Spin models with critical ground space degeneracy from Lie algebra relations

AI summary

Overview

Research area: Quantum many-body theory and tensor networks — specifically exactly solvable spin models derived as parent Hamiltonians of Matrix Product States (MPS), with Lie algebra representation theory as the organizing principle.

Technical level: Advanced. The paper assumes familiarity with MPS, parent Hamiltonians, injectivity, Lie algebras, and representation theory (irreps, Casimir elements, Vogel parameters).

Scope (one sentence): The paper constructs an SO(3)-symmetric nearest-neighbor parent Hamiltonian from the generators of 𝔰𝔬(3) and the identity, proves that its ground space grows algebraically — quadratically for open boundary conditions and linearly for periodic boundary conditions — and generalizes the construction to SO(d) and other compact simple Lie groups.

What This Paper Is About

Given any MPS, one can build a "parent Hamiltonian" that has the MPS in its ground space, and under standard conditions (injectivity, sufficiently large Hamiltonian range) that ground space is unique and gapped. This paper asks what happens when the range is deliberately shrunk to the injectivity length itself, for a special MPS whose building blocks are the generators of the Lie algebra 𝔰𝔬(3) together with the identity matrix. The authors show that this small parent Hamiltonian encodes the quadratic relations of the Lie algebra so precisely that its ground space degeneracy scales algebraically with system size, producing a critical, gapless model rather than a unique ground state.

Key Contributions

1. A Lie-algebraic MPS construction. The authors define a 4-dimensional on-site Hilbert space with orthonormal basis {x, y, z, e}, where the three matrices A_x, A_y, A_z are the adjoint-action matrices of 𝔰𝔬(3) ≅ 𝔰𝔲(2) and A_e is the 3×3 identity. The resulting MPS tensor lives in ℳ₃ ⊗ ℋ and is shown to be injective with injectivity length 2.

2. Exact characterization of the open-boundary ground space. For open boundary conditions (OBC), the ground space of the two-site parent Hamiltonian on n particles is shown to have dimension (n+1)², decomposing into exactly one copy of each integer-spin irreducible representation of SO(3) — that is, one irrep of each odd dimension.

3. Periodic-boundary analysis and gaplessness. For periodic boundary conditions (PBC), the ground space degeneracy is linear and consists of a singlet (the defining MPS itself) plus a ferromagnet (the irrep of maximal spin), which implies gapless spin-wave excitations.

4. Generalization to compact simple Lie groups. The construction and the OBC ground-space analysis are extended to SO(d) and other compact simple Lie groups, with the degeneracy formula proved using Vogel's universal parameters for the eigenvalues of the adjoint action of the split Casimir element on the symmetric square of a simple Lie algebra.

Main Findings

• The model is SO(3)-symmetric under the adjoint action. The MPS tensor satisfies (𝟙 ⊗ ψ(g))(A) = −(ad(A_g) ⊗ 𝟙)(A) for every g ∈ 𝔰𝔬(3), so the symmetry representation on the physical space is the spin 0 ⊕ 1 representation.

• The MPS is injective with injectivity length 2. Because injectivity length is 2, the standard theorem guarantees that the 3-site parent Hamiltonian (one site above the injectivity length) has a unique ground space and is gapped.

• The 2-site parent Hamiltonian is the source of the critical behavior. The model studied is the parent Hamiltonian at range l = 2, equal to the injectivity length k = 2, so the uniqueness theorem does not apply. The algebraic ground-space scaling is attributed to this deliberately small range.

• Open boundary conditions give quadratic degeneracy. The ground space on n particles has dimension (n+1)², containing one irrep of each odd dimension (integer spin) of SO(3).

• Periodic boundary conditions give linear degeneracy. The ground space consists of the singlet MPS and the maximal-spin (ferromagnetic) irrep, yielding gapless excitations.

• This is a distinct degeneracy class. The authors contrast their result with Schuch et al. (2025), where small parent Hamiltonians of injective MPS were shown to generically have unique ground states, with deviations constructed only for exponentially growing ground-space degeneracy. Their model instead shows critical (algebraic) scaling.

• Comparison to the Motzkin chain. Quadratic ground-space scaling also appears in the Motzkin chain without boundary terms, but the present model differs in having a symmetry group directly tied to the degeneracy and in being a parent Hamiltonian of an MPS.

• The proof technique is Lie-algebraic. The OBC degeneracy formula for general compact simple Lie groups relies on Vogel's universal parameters governing the split Casimir element's adjoint action on the symmetric square of the algebra.

• Subproduct system perspective. The OBC ground spaces of frustration-free Hamiltonians naturally form a subproduct system; those obtained here are Lie-algebra equivariant by construction, and the paper analyzes how this equivariance controls the dimension growth.

