![[OSU Physics logo]](/Images/osu-physics.gif)
Silicon - the semiconductor that is the mainstay of the contemporary computer industry - also has many attractive properties for future quantum logical devices. Specifically, both electron and nuclear spins situated in Si can have extremely long lifetimes (t1@1 hour), and are consequently highly immune from decoherence induced by environmental noise. A Si-based quantum computer will require the development of devices capable of performing one and two qubit logical operations, as well as the initialization and measurement of single spin qubits. I will describe a quantum computer architecture in which the nuclear spin qubits are situated on 31P donors in a Si host, located beneath metal gates capable of controlling the positions of the electrons bound to the donors. The hyperfine interaction between the electron and nuclear spins is proportional to the probability that the electron is at the nuclear site, and consequently applied gate voltages can regulate the strength of the interactions between electron and nuclear spins.In the proposed architecture quantum logical operations are effected by a combination of voltages applied selectively to gates above specific sites and globally applied radio frequency magnetic fields, which induce transitions between states of the spin system that are resonant with the applied field. Qubit measurement is performed by first transferring the nuclear spin to an electron and then measuring the electron spin. The Pauli Principle requires that the charge configuration of a two-electron system must depend on whether the electrons are in a mutual singlet or triplet spin state. Thus, sensitive electronic probes of a two-electron system (single electron transistors for example) can be used to infer the spin state of the electrons under appropriate circumstances.I will describe the experimental program currently underway in my laboratory to develop devices to demonstrate the ideas presented above, particularly devices for measuring single electron spins. The most obvious obstacle to the realization of the proposed architecture is the need to deposit P donors into Si in well-defined positions, and I will present possible solutions to this problem. Finally, I will discuss what I consider to be the most promising approaches to the eventual development of a large-scale quantum computer in Si.
For more information, follow this link.