🔢 mathematics

Matrix-product state skeletons in Onsager-integrable quantum chains

This paper extends the concept of dense Matrix-product state (MPS) skeletons from free-fermion models to interacting $N$ -state Onsager-integrable chiral clock chains by constructing MPS that form a dense skeleton in gapped regions and serve as exact eigenstates in specific spectral sectors, thereby enabling closed-form calculations of disorder parameters and revealing new excited states through the Onsager algebra.

Original authors: Imogen Camp, Nick G. Jones

Published 2026-06-19

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Imogen Camp, Nick G. Jones

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 a vast, complex landscape representing the behavior of a quantum chain—a line of tiny magnets that can point in different directions. Physicists call this landscape a "phase diagram." In some parts of this landscape, the magnets settle into a calm, predictable state (a "gapped" region). In other parts, they are chaotic and fluctuating wildly (a "gapless" region).

For decades, scientists have struggled to map the calm parts of this landscape with perfect precision. They have powerful tools to approximate the state of the magnets, but these tools are like blurry photographs: they get the general picture right, but they miss the fine details.

This paper, by Imogen Camp and Nick G. Jones, introduces a new way to see the landscape clearly. They discovered a hidden "skeleton" running through the calm regions of specific quantum chains.

The "Skeleton" Analogy

Think of the phase diagram as a dense forest. Usually, to understand the forest, you have to guess what the trees look like based on blurry photos.

The authors found a network of special, perfectly clear paths running through this forest. These paths are the MPS Skeletons.

The Path: Along these specific paths, the quantum magnets settle into a state that can be described with perfect mathematical precision using a tool called a "Matrix-Product State" (MPS). It's like having a high-definition, 3D blueprint of the forest floor.
The Density: These paths are so numerous and closely spaced that you can never be more than a tiny step away from one. If you want to know what the forest looks like at a random spot, you can find a path right next to it that gives you an almost perfect answer.

From Simple to Complex

Previously, scientists only knew how to draw these perfect blueprints for "free-fermion" models. You can think of these as simple, non-interacting toys where each magnet acts independently.

This paper is a breakthrough because it extends this ability to interacting systems. In these systems, the magnets talk to each other; the state of one affects its neighbors. It's like moving from a room of people standing silently to a room where everyone is having a complex conversation. The authors show that even in this noisy, interacting world, there are still these hidden, perfectly calculable paths (the skeleton) where the conversation follows a strict, solvable pattern.

The "Onsager" Key

The specific type of quantum chain they studied is called the "chiral clock model." These models are special because they obey a set of mathematical rules known as the Onsager algebra.

The authors used this algebra like a master key. They showed that if you arrange the "ingredients" of the quantum chain (the coefficients in their equations) in a specific mathematical shape (a perfect square), the system unlocks a state that can be written down exactly.

The Recipe: They found that if you mix the ingredients in a specific way (mathematically, if a polynomial looks like a perfect square), you get a "ground state" (the lowest energy, most stable state) that is perfectly solvable.
The Excited States: They didn't just find the calmest state; they also found a set of "excited states" (slightly more energetic states) that are also perfectly solvable along these paths. This is like finding not just the floor of the building, but also the perfectly defined stairs leading up to the first floor.

What This Means for the Reader

Exact Answers, Not Guesses: For a huge class of interacting quantum systems, the authors can now write down the exact state of the system, rather than just approximating it.
A Map for the Future: Because these "skeleton" paths are so dense, they provide a powerful method to approximate the behavior of any system in these calm regions. If you want to know how a specific quantum chain behaves, you can find a "skeleton" path very close to it and use that exact solution as a nearly perfect estimate.
New Tools for Correlations: The paper also uses this method to calculate a specific property called the "disorder parameter" (a way to measure how disordered the system is). They found a clean, closed-form formula for this in these interacting systems, something that was previously only known for the simpler, non-interacting cases.

What They Did Not Do

It is important to stick to what the paper actually claims:

They did not apply this to real-world clinical uses or specific quantum computers yet.
They did not claim to solve the entire phase diagram; they specifically focused on the "gapped" (calm) regions surrounding certain fixed points.
They did not claim that every point in the landscape has an exact solution, only that the solutions are dense enough to approximate any point very well.

In short, the authors have built a set of "perfectly clear windows" into a complex quantum world. While the world outside the windows is still complicated, these windows are so numerous and close together that we can now see the whole picture with unprecedented clarity.

Technical Summary: Matrix-Product State Skeletons in Onsager-Integrable Quantum Chains

Problem Statement
While integrability provides analytical insights into special families of Hamiltonians and Matrix-Product States (MPS) offer powerful tools for approximating ground states in gapped 1D systems, the intersection of these two concepts remains under-explored, particularly beyond free-fermion models. Previous work established that for free-fermion chains (specifically the BDI and AIII classes), the set of Hamiltonians with exact MPS ground states forms a "dense skeleton" within the phase diagram. This skeleton allows for the analytic approximation of generic ground states via a sequence of exactly solvable models. However, for interacting systems, such as the $N$ -state Onsager-integrable chiral clock models ( $N > 2$ ), the existence and structure of such an MPS skeleton were unknown. The challenge lies in the fact that while Onsager-integrable models possess exact solutions via functional equations, their ground states are generally not exact MPS, and the phase diagrams for $N > 2$ contain extended gapless regions, unlike the $N=2$ case.

