Research teams from Harvard University in Cambridge, Mass., and from Georgia Institute of Technology in Atlanta have independently developed quantum memories, clouds of atoms that store single photons and allow them to be retrieved on demand. The groups connected two such memories via an optical fiber, generated a single photon at one site and transferred that photon to the other. They then retrieved the photon and verified that its quantum properties were preserved after storing it for a time at the second site.Researchers have demonstrated a primitive quantum memory and a simple quantum network. A photon generated at the dense cloud of rubidium atoms at site A traveled to the cloud at site B over a connecting optical fiber. The photon was converted into an atomic excitation, stored for a time and then retrieved with its characteristics preserved. ©2005 Nature Publishing Group.Matthew D. Eisaman, a graduate student at Harvard, noted that the capabilities demonstrated by these primitive quantum components could prove fundamental to future applications. “Quantum networks will most likely be an integral part of any future realizations of quantum computing or quantum communications,” he said.Securing communicationsCurrent quantum information processing involves single photons. They are produced in a particular quantum state, such as a given spin, at one location. They then travel down a fiber to another location, where they are interrogated. Such methods are finding application in quantum key distribution, a technique that renders communications secure by ensuring that cryptographic keys cannot be intercepted by a third party without detection.However, quantum information is fragile and can be lost because of environmental influences, limiting how far photons can travel. What is needed is a so-called quantum repeater that would swap quantum states along shorter segments of the optical path and thereby extend the distance over which communication fidelity could be preserved.The new research illustrates how the memory for this could be constructed using electromagnetically induced transparency. In exploiting this effect, the investigators fired a laser into a cloud of rubidium atoms to create a single photon. A control beam rendered a second cloud transparent and allowed the photon from the first to enter the dense ensemble of atoms after traveling over a connecting fiber. When the control beam was turned off, the second cloud became opaque and trapped the photon. Sometime later, the researchers turned the beam back on and freed the photon. Tests by the investigators proved that photons stored and retrieved in this way retained their quantum properties.The Harvard group used a room-temperature cloud, while the Georgia Tech group used atoms at near absolute zero. Achieving an ultracold cloud involved magneto-optic traps and was technically more complex than the room-temperature approach, which might make it harder to scale up to a full, multinode quantum network. On the other hand, noted Alex Kuzmich, an assistant physics professor at Georgia Tech, cold atoms work better in this application. He cited this as one reason for his group’s ability to verify the preservation of both quantum and particlelike properties of stored photons.Although encouraging, these results highlight some problems. The photon generation rate, at a few seconds per photon, is currently too low, and the storage time of 11 μs or so is too short for practical applications. Both of these issues are the target of ongoing research.The next step involves the entanglement of two remote atomic quantum bits, or qubits, which a team at California Institute of Technology in Pasadena and at Bell Labs in Murray Hill, N.J., demonstrated for a distance of 2.8 m (see “Entanglement Demonstrated Between Remote Ensembles,” Photonics Spectra, January). The Georgia Tech researchers subsequently have accomplished this for more than 5 m.However, Kuzmich knows that much more work must be done before quantum memories are really useful. “The ultimate goal is the quantum repeater, but that will require further fundamental scientific and technological breakthroughs,” he said.