Obstructed Cooper pairs in flat band systems - weakly-coherent superfluids and exact spin liquids

This paper demonstrates that in line-graph lattices with strong attractive interactions, doped charges form obstructed Cooper pairs whose motion is frustrated by destructive interference, causing the pair kinetic energy to vanish and resulting in a disorder-free, interaction-driven localization that yields an exact spin liquid ground state with topological order.

The Gist

The Big Picture: When Superconductivity Gets "Stuck"

Imagine a superconductor as a super-highway where electrons travel in pairs (called Cooper pairs) without any friction. Usually, if you push these pairs, they zoom forward effortlessly. This ability to flow is called superfluid stiffness.

In most materials, if you make the electrons stick together very tightly (strong attraction), they become heavy and slow down, but they can still move. However, this paper discovers a strange new type of superconductor where the pairs get so "stuck" that they can't move at all, even though they are perfectly paired up. The authors call these "Obstructed Cooper pairs."

The Setting: A Tricky Dance Floor

To understand why this happens, imagine the electrons aren't moving on a smooth floor, but on a very specific, tricky dance floor called a Checkerboard Lattice.

1. The Flat Band: In this dance floor, the "energy landscape" is perfectly flat. Usually, particles need a slope to roll down and gain speed. Here, the floor is flat, so they don't naturally roll anywhere.
2. The Strong Grip: The electrons are holding hands very tightly (strong attraction). In normal physics, this makes them heavy and slow, but they can still shuffle forward.

The Magic Trick: Destructive Interference

Here is the twist: When these tightly holding pairs try to take a step to a new spot on the dance floor, they encounter a traffic jam caused by their own wave nature.

• The Analogy: Imagine two people trying to walk through a doorway. One person tries to walk through the left side, and the other tries to walk through the right side. But because they are quantum waves, their "steps" cancel each other out perfectly. It's like noise-canceling headphones, but for movement.
• The Result: The pair tries to move left, but the wave cancels it. It tries to move right, but the wave cancels that too. The pair is trapped in a tiny, localized box. They are vibrating in place but cannot travel.

The authors call this Destructive Interference. Because of this, the "stiffness" (the ability to carry current) drops to zero.

The Two Scenarios

The paper looks at this phenomenon in two different ways:

1. The Low-Density Crowd (The "Empty Dance Floor")
Imagine only a few pairs are on the dance floor.

• The Trap: Each pair finds a specific spot where it can sit comfortably. Because of the tricky geometry of the floor, if a pair sits in one spot, it creates a "shadow" that prevents it from hopping to the next spot.
• The Outcome: The pairs form Compact Localized States (CLS). Think of them as molecules that are stuck to a specific tile. They can't move, so the material acts like an insulator, not a superconductor, even though the pairs are there.

2. The Quarter-Filled Crowd (The "Packed Dance Floor")
Imagine the floor is exactly 25% full.

• The Puzzle: This setup turns out to be mathematically identical to a famous puzzle called the Quantum Dimer Model.
• The Surprise: In this state, the material doesn't just stop moving; it enters a Spin Liquid state. This is a weird quantum state where the electrons are entangled over long distances, like a giant, invisible web.
• Fractionalization: If you try to add or remove a pair (a "hole"), it doesn't just move as a whole. It breaks apart into smaller, free-floating pieces called holons. It's like throwing a brick into a pond, and instead of the brick sinking, it shatters into tiny, invisible dust motes that float away independently.

Why Does This Matter?

1. It Defies Expectations:
Usually, scientists think that if you make electrons stick together strongly, they eventually become a superconductor (or a superfluid). This paper shows that under specific conditions, strong sticking actually kills the superconductivity. The pairs exist, but they are paralyzed.

2. It's Not Disorder:
Usually, when electrons get stuck, it's because the material is dirty or has impurities (disorder). Here, the material is perfectly clean and ordered. The "stuckness" comes purely from the geometry of the lattice and the strength of the interaction. It's a "disorder-free localization."

3. New Materials:
The authors suggest this could happen in "anti-cuprate" materials (related to high-temperature superconductors) or in new quantum simulators built with atoms. If scientists can build these materials, they might see:

• Zero Stiffness: The material won't respond to magnetic fields the way a normal superconductor does.
• Dark Pairs: The pairs won't react to electric fields.
• Topological Order: A hidden, robust quantum order that could be useful for quantum computing.

The Takeaway

This paper reveals a hidden "trap" in the quantum world. If you build a specific type of lattice and make electrons pair up very strongly, you don't get a super-highway for electricity. Instead, you get a quantum traffic jam where the cars (pairs) are locked in place by their own wave nature.

It's a beautiful example of how, in the quantum world, being "strongly connected" doesn't always mean "moving together." Sometimes, it means getting stuck in a perfect, frozen dance.

Technical Summary

1. Problem Statement

The paper addresses a fundamental conceptual hurdle in the theory of superconductivity within partially filled flat bands.

• The Conventional View: In standard weak-coupling superconductors, the superfluid stiffness ($D_s$) is determined by the electronic band structure. In the strong-coupling limit (large attractive interaction $U$), pairs become localized, and their kinetic energy (and thus stiffness) is expected to scale inversely with the binding energy ($D_s \propto 1/U$) via second-order virtual hopping processes.
• The Flat Band Challenge: In systems with exactly flat bands (e.g., line-graph lattices like the checkerboard or Kagome lattices), the absence of a Fermi surface and the presence of destructive interference in hopping pathways challenge this picture.
• The Core Question: Does the superfluid stiffness vanish as $O(U^{-1})$ in the strong-coupling limit for flat band systems, or can it vanish even faster due to specific interference effects? The authors investigate whether strong pairing interactions can lead to a "disorder-free localization" of Cooper pairs, resulting in a superconductor with zero superfluid stiffness at the leading order.