Methodology
The authors investigate the family of Hamiltonians $H_A[\{t_m\}, N] = \sum_m t_m A_m$ , where $A_m$ are generators of the Onsager algebra acting on $N$ -state chiral clock chains. The core methodology involves:

Algebraic Construction: Utilizing the Onsager algebra relations and pivot procedures (unitary transformations generated by $A_m$ ) to map generic Hamiltonians on the "skeleton" to fixed-point Hamiltonians ( $A_k$ ) with known ground states.
Polynomial Characterization: Defining a Laurent polynomial $f(z) = \sum t_m z^m$ associated with the Hamiltonian. The authors identify a subset of this family where $f(z) = \pm z^p g(z)^2$ for a polynomial $g(z)$ . This condition generalizes the $N=2$ free-fermion skeleton condition.
Iterative MPO Transformation: Constructing the exact eigenstate by applying a sequence of Matrix-Product Operators (MPOs), $M(k) = \exp(\mp \beta_k A_{p+k})$ , to the ground state of a fixed-point Hamiltonian ( $| \psi_p \rangle$ ). The parameters $\beta_k$ are derived from the coefficients of $g(z)$ using an algorithmic procedure (identical to the Schur-Cohn algorithm).
Spectral Analysis: Analyzing the spectrum of the transformed Hamiltonian to determine the regions where the constructed eigenstate is the true ground state. This involves examining the spectral gap and the behavior of zeros of $f(z)$ on the unit circle.

Key Contributions and Results

Existence of an MPS Skeleton for Interacting Models: The paper demonstrates that for $N > 2$ chiral clock models, Hamiltonians satisfying $f(z) = \pm z^p g(z)^2$ possess exact MPS ground states (or excited states) in gapped regions surrounding the fixed-point generators $A_k$ . This constitutes the first identification of an MPS skeleton in an interacting family of Onsager-integrable models.
Exact Eigenstate Construction (Result 1): For any Hamiltonian in the family where $f(z) = \pm z^p g(z)^2$ , the authors construct an exact eigenstate $|\phi\rangle = M(d)\dots M(1) |\psi_p^\pm\rangle$ . This state is an exact MPS with finite bond dimension, provided the coefficients derived from $g(z)$ satisfy $|b_k| \neq 1$ .
Ground State Identification (Result 2): The constructed eigenstate is proven to be the exact ground state of the Hamiltonian within the set $S$ , defined as the union of gapped regions connected to the fixed-point Hamiltonians $A_k$ via gapped paths. Outside these regions (in gapless sectors), the state remains an eigenstate but is no longer the ground state.
Density of the Skeleton (Result 3): The set of Hamiltonians on the skeleton is dense within the gapped regions $S$ . For any generic Hamiltonian in $S$ , its ground state energy density can be approximated arbitrarily well by a sequence of Hamiltonians on the skeleton. This provides an analytic approach to approximating ground states in interacting Onsager-integrable chains.
Excited States (Result 4): The authors identify a set of exact MPS excited states corresponding to low-lying states in non-zero momentum sectors. These are constructed similarly to the ground state but act on specific excited sectors of the fixed-point Hamiltonians.
Disorder Parameter Calculation: As an application, the authors derive a closed-form expression for the disorder parameter in a specific family of interacting models ( $H = A_0 + 2aA_1 + a^2A_2$ ) on the skeleton, generalizing known results from the $N=2$ Ising case to arbitrary even $N$ .

Significance and Claims
The paper claims to partially "expose" the MPS skeleton in interacting Onsager-integrable systems, extending the concept from free-fermion classes to interacting chiral clock models. The significance lies in:

Analytic Approximation: Providing a method to approximate ground states in gapped regions of interacting models using a dense set of exactly solvable MPS states, bypassing the need for numerical MPS optimization or Bethe ansatz diagonalization for these specific regions.
Algebraic Generality: Highlighting that many results rely primarily on the algebraic structure of the Onsager algebra rather than the specific representation, suggesting potential applicability to other Onsager-integrable Hamiltonians.
New Perspective on Eigenvectors: Offering a direct construction of MPS eigenstates that is largely decoupled from the complex functional equations typically used to solve these models, providing a new perspective on the eigenvector structure of interacting chiral clock chains.

The authors remain modest regarding the scope, noting that the condition $f(z) = \pm z^p g(z)^2$ is sufficient but not necessarily necessary for exact MPS ground states, and that the construction excludes a measure-zero set of points where the gap closes or coefficients diverge. They also note that while the skeleton is dense in the gapped regions $S$ , the phase boundaries themselves (particularly the gapless regions) are not fully characterized by this method.