🔬 mesoscale physics

Brownian Motion in Orthogonal and Symplectic Groups

This paper establishes a unified diagrammatic framework to characterize Brownian ensembles for orthogonal and symplectic random matrices, computing polynomial averages and constructing an invariant interpolation for the $\mathrm{SO}^-(q)$ sector to advance understanding in quantum chaos, transport, and information scrambling.

Original authors: Zhiyang Tan, Piet W. Brouwer

Published 2026-07-07

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Zhiyang Tan, Piet W. Brouwer

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are watching a drop of ink spread out in a glass of water. At first, the ink is a tight, organized blob. As time passes, it swirls, mixes, and eventually becomes a uniform, featureless gray. In the world of quantum physics, scientists use a similar concept called Brownian Motion to describe how complex systems (like tiny particles or quantum computers) lose their initial order and become completely random.

This paper by Zhiyang Tan and Piet W. Brouwer is like a new, highly detailed instruction manual for predicting exactly how this "mixing" happens in two specific types of quantum systems: those that behave like mirrors (Orthogonal) and those that behave like spinning tops (Symplectic).

Here is a breakdown of their work using everyday analogies:

1. The Goal: Predicting the "Mixing"

In physics, when a system is perfectly chaotic, its behavior follows a specific, universal pattern known as the "Haar measure." Think of this as the "perfectly mixed" state where every possible outcome is equally likely.

The authors wanted to know: If we start with a specific, ordered quantum state and let it evolve randomly, how long does it take to reach that "perfectly mixed" state? And more importantly, can we calculate the statistics of this process at every step?

To do this, they developed a diagrammatic framework. Imagine trying to solve a complex math problem by drawing pictures instead of writing equations. They created a set of visual "rules" (like a board game) where lines and loops represent mathematical operations. By following these rules, they could calculate the average behavior of these systems up to a high level of complexity (fourth order).

2. The Two Types of "Mirrors"

The paper focuses on two specific types of quantum symmetry:

Orthogonal (The Mirror): Imagine a system where the physics works the same if you look at it in a mirror. Mathematically, these are "real" numbers.
Symplectic (The Spinning Top): Imagine a system with a specific kind of "handedness" or spin (like electrons). These involve "quaternion" numbers, which are more complex than regular numbers.

The authors showed that even though these two systems are different, they can be described using the same underlying "mixing" logic, just with a few sign changes (like flipping a switch from positive to negative).

3. The "Broken" Mirror Problem

Here is a tricky part the authors solved.
Imagine a door that can only open one way. If you start at the "Open" position (the identity matrix), you can walk through the door and wander around the room (the group of matrices with determinant +1). But you can never reach the "Closed" side of the room (matrices with determinant -1) because you can't walk through a wall.

In the world of Orthogonal matrices, the "room" is split into two disconnected halves. Standard Brownian motion (starting from a clean slate) only explores the first half. It never reaches the second half.

The Solution: The authors figured out how to "teleport" the system to the second half at the very beginning. By starting with a specific, pre-arranged "seed" (a matrix with a determinant of -1) and then letting the random mixing begin, they created a path that connects the two halves. This allows them to study the entire "room," not just one side. They call this an "interpolation," which is like building a bridge between a specific starting point and total randomness.

4. What They Calculated

Using their picture-based rules, the team calculated specific "averages" (mathematical summaries) of how these systems behave. They didn't just look at the final result; they looked at the journey.

They calculated how the system's "memory" of its starting point fades away over time.
They found that the time it takes to become fully random scales with the size of the system in a predictable way (specifically, it grows logarithmically, meaning it gets slower as the system gets bigger, but not too much slower).

5. Real-World Applications Mentioned

The paper connects these abstract math problems to three specific physical fields:

Quantum Information: They used their results to measure how fast "scrambling" happens. If you drop a piece of information into a quantum system, how quickly does it get mixed up so you can't find it anymore? Their math helps predict this speed.
Quantum Chaos: They looked at the "Spectral Form Factor," which is a way of listening to the "music" of a quantum system to see if its notes (energy levels) are repelling each other like magnets (a sign of chaos).
Quantum Transport: They applied their findings to a junction between two special superconductors (materials that conduct electricity with zero resistance). They calculated how heat flows through this junction when the electrons are bouncing around randomly due to disorder.

Summary

In short, Tan and Brouwer built a new set of "visual tools" to track how quantum systems lose their order and become random. They solved a specific puzzle about how to study the "hidden side" of mirror-like quantum systems and provided precise formulas for how fast these systems mix. This helps physicists understand everything from how quantum computers scramble data to how heat moves through exotic superconducting wires.

Technical Summary: Brownian Motion in Orthogonal and Symplectic Groups

Problem Statement
Matrix Brownian motion serves as a fundamental framework for analyzing crossover ensembles in quantum chaos, quantum transport, and the thermalization of many-body systems. While the Brownian motion of unitary matrices (leading to the Circular Unitary Ensemble, CUE) is well-established, there is a need for a unified, explicit characterization of Brownian ensembles for orthogonal and symplectic matrices. These ensembles describe chaotic systems with particle-hole symmetry, where the underlying Hamiltonian is anti-symmetric (orthogonal case) or anti-self-dual (symplectic case). A specific topological challenge arises in the orthogonal case: the standard Brownian process starting from the identity matrix is confined to the special orthogonal group $SO(q)$ (determinant +1) and cannot access the disconnected component $SO^-(q)$ (determinant -1). This paper addresses the calculation of polynomial averages for these ensembles and proposes a method to interpolate between the disconnected sectors.

