Cavity-induced backaction in Purcell-enhanced photon emission of a single ion in an ultraviolet fiber cavity

We study the behavior of a single laser-driven trapped ion inside a microscopic optical Fabry-Perot cavity. In particular, we demonstrate a fiber Fabry-Perot cavity operating on the principal $S_{1/2}\to P_{1/2}$ electric dipole transition of an ${\mathrm{Yb}}^{+}$ ion at 369 nm with an ion-cavity coupling strength of $g=2\pi \times 67\left(1\right)$ MHz. We employ the cavity to study the generation of single photons and observe cavity-induced backaction in the Purcell-enhanced emission of photons. Tuning of the amplitude and phase between the driving field and the cavity field built up from photons scattered into the cavity mode by the ion allows us to enhance or suppress the total rate of photon emission from the ion-cavity system.

Figure 1

(a) Schematic setup comprising an end-cap trap and the fiber cavity. (b) ${}^{174}\mathrm{Yb}^{+}$ ion level structure. (c) Loss increase of the cavity coating vs time under ultrahigh vacuum conditions measured from the decrease of the cavity incoupling efficiency. The increase fits the exponential model described in [26] with a time constant of ${\tau }_{\text{loss}}=123\pm 15$ days. The error bars correspond to a 2% statistical measurement error of the cavity incoupling, from which the losses are calculated considering a fiber-cavity mode matching of $(50.5\pm 3.4)\%$.

Figure 2

(a) Photon rate scattered transversely to the cavity mode vs atom-laser detuning. The solid line is a Lorentzian fit to the data from which the Purcell enhanced linewidth of the atomic transition can be retrieved (see main text). The larger scattering of the data at positive detuning comes from Doppler heating of the ion during the measurement cycle. The saturation parameter for this measurement is $s=2.8$, from which follows ${\mathrm{\Gamma }}^{\prime }=2\pi \times 24\left(5\right)$. (b) Second-order correlation function of the photons emitted from the cavity tuned into resonance with the ion (${\mathrm{\Delta }}_a={\mathrm{\Delta }}_c$) and a drive laser detuning of ${\mathrm{\Delta }}_a=-\mathrm{\Gamma }$ at $s=14$. The solid line is a fit to the data (see text), with ${\mathrm{\Gamma }}^{\prime }$ obtained from (a) and the Rabi frequency and average photon production rate as free parameter. Inset: Experimental setup. BS: beam-splitter, SPCM: single-photon counting module, SM: single-mode.

Figure 3

(a) Two-level system in a cavity driven by a laser transverse to the cavity. Photodetectors monitor both rates $P_c$ and $P_{4\pi -c}$. A magnetic field of about 2G is applied along the cavity axis such that the cavity only mediates $\sigma$ transitions and the drive laser polarization is adjusted perpendicular to it. (b) Cavity output. (c) Free-space emission. (d) Total photon emission. (e) Minimum of the free-space scattering ${\mathrm{\Delta }}_{c,-}$ for different detunings of the drive laser from the atomic resonance. The solid line is from Eq. (1) and the dashed line describes the linear dependance of the minimum for a drive laser detuning close to atomic resonance discussed in the main text.