Imagine a stadium full of people performing "the wave" in perfect unison. If you want to model this crowd, you don't need to track every individual's movements or thoughts; you just need a single equation that describes the collective propagation of the wave.
This is the core intuition behind standard mean-field theories, such as the Gross-Pitaevskii equation, used to describe Bose-Einstein Condensates (BECs). Mean-field theory operates under a strict single-orbital constraint: it assumes that every single boson in an ultracold gas occupies the exact same quantum state. For decades, this simplification was the bedrock of cold-atom physics.
But what happens when the crowd splits, and people begin executing multiple different waves simultaneously? In quantum mechanics, this is known as condensate fragmentation. When a system fragments, particles distribute themselves across multiple macroscopically occupied quantum states (or natural orbitals), causing standard mean-field approximations to break down completely.
The Rigidity of the Single-Orbital Constraint
Why does mean-field theory fail when a condensate fragments? The answer lies in what the theory forces you to ignore: many-body correlations.
By restricting all particles to a single orbital, mean-field theory assumes the system remains perfectly coherent. As Alon et al. demonstrated, this assumption collapses as inter-particle interactions strengthen or the geometry of the trapping potential changes. When a system undergoes fragmentation, it trades its single-state purity to distribute particles across multiple orbitals, a reality that single-orbital equations lack the mathematical degrees of freedom to describe.
Crucially, this failure is often invisible if you only look at the physical shape of the cloud. Streltsov et al. showed that while mean-field theories can accurately predict the real-space spatial density of a fragmented gas in a double-well potential, they fail spectacularly when predicting momentum-space behavior. Because mean-field models cannot capture the complex many-body correlations that develop during fragmentation, they miss a massive loss of second-order coherence in momentum space, which manifests as highly volatile, oscillating correlation functions that simpler theories smooth away.
Destabilizing the Josephson Junction
The breakdown of mean-field theory is not just a subtle mathematical mismatch; it fundamentally alters the macroscopic behavior of quantum devices, such as bosonic Josephson junctions (a BEC split between a double-well potential).
According to standard mean-field textbooks, if you place a large number of interacting bosons into one side of a double well, repulsive interactions will lock them in place—a phenomenon called macroscopic self-trapping. However, exact many-body simulations by Streltsov et al. revealed that this self-trapping is an illusion of the approximation.
In reality, the system rapidly fragments into multiple occupied states within just a few oscillation cycles. This fragmentation distributes the particles, which reduces the effective quantum pressure and repulsive forces holding them back. As a result, the particles tunnel across the barrier far more easily than mean-field theory predicts, bypassing the expected trapping thresholds entirely.
Furthermore, this breakdown is not something that can be engineered away by adding more particles or weakening interactions. Streltsov et al. discovered a principle of universal fragmentation: if the product of the particle number and interaction strength is held constant, condensates of any size fragment to the exact same degree over long time scales. The Gross-Pitaevskii equation is fundamentally unstable over long durations because many-body effects inevitably drive the collapse of simple density oscillations.
Coherence Decoupling in Open Systems
The limitations of mean-field theory extend beyond closed traps to open systems where particles escape into space. When an interacting many-body system tunnels through a potential barrier into open space, one might intuitively expect a continuous, coherent stream of matter.
By modeling this decay from first principles, Lode et al. showed that the act of tunneling itself triggers fragmentation. As particles interact and escape, the system splits into distinct quantum states. Interestingly, spatial coherence is preserved within the remaining trapped source and within the escaping jets of particles, but coherence is completely lost between the source and the jets. Because mean-field approximations assume absolute coherence across the entire system, they cannot predict this spatial decoupling or the multi-peak momentum distributions that emerge from it.
The Illusion of the Average: Single-Shot Measurements
If mean-field theory is so qualitatively flawed in these regimes, why did it escape detection in early experiments? The answer comes down to how data is collected in the laboratory.
Traditional quantum simulations calculate the expectation value of the single-particle density—effectively averaging the results of thousands of experimental runs. In a fragmented state, this average is smooth and featureless, matching the predictions of mean-field theory.
To bridge this gap, Kasevich developed an algorithm to simulate "single-shot" experimental images by sampling the full $N$-particle probability density.
The results are telling: while the average density profile of a fragmented state looks uniform, any individual single-shot image reveals jagged interference fringes, random vortices, and spontaneous symmetry breaking. The smooth profile predicted by mean-field theory does not represent the physical reality of any single experiment; rather, it is a statistical artifact of averaging over massive, intrinsic many-body fluctuations driven by fragmentation.
Key Foundational Work
Reduced density matrices and coherence of trapped interacting bosons (2008) — contributed 5 concept elements to this direction.
Continuous configuration-interaction for condensates in a ring (2004) — contributed 1 concept elements to this direction.