---
title: Spin-SPT Phases Overview
url: https://www.emergentmind.com/topics/spin-spt-phases
type: topic
---

# Spin-SPT Phases Overview

A symmetry-protected topological (SPT) phase in a quantum spin system is a gapped, short-range-entangled phase of matter that is nontrivial only in the presence of a protecting symmetry. Spin-SPT phases are realized in integer or half-integer spin chain and ladder systems, as well as in higher-dimensional quantum magnets, and are classified by projective representations of the symmetry group. These phases are sharply distinct from both symmetry-breaking phases (which have a degenerate manifold and local order parameter) and from long-range entangled (intrinsically topological) phases that do not require symmetry protection. Distinguished by robust edge states and nonlocal string order, spin-SPT phases exemplify a framework in which group cohomology labels a complete set of 1D gapped quantum phases under on-site and antiunitary (time-reversal) symmetries, as well as with point-group symmetries and their combinations [1103.3323, 2110.04671, 1105.6021].

## 1. Cohomological Classification of 1D Spin-SPT Phases

The modern classification of 1D spin-SPT phases associates each phase with an equivalence class of 2-cocycles in the group cohomology group $H^2(G,U(1))$, where $G$ is the on-site symmetry group (possibly including time reversal $\mathcal{T}$ or parity $P$). For a gapped, $G$-invariant Hamiltonian, its ground state can be mapped to a matrix product state (MPS), whose virtual (edge) degrees of freedom inherit a projective representation $R(g)$:
\[
R(g)\,R(h) = \omega(g,h)\,R(gh), \qquad \omega(g,h) \in U(1)
\]
with associativity enforcing the 2-cocycle condition:
\[
\omega(g,h)\,\omega(gh,k) = \omega(g,hk)\,\omega(h,k)
\]
By identifying two cocycles related by a coboundary, the cohomology group $H^2(G,U(1))$ emerges [1103.3323, 2110.04671]. Each element $[\omega]\in H^2(G,U(1))$ labels a distinct SPT phase protected by $G$.

For paradigmatic cases:
- $G=Z_2\times Z_2$ (e.g., $\pi$ rotations of spin-1 about $x$ and $z$): $H^2(Z_2\times Z_2,U(1))=\mathbb{Z}_2$ (the AKLT/Haldane phase),
- $G=Z_2^{\mathcal{T}}$ (time-reversal alone): $H^2(Z_2^\mathcal{T},U(1))=\mathbb{Z}_2$,
- $G=SO(3)$: $H^2(SO(3),U(1))=\mathbb{Z}_2$ (spin-$1$ Haldane vs. trivial),
- General $G$: $H^2(G,U(1))$ may be a product of many $\mathbb{Z}_2$ factors, with antiunitary and parity symmetries yielding additional invariants [1103.3323, 2110.04671, 1105.6021].

## 2. Physical Characterization: Edge States, String Order, and SPT Invariants

The hallmark of a nontrivial spin-SPT phase is the emergence of robust, symmetry-protected edge states in open boundary conditions:
- Nontrivial projective representations at the ends (e.g., spin-$1/2$ Kramers doublet under $\mathbb{Z}_2\times\mathbb{Z}_2$ in the Haldane chain; projective action of $G$ on boundary modes).
- These edge degeneracies resist any symmetric, gapped boundary perturbation; breaking the protecting symmetry, or closing the gap, is required to remove the degeneracy [2110.04671].

A powerful nonlocal diagnostic is the string order parameter. For instance, in the Haldane phase ($G=\mathbb{Z}_2\times\mathbb{Z}_2$):
\[
O^Z_{\rm string}(j,k)=\langle S^z_j \exp[i\pi\sum_{l=j+1}^{k-1} S^z_l] S^z_k\rangle
\]
is nonzero in the SPT phase and vanishes in trivial (or symmetry-breaking) phases [1103.3323].

In SU(2)- or SO(3)-symmetric chains, the distinction between the SPT (Haldane) phase and the trivial phase is also encoded in the "multiplet" structure of the entanglement spectrum: Haldane phases yield half-integer virtual bonds, while trivial phases restrict to integer representations [1306.5671]. The "entanglement gap" $G=S^I-S^H$ (difference between integer- and half-integer-bond entropies) is quantized: $G=+1$ in SPT, $G=-1$ in the trivial phase.

## 3. Symmetry Scenarios and Their SPT Classification

The protection of a spin-SPT phase critically depends on the symmetry group $G$.

### On-site Unitary ($SO(3)$, $SU(2)$, $Z_2\times Z_2$):

- $H^2$ classification applies as above.
- Haldane phase: $SO(3)$-protected, projective end states, nontrivial entanglement structure.

### Time-Reversal and Inversion:

- For antiunitary $T$ or spatial symmetries (reflection, inversion), the group structure is enriched, and the cohomology must be "twisted" to accommodate complex conjugation [1103.3323, 2110.04671].
- For example, $H^2(D_2+T, U(1))=Z_2^4$ gives 16 classes in $S=1$ chains with $D_2$ (three orthogonal spin-inversion axes) and time-reversal [1105.6021].

