Evolution of superconductivity from charge clusters to stripes in the $t$-$t'$-$J$ model

Using finite-temperature tensor network simulations on the $t$-$t'$-$J$ model, this study reveals that superconductivity in underdoped cuprates evolves from pairing correlations localized on intermediate-temperature hole clusters to a coherent, system-wide state in the ground-state stripe phase, providing a microscopic explanation for experimental observations of local pairing above $T_c$ and charge clustering.

The Gist

Imagine a crowded dance floor at a party. This is the world of high-temperature superconductors (like cuprates), where electrons are the dancers. Usually, these dancers are so stubborn and repulsive that they can't move in sync; they just bump into each other, creating resistance (heat). But under the right conditions, they suddenly lock arms and glide across the floor without any friction at all. This is superconductivity.

For decades, scientists have been trying to figure out exactly how these electrons decide to lock arms. This paper, by Sinha, Karlsson, Ulaga, and Wietek, acts like a high-speed camera, taking snapshots of the dance floor at different temperatures to see the story unfold.

Here is the story they tell, broken down into simple steps:

1. The Setting: A Chaotic Dance Floor

The researchers used a mathematical model (the t–t'–J model) to simulate a grid of electrons. Think of this grid as a giant checkerboard.

• The Problem: At high temperatures, the electrons are chaotic. They are "doped" (meaning some extra holes or empty spaces are added), but they are scattered everywhere.
• The Goal: To see how they organize themselves as the room cools down.

2. The Middle Stage: The "Puddles" Form

As the room cools to an intermediate temperature, something interesting happens. The electrons don't just spread out evenly. Instead, the "holes" (the empty spaces where electrons aren't) start to stick together.

• The Analogy: Imagine rain falling on a dry, dusty floor. At first, the water is just a mist. But as it settles, it forms puddles.
• What the paper found: The holes form these "puddles" or clusters. They are like little islands of activity in a sea of calm.
• The Magic: Here is the big discovery: Pairing happens inside these puddles. The electrons start holding hands only inside these specific clusters. It's as if the dancers only found their partners when they were stuck inside a small, crowded corner of the dance floor. The rest of the floor is still chaotic and unpaired.

3. The Final Stage: The "Stripe" Superhighway

As the room gets even colder, the "puddles" don't disappear; they reorganize. They line up to form long, parallel lines.

• The Analogy: Imagine those scattered puddles of water on the floor suddenly merging and aligning into long, straight canals or stripes.
• What happens to the pairs: The electron pairs that were previously stuck inside the small puddles now realize they can talk to each other across the canals. They stop being isolated and start moving in unison across the entire floor.
• The Result: The "locking" of the pairs spreads out. The dance becomes a perfectly synchronized wave moving across the whole system. This is the superconducting stripe phase.

4. The Key Insight: "Pair-Charge Locking"

The authors call the connection between the holes and the pairs "pair-charge locking."

• Think of it like this: You can't have a dance party without dancers. In this material, the "dancers" (electron pairs) are physically glued to the "dance floor" (the hole clusters). You can't have the superconductivity without first having these clusters of holes.
• The paper shows that the superconductivity doesn't just appear out of nowhere; it grows out of these messy, localized clusters.

Why Does This Matter?

For a long time, scientists debated whether superconductivity happened in the "empty" spaces or the "crowded" spaces.

• Old Idea: Maybe the pairs form in the quiet, empty background.
• This Paper's Conclusion: No! The pairs form first in the crowded, messy clusters (the puddles) at higher temperatures. Then, as it gets colder, these local pairs link up to create the global superconducting state.

Connecting to Real Life

The authors compare their findings to real-world experiments:

• STM (Scanning Tunneling Microscopy): This is like a super-microscope that looks at the surface of materials. It has seen these "puddles" and "stripes" in real copper-oxide materials. The paper explains why we see them: because the electrons naturally want to clump together before they can flow freely.
• NMR (Nuclear Magnetic Resonance): This technique detects how atoms are moving. It has also seen signs of these "islands" of charge.

The Bottom Line

This paper tells us that superconductivity is a journey from chaos to order.

1. Hot: Chaos.
2. Warm: Electrons form local clusters (puddles) and start pairing up inside them.
3. Cold: These clusters line up into stripes, and the pairs connect across the whole system, creating a frictionless flow.

It's a beautiful picture of how a messy, disordered system can self-organize into a perfect, magical state of superconductivity, all starting with the simple act of electrons sticking together in small groups.

Technical Summary

1. Problem Statement

The paper addresses the fundamental mechanism of high-temperature superconductivity in cuprates, specifically focusing on the pseudogap regime and the competition/coexistence between charge order (stripes) and superconductivity.

• The Core Question: How does superconductivity emerge from a Mott insulating parent state? Specifically, what is the role of pairing correlations in the intermediate-temperature regime where static stripe order has "melted" but charge inhomogeneities persist?
• The Debate: Does pairing primarily occur in the hole-poor antiferromagnetic (AFM) background (spin-fluctuation scenarios) or on the hole-rich regions themselves?
• The Model: The authors investigate the $t-t'-J$ model, a minimal model for cuprates, at doping $p=1/16$, $J/t=0.4$, and next-nearest-neighbor hopping $t'/t=0.2$. Previous ground-state studies indicated intertwined stripe and superconducting tendencies, but the finite-temperature evolution of these states remained unclear.

2. Methodology

The study employs finite-temperature tensor network simulations using Minimally Entangled Typical Thermal States (METTS).

