Exact Rokhsar-Kivelson points hosting first-order phase transitions in S=1 compass models

arXiv:2611.05326 · cond-mat.stat-mech, cond-mat.str-el · Submitted 2026-11-03 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Condensed Matter. The day in condensed matter: materials, phases and the measurements under them.

Kai: We're looking at the paper: "Exact Rokhsar-Kivelson points hosting first-order phase transitions in S=1 compass models".

Mira: Here's the gist of the paper. Exact Rokhsar–Kivelson points hosting first-order phase transitions in S = 1 compass models is studied on square and kagome lattices with bond-dependent bilinear and biquadratic interactions to determine exact ground-state manifolds and locate exact Rokhsar–Kivelson (RK) points that host first-order transitions Model Description The study focuses on S = 1 compass models on the square and kagome lattices featuring bond-dependent bilinear and biquadratic interactions parameterized by an angle θ.

Kai: The Hamiltonian is given by Hˆ (θ) = cos θ X b Sˆαb ib Sˆαb jb + sin θ X b Sˆαb ib Sˆαb jb 2, where the sum runs over all bonds b of the lattice and θ parametrizes the relative strengths of the bilinear and biquadratic terms The square lattice contains only xand y-type bonds, whereas the kagome lattice contains all three bond types.

Mira: Ground State Structure The ground-state manifold is known exactly throughout the interval π/4 ≤ θ ≤ 3π/4.

Kai: In the interior region, π/4 < θ < 3π/4, both models have two product states in the ground state manifold which are annihilated by every term in the Hamiltonian.

Paper summary: Kai: So to summarize this paper, they are looking at S one compass models on square and kagome lattices with bond dependent bilinear and biquadratic interactions parameterized by an angle theta. Their thesis is that they can find the exact ground state manifolds for both geometries throughout the interval pi over four up to three pi over four.

Mira: They claim that at the endpoints, theta equals pi over four and three pi over four, these are exact Rokhsar–Kivelson points. This means one ground state per dynamically connected component of a configuration graph G in the S one Cartesian basis.

Lev: What matters is that they also show that both models have m equals order L independent and mutually commuting Z two line charges, which forces a ground state degeneracy of at least two m at those RK endpoints. That’s a structural feature they found.

Kai: And then in the interior region, between pi over four and three pi over four, both models have exactly two product states in the ground state manifold that minimize the energy because of positive semidefiniteness.

Mira: The crucial finding is that this interior region hosts a first order transition into these two product states with long range order. They found that the ground state energy density is zero throughout this interval, but its derivative at theta equals pi over four shows a jump, establishing the first-order nature of that transition.

Lev: So for us, the implication is that these models are not just smoothly varying systems; they have sharp structural changes in their ground states depending on how you tune those interaction parameters. That’s something we need to account for when designing control sequences.

Kai: They also provided a specific mathematical representation for the ground state wavefunctions using a PEPS structure with bond dimension two, which is useful for understanding the entanglement structure of these exact states.

Mira: This work matters because it provides an exact characterization of where these transition points lie and how those transitions are manifested in terms of configuration space connectivity, which is a direct link between the physics and the geometry of the lattice.

Lev: It’s not just about finding points; it’s about understanding the phase boundaries precisely enough that we know exactly where to look for sharp changes in our simulations or experiments.

Conclusion: Kai: Looking at the full picture of this paper, "Exact Rokhsar–Kivelson points hosting first-order phase transitions in S=one compass models," we see that the authors managed to precisely map out the exact ground state behavior for these spin models across a significant range of interaction parameters <ref:2611.05326#pg1,Exact Rokhsar–Kivelson points hosting first-order phase transitions in S=1>.

Mira: The core implication is tying the geometry of the lattice directly to the degeneracy of those states at specific points, showing how those specific geometric configurations dictate where first order transitions occur.

Lev: For us, it’s about knowing that these structural changes are not just theoretical curiosities; they are real features that would require very careful tuning if we were trying to build a device based on this physics. It gives us the map of where the phase boundaries are sharpest.