2. Methodology

The authors employ a multi-faceted theoretical approach combining perturbation theory, exact diagonalization, and mapping to solvable models:

• Model System: They study an attractive Hubbard model on a checkerboard line-graph lattice. This lattice is chosen because its non-interacting band structure hosts an exactly flat band when nearest-neighbor (NN) and next-nearest-neighbor (NNN) hopping integrals are equal ($W=0$).
• Strong-Coupling Expansion: They perform a Schrieffer-Wolff transformation in the limit $U \gg t$ (where $t$ is the hopping integral). This projects the fermionic model onto a low-energy effective Hamiltonian of hard-core bosons (representing local Cooper pairs) with nearest-neighbor hopping and repulsion.
• Mean-Field Analysis: They calculate the superfluid stiffness using BCS mean-field theory for both $s$-wave (symmetric) and $d$-wave (antisymmetric) pairing channels to identify the asymptotic scaling behavior.
• Exact Diagonalization & Mapping:
  • They analyze the single-particle and many-body spectra of the effective boson Hamiltonian.
  • At quarter-filling, they map the system to a Quantum Dimer Model (QDM).
  • They demonstrate that the system maps to the exactly solvable Rokhsar-Kivelson (RK) point of the QDM, allowing for an analytical determination of the ground state.
• Current Response Analysis: They explicitly calculate the diamagnetic and paramagnetic current responses to prove the vanishing of the superfluid stiffness.

3. Key Contributions and Results

A. Obstructed Cooper Pairs and Vanishing Stiffness

The central finding is the existence of "Obstructed Cooper Pairs."

• Destructive Interference: In the strong-coupling limit, the motion of local Cooper pairs on the line-graph is frustrated by destructive interference.
• Scaling Law: Unlike conventional strong-coupling superconductors where $D_s \propto U^{-1}$, the authors show that for the $d$-wave pairing channel on the checkerboard lattice, the leading-order contribution to the kinetic energy vanishes identically.
• Result: The superfluid stiffness scales as $D_s \propto U^{-3}$. This implies that the critical temperature ($T_c$) and stiffness are orders of magnitude lower than in systems with unobstructed hopping.

B. Compact Localized States (CLS) and Dark Eigenstates

• Compact Localized States (CLS): The effective pair-hopping Hamiltonian possesses exact eigenstates that are compactly localized on a single plaquette (a "molecular orbital" of the pair). These states are "dark" because they do not couple to uniform vector potentials (zero Drude response).
• Many-Body Degeneracy: At low densities, the system exhibits an extensive ground state degeneracy. Bosons can occupy non-overlapping CLSs without energy cost, leading to a manifold of degenerate ground states.
• Kinetic Stabilization of $d$-wave: Counter-intuitively, the kinetic energy of the pairs favors the orbitally antisymmetric $d$-wave condensate over the $s$-wave condensate in the strong-coupling limit, despite the attractive interaction usually favoring $s$-wave.

C. Mapping to Quantum Spin Liquids

At quarter-filling (one boson per plaquette in the dual picture), the system maps to a Quantum Dimer Model:

• Rokhsar-Kivelson Point: The effective Hamiltonian maps exactly to the RK point of the QDM.
• Ground State: The ground state is a $d$-wave Resonating Valence Bond (RVB) spin liquid. This state possesses:
  • Long-range entanglement.
  • Topological order (specifically $Z_2$ order on non-bipartite lattices).
  • Deconfined Holon Excitations: Upon doping (adding or removing a pair), the charge fractionalizes into deconfined holons (charge-$e$, spin-0 partons) that move freely without an energy cost to leading order.
• Many-Body Scars: The authors identify extensive degeneracies in the many-body spectrum arising from "interference-caged" states, analogous to many-body flat bands in Fock space.

D. Microscopic Origin

The authors propose a microscopic mechanism where strong antiferromagnetic interactions on a square lattice, combined with charge-dependent exchange, naturally generate an attractive Hubbard model on the line-graph. This provides a high-energy origin for the "obstructed" physics without requiring fine-tuned disorder.

4. Significance and Implications

• New Mechanism for Localization: The paper establishes a disorder-free mechanism for interaction-driven localization. Strong pairing interactions can collapse the kinetic energy of Cooper pairs, preventing superfluidity even in a clean system.
• Bridge Between Fields: It creates a rigorous theoretical bridge between strong-coupling superconductivity and frustrated magnetism, showing that a zero-stiffness superconductor can be an exact spin liquid.
• Experimental Predictions:
  • Spectral Weight: Obstructed pairs should exhibit spectral weight on the edges of the lattice with specific sign structures (detectable via Scanning Josephson Interferometry).
  • Two-Photoemission (2e-ARPES): The flat pair dispersion and the symmetry character of the obstructed pairs can be directly imaged using coincidence two-photoemission spectroscopy.
  • Material Candidates: The physics is relevant for "anti-cuprate" materials (where transition metals are on the links of a square lattice) and moiré heterostructures.
• Theoretical Impact: It challenges the conventional wisdom that strong coupling always leads to a Bose-Einstein Condensate (BEC) with finite stiffness, showing instead that specific lattice geometries can lead to a "weakly-coherent" superfluid or a spin liquid with deconfined excitations.

Conclusion

The paper demonstrates that in flat band systems with specific lattice geometries (line-graphs), strong attractive interactions do not simply lead to localized bosons with $1/U$ stiffness. Instead, destructive interference creates obstructed Cooper pairs with zero leading-order stiffness ($O(U^{-3})$). At quarter-filling, this system realizes an exact $d$-wave RVB spin liquid with deconfined holons, offering a novel paradigm for interaction-driven localization and topological order in superconducting systems.