Structural and magnetic properties of granular Co-Pt multilayers with perpendicular magnetic anisotropy

We present a study of granular Co-Pt multilayers by means of high-resolution transmission electron microscopy (HRTEM), extended x-ray absorption fine structure (EXAFS), SQUID-based magnetic measurements, anomalous Hall effect (AHE), and x-ray magnetic circular dichroism (XMCD). We describe these granular films as composed of particles with a pure cobalt core surrounded by an alloyed Co-Pt interface, embedded in a Pt matrix. The alloy between the Co and Pt in these granular films, prepared by room temperature sputter deposition, results from interdiffusion of the atoms. The presence of this alloy gives rise to a high perpendicular magnetic anisotropy (PMA) in the granular films, as consequence of the anisotropy of the orbital moment in the Co atoms in the alloy, and comparable to that of highly-ordered CoPt $L{1}_0$ alloy films. Their magnetic properties are those of ferromagnetically coupled particles, whose coupling is strongly temperature dependent: at low temperatures, the granular sample is ferromagnetic with a high coercive field; at intermediate temperatures the granular film behaves as an amorphous asperomagnet, with a coupling between the grains mediated by the polarized Pt, and at high temperatures, the sample has a superparamagnetic behavior. The coupling/decoupling between the grains in our Co-Pt granular films can be tailored by variation of the amount of Pt in the samples.

Figure 1

HRTEM images in cross section configuration for samples with $t_{\mathrm{Co}}=$ 0.7 nm and (a) $t_{\mathrm{Pt}}=0.6$, (b) 1.5, (c) 4.5, and (d) 6.0 nm. Fast Fourier transform (FFT) at selected zones of the images are shown, with the spots indexed according to the interplanar distances of a fcc cell in all cases. A Co-Pt particle is highlighted in (a) and (b). Pt NPs are marked in (c) and (d).

Figure 2

STEM images in plan view configuration for the samples with $t_{\mathrm{Co}}=$ 0.7 nm, and (a) $t_{\mathrm{Pt}}=$ 0.6 and (b) 1.5 nm.

Figure 3

XANES at the Co K edge for the Co-Pt granular films and comparison with that of a Co foil for reference. (Inset) Variation of the areas of peaks I and II in the figure as a function of $t_{\mathrm{Pt}}$.

Figure 4

(a) Fourier transform of EXAFS signal at the Co $K$ edge on the Co-Pt samples. The measurements on the Co foil and bare Co nanoparticles with $t_{\mathrm{Co}}=$ 0.7 nm in are shown for comparison. (Inset) Zoom of the first coordination shell in the particulate systems. (b) Best fits to the first coordination shell in (a) for the Co-Pt nanoparticles in R space. (c) Contribution of the first coordination shell to the EXAFS signal for the Co-Pt nanoparticles along with its best fit. In (b) and (c), symbols represent experimental data and solid lines their best fits; curves have been vertically shifted for clarity.

Figure 5

XANES spectra at the Pt $L_3$ edge for the Co-Pt granular multilayer samples, along with the absorption of the Pt foil plotted for reference. (Inset) Details of the variation of the white line intensity.

Figure 6

(a) Fourier transform of EXAFS signal at the Pt $L_3$ edge on the Co-Pt granular films studied and comparison with that of the Pt foil. (b) Best fits to the first coordination shell in (a) for the Co-Pt nanoparticles in R space. (c) Contribution of the first coordination shell to the EXAFS signal for the Co-Pt nanoparticles along with their best fits. In (b) and (c) symbols represent experimental data and solid lines their best fits; curves have been vertically shifted for clarity.

Figure 7

Direct current (dc) susceptibility curves measured under an applied field of $H_{\mathrm{dc}}=$ 200 Oe perpendicular and parallel to the substrate plane for the Co-Pt sample with $t_{\mathrm{Pt}}=1.5$ nm. (b) $M\left(T\right)$ curves measured for the same sample at various fields applied in the parallel direction. ($\cdots$) $M(T_f,H_{\mathrm{dc}})$, superparamagnetic-soft ferromagnetic phase separation line. (c) $M\left(T\right)$ with the field applied in the perpendicular direction. (- - -) $M(T_1,H_{\mathrm{dc}})$ soft-hard ferromagnetic phase separation line. Values of $H_{\mathrm{dc}}$ are indicated beside each curve.