Methodology
The authors develop a unified diagrammatic framework based on the Brauer algebra to compute polynomial averages of the form $\langle U_{i_1 i'_1}(t) \dots U_{i_k i'_k}(t) \rangle$ for $k \le 4$ .

Tensor Algebra: The evolution of the moment operator $U_k(t) = \langle U(t)^{\otimes k} \rangle$ is governed by a linear differential equation $U_k(t) = e^{L_k t}$ . The generator $L_k$ is derived by expanding the time-ordered exponential for small $t$ and averaging over the random Hamiltonian $H(t)$ .
Diagrammatic Representation: The authors generalize the diagrammatic approach used for unitary Brownian motion to the orthogonal and symplectic cases. They define rank- $2k$ $2 k$ tensors ( $I_k, S_k, X_k$ $I_{k}, S_{k}, X_{k}$ , and higher-order variants) represented by diagrams involving Kronecker deltas and the involution matrix $Z$ $Z$ .
- In the orthogonal case, $Z$ is symmetric, and diagrams are unoriented.
- In the symplectic case, $Z$ is antisymmetric, requiring oriented lines where reversing orientation introduces a negative sign.
Algebraic Calculation: The authors identify the commutative algebra generated by these tensors. For $k=2$ , the algebra is spanned by three elements; for $k=3$ , it requires five basis elements (including $S'_3$ and $X'_3$ ); for $k=4$ , it spans 12 elements. By constructing the multiplication tables (represented as matrices $M_k$ ), they diagonalize the generator $L_k$ to obtain explicit analytical expressions for the moment operators $U_k(t)$ .
Interpolation for $SO^-(q)$ : To address the topological obstruction preventing access to the $SO^-(q)$ sector, the authors introduce a non-trivial initial condition $V = U(0)$ . Specifically, they consider an ensemble where $V$ is a Haar-random special orthogonal matrix conjugated by a diagonal matrix with a single $-1$ entry. The full moment operator is then constructed as $U_k(t) = V_k \tilde{U}_k(t)$ , where $\tilde{U}_k(t)$ is the standard Brownian evolution.

Key Contributions and Results

Explicit Polynomial Averages: The paper provides explicit analytical expressions for the moment operators $U_k(t)$ and trace moments $R_{i_1, \dots, i_m}$ up to the fourth order ( $k=4$ ) for both orthogonal and symplectic Brownian ensembles. These expressions are given in terms of exponential decay modes dependent on the dimension $q$ and time $t$ .
Duality Transformation: The authors identify a duality transformation ( $t \to -t, q \to -q$ , and sign flips for specific tensors) that relates the results of the orthogonal and symplectic cases, allowing the symplectic results to be derived directly from the orthogonal ones.
Interpolation Ensemble: A rigorous construction is presented for an ensemble that interpolates between a "minimal" distribution on the disconnected $SO^-(q)$ sector and the Haar measure on $SO^-(q)$ in the long-time limit. Crucially, this interpolating ensemble maintains orthogonal invariance at all times $t$ , unlike previous approaches that might only achieve invariance asymptotically.
Trace Moments: The paper derives explicit formulas for trace moments up to $k=3$ (with $k=4$ provided in the appendix), which encode the spectral statistics of the evolving matrices.

Applications and Significance
The authors demonstrate the utility of their derived moment operators in three distinct fields:

Quantum Information: The results are applied to calculate frame potentials $F^{(k)}(t)$ , which measure the distance of the ensemble from forming a $k$ -design. The analysis shows that the characteristic scrambling time for convergence to the Haar measure scales as $\sim \log q$ (after rescaling time by $q$ ), consistent with unitary cases. The framework also supports calculations for out-of-time-order correlators (OTOCs), second Rényi entropies, and shadow norms for classical shadow tomography.
Quantum Chaos: The trace moments are used to evaluate the spectral form factor $K(t)$ and its variance, which diagnose level repulsion and spectral correlations.
Quantum Transport: The framework is applied to model disorder-induced scattering between co-propagating chiral Majorana modes at the junction of two topological superconductors (symmetry classes D and C). The authors derive the average and variance of the heat conductance $G_Q$ as a function of the Brownian time (proportional to junction length).

Significance
The paper claims to provide a "controlled route" to non-equilibrium spectral dynamics for systems with particle-hole symmetry. By extending the diagrammatic machinery to orthogonal and symplectic groups, the work offers a tractable analytical tool for studying operator spreading, thermalization, and transport in systems where time-reversal or particle-hole symmetry is preserved. The construction of the $SO^-(q)$ interpolation ensemble is highlighted as a solution to a topological limitation in standard Brownian motion models, enabling the study of dynamics in disconnected symmetry sectors while preserving invariance throughout the evolution. The authors position this work as a foundational step for future studies on measurement-induced phase transitions and noise-induced transitions in monitored quantum systems.