Strengthened correlations near [110] edges of $d$-wave superconductors in the t-J model with the Gutzwiller approximation

Using the Gutzwiller approximation within the t-J model, this study reveals that [110] edges of d-wave superconducting slabs exhibit locally strengthened correlations and reduced hopping that draw quasiparticle charge toward the boundary, thereby weakening local superconductivity, suppressing zero-energy Andreev bound states, and preventing the formation of an extended s-wave component across a wide range of hole dopings.

The Gist

Imagine a high-tech city made of tiny, dancing electrons. In most materials, these electrons move around freely, like people walking through an open park. But in a special class of materials called cuprates (which are high-temperature superconductors), the electrons are like a very crowded, chaotic mosh pit. They are so packed together that they constantly bump into and interact with each other. This is what physicists call "strong correlations."

In this city, the electrons usually pair up to form a supercurrent—a frictionless flow of electricity. These pairs dance in a specific pattern called d-wave symmetry. You can visualize this pattern like a four-leaf clover: the dance is strong along two opposite directions but weak (or non-existent) in the other two.

The Problem: The Edge of the City

The researchers in this paper decided to look at the "edge" of this electron city. Specifically, they looked at a cut made at a 45-degree angle (the [110] edge).

In a perfect, empty park (a simple theory), if you cut a d-wave city at this angle, the "clover" pattern gets smashed against the wall. The dance breaks down right at the edge. Usually, physicists thought this would create a chaotic mess where the electrons might try to change their dance style entirely, perhaps switching to a different pattern (like an "s-wave" or a circle) to survive.

The Discovery: The Crowd Pulls Together

However, this paper used a sophisticated new method (the Gutzwiller approximation) to simulate what happens when you account for the fact that the electrons are really crowded and interact strongly.

Here is what they found, using a simple analogy:

1. The Edge Becomes a Magnet for People
When the researchers simulated the edge, they discovered that the electrons didn't just scatter away. Instead, the "crowd" was drawn toward the edge. Imagine a crowded room where, suddenly, everyone near the exit feels a magnetic pull and crowds even tighter against the door.

• The Result: The electron density (how many people are in a spot) increases right at the edge.

2. The "Mott" Insulator State
Because the crowd at the edge became so dense, the electrons got stuck. They were so packed together that they couldn't move anymore. In physics terms, the edge turned into a Mott insulator—a state where electricity stops flowing because the traffic is too jammed.

• The Metaphor: It's like a highway exit ramp where the cars get so bunched up that they come to a complete standstill. The "hopping" ability of the electrons (their ability to jump from one spot to another) was severely reduced.

3. The Dance Gets Weaker, Not Stronger
You might think that if the electrons are closer, they would dance better. But because they were so stuck in the "insulator" jam, their ability to form the superconducting pairs actually got weaker at the edge.

• The Twist: In older, simpler theories, scientists thought the edge would break the d-wave pattern and allow a new, circular (s-wave) pattern to form. But because the electrons were so strongly correlated (so tightly packed), the "d-wave" pattern held on stubbornly, even though it was weakened. The new "circular" dance never got a chance to start.

4. The Ghosts Disappear
In these superconductors, the broken edge usually creates special "ghost" states called Andreev bound states. Think of these as ghostly echoes of the electrons that sit right at zero energy, creating a distinct signal that scientists can measure.

• The Finding: Because the edge became so crowded and "insulating," these ghostly echoes were suppressed. Their signal became much fainter than expected. It's as if the crowd at the door was so thick that the echoes couldn't get out to be heard.

Why Does This Matter?

This study is important because it corrects a long-held belief.

• Old View: The edge of a superconductor is a place where the rules break down, and new, exotic states of matter (like time-reversal symmetry breaking) might spontaneously appear.
• New View: The strong interactions between electrons actually stabilize the edge. The electrons pull together, get stuck, and suppress the exotic changes. The "edge" is less chaotic and more like a traffic jam that prevents the system from rearranging itself into something new.

The Takeaway

The paper tells us that in these complex, crowded quantum materials, you can't just look at the edge in isolation. The "crowd" (strong correlations) pulls the electrons to the boundary, jams them up, and prevents them from changing their dance style. This means that the exotic, mysterious behaviors we hoped to find at the edges of these superconductors might be much harder to find than we thought, because the electrons are too busy jamming up the exit to let anything new happen.

Technical Summary

1. Problem Statement

The paper investigates the behavior of strongly correlated $d$-wave superconductors, specifically focusing on the physics at the edges of a slab cut at a $45^\circ$ angle relative to the crystallographic axes (the [110] orientation).

• Context: In weak-coupling theories (like standard Bogoliubov-de Gennes), [110] edges of $d$-wave superconductors are known to host zero-energy Andreev bound states (ABS). The suppression of the $d$-wave order parameter at these edges often leads to the spontaneous formation of a subdominant extended $s$-wave component, potentially breaking time-reversal symmetry.
• The Gap: Previous studies often assumed uniform electron densities or used weak-coupling approximations. However, high-temperature superconductors (cuprates) are strongly correlated systems emerging from a Mott insulating state. The authors aim to determine how strong electron-electron correlations and the resulting charge redistribution at the edge affect the superconducting order, the nature of the edge states, and the stability of competing symmetry-broken phases.

