Unraveling the influence of electronic and magnonic spin-current injection near the magnetic ordering transition of IrMn metallic antiferromagnets

Although spin injection at room temperature in an IrMn metallic antiferromagnet strongly depends on the transport regime, and is more efficient in the case of magnonic transport, in this article, we present experimental data demonstrating that the enhanced efficiency of spin injection caused by spin fluctuations near the ordering temperature can be as efficient for the electronic and magnonic transport regimes. By selecting representative interacting environments, we also demonstrated that the amplification of spin injection near the ordering temperature of the IrMn antiferromagnet is independent of exchange coupling with an adjacent NiFe ferromagnet. In addition, our findings confirm that the spin current carried by magnons penetrates deeper than that transported by conduction electrons in IrMn. Finally, our data indicates that the value of the ordering temperature for the IrMn antiferromagnet is not significantly affected by either the electronic or magnonic nature of the spin-current probe, or by exchange coupling.

Figure 1

(a), (b) Spin-pumping experiments: out-of-equilibrium magnetization ($M$) dynamics of the NiFe ferromagnet pumps electronic ($I_S^{el}$) and magnonic spin currents ($I_S^{\mathrm{mag}}$). Compared to a NiFe/IrMn bilayer (a), potential transmission of the magnonic spin current in an NiFe/Cu/IrMn trilayer (b) involves additional electron-magnon conversion at interfaces since the nonmagnetic Cu only allows electronic transport. (c) Representative series of differential absorption spectra ($d{\chi }^{\prime \prime }/dH$ vs $H$) measured at different temperatures ($T$). The data correspond to a series of measurements for a NiFe(8)/IrMn(1.2) bilayer (nm). The lines were fitted to the data using a Lorentzian derivative. The peak-to-peak linewidth ($\mathrm{\Delta }{H}_{pp}$) provides information on the amount of spin current transmitted and absorbed by the IrMn antiferromagnet (${\alpha }^p$). (d) Representative frequency dependence of $\mathrm{\Delta }{H}_{pp}$ measured at 300 K. The lines are linear fit. The data correspond to measurements for NiFe(8)/IrMn(0.6) and NiFe(8)/IrMn(1.2) bilayers (nm).

Figure 2

Temperature ($T$) dependence of (a), (b) the NiFe layer's Gilbert damping (α), (c), (d) the IrMn antiferromagnet contribution to NiFe damping (${\alpha }^p=\alpha -{\alpha }^{\mathrm{ref}}$), and (e),(f) the NiFe resonance field ($H_{\mathrm{res}}$) as a function of the IrMn antiferromagnet's thickness ($t_{\mathrm{IrMn}}$) for two representative series of samples: NiFe(8)/Cu(3)/IrMn($t_{\mathrm{IrMn}}$) trilayers and NiFe(8)/IrMn($t_{\mathrm{IrMn}}$) bilayers (nm). (a) and (c) are adapted from our previous work [33] to allow comparison. $\delta {\alpha }^p$ denotes the extra contribution to damping due to the magnetic phase transition of the IrMn antiferromagnet and $T_{\mathrm{crit}}^{\mathrm{IrMn}}$ stands for the corresponding critical temperature. In (c), data were shifted vertically to facilitate reading, the native values are (0.2, 1.4, 0.75, and 1.25) $\times {10}^{-3}$ for $t_{\mathrm{IrMn}}=0.6$, 0.8, 1, and 1.2 nm.

Figure 3

IrMn thickness ($t_{\mathrm{IrMn}}$)dependence of (a) the contribution to damping due to the magnetic phase transition of the IrMn antiferromagnet ($\delta {\alpha }^p$), and (b) the corresponding critical temperature ($T_{\mathrm{crit}}^{\mathrm{IrMn}}$) for NiFe(8)/IrMn($t_{\mathrm{IrMn}}$) bilayers (nm) and NiFe(8)/Cu(3)/IrMn($t_{\mathrm{IrMn}}$) trilayers (determined after subtraction of a baseline in a way that it follows the natural trend of the signal [33], open symbol, or considering a constant baseline, dotted signal). In (a), the dashed line corresponds to a constant fit and the straight line to a linear fit of the data constrained to pass through (0,0). In (b), line fitting was based on the equation presented in Ref. [40] in the thin-layer regime.

