Extended Hubbard model on fractals: d-Wave superconductivity and competing pairing channels

This study demonstrates that fractal geometries, such as the Sierpiński carpet and gasket, act as selective filters for pairing symmetries in the extended Hubbard model by geometrically frustrating sign-changing order parameters to suppress d-wave superconductivity while enhancing extended s-wave or hybrid pairing channels.

The Gist

Imagine you are a city planner trying to build the most efficient subway system possible. In a normal city (a standard crystal), the streets form a perfect grid. You know exactly how the trains (electrons) should move to get from A to B.

Now, imagine you decide to build this subway system on a fractal. A fractal is a shape that looks the same no matter how much you zoom in, like a coastline or a snowflake. It's full of holes, twists, and missing pieces. It's a "broken" city where some streets just end abruptly, and neighborhoods are disconnected in weird ways.

This paper asks a big question: If we build a superconductor (a material that conducts electricity with zero resistance) on this broken, fractal city, what happens to the way the electrons pair up?

Here is the breakdown of their discovery, using some everyday analogies:

1. The "Dance" of Electrons

In a superconductor, electrons don't run around alone; they dance in pairs (called Cooper pairs). The way they dance is called the "symmetry" of the pairing.

• S-Wave (The Simple Waltz): Imagine a couple holding hands and spinning in a perfect circle. They look the same from every angle. This is the "s-wave." It's simple, uniform, and doesn't care much about the direction they are facing.
• D-Wave (The Crossed-Arms Tango): Now imagine a couple where one partner holds a sign saying "Up" and the other says "Down." If they turn 90 degrees, the signs flip. This is "d-wave." It's complex and relies on a perfect grid to work. It needs a specific "cross" shape to function correctly.

2. The Problem with the Fractal City

The researchers looked at two types of fractal cities:

• The Sierpinski Carpet: A square grid with holes punched out of it, over and over again.
• The Sierpinski Gasket: A triangle with holes punched out.

The Carpet Disaster (The Broken Cross):
On a normal square grid, the "d-wave" dance works perfectly because every street corner has four roads (North, South, East, West). The dancers can easily arrange themselves: "You go North (Up), I go South (Down)."
But on the Sierpinski Carpet, the city planners removed whole blocks. Suddenly, a dancer who needs to go North finds a giant hole where the street used to be. The "cross" is broken.

• The Result: The complex "d-wave" dance falls apart. The dancers get frustrated because the geometry won't let them hold the required positions. The city forces them to switch to the simpler "s-wave" dance (the waltz), which doesn't need a perfect cross. Surprisingly, this switch actually makes the superconductivity stronger and more stable in this broken city than it was in the perfect one.

The Gasket Surprise (The Hybrid Dance):
On the Triangular Gasket, the holes are different. Instead of just breaking the dance, the missing streets force the dancers to mix their styles.

• The Result: The electrons don't just do the simple waltz or the complex tango. They invent a hybrid dance (a mix of s, d, and even a "chiral" d+id style). It's like a dance troupe improvising a new routine that combines the best moves of both styles. This new hybrid dance turns out to be incredibly efficient, making the superconductivity even stronger than before.

3. The "Filter" Effect

The main takeaway is that geometry acts like a filter.

• If you have a perfect grid, the complex, high-energy dances (d-wave) are the winners.
• If you have a fractal grid with holes, the complex dances get "frustrated" and lose.
• The fractal geometry selects which dance works best. It suppresses the dances that need perfect symmetry and boosts the ones that can adapt to the chaos.

4. Why Does This Matter?

You might ask, "Who cares about fractal cities?"

• Real-World Tech: Scientists can now build tiny, atom-sized structures using microscopes (like placing individual atoms to build a fractal).
• Superconductors: We want superconductors that work at higher temperatures (so we don't need expensive liquid helium to cool them).
• The Discovery: This paper suggests that by intentionally designing materials with fractal shapes (holes and weird patterns), we might be able to trick electrons into pairing up more strongly, potentially leading to room-temperature superconductors.