2. Methodology

The authors employ the $t-J$ model solved using the Statistically Consistent Gutzwiller Approximation (SGA).

• Hamiltonian: The system is described by the $t-J$ Hamiltonian, which includes nearest-neighbor hopping ($t$) and antiferromagnetic superexchange ($J$), subject to Gutzwiller projection operators that forbid double occupancy of lattice sites.
• Gutzwiller Approximation (SGA):
  • Instead of a simple mean-field treatment, the SGA introduces renormalization factors ($g_t, g_s$) that depend on local electron occupations ($n_i$).
  • These factors renormalize the hopping amplitude ($t \to g_t t$) and the superexchange coupling ($J \to g_s J$).
  • The approach uses Lagrange multipliers to enforce self-consistency between the mean fields (hopping $\chi$, pairing $\Delta$) and the local electron density ($n$).
• Geometry: The study models a slab of width $N_x = 100$ sites with [110] edges. The system is translationally invariant along the edge tangent ($y'$) but allows for spatial variation perpendicular to the edge ($x'$).
• Parameters: Calculations are performed for hole dopings $\delta \in [0.05, 0.2]$ (covering underdoped to slightly overdoped regimes) with $J = 0.25|t|$ and low temperature ($\beta = 500$).
• Comparison: The authors compare their full self-consistent results (where charge redistributes) against a simplified approximation where electron density and hopping are assumed to remain at their bulk values throughout the slab (similar to previous works like Tanuma et al.).

3. Key Contributions and Results

A. Charge Redistribution and Strengthened Correlations

• Charge Accumulation: The most significant finding is that quasiparticle charge is drawn to the edges. As the system approaches the edge, the local electron density ($n_a$) increases.
• Proximity to Mott Insulator: At low doping ($\delta = 0.05$), the edge electron density approaches unity ($n \approx 1$), effectively pushing the edge region toward a Mott insulating state.
• Renormalization: This increased density causes the Gutzwiller factor ($g_t$) to decrease significantly near the edge. A lower $g_t$ signifies strengthened correlations and reduced effective hopping amplitudes on the bonds near the edge.

B. Impact on Superconducting Order

• Pairing Amplitude vs. Order Parameter:
  • The bare pairing amplitude in the uncorrelated state ($\Delta_{uncorr}$) is actually enhanced near the edge at low doping ($\delta=0.05$) due to strong interactions.
  • However, the physical superconducting order parameter ($\Delta_{sc} = g_t \Delta_{uncorr}$) is suppressed at the edge for all doping levels. This suppression is driven by the vanishing Gutzwiller factor ($g_t$) caused by the charge accumulation.
• Doping Dependence:
  • At high doping ($\delta=0.2$), the pairing amplitude is suppressed near the edge (consistent with weak coupling), but less so than in uniform models.
  • At low doping ($\delta=0.05$), the pairing amplitude is enhanced, yet the physical order parameter remains suppressed due to correlation effects.

C. Fate of Andreev Bound States and Symmetry Breaking

• Zero-Energy States: The zero-energy flat band of Andreev bound states (ABS) survives the inclusion of strong correlations.
• Spectral Weight Reduction: Crucially, the spectral weight of these zero-energy states is substantially reduced near the edge. This is a direct consequence of the Gutzwiller factor $g^N_a$ (related to $g_t$) suppressing the local density of states (LDOS) as the edge becomes more correlated.
• No Extended $s$-wave Formation: In weak-coupling theories, the suppression of the $d$-wave component at the edge allows a subdominant extended $s$-wave component to form (leading to time-reversal symmetry breaking).
  • Result: In this strongly correlated SGA model, the extended $s$-wave component does not form. The robustness of the $d$-wave pairing amplitude (due to charge redistribution) and the suppression of the spectral weight prevent the instability required for the $s$-wave component to emerge.

4. Significance and Conclusion

• Robustness of $d$-wave: The study demonstrates that strong correlations stabilize the $d$-wave symmetry at [110] edges against the formation of competing $s$-wave components, challenging predictions based on weak-coupling theories.
• Edge Physics: The edge of a strongly correlated superconductor is not merely a boundary where order vanishes; it is a region of enhanced correlations and charge accumulation that fundamentally alters the local electronic structure.
• Experimental Implications:
  • The substantial reduction in the spectral weight of zero-energy ABS suggests that experimental signatures (like Zero-Bias Conductance Peaks in tunneling experiments) might be weaker or more difficult to observe in strongly correlated regimes than predicted by standard theories.
  • The lack of spontaneous $s$-wave formation implies that time-reversal symmetry breaking at these edges may be less robust or require different mechanisms (e.g., magnetic ordering) than previously thought in the context of pure superconductivity.

In summary, the paper establishes that charge redistribution driven by strong correlations is a critical mechanism that locally strengthens correlations at [110] edges, suppresses the physical superconducting order, reduces the visibility of Andreev bound states, and inhibits the formation of subdominant $s$-wave order parameters in high-$T_c$ superconductors.