Kai: So, in simple terms, the paper shows that for these S one compass models, there’s a specific set of angles where you get exact Rokhsar–Kivelson points with specific degeneracy, and around those points, you get sharp first order transitions into stable regions of product states.

Mira: It moves beyond just observing phase transitions; it gives us the exact mathematical framework—the configuration graph G—that explains why those transitions happen in relation to the connectivity of the lattice structure itself.

Lev: What this means for future work is that if we want to predict these sharp transition points in more complex systems, we need tools that can handle this type of exact mapping between geometry and ground state properties.

Kai: It’s a strong result because it gives us an exact blueprint for where the interesting physics happens in these specific quantum systems, even if scaling up to larger system sizes with higher bond dimensions is still a challenge.

Mira: Overall, this work establishes a very rigorous connection between topological concepts like Rokhsar–Kivelson points and the actual measurable phase transitions in these physical spin models.

Jay Pandey

Department of Theoretical Physics, Tata Institute of Fundamental Research, Mumbai

cond-mat.stat-mech, cond-mat.str-el

Submitted: 2026-11-03

Updated: 2026-11-03

Comments: 9+2 pages, 4 figures. Comments are welcome!

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 87/100

The gist: Exact Rokhsar–Kivelson points hosting first-order phase transitions in S = 1 compass models is studied on square and kagome lattices with bond-dependent bilinear and biquadratic interactions to…

Key concepts

S=1 compass models
These are physical systems described by a specific quantum Hamiltonian where spins (S=1) interact via bilinear and biquadratic terms. The interactions depend on the bond connecting two sites, which is controlled by an angle theta, allowing for rich phase behavior.
Rokhsar–Kivelson (RK) points
These are specific points in parameter space where a quantum system exhibits a special type of degeneracy in its ground state. In this study, they are identified as endpoints where the system transitions into an interior phase with long-range order.
First-order transition
This describes a sudden, discontinuous change in the physical properties of the system when a parameter (like theta) is varied. The paper shows that at these RK points, varying theta causes a sharp jump in the system's energy density, indicating a first-order phase transition.
Ground-state manifold
This refers to all possible quantum states that have the lowest possible energy for a given configuration of parameters. The study precisely maps out this manifold across different lattice structures and interaction strengths.

Terminology

Summary

Exact Rokhsar–Kivelson points hosting first-order phase transitions in S = 1 compass models is studied on square and kagome lattices with bond-dependent bilinear and biquadratic interactions to determine exact ground-state manifolds and locate exact Rokhsar–Kivelson (RK) points that host first-order transitions

Model Description

The study focuses on S = 1 compass models on the square and kagome lattices featuring bond-dependent bilinear and biquadratic interactions parameterized by an angle θ. The Hamiltonian is given by Hˆ (θ) = cos θ X b Sˆαb ib Sˆαb jb + sin θ X b Sˆαb ib Sˆαb jb 2, where the sum runs over all bonds b of the lattice and θ parametrizes the relative strengths of the bilinear and biquadratic terms The square lattice contains only xand y-type bonds, whereas the kagome lattice contains all three bond types. p. 1

Where the paper says it

Each bond b of the two lattices is assigned an orientation-dependent label αb ∈ {x, y, z} and it couples the corresponding component of the S = 1 spins located at its endpoints ib and jb through bilinear and biquadratic interactions: Hˆ (θ) = cos θ X b Sˆαb ib Sˆαb jb + sin θ X b Sˆαb ib Sˆαb jb 2 , (1) where the sum runs over all bonds b of the lattice and θ parametrizes the relative strengths of the bilinear and biquadratic terms. The square lattice contains only xand y-type bonds, whereas the kagome lattice contains all three bond types, as illustrated in Fig.