Summary Analogy

Think of the electrons as a marching band.

• On a perfect grid, they can do a complex, synchronized routine that requires everyone to be in a specific spot.
• On a fractal grid, the stage is full of holes. The complex routine fails because people fall into the holes.
• The Twist: Instead of giving up, the band leader (the fractal geometry) forces them to switch to a simpler, more robust routine. Or, in the case of the triangle, they invent a brand-new, hybrid routine that is actually better than the original.

The paper proves that shape matters. By changing the shape of the material from a perfect square to a fractal, you can fundamentally change how electricity flows, potentially unlocking the next generation of super-powerful electronics.

Technical Summary

1. Problem Statement

The paper investigates how fractal geometry influences the stability and competition of superconducting order parameters with non-trivial spatial symmetries (specifically $d$-wave and extended $s$-wave).

While previous work established that fractal lattices (like the Sierpiński gasket) can enhance the critical temperature ($T_c$) of conventional $s$-wave superconductivity, it remained unclear how fractals affect pairing channels where the order parameter changes sign (e.g., $d_{x^2-y^2}$ or $d_{xy}$). The central hypothesis is that the unique boundary structure and "geometric frustration" of fractals—where bulk and edge interpenetrate—may selectively suppress sign-changing order parameters while allowing uniform-phase channels to survive or amplify.

2. Methodology

The authors employ the Extended Hubbard Model on various fractal lattices, solved using Bogoliubov-de Gennes (BdG) mean-field theory.

• Model Hamiltonian:
  The system is described by a tight-binding Hamiltonian with on-site attraction ($U < 0$) and nearest-neighbor attraction ($V < 0$):
  $$H = -t \sum_{\langle i,j \rangle, \sigma} c^\dagger_{i,\sigma}c_{j,\sigma} - \mu \sum_{i,\sigma} n_{i,\sigma} + \frac{U}{2} \sum_{i,\sigma} n_{i,\sigma}n_{i,\bar{\sigma}} + \frac{V}{2} \sum_{\langle i,j \rangle, \sigma, \sigma'} n_{i,\sigma}n_{j,\sigma'}$$
  This model is minimal for hosting $d$-wave pairing.

• Numerical Approaches:

  1. Self-consistent BdG: Used for moderate system sizes. The authors solve for pairing amplitudes ($\Delta_{ij}$) and charge densities ($n_i$) iteratively. To handle convergence issues, they employ advanced mixing schemes inspired by machine learning (RMSprop, Adagrad).
  2. Kernel Polynomial Method (KPM): Used for larger fractal generations. This method expands the spectral density using Chebyshev polynomials to compute mean fields without explicit diagonalization of the Hamiltonian, allowing access to larger system sizes.

• Symmetry Constraints:
  To probe specific channels, the authors use symmetry-constrained initial seeds:

  • Pure $d$-wave: Enforced $\Delta_{ii}=0$ and specific sign patterns on bonds.
  • Extended $s$-wave: Uniform sign on all bonds.
  • Mixed states ($s+d$, $s+d+id$): Superpositions of components.

• Validation:
  The methods were benchmarked against known Quantum Monte Carlo (QMC) results for the square lattice, confirming the mean-field approach captures the correct competition between pairing symmetries and charge ordering.

3. Key Results

The study analyzes three primary lattice geometries: the Square Lattice, the Sierpiński Carpet (infinitely ramified), and the Triangular/Honeycomb Sierpiński Gaskets (finitely ramified).

