $\mathrm{Ir}/\mathrm{FeV}\left(t_{\mathrm{FeV}}\right)/\mathrm{Cu}$ systems. Solid lines refer to the linear fit and symbols are experimental data.