Tuning of spin relaxation and the Kondo effect in copper thin films by ionic gating

Spin relaxation length is a fundamental material parameter that influences all aspects of spin dependent transport. The ability to tune the spin relaxation length leads to novel spintronic phenomena and functionalities. We explore the tunability of the spin relaxation length in the mesoscopic Cu channels of nonlocal spin valves by using the ionic gating technique via a ${\mathrm{Li}}^{+}$ containing solid polymer electrolyte. At 5 K, the Cu spin relaxation length ${\lambda }_{\mathrm{Cu}}$ is tuned reversibly between 670 and 410 nm and the Cu resistivity ${\rho }_{\mathrm{Cu}}$ is tuned by 9%. The strength of the Kondo effect due to the Fe impurities in Cu is tuned by one order of magnitude. At 295 K, ${\lambda }_{\mathrm{Cu}}$ is tuned between 380 and 300 nm and ${\rho }_{\mathrm{Cu}}$ is tuned by 7%. A gradual amplification of the tuning ranges by repeated gate cycling is observed and clearly suggests an electrochemical origin. Tunable spin relaxation in simple metals enriches functionalities in metal-based spintronics and shines light on fundamental spin relaxation mechanisms.

Figure 1

(a) A cartoon illustration of ionic gating on a NLSV. (b) The SEM image of a NLSV with a side gate. $R_s$ versus $B$ curves of the NLSV device with $L=800\mathrm{nm}$ at 5 K after cooling the device from 295 K under $V_g$ of (c) +1.2 V, (b) 0 V, and (c) −1.5 V.

Figure 2

(a) The spin signal and (b) the Cu resistivity as a function of $T$ as the NLSV is cooled under various $V_g$. Solid lines in (b) are fits by using the Bloch-Grüneisen term for phonon resistivity. (c) Derived ${\lambda }_{\mathrm{Cu}}$ as functions of $T$ under various $V_g$ values.

Figure 3

$R_s$ versus $B$ curves for a NLSV with an Al channel under $V_g$ of (a) +3.5 V and (b) −3.5 V at 295 K. $R_s$ versus $B$ curves for the Cu-based device under $V_g$ of (c) +1.2 V and (d) −1.5 V at 295 K.

Figure 4

(a) The plot of ${\rho }_{\mathrm{Cu}}\left(T\right)-{\rho }_{\mathrm{Cu}}\left(5\mathrm{K}\right)$ versus $T$ between 5 and 22 K for the Cu channel under various $V_g$ values. (b) The spin relaxation rate ${\tau }_s^{-1}$ versus $T$ between 5 and 250 K under various $V_g$ values. All solid lines are fitting lines.

Figure 5

(a) The measured $\mathrm{\Delta }{R}_s$ (left vertical axis) and applied $V_g$ (right vertical axis) as functions of gating time $t_g$ at 295 K. (b) The measured ${\rho }_{\mathrm{Cu}}$ (left vertical axis) and applied $V_g$ (right vertical axis) as a function of $t_g$ at 295 K.

Figure 6

(a) The tuning range of spin signals $\mathrm{\Delta }{R}_s^{+}-\mathrm{\Delta }{R}_s^{-}$ between positive and negative polarities of $V_g$ for each complete gate cycle as a function of $t_g$ for the NSLV at 295 K. Different symbols indicate the range of applied $V_g$. (b) Hysteresis loop of $\mathrm{\Delta }{R}_s$ versus $V_g$ around $t_g=12\mathrm{h}$. (c) Hysteresis loop of $\mathrm{\Delta }{R}_s$ versus $V_g$ around $t_g=31\mathrm{h}$.