Recently, a team led by Prof. GUO Guangcan from the University of Science and Technology of China (USTC) of the Chinese Academy of Sciences (CAS), collaborating with Prof. Marc-Olivier Renou from the Institute of Photonic Sciences, Spain, verified genuine multipartite nonlocality based on Local operation and shared randomness (LOSR), and demonstrated experimentally that bipartite nonlocal correlation or tripartite nonlocal correlation could not explain all possible correlations in nature. The study was published in Physical Review Letters.
Quantum nonlocality describes the strong nonlocal correlations between particles and could lead to critical applications in quantum information. For multibody system, genuine multipartite nonlocality has been recognized as the strongest form of multipartite quantum correlation. The examination of genuine multipartite nonlocality relies on the violation of Bell’s inequality within the local operation and classical communications (LOOC) framework. However, scientists usually postulate that the probability distribution of joint measurements between different observers obeys the non-signaling condition in examining multipartite nonlocality, while classical communication is forbidden. Therefore, the LOCC-based definition of genuine multipartite nonlocality is not restrictive.
In quantum resource theory, LOCC is not the only free operation that can be used to define entangled resources. When classical communication is excluded, a more natural definition is based on the theoretical framework of LOSR. In the experiment, the research team first verified the nonlocality of tripartite entanglement. The LOSR model of the tripartite system not only allows shared randomness between three bodies, but also allows unrestricted correlation between either two. That is, their correlation could be a quantum correlation or a correlation that is allowed by generalized probabilistic theory, which does not rely on quantum mechanical descriptions.
The team improved the LOSR nonlocality inequality proposed in previous theoretical work and presented a more robust inequality for the white noise commonly found in the multiphoton experiment. They experimentally verified the genuine tripartite LOSR nonlocality by a violation of more than 26.3 standard deviations by using the self-developed “sandwich” high-fidelity multiphoton entanglement source. Furthermore, the team experimentally violated the genuine quadripartite LOSR inequality, which ruled out the local model based on shared randomness and generalized tripartite entanglement, demonstrating genuine quadripartite nonlocality.
(a) LOSR local model; (b) New LOSR inequality for white noise. (Image by Prof. GUO Guancan’s team)
This work shows that bipartite nonlocal correlation or tripartite nonlocal correlation could not explain all possible correlations in nature. The inequality the team proposed can be generalized to any number of parties. And the method of preparing high-fidelity entangled sources provides the basis for experimentally demonstrating the nonlocality of a more extensive quantum system in the future.
Paper link: https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.129.150402
(Written by MA Xuange, edited by ZHANG Yihang, USTC News Center)