Abstract
"Spintronics" is used to describes electronics based on control over the spin of the electron rather than the charge; instead of having the charge as the carrier of information, the electrons spin-state should be the target for control and detection in a given device. The spin qubit, one of the contenders in the race to build a large-scale quantum computer, is such a component, and research aiming to build, manipulate and couple spin qubits is looking at many materials systems to nd one where the requirements for fast control and long coherence time can be combined with ecient coupling between distant qubits. This thesis presents electric measurement on two of the materials systems currently at the forefront of the spin qubit race, namely InAs nanowires and GaAs/AlGaAs heterostructures.
For the InAs nanowires we investigate dierent gating geometries towards the goal of dening stable quantum dots in the nanowire at mK temperatures. Three types of local gates are analyzed; narrow gates (50100nm) located on top of or below the nanowire, and wide gates overlapping the interfaces between nanowire and source and drain electrodes. We nd that applying negative potentials to the local gate electrodes induces tunable barriers of up to 0:25 eV. From the temperature dependence of the conductance, the barrier height is extracted and mapped as a function of gate voltage. Top and bottom gates are similar to each other in terms of electrostatic couplings (lever arms 0:10:2 eV=V) and threshold voltages for barrier induction (Vg 􀀀1 to 􀀀2V), but low temperature gate sweeps suggest that device stability could be affected by the differences in device processing for the two gate geometries.
For the GaAs heterostructure we investigate two new ideas for realizing spin qubits in lateral quantum dots.
First, we incorporate ferromagnetic metal in the depletion gates making them double as micro-magnets supplying magnetic eld gradients allowing spin qubit operation. We demonstrate full tunability of the electron occupation with the magnetic gate structure, combined with a magnetic eld gradient of 25mT determined by electric dipole spin resonance (EDSR). The EDSR mechanism could not be determined with certainty. For the same device, we investigate using single, unpaired electron spins in multi-electron quantum dots for spin qubits; qubit coherence in this regime could benefit from screening effects. We find that simple, alternating spin filling is not followed. Furthermore, measurement of the exchange splitting, J, indicate two magnetic field dependent transitions lifting spin blockade which is likewise inconsistent with the simplest model for spin filling. The effect of the magnetic field gradients from the micro-magnet could play a role in the observed differences between the multi- and the few-electron double dots.
For the InAs nanowires we investigate dierent gating geometries towards the goal of dening stable quantum dots in the nanowire at mK temperatures. Three types of local gates are analyzed; narrow gates (50100nm) located on top of or below the nanowire, and wide gates overlapping the interfaces between nanowire and source and drain electrodes. We nd that applying negative potentials to the local gate electrodes induces tunable barriers of up to 0:25 eV. From the temperature dependence of the conductance, the barrier height is extracted and mapped as a function of gate voltage. Top and bottom gates are similar to each other in terms of electrostatic couplings (lever arms 0:10:2 eV=V) and threshold voltages for barrier induction (Vg 􀀀1 to 􀀀2V), but low temperature gate sweeps suggest that device stability could be affected by the differences in device processing for the two gate geometries.
For the GaAs heterostructure we investigate two new ideas for realizing spin qubits in lateral quantum dots.
First, we incorporate ferromagnetic metal in the depletion gates making them double as micro-magnets supplying magnetic eld gradients allowing spin qubit operation. We demonstrate full tunability of the electron occupation with the magnetic gate structure, combined with a magnetic eld gradient of 25mT determined by electric dipole spin resonance (EDSR). The EDSR mechanism could not be determined with certainty. For the same device, we investigate using single, unpaired electron spins in multi-electron quantum dots for spin qubits; qubit coherence in this regime could benefit from screening effects. We find that simple, alternating spin filling is not followed. Furthermore, measurement of the exchange splitting, J, indicate two magnetic field dependent transitions lifting spin blockade which is likewise inconsistent with the simplest model for spin filling. The effect of the magnetic field gradients from the micro-magnet could play a role in the observed differences between the multi- and the few-electron double dots.
| Originalsprog | Engelsk |
|---|
| Forlag | The Niels Bohr Institute, Faculty of Science, University of Copenhagen |
|---|---|
| Status | Udgivet - 2012 |