Separability and entanglement of resonating valence-bond states

This paper investigates the separability and entanglement properties of Rokhsar-Kivelson and resonating valence-bond states across arbitrary lattices, proving that disconnected subsystems are exactly or effectively separable while providing exact expressions for entanglement in adjacent subsystems, thereby demonstrating that these quantum spin liquid and critical states exhibit vanishing entanglement between spatially separated regions.

The Gist

Imagine you have a giant, complex puzzle made of tiny magnets (spins) that can point up or down. In the quantum world, these magnets don't just sit there; they dance in a superposition of all possible arrangements at once. This paper investigates a specific type of quantum dance called Resonating Valence Bond (RVB) states and Rokhsar-Kivelson (RK) states.

The big question the authors ask is: If you take two separate groups of these dancing magnets and look at them, are they "entangled" (deeply connected in a spooky quantum way), or are they just independent neighbors?

Here is the breakdown of their findings using simple analogies.

1. The Players: The Quantum Dance Floor

• The Dimer Model (RK States): Imagine a floor covered in tiles. You have to cover the whole floor with dominoes (dimers) so that every spot is covered, and no dominoes overlap. In the quantum version, the system is in a state where it is simultaneously every possible way you could cover the floor with dominoes.
• The RVB State: Now, imagine the tiles are people holding hands. A "valence bond" is a pair of people holding hands. The RVB state is a quantum superposition where everyone is holding hands in every possible pairing arrangement simultaneously.

2. The Setup: The "Island" Experiment

The researchers imagine cutting this giant dance floor into three pieces:

• Island A1: A group of magnets on the left.
• Island A2: A group of magnets on the right.
• The Ocean (Region B): A huge buffer zone of magnets separating the two islands.

The key rule: Island A1 and Island A2 are disconnected. They don't touch. They are separated by the Ocean.

3. The Big Discovery: "Spooky Action" Dies Out

In quantum mechanics, "entanglement" is like a secret telepathic link. If two particles are entangled, measuring one instantly tells you something about the other, no matter how far apart they are.

The authors proved something surprising: If you separate two groups of these quantum magnets with a buffer zone, the "telepathic link" (entanglement) between them effectively vanishes.

• For the Domino Models (RK States): If the islands are truly disconnected, they are 100% separable. It's as if they are two strangers in different rooms who have never met. There is zero quantum connection between them. The math proves this exactly, even on a computer simulation of a grid.
• For the Hand-Holding Models (RVB States): This is slightly more complex because the "hands" (spins) can reach across the gap. However, the authors found that the connection drops off exponentially fast as the distance increases.
  • Analogy: Imagine trying to whisper a secret across a canyon. If the canyon is small, you might hear it. But if you double the width, the sound doesn't just get half as loud; it gets way quieter. If you triple the width, it's practically silent.
  • In these quantum states, if you double the distance between the islands, the "quantum whisper" (entanglement) becomes almost non-existent. Even if the islands are relatively close compared to their size, the connection is so weak it's negligible.

4. The "Logarithmic Negativity" Meter

How do you measure if two things are entangled? The paper uses a tool called Logarithmic Negativity.

• Think of this as a "Spookiness Meter."
• If the meter reads Zero, the two groups are not entangled (they are separable).
• If it reads High, they are deeply entangled.

The paper shows that for disconnected islands in these specific quantum states, the Spookiness Meter reads Zero (or so close to zero it doesn't matter).

5. Why This Is Surprising

Usually, in quantum physics, if you have a "critical" system (one that is on the edge of changing phases, like water about to boil), things stay connected over long distances. You'd expect the islands to still be entangled even if they are far apart.

But these RVB and RK states are special. They are like Quantum Liquids that are "stiff" against long-distance entanglement.

• The Analogy: Imagine a crowd of people. In a normal crowd, if someone shouts, the news travels across the room. In this specific quantum crowd, if you split the room in half with a wall, the two halves stop "talking" to each other quantum mechanically, even if the wall is thin. They only talk to their immediate neighbors.

