Observing Bell Inequality Violation Beyond the Qubit Bound in a Spinor Bose--Einstein Condensate

Quantum mechanics is famous for allowing particles to be linked in ways that seem to defy common sense. When two particles are entangled, measuring one instantly reveals information about the other, no matter how far apart they are. This phenomenon, known as nonlocality, was first highlighted by physicist John Bell, who proposed a way to test whether these connections could be explained by hidden, pre-existing rules or if they truly require a strange, non-classical reality. For decades, scientists have used this test to prove that nature is indeed quantum, but they have mostly done so with simple, two-state systems, like coins that can only be heads or tails. While these two-state systems are useful, they represent a very limited view of the quantum world. Many real-world quantum systems are more complex, capable of existing in three or more states at once, much like a die that can show one, two, three, four, five, or six sides. Understanding whether these more complex, multi-state systems can also exhibit these profound nonlocal connections has been a major open question, especially when dealing with huge groups of particles where individual tracking is impossible.

A team of researchers has now answered this question by demonstrating that a large cloud of atoms can display these complex quantum links in a way that simple two-state systems never could. Working with a spin-1 Bose-Einstein condensate, a state of matter where roughly 31,000 rubidium atoms behave as a single quantum entity, the scientists created a specific type of quantum state known as spin-nematic squeezing. In this state, the atoms are not just linked; their internal properties are correlated in a way that reduces uncertainty in certain measurements below what is possible for any collection of simple two-state particles. By measuring the collective behavior of the entire cloud without looking at individual atoms, the researchers observed a violation of a Bell inequality that surpassed the theoretical limit for qubits, the standard two-state units of quantum information. This result provides direct evidence that the atoms are acting as genuine three-level systems, or qutrits, and that their correlations are fundamentally more complex than anything achievable with simple binary quantum bits.

The experiment took place in a laboratory where the researchers trapped a cloud of rubidium-87 atoms in a focused beam of light, cooling them until they formed a Bose-Einstein condensate. Initially, all the atoms were prepared in a specific, calm state where they occupied the middle energy level of their internal structure. To create the necessary quantum links, the researchers used a technique called spin-mixing dynamics. By carefully adjusting magnetic fields and microwave frequencies, they triggered collisions between the atoms that caused pairs of them to spontaneously change their internal states, moving from the middle level to the upper and lower levels simultaneously. This process, which lasted for about 95 milliseconds, generated a highly entangled state where the atoms were squeezed together in a way that minimized random noise in specific directions. The researchers then measured the collective properties of the entire cloud, looking for patterns in how the atoms were distributed among their three possible internal states.

The key to their discovery was a mathematical test, known as a Bell witness, designed to distinguish between simple two-state correlations and more complex three-state ones. If the atoms were behaving as simple two-state systems, the results of their measurements would have stayed within a specific numerical boundary. However, the data collected from the rubidium cloud clearly crossed this boundary, dipping well below the limit that any collection of two-state particles could ever reach. The researchers calculated that to produce the observed results, nearly half of the atoms in the cloud had to be described as genuine three-level systems. This finding rules out the possibility that the observed correlations were just a complicated version of simple two-state entanglement. Instead, it confirms that the cloud possessed a richer, higher-dimensional quantum structure that is intrinsic to the spin-1 nature of the atoms.

This achievement is significant because it proves that high-dimensional quantum correlations can be detected in macroscopic systems using only collective measurements. In many previous experiments, verifying such complex quantum states required the ability to control and measure individual particles one by one, a task that becomes impossible with clouds containing tens of thousands of atoms. Here, the researchers showed that by looking at the group as a whole, they could certify the dimensionality of the quantum connections without needing to see the individual atoms. The experiment also highlighted the importance of squeezing in two specific directions simultaneously, a condition that was essential for revealing the three-level nature of the correlations. When the researchers deliberately removed the ability of the system to access its full three-dimensional structure, the special quantum violation disappeared, confirming that the effect relied on the full complexity of the spin-1 atoms.

The work establishes spinor Bose-Einstein condensates as a powerful platform for exploring the limits of quantum theory in large, interacting systems. It demonstrates that the strange, nonlocal connections predicted by quantum mechanics are not limited to simple binary systems but can thrive in more complex, multi-level environments. This opens new avenues for testing the foundations of physics and could lead to improved methods for quantum sensing and information processing that go beyond the capabilities of standard qubit-based technologies. By showing that coarse-grained measurements are sufficient to certify high-dimensional quantum resources, the study provides a practical path forward for investigating the rich, multi-layered structure of quantum correlations in the macroscopic world.