Fermionic SPT Phases

Updated 17 June 2026

Fermionic SPT phases are short-range-entangled fermionic systems that exhibit symmetry-protected edge states and nonlocal entanglement.
They are classified using approaches such as group supercohomology, spin cobordism, and fixed-point wavefunctions that integrate fermion-parity and projective symmetry effects.
Key diagnostics include nonlocal order parameters, entanglement spectra, and emergent boundary supersymmetry, providing practical signatures for experimental detection.

Fermionic symmetry protected topological (FSPT) phases are short-range-entangled, symmetry-enforced quantum phases of interacting fermions, distinguished from trivial product states by the presence of protected edge states, nonlocal entanglement structure, and topological responses which are robust as long as certain symmetries are preserved. Unlike their bosonic SPT or free-fermion counterparts, FSPT phases require a careful consideration of the interplay between the fermion-parity superselection structure, projective symmetry realizations, and interactions. Over the past decade, the field has developed rigorous classification schemes—most notably, group supercohomology, spin-cobordism theory, and fixed-point wavefunction constructions via finite-depth symmetric local unitary circuits—that map out the allowed types of FSPT order in 1D, 2D, and 3D in terms of group cohomology and cobordism invariants. Many FSPTs manifest intrinsically interacting “beyond free-fermion” behaviors and exhibit signatures such as boundary degeneracies, supersymmetry in edge spectra, and exotic anyonic braiding in gauged versions of the theory.

1. Foundational Concepts and Classification Frameworks

Fermionic SPT phases are defined as gapped, short-range-entangled ground states of interacting fermion systems, distinguished up to finite-depth fermionic symmetric local unitary (FSLU) transformations that commute with a specified symmetry group $G_f$ . The central subgroup $\mathbb{Z}_2^f$ generated by fermion-parity $P_f$ is always present, so $G_f$ is a (possibly nontrivial) central extension of a bosonic group $G_b$ : $1 \to \mathbb{Z}_2^f \to G_f \to G_b \to 1$ classified by an extension cocycle $\omega_2\in H^2(G_b,\mathbb{Z}_2)$ , which sets important constraints for SPT realization (Wang et al., 2018).

Three major approaches have become central:

Generalized Group Cohomology (Supercohomology): Generalizes bosonic SPT classification by including both complex fermion decoration data ( $n$ -cochains) and bosonic SPT cocycles, subject to obstruction conditions and quotienting out trivializations (Wang et al., 2018, Wang et al., 2017).
Atiyah–Hirzebruch Spectral Sequence (AHSS) and Spin Cobordism: Classification by torsion in cobordism groups, unifying SPT order as generalized cohomology with respect to symmetry and spacetime structure (Spin/Pin/Pin $^\pm$ ) (Kapustin et al., 2014, Ren et al., 2023).
Fixed-Point Wavefunction and FSLU Circuits: Construct FSPT wavefunctions by domain-wall decoration (e.g., of Kitaev chains, complex fermions) and implement F-move local unitaries, organizing consistency and stacking laws according to the AHSS layers (Wang et al., 2018, Ren et al., 2023).

In 1D, the full classification with finite symmetry is given by a generalized Wall group law: $\mathbb{Z}_2 \times H^1(G, \mathbb{Z}_2) \times H^2(G,U(1)_\mathfrak{p})$ with uniquely twisted stacking (Bourne et al., 2020).

In higher dimensions, the group structure involves a spectral sequence built from $\mathbb{Z}_2^f$ 0 and further decorated with obstruction-free subgroups and trivializations (Ren et al., 2023, Wang et al., 2017).

2. One-Dimensional Fermionic SPT Phases

In 1D, FSPT phases are sharply characterized by the interplay between symmetry fractionalization (projective boundary representations) and fermion parity. For generic finite symmetry $\mathbb{Z}_2^f$ 1, phases are labeled by

$\mathbb{Z}_2^f$ 2

corresponding to “fermion decoration” and bosonic SPT layers (Niu et al., 2023).

Canonical models include the Kitaev Majorana chain ( $\mathbb{Z}_2^f$ 3 index), BDI-class chains with time-reversal ( $\mathbb{Z}_2^f$ 4 reduction due to interactions (Shapourian et al., 2016)), and interacting chains with nontrivial projective action classified by group cohomology (Tang et al., 2012, Montorsi et al., 2016). All possible phases can be constructed as irreducible fermionic matrix-product states (fMPS) labeled by $\mathbb{Z}_2^f$ 5 with twisted stacking laws (Bourne et al., 2020). Each phase is uniquely detected by “strange correlators” constructed from bulk wavefunctions (Niu et al., 2023), nonlocal string order parameters (Montorsi et al., 2016), and partial reflection/transpose invariants quantized by bulk geometry (Shapourian et al., 2016).

