Spectral design principles for local-excitation retention in impurity-assisted atomic arrays

This paper establishes spectral design principles for enhancing local-excitation retention in impurity-assisted atomic arrays by utilizing biorthogonal eigenmode decomposition to reveal that survival dynamics depend on both eigenmode decay rates and initial-state overlaps, leading to a surrogate objective that successfully guides the inverse design of nontrivial aperiodic atomic configurations.

The Gist

The Big Picture: The "Quiet Party" in a Crowd

Imagine you are at a crowded party (the atomic array). Suddenly, one person in the middle (the storage atom) starts shouting a secret message (the excitation).

In a normal room, that person would shout, and the sound would quickly fade away as it bounces off walls and dissipates. But in this quantum party, the people are arranged in a very specific way. If they stand in the right spots, they can "talk" to each other in a way that cancels out the sound waves, trapping the secret message inside the group for a very long time. This is called subradiance (or "quietness").

The goal of this research is to figure out how to arrange the crowd so that the secret message stays trapped for as long as possible, without leaking out.

The Problem: It's Not Just About the "Quietest" Person

Previous scientists thought the solution was simple: "Find the arrangement where the group is the quietest possible." They looked for the single "slowest decay rate" (the longest time it takes for the sound to fade).

The authors of this paper say: "That's not enough!"

Here is why:
Imagine the crowd is made up of different "choirs."

• Choir A is very quiet (slow decay), but the person shouting the secret doesn't know how to sing with them. The secret gets lost immediately.
• Choir B is moderately quiet, but the person shouting fits perfectly with them. The secret stays.
• Choir C and Choir D are both very quiet, and the person shouts to both of them at the same time. Because they are slightly out of sync, they start arguing (interfering). This causes the secret to bounce back and forth wildly, making the "loudness" go up and down like a rollercoaster. This is bad because if you check the secret at the wrong moment, it might look like it's gone.

The Lesson: You don't just want the quietest group. You want the group where the secret fits perfectly into one specific, quiet choir, and you want to avoid having the secret split between two choirs that will start arguing.

The Solution: A "Spectral Design" Recipe

The authors created a new "recipe" (a mathematical formula) to design the perfect crowd arrangement. They call it a Spectral Surrogate.

Think of this recipe as a scorecard for designing the party layout. It checks two things:

1. The Volume: Is the chosen choir quiet enough? (Low decay rate).
2. The Connection: Does the person shouting fit perfectly with that choir? (High overlap).

If the scorecard sees that the secret is split between two choirs (which causes the "rollercoaster" effect), it gives a bad score. It forces the design to focus on funneling the secret into one single, super-quiet choir.

The Experiment: Rearranging the Furniture

To test this, the researchers used a computer to "move the furniture" (the atoms) around.

• The Constraint: You can't put the furniture too close together (like you can't put two chairs in the same spot). This is a real-world rule called the minimum-distance constraint.
• The Result: They started with a simple circle of atoms. The computer rearranged them into weird, non-repeating patterns (aperiodic configurations).

These new patterns looked nothing like a perfect circle or a grid. They looked like a sunflower or a random scattering. But when they tested them:

• The secret message stayed trapped much longer than in the old shapes.
• The "rollercoaster" bouncing stopped. The message just slowly faded away, which is exactly what you want for a memory.

Why This Matters (The "So What?")

This research gives us a new way to build Quantum Memories.

• Old way: Try to find the absolute quietest spot.
• New way: Use this "scorecard" to arrange atoms so the information flows smoothly into one long-lived state without getting confused by interference.

It's like realizing that to keep a secret safe, you don't just need a soundproof room; you need a room where the person telling the secret knows exactly which wall to whisper against, and no one else is whispering back at the same time.

Summary in One Sentence

The paper teaches us that to store quantum information for a long time, we shouldn't just look for the quietest group of atoms, but rather design a specific, slightly messy arrangement that funnels all the energy into one single, quiet "mode" while avoiding the chaotic bouncing that happens when energy splits between multiple modes.

Technical Summary

1. Problem Statement

The paper addresses the challenge of local-excitation retention in cooperative atomic systems, specifically in an "impurity-assisted" setting. In this scenario, a single "storage atom" is initially excited and coupled to a surrounding array of ground-state atoms. The goal is to retain the excitation in the storage atom for as long as possible without external write or retrieval fields, relying solely on the cooperative radiative effects (subradiance) of the array.

Key Challenges Identified:

• Multimode Dynamics: The survival of the excitation is not determined solely by the smallest collective decay rate (the "slowest" mode). Instead, the dynamics involve multiple interfering collective modes.
• Oscillatory Behavior: When multiple long-lived modes have comparable weights (overlaps with the initial state), they interfere, causing large oscillations in the excitation probability. This makes the retention time highly sensitive to the measurement time, hindering reliable storage.
• Limitations of Existing Metrics: Previous optimization efforts often focused on minimizing the minimum decay rate alone, which fails to predict performance in multimode regimes where overlap strength and interference are critical.

2. Methodology

The authors employ a spectral design approach based on the eigenmode decomposition of an effective non-Hermitian Hamiltonian.

