The successful implementation of a scalable, fault-tolerant quantum computer would introduce a type of information processing more powerful than any available today. Reciprocally, the discovery of a fundamental obstacle to such a system would be an important advance in the foundations of quantum theory. No such fundamental obstacles are currently known, but neither has any architecture been shown to be experimentally scalable
Many technologies have been considered for finding such an architecture; in this work I focus on nuclear spins in semiconductors. Semiconductors provide promising optical means for polarizing and measuring small nuclear spin ensembles, which are tasks that pose critical challenges to quantum computers based on nuclear magnetic resonance (NMR). At the same time, semiconductor nuclei are sufficiently coherent quantum oscillators to allow complex information processing using resonant radio-frequency pulse sequences. In particular, the isotopically clean and magnetically quiet environment of pure, high quality, bulk single-crystal silicon provides a nuclear environment allowing what may be the longest absolute coherence time of any solid-state qubit currently under consideration. I have experimentally tested this claim using high-power NMR pulse sequences to eliminate inhomogeneous dephasing and dipolar evolution among an ensemble of 29Si nuclei in isotopically modified silicon crystals. Intrinsic decoherence processes are only observed in polycrystalline silicon, where 1/f charging noise processes are likely to blame. In high-quality single crystal samples, nuclear coherence persists for over 25 seconds, a timescale limited only by pulse sequence imperfections
I will discuss an architecture that takes advantage of this clean nuclear environment, but I will also address its scalability limitations due to silicon's poor optical characteristics. These limitations will suggest new experiments employing nuclear spins in optically controlled semiconductor quantum dots, which may hold more promise for future scalable quantum computer architectures
