Quantentechnologie

Optical cavities with coupled atoms are a promising platform as the nodes of future quantum networks, enabling interchange of information between single atoms and single photons. In particular, fiber-based high-bandwidth cavities offer convenient and efficient routing of quantum information, in an interesting regime that combines strong coupling with the atom and high-rate information exchange with the quantum channel. However, critical control of coupled atoms for the quantum node operation, is hindered by the strong Purcell effect, the miniaturized geometry and the high-bandwidth property of such cavities. In this thesis, I report on my contributions towards a high level of control of individual atoms coupled to a high-bandwidth cavity. To this end, three new experimental techniques were developed specially adapted to high-bandwidth fiber cavities, with the following specific goals: (i) intracavity ground-state cooling of single atoms; (ii) atom position detection by fluorescence imaging independent from the cavity transition; (iii) cavity loading of small atomic ensembles with increased density.

In the first part of this work, I present the experimental setup, consisting of a fiber Fabry-Pe ́rot cavity (FFPC) coupled to 87Rb atoms, and the necessary experimental apparatus to operate the system in a stable manner. I start by motivating the advantage of high-bandwidth cavities with a brief discussion on cavity-mediated light-matter interfacing, and the peculiar strong coupling regime. Then, I give an overview of the complete system with emphasis on the recent technical upgrades, such as an improved cavity stabilization, an upgraded Raman laser setup with a linewidth-reduced DBR laser, and a new cavity-compatible imaging system. Lastly, I introduce the basic experimental toolbox for atomic control that we employ to operate the atom-cavity module: (i) cavity-based atom detection; (ii) cooling with a magneto-optical trap (MOT) and trapping with a 3D lattice; (iii) state initialization by optical pumping; (iv) Raman hyperfine manipulation; (v) position detection by imaging. Most of my work was to extend such basic toolbox for an improved atomic control, with the techniques presented in the next chapters.

In Chapter 3, I report successful cooling of a single 87Rb atom to its one-dimensional motional ground state while coupled to the FFPC, by degenerate Raman sideband cooling (dRSC). We overcome the challenge of cooling in such high-bandwidth atom-cavity modules, by adapting the degenerate dRSC technique to our cavity and lattice geometry. Raman cooling transitions are driven by the trapping lattice and repumping by the intracavity probe field, without the need of additional lasers and activated by the magnetic bias field. The resource-efficient and simple implementation is a highlight.

In Chapter 4, I present a newly implemented method in our system for successful fluorescence imaging of small atomic ensembles coupled to a high-bandwidth FFPC, that overcomes the inhibiting Purcell effect and the restricted optical access. It is based on techniques from the field of quantum gas microscopes and relies on the detection of repumper fluorescence on the D1 line generated by three-dimensional (3D) continuous Raman sideband cooling (cRSC). Thus, it remains fully independent from the cavity on the D2 line, for simultaneous operation of the atom-cavity node and position detection of the atoms. It requires only a single free-space beam together with intra-cavity fields, ideal for platforms with limited optical access, e.g. miniaturized quantum optical devices. The repumper-induced differential light shifts and the heating by dipole-force fluctuations (DFFs) are also analyzed.

In Chapter 5, I introduce a novel and simple method to load the intracavity lattice: the drive-through loading. It only relies on the dynamic control of intensity and phase of one lattice arm that works as a conveyor belt between the MOT and the intracavity lattice. I discuss the working principle of the technique, demonstrate that its efficiency, and show its tuning capability of the cavity-coupled atom number. In the last chapter, I summarize the advances presented here that extend the toolbox for control and manipulation of atom-cavity systems, impacting in the development of quantum networks. The three new techniques presented here, with a future implementation of single-atom addressing, pave the way for creating atomic arrays with predefined number and positions in the cavity: a cavity-quantum register.

High-bandwidth, fiber-based optical cavities are a promising building block for future quantum networks. They are used to resonantly couple stationary qubits such as single or multiple atoms with photons routing quantum information into a fiber network at high rates. In high-bandwidth cavities, standard fluorescence imaging on the atom-cavity resonance line for controlling atom positions is impaired since the Purcell effect strongly suppresses all-directional fluorescence. Here, we restore imaging of ⁸⁷Rb atoms strongly coupled to such a fiber Fabry-Pérot cavity by detecting the repumper fluorescence which is generated by continuous and three-dimensional Raman sideband cooling. We have carried out a detailed spectroscopic investigation of the repumper-induced differential light shifts affecting the Raman resonance, dependent on intensity and detuning. Our analysis identifies a compromise regime between imaging signal-to-noise ratio and survival rate, where physical insight into the role of dipole-force fluctuations in the heating dynamics of trapped atoms is gained.

We report on vibrational ground-state cooling of a single neutral atom coupled to a high-bandwidth Fabry-Pérot cavity. The cooling process relies on degenerate Raman sideband transitions driven by dipole trap beams, which confine the atoms in three dimensions. We infer a one-dimensional motional ground-state population close to $90\%$ by means of Raman spectroscopy. Moreover, lifetime measurements of a cavity-coupled atom exceeding 40 s imply three-dimensional cooling of the atomic motion, which makes this resource-efficient technique particularly interesting for cavity experiments with limited optical access.

We demonstrate the storage of 5 ns light pulses in a single rubidium atom coupled to a fiber-based optical resonator. Our storage protocol addresses a regime beyond the conventional adiabatic limit and approaches the theoretical bandwidth limit. We extract the optimal control laser pulse properties from a numerical simulation of our system and measure storage efficiencies of (8.1±1.1)%, in close agreement with the maximum expected efficiency. Such well-controlled and high-bandwidth atom-photon interfaces are key components for future hybrid quantum networks.

We employ active feedback to stabilize the frequency of single photons emitted by two separate quantum dots to an atomic standard. The transmission of a rubidium-based Faraday filter serves as the error signal for frequency stabilization. We achieve a residual frequency deviation of <30 MHz, which is less than 1.5% of the quantum dot linewidth. Long-term stability is demonstrated by Hong-Ou-Mandel interference between photons from the two quantum dots. Their internal dephasing limits the expected visibility to V = 40%. We observe V_lock = (41±5)% for frequency-stabilized dots as opposed to V_free = (31±7)% for free-running emission. Our technique reaches the maximally expected visibility for the given system and therefore facilitates quantum networks with indistinguishable photons from distributed sources.