I arrived in Oxford in 1985 to read chemistry at Corpus Christi College and then moved to St Johns College for my DPhil. studies in magnetic resonance. I moved to Merton College as a Junior Research Fellow working on applications of magnetic resonance in biochemistry, and spent a year at the University of California at Berkeley studying fundamental topics in magnetic resonance. On my return to Oxford I became interested in quantum information theory and moved to the department of physics, returning to Corpus Christi College as a research fellow. In 2002 I was appointed as University Lecturer and Tutorial Fellow in Physics at Brasenose.
As organising tutor I oversee all physics teaching in Brasenose, and personally teach about a quarter of the course, including mechanics, relativity, quantum mechanics, atomic physics, and quantum information theory.
Quantum Information Processing and Nuclear Magnetic Resonance.
Quantum information processing is the combination of quantum mechanics (the underlying physical theory describing the universe at the atomic scale) and classical information theory (the basic theory behind computers and telecommunication systems) to produce entirely new types of information processing techniques and devices. The most famous example is quantum computers, which could in principle solve certain problems almost infinitely faster than current classical devices.
Although the theory of QIP is fairly well understood, actually building useful quantum computers is proving very difficult. Liquid state Nuclear Magnetic Resonance (NMR) provides an ideal test bed for building small devices and studying how they work, and we have been following this route for about ten years.
Early research concentrated on implementing simple quantum algorithms, but subsequently we have concentrated on more fundamental questions: firstly, how one can perform accurate quantum logic with imperfect equipment; secondly, how one can prepare NMR quantum computers in pure initial states without cooling them down to very low temperatures; and thirdly, how one can implement quantum logic gates without the need to selectively address individual nuclei. More recently we have begun to consider the use of electron spin resonance to perform similar experiments with electron spins instead of nuclear spins. Although significantly more challenging this is more likely to lead to useful devices.