The two recent breakthroughs exploit the quantum properties of nuclear spins, which can point "up" or "down" relative to an applied electric or magnetic field. A team of physicists from IBM, Stanford University and MIT in the US used nuclear magnetic resonance to manipulate the spins of hydrogen and carbon-13 nuclei in a chloroform (CHCl3) molecule (Nature 393 143). Using a sequence of electromagnetic pulses they were able to perform a quantum algorithm to determine whether an unknown mathematical function was constant or balanced. However, it will be difficult to scale up this approach to a working quantum computer.

A more realistic approach would be to design a silicon-based quantum computer - and this is what Bruce Kane of the University of New South Wales in Sydney, Australia, has proposed (Nature 393 133). The "quantum bits" in Kane's proposal are isolated phosphorous ions in a silicon crystal. The state of the spins are set by voltages applied to metal contact gates, and their interactions are controlled by other metal gates. Kane has also devised ways for getting all of the spins to point in the same direction at the start of a calculation, and for reading out the results at the end - two of the biggest challenges in quantum computation. Kane admits that building such a quantum computer will involve "substantial challenges", but he points out that the electronics industry is already working on similar problems as it prepares for the next generation of conventional chips. The main technological challenges are to find ways to accurately position the phosphorus ions and to reduce defects in the device.