Physicists usually build entanglement from the ground up, by forcing multiple atoms, ions, or photons into close contact. Now, following up on a 2003 proposal by Anders Sørensen and Klaus Mølmer, researchers at the Max Planck Institute of Quantum Optics in Germany have sculpted their way to entanglement by removing undesired components of the wavefunction that describes two atoms.
Stephan Welte and colleagues loaded a pair of rubidium atoms into a 0.5-mm-long cavity and optically pumped both atoms into a given state—for example, spin down. A laser pulse then rotated the atoms, creating a superposition of up and down states in each atom but still no special relationship between the atoms. To initiate entanglement, the researchers injected linearly polarized photons to bounce around the cavity. (In the illustration below, photons are shown in blue and atoms in red; the inset is of a fluorescence image from a trial.) A measured shift in one photon’s polarization indicated that it had encountered an atom in the up state; a second photon’s polarization shift signaled a brush with an atom in the down state. The photon interventions served to chip away the down–down and up–up components from the wavefunction of the atomic pair. Left behind were wavefunction states in which the rubidium atoms could not be described separately—the atoms were maximally entangled.
Because the wavefunction-carving technique relies on photons rather than interatomic forces, it should work even if the atoms are spaced far apart in a large cavity. Researchers also should be able to entangle two qubits in a larger ensemble of atoms. Welte and his colleagues hope to demonstrate that the technique could be useful in quantum repeaters, which could preserve and relay the entanglement between qubits spaced hundreds of kilometers apart. (S. Welte et al., Phys. Rev. Lett., in press.)
