Fig. 1: Material stack, devices, and valley splitting measurements.
From: Atomic fluctuations lifting the energy degeneracy in Si/SiGe quantum dots
a, b High-angle annular dark field scanning transmission electron microscopy (HAADF-STEM) of 28Si/SiGe quantum wells A and B, respectively. c, d Schematic cross-section of a heterostructure with gate layout and false-coloured scanning electron microscope image of a double quantum dot, respectively. Q1 and Q2 are the quantum dots defined through confinement potentials (schematic, grey line) formed below plunger gates P1 and P2. CS is a nearby quantum dot used as a charge sensor. e Typical stability diagram of a double quantum dot formed by plunger gates P1 and P2 and measured by a nearby charge sensor (CS in d). f Close-up of the stability diagram in the few-electron regime. g Typical magnetospectroscopy of the (1,0) → (2,0) transition, used to measure singlet-triplet splittings. An offset of 1082 mV is subtracted for clarity from the gate voltage applied to P2. Black lines show the location of the maximum of the differentiated charge-sensor signal (dISD/dP2) of the electron charging transition. Red lines show a fit to the data, from which we extract the kink position BST. The valley splitting Ev is given by gμBBST, where g = 2 is the gyromagnetic ratio and μB is the Bohr magneton. h Experimental scatter plots of the valley splittings for quantum wells A (magenta) and B (green), with thick and thin horizontal black lines denoting the mean and two-sigma error bars. For quantum well B, the data point EV = 0 μeV indicates that the kink in magnetospectroscopy associated with valley splitting was not observed and, consequently, that the valley splitting is below the lower bound of about 23 μeV set by our experimental measurement conditions (see Supplementary Fig. 6 and Supplementary Table 1).
