Subsystem-Protected Topological Orders (SSPTs)

• Subsystem-protected topological orders (SSPTs) are quantum phases defined by robust, symmetry-protected phenomena arising from lower-dimensional subsystem symmetries.
• They exhibit distinct bulk, boundary, and entanglement signatures analyzed via cohomological classifications and anomaly indicators.
• Research in SSPTs reveals dualities with fracton orders and has led to innovative lattice models demonstrating nontrivial topological properties.

A subsystem-protected topological order (SSPT) is a phase of quantum matter in which robust, symmetry-protected phenomena arise not from a global symmetry but from a symmetry acting independently on lower-dimensional rigid subsystems—such as lines or planes—of the many-body system. Unlike conventional global SPT phases, which are protected by the invariance of the full system under a single group, SSPTs are protected by an extensive set of subsystem symmetries, dramatically enriching their boundary, bulk, and entanglement structures. The study of SSPTs has led to paradigmatic models in two and higher dimensions, a robust cohomological classification for strong 2D orders, a systematic understanding of their diagnostics, and deep connections to fracton topological orders.

1. Subsystem Symmetries and the Definition of SSPT Phases

Subsystem symmetries are defined by their action on lower-dimensional rigid submanifolds of a lattice. In a 2D square lattice with onsite abelian symmetry group $G_s$, the line-like subsystem symmetry is generated by

$$S^v_x(g) = \prod_{y=-\infty}^{+\infty} u_{xy}(g), \qquad S^h_y(g) = \prod_{x=-\infty}^{+\infty} u_{xy}(g)$$

for every $x$ or $y$ and each $g \in G_s$; $u_{xy}$ is a faithful representation at site $(x, y)$. These symmetries act as products of onsite transformations applied rigidly along rows or columns, forming a symmetry group whose size grows linearly with system size.

A two-dimensional SSPT phase is a gapped, short-range entangled state invariant under such subsystem symmetries, and which cannot be transformed to a trivial product state by any finite-depth circuit of local unitaries preserving the symmetries. A crucial distinction emerges between:

• Weak SSPTs: Phases that can be built from decoupled arrays of 1D SPT chains—each protected by the associated global symmetry—with minimal or trivial 2D entanglement structure.
• Strong SSPTs: Phases that exhibit genuinely 2D topological features, not present in any stack or product of 1D SPTs, and which cannot be disentangled by linearly-symmetric local unitaries (LSLUs). Strong SSPTs manifest nontrivial boundary phenomena and unique bulk invariants (Devakul et al., 2018, Tantivasadakarn et al., 2019).

This strong/weak dichotomy extends to higher dimensions and to subsystems of various co-dimensions (lines, planes, fractals).

2. Cohomological Classification of Strong SSPT Orders

Strong 2D SSPT phases with abelian onsite symmetry $G_s$ are classified by the cohomological group

$$\mathcal{C}[G_s] = \frac{H^2[G_s^2, U(1)]}{(H^2[G_s, U(1)])^3}$$

where $H^2[G, U(1)]$ is the standard group cohomology (Devakul et al., 2018, Jia et al., 28 May 2025). The numerator $$S^v_x(g) = \prod_{y=-\infty}^{+\infty} u_{xy}(g), \qquad S^h_y(g) = \prod_{x=-\infty}^{+\infty} u_{xy}(g)$$0 classifies projective representations on the edge under $$S^v_x(g) = \prod_{y=-\infty}^{+\infty} u_{xy}(g), \qquad S^h_y(g) = \prod_{x=-\infty}^{+\infty} u_{xy}(g)$$1 symmetry acting separately on left/right boundaries, while the denominator modds out equivalences generated by stacking 1D SPT chains on either side or along the diagonal subgroup. This quotient precisely captures "intrinsic" or "strong" 2D SSPT data, eliminating all weak (stackable) contributions.

Explicitly:

• For $$S^v_x(g) = \prod_{y=-\infty}^{+\infty} u_{xy}(g), \qquad S^h_y(g) = \prod_{x=-\infty}^{+\infty} u_{xy}(g)$$2, $$S^v_x(g) = \prod_{y=-\infty}^{+\infty} u_{xy}(g), \qquad S^h_y(g) = \prod_{x=-\infty}^{+\infty} u_{xy}(g)$$3, $$S^v_x(g) = \prod_{y=-\infty}^{+\infty} u_{xy}(g), \qquad S^h_y(g) = \prod_{x=-\infty}^{+\infty} u_{xy}(g)$$4, so $$S^v_x(g) = \prod_{y=-\infty}^{+\infty} u_{xy}(g), \qquad S^h_y(g) = \prod_{x=-\infty}^{+\infty} u_{xy}(g)$$5.
• For $$S^v_x(g) = \prod_{y=-\infty}^{+\infty} u_{xy}(g), \qquad S^h_y(g) = \prod_{x=-\infty}^{+\infty} u_{xy}(g)$$6, $$S^v_x(g) = \prod_{y=-\infty}^{+\infty} u_{xy}(g), \qquad S^h_y(g) = \prod_{x=-\infty}^{+\infty} u_{xy}(g)$$7, $$S^v_x(g) = \prod_{y=-\infty}^{+\infty} u_{xy}(g), \qquad S^h_y(g) = \prod_{x=-\infty}^{+\infty} u_{xy}(g)$$8 giving $$S^v_x(g) = \prod_{y=-\infty}^{+\infty} u_{xy}(g), \qquad S^h_y(g) = \prod_{x=-\infty}^{+\infty} u_{xy}(g)$$9 (Devakul et al., 2018, Jia et al., 28 May 2025).

The group structure is derived from the pattern of projective commutation phases encoded at the boundaries and is realized by explicit lattice cluster-state models.