Ground State Structure

The ground-state manifold is known exactly throughout the interval π/4 ≤ θ ≤ 3π/4. In the interior region, π/4 < θ < 3π/4, both models have two product states in the ground state manifold which are annihilated by every term in the Hamiltonian. p. 0

Where the paper says it

Both models possess m = O(L) independent and mutually commuting conserved Z2 line charges, where L is the linear system size, forcing a ground-state degeneracy of at least 2m at the RK endpoints. Throughout the interior region, π/4 < θ < 3π/4, the ground-state manifold on each lattice is instead spanned by two product states, both annihilated by the Hamiltonian.

At both endpoints (θ = π/4 and 3π/4) on the two lattices, a product of on-site unitary operators maps the Hamiltonian to the Laplacian of a configuration-graph G in the S = 1 Cartesian basis. p. 1
Where the paper says it

Each is annihilated by every term in the Hamiltonian, whose positive semidefiniteness in this interval guarantees that these states minimize the energy. At both endpoints (θ = π/4 and θ = 3π/4) on the two lattices, a product of on-site unitary operators maps the Hamiltonian to the Laplacian of a configuration-graph G [31, 32] whose vertices represent product configurations in the S = 1 Cartesian basis.

RK Point Physics

The endpoints are exact Rokhsar–Kivelson (RK) points, with one ground state per dynamically connected component of G. Both models possess m = O(L) independent and mutually commuting conserved Z2 line charges, where L is the linear system size, forcing a ground-state degeneracy of at least 2m at the RK endpoints. p. 0

Where the paper says it

Each such state is an equal-amplitude superposition in the transformed basis and acquires configuration-dependent phases upon transforming back to the original basis. Both models possess m = O(L) independent and mutually commuting conserved Z2 line charges, where L is the linear system size, forcing a ground-state degeneracy of at least 2m at the RK endpoints.

Throughout the interior region, π/4 < θ < 3π/4, each RK endpoint also hosts a first-order transition into this intervening interior phase spanned by the two product states with long-range order. p. 0
Where the paper says it

Throughout the interior region, π/4 < θ < 3π/4, the ground-state manifold on each lattice is instead spanned by two product states, both annihilated by the Hamiltonian. On both lattices, each RK endpoint also hosts a first-order transition into this intervening interior phase spanned by the two product states with long-range order.

Exact Diagonalization and Transition

The ground-state energy density e0 = E0/Nb is zero throughout π/4 ≤ θ ≤ 3π/4. The bound forces a nonzero left-hand slope at θc = π/4 of the energy density e0 = E0/Nb, whereas its right-hand slope vanishes, establishing a first-order transition. p. 5

Where the paper says it

By contrast, E0(θ) = 0 throughout π/4 ≤ θ ≤ 3π/4. The bound forces a nonzero left-hand slope at θc = π/4 of the energy density e0 = E0/Nb, whereas its right-hand slope vanishes, establishing a first-order transition.

Exact diagonalization confirms that the ground-state energy density e0 is zero for π/4 < θ < 3π/4 and shows a discontinuity in de0/dθ at θ = π/4, consistent with the predicted first-order transition. p. 5
Where the paper says it

V, and shown in the schematic in Fig. The insets show a discontinuity in de0/dθ at θ = π/4, consistent with the first-order transition predicted in Sec.

Ground State Representation

The ground state wavefunctions have a natural PEPS representation [33–36] of bond dimension two on both the lattices. The normalized ground state associated with a connected component F of G is ΨS,F ⟩ = 1/p F X s∈F i nz(s) (−1) n B z(s)s⟩. p. 4

Where the paper says it

Let n µ λ (s) denote the number of sites on sublattice µ carrying state λ in configuration |s⟩, and let nz(s) be the total number of z states present in |s⟩. The normalized ground state associated with a connected component F of G is |ΨS,F ⟩ = 1 p |F| X s∈F i nz(s) (−1) n B z (s) |s⟩, (8) |ΨK,F ⟩ = 1 p |F| X s∈F i nz+n B x +n C y (−1) n B x +n B z +n C y +n C z |s⟩, (9) on the square and kagome lattices, respectively.

