Large exchange bias in polycrystalline MnN/CoFe bilayers at room temperature

We report on the new polycrystalline exchange bias system MnN/CoFe, which shows exchange bias of up to 1800 Oe at room temperature with a coercive field around 600 Oe. The room-temperature values of the interfacial exchange energy and the effective uniaxial anisotropy are estimated to be $J_{\mathrm{eff}}=0.41\mathrm{mJ}/{\mathrm{m}}^2$ and $K_{\mathrm{eff}}=37\mathrm{kJ}/{\mathrm{m}}^3$. The thermal stability was found to be tunable by controlling the nitrogen content of MnN. The maximum blocking temperature exceeds $325{}^{\circ }\mathrm{C}$, however the median blocking temperature in the limit of thick MnN is $160{}^{\circ }\mathrm{C}$. Good oxidation stability through self-passivation was observed, enabling the use of MnN in lithographically defined microstructures. As a proof of principle we demonstrate a simple giant magnetoresistance stack exchange biased with MnN, which shows clear separation between parallel and antiparallel magnetic states. These properties come along with a surprisingly simple manufacturing process for the MnN films.

Figure 1

(a) X-ray diffraction spectrum of a Ta 10/MnN 32 film annealed at $300{}^{\circ }\mathrm{C}$ for 15 min. (b) Lattice constants, (c) full width at half maximum (FWHM) of the rocking curves of the (002) peaks, (d) perpendicular grain sizes, and (e) microstrain of as-deposited and annealed films as functions of the MnN thickness. (f) Lattice constants as functions of annealing temperature for two MnN film thicknesses. The inset shows the lattice constant as a function of the ${\mathrm{N}}_2$ flow in the sputtering gas. The dashed lines in (b) and (f) represent the lattice constant $a$ from Ref. [16].

Figure 2

(a) MOKE loops parallel and perpendicular (in the sample plane) to the field cooling direction for a sample with $t_{\mathrm{MnN}}=32$ and $t_{\mathrm{CoFe}}=1.6\mathrm{nm}$. (b) The dependence of $H_{\mathrm{EB}}$ and $H_{\mathrm{C}}$ on the MnN thickness with $t_{\mathrm{CoFe}}=1.8\mathrm{nm}$. (c) The dependence of $H_{\mathrm{EB}}$ and $H_{\mathrm{C}}$ on the CoFe thickness with $t_{\mathrm{MnN}}=30\mathrm{nm}$. The samples were annealed at $325{}^{\circ }\mathrm{C}$ for 15 min. Dotted lines are guides to the eye.

Figure 3

Deposition pressure dependence of (a) exchange bias and coercive field ($t_{\mathrm{MnN}}=30$ and $t_{\mathrm{CoFe}}=1.8\mathrm{nm}$, annealing at $325{}^{\circ }\mathrm{C}$, 15 min). (b) Surface roughness (root mean square) from x-ray reflectivity, (c) rocking curve FWHM of the (002) x-ray reflection, and (d) microstrain parameter ${\varepsilon }_{\mathrm{MnN}}$ for $t_{\mathrm{MnN}}=30\mathrm{nm}$ without annealing.

Figure 4

Annealing temperature dependences of $H_{\mathrm{EB}}$ and $H_{\mathrm{C}}$ for different nitrogen partial pressure ratios during the reactive sputtering of MnN. The total pressure was held at $2.3\times {10}^{-3}\mathrm{mbars}$. Film thicknesses were $t_{\mathrm{CoFe}}=1.8$ and $t_{\mathrm{MnN}}=32\mathrm{nm}$.

Figure 5

Results from reversed field cooling experiments with four different MnN thicknesses. (a) Normalized exchange bias field $H_{\mathrm{EB}}$, (b) unblocked ratio, and (c) derivative of the unblocked ratio versus reversal temperature. Initial and reversal fields were 6.5 kOe.

Figure 6

(a) ${\mathrm{MnO}}_x$ thickness after annealing a Ta 10/MnN 30 bilayer in air for 1 h. The fit represents Eq. (3). (b) GMR loops at room temperature of (I) $\mathrm{Ta}10/\mathrm{MnN}30/\mathrm{CoFe}2.2/\mathrm{Cu}1.8/\mathrm{CoFe}2.2/\mathrm{Ta}0.5/{\mathrm{Ta}}_2{\mathrm{O}}_{5}1.0(250{}^{\circ }\mathrm{C},15\min )$ and (II) $\mathrm{Ta}10/\mathrm{MnN}30/\mathrm{CoFe}2.2/\mathrm{Cu}2.8/\mathrm{CoFe}2.2/\mathrm{Ta}0.5/{\mathrm{Ta}}_2{\mathrm{O}}_{5}1.0(325{}^{\circ }\mathrm{C},15\min )$.