3. Prototypical Models and Lattice Realizations

The canonical example of a strong 2D SSPT is the cluster state on a square lattice with two qubits per site: $x$0 This model, protected by $x$1 symmetry acting along each line, realizes the nontrivial cohomology class $x$2 (Devakul et al., 2018, Jia et al., 28 May 2025). Similar construction applies for general $x$3 using Pauli or Weyl operators.

In 3D, planar and fractal subsystem symmetries yield higher-dimensional cluster-type models (e.g., X-cube models, Haah's code) as "ungaugings" or duals of SSPT parents (Tantivasadakarn et al., 2019, Devakul et al., 2019, Schmitz, 2019). These provide a connection between subsystem-protected phases and fracton orders.

4. Bulk, Boundary, and Entanglement Phenomena

Strong SSPTs display distinctive signatures in their edge and entanglement structure:

• Boundary degeneracy: Open boundaries host projective representations of the subsystem symmetry group that cannot be trivialized locally, resulting in an extensive, symmetry-protected degeneracy—typically exponential in boundary length (1803.02369, Devakul et al., 2018, Tantivasadakarn et al., 2019).
• Spurious topological entanglement entropy (TEE): On a cylinder or region bipartition, the entanglement entropy satisfies

$x$4

where $x$5 is nonzero only for strong SSPTs, even though the bulk is short-range entangled. For a 2D cluster state, $x$6 is universal throughout the phase (Devakul et al., 2018, Stephen et al., 2019). This is robust under any perturbation respecting the subsystem symmetry.

• Entanglement spectrum: Exact degeneracy or gapless structures in the entanglement Hamiltonian are directly determined by the mixed projective anomaly at the boundary (Ding et al., 2024). Such spectral features distinguish strong from weak and intrinsic SSPT phases.

5. Detection, Diagnostics, and Response Theory

Physical and computational diagnostics of SSPT order utilize a combination of entanglement, correlation, and anomaly indicators:

• LSLU invariants: Nonzero values of the bulk invariant $x$7, not removable by any linearly-symmetric local unitaries, signal strong order (Devakul et al., 2018).
• Strange correlators: The overlap between a trivial product state and the SSPT ground state with operator insertions exhibits long-range order (or plateau) only in the presence of nontrivial SSPT order (Zhou et al., 2022). This provides a bulk detection method, accessible via quantum Monte Carlo even under periodic boundaries.
• Anomaly indicators: Exact mixed 't Hooft anomalies between adjacent subsystem symmetries, detectable through PEPS transfer matrix eigenvalues, distinguish weak, strong, and "intrinsic" SSPT phases (Ding et al., 2024). In mixed-state or disordered scenarios, these anomalies persist and remain well-defined.

A summary table of representative diagnostics is given below:

Diagnostic	Weak SSPT	Strong SSPT	Intrinsic SSPT
LSLU invariant β	1 (triv)	≠1 (nontriv)	may vanish
Spurious TEE γ	0	>0	variable
Strange correlator	decaying	plateau/long	pattern-dependent
Boundary anomaly	absent	present	mixed only

6. Higher-Dimensional and Generalized SSPT Orders

Subsystem symmetries can protect SPT order in $x$8 spatial dimensions, involving planar (e.g., 3D cubic lattice with $x$9-plane subsystem symmetry), fractal, or higher-form symmetries (Devakul et al., 2019, Tantivasadakarn et al., 2019, Schmitz, 2019). Classification in 3D uses generalizations of the 2D cohomological quotient, with strong 3D SSPTs characterized by invariants not possible in stacked lower-dimensional systems.

Moreover, HOCA-generated SSPT phases realize subsystem-protected order protected by line, plane, fractal, or even chaotic subsystems (Zhang et al., 2023). These can be distinguished by multi-point strange correlators and are captured by new algebraic and geometric invariants.

Subsystem-protected topology admits non-invertible symmetry actions, higher-codimension protected modes, and coexisting topological and symmetry-breaking orders (Furukawa, 16 May 2025, Mana et al., 3 Mar 2026, Kim et al., 2021). Symmetry topological field theory (SymTFT) has recently been developed to give a holographic (bulk-boundary) description of subsystem Symmetry-Protected Topological phases and recover the full cohomological classification (Jia et al., 28 May 2025).

7. Duality and Relation to Fracton Topological Orders

There is a systematic duality between strong SSPTs and fracton topological orders:

• Ungauging (SSPT to fracton): Gauging the subsystem symmetry in a strong SSPT produces a dual fracton order, with logical operators and subextensive ground state degeneracy reflecting the underlying SSPT data (Tantivasadakarn et al., 2019, Schmitz, 2019, Devakul et al., 2019, Schmitz, 2019).
• Distillation (fracton from SSPT layers): Type-II fracton codes such as Haah's cubic code can be obtained as "distilled" long-range entangled phases from stacks of 2D SSPT layers (Schmitz, 2019).
• Classification link: The classification of strong fracton codes exactly matches that of strong SSPT parents under this duality, up to lattice translation breaking (Tantivasadakarn et al., 2019, Devakul et al., 2019).

Subsystem SymTFT gives a field-theoretic realization of the boundary-bulk correspondence and the subtle structure of foliation and anomaly in these phases (Jia et al., 28 May 2025).

---

The study of subsystem-protected topological orders has led to a comprehensive understanding of the classification, model realization, diagnostics, bulk-boundary correspondence, and dualities with fracton orders, with symmetry topological field theory providing a unified bulk characterization (Devakul et al., 2018, Tantivasadakarn et al., 2019, Jia et al., 28 May 2025, Devakul et al., 2019, Ding et al., 2024, 1803.02369). These results continue to drive progress in quantum information, condensed matter physics, and the general theory of quantum phases beyond conventional SPT and topological order frameworks.