Improving Single Excitation Fidelity in Rydberg Superatoms for Efficient Single Photon Emission

1. Problem Statement

The paper addresses the challenge of generating deterministic single photons from a Rydberg atomic ensemble coupled to an optical cavity. While such systems offer a scalable pathway for quantum networks, their performance is fundamentally limited by decoherence, specifically:

• Imperfect Rydberg Blockade: In large ensembles, the blockade is not perfect, leading to unwanted double excitations (two atoms excited to Rydberg states simultaneously). These double excitations degrade the indistinguishability of emitted photons.
• Thermal Dephasing: Finite atomic temperatures cause random motion, leading to non-Markovian dephasing that destroys the phase coherence required for directional photon emission.
• Current Limitations: Previous experimental benchmarks (e.g., Ref. [25]) achieved a single excitation probability of roughly 77%, with significant population loss into leakage channels (double excitations and thermal continua).

The core objective is to enhance the fidelity of preparing the collective single-excitation state ($|R_0\rangle$) by suppressing leakage into doubly-excited states ($|S_k\rangle$) and mitigating decoherence effects through advanced pulse shaping.

2. Methodology

The authors employ a combination of analytical modeling, numerical optimization, and quantum control theory:

• System Modeling:

  • The system is modeled using a Lindblad master equation within a truncated Hilbert space.
  • To capture non-Markovian thermal dephasing, the model includes a subset of asymmetric singly-excited states ($|R_j\rangle$) that couple to a thermal continuum ($M_{th}$).
  • To model imperfect blockade, the system includes doubly-excited states ($|S_k\rangle$) that couple to an interaction-induced continuum ($M$).
  • A convergence analysis determined that a basis size of $n_{max} = 8$ (8 asymmetric and 8 doubly-excited states) is sufficient to accurately capture the dynamics while remaining computationally efficient.

• Control Strategy: DRAG (Derivative Removal by Adiabatic Gate):

  • The authors adapt the DRAG technique, originally developed for superconducting qubits, to the Rydberg platform.
  • Mechanism: DRAG introduces a counter-diabatic term (an out-of-phase component) to the primary control pulse. This term cancels out diabatic transitions to leakage states.
  • Pulse Design:
    • Primary Pulse: A sine-squared pulse ($\Omega_I$).
    • Perturbative DRAG: Adds a term proportional to the time derivative of the primary pulse ($\dot{\Omega}_I$).
    • Non-perturbative DRAG: Uses a Givens rotation formalism to handle strong driving fields where the Rabi frequency approaches the energy gap to leakage states.
  • Optimization: The DRAG coefficient ($\alpha$), pulse amplitude ($A$), and laser detuning ($\Delta_d$) are optimized using a two-step procedure (global search followed by local gradient-based refinement) to minimize population in leakage states.

• Benchmarking:

  • The DRAG results are compared against GRAPE (Gradient Ascent Pulse Engineering), a fully numerical optimal control method, to determine if DRAG is near the theoretical control limit.

3. Key Contributions

• Cross-Platform Adaptation: Successfully adapted the DRAG technique from superconducting circuits to a neutral atom Rydberg platform, demonstrating its versatility in suppressing leakage in multi-level atomic systems.
• Identification of Dominant Leakage: Analyzed the system dynamics to identify that double excitations (specifically state $|S_3\rangle$ or $|S_4\rangle$ depending on parameters) are the dominant source of irreversible population loss on short timescales, rather than thermal dephasing.
• Parameter Regime Optimization: Identified a specific regime of pulse duration ($T$) and atomic ensemble radius ($\sigma$) that balances competing constraints:
  • Short pulses reduce thermal dephasing but require high power, increasing double excitations.
  • Small radii strengthen the blockade but increase collisional losses.
  • The optimal regime found is $T = 0.5 \, \mu\text{s}$ and $\sigma = 4.3 \, \mu\text{m}$.
• Experimental Feasibility: Demonstrated that DRAG produces smooth, analytically defined pulse shapes that are easier to implement experimentally compared to the complex, oscillatory pulses often generated by GRAPE.

4. Results

• Fidelity Improvement:

  • Using standard sine-squared pulses with the experimental parameters ($T=0.25\,\mu\text{s}, \sigma=4.6\,\mu\text{m}$), the single excitation probability is 85.1%.
  • Applying non-perturbative DRAG to these parameters improves the probability to 87.7%.
  • By optimizing the system parameters (moving to $T=0.5\,\mu\text{s}, \sigma=4.3\,\mu\text{m}$) and applying non-perturbative DRAG, the single excitation probability reaches 91.9%.
  • This represents a significant improvement over the 77% theoretical benchmark reported in previous work (Ref. [25]).

• Leakage Suppression:

  • DRAG effectively suppresses population transfer to the dominant leakage state ($|S_k\rangle$) and the associated continuum ($M$).
  • In the optimized regime, leakage into the interaction-induced continuum is reduced from ~10.8% (experimental params) to ~3.8%.

• Benchmarking against GRAPE:

  • GRAPE optimization yielded results nearly identical to DRAG (92.2% vs. 91.9% for optimized parameters).
  • This confirms that DRAG operates close to the fundamental optimal control limit imposed by the system's intrinsic decoherence, despite using far fewer tunable parameters (3 vs. 150+).

• Dephasing-Free Limit:

  • In a theoretical dephasing-free limit, DRAG achieved 97.1% fidelity, indicating that the current bottleneck is the rapid Markovian dephasing (population loss to the continuum before the DRAG correction can act), not the control strategy itself.

5. Significance

• High-Fidelity Single Photon Sources: The work provides a roadmap for achieving near-deterministic single-photon emission, a critical requirement for scalable quantum communication and distributed quantum computing.
• Robustness to Decoherence: It demonstrates that analytical control techniques like DRAG can effectively mitigate leakage even in systems dominated by rapid, non-unitary decoherence, challenging the notion that such systems require complex numerical control.
• Practical Implementation: By showing that DRAG pulses are smooth and experimentally feasible, the paper bridges the gap between theoretical optimal control and practical implementation in cold atom experiments.
• Future Directions: The authors suggest that further improvements could be achieved by increasing the principal quantum number ($n$) to strengthen the Rydberg blockade or by spatially filtering the atomic ensemble to remove weakly interacting atoms in the tails of the density distribution.

In conclusion, this paper establishes DRAG pulse shaping as a highly effective, experimentally viable method for enhancing the fidelity of Rydberg superatom excitation, pushing single-photon source performance from ~77% to nearly 92%, thereby approaching the fundamental limits set by decoherence.