${\mathrm{C}\mathrm{o}}_{1-x}{\mathrm{F}\mathrm{e}}_x{\mathrm{S}}_2$: A Tunable Source of Highly Spin-Polarized Electrons

In the emerging field of spin-electronics ideal ferromagnetic electron sources would not only possess a high degree of spin polarization, but would also offer control over the magnitude of this polarization. We demonstrate here that a simple scheme can be utilized to control both the magnitude and the sign of the spin polarization of ferromagnetic ${\mathrm{C}\mathrm{o}\mathrm{S}}_2$, which we probe with a variety of techniques. The position of the Fermi level is fine-tuned by solid solution alloying with the isostructural diamagnetic semiconductor ${\mathrm{F}\mathrm{e}\mathrm{S}}_2$, leading to tunable spin polarization of up to 85%.

Figure 1

(a) Schematic depiction of the basic concept of composition control of $E_F$ and $P$ in ${\mathrm{C}\mathrm{o}}_{1-x}{\mathrm{F}\mathrm{e}}_x{\mathrm{S}}_2$. (b) Calculated DOS for $x=0.00$, 0.125, and 0.25. The value of $P$ in each case is shown underneath the figure. $E_F$ is taken as the zero point of energy.

Figure 2

$x$ dependence of the experimental lattice parameter. Solid points are data, while the blue or dark grey dotted line is Vegard's law. Lower inset: X-ray powder diffraction from a representative sample ($x=0.10$). Upper inset: Crystal structure of ${\mathrm{C}\mathrm{o}}_{1-x}{\mathrm{F}\mathrm{e}}_x{\mathrm{S}}_2$. Blue or dark grey circles are the Co(Fe) atoms; red or light grey circles represent the S atoms.

Figure 3

$x$ dependence of (a) the theoretical $P$, (b) the theoretical $M_S$, (c) the experimental $M_S$ (assuming no moment on Fe), (d) the high field (9 T) MR at $\approx 0.5T_C$, and (e) the experimental AMR at 4.2 K. AMR is defined as $({\rho }_{\parallel }-{\rho }_{\perp })/\rho (0)$, where ${\rho }_{\parallel }$ is the resistivity measured with field parallel to current, ${\rho }_{\perp }$ is the resistivity measured with field perpendicular, and $\rho (0)$ is the zero field value.

Figure 4

(a) 4.2 K conductance (normalized to the normal state), as a function of voltage (normalized to the superconducting gap), for $x=0.00$, 0.08, 0.15, and 0.30. Points represent experimental data, while the solid lines are fits as described in the text. Fitting parameters are $x=0.00$, $P=57\%$, $Z=0.00$; $x=0.08$, $P=56\%$, $Z=0.09$; $x=0.15$, $P=80\%$, $Z=0.28$; and $x=0.30$, $P=73\%$, $Z=0.09$. Inset: $x$ dependence of (b) the magnitude of the theoretical $P$, and (c) the experimentally determined $P$.