Emergent boundary supersymmetry is a universal property of all intrinsically fermionic SPTs with $\mathbb{Z}_2^f$ 6 nontrivial: the open-chain boundary acts as $\mathbb{Z}_2^f$ 7 supersymmetric quantum mechanics with Bose–Fermi paired degeneracy protected by the symmetry anomaly (Prakash et al., 2020, Prakash et al., 2020).

Table: 1D FSPT Classifications for Common $\mathbb{Z}_2^f$ 8

Symmetry $\mathbb{Z}_2^f$ 9	Classification Group	Key Signature
$P_f$ 0	$P_f$ 1	Majorana chain
$P_f$ 2	$P_f$ 3	Two indices (parity, projective)
BDI ( $P_f$ 4)	$P_f$ 5	$P_f$ 6 reduction

Physical detection can involve edge-state degeneracies, entanglement spectrum analysis, and bulk many-body invariants from crosscap partition functions or nonlocal operations (Shapourian et al., 2016, Niu et al., 2023).

3. Two-Dimensional Fermionic SPT Phases

Fermionic SPT phases in 2D are classified by cohomology involving complex fermion, Kitaev-chain, and bosonic decorations, subject to obstruction-free conditions derived from the symmetry extension structure and self-consistency of domain-wall stacking (Wang et al., 2018, Ren et al., 2023). For finite Abelian $P_f$ 7, a complete classification can be conjectured by gauging the symmetry, calculating multi-vortex braiding invariants, and solving constraint equations (Wang et al., 2016). The Abelian stacking group incorporates both free-fermion and bosonic SPT building blocks but admits “intrinsically interacting” FSPT sectors that cannot be realized by stacking such blocks (e.g., for $P_f$ 8) (Wang et al., 2016).

The $P_f$ 9-matrix-based Abelian Chern–Simons formalism allows explicit construction and diagnosis of root phases, including realization of gapless Majorana edge theories and edge Luttinger liquids with symmetry anomalies that cannot be symmetrically gapped (intrinsic ingappability) (Ning et al., 2019). Some SPTs feature protected gapless edge Luttinger liquids with nontrivial symmetry transformations unrealizable in free-fermion sectors (Sullivan et al., 2019).

Fermionic crystalline SPTs (with rotation, inversion, or reflection symmetries) are understood using the crystalline equivalence principle: a crystalline FSPT is mapped to an internal-symmetry FSPT with a modified extension structure. Such phases can exhibit protected Majorana modes at rotation centers or hinges, or feature nontrivial LSM-type constraints (Cheng et al., 2018, Ren et al., 2023).

4. Three-Dimensional Fermionic SPT Phases and Cobordism

In 3D, FSPT phases demand a hierarchical classification. Generalized group supercohomology yields two short exact sequences that incorporate bosonic $G_f$ 0, supercohomology $G_f$ 1 (decorated complex fermions), and Majorana-chain $G_f$ 2 layers, together with explicit obstruction equations arising from fermionic pentagon/hexagon consistency (e.g., $G_f$ 3 and the Majorana hexagon obstruction) (Wang et al., 2017, Wang et al., 2018). This structure is reflected in spin-cobordism groups: for example,

$G_f$ 4

where the torsion part classifies distinct SPTs and the free part encodes anomalous response terms (Kapustin et al., 2014). For $G_f$ 5, the familiar periodic table $G_f$ 6 classification of 3D class DIII free-fermion topological superconductors is reduced to $G_f$ 7 under interactions, as revealed by gravitational and CP (or reflection)-anomaly analysis of boundary conformal field theory partition functions (Hsieh et al., 2015, Shapourian et al., 2016, Wang et al., 2014).

The surface topological order in 3D FSPTs is constrained by bulk anomalies and can feature symmetry-enforced gaplessness, anomalous anyon content (e.g., $G_f$ 8, T–Pfaffian), and protected surface CFTs (Wang et al., 2014, Hsieh et al., 2015).

