Scientists at the Department of Energy's Los Alamos National Laboratory (Los Alamos, NM) and the University of Queensland's Centre for Quantum Computer Technology in Australia have made an advance in the quest for a functional quantum computer by exploiting quanta of light or photons in a novel way.
A functional quantum computer could solve certain large mathematical problems and crack secret codes at speeds faster than today's fastest supercomputers. If quantum computers can be built, they can factor large numbers, making them extremely useful for decoding information encrypted by means of currently standard methods.
Los Alamos researchers propose to use quanta of light or photons, the smallest unit of electromagnetic energy, as the basic elements for quantum information processing. Previous proposals based on photons required a crucial ingredient: nonlinear optical elements that allow photons to interact with each other. While such elements have been used for proof-of-principle demonstrations, they suffer from a fatal flaw: they are much too weak to be combined usefully for quantum computation.
Up to now researchers believed that the only feasible option for a photon-based quantum computer was to make the nonlinear elements stronger by several orders of magnitude, which seemed a difficult task. Now Emanuel Knill and Raymond Laflamme of Los Alamos and Gerard Milburn of the University of Queensland have proposed a different approach (see Nature, January 4, 2001). Their idea, which relies on using the high sensitivity of single-photon detection, exploits the detection results to simulate the effects of nonlinear elements. Although this process results in apparently irreversible loss of the "quantumness" of the system, the researchers have demonstrated that this loss is preventable with the use of quantum error correction.
The proposed device reportedly has several advantages over its rivals. One advantage is that it can work at room temperature, which potentially makes these devices as accessible as personal computers. Also, it is based on existing technology: beam splitters, phase shifters, single photon sources and detectors. These, however, need to operate at higher precision than currently available.
"It was widely believed that optics without nonlinear elements is no more powerful than currently available, classical computers," said Knill. "Although the measurements in our scheme irreversibly alter the system, one can still quantum compute usefully. The unwanted effect of measurements can be considered as an error on the system, and as long as both the location and the type of error are known, the system is surprisingly resilient. This discovery is surprising and unexpected, and we think that it provides a useful blueprint for quantum computers. The challenge will be to put our idea into practice."
A three-qubit quantum computer was demonstrated by Knill, Laflamme and their collaborators at Los Alamos in 1998 using nuclear magnetic resonance with trichloroethylene molecules; they built the first seven qubit device in 2000.