6. The Takeaway for the Future

The authors conclude that for these specific types of quantum matter (which are candidates for "Quantum Spin Liquids"—a mysterious state of matter that never freezes):

1. No Long-Range Ghosts: You don't need to worry about spooky quantum connections between two disconnected parts of the material.
2. Classical-Like Behavior: Even though the system is quantum, the way these separated parts relate to each other looks very much like classical physics (where things only affect their immediate neighbors).
3. Scalability: This holds true whether the system is small or the size of the universe.

In a nutshell: The paper proves that in these exotic quantum materials, if you separate two chunks with a gap, they become independent. The "quantum glue" that usually binds the universe together doesn't stretch across the gap in these specific states. It's a comforting result for physicists trying to build quantum computers, as it suggests these states might be easier to control and less prone to "spooky" errors spreading across the whole system.

Technical Summary

1. Problem Statement

The paper addresses the fundamental challenge of characterizing entanglement and separability in quantum many-body systems, specifically focusing on Rokhsar-Kivelson (RK) states and Resonating Valence-Bond (RVB) states.

• Context: These states are central to understanding quantum spin liquids and quantum critical matter. While entanglement entropy is well-studied, determining whether a mixed state (obtained by tracing out a subsystem) is separable (i.e., not entangled) is computationally hard (NP-hard).
• Specific Gap: Previous work established that continuum RK states are separable for disjoint regions due to locality. However, it was unclear if this holds for lattice RK states (specifically dimer models) and general RVB states (spin singlets) on arbitrary graphs, particularly regarding multipartite separability and the behavior of the logarithmic negativity (a measure of entanglement for mixed states).
• Key Question: Do disjoint subsystems in RK and RVB states share quantum entanglement, or are they effectively separable?

2. Methodology

The authors employ a combination of exact analytical derivations, statistical mechanics mapping, and graphical methods to analyze the reduced density matrices ($\rho_{A_1 \cup A_2}$) of subsystems.

• RK States (Dimer Models):

  • Definition: RK states are defined as equal-weight superpositions of all allowed configurations of a statistical model (e.g., dimer coverings) on a graph.
  • Tripartition: The system is divided into two disjoint subsystems ($A_1, A_2$) and a buffer region ($B$).
  • Reduced Density Matrix (RDM): The authors derive the RDM by tracing out $B$. They utilize the locality of the energy functional and the local constraints of the dimer model (e.g., the ice rule) to decompose the wavefunction.
  • Separability Criterion: They check if the RDM can be written as a convex combination of product states ($\rho = \sum p_i \rho_i^{A_1} \otimes \rho_i^{A_2}$).
  • Logarithmic Negativity: Calculated using the replica trick ($E = \lim_{n \to 1/2} \log \text{Tr}|\rho^{T_1}|^{2n}$) and partial transpose properties.

• RVB States (Spin Models):

  • Definition: Superpositions of spin singlet coverings (valence bonds) on vertices.
  • Overlap Calculation: Unlike RK states, RVB basis states are not orthogonal. The authors use transition graphs (superimposing two singlet configurations to form closed loops and strings) to compute overlaps $\langle \gamma | \gamma' \rangle = N^{n_\ell - N/2}$.
  • Graphical Analysis: They analyze the reduced density matrix by classifying transition graphs into those with "strings" connecting $A_1$ and $A_2$ through $B$ and those that do not.
  • Symmetry Arguments: They investigate the symmetry of the RDM under partial transposition ($\rho^{T_1}$) to determine separability and negativity.

3. Key Contributions and Results

A. Rokhsar-Kivelson (RK) States

1. Exact Separability for Dimer RK States:

   • For dimer RK states on arbitrary tileable graphs, the authors prove that the reduced density matrix of two disconnected subsystems is exactly separable.
   • Mechanism: Due to local constraints, the boundary configurations of $A_1$ and $A_2$ compatible with a bulk configuration in $B$ are mutually exclusive or form equivalence classes that allow the RDM to be written as a sum of product states.
   • Implication: There is zero bipartite and multipartite entanglement between disjoint regions in dimer RK states, even on the lattice (no thermodynamic limit required).

