Skip to main content
QUICK REVIEW

[Paper Review] Deterministic Storage and Retrieval of Telecom Quantum Dot Photons Interfaced with an Atomic Quantum Memory

S. E. Thomas, L. Wagner|arXiv (Cornell University)|Mar 7, 2023
Quantum Information and Cryptography4 citations
TL;DR

This paper demonstrates the first deterministic storage and on-demand retrieval of single photons from an InAs semiconductor quantum dot at telecom wavelengths (1529.3 nm) into a hot rubidium vapor atomic quantum memory using the ORCA protocol. With a total internal memory efficiency of 12.9±0.4% and a signal-to-noise ratio of 18.2±0.6, the work establishes a critical hybrid quantum interface for scalable photonic quantum networks and quantum repeaters.

ABSTRACT

A hybrid interface of solid state single-photon sources and atomic quantum memories is a long sought-after goal in photonic quantum technologies. Here we demonstrate deterministic storage and retrieval of photons from a semiconductor quantum dot in an atomic ensemble quantum memory at telecommunications wavelengths. We store single photons from an InAs quantum dot in a high-bandwidth rubidium vapour based quantum memory, with a total internal memory efficiency of $(12.9 \pm 0.4) \%$. The signal-to-noise ratio of the retrieved photons is $18.2\pm 0.6$, limited only by detector dark counts. This demonstration paves the way to quantum technologies that rely on distributed entanglement, and is especially suited for photonic quantum networks.

Motivation & Objective

  • Enable deterministic, high-fidelity storage and on-demand retrieval of single photons from semiconductor quantum dots in atomic quantum memories.
  • Address the challenge of matching quantum dot emission bandwidth (GHz) to atomic memory bandwidth for efficient quantum interface.
  • Achieve operation at telecom wavelengths (1529.3 nm) to enable low-loss transmission over long-distance optical fibers.
  • Overcome limitations of prior systems by demonstrating a hybrid quantum light-matter interface with high signal-to-noise ratio and low noise floor.
  • Pave the way for distributed quantum networks and quantum repeater architectures based on quantum dot sources and atomic ensembles.

Proposed method

  • Utilize the off-resonant cascaded absorption (ORCA) protocol in a hot rubidium vapor to store and retrieve photons at 1529.3 nm.
  • Employ spectral and temporal filtering to match the quantum dot photon emission (Gaussian-shaped, FWHM ≈ 300 ps) to the memory's 1 GHz bandwidth.
  • Apply an electro-optic modulator (EOM) to shape the photon wavepacket from an exponential decay (1/e ≈ 0.85 ns) into a Gaussian with FWHM = 300 ps.
  • Use a Fabry-Perot etalon for spectral filtering, tuned to transmit at 1529.3 nm with 85% transmission for CW light and 3.0±0.5% for inhomogeneously broadened single photons.
  • Implement a time-resolved detection scheme with a 500 ps integration window to balance signal-to-noise ratio and photon count rate.
  • Measure memory efficiency using a calibrated single-photon detector and account for optical losses via a detailed loss budget (total transmission η_optical = 0.153±0.009).
Figure 1: Schematic of the experimental set-up for the quantum dot - quantum memory interface. (a) Energy level scheme for the Telecom ORCA quantum memory protocol in rubidium vapour. (b) Scheme of the semiconductor QD sample with semiconductor bottom DBR, metamorphic buffer (MMB) and oxide top DBR.
Figure 1: Schematic of the experimental set-up for the quantum dot - quantum memory interface. (a) Energy level scheme for the Telecom ORCA quantum memory protocol in rubidium vapour. (b) Scheme of the semiconductor QD sample with semiconductor bottom DBR, metamorphic buffer (MMB) and oxide top DBR.

Experimental results

Research questions

  • RQ1Can single photons from a semiconductor quantum dot be deterministically stored and retrieved in an atomic quantum memory at telecom wavelengths?
  • RQ2What is the achievable memory efficiency and signal-to-noise ratio when matching GHz-bandwidth quantum dot emission to a high-bandwidth atomic memory?
  • RQ3How does temporal and spectral filtering affect the compatibility and performance of the quantum dot–memory interface?
  • RQ4To what extent does the noise floor of the ORCA protocol limit the performance of the hybrid quantum interface?
  • RQ5Can this system enable scalable, long-distance quantum networks through efficient, on-demand photon storage and retrieval?

Key findings

  • The total internal memory efficiency was measured at (12.9±0.4)%, representing a key milestone in deterministic quantum memory operation with solid-state sources.
  • The signal-to-noise ratio of the retrieved photons reached 18.2±0.6, limited only by detector dark counts, indicating high-fidelity retrieval.
  • The system achieved a 500 ps integration window compromise yielding a count rate of (22±1) s⁻¹ and a signal-to-noise ratio of 18.2±0.6.
  • After accounting for temporal filtering losses, the memory efficiency reduced to (10.9±0.3)% when using the full input pulse, confirming robustness to filtering effects.
  • The spectral filtering efficiency for inhomogeneously broadened single photons was measured at (3.0±0.5)%, consistent with theoretical expectations.
  • The optical loss budget was quantified with a total transmission of η_optical = 0.153±0.009 from source to detectors, with major losses from EOM, fiber coupling, and filters.
Figure 2: Characterisation of single-photon source (a) Spectrum of photons from QD without (blue) and with (red) spectral filtering from the etalon, normalised to the maximum intensity. The dashed black line shows the target wavelength of the memory. (b) The second-order autocorrelation, $g^{(2)}(\t
Figure 2: Characterisation of single-photon source (a) Spectrum of photons from QD without (blue) and with (red) spectral filtering from the etalon, normalised to the maximum intensity. The dashed black line shows the target wavelength of the memory. (b) The second-order autocorrelation, $g^{(2)}(\t

Better researchstarts right now

From reading papers to final review, dramatically reduce your research time.

No credit card · Free plan available

This review was created by AI and reviewed by human editors.