Deterministic Single Photons via Conditional Quantum Evolution

A source of deterministic single photons is proposed and demonstrated by the application of a measurement-based feedback protocol to a heralded single-photon source consisting of an ensemble of cold rubidium atoms. Our source is stationary and produces a photoelectric detection record with sub-Poissonian statistics.

Figure 1

Schematic of experimental setup, with the inset showing the atomic level scheme (see text).

Figure 2

$g_{\mathrm{si}}$ (a) and $\alpha$ (b) vs $p_1$, for $\tau =80\mathrm{ns}$. The solid lines are based on Eqs. (3, 4), with a nearly negligible background [5]. The inset shows $g_{\mathrm{si}}$ vs storage time $\tau$. The full curve is a fit of the form $1+B\exp (-{\tau }^2/{\tau }_c^2)$ with $B=16$ and ${\tau }_c=31.5\mu \mathrm{s}$ as adjustable parameters.

Figure 3

$g_D^{(2)}(0)$ as a function of $N$ (a) and $p_1$ (c); ${\eta }_D$ as a function of $N$ (b) and $p_1$ (d). In (a) and (b) $p_1=0.003$ (about $6\times {10}^5$ photons per write pulse were used), whereas in (c) and (d) $N=150$. The full curves are based on Eq. (5) with the values of efficiencies and coherence times given in the text, with, however, ${\eta }_D$ multiplied by an empirical factor of $2/3$. We believe this reduced efficiency is due to imperfect switching of the read light in the feedback-based protocol (we note that there are no other adjustable parameters in the theory). Deviations from the theory in (c) and (d), could be explained by drifts in the residual magnetic field and read light leakage.