A. Square Lattice vs. Sierpiński Carpet

• Square Lattice: At half-filling with nearest-neighbor attraction, the system exhibits a robust $d_{x^2-y^2}$ superconducting dome with high $T_c$, while extended $s$-wave is weak.
• Sierpiński Carpet: The fractal geometry destabilizes pure $d$-wave superconductivity.
  • The "cross" configuration required for $d$-wave (four bonds with alternating signs) is broken by the removal of sites.
  • This creates geometric frustration, preventing the formation of a coherent global $d$-wave order.
  • Result: The system reorganizes into a state dominated by extended $s$-wave pairing. Remarkably, the $T_c$ for this extended $s$-wave state is enhanced compared to the regular square lattice, even though the carpet is infinitely ramified (a property previously thought to suppress $T_c$ enhancement for pure $s$-wave).
  • If pure $d$-wave is strictly forced, the solution becomes a geometrically modified state with global sign changes but local $s$-like characteristics.

B. Triangular Sierpiński Gasket

• Regular Triangular Lattice: Supports a competition between extended $s$-wave and a 2D $d$-wave subspace (allowing real $d$ and chiral $d+id$ states).
• Sierpiński Gasket:
  • Bond removal mixes the $s$ and $d$ channels locally.
  • Result: The stable ground state is a mixed $s + d + id$ state.
  • Unlike the carpet, the $d$-wave subspace is not eliminated but reshaped. The system exhibits a significant enhancement in $T_c$ and gap amplitude compared to the regular lattice, with only a modest narrowing of the doping window.

C. Honeycomb Sierpiński Gasket

• Regular Honeycomb Lattice: Shows symmetric $s$-wave domes due to its bipartite nature.
• Sierpiński Gasket: The fractal geometry preserves the bipartite symmetry.
  • Result: No new mixed states emerge. The pairing remains purely $s$-wave.
  • The primary effect is a quantitative enhancement of $T_c$ and the gap magnitude, similar to previous findings for on-site $s$-wave on finitely ramified fractals.

D. Phase Stiffness and Coherence

• Calculations of superfluid stiffness ($D_s$) confirm that the condensates maintain macroscopic phase coherence across the fractal structures.
• While the Mermin-Wagner theorem suggests no long-range order in 2D at finite $T$, the authors argue that for finite-generation fractals (physically realizable), the hierarchical structure introduces a cutoff length scale that suppresses vortex proliferation, allowing for quasi-long-range order at finite temperatures.

4. Physical Mechanisms

The core mechanism identified is geometric frustration of sign-changing order parameters.

• Orthogonality Breaking: In regular lattices, $s$-wave and $d$-wave channels are orthogonal. In fractals, the removal of bonds breaks the local coordination shells (e.g., the "star" of bonds), causing the Gram matrix of channel form factors to become non-diagonal.
• Channel Mixing: This loss of orthogonality forces the system to mix channels.
  • On the Carpet, the frustration is severe enough to kill the $d$-channel entirely, favoring the robust $s$-channel.
  • On the Gasket, the mixing creates a stable hybrid state ($s+d+id$) that benefits from the fractal density of states without losing the topological protection of the $d$-subspace.

5. Significance and Conclusion

• Fractals as Symmetry Filters: The paper demonstrates that fractal geometry acts as a selective filter for pairing symmetries. It does not uniformly boost superconductivity but rather reshapes the competition between channels based on the compatibility between the order parameter's symmetry and the lattice topology.
• Beyond Ramification: The results challenge the simple dichotomy that only finitely ramified fractals enhance superconductivity. The Sierpiński carpet (infinitely ramified) enhances extended $s$-wave when competing against $d$-wave, whereas it suppresses pure on-site $s$-wave.
• Design Parameter: Lattice topology is established as a tunable parameter for engineering superconducting states, offering a route to stabilize exotic mixed states (like $s+d+id$) or enhance $T_c$ without changing chemical doping or interaction strengths.
• Experimental Relevance: With recent advances in atomically precise fabrication (STM, molecular assembly), these predicted fractal superconducting states are within experimental reach.

In summary, the work provides a comprehensive theoretical framework showing that fractal geometry fundamentally alters the superconducting phase diagram, stabilizing mixed pairing states and enhancing critical temperatures through the interplay of geometric frustration and channel mixing.