2. General RK States with Local Constraints:

   • For more general RK states (where the underlying model has local constraints but not necessarily equal weights), the reduced density matrix is exactly separable on the square lattice if boundaries have no concave angles.
   • For arbitrary graphs or concave boundaries, the state is separable in the thermodynamic limit. The non-separable terms are suppressed by the ratio of boundary energy to bulk energy.

3. Logarithmic Negativity:

   • Disjoint Subsystems: Any local RK state has zero logarithmic negativity for disjoint subsystems, even if the density matrix is not exactly separable (i.e., they are "bound entangled" states where entanglement cannot be distilled).
   • Adjacent Subsystems: The authors derive an exact expression for the logarithmic negativity of adjacent subsystems in terms of the partition functions of the underlying statistical model. They show it satisfies an area law (proportional to the boundary length).

B. Resonating Valence-Bond (RVB) States

1. Separability for Disconnected Subsystems:

   • For SU(2) RVB states, the reduced density matrix of two disconnected subsystems is separable up to exponentially small terms of order $O(2^{-d/2})$, where $d$ is the lattice distance between the subsystems.
   • Scaling Limit: Crucially, separability holds in the scaling limit ($d, L \to \infty$) even for arbitrarily small ratios $d/L$. This is a novel feature distinguishing RVB states from standard Conformal Field Theories (CFTs), where entanglement usually decays as a power law or depends on the ratio $d/L$.

2. Logarithmic Negativity:

   • The logarithmic negativity for disjoint RVB subsystems is exponentially suppressed with distance: $E(A_1:A_2) \sim O(2^{-d/2})$.
   • This behavior is non-generic: It holds regardless of whether the system is gapped or gapless (critical). Even for critical RVB states (which usually exhibit power-law correlations), the entanglement between disjoint regions vanishes exponentially.

3. Generalization to SU(N):

   • Results generalize to SU(N) RVB states. The non-separable terms scale as $O(N^{-d/2})$.
   • In the limit $N \to \infty$, the RVB states become exactly orthogonal dimer states, recovering the exact separability of RK states.

C. Multipartite Separability

• The authors extend their analysis to $k$ disjoint subsystems.
• RK States: Exactly $k$-separable for dimer models; separable in the thermodynamic limit for general local constraints.
• RVB States: $k$-separable up to terms of order $O(2^{-d_{\min}/2})$, where $d_{\min}$ is the minimum distance between any pair of subsystems.

4. Significance and Implications

• Nature of Quantum Spin Liquids: The results suggest that while RVB states possess long-range entanglement (topological order) preventing deformation into a product state, this entanglement is not shared between spatially separated disjoint regions in the form of distillable entanglement. The entanglement is "bound" or localized.
• Distinction from CFTs: The exponential suppression of negativity in RVB states, even in the critical (gapless) regime, contrasts sharply with standard CFTs where negativity decays algebraically. This highlights the unique nature of quantum criticality in spin liquids.
• Bound Entanglement: The finding that RK states have zero negativity but may not be strictly separable (in the thermodynamic limit for general constraints) identifies them as bound entangled states. This implies their entanglement cannot be distilled into pure singlets via local operations and classical communication (LOCC).
• Measurement-Induced Entanglement (MIE): The authors speculate that the logarithmic negativity in these sign-free states is smaller than the MIE, suggesting a hierarchy of entanglement measures in non-negative wavefunctions.
• Universality: The results hold for arbitrary lattices and a wide class of RK/RVB states, providing a unified framework for understanding entanglement in dimer and spin liquid models.

Conclusion

The paper rigorously demonstrates that disconnected subsystems in RK and RVB states are effectively separable (or possess only bound entanglement with vanishing negativity). This separability is exact for dimer models and exponentially accurate for spin models, persisting even in the scaling limit for critical systems. These findings refine the understanding of entanglement structure in quantum spin liquids, distinguishing between global topological order and local bipartite/multipartite entanglement between disjoint regions.