Summary form only given. Long-term storage of quantum information is a long standing goal since decades. It is particularly motivated by the fact that quantum network implementations strongly rely on quantum memories for light [1]. Long storage time and high efficiency are the main two figures of merit for a qubit memory. However, although quantum memories have been implemented in a large variety of physical systems, it turns out to be challenging to simultaneously fulfill those two main requirements.In our setup, a single 87Rb atom is trapped in a two-dimensional optical lattice inside a high-finesse cavity. Based on seminal work [2], the photonic polarization qubit a|L) + ß|R) described in the left/right polarization basis is mapped onto the atomic state |F = 2,= -1) + ß|F = 2, mf = +1) via a stimulated Raman adiabatic passage (STIRAP) as depicted in step I) of Fig. 1: The readout efficiency of this quantum memory protocol is basically constant for our typical coherence time. This is in contrast to systems based on atomic ensemble where decoherence affects the collective excitation reemission. However, fluctuations of the magnetic field deteriorate the coherence of the superposition. Indeed, the two levels we use are split by the Zeeman effect and thus the free evolution of the system will accumulate a random phase increasing with time. Typically, the coherence information of the qubit is lost after 200μs.
Summary form only given. Long-term storage of quantum information is a long standing goal since decades. It is particularly motivated by the fact that quantum network implementations strongly rely on quantum memories for light [1]. Long storage time and high efficiency are the main two figures of merit for a qubit memory. However, although quantum memories have been implemented in a large variety of physical systems, it turns out to be challenging to simultaneously fulfill those two main requirements.In our setup, a single 87Rb atom is trapped in a two-dimensional optical lattice inside a high-finesse cavity. Based on seminal work [2], the photonic polarization qubit a|L) + ß|R) described in the left/right polarization basis is mapped onto the atomic state |F = 2,= -1) + ß|F = 2, mf = +1) via a stimulated Raman adiabatic passage (STIRAP) as depicted in step I) of Fig. 1: The readout efficiency of this quantum memory protocol is basically constant for our typical coherence time. This is in contrast to systems based on atomic ensemble where decoherence affects the collective excitation reemission. However, fluctuations of the magnetic field deteriorate the coherence of the superposition. Indeed, the two levels we use are split by the Zeeman effect and thus the free evolution of the system will accumulate a random phase increasing with time. Typically, the coherence information of the qubit is lost after 200μs.
The long-lived, efficient storage and retrieval of a qubit encoded on a photon is an important ingredient for future quantum networks. Although systems with intrinsically long coherence times have been demonstrated, the combination with an efficient light-matter interface remains an outstanding challenge. In fact, the coherence times of memories for photonic qubits are currently limited to a few milliseconds. Here we report on a qubit memory based on a single atom coupled to a high-finesse optical resonator. By mapping and remapping the qubit between a basis used for light-matter interfacing and a basis which is less susceptible to decoherence, a coherence time exceeding 100 ms has been measured with a time-independant storage-and-retrieval efficiency of 22%. This demonstrates the first photonic qubit memory with a coherence time that exceeds the lower bound needed for teleporting qubits in a global quantum internet.
Optical quantum memory plays an important role in scaling-up linear optical quantum computations and longdistance quantum communication. For effectively realizing such tasks, a long-lived and highly-efficient quantum memory is needed. The dynamic electromagnetically-induced-transparency (EIT) process can be used for completing an absorptive storage scheme in an atomic ensemble. In such a process, the quantum states of coming single photons can be coherently transformed into spin waves associated with coherences between atomic ground levels via switching off controlling light beam. For storing a single-mode optical signal, a pair of ground levels is involved. While for storing an optical polarization qubit, i.e., two orthogonal polarization modes, the coherence between two pairs of ground levels will be involved. Also, to obtain a high efficiency in the EIT optical storage, the optical-depth of the cold ensemble should be high. For prolonging the coherent time of the spin waves stored in atomic ensemble, decoherence between spin waves due to atomic motion and non-uniform Zeeman shift of ground levels should be effectively suppressed. Recently, a long-lived and highly-efficient optical quantum memory for single-mode storage in a high-optical-depth cold atom ensemble has been experimentally demonstrated via the gradient echo memory scheme (2016 Optical 3 100). While, for the optical polarization qubit storage, a long lifetime (in ms) and high-fidelity EIT storage experiment has been achieved by our group, but the storage efficiency in the experiment is very low (8%) due to lower optical depth of the cold ensemble (2013 Phys. Rev. Lett. 111 240503). The storage efficiency in long-lived storage of two orthogonal polarization modes still needs further improving. Here in this paper we demonstrate an experiment of long-lived and highly-efficient storage of two optical orthogonal polarization modes in a high optical-depth cold atomic ensemble via dynamic EIT process. For achieving a long lifetime in the storage experiment, we follow the two steps, which are used in our previous work (2013 Phys. Rev. Lett. 111 240503). 1) We make the signal and writing-reading light beams collinearly pass through the cold-atom cloud along the z direction to suppress the decoherence between the spin waves due to atomic motion. 2) We apply a moderate magnetic field (13.5 G) to the cold-atom ensemble to lift Zeeman degeneracy. So, the magnetic-field-sensitive transitions are removed from EIT system and the two optical orthogonal polarization modes are stored as two magnetic-field-insensitive spin waves. In contrast to our previous experiment, we finish the storage in the high optical-depth cold atomic ensemble. To obtain such a high optical-depth cold atomic ensemble, we expand the diameters of the trapping laser beams and use a pair of rectangular magnetic coils in a magnetic optical trap (MOT) to prepare a cigar-shaped cold atomic ensemble. The MOT magnetic field is further compressed, and then the optical-depth of the cold atomic ensemble increases up to ~11 in the present experiment, which allows us to achieve a storage efficiency of 30%, which exceeds the previous value (8%). At an MOT repetition rate of 10 Hz, the measured zero-delay storage efficiencies for the two orthogonal polarization modes are symmetric, which go up to ~30%. The 1/e-folding lifetimes of the two orthogonal polarization modes rise up to 3 ms. We also measure the dependence of the zero-delay retrieval efficiency on the MOT repetition rate F and find that the storage efficiency is still more than 20% when the repetition rate F is 50 Hz. The present results will allow one to achieve a long lifetime and highly-efficient quantum memory for photonic polarization qubit and then find applications in scaling-up linear-optical quantum computations and long-distance quantum communication.
The realization of a large-scale quantum network of interconnected quantum nodes is one of the most promising directions in quantum technology. Such a network distributes entanglement between remote matter qubits via single photons in optical fibers, enabling quantum repeaters and long-distance secure communication and supporting emerging applications such as distributed quantum computing and network-enhanced metrology. To realize this vision, proposