Theory of post-selected entanglement transitions in monitored bosons

In the quantum world, the act of looking at a system changes it. This is not merely a philosophical curiosity but a fundamental law of physics: when we measure a particle, we force it to choose a state, collapsing a cloud of possibilities into a single reality. For decades, physicists have studied how this constant "watching" affects groups of particles, particularly those that are easy to count, like electrons or artificial atoms that can hold only one or two units of energy. In these systems, a fascinating battle plays out between the natural tendency of particles to spread out and entangle with one another, and the tendency of measurements to pin them down and separate them. The result is a phase transition, a sudden shift in how the particles behave, much like water freezing into ice.

However, a vast and complex territory remained largely unexplored: systems made of bosons, a type of particle that can pile up in the same place in unlimited numbers. Unlike their fermionic cousins, bosons are naturally prone to clustering, and their behavior is notoriously difficult to predict when they are being watched. For a long time, scientists believed that if these bosonic systems were not interacting with each other, the act of measurement would simply strip away all their complexity, leaving them in a simple, unentangled state regardless of how they were set up. The question lingered: could a system of free, non-interacting bosons, if initialized in a specific way, still exhibit a dramatic shift from a highly entangled state to a simple one, driven solely by the rate at which it is observed?

A team of researchers has now answered this question with a resounding yes. By developing a new mathematical framework to track the fate of bosons under continuous observation, they discovered that these systems can indeed undergo a sharp phase transition. The key lies in how the particles are initially arranged. When a specific type of bosonic system is started with particles sitting on individual sites and then subjected to a steady stream of measurements, the outcome depends entirely on the strength of that monitoring. If the observation is weak, the particles remain in a chaotic, highly entangled state where information is shared across the entire system. But if the observation becomes strong enough, the particles suddenly collapse into a tiny, microscopic number of states, effectively condensing into a single, quiet mode. This shift changes the way the system stores information, moving from a scaling that grows with the size of the system to one that grows only with the logarithm of its size.

To uncover this phenomenon, the researchers turned to a theoretical model involving a one-dimensional lattice, a grid-like structure where particles can hop from site to site. They focused on a specific setup where the bottom row of sites is constantly monitored for the presence of particles, but with a twist: they only considered the rare, post-selected scenarios where no particles were ever detected. In the language of quantum mechanics, this "no-click" trajectory is equivalent to the system evolving under a special, non-standard type of energy rule that causes certain states to decay faster than others. The team realized that standard mathematical tools, which work well for simpler particles, failed here because the initial state of the bosons was inherently complex and did not follow the usual rules of symmetry.

They overcame this hurdle by creating a new method that avoids the need for complex, repetitive calculations. Instead, they expressed the entanglement of the system using a specific type of matrix calculation known as a permanent, which captures the unique way bosons exchange places. This allowed them to track the system's evolution with high precision. Their simulations revealed a clear dividing line in the behavior of the system. When the connection between the monitored and unmonitored parts of the lattice was strong, or the monitoring rate was low, the system behaved like a typical free system, maintaining a high level of entanglement that scaled with the total number of particles. This is the volume-law phase, where the complexity of the system is vast and distributed.

However, as the researchers increased the rate of monitoring or adjusted the connections between the lattice sites, the system crossed a critical threshold. In this new regime, the mathematical spectrum of the system reorganized itself. The imaginary parts of the energy levels, which dictate how quickly different modes of the system decay, became spread out. This dispersion acted as a filter, causing almost all the particle states to vanish rapidly, leaving only a single, slowest-decaying mode behind. The bosons, driven by the measurement back-reaction, dynamically condensed into this single surviving state. This is the log-volume phase, where the entanglement is drastically reduced, scaling only with the logarithm of the system size.

The researchers confirmed that this transition is not an artifact of their specific model but a fundamental feature of how free bosons respond to observation. They introduced a diagnostic tool that can predict this transition without needing to perform the incredibly difficult calculations of the full system. By looking at the rank of a specific matrix that describes how the particles overlap, they could determine whether the system would remain in a complex, entangled state or collapse into a simple one. This rank drops to one in the condensed phase, signaling that the entire system has effectively merged into a single mode.

This work establishes a distinct mechanism for measurement-induced phase transitions that is unique to bosonic systems. It challenges the previous assumption that free bosons would always succumb to the disentangling power of measurement. Instead, it shows that under the right conditions, the interplay between the initial state and the monitoring process can drive the system into a state of dynamic condensation. The findings provide a computationally efficient way to diagnose these transitions and offer a concrete protocol for observing them in future experiments with superconducting cavities or other bosonic platforms. The study suggests that the dimensionality of the space of surviving modes is the critical factor determining the complexity of the system, a principle that could reshape our understanding of how quantum information is preserved or lost in the presence of observation.