Figure 4

Temperature ($T$) dependence of (a), (b) the NiFe layer's Gilbert damping ($\alpha$), and (c), (d) the NiFe resonance field ($H_{\mathrm{res}}$), for a range of NiFe ferromagnet thicknesses ($t_{\mathrm{NiFe}}$), as recorded for two representative series of samples: NiFe($t_{\mathrm{NiFe}}$)/IrMn(0.6) and NiFe($t_{\mathrm{NiFe}}$)/IrMn(1.2) bilayers (nm). The lines in (c), (d) correspond to a fit to the high-temperature data for the NiFe(8)/IrMn(0.6) bilayer, using the Kittel equation and discarding the exchange bias terms.

Figure 5

Temperature ($T$) dependence of (a), (b) the exchange bias coupling field $\left({\mathrm{H}}_{\mathrm{E},\mathrm{st}}\right)$, and (c,d) the coercive field $\left(H_{C,\mathrm{st}}\right)$ when the NiFe ferromagnet thickness ($t_{\mathrm{NiFe}}$) is varied for two representative series of samples: $\mathrm{NiFe}\left(t_{\mathrm{NiFe}}\right)/\mathrm{IrMn}\left(0.6\right)$ and $\mathrm{NiFe}\left(t_{\mathrm{NiFe}}\right)/\mathrm{IrMn}\left(1.2\right)$ bilayers (nm). The inset in (b) shows representative hysteresis loops at various temperatures with the example of the NiFe(8)/IrMn(1.2) bilayer.

Figure 6

Temperature ($T$) dependence of (a) the rotational anisotropy ($H_{\mathrm{rot}}$) calculated from the data in Figs. 4 and 5, for a range of NiFe ferromagnet thicknesses ($t_{\mathrm{NiFe}}$), as recorded for two representative series of samples: $\mathrm{NiFe}\left(t_{\mathrm{NiFe}}\right)/\mathrm{IrMn}\left(0.6\right)$ and $\mathrm{NiFe}\left(t_{\mathrm{NiFe}}\right)/\mathrm{IrMn}\left(1.2\right)$ bilayers (nm). Line fitting is discussed in the text along with the corresponding universal behavior of $H_{\mathrm{rot}}{t}_{\mathrm{NiFe}}/t_{\mathrm{IrMn}}$ with $T/T_{\mathrm{rot}}$ plotted in (b). The line in (b) is a visual guide.

Figure 7

Temperature ($T$) dependence of (a) the NiFe layer's resonance field ($H_{\mathrm{res}}$), and (b) the peak-to-peak linewidth of the NiFe absorption spectrum ($\mathrm{\Delta }{H}_{pp}$), for a bias field applied in- and out-of- the plane of the sample, as recorded for two representative samples: NiFe(8)/IrMn(0.6) and NiFe(8)/IrMn(1.2) bilayers (nm). The straight lines in (a) correspond to a fit to the data, using the Kittel equations and including the exchange bias terms for the low-temperature data. For the sake of comparison, the dashed lines in (a) correspond to a fit for the NiFe(8)/IrMn(0.6) bilayer, discarding the exchange bias terms.

Figure 8

(a) NiFe thickness ($t_{\mathrm{NiFe}}$) dependence of (a) the IrMn antiferromagnet contribution to NiFe damping (${\alpha }^p$) measured at $T=T_{\mathrm{crit}}^{\mathrm{IrMn}}$ for $\mathrm{NiFe}\left(t_{\mathrm{NiFe}}\right)/\mathrm{IrMn}\left(t_{\mathrm{IrMn}}\right)$ bilayers with $t_{\mathrm{IrMn}}=0.6$, 0.8, 1, and 1.2 nm, and at $T=300\mathrm{K}$ when relevant, i.e., for $t_{\mathrm{IrMn}}=0.6$ nm. The lines are visual guides. (b) Corresponding NiFe thickness-dependence of $T_{\mathrm{crit}}^{\mathrm{IrMn}}$. Data for inverted $\mathrm{IrMn}\left(1.2\right)/\mathrm{NiFe}\left(t_{\mathrm{NiFe}}\right)$ bilayers with $t_{\mathrm{NiFe}}=25$ and 50 nm are plotted for comparison. Line fitting was based on the equation presented in Ref. [40] in the thin-layer regime and returned a phenomenological spin-spin correlation length $n_0$. Inset: $n_0$ vs $t_{\mathrm{NiFe}}$.