Skip to main content

Non-Classical Correlations and Entanglement in Twin-Atom Beams

Final Report Summary - CORENT (Non-Classical Correlations and Entanglement in Twin-Atom Beams)

One of the major objectives of quantum atom optics (QAO) is the production and observation of entanglement in pairs of atoms (twin atoms), with the aim of realising EPR- or Bell-type experiments. There are several reasons that motivate this objective. From the strict perspective of settling the conflict between quantum theory and local realism, massive objects such as atoms have the advantage of being closer to the original formulation of the EPR paradox, and (maybe) even more bound to obey local reality principles. To this regard, the most appealing situation would be that where the momentum state of two atoms is entangled, as in the Rarity and Tapster experiments. There also exists alternatives to standard quantum theory that are not ruled out by any Bell inequality, and which might be tested by experiments using atoms. This is the case, for example, of theories of spontaneous decoherence or gravitational nonlinearity. Finally, from a more practical perspective, entanglement is the central resource of quantum computing, and it can be used to greatly enhance the sensitivity of interferometric measurements, which are at the heart of the most precise clocks or inertial sensors. Atoms are a promising platform to develop such applications, and another important task of QAO is to provide the required experimental tools.

While the vast majority of QAO experiments so far used the internal degrees of freedom to carry the quantum properties, the CORENT project was aimed at exploring a radically different route by focusing on pairs of atoms with an entangled momentum state. The strategy was to build on previous work of the group led by C. Westbrook at the Laboratoire Charles Fabry on a promising source of twin-atom beams based on a dynamical instability in a Bose-Einstein condensate (BEC) of metastable Helium-4. Interferometers of increasing complexity would then be formed by applying successive optical lattices causing the Bragg diffraction of the atoms. The main objective we assigned to the project was to demonstrate the presence of entanglement between the two atomic beams in a regime where the beams are populated by many atoms. The regime where a single atom pair is emitted, much more difficult to reach, was left for the very end of the project, and should have rather been the focus of later studies.

Briefly stated, we succeeded beyond our expectations as we were able to demonstrate unambiguous quantum effects in the few particles per mode limit already in the second year of the project. The work was described in two articles, one published in the Nature journal in 2015 [1], and one submitted to the Physical Review Letters journal in 2017 [2]. In the first article we reported an experiment closely analoguous to the celebrated Hong-Ou-Mandel experiment. We realised a two-inputs/two-outputs interferometer and sent in atom pairs created by our source. The two atoms of a pair enter the interferometer each by a different input port. They are then recombined on an atomic beam-splitter and can exit the interferometer in the two outputs ports with an a priori equal probability. Classically, the probability for an atom to exit in one output or the other is completely independent from the fate of the other atom. But quantum mechanically this is no longer the case. An interference between the two-particle probability amplitudes results in the cancellation of the probability to see the two atoms exit in separate outputs. In other words, the atoms are seen to always exit in the same port, even though this port changes randomly from one measurement to another. This paradigmatic quantum optics effect probably is the simplest manifestation of a two-partical interference and was never observed for atoms so far. Our experiment therefore had a strong impact on the research field, and even attracted some attention from the media (see "Dissemination activities").

These unanticipated results motivated a change of priority for the project. Taking advantage of the multimode character of our source of atom pairs, we started working on the implementation of a four-inputs/four-outputs interferometer to perform a test of a Bell inequality with momentum entangled atoms, in the spirit of the experiment performed with photons by Rarity and Tapster in 1990. Such experiment is even more difficult because it requires a high momentum selectivity and a precise control of the relative phase between the two branches of the interferometer. We spent a long time testing and calibrating the different elements of the interferometer one by one. Unfortunately, due to unrelated technical issues, we were forced to delay its realization for a year. We however managed to perform a first proof-of-principle experiment in a simplified configuration where the relative phase can not be varied at will. We could observe quantum interferences that hint at the presence of entanglement, but do not yet constitute a rigorous proof. This work gave us confidence that our approach enables a test of a Bell inequality for massive particles with entangled momenta. We have resumed our progression towards this goal about two months before the end of the CORENT project and are currently pursuing this work.

Beside its scientific objectives, the CORENT project also served the purpose of reintegrating the fellow, Marc Cheneau, in the French research community after his post-doctoral stay. In this domain also the project was successful. The fellow's researcher position at the Charles Fabry Laboratory was secured by CNRS in January 2014 and advanced in January 2017. Since the beginning of the project, the fellow tooke an active part in the life of his host institute as an elected member of the laboratory council. He was offered the freedom to develop a new research activity on his own, which was made possible when he was awarded an ERC starting grant in 2015. The DYNAMIQS project, aimed at studying the relaxation dynamics of quantum gases, started in 2016. The project is conducted by a newly formed team of two PhD students and one engineer, supervised by the fellow. Himself has been working on both the CORENT and the DYNAMICS project during the last 18 months of the CORENT project.

[1] R. Lopes et al., Nature 520, pp 66-68 (2015)
[2] P. Dussarrat et al., available at arXiv:1707.01279 [cond-mat, physics:quant-ph]