Lattice Unitarity: Saturated Collisional Resistivity in Hubbard Metals

This study demonstrates that ultracold fermions in a three-dimensional optical lattice exhibit a saturation of collisional resistivity at a value independent of interaction strength in the strongly interacting metallic regime, a phenomenon quantitatively explained by a renormalized two-body scattering matrix model.

The Gist

Imagine a crowded dance floor where everyone is trying to move from one side to the other. In a normal room, if you bump into someone, you might stumble, but you keep moving. Now, imagine this dance floor is made of a grid of tiny, invisible boxes (a "lattice"), and the dancers are ultra-cold atoms that are so cold they act like waves rather than solid balls.

This paper is about a team of scientists who put these atoms on a dance floor and tried to push them with a gentle, rhythmic shove (an electric current). They wanted to see how much the atoms resisted moving (resistivity) when they started bumping into each other really hard.

Here is the story of what they found, explained simply:

1. The Setup: A Perfect Dance Floor

The scientists used a special laser setup to create a 3D grid of light. Into this grid, they placed potassium atoms. They could tune how much these atoms "hated" each other (their interaction strength).

• Low Hating: When the atoms barely noticed each other, they moved easily, like people walking through an empty hallway.
• High Hating: When they turned up the "hate," the atoms started bumping into each other violently. In normal physics, you'd expect that if they bump harder, they should get stuck and stop moving almost entirely. The resistance should go up forever.

2. The Surprise: The "Speed Limit" of Resistance

The scientists expected that as they made the atoms bump harder and harder, the resistance would keep climbing. It's like trying to run through a crowd that keeps getting more aggressive; you'd think you'd eventually get stuck in place.

But that's not what happened.

They found that the resistance hit a ceiling. No matter how much they increased the "bumping" (interaction strength), the atoms stopped getting slower. The resistance saturated. It hit a maximum speed limit and refused to go any higher.

The Analogy: Imagine you are trying to push a shopping cart through a crowd.

• Normal Physics: If the crowd gets more aggressive, you get stuck.
• This Experiment: It's as if the crowd has a rule: "We can bump into each other as hard as we want, but we can't block the path more than this specific amount." The atoms found a way to keep flowing, even when they were colliding with maximum force.

3. Why Did This Happen? The "Quantum Traffic Jam"

Why did the resistance stop growing? The scientists realized it's because of the grid itself.

In free space (no grid), if particles collide, they can scatter in any direction. But on a grid, the atoms are confined to specific paths. When they collide, they can't just scatter anywhere; they have to follow the rules of the grid.

The scientists discovered a concept they call "Lattice Unitarity."

• Think of it like a highway with a strict speed limit. Even if the drivers (atoms) are driving recklessly (strong interactions), the road itself (the lattice) limits how much they can slow down the traffic flow.
• The "bumping" becomes so intense that the atoms essentially forget they are individual particles and start moving as a coordinated wave. The grid forces them to find a way to move, preventing the traffic jam from becoming total gridlock.

4. The Temperature Factor

They also checked what happens when the atoms get hotter (more energetic).

• Cold Atoms: They move slowly and follow the grid rules closely.
• Hot Atoms: As they got hotter, the resistance started to rise again, but this time it was because the atoms were moving so fast they were hitting the "walls" of the grid more often. It was like the dance floor getting so crowded with energetic dancers that they started tripping over each other again, but in a predictable, linear way.

5. Why Does This Matter?

This discovery is a big deal for two reasons:

1. It Solves a Puzzle: For a long time, physicists have been trying to understand "bad metals"—materials that conduct electricity poorly and don't follow the usual rules. This experiment shows that even in the most chaotic, strongly interacting systems, there is a fundamental limit to how bad the conductivity can get.
2. A New Benchmark: The scientists didn't just guess this; they built a mathematical model that perfectly predicted what they saw. It's like building a simulation of a traffic jam and having it match real-life data perfectly. This gives us a new tool to understand how electrons move in complex materials, like the ones used in superconductors or quantum computers.

The Takeaway

In simple terms: Even when particles are fighting each other as hard as physics allows, the structure of the universe (the lattice) puts a cap on how much they can slow down. There is a "maximum resistance," and once you hit it, making the particles fight harder doesn't make them stop moving any more. It's a fundamental speed limit for chaos.

Technical Summary

1. Problem Statement

The study addresses the fundamental question of transport properties in strongly interacting fermionic systems, specifically within the Fermi-Hubbard model (HM). While the HM is a cornerstone for understanding high-temperature superconductivity and Mott insulators, its transport behavior in the strongly interacting, high-temperature regime ($U \gtrsim t$ and $T \gtrsim t$, where $U$ is on-site interaction and $t$ is tunneling energy) remains challenging to characterize theoretically and experimentally.

Key challenges include:

• Breakdown of Perturbation Theory: Standard perturbative approaches (like the Born approximation) predict that resistivity should scale quadratically with interaction strength ($\rho \propto U^2$). However, in the unitary limit (infinite scattering length), physical cross-sections must remain bounded due to unitarity constraints.
• Lattice vs. Free Space: In free space, the unitary limit is reached when the scattering length diverges. In an optical lattice, the discrete band structure and momentum dependence of the scattering matrix ($T$-matrix) complicate this, raising the question of whether a "lattice unitarity" bound exists and how it manifests in transport.
• Lack of Microscopic Benchmarks: There is a scarcity of experimental data where transport coefficients in strongly correlated atomic systems can be quantitatively compared to first-principles calculations without free fitting parameters.