Figure 8

Real part of the ac susceptibility for the sample with $t_{\mathrm{Co}}=0.7$ nm and $t_{\mathrm{Pt}}=1.5$ nm measured with the applied ac field in the parallel configuration as a function of temperature using different excitation frequencies. (Inset) Analogous ac susceptibility curves measured for the same sample in the perpendicular configuration.

Figure 9

Magnetic phase diagram of the Co-Pt sample with $t_{\mathrm{Co}}=0.7$ nm, $t_{\mathrm{Pt}}=1.5$ nm and $N=25$, as deduced from $M(T,H_{\mathrm{dc}})$ measurements.

Figure 10

Magnetization curves measured for the sample with $t_{\mathrm{Co}}=0.7$ nm and $t_{\mathrm{Pt}}=1.5$ nm, with $H_{\mathrm{dc}}$ applied perpendicular () and parallel ($\blacksquare$) to the substrate: (a) $T=5$, (b) 200, and (c) 400 K. The dashed line in (c) is the fitted Langevin curve with $\langle D\rangle =3.5$ nm and $\sigma =0.4$ nm.

Figure 11

Temperature dependence of the coercive field measured with $H_{\mathrm{dc}}$ applied perpendicular to the substrate plane for the samples with $t_{\mathrm{Co}}=0.7$ nm and (a) () $t_{\mathrm{Pt}}=1.5$, (b) () $4.5$, and (b) () $6.0$ nm.

Figure 12

Average of the two branches of the hysteresis curves measured for the sample with $t_{\mathrm{Co}}=0.7$ nm and $t_{\mathrm{Pt}}=1.5$ nm, with $H_{\mathrm{dc}}$ applied perpendicular () and parallel ($\blacksquare$) to the substrate at $T=5$ K.

Figure 13

AHE hysteresis loops measured at $T=5$ K and $H_{\mathrm{dc}}$ perpendicular to the substrate for the samples with $t_{\mathrm{Co}}=0.7$ nm and (a) $t_{\mathrm{Pt}}=1.5$ and (b) $6.0$ nm. Inset in (b): details of the first quadrant of the hysteresis loop in (b).

Figure 14

Comparison of the $H_{\mathrm{C}}$ vs $T$ obtained by SQUID () and AHE () measurements with $H_{\mathrm{dc}}$ applied perpendicular to the substrate plane of the sample with $t_{\mathrm{Co}}=0.7$ nm and $t_{\mathrm{Pt}}=1.5$ nm. Dashed line: guide to the eye.

Figure 15

Co $K$-edge XMCD signal measured in the Co-Pt granular film ($▴$) with $t_{\mathrm{Co}}=$ 0.7 nm and $t_{\mathrm{Pt}}=$ 1.5 nm and comparison to that of a Co-Pt $L{1}_0$ alloy () from Ref. [49] and for bare Co NPs () with the same $t_{\mathrm{Co}}$. Solid line corresponds to a fit of the XMCD for the Co-Pt granular film, as a linear combination of the other two curves. Curves are shifted vertically for the sake of clarity.

Figure 16

() Experimental data from Ref. [49] and () calculated XMCD signal at the Co $K$ edge for a Co-Pt $L{1}_0$ alloy. () The calculated XMCD for a $A1$ Co-Pt alloy is also shown for comparison.

Figure 17

Co $L_{2,3}$-edge spectra of the $t_{\mathrm{Co}}=0.7$ nm and $t_{\mathrm{Pt}}=1.5$ nm sample. (a) ($-$) Normalized XAS spectra mean curve measured at normal incidence. () The double-step function used is also indicated. () Integrated white line. (b) Angle dependent normalized XMCD spectra: () normal incidence and () $60{}^{\circ }$. The integrated area of each XMCD signal is also plotted.

Figure 18

() Normalized XANES (a) and () XMCD (b) at the Pt $L_{2,3}$ edges in the Co-Pt sample with $t_{\mathrm{Co}}=$ 0.7 nm and $t_{\mathrm{Pt}}=$ 1.5 nm. () XMCD integrated area. (- -) The XANES spectrum of Au $L_{2,3}$ edges is shown in (a) for comparison. The higher limit for the $p$ and $q$ integrals, used for the sum rules analysis, is given in (b).

Figure 19

Comparison of the magnetic hysteresis loops measured at $T=12$ K with different techniques in the Co-Pt granular sample with $t_{\mathrm{Co}}=$ 0.7 nm and $t_{\mathrm{Pt}}=$ 1.5 nm: () SQUID magnetometry, () AHE, () XMCD at the Pt $L_3$ edge, and () XMCD at the Co $K$ edge.