Figure 4: SQUID oscillation model and high-temperature behaviour.
From: Evidence for an anomalous current–phase relation in topological insulator Josephson junctions
(a) Observed (red squares) SQUID oscillations for VTG=−18 V and 20 mK, compared with theoretical (black) curve based on a toy model CPR. (b) Theoretical CPR used to derive the theoretical SQUID oscillations in a. We add the q=0 and q=0.1 modes (in normalized units in which the velocity υ and energy gap Δ0 are set equal to 1) to a purely sinusoidal CPR (shown in blue). The resulting composite CPR is shown in black, with the contribution from the q=0 and q=0.1 modes shown in red. Note that the subtle nonsinusoidal behaviour leads to detectable levels of node-lifting in SQUID oscillations while possibly being obscured in direct measurement of CPR. The theoretical SQUID oscillations were rescaled by sin(πB/B0)/(πB/B0) to mimic Fraunhofer-like decay of critical current due to single-junction diffraction. (c) Critical current versus magnetic field for 20 mK (black), 400 mK (red), and 800 mK (blue) at VTG=−18 V, showing a discernible change in SQUID modulation depth. For each temperature, the critical current is normalized by the value at B=0.146 mT. As the temperature increases, the current at the SQUID nodes decreases, indicating that the CPR is reverting to a conventional form (that is, sinusoidal). (d) Comparison of Ic versus B traces at VTG=−18 V for 20 mK (black) and 800 mK (red). Inset shows the same data as the 800 mK in detail to show the diffraction pattern nodes (blue arrows).
