arXiv:2610.23536 · IEEE International Conference on Quantum Computing and Engineering (QCE)
Temperature dependence of a warm ensemble based memory on quantum communication rates
Kenny Gregory, Khaled Mnaymneh, Connor Kupchak, Andrew MacRae
Abstract
The success of long-distance quantum communication will hinge on quantum repeater nodes equipped with robust and scalable quantum memories. Such memories must possess specific operating characteristics to make these nodes practical. An attractive option for these memory devices is an atomic warm vapor system operating under the conditions of electromagnetically induced transparency (EIT). These systems are easy to implement and require relatively low equipment overhead. A key metric in atomic-based quantum memory performance is optical depth, as it is closely tied to the strength of light-matter coupling. Here, we simulate and measure the optical depth-based performance of an EIT quantum memory using a warm atomic vapor cell. We then translate the simulated and measured optical depth and efficiency metrics into expected entanglement distribution rates. Our analysis evaluates the feasibility of these memories in optical quantum repeater nodes, which are essential for robust long-distance quantum communication.
AI Summary
The paper examines how temperature-dependent optical depth and storage efficiency in a warm-vapour EIT memory affect projected entanglement-distribution rates in a fibre-based quantum repeater chain. It combines characterization of a portable $^{87}\mathrm{Rb}$ memory operating at 795 nm with semiclassical Maxwell–Bloch simulations and an analytical repeater-rate model. Measurements over $55$–$75\,^{\circ}\mathrm{C}$ give optical depths $D\approx1.2$–$2.4$, increasing approximately exponentially with temperature, while storage efficiency shows a saturating response. Simulations include Doppler broadening, spontaneous emission and fixed ground-state dephasing. Example calculations for a $2\,\mu\mathrm{s}$ storage interval yield efficiencies of approximately $5\%$, $9\%$ and $16\%$ at $D=2,3,5$, respectively; higher-depth simulations give roughly $20\%$–$60\%$ efficiency over $D=7$–$60$. These efficiencies are inserted into a model with equal-length links, midpoint Bell-state measurements and memory-assisted asynchronous link generation. Rate contours for two- and four-link chains illustrate trade-offs among optical depth, distance and repeater number. Measured efficiencies produce common-rate contours spanning approximately 9–52 km; higher simulated efficiencies are used to project operation beyond 250 km, rather than demonstrate entanglement distribution experimentally. The rate calculations assume infinite memory lifetime, compatible repetition rates, indistinguishable photons, $90\%$ detector efficiency and unity frequency-conversion and additional coupling efficiencies. Memory noise, finite coherence time and their temperature dependence are deferred to future work. The paper also notes reabsorption and four-wave mixing as possible limitations at higher optical depth; it does not establish performance exceeding the repeaterless bound.
Central claim may not hold: The communication-rate predictions assume infinite memory lifetime and ideal conversion, coupling, and photon indistinguishability, so they support only conditional feasibility rather than demonstrated operation of the characterized memory in a repeater. The fibre-loss expressions omit the division by 10 required when attenuation is specified in dB/km, making the printed distance-dependent loss exponent ten times too large. The asynchronous rate formula omits the waiting-time overhead for all elementary links to become ready, which remains present even with infinite-lived memories. The stated ground-state dephasing rate suppresses stored-signal efficiency to at most approximately 5.3e-7 over the stated 2-microsecond storage interval, contradicting the reported simulated efficiencies of 5–16%.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Journal reference
- 2025 IEEE International Conference on Quantum Computing and Engineering (QCE), Albuquerque, NM, USA, 2025, pp. 996-1002
- DOI
- 10.1109/QCE65121.2025.00111
- Comments
- 7 pages, 5 figures