@article{2018-macha,
	Abstract = {<p>
	In this work I theoretically investigate and experimentally realize the storage of short light-pulses in a fiber-based atom-cavity system. Our miniaturized optical resonator &ndash; with seven times the natural atomic linewidth and a small mode volume &ndash; simultaneously ensures a high bandwidth and operation in the strong-coupling regime. In particular, it enables the storage of light pulses with on average one photon and a temporal extent of less than 10 ns, which is more than a factor of two shorter than the atomic excited state lifetime of rubidium. We obtain a storage efficiency of 8%, consistent with both cavity losses and the employed level scheme.<br />
	 In order to improve the coupling and number of measurements for which a single atom can be recycled, we use dipole-trap assisted, degenerate Raman sideband cooling and a further development of our carrier-free Raman sideband cooling scheme, which permits a three-dimensional ground state population of 70%. The new techniques increase the measurement repetition rate by two orders of magnitude to &sim; 2 kHz. Moreover, for the first time we achieve a Zeeman state preparation fidelity above 95% in our experiment.<br />
	 On this basis, I present the deterministic generation of single photons in the near-adiabatic limit. By shaping the control laser pulse, we do not only show that we can control the temporal waveform of retrieved photons, but also reach a faster extraction from the cavity-coupled atom than possible in free-space. The quantum nature of the retrieved light is verified by measuring a second-order correlation function, which yields the expected antibunching. Moreover, the generation of photons in the cavity mode with an efficiency exceeding 66% is used as a fast hyperfine-state detection method, since our traditional, non-destructive state detection via a probe laser is no longer applicable in a Raman configuration due to the absence of a cycling transition. In order to realize Raman coupling between the two hyperfine ground states, we develop a scheme for shifting the cavity resonance frequency between two hyperfine transitions. During the scan, we are furthermore able to determine the atom-cavity coupling strength via the vacuum Rabi splitting in each individual measurement &ndash; a useful tool for post-selection of acquired data sets.<br />
	 By employing a numerical simulation based on a full quantum-mechanical master equation, I find the strategy to store a coherent laser pulse with the maximum possible efficiency for a given system. Although the cavity input field is treated classically, our simulation model is able to calculate efficiencies for a pure single-photon Fock-state input. Moreover, numerical optimal control methods enable us to find control pulses with storage efficiencies slightly above those achieved for temporally-scaled adiabatic control pulses. For our specific system, we finally demonstrate the non-adiabatic storage of a short, coherent light pulse.<br />
	 The ability to interact with pulses of high bandwidths encourages quantum hybrid experiments with quantum dots as single-photon sources. In this context, the stabilization of their emission frequency to an atomic transition is required. In collaboration with the IFW Dresden, I present a technique to counteract long-term frequency drifts by applying rate-based feedback to a strain-tunable quantum dot, which results in frequency deviations smaller than 1.5% of its emission linewidth. By simultaneously stabilizing the emission frequency of two quantum dots in separate cryostats, we enhance their two-photon interference visibility in a Hong-Ou-Mandel measurement from 31% to 41%, which corresponds to the maximum reachable visibility for the given emitters. Frequency-stable, efficient photon sources together with atom-cavity based quantum memories may facilitate the realization of quantum networks.</p>},
	Author = {Macha, T.},
	Journal = {},
	Pages = {},
	Title = {{ Storage of Short Light Pulses in a Fiber–Based Atom–Cavity System}},
	Volume = {},
	Year = {2018}
}