Preparing high purity entangled states for NMR quantum computing
by
Hilary Carteret
LITQ, DIRO, Universite de Montreal, Montreal, Quebec
Coauthors: M.S. Anwar, D. Blazina, S.B. Duckett, T.K. Halstead, J.A. Jones, C.M. Kozac and R.J.K. Taylor

A number of objections have been raised against the liquid-phase nuclear magnetic resonance (NMR) implementation of a quantum computer. In 1997 Warren argued that efficient NMR quantum computation required temperatures far below 1K. In 1998, Braunstein et al. claimed that liquid-phase NMR is incapable of universal quantum computation because it cannot produce entanglement, and thus the computer could be described by local hidden variable models.

This presentation will give an introduction to para-hydrogen induced polarization (PHIP) and show how this can be used to prepare a two-qubit system in an almost pure, entangled state. To achieve polarizations comparable to our observed value of 0.916 +/- 0.019 (out of a maximum possible value of 1) by thermal means would require cooling the system to 6.4mK in the 9.6 T field used, or a magnetic field of 0.45 MT at room temperature. Our states had an entanglement of formation of 0.822.

This method clearly outperforms other liquid-phase initialization methods for producing both polarization and entanglement, regardless of the arguments over whether or not entanglement is really necessary for universal quantum computation.

Date received: April 1, 2004