Review article: Quantum computing based on semiconductor nanowires
Topological phases: Solid state Majorana fermions
Majorana fermions have the unique property that they are equal to their own anti-particles. Such fermions were heavily looked for but never observed among elementary particles. Recently, Majorana fermions were proposed theoretically as quasiparticles in solid state systems with a topological phase. I am working on Majorana fermions in semiconductor nanowires combined with superconductors. Majorana fermions are important in the context of topological quantum computing, which promises decoherence-free manipulation of quantum information.
Quantum information: Qubits in semiconductor nanowires
Semiconductor nanowires are flexible building blocks for nanoelectronics. It has recently become possible to confine single electrons in nanowires using electric fields from nearby gates (see figure). These electrons are natural quantum bits. We operate qubits in InAs and InSb nanowires with the qubit basis being the two spin-orbital states of single electrons. These states are the hybrids of spin and motion, and we take advantage of both. We can store information in the spin state, but manipulate the qubit using the orbital part of the wavefunction, by simply oscillating the electron back and forth inside the nanowire. Spin-orbit qubits are controlled by electric fields, which is the preferred method of control for nanostructures. Our qubits are individually addressable due to gate-tunable g-factors. We perform universal and fast coherent control and apply dynamical decoupling techniques to extend coherent evolution times.
Spintronics: Spin currents in two-dimensional electron gases
At the University of British Columbia we have generated and detected pure spin currents using quantum point contacts. Quantum point contacts are narrow gate-defined constrictions in the two-dimensional electron gas. At high magnetic fields they transmit only one spin orientation, which makes them perfect spin injectors and detectors (see figure for our spin current signal). In long micron-wide channels we have achieved spin relaxation length exceeding 100 microns. This makes pure spin currents in GaAs two-dimensional electron gases attractive both for spintronic technology and as a probe of spin polarization and dynamics in mesoscopic systems.
We have used spin currents to observe a new type of spin resonance due to ballistic bouncing of electrons. This ballistic spin resonance is mediated by spin-orbit interaction and does not require time-dependent driving fields. We have then used it to measure the anisotropy of spin-orbit interaction in a GaAs 2D electron gas. Furthermore, we have demonstrated a spintronic device, field effect spin switch, in which spin relaxation length is controlled by gate voltage. In the future the same principle can be used to achieve coherent spin rotations using only static magnetic fields.
Bouncing spins by Lieven Vandersypen
During my PhD research at the University of Illinois at Urbana-Champaign I studied superconductor-ferromagnet-superconductor Josephson π-junctions. Josephson effect is the flow of supercurrent through a thin layer of non-superconducting material placed between two superconductors. This supercurrent depends on the phase difference between the two superconducting wavefunctions. While typically the phase difference across a Josephson junction is zero when there is no supercurrent flowing, π-junctions offer an interesting twist: the superconducting wavefunction prefers to have a phase shift of π.
Because of fluxoid quantization a π-junction causes a spontaneous current if embedded in a superconducting loop. The left and right circulating spontaneous currents are energetically degenerate, which may be useful for flux qubits in the future. We mapped out the transition into a π-junction state and imaged spontaneous currents in the arrays of π-junction loops using a scanning SQUID microscope (see figure). The mechanisms of the π-junction state are abundant and illuminate the electronic and spin interactions that underlie superconductivity. In Urbana we have used the exchange interaction in the ferromagnetic barrier to induce oscillations of the Cooper pair wavefunction.