Spin states of a three-electron system can encode one quantum bit in such a way that universal qubit control by voltage pulses becomes possible. In this work, a triple-dot qubit is operated in a highly symmetric charge configuration, using resonant radio-frequency voltage pulses applied to one gate electrode. The authors show that careful symmetrization reduces the susceptibility of the qubit frequency to voltage fluctuations by more than an order of magnitude, compared to conventional triple-dot spin qubits. Combining voltage control with symmetric operating regimes may lead to multiqubit processors based on semiconducting spin qubits that at the same time alleviate decoherence due to electrical noise.
A quantum computer would exploit the phenomena of quantum superposition and entanglement in its functioning and with them offer pathways to solving problems that are too hard or complex to even the best classical computers built today. The implementation of a large-scale working quantum computer could bring about a change in our society rivaling the one started by the digital computer. However, the field is still in its infancy and there are many theoretical and practical issues needing to be solved before large-scale quantum computing can become reality. In digital computers, data is stored in bits. The quantum equivalent is called a qubit (quantum bit) and it is basically a quantum mechanical two-level system that can be in a superposition of its two basis states. There are many different proposals for implementing qubits, but one of the most promising ones is to encode the qubit using electron spins trapped in semiconductor quantum dots. Singlet-triplet qubits are spin qubits where the two-electron spin eigenstates are used as the qubit's basis. The required one and two-qubit operations have already been demonstrated experimentally by several research groups around the world in this qubit architecture. The most severe factor limiting the implementation of larger systems of qubits is decoherence. The qubits are not isolated systems, they interact with their environment, which can lead to the loss of quantum information. Few-electron systems can be simulated accurately using first principle methods that become too taxing when the particle number increases. The topic of this thesis is the simulation of quantum dot singlet-triplet qubit systems using accurate exact diagonalization based methods. The emphasis is on the realistic description of qubit operations, both single-qubit ones and those involving the interaction between neighboring qubits. The decoherence effects are also discussed alongside with certain proposals to alleviate their effects.
In Part I of this article, we introduced the notion of encoding binary bit information in the spin polarization of a single electron confined in a quantum dot and placed in a magnetic field. The spin polarization becomes bistable in a magnetic field and the spin can point either parallel to the field or antiparallel to it. These two polarizations encode the classical binary bits 0 and 1. We showed that since the bits can be switched by simply flipping the electron's spin without physically moving the electron in space and causing a current, the energy dissipated during the bit flip operation could be made very small. In Part II we discuss the notion of using a single electron's spin as a quantum bit (or qubit) that represents information in quantum computers. Just as spin is superior to charge as a vehicle to host classical bit information in classical computers it turns out that spin is superior to charge to host quantum information as well. That is why spin-based quantum information processing is gradually becoming the staple of scalable solid state versions of quantum computers.