⚛️ high-energy theory

Matrix product state classification of 1D multipole symmetry protected topological phases

This paper systematically classifies one-dimensional bosonic symmetry-protected topological phases protected by spatially modulated multipole symmetries using matrix product states, revealing that the classification for $r$ -pole symmetries is determined by distinct components of second group cohomology groups encoding boundary projective representations.

Original authors: Takuma Saito, Weiguang Cao, Bo Han, Hiromi Ebisu

Published 2026-01-15

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Takuma Saito, Weiguang Cao, Bo Han, Hiromi Ebisu

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 building a long chain of Lego bricks. In the world of physics, these bricks are tiny particles, and the way they snap together determines the "phase" of matter—whether it's a solid, a liquid, or something stranger.

For a long time, physicists had a standard rulebook for these phases. But recently, they discovered a new, exotic type of Lego chain where the rules change depending on where you are in the chain. This is what the paper calls "spatially modulated symmetry."

Here is a simple breakdown of what the authors, Takuma Saito and his team, did in this paper:

1. The New Rulebook: "Multipole" Symmetries

Usually, if you have a symmetry (like rotating a shape and it looks the same), it applies everywhere equally. Think of a spinning top; it looks the same no matter where you look at it.

But in this paper, the authors look at a special kind of symmetry called multipole symmetry.

The Analogy: Imagine a row of people holding hands.
- Normal Symmetry (Monopole): Everyone holds hands with the same strength. If you swap two people, the chain looks the same.
- Dipole Symmetry: The strength of the handshake depends on where you are. Maybe the people at the start hold hands tightly, and the people at the end hold hands loosely, but the pattern of tightness and looseness is preserved.
- Quadrupole and Higher: The pattern gets even more complex. The "rules" of how the particles interact change based on their position in the chain (like a mathematical curve drawn over the chain).

The authors focus on these "position-dependent" rules, specifically looking at 1D chains (one-dimensional lines of particles).

2. The "Black Box" Problem: Matrix Product States

To understand these complex chains, the authors used a mathematical tool called Matrix Product States (MPS).

The Analogy: Imagine you have a very long, complex machine (the quantum chain), but you can only see the outside. You can't see the gears inside. However, you know that if you push a button on the left end, something specific happens on the right end.
The authors used MPS to act like a "decoder ring." It allows them to look at the "gears" inside the machine (the mathematical structure of the chain) without having to simulate the whole thing. They used this to figure out how these weird, position-dependent symmetries behave.

3. The Secret at the Ends: Edge States

The most exciting part of their discovery is what happens at the ends of the chain.

The Analogy: Think of a long, quiet hallway. In the middle of the hallway, everything is calm and follows the rules. But at the very ends of the hallway, the rules get weird. The "walls" at the ends start dancing to a different beat.
In physics, these are called edge states. The authors found that when you apply these special "multipole" symmetries, the ends of the chain don't just sit there; they form a special, "projective" relationship. It's like the two ends of the chain are holding hands in a secret code that the middle of the chain doesn't know about.

4. The Classification: Sorting the Exotic Phases

The main goal of the paper was to classify these phases.

The Analogy: Imagine you have a huge box of different types of alien toys. Some are red, some are blue, some have wheels, some have wings. The authors wanted to create a filing system to sort them all.
They found that you can sort these phases based on a mathematical concept called Group Cohomology.
- Think of this as a "fingerprint" for the phase.
- They discovered that for a chain with a "rank-r" symmetry (where $r=0$ is normal, $r=1$ is dipole, $r=2$ is quadrupole, etc.), the "fingerprint" is determined by specific parts of a mathematical formula.
- The Result: They created a formula (Equation 3.10 and 3.11 in the paper) that tells you exactly how many different types of these exotic phases exist for any given symmetry. It's like saying, "If you have a quadrupole symmetry, there are exactly this many unique ways the ends of the chain can dance."

5. Building the Models

Finally, the authors didn't just do the math; they showed how to actually build these phases in a theoretical "lab."

They constructed specific lattice models (mathematical blueprints for chains of particles) that realize these phases.
They showed three specific examples for "quadrupole" symmetry (rank 2):
1. Monopole-Quadrupole: A mix of simple and complex rules.
2. Dipole-Quadrupole: A mix of medium and complex rules.
3. Quadrupole-Quadrupole: A mix of complex and complex rules.
Each of these blueprints creates a unique "dance" at the ends of the chain, proving that their classification system works.

Summary

In short, this paper is a cataloging project for a new type of quantum matter.

They looked at chains where the rules change based on position (multipole symmetry).
They used a mathematical tool (MPS) to peek inside the chain.
They discovered that the ends of these chains have secret, unique behaviors (edge states).
They created a mathematical "filing system" to count and sort all the possible unique behaviors.
They built theoretical models to prove these behaviors actually exist.

They didn't invent a new material to build a phone or a battery; instead, they provided the theoretical map that future scientists can use to understand and potentially find these strange, exotic phases of matter in the real world.

Technical Summary: Matrix Product State Classification of 1D Multipole Symmetry Protected Topological Phases

1. Problem Statement

The classification of phases of matter beyond the Landau-Ginzburg paradigm has been a central theme in condensed matter physics, particularly regarding Symmetry Protected Topological (SPT) phases. While SPT phases protected by global symmetries are well-understood via group cohomology, the landscape of phases protected by spatially modulated symmetries remains largely unexplored. These symmetries, where the action on local degrees of freedom depends on position, were originally proposed in the context of fracton phases.

