⚛️ quantum physics

The Lieb-Robinson correlation function for long disordered transverse-field Ising chains

This paper employs a linearly-scaling computational method to calculate the Lieb-Robinson correlation function in long disordered transverse-field Ising chains, revealing how increasing disorder induces localization that halts quantum information propagation and characterizing the resulting correlation length's behavior across the phase transition.

Original authors: Brendan J. Mahoney, Craig S. Lent

Published 2026-10-14

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Brendan J. Mahoney, Craig S. Lent

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

In the quantum world, information does not travel like a ripple in a pond or a flash of light in a vacuum. Instead, it spreads through a complex web of connections between tiny particles, governed by rules that are fundamentally different from our everyday experience. For decades, physicists have known that even without the speed limit of light, there is a strict boundary to how fast information can move through a chain of interacting quantum bits. This boundary, known as a light cone, acts like an invisible wall: inside it, particles can influence one another; outside it, they remain completely unaware of each other's existence. Understanding how this boundary behaves is crucial for building future quantum computers, which rely on the precise movement of information across large arrays of qubits. However, calculating exactly how this information spreads in a system with thousands of particles has been a nearly impossible task, as the mathematical complexity grows so fast that it overwhelms even the most powerful supercomputers.

A team of researchers at the University of Notre Dame has now cracked this problem for a specific type of quantum chain, revealing how disorder in the material can completely halt the flow of information. By developing a new computational method that scales efficiently with the size of the system, they were able to simulate chains containing over a thousand qubits, a feat previously out of reach. Their work focuses on a model known as the transverse-field Ising chain, a simplified but powerful representation of how magnetic spins interact. In a perfect, orderly chain, they confirmed that quantum correlations race down the line at a constant speed, creating a clear, expanding front of influence. But when they introduced randomness into the connections between the qubits, the story changed dramatically. The disorder acted like a series of potholes and roadblocks, causing the information to get stuck and localized near its starting point.

The researchers found that as the level of disorder increased, the smooth, diagonal edge of the information front began to bend upward, eventually becoming a vertical wall. This vertical shape signifies that the information has stopped moving entirely; it is trapped in a small region, unable to propagate further down the chain. This phenomenon, a form of localization, means that in a sufficiently disordered system, a quantum signal sent from one end will never reach the other, no matter how much time passes. The team measured exactly how far the information could travel before stopping, defining a specific distance called the correlation length. They discovered that this distance depends heavily on the strength of the disorder and the average strength of the connections between the qubits. In the phase where the spins are disordered, the distance the information travels grows with the square of the coupling strength but shrinks rapidly as the disorder increases.

Perhaps most surprisingly, the researchers observed that the behavior of this complex quantum system could be mapped directly onto a much simpler model of a single particle moving through a disordered lattice. This connection allowed them to translate their findings into a language that bridges the gap between complex many-body physics and single-particle behavior. They also examined what happens at the critical point where the system shifts from one phase to another. Here, they found that the distance the information travels reaches a peak, suggesting that the system is most vulnerable to spreading information right at the moment of transition. However, they also noted a subtle but important detail: while the average behavior of the system suggests a certain level of connectivity, rare, specific arrangements of disorder can create "lucky" paths where information travels much further than usual. These rare events skew the average, creating a divergence between what is typical and what is mathematically average, a nuance that could be vital for designing robust quantum devices.

The study confirms that while quantum information can move swiftly in a clean environment, the introduction of randomness acts as a powerful brake, capable of freezing the flow of information entirely. This has profound implications for the design of quantum computers, where maintaining the flow of information is essential. If the connections between qubits are too disordered, the computer may become unable to process data across its full length. The researchers' ability to simulate these systems with such precision provides a new tool for predicting how real-world quantum materials will behave, offering a clear path forward for engineers who must navigate the delicate balance between order and chaos in the quantum realm.

Technical Summary: The Lieb-Robinson Correlation Function for Long Disordered Transverse-Field Ising Chains

Problem Statement
The propagation of quantum information and correlations in many-body systems is fundamentally constrained by the Lieb-Robinson (LR) bound, which establishes a finite velocity for the spread of operator support. While considerable theoretical work has refined bounds on the LR correlation function ( $C_{A,B}$ ) for various systems, directly calculating the actual value of this function for large systems remains computationally intractable. Standard state-based methods require a state space that grows exponentially with system size ( $2^{N_q}$ ), limiting direct calculations to relatively small chains (e.g., $N_q \approx 22$ ). Furthermore, the impact of disorder on these correlations—specifically in the context of Anderson and many-body localization—requires analysis of systems large enough to observe localization phenomena, which has been difficult to achieve with existing computational techniques.

