Realization of efficient tuning of the Fermi level in iron-based ferrimagnetic alloys

The Stoner criterion allows only three single elements possessing room-temperature (RT) ferromagnetism: cobalt (Co), nickel (Ni), and iron (Fe). Although these three elements have played central roles in magnetism-based materials, their large work function ($4.5\sim 5.2\mathrm{eV}$) is becoming a non-negligible obstacle for realization of spin devices using nonmetallic materials with finite energy gaps, because injection of electron spins into these nonmetallic materials is strongly hampered due to the large Schottky barrier height. Hence, a novel ferromagnetic or ferrimagnetic material simultaneously possessing RT ferromagnetism or ferrimagnetism and high Fermi energy is strongly required. Here, we show that an Fe-based alloy, iron-gadolinium (FeGd), allows circumvention of the obstacle. Surprisingly, only 20% of Gd incorporation in Fe dramatically modulates the Fermi energy from −4.8 to −3.0 eV, which is the largest modulation in all metallic alloys reported thus far. The coexistence of ferrimagnetism and nonzero spin polarization at RT of FeGd supports its abundant potential for future applications in low-carrier-density materials such as monolayer, organic, and nondegenerate inorganic semiconductors.

Figure 1

(a) Results of the UPS measurements. The $x$-axis represents the energy of the electron on the basis of the vacuum level, and the $y$-axis shows the intensity in units of counts per second (CPS). The UPS spectra of ${\mathrm{Fe}}_{100}{\mathrm{Gd}}_0, {\mathrm{Fe}}_{83}{\mathrm{Gd}}_{17}, {\mathrm{Fe}}_{80}{\mathrm{Gd}}_{20}$, and ${\mathrm{Fe}}_0{\mathrm{Gd}}_{100}$ samples were obtained from the single-composition samples, whereas the other spectra were obtained sequentially from the gradated-composition sample. The red dashed lines represent the Fermi energy estimated by fitting using convolution of the Fermi–Dirac function and the Gaussian function. The blue arrows indicate the UPS peaks that are identical to the energy of the Fe $3d$ band. The green and red arrows indicate the UPS peaks that are identical to the energy of the Gd $4f$ band. (b) Comparison of the composition dependence of the WF among ferromagnetic/ferrimagnetic alloys. The $x$-axis shows the surface composition, and the $y$-axis shows the measured WFs. The green and blue plots represent the WF values of ${\mathrm{Ni}}_{100\text{−}x}{\mathrm{Fe}}_x$ and ${\mathrm{Co}}_{100\text{−}x}{\mathrm{Fe}}_x$. The pink plots represent the WF values of ${\mathrm{Fe}}_{100\text{−}x}{\mathrm{Gd}}_x$. The filled (open) circles represent the results of the gradated (single)-composition sample. (c) Comparison of the WF modulation efficiency η based on the composition of binary alloys for a variety of alloys. η is the change in the WF due to adding a low WF element of 1%.

Figure 2

Magnetization curves of ${\mathrm{Fe}}_{100\text{−}x}{\mathrm{Gd}}_x$ films measured at RT by using a vibrating sample magnetometer for (a) Gd, (b) ${\mathrm{Fe}}_{79}{\mathrm{Gd}}_{21}$, (c) ${\mathrm{Fe}}_{86}{\mathrm{Gd}}_{14}$, and (d) Fe single-composition samples. An in-plane magnetic field was applied to the sample. (e) Saturation magnetization $M_{\mathrm{s}}$ obtained by magnetization curves as a function of $x$. The pink area is the magnetic compensation area reported in Refs. [39, 40]. (f) Temperature evolution of magnetization, $M$ at $B_x=1\mathrm{T}$ normalized by $M$ at 300 K for ${\mathrm{Fe}}_{100\text{−}x}{\mathrm{Gd}}_x$ single-composition samples. The inset shows the Curie temperature $T_{\mathrm{c}}$ as a function of $x$.

Figure 3

(a) A schematic of the device structure for the measurements of resistivity and anomalous Hall effect. (b) Temperature dependences of resistivity ${\rho }_{xx}$, for ${\mathrm{Fe}}_{100\text{−}x}{\mathrm{Gd}}_x$ single-composition samples with various $x$. (c) Temperature dependences of ${\rho }_{xx}$ normalized by ${\rho }_{xx}$ at 300 K.

Figure 4

[(a)–(d)] Results of anomalous Hall measurements of the Hall bar-shaped ${\mathrm{Fe}}_{100\text{−}x}{\mathrm{Gd}}_x$ single-composition samples at various temperatures. The Hall voltage was measured under the application of an out-of-plane magnetic field. A dc charge current of 0.1 mA was applied. The device geometry and current–voltage configuration are shown in Fig. 3.

Figure 5

[(a)–(d)] Semi-log plots of the current-rectifying characteristics for the ${\mathrm{Fe}}_{100\text{−}x}{\mathrm{Gd}}_x$ single-composition samples measured at 300 K. A gold (Au) antimony (Sb) ohmic contact was fabricated at the backside of the $n$-Si substrate. The device structure is shown in the inset of (a). The size of the ${\mathrm{Fe}}_{100\text{−}x}{\mathrm{Gd}}_x$ electrodes was $0.29\mathrm{m}{\mathrm{m}}^2$.