2. Methodology

The authors utilized ultracold fermionic $^{40}$K atoms trapped in a three-dimensional cubic optical lattice to simulate the single-band Hubbard model.

• Experimental Setup:

  • System: Spin-balanced mixture of $^{40}$K in a lattice with period $a_L = 0.53$ µm and depth $2.5 E_R$.
  • Regime: Low filling ($n \lesssim 0.1$), ensuring the system is strongly interacting ($U \gtrsim t$) but weakly correlated (metallic), allowing comparison with two-body physics.
  • Interaction Tuning: Magnetic Feshbach resonance near 202 G was used to tune the $s$-wave scattering length $a_S$, varying $U/t$ from $\approx 1$ to $\approx 6$.
  • Temperature Control: Samples were prepared at temperatures $T/t$ ranging from $\approx 1$ to $\approx 4$.

• Transport Measurement Technique:

  • Global Conductivity: Instead of local current measurements, the team measured global conductivity $\sigma(\omega)$ by applying a sinusoidal force via displacement of the optical dipole trap (XDT).
  • In Situ Imaging: The center-of-mass response of the atomic cloud was tracked using fluorescence imaging. The response was decomposed into in-phase ($S$) and out-of-phase ($C$) components relative to the drive.
  • Extraction: Real ($\text{Re}[\sigma]$) and imaginary ($\text{Im}[\sigma]$) conductivity were derived, and resistivity $\rho(\omega) = 1/\sigma(\omega)$ was calculated.
  • Kubo Fit: Spectra were fitted using a Kubo-type response function with a phenomenological relaxation time $\tau_Q$ to extract the current dissipation rate $\Gamma = \tau_Q^{-1}$.

• Theoretical Framework:

  • Kinetic Theory: A Boltzmann equation approach was developed using the method of moments.
  • Non-Perturbative $T$-Matrix: The collision integral utilized the exact two-body $T$-matrix in a lattice, calculated via a generalized Fermi golden rule. This accounts for the full momentum and energy dependence of scattering, going beyond the Born approximation ($T \approx U$).
  • Ansatz: A shifted equilibrium distribution ansatz was used to derive equations of motion for the center of mass, linking the microscopic collision rate to the macroscopic dissipation rate $\Gamma$.

3. Key Contributions

1. Observation of Resistivity Saturation: The paper provides the first experimental evidence that current dissipation in a strongly interacting lattice metal saturates as interaction strength increases, deviating significantly from the perturbative $U^2$ scaling.
2. Definition of "Lattice Unitarity": The authors define a "lattice unitarity" limit where the scattering amplitude is bounded by the lattice structure itself. Unlike free space, this bound is momentum-dependent, meaning the thermal ensemble does not reach the theoretical maximum scattering cross-section even at $U \to \infty$.
3. Quantitative Benchmark: The study achieves a rare quantitative agreement between experimental transport data and a first-principles calculation using the full non-perturbative $T$-matrix, with no free fitting parameters for the interaction strength or temperature dependence.
4. Separation of Mechanisms: The work distinguishes between the saturation of the scattering rate (dynamical) and the renormalization of the effective mass, clarifying that the observed saturation is a collisional effect.

4. Key Results

• Saturation of Dissipation Rate:

  • As $U/t$ increased from $\approx 1$ to $\approx 6$, the measured current dissipation rate $\Gamma$ increased initially but then saturated, becoming nearly independent of $U$.
  • At the highest interaction strengths, the dissipation rate reached roughly one-third of the theoretical "lattice unitary" bound.
  • The measured resistivity deviated from the first Born approximation ($\rho \propto U^2$) by a factor of up to seven in the strong coupling regime.

• Temperature Dependence:

  • In the saturated regime ($U^2 \gg t^2$), the real resistivity $\text{Re}[\rho]$ was found to increase linearly with temperature ($T$).
  • This linear scaling is attributed to the temperature dependence of the effective mass $m^*$ and the collision rate $\Gamma$, rather than a saturation of the scattering cross-section itself.
  • The data confirmed that Umklapp scattering (momentum transfer to the lattice) dominates the resistivity in this regime.

• Theoretical Validation:

  • The dissipation model, utilizing the full lattice $T$-matrix, perfectly reproduced the experimental data across the entire range of $U$ and $T$.
  • The model predicted a dimensionless dissipation efficiency $C_\Gamma$ that matched the observed values, confirming that the saturation is a consequence of the bounded nature of the $T$-matrix in a periodic potential.

5. Significance

• Fundamental Physics: This work provides a clear microscopic understanding of bounded resistivity in low-density metals. It demonstrates that in a lattice, the unitarity limit is not a simple divergence of scattering length but a complex interplay between interaction strength and band structure.
• Benchmark for Strongly Correlated Systems: The quantitative agreement between experiment and the non-perturbative two-body model serves as a critical benchmark for theories of strongly correlated electronic systems, including high-$T_c$ superconductors and strange metals.
• Methodological Advance: The technique of measuring global conductivity via in situ imaging in a disorder-free optical lattice offers a powerful tool for studying transport in regimes inaccessible to solid-state materials (e.g., without phonons or impurities).
• Future Directions: The results open pathways for studying hydrodynamic transport, the emergence of "bad metal" behavior, and transport in lower-dimensional Hubbard systems, as well as probing the limits of the single-band approximation at higher densities.

In summary, Corapi et al. have experimentally demonstrated that strong interactions in a lattice do not lead to infinite resistivity but rather to a saturated, interaction-independent dissipation rate, governed by the unique "lattice unitarity" of Bloch wave scattering. This finding bridges the gap between microscopic scattering theory and macroscopic transport in quantum materials.