Methodology
The authors employ a recently developed "Operator Pauli Walk" (OPW) method to calculate the LR correlation function directly. This approach avoids the exponential explosion of the full Hilbert space by exploiting the specific structure of the Transverse-Field Ising Model (TFIM) Hamiltonian.

Operator Space Reduction: Instead of evolving a quantum state, the method tracks the evolution of the operator $Z_1(t)$ in the Heisenberg picture. For the TFIM, the nested commutators required to evolve $Z_1(t)$ explore only a structured subspace of Pauli strings with dimension $2N_q$ , rather than the full $4^{N_q}$ space.
Matrix Formulation: This dynamics is mapped onto a discrete walk on an operator graph, represented by a sparse, tridiagonal adjacency matrix $A$ of size $2N_q \times 2N_q$ . The LR correlation function is derived from the matrix exponential of this adjacency matrix:
$C_k(t) = 2 \sqrt{\sum_{m=2k}^{2N_q} \left| \left[ \exp\left(-\frac{2\pi t}{\tau} A\right) \right]_{1,m} \right|^2}$
Disorder Implementation: The method is extended to disordered chains by allowing the coupling strengths $J'_k$ to vary spatially. This modifies the weights of the adjacency matrix edges while maintaining the linear scaling of the matrix size. The authors calculate disorder-averaged correlation functions by sampling random coupling configurations from a uniform distribution and computing the arithmetic mean ( $\bar{C}$ ) and geometric mean ( $C_{typ}$ ).
System Scale: This linear scaling ( $O(N_q)$ ) enables the computation of LR correlations for chains exceeding 1,000 qubits, a regime previously inaccessible for direct calculation.

Key Results
The study investigates the propagation of quantum correlations in TFIM chains across paramagnetic ( $J' < 1$ ) and ferromagnetic ( $J' > 1$ ) phases, with and without random coupling disorder.

Propagation and Localization:
- In the absence of disorder, the correlation front propagates ballistically with a velocity determined by the coupling strength, forming a linear light cone.
- The introduction of disorder impedes this propagation. As disorder strength ( $\Delta J'$ ) increases, the light-cone isocontours bend upward and eventually become vertical, indicating that the correlation front halts. This signifies the localization of quantum correlations.
- At late times, the correlation function becomes stationary (time-independent), confirming the cessation of information propagation down the chain.
Correlation Length and Disorder:
- The spatial decay of correlations is characterized by a correlation length $\lambda$ .
- In the paramagnetic phase, $\lambda$ decreases as disorder increases. The authors find that $\lambda$ scales quadratically with the inverse of the disorder level ( $\lambda \sim 1/(\Delta J')^2$ ).
- In the ferromagnetic phase, correlations penetrate farther than in the paramagnetic phase due to higher group velocities, but disorder similarly induces localization.
- The correlation length peaks at the quantum phase transition ( $\delta \approx 0$ ), where the quasiparticle excitation energy vanishes.
Statistical Properties (Arithmetic vs. Geometric Means):
- The paper highlights a divergence between the arithmetic mean ( $\bar{C}$ ) and the geometric mean ( $C_{typ}$ ) of the correlation function, particularly deep in the chain where correlation values are small.
- The arithmetic mean is dominated by rare "lucky" samples with significantly higher correlation values, while the geometric mean reflects the typical behavior. This divergence is attributed to the multiplicative nature of scattering from disorder and is observed even in simpler tight-binding models, suggesting it is a general feature of disordered 1D systems rather than a unique signature of Griffiths phases in the TFIM.
Mapping to the SSH Model:
- The authors demonstrate a quantitative mapping between the TFIM LR correlation function and the probability density of a single particle in a disordered Su-Schrieffer-Heeger (SSH) tight-binding model. The adjacency matrix governing the operator walk in the TFIM is shown to be equivalent (via a gauge transformation) to the SSH Hamiltonian.

Significance and Claims
The paper claims that the Operator Pauli Walk method provides a computationally efficient tool for characterizing quantum information propagation in large, disordered many-body systems. By enabling calculations for systems with over 1,000 qubits, the method reveals the transition from ballistic propagation to localized behavior in disordered Ising chains. The work establishes that increasing disorder causes the localization of quantum correlations, effectively halting the propagation of information. Furthermore, the paper provides a specific characterization of the correlation length's dependence on disorder and coupling strength, and offers a rigorous connection between the many-body LR correlation function and single-particle tight-binding physics, allowing for the application of single-particle localization insights to many-body correlation dynamics.

More like this