• Physical Model:

  • A system of $N$ two-level atoms interacting with the vacuum electromagnetic field.
  • The dynamics are governed by a non-Hermitian Hamiltonian $\hat{H} = \sum_{jj'} (J_{jj'} - i\frac{1}{2}\Gamma_{jj'}) \hat{\sigma}_{eg}^j \hat{\sigma}_{ge}^{j'}$, where $J_{jj'}$ is the cooperative Lamb shift and $\Gamma_{jj'}$ is the cooperative decay rate.
  • The initial state is a single excitation localized on a central "storage atom."

• Biorthogonal Eigenmode Decomposition:

  • The survival amplitude is expressed as a sum over collective eigenmodes: $\langle \psi(0)|\psi(t)\rangle = \sum_\ell w_\ell e^{-i\kappa_\ell t}$.
  • Here, $\kappa_\ell$ are complex eigenvalues (with decay rates $\Gamma_\ell = -2\text{Im}(\kappa_\ell)$), and $w_\ell$ are complex weights representing the overlap between the initial state and the right/left eigenstates.
  • The survival probability $p_e(t)$ depends on both the decay rates $\Gamma_\ell$ and the weights $|w_\ell|^2$.

• Spectral Surrogate Objective:

  • To avoid direct time-domain optimization (which is sensitive to oscillation phases), the authors introduce a differentiable spectral surrogate cost function ($F$):
    $$F = \alpha \sum_\ell p_\ell \log(\Gamma_\ell/\gamma_0) - \beta \sum_\ell p_\ell \log p_\ell$$
    where $p_\ell = |w_\ell| / \sum |w_\ell|$.
  • Term 1 (Weighted Decay): Minimizes the decay rate of modes that have significant overlap with the initial state.
  • Term 2 (Entropy): Minimizes the Shannon entropy of the weight distribution, promoting the concentration of the initial excitation into a single dominant subradiant mode to suppress multimode interference.

• Inverse Design:

  • The surrogate is used in a constrained optimization scheme to find optimal atomic positions.
  • Constraints: Minimum interatomic distance ($r_{min}$) to ensure experimental feasibility.
  • Algorithm: Sequential Least Squares Quadratic Programming (SLSQP) starting from perturbed ring configurations.

3. Key Contributions

1. Identification of Spectral Criteria: The paper establishes that successful local-excitation retention requires a dual condition: (a) a small decay rate for the relevant mode, and (b) a high overlap (weight) of the initial state with that specific mode. The smallest decay rate alone is insufficient.
2. Explanation of Oscillations: The authors analytically demonstrate that large oscillations in excitation probability arise when two or more long-lived modes have comparable weights. The oscillation frequency is determined by the difference in the real parts of the eigenvalues.
3. Development of a Spectral Surrogate: They propose a compact, physically motivated objective function that balances lifetime extension and the suppression of multimode interference, enabling efficient inverse design.
4. Proof-of-Principle Inverse Design: The method successfully generates non-trivial, aperiodic atomic configurations that outperform standard periodic (square lattice) and simple aperiodic (sunflower, ring) geometries under experimental constraints.

4. Results

• Comparison of Geometries:
  • Square Lattice: Contains the mode with the absolute minimum decay rate, but the initial excitation is distributed across many modes (low weight on any single mode), leading to poor retention.
  • Sunflower/Ring: Exhibit a dominant mode with a large weight. The sunflower geometry showed the best retention due to a combination of a small decay rate and high weight, though it still exhibited some oscillations.
• Optimized Configurations:
  • Starting from a ring geometry, the optimization produced aperiodic, concentric structures that differ significantly from the initial seed.
  • Performance: For $N=12$ atoms and a minimum distance $r_{min}/\lambda_0 = 0.1$, the optimized structure achieved a storage atom excitation probability of 0.65 at time $t^*$ (where $\gamma_0 t^* = 30$), compared to 0.14 for the reference ring and 0.01 for the square lattice.
  • Mode Characteristics: The optimized structures concentrate the initial weight ($|w_{max}|^2 \approx 0.65$) onto a single mode with a decay rate reduced by two orders of magnitude ($\Gamma/\gamma_0 \approx 2.4 \times 10^{-5}$) compared to the reference.
• Physical Mechanism: Analysis of the dominant eigenmodes reveals an out-of-phase (destructive interference) pattern between the central storage atom and surrounding atoms. This suppresses far-field radiation, effectively trapping the excitation.
• Robustness: The optimized structures remain robust against small positional fluctuations (up to 1% of the wavelength), maintaining high retention probabilities.

5. Significance

• Paradigm Shift in Design: The work moves beyond simple "minimum decay rate" optimization to a holistic spectral design approach that accounts for mode overlaps and interference. This is crucial for multimode quantum systems.
• Experimental Feasibility: By incorporating minimum-distance constraints and generating aperiodic configurations, the study bridges the gap between theoretical quantum control and experimental implementation in optical lattices or nanophotonic waveguides.
• Quantum Memory Applications: The findings provide a blueprint for designing high-bandwidth, long-lived quantum memories where photon storage fidelity is enhanced by engineering the spectral properties of the atomic array rather than just the geometry.
• Generalizability: The spectral surrogate framework can be applied to other cooperative light-matter systems where balancing long lifetimes with robustness against interference is required.

In conclusion, the paper demonstrates that spectral engineering—specifically funneling excitation into a single, long-lived, high-overlap mode while suppressing competing modes—is the key to maximizing local-excitation retention in impurity-assisted atomic arrays.