On-Demand Millisecond Storage of Spectro-Temporal Multimode Telecom Photons

The realization of scalable quantum networks for distribution of entanglement over long distances hinges on quantum repeaters. To outperform the exponential transmission loss in optical fibers, quantum repeaters must employ multiplexing schemes in the temporal, spectral, or spatial domain. The performance of such a multiplexed scheme is contingent on efficient quantum memories offering both extended storage times and large multimode capacities. In this work, we experimentally demonstrate such a memory operating at telecom wavelength using an Er$^{3+}$:Y${2}$SiO${5}$ crystal. Using single-photon detectors, we record on-demand storage and recall of weak coherent pulses for up to $1$ ms, exceeding that of previously reported quantum memories based on Er$^{3+}$. The memory exhibits an efficiency of 10.36% at 300 $μ$s storage time with a signal-to-noise ratio of $10.9$. We further showcase its multimode capacity by storing 20 temporal and 3 spectral modes simultaneously with on-demand and selective recall capabilities, essential for a scalable quantum repeater architecture.

💡 Research Summary

The paper presents a significant advance in quantum memory technology by demonstrating an on‑demand, millisecond‑scale storage device operating at the telecom C‑band (≈1550 nm) using a 5 ppm ¹⁶⁷Er³⁺:Y₂SiO₅ crystal. The authors employ the Chirped Pulse Phase Encoding (CPPE) protocol, which uses two strong, adiabatically chirped control pulses to invert the atomic ensemble, suppress the primary photon‑echo, and retrieve the stored weak coherent pulses at a programmable time. The control pulses have a sech amplitude envelope and a linear frequency sweep, providing robustness against power and frequency fluctuations. The crystal is cooled to ~500 mK in a He‑3 cryostat and subjected to a 3 T magnetic field oriented at 135° in the D₁–D₂ plane, which splits the spin sub‑levels and enables long optical coherence.

Experimentally, weak coherent probe pulses (0.75 µs FWHM Lorentzian shape) with an average photon number of ~720 before the crystal (≈144 photons actually absorbed after accounting for ~20 % total transmission) are stored. The memory exhibits an exponential decay of efficiency with storage time, yielding 10.36 % efficiency and a signal‑to‑noise ratio (SNR) of 10.9 at 300 µs, and 1.42 % efficiency with SNR 1.5 at 1 ms. The measured optical coherence time is T₂ = 858 ± 80 µs. Noise is dominated by spontaneous emission from imperfect π‑pulses of the chirped control fields. The authors note that laser frequency instability broadens the retrieved echo, which could be mitigated by cavity locking.

Beyond single‑mode performance, the work showcases multimode capacity in both temporal and spectral domains. A train of 25 temporally separated pulses (7 µs spacing) is stored for 800 µs with an average efficiency of 2.67 % and SNR 9.18. Spectrally, three independent memory channels are created within the inhomogeneous broadening, spaced by 4 MHz, each storing the same temporal train. Selective recall is achieved by gating the output with acousto‑optic modulators after the second control pulse, allowing any of the three spectral channels to be retrieved independently. The per‑channel efficiencies range from 1.20 % to 1.66 %. Combining the 20 temporal modes with the three spectral channels yields a total multimode capacity of Mₜ × Mₛ ≈ 60 modes, which meets the threshold identified for practical quantum repeater operation.

The authors also develop a simple analytical model for entanglement distribution success probability in a quantum repeater architecture. The model incorporates memory efficiency η₀, coherence time T₂, storage time Tₛ (set by link distance L, number of elementary links nₗ, and light speed in fiber), and multimode capacity M. Using realistic parameters (η₀ ≥ 65 %, T₂ ≥ 3 ms, M ≥ 60), the model predicts that a repeater with elementary links of ~200 km can outperform direct transmission, provided the memory can store for ≈1 ms. While the current experiment achieves η₀ ≈ 23 % and T₂ ≈ 858 µs, the authors argue that improvements such as increasing optical depth, impedance matching, and embedding the crystal in an optical cavity could raise η₀ above the required 65 % and extend T₂ to the millisecond regime.

In summary, this work delivers the first demonstration of an Er³⁺‑based quantum memory that simultaneously offers millisecond‑scale on‑demand storage and combined temporal‑spectral multimode capacity at telecom wavelengths. These results bridge a critical gap toward scalable, long‑distance quantum networks and provide a clear pathway for further enhancements that could render such memories practical components of future quantum repeaters.