Extended Data Fig. 2: Magnetoresistance and hysteresis at additional filling factors and with perpendicular magnetic field.

a, Lineslice of longitudinal resistance Rxx as a function of bottom gate voltage VBG (the top axis indicates the filling factor v) in device A. The dashed lines identify the dopings at which the following magnetoresistance measurements are done. The triangle, square, and circle represent a hole-doped bias, a charge neutrality point (CNP) bias, and an electron-doped bias respectively. b, c, The longitudinal resistance Rxx and magnetoresistance (MR = [Rxx(B) − Rxx(B = 0)]/Rxx(B = 0)) as a function of in-plane magnetic field B∥ at dopings identified by the symbols on the upper right-hand corner. The hysteresis and the MR is relatively less prominent at the CNP (compare ≃ 5% MR in c to that of ≃ 40% MR at B∥ = 0.5 T in b). d, e, Rxx and MR as a function of perpendicular magnetic field B⊥ at dopings identified by the symbols on the upper right-hand corner. The hysteresis with B⊥ is not as striking as with B∥ however, it is present in subpanel d. The arrows indicate the direction of the magnetic field sweep. This serves as a direct proof that indeed the superconducting phase is surrounded by phases that have magnetic ordering. We further present Hall resistance hysteresis data in the Supplementary Section XIV as additional evidence of broken time-reversal symmetry due to proximal magnetic order.