Specifically, this work addresses the systematic classification of one-dimensional (1D) bosonic SPT phases protected by multipole symmetries. Multipole symmetries involve the conservation of multipoles (e.g., dipole, quadrupole) and impose substantial constraints on system dynamics, leading to phenomena such as Hilbert space fragmentation. While explicit lattice models for dipole SPTs in (1+1)d and defect network constructions in (2+1)d have been proposed, a systematic classification of SPT phases with general Abelian multipole symmetries (specifically rank- $r$ ) in 1D, including their lattice model realizations, has not been fully established.

2. Methodology

The authors employ the Matrix Product State (MPS) formalism to construct and classify these phases. The methodology proceeds through the following steps:

Definition of Symmetries: The authors define rank- $r$ multipole symmetry operators $U^{(r)}_g$ for a finite Abelian group $G$ acting on a spin chain $\Lambda$ . The operator is defined as:
$U^{(r)}_g := \prod_{x \in \Lambda} (g_x)^{\binom{x+r-1}{r}}$
where the exponent is an integer-valued polynomial. The presence of translation symmetry $T$ implies that a system with rank- $r$ multipole symmetry also possesses all lower-rank multipole symmetries ( $0 \le k \le r$ ).
Bulk Symmetry Action: Using the fundamental theorem of MPS, the authors analyze how symmetry operators act on the MPS tensors $A$ in the bulk. For a unique gapped ground state, the action of the global symmetry induces unitary operators $X^{(k)}_g$ on the virtual (bond) space. Through an inductive argument, they establish that the action of a rank- $(k+1)$ symmetry operator on the MPS tensor relates to the rank- $k$ virtual operators via:
$X^{(k)\dagger}_g A(x) \doteq X^{(k+1)\dagger}_g A(x) X^{(k+1)}_g$
where $\doteq$ denotes equality up to a $U(1)$ phase.
Boundary Analysis and Projective Representations: By introducing open boundaries, the authors investigate the action of symmetry operators on the edge modes. The symmetry operators $U^{(k)}_g$ induce projective representations on the boundaries, characterized by unitary operators $X^{(k)}_g$ . The classification relies on analyzing the commutation relations of these boundary operators:
$X^{(k)}_g X^{(l)}_h = e^{i\theta_{g,h}(k,l)} X^{(l)}_h X^{(k)}_g$
The phase factors $\theta_{g,h}(k,l)$ must satisfy consistency conditions derived from the commutativity of the global symmetry operators on the entire chain.
Cohomological Classification: The constraints on the phase factors $\theta_{g,h}(k,l)$ are analyzed using group cohomology. The authors derive specific relations (Proposition 1) that reduce the number of independent phases. They utilize the Künneth formula to decompose the second cohomology group of the product group $G^{\times(r+1)}$ and identify the distinct topological invariants that label the SPT phases.

3. Key Contributions and Results

A. Systematic Classification Formula

The primary result is a systematic classification of rank- $r$ multipole SPT phases in 1D. The distinct phases are labeled by the quotient of cohomology groups, as summarized in Equation (3.10):

$C^{(r)}(G) \cong \begin{cases} \left[ H^2(G, U(1)) \oplus \frac{H^1(G, H^1(G, U(1)))}{H^2(G, U(1))} \right]^{\oplus m} & (r = 2m - 1) \\ [H^2(G, U(1))]^{\oplus (m+1)} \oplus \left[ \frac{H^1(G, H^1(G, U(1)))}{H^2(G, U(1))} \right]^{\oplus m} & (r = 2m) \end{cases}$

Alternatively, this can be expressed recursively via Equation (3.11):
$C^{(r)}(G) = \frac{H^2(G^{\times(r+1)}, U(1))}{[C^{(0)}(G)] \oplus 2[C^{(1)}(G)] \oplus \cdots \oplus 2[C^{(r-1)}(G)]}$

This formulation indicates that the classification of rank- $r$ phases can be inductively derived from lower-rank classifications ( $r=0, 1, \dots, r-1$ ).

For $r=0$ (global symmetry), the result recovers the standard $H^2(G, U(1))$ classification.
For $r=1$ (dipole symmetry), the result aligns with previous findings in [62].

B. Explicit Lattice Realizations

The authors provide concrete lattice model constructions (stabilizer Hamiltonians) for the case of quadrupole symmetry ( $r=2$ ) with $G = \mathbb{Z}_N \times \mathbb{Z}_M$ . They demonstrate three distinct classes of SPT phases, each corresponding to a specific component of the classification:

Monopole-Quadrupole SPTs: Characterized by the commutation between charge and quadrupole symmetries, classified by $H^2(G, U(1))$ .
Dipole-Quadrupole SPTs: Characterized by the commutation between dipole and quadrupole symmetries, classified by the quotient term $\frac{H^1(G, H^1(G, U(1)))}{H^2(G, U(1))}$ .
Quadrupole-Quadrupole SPTs: Characterized by the commutation between two quadrupole symmetries, classified by $H^2(G, U(1))$ .

These models utilize clock and shift operators with specific decorations to realize the required projective representations at the boundaries.

4. Significance and Claims

The paper claims to establish the first systematic classification of 1D bosonic SPT phases protected by Abelian multipole symmetries. By utilizing the MPS formalism, the authors explicitly link the boundary projective representations to the bulk topological order, identifying the distinct phases via group cohomology.

The authors state that their results:

Deepen the understanding of the infrastructure of modulated symmetries.
Provide a foundation for classifying topological phases protected by other types of modulated symmetries.
Offer a recursive framework (Eq. 3.11) that connects higher-rank multipole SPTs to lower-rank ones.

The work is presented as a foundational step, noting that while they focus on Abelian groups and 1D systems, the approach suggests pathways for generalization to higher dimensions and non-Abelian or non-invertible symmetries, though these are left for future work. The authors explicitly defer the treatment of twisted boundary conditions and bulk-boundary correspondence on closed manifolds to separate studies.