Methodology in Plain English

The authors start from the three 3×3 matrices that describe how the generators x, y, z of 𝔰𝔬(3) act on the algebra itself through commutation — the adjoint representation — and add the identity matrix as a fourth on-site operator. Stacking these four matrices as the local tensor of an MPS gives a state whose structure literally encodes the Lie algebra's quadratic relations.

From this tensor they build the nearest-neighbor parent Hamiltonian: each term is the orthogonal projector onto the complement of the two-site subspace spanned by the MPS, and the full Hamiltonian is a sum of such projectors along the chain. Because the Hamiltonian is a sum of non-negative projectors, its ground space is exactly the set of states annihilated by every term, which can be written as an intersection of local constraints.

The ground space is then identified by decomposing the relevant tensor products into irreducible representations of SO(3) and counting how many copies of each irrep appear, rather than by numerically diagonalizing the Hamiltonian. For PBC the authors identify which representations survive the periodic closure of the chain. The generalization replaces 𝔰𝔬(3) by an arbitrary compact simple Lie algebra and uses Vogel's parameters — a universal way of writing Casimir eigenvalues — to obtain a closed degeneracy formula. No numerical simulations, benchmark datasets, or experimental comparisons are reported in the paper.

Why This Matters

The work sharpens the boundary between "small parent Hamiltonians that are still well behaved" and "small parent Hamiltonians that are genuinely critical." Since small parent Hamiltonians are attractive for analytical work and for variational ansätze, knowing that algebraic ground-space growth is possible — and having an explicit, symmetry-controlled example — gives the field a new reference point beyond unique and exponentially degenerate cases.

Potential applications (as general relevance, not claims made in the paper):

• Benchmarking tensor-network algorithms that assume unique or gapped ground states, by providing exactly solvable counterexamples.
• Designing and analyzing quantum simulators of spin chains with tunable boundary conditions and controlled degeneracies.
• Studying ferromagnetic ground spaces and gapless spin-wave excitations in solvable lattice models.
• Providing solvable testbeds for Lie-algebraic symmetries in quantum many-body systems, useful in mathematical physics and representation theory.

Industry relevance: The results are foundational rather than device-oriented. They matter indirectly to quantum computing and quantum simulation groups that rely on tensor-network methods for state preparation, circuit compression, and classical simulation, and to groups developing symmetry-aware algorithms where representation-theoretic structure determines computational cost.

Future Directions

1. Periodic boundary conditions for general Lie groups. The paper analyzes PBC in detail for SO(3); extending the linear-degeneracy result and the singlet-plus-ferromagnet picture to SO(d) and other compact simple Lie groups is a natural open problem.

2. Classifying algebraic degeneracy for small parent Hamiltonians. The work adds one data point (algebraic scaling) to a short list (unique, exponential, algebraic); a systematic classification of which MPS tensors produce which scaling laws remains open.

3. Stability and perturbations. Whether the critical ground-space structure survives symmetry-breaking perturbations, longer-range terms, or boundary defects, and how it connects to physical criticality, is not settled.

4. Subproduct systems and equivariant dimension growth. The authors point to Lie-algebra-equivariant subproduct systems as the organizing structure; a general theory relating equivariance to degeneracy growth could yield further solvable families.

Target Audience

Quantum many-body theorists and tensor-network researchers interested in exactly solvable models and parent Hamiltonians; mathematical physicists working on representation theory and operator algebras; and graduate students with a background in MPS, Lie algebras, and representation theory who want a worked example of how algebraic structure controls ground-space degeneracy. Readers seeking numerical benchmarks, experimental proposals, or algorithmic performance comparisons will not find them here.

Authors’ abstract

We introduce an $\mathrm{SO}(3)$-symmetric spin model derived from the algebraic structure of $\mathfrak{so}(3)$: It is obtained as the nearest-neighbor parent Hamiltonian of a Matrix Product State (MPS) built from the generators of the Lie algebra $\mathfrak{so}(3)$ and the identity matrix, and therefore encodes the quadratic relations of the Lie algebra. We characterize the ground space structure of the resulting model and show that for open boundary conditions (OBC), it exhibits a quadratic ground space degeneracy, given by one irreducible representation (irrep) of each odd dimension (integer spin). For periodic boundary conditions (PBC), it has a linear ground space degeneracy, consisting of a singlet --- namely, the MPS underlying the model --- and a ferromagnet (the irrep with maximal spin), and thus exhibits gapless excitations. We also generalize the construction and the analysis of the OBC ground space to $\mathrm{SO}(d)$ and other compact simple Lie groups. The MPS on which the model is based is an injective MPS, and therefore, its 3-site parent Hamiltonian has a unique ground space and is gapped. The observed critical behavior thus has its origin in the small parent Hamiltonian considered. The model's algebraic ground space scaling thus provides an example distinct from that reported in (Schuch et al., 2025, arXiv:2503.10767), where it was shown that small parent Hamiltonians of injective MPS generically have unique ground states, and examples with an exponential ground space degeneracy were given.