Detection of photon-level signals embedded in sunlight with an atomic photodetector

TL;DR

The paper tackles the problem of detecting photon-level signals in daylight by deploying a quantum jump photodetector based on a single trapped rubidium atom. It develops a rate-equation framework to describe sunlight-atom interactions and experimentally demonstrates detection of narrow-band photons embedded in strong solar background, with quantitative agreement between data and theory. A channel-capacity analysis shows meaningful information rates under daylight conditions, suggesting practical relevance for daylight LIDAR, remote magnetometry, and free-space communications. The approach provides a pathway to robust daytime quantum and classical optical sensing using inherently narrow-band atomic filters, with potential applicability to other atomic systems and wavelengths.

Abstract

The detection of few-photon signals in a broadband background is an extreme challenge for photon counting, requiring filtering that accepts a narrow range of optical frequencies while strongly rejecting all others. Recent work [Zarraoa et. al, Phys. Rev. Res. 6, 033338 (2024)] demonstrated that trapped single atoms can act as low dark-count narrow-band photodetectors. Here we show that this ``quantum jump photodetector'' (QJPD) approach can also detect photon-level signals embedded in strong sunlight. Using a single rubidium atom as a QJPD, we count arrivals of individual narrow-band laser photons embedded in sunlight powers of order $10^{10}$ photons/s. We derive a rate-equation model for the atom's internal-state dynamics in sunlight, and find quantitative agreement with experiment. Using this model, we calculate the channel capacity over a noisy communication channel when sending weak coherent states and detecting them in the presence of sunlight, achieving a representative rate of 0.5 bits per symbol when sending 150 probe photons per 10 ms time-bin, embedded in 1 nW of sunlight (of order $10^{10}$ photons/s in the visible and near-infrared bands). The demonstration may benefit background-limited applications such as daytime light detection and ranging (LIDAR), remote magnetometry, and free-space classical and quantum optical communications.

Figures (4)

• Figure 1: (a) Visual schematic of potential applied scenario: single-photon signals from a satellite are detected using an atomic detector (a single $^{87}$Rb atom) on top of a strong broadband background, e.g., sunlight. (b) Emission lines of $^{87}$Rb with ground-state-connected transitions at 420 nm, 780 nm and 795 nm (top) and sunlight spectrum after passing Earth's atmosphere (bottom). Red solid line marks the probe frequency at 780nm. (c) Energy levels of $^{87}$Rb with the relevant probe- and sun-driven transition rates.
• Figure 2: Experimental setup and sunlight spectrum. (a) Sunlight and probe are combined using two fiber beam-splitters (BSs) before being sent to the atom. (b) Sunlight is collected into a single-mode fiber using a collimator mounted on the tracking base of a telescope situated on the roof of the building and sent to the lab via long fibers. (c) Comparison between the spectral radiance of sunlight past the Earth's atmosphere (yellow shaded curve) and the one measured in the lab before the chamber, shown in red. Two black vertical lines at wavelengths 780nm and 795nm are included for visual reference.
• Figure 3: Sunlight combined with probe saturation curves. Data points show the probability of the atom being in $\ket{2}$ after the exposure time of 10ms and its uncertainty, for sunlight combined with probe photons at 0ph / 10 m s (black), 49 +- 6ph / 10 m s (blue) and 295+-16ph / 10 m s (red). Solid grey line show data fitted with Eq. (\ref{['eq:fit_sun_only']}) with parameters $N_2^\text{sat,exp} = 0.66 +- 0.03$ , $b_\text{exp} = 9 +- 2n W^{-1} s^{-1}$. Dashed lines plot the theoretical expectation of Eq. (\ref{['eq:population_dynamics_solution']}) without any free parameters, where the rates were calculated using Eq. (\ref{['eq:R_probe_etaQJ']}) with $N_2^\text{sat,exp}$ and $b_\text{exp}$ from the fitted Sun data, $\eta_\text{QJ}=8.5+-0.5e-3$ and corresponding probe photon numbers for low probe (blue) and high probe (red).
• Figure 4: Calculated channel capacity for a binary communication channel in presence of background photons for the atomic detector.