### Ladders and Higher Spins:

- In $SU(n)$- or $SO(n)$-symmetric chains, SPT phases exhibit a richer structure; for even $n$, an SPT$^2$ phase exists with two exactly degenerate entangled MPS states (exponentially split in finite size) [2504.10919].
- For $S>1$ (such as spin-2 chains), effective spin-1 SPT order arises as an intermediate phase, classified by projective representations under a reduced $Z_2\times Z_2 + T$ symmetry [1412.3370, 1511.06338].

## 4. Explicit Models, Hamiltonians, and Realizations

Canonical models realizing spin-SPT phases include:

| Phase           | Minimal Model (Hamiltonian)                  | Edge States                    | Key Symmetry Group            |
|-----------------|----------------------------------------------|-------------------------------|-------------------------------|
| Haldane (AKLT)  | $H_{\rm AKLT} = \sum_j \mathbf{S}_j \cdot \mathbf{S}_{j+1} + \frac{1}{3} (\mathbf{S}_j \cdot \mathbf{S}_{j+1})^2$ | Spin-½ doublet                | $SO(3)$, $\mathbb{Z}_2\times\mathbb{Z}_2$, $\mathcal{T}$ |
| $D_2\times\mathbb{Z}_2^T$ SPT | Specialized S=1 ladder models [1105.6021] | Various, including quadruplets| $D_2$ (x,y,z 180° rotations), $\mathcal{T}$ |
| SPT$^2$ (SO(4), even-$n$)   | SO(4) bilinear–biquadratic chain [2504.10919]          | Exponential doublet splitting | $SO(4)$                      |
| Parity-broken dimer SPTs    | Zigzag XXZ chain, e.g. $J_1$, $J_2$, $\Delta$ [1406.5382, 1810.00325] | Kramers doublets              | $D_2$, inversion, $\mathcal{T}$ |

Characteristic parent Hamiltonians are often built as sums of two-site projectors or deformations of the AKLT model, with anisotropies or multi-spin couplings selecting among different SPT classes [1101.5680, 1204.5162, 1503.05835].

## 5. Phase Diagrams and Transitions Between SPT Phases

Spin-SPT phase diagrams often exhibit transitions between distinct SPT classes, trivial (product) states, and symmetry-breaking phases:

- For frustrated spin-½ chains and ladders, both SPT–SPT direct transitions (via Gaussian critical lines with $c=1$, as in [1406.5382]), first-order transitions, and symmetry-breaking intermediate phases may occur [1503.05835].
- In higher-spin chains (e.g., spin-2), effective spin-1 SPT phases emerge as intermediate phases and display continuous critical lines with $c=2$ (Majorana fermion field theory), or $c=1$ with sine-Gordon bosons [1511.06338, 1412.3370].
- In two-leg spin ladders with $D_2\times\sigma$ symmetry, a full $\mathbb{Z}_2^3$ classification is manifest, and transitions are generically first-order or go via a trivial intermediate phase when only discrete symmetries are present [1503.05835].
- The presence of continuous or emergent symmetries at criticality is required for continuous SPT–SPT transitions; otherwise, transitions generally fragment via first-order jumps or intervening trivial/symmetry-breaking phases.

## 6. Experimental and Numerical Probes of Spin-SPT Phases

Experimental signatures include:
- Edge spectroscopy detecting protected end multiplets (e.g., via ESR/optical/fluorescence methods).
- Low-temperature Curie tails in susceptibility for specific field directions (depending on the projective class) [1101.5680].
- Magnetization or string-order measurements via quantum gas microscopes or microwave absorption [1810.00325].
- Entanglement spectrum degeneracy and nonlocal correlators (string order, dimer string) in numerical DMRG, iTEBD, and MPS simulations [1306.5671, 1406.5382, 1503.05835, 1412.3370].

Numerical tensor-network techniques provide access to:
- Direct computation of topological invariants (cohomology labels) from Schmidt decompositions.
- Extraction of entanglement entropy scaling and identification of central charge at critical transitions.
- Mapping of phase diagrams and location of triple points, as in realistic spin-1 models [2406.19372].

## 7. Generalizations and Connections

- The general group-cohomology approach extends to arbitrary finite $G$ and higher dimensions: $H^2(G,U(1))$ classifies 1D SPT, $H^3(G,U(1))$ classifies 2D SPT [2110.04671, 1205.7024].
- Periodic driving can stroboscopically realize and detect SPT order; for instance, high-frequency Floquet systems engineer effective SPT Hamiltonians with emergent symmetry [1503.07871].
- SPT phases are robust computational resource states for measurement-based quantum computation; the Lie group of executable logical gates is determined by the cohomology class itself and is uniform throughout an SPT phase [1611.08053].
- In quasi-1D or high-$n$ symmetry systems (e.g., SO($n$)), rich even/odd structures in SPT phase diagrams and ground-state degeneracies are firmly established and quantifiable with modern DMRG extensions [2504.10919].

In summary, spin-SPT phases constitute an extensive and rigorously classified family of gapped, short-range-entangled phases in quantum spin systems, characterized by projective representations and group cohomology invariants, with robust and quantifiable experimental and computational signatures. The theory is foundational for understanding topological phases of quantum matter, their transitions, and their implications for computation, protected transport, and quantum information [1103.3323, 2110.04671, 1101.5680, 1306.5671, 2504.10919, 1406.5382, 1503.05835, 1105.6021].

Source: https://www.emergentmind.com/topics/spin-spt-phases