5. Intrinsically Interacting and Beyond-Free-Fermion FSPTs

A salient feature of fermionic SPT order is the existence of phases that cannot be constructed from free-fermion band insulators, nor from stacking with bosonic SPTs—even in interacting systems. For instance:

The “fermionic Haldane phase” in 1D spinless Hubbard ladders is a $G_f$ 9 SPT protected by $G_b$ 0 with edge spin-½ degrees of freedom and an entanglement spectrum signature of projective representations (Ning et al., 2014).
In 2D, the aforementioned “intrinsically interacting” phases exist for $G_b$ 1 and exhibit three-loop braiding invariants that cannot be realized in free-fermion systems or via BSPT embedding (Wang et al., 2016, Ning et al., 2019).
In 2D and 3D with crystalline symmetry, SPTs protected by $G_b$ 2 rotation plus internal symmetries require strongly interacting constructions, detected by block-state fusion and protected by nontrivial stacking and gluing constraints (Cheng et al., 2018).

These phases can be diagnosed by bulk many-body invariants (e.g., crosscap partition function phases), strange correlators, and symmetry-twisted ground-state properties, with no free-fermion or bosonic SPT realization (Montorsi et al., 2016, Shapourian et al., 2016, Niu et al., 2023).

6. Stacking Laws, Obstruction Theory, and Higher-Dimensional Generalizations

The stacking group structure of interacting FSPT phases is more intricate than the bosonic case, due to the pairing of complex-fermion and Kitaev-chain decorations and the necessity to preserve all obstruction and trivialization relations at each AHSS page. Explicit stacking laws have been derived up to 2+1D for both unitary and antiunitary symmetries, and in all 17 wallpaper groups for crystalline FSPTs (Ren et al., 2023). The stacking rules involve nontrivial cocycle additions, modifications by symmetry extension classes $G_b$ 3, and explicit phase factors (see Eqs. (4.20)–(4.22.2), (5.15.1)–(5.15.4) of (Ren et al., 2023)).

Decorated domain-wall constructions extend to higher dimensions, incorporating higher-degree cochains and membrane decorations, with obstruction-free conditions tracked by spectral sequences and cobordism. Future generalizations to arbitrary dimension demand a complete group supercohomology or spin-cobordism approach, which is conjectured to be exhaustive for $G_b$ 4 and predictive for yet unexplored high-dimensional SPT phenomena (Wang et al., 2017, Ren et al., 2023, Kapustin et al., 2014).

7. Physical Diagnostics, Signatures, and Experimental Implications

FSPT phases are characterized and detected by a set of robust signatures:

Boundary signatures: presence of protected zero modes, edge-state degeneracy, and anomalous symmetry realization (e.g., projective representation, boundary supersymmetry) (Prakash et al., 2020).
Nonlocal order: nonlocal string/Haldane parity correlators serve as bulk diagnostics of SPT order and track phase transitions (Montorsi et al., 2016).
Many-body invariants: partial reflection/transpose overlaps and crosscap partition functions yield quantized phases matching the SPT index (Shapourian et al., 2016, Niu et al., 2023).
Strange correlators: long-range or oscillatory strange correlations in wavefunction overlaps diagnose both fermionic and bosonic invariants (Niu et al., 2023).
Entanglement spectrum: Schmidt degeneracy pattern reveals the topological sector (Montorsi et al., 2016, Ning et al., 2014).
Bulk responses: gauging symmetries and analyzing resulting anyonic braiding properties (via the $G_b$ 5-matrix or statistical invariants) distinguish FSPT phases, especially in 2D (Wang et al., 2016, Ning et al., 2019).
Crystalline response: lattice models with reflection, rotation, or inversion symmetry realize protected corner, hinge, or center modes depending on central extension structure (Cheng et al., 2018, Chen et al., 2016).

Numerical and experimental platforms such as cold atom quantum simulators, mesoscopic wires, and interacting ladder systems have been proposed for the physical realization and measurement of these signatures (Montorsi et al., 2016, Ning et al., 2014, Chen et al., 2016).

In summary, fermionic symmetry protected topological phases form a vast landscape determined by group extensions, rich cohomological data, and nontrivial interaction effects. Their classification weaves together algebraic topology, quantum many-body physics, and symmetry analysis, providing unified principles that encompass and predict both “classical” free-fermion topological states and genuinely new interacting states of finite and crystalline symmetry. Current frontiers include the exhaustive classification in spatial dimensions $G_b$ 6, explicit construction of beyond-supercohomology FSPT phases, and the physical realization and measurement of their distinguishing topological signatures (Wang et al., 2017, Ren et al., 2023, Kapustin et al., 2014).