Controlling EPR and Bell correlations in BECs (CEBBEC)

The key element for most of quantum technologies is entanglement, a kind of correlation which only exists in quantum mechanics. For instance, in quantum sensing entanglement improves the sensitivity of interferometric measurements: the relative phase of two interfering waves is measured more precisely when the waves are initially entangled. This method is applied for example for photonic waves on the LIGO gravitational wave interferometer. In the CEBBEC project we aim at generating entanglement between two spatially separated atomic clouds containing thousands of atoms. These entangled atoms will then serve for improving the sensitivity of an atomic interferometer performing inertial sensing. CEBBEC is a collaboration of three experimental groups from Germany, France and Austria and two theoretical groups from Spain and Italy. We will explore different experimental approaches based on the same principle: the interactions present in a dense ultracold atomic induce pairwise emission of entangled atoms. Depending on the experimental configuration, either the momenta of the atoms or their spins are entangled. In this last case, the entanglement needs to be transferred to the momenta in order to obtained separated entangled clouds. We will look for the approach providing the most robust and interferometrically useful entanglement and implement it in an inertial sensor. This experimentally challenging project also requires to extend our theoretical understanding of entanglement: While in most experiments with entangled atoms the entanglement is shared between all atoms and diluted such that when splitting the atomic system in two parts the entanglement between these parts is always weak, in the CEBBEC project we are dealing with two parts which are maximally entangled, since each atom is mainly entangled with its partner and only weakly with atoms of other pairs. Combining experimental investigations and theoretical efforts, we will develop new concepts to describe and experimentally measure this type of entanglement.

Photon pairs play an important role in many fundamental quantum experiments and in many applications in quantum communication, quantum metrology and general quantum information technology. In the Vienna part of CEBBEC we studied how to create pairs of propagating atomic matter waves. Photon pairs are created in a nonlinear crystal when a UV photon 'converts' to two infrared photons, preserving energy and momentum conservation. We create our atom pairs from a Bose Einstein Condensate (BEC) that sits in an transvers excited quantum state in a elongated (one dimensional) trap. The quantum statistics of Bosons determines that the only way the atoms can fall back to the motional ground state is if they 'decay down' in pairs. Energy/momentum conversation led to the atoms in the pair being emitted with opposite momenta. In our experiment the elongated (1D) trap was two parallel 1D traps arranged in a 1D double well and the atoms decay down into a superposition of the two wells. Again, quantum statistics forces the atoms to decay into a single Bell state. This Bell state was verified by two particle interference. That is each atom individually shows NO interference pattern, but in pair correlations one observes joint 'two-particle' interference. The strong interactions of the atoms and the ability of detailed design of potentials on the atom chip has significant advantages: whereas for photons the efficiency to create a twin photon is exceedingly small (<10e-6) we can convert up to 50% of the atoms in the BEC into twin atom pairs. The atoms then propagate in single mode guides, and all can be used. In the future we will improve our setup to be able to probe Bells inequalities with propagating matter waves, and extend the simple double well into integrated circuits for entangled matter waves.