Enhancement of Josephson Critical Currents in Ferromagnetic ${\mathrm{Co}}_{40}$${\mathrm{Fe}}_{40}$${\mathrm{B}}_{20}$ by Thermal Annealing

The electrical and structural properties of ${\mathrm{Co}}_{40}{\mathrm{Fe}}_{40}{\mathrm{B}}_{20}$ ($\mathrm{Co}\text{−}\mathrm{Fe}\text{−}\mathrm{B}$) are tunable by thermal annealing. This is key to the optimization of $\mathrm{Co}\text{−}\mathrm{Fe}\text{−}\mathrm{B}$-based spintronic devices, where the advantageously low magnetic coercivity, high spin polarization, and controllable magnetocrystalline anisotropy are utilized. Here, we report $\mathrm{Nb}/\mathrm{Co}\text{−}\mathrm{Fe}\text{−}\mathrm{B}/\mathrm{Nb}$ Josephson devices and demonstrate an enhancement of the critical current by up to 700% following thermal annealing due to increased structural ordering of the $\mathrm{Co}\text{−}\mathrm{Fe}\text{−}\mathrm{B}$. The results demonstrate that $\mathrm{Co}\text{−}\mathrm{Fe}\text{−}\mathrm{B}$ is a promising material for superconducting quantum spintronic devices.

Figure 1

(a) J_c(H) pattern for a $\mathrm{Nb}$(300 nm)/$\mathrm{Co}\text{−}\mathrm{Fe}\text{−}\mathrm{B}($3.5 nm)/$\mathrm{Nb}$(300 nm) nanopillar before (solid lines) and after (dashed lines) thermal annealing at 400 °C for 30 min. The red (solid and dashed) line shows J_c with increasing H and the blue (solid and dashed) line shows J_c with decreasing H. The inset shows a schematic diagram of a nanopillar. (b) In-plane magnetic hysteresis (δ) for $\mathrm{Nb}$(300 nm)/$\mathrm{Co}\text{−}\mathrm{Fe}\text{−}\mathrm{B}($d${}_F$)/$\mathrm{Nb}$(300 nm) nanopillars before (black diamonds) and after (red circles) thermal annealing. The vertical error bars represent the statistical scatter of δ for multiple nanopillars on the same circuit. The black line is a least-squares regression-line fit for the nanopillars before thermal annealing, giving a volume saturation magnetization of (643 ± 21) emu cm⁻³ and a magnetically dead layer thickness at each $\mathrm{Nb}/\mathrm{Co}\text{−}\mathrm{Fe}\text{−}\mathrm{B}$ interface of (0.33 ± 0.09) nm. (c) An estimate of the coercive field (H_c) of the nanopillars versus d${}_F$ and (d) the magnetic field periodicity (n) of I_c(H) versus d${}_F$ before (black diamonds) and after (red circles) thermal annealing. All data at 1.6 K.

Figure 2

(a) AR_N versus d${}_F$ before (black diamonds) and after (red circles) thermal annealing. The black line shows a least-squares regression-line fit for the nanopillars before annealing giving ${\rho }_F$ ≈ (88 ± 46) µΩ cm and 2AR${}_{\mathrm{Nb}/\mathrm{Co}\text{−}\mathrm{Fe}\text{−}\mathrm{B}}$ = (4.4 ± 1.4) fΩ m². For the annealed nanopillars, ${\rho }_F$ ≈ (65 ± 57) µΩ cm and 2AR${}_{\mathrm{Nb}/\mathrm{Co}\text{−}\mathrm{Fe}\text{−}\mathrm{B}}$ = (2.1 ± 1.7) fΩ m² from a least-squares regression-line fit (red line). (b) J_c versus d_F before (black diamonds) and after thermal annealing (red circles) at 400 °C for 30 min. The inset shows the relative change in J_c following thermal annealing [(J_c,annealed− J_c,as-grown)/J_c,as-grown] versus d${}_F$. (c) I_cR_N versus d${}_F$ where the dashed lines are least-squares regression-line fits giving a coherence length in $\mathrm{Co}\text{−}\mathrm{Fe}\text{−}\mathrm{B}$ of ${\xi }_F$ = (0.93 ± 0.02) nm and (1.00 ± 0.04) nm, for the nanopillars before and after thermal annealing, respectively. The inset shows the relative change in I_cR_N following thermal annealing [(I_cR_N,annealed− I_cR_N,as-grown)/I_cR_N,as-grown] versus d${}_F$. The error bars in AR_N, J_c, and I_cR_N represent the statistical scatter for multiple nanopillars. All data at 1.6 K.