Further Connections

The construction realizes the equal-weight case of the stochastic-matrix-form framework of Ref. [25], applied separately to each connected fragment. p. 6

Where the paper says it

At the RK endpoints, our construction realizes the equal-weight case of the stochastic-matrix-form framework of Ref. [25], applied separately to each connected fragment.

p. 6
Where the paper says it

At the RK endpoints, our construction realizes the equal-weight case of the stochastic-matrix-form framework of Ref. [25], applied separately to each connected fragment.

The adjacency-degree representation of the Hamiltonian in Eq. p. 0
Where the paper says it

Here we establish a direct route to RK physics in two-dimensional S = 1 compass models having only nearest-neighbour two-site bilinear and biquadratic interactions. On compatible geometries, a product of sublattice-dependent on-site unitaries brings the full Hamiltonian into an adjacency-degree representation on a configuration graph in the Cartesian basis.

(7) also makes the analogy with QDMs explicit. p. 6
Where the paper says it

The adjacency-degree representation of the Hamiltonian in Eq. (7) also makes the analogy with QDMs explicit [3–7].

At the RK endpoints, our construction realizes the equal-weight case of the stochastic-matrix-form framework of Ref. p. 6
Where the paper says it

Determining whether such behavior is confined to fine-tuned points or can persist throughout an extended phase remains an open question. At the RK endpoints, our construction realizes the equal-weight case of the stochastic-matrix-form framework of Ref.

[25], applied separately to each connected fragment. The Laplacian representation at the RK endpoints also identifies continuous families of local Hamiltonians with exactly the same ground-state manifold. p. 6
Where the paper says it

Finding exact excited states is another direction to be explored in the future. The Laplacian representation at the RK endpoints also identifies continuous families of local Hamiltonians with exactly the same ground-state manifold.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved AI can perform exact ground-state calculations for S=1 compass models on square and kagome lattices by mapping them to configuration graphs using a product of on-site unitaries that transforms the Hamiltonian into an adjacency-degree representation (Eq. 7). This allows for the determination of the exact ground state manifold throughout the entire interval pi/4 to 3pi/4.

  2. Improved AI can predict first-order phase transitions at RK points by analyzing the ground state energy density e0(theta) as a function of theta, showing a discontinuity in its first derivative at theta = pi/4 and 3pi/4. This prediction is supported by the bound E0(theta) = 4/9 Nb sin θ − cos θ for 0 ≤ theta < pi/4.

  3. Improved AI can construct exact ground state wavefunctions via a PEPS representation, specifically using the formula in Eq. (AP11). This allows the AI to determine the normalized ground state amplitude of any configuration s within a specific line-charge sector q, which is crucial for understanding subsystem symmetry induced Hilbert space decomposition.

  4. Improved AI can characterize the structure of the configuration graph G and its connected components based on conserved Z2 line charges. The analysis shows that each connected component of G supports one state in the ground-state manifold, enabling the AI to determine the full ground-state manifold therefore reduces to characterizing the connected components of the configuration-graph.

Abstract

We study S=1 compass models on the square and kagome lattices with bond-dependent bilinear and biquadratic interactions parametrized by an angle θ. For compatible geometries, we determine the exact ground-state manifolds throughout the interval π/4 at mostθ at most3π/4. At each endpoint, a product of on-site unitaries maps the Hamiltonian to the Laplacian of a configuration graph G in the S=1 Cartesian basis. These endpoints are exact Rokhsar-Kivelson (RK) points, with one ground state per dynamically connected component of G. Each such state is an equal-amplitude superposition in the transformed basis and acquires configuration-dependent phases upon transforming back to the original basis. Both models possess m=O(L) independent and mutually commuting conserved Z 2 line charges, where L is the linear system size, forcing a ground-state degeneracy of at least 2 m at the RK endpoints. Throughout the interior region, π/4<θ<3π/4, the ground-state manifold on each lattice is instead spanned by two product states, both annihilated by the Hamiltonian. On both lattices, each RK endpoint also hosts a first-order transition into this intervening interior phase spanned by the two product states with long-range order.

Sources

Related papers