• Algorithm: METTS generates a Markov chain of pure states ($|\psi_s\rangle$) that sample the thermal density matrix. Thermal expectation values are approximated by averaging over these snapshots: $\langle O \rangle_T \approx \frac{1}{N_s} \sum \langle \psi_s | O | \psi_s \rangle$.
• Geometry: Simulations are performed on cylindrical geometries with open boundaries along $\hat{x}$ (length $L$) and periodic boundaries along $\hat{y}$ (circumference $W$). Systems studied include $16\times6$, $24\times4$, and $32\times4$ cylinders.
• Key Diagnostics:
  1. Pairing Analysis: The authors analyze the two-particle reduced density matrix (2RDM) for singlet pairing ($\rho_2$). Following the Penrose-Onsager-Yang criterion, they examine the eigenvalues ($\epsilon_n$) and eigenvectors (pair wavefunctions $\chi_n$) of $\rho_2$. A leading eigenvalue scaling linearly with system size indicates a condensate.
  2. Charge Clustering: They perform a snapshot-level analysis of hole density ($n_h$) to identify "hole-rich" clusters. A threshold-based algorithm defines connected components of sites where hole density exceeds an adaptive cutoff.
  3. Pair-Charge Locking: A correlation coefficient ($\Lambda$) is calculated between the local hole density and the amplitude of the leading pair wavefunction to quantify if pairing is localized on charge clusters.
  4. Order Parameters: Static spin ($S_s$) and charge ($S_c$) structure factors are computed to track AFM and stripe correlations.

3. Key Contributions

The paper provides a microscopic, real-space picture of the evolution from a disordered, charge-clustered state to a coherent superconducting stripe phase.

• Identification of a Precursor State: The authors identify an intermediate-temperature regime where doped holes form mesoscopic clusters (nanoscale phase separation) without macroscopic phase separation.
• Localization of Pairing: They demonstrate that in this intermediate regime, pairing correlations are tightly localized on these hole-rich clusters, rather than being uniform or residing in the AFM background.
• Mechanism of Coherence: The study elucidates the transition mechanism: as temperature decreases, these localized pair wavefunctions on individual clusters hybridize and tunnel through the AFM barriers, eventually becoming delocalized and coherent across the entire system to form the ground-state stripe superconductor.
• Fragmented Condensate: The work characterizes the finite-temperature superconducting state as a "fragmented condensate," where multiple leading eigenvalues of the 2RDM are significant, corresponding to the number of charge clusters.

4. Detailed Results

A. Evolution of Pairing Eigenvalues

• Intermediate Temperatures ($T/t \gtrsim 0.2$): The leading eigenvalues of the singlet 2RDM are separated from the rest of the spectrum, but the splitting ($w = \epsilon_1 - \epsilon_3$) is near zero. The corresponding pair wavefunctions are spatially localized on specific hole clusters.
• Low Temperatures ($T/t \lesssim 0.2$): As the system cools, the splitting $w$ increases sharply. This indicates the onset of inter-cluster tunneling (Josephson coupling). The leading pair wavefunction evolves from a localized cluster mode to a coherent, system-spanning $d$-wave-like structure.
• Scaling: At low temperatures, the total condensate weight ($\epsilon_{tot}$) scales algebraically with system length ($L^\nu$), consistent with the formation of a quasi-1D condensate.

B. Charge Clustering and "Forestalled" Phase Separation

• At high temperatures, holes are dispersed ($m=1$).
• Upon cooling, holes aggregate into larger clusters (increasing mean cluster size $\bar{m}$).
• Crucially, this aggregation does not immediately lead to macroscopic phase separation. Instead, it forms a forestalled phase separation where charge clusters fluctuate.
• At the lowest temperatures, these clusters reorganize into static stripes (period-8 charge order), and the mean cluster size stabilizes.

C. Pair-Charge Locking

• The correlation coefficient $\Lambda$ between hole density and the pair wavefunction amplitude is strongly positive at low temperatures and decreases as temperature rises.
• This confirms that the superconducting order parameter is "locked" to the charge inhomogeneities. The pairing nucleates on the holes; the holes do not merely sit in a pre-existing superconducting sea.

D. Temperature Regimes

1. $T/t > 0.2$: Weak correlations, no significant clustering, negligible pairing.
2. $0.03 < T/t < 0.2$ (Pseudogap/Intermediate): Strong AFM correlations build up. Holes form fluctuating clusters. Pairing is strong but localized on these clusters (fragmented condensate).
3. $T/t < 0.03$: Stripe order sets in. Clusters merge into stripes. Pairing becomes delocalized and coherent across the system.

5. Significance and Implications

• Unifying Experimental Observations: The findings provide a theoretical basis for several experimental anomalies in cuprates:
  • STM: Explains the observation of "nanoscale phase separation" and local pairing gaps above $T_c$ (e.g., in BSCCO), showing that pairing can exist locally on clusters before global coherence is established.
  • NMR: Supports evidence of charge clustering and "disconnected metallic islands" in lightly doped systems.
  • Strange Metal: Suggests that charge inhomogeneity is a crucial ingredient for the strange metal regime, supporting recent mesoscopic theories.
• Microscopic Picture: The paper proposes a clear narrative: Magnetic correlations at intermediate temperatures drive hole attraction $\rightarrow$ formation of hole clusters $\rightarrow$ local pairing on these clusters $\rightarrow$ tunneling/hybridization between clusters $\rightarrow$ global coherent superconductivity (stripes).
• Methodological Advance: It demonstrates the power of METTS in resolving the interplay between charge order and superconductivity at finite temperatures, a regime difficult to access with traditional ground-state methods like DMRG.

In conclusion, the paper establishes that the parent state of the superconducting stripe phase is a disordered, charge-inhomogeneous state with localized pairs, and that the transition to superconductivity is a process of establishing coherence between these pre-formed local pairs.