Quantum-preserving telecom conversion of atomic biphotons

Abstract

We experimentally demonstrate telecom frequency conversion of atomic biphotons using a diamond-type atomic ensemble. By spectrally engineering heralded photons and optimizing the atomic converter, efficient conversion is achieved while preserving the temporal waveform and nonclassical antibunching behavior. The converted photons retain strong quantum correlations and well-defined wavepackets, demonstrating preservation of dynamical quantum properties beyond photon-statistics-based benchmarks. The measured performance agrees with a microscopic model that captures the spectral acceptance and parameter dependence of the converter. These results establish a practical interface between atomic photon sources and telecom fiber networks, enabling quantum interference and distributed quantum communication.

Figures (4)

• Figure 1: Experimental architecture of the atomic biphoton source and the telecom frequency conversion platform. (a) and (b) show the energy-level configurations of the double-$\Lambda$ SFWM source and the diamond-type conversion process. The fields $\Omega_1$, $\Omega_2$, $\hat{a}_p$, $\hat{a}_t$, $\Omega_c$, $\Omega_d$, and $\hat{a}_s$ denote pump 1, pump 2, probe, trigger, coupling, driving, and the converted signal field. (c) and (d) illustrate the corresponding experimental implementations. F, filter; DM, dichroic mirror; EFS, etalon filter set; HWP, half-wave plate; L, lens; M, mirror; PBS, polarizing beam splitter; POL, polarizer; QWP, quarter-wave plate; SPCM, single-photon counting module; SMF, single-mode fiber; MMF, multi-mode fiber. (e) Timing sequence illustrating the synchronization between heralded photon preparation and the conversion process.
• Figure 2: Benchmarking of the heralded single-photon source. (a) Normalized cross-correlation $g^{(2)}_{t\text{-}p}(\tau)$ between the trigger and probe photons, demonstrating strong temporal quantum correlations. (b) Conditional autocorrelation $g^{(2)}_{p\text{-}p|t}(\tau)$ of the heralded probe photon, confirming sub-Poissonian statistics and high single-photon purity. Circles and squares represent experimental data and solid curves show theoretical predictions. The parameters are ${\rm OD} = 8$, $\Omega_1 = 0.9\Gamma$, $\Omega_2 = 4\Gamma$, and $\Delta_1 = -4\Gamma$. Error bars indicate one standard deviation.
• Figure 3: Verification of quantum preservation after frequency conversion. (a) Normalized cross-correlation $g^{(2)}_{t\text{-}s}(\tau)$ between the trigger and converted signal photons, showing preserved temporal correlations after conversion. (b) Conditional autocorrelation $g^{(2)}_{s\text{-}s|t}(\tau)$ of the heralded telecom photon, confirming preservation of single-photon purity. Circles and squares denote experimental data, solid curves show theoretical predictions, and dashed curves indicate the pre-conversion results in Fig. \ref{['fig:biphoton']}. The parameters are ${\rm OD} = 110$, $\Omega_c = 20\Gamma$, $\Omega_d = 12\Gamma$, $\Delta_p = -4\Gamma$, $\Delta_c = 8\Gamma$, and $\Delta_d = -5\Gamma$.
• Figure 4: High-efficiency telecom conversion enabled by spectral matching. Signal conversion efficiency and probe transmission as functions of the driving field strength. Circles denote probe transmission and squares represent signal conversion efficiency. Solid curves are theoretical predictions. The optimized source parameters are ${\rm OD} = 17$, $\Omega_1 = 1.8\Gamma$, $\Omega_2 = 2\Gamma$, and $\Delta_1 = -4\Gamma$. The conversion parameters are ${\rm OD} = 120$, $\Omega_c = 23\Gamma$, $\Delta_p = -4\Gamma$, $\Delta_c = 8\Gamma$, and $\Delta_d = -5\Gamma$. The reduced photon bandwidth ensures near-complete spectral overlap with the conversion window and enables high-efficiency conversion close to the steady-state limit. These results confirm that the efficiency reduction in the broadband regime mainly originates from spectral mismatch.