Fermionic Gaussian Scrooge Ensembles in Deep Thermalization

In the quantum world, particles do not simply sit still; they exist in a state of constant, intricate connection known as entanglement. When a large group of these particles interacts chaotically, they spread their information so thoroughly that any small piece of the system begins to look like a random sample of the whole. This process, called thermalization, is how a hot cup of coffee eventually reaches the same temperature as the room. However, physicists have recently discovered a deeper layer to this story. If you take a snapshot of a chaotic system by measuring part of it, the remaining unmeasured part does not just settle into a simple average state. Instead, it forms a specific collection of possible states, each with a certain probability, that turns out to be the most random arrangement possible given the information you have. This collection is known as the Scrooge ensemble, a concept that describes how chaos creates a universal kind of randomness.

But not all quantum systems are chaotic. Some, known as free-fermion systems, follow much stricter rules. In these systems, the particles do not interact with each other in complex ways, and their behavior is constrained to a specific mathematical shape called a Gaussian state. For a long time, scientists wondered if the deep thermalization phenomenon—the emergence of that universal random collection—could happen in these orderly, non-chaotic systems. If it did, the resulting collection of states would have to look different from the chaotic version, respecting the unique rules of these free particles. This question remained unanswered until a new study by Ning Sun and Pengfei Zhang at Fudan University and the Hefei National Laboratory. They set out to define what this universal randomness looks like for free fermions and to prove that it actually emerges in real quantum systems.

The researchers began by proposing a new theoretical framework they call the fermionic Gaussian Scrooge ensemble. To understand this, imagine a standard deck of cards where the order is completely random; this is what happens in chaotic systems. Now, imagine a deck where the cards are still shuffled, but they must stay in a specific pattern, like all red cards grouped together. This patterned randomness is what the new ensemble describes. The team defined this collection mathematically by taking the standard random distribution for these specific particles and distorting it based on the average state of the system. A key feature of their discovery is that this new ensemble possesses a hidden symmetry involving how different copies of the system rotate relative to one another, a property that distinguishes it from the chaotic version. They showed that this structure is the correct target for deep thermalization in any system of free fermions.

To prove that this theoretical idea is real, the authors demonstrated its emergence in two very different physical settings. First, they analyzed a specific theoretical model known as the SYK2 model, which describes a collection of particles governed by a quadratic Hamiltonian with random couplings. They performed a rigorous mathematical proof showing that if you measure part of this system at any point in time, the remaining part perfectly matches their new fermionic Gaussian Scrooge ensemble. This was not just a long-term trend; it held true at every single moment of the system's evolution. The mathematics revealed that the system naturally organizes itself into this specific random pattern because of a continuous family of solutions in the underlying equations, a feature that arises from the way the particles are connected.

Second, the team turned to computer simulations to see if this phenomenon appeared in more generic, random systems. They modeled a one-dimensional chain of particles evolving through a series of random gates that preserved the total number of particles, a condition known as charge conservation. They started with various initial configurations, including some where particles were paired up in specific ways and others where they were distributed differently. As they let the system evolve for longer and longer times, they measured the remaining particles and compared the results to their new theoretical ensemble. The data showed a clear convergence: as time passed, the collection of states generated by the random circuit approached the fermionic Gaussian Scrooge ensemble. This held true across different starting conditions, suggesting that this behavior is a universal rule for this class of systems, not a fluke of a specific setup.

The study also addressed a common question about what happens when measurements reveal specific details, such as the exact number of particles in a section. In chaotic systems, such measurements can sometimes force the system into a very specific, less random state. However, the researchers found that in these free-fermion systems, the natural fluctuations of the particles prevent the measurements from locking the remaining system into a rigid pattern. Even when the number of particles is conserved, the system retains enough freedom to explore the full range of the new Gaussian Scrooge ensemble. This finding clarifies that the universal randomness of deep thermalization is robust, surviving even when the system is constrained by physical laws like charge conservation.

By establishing this new ensemble, the work provides a complete description of how free-fermion systems thermalize at a deep level. It bridges the gap between the known behavior of chaotic systems and the previously unexplored territory of non-interacting ones. The results confirm that while the rules of the game change when particles stop interacting chaotically, the game still ends in a state of maximum randomness, just a different kind of randomness than before. This insight helps physicists understand the fundamental limits of how information spreads in quantum matter, offering a clearer picture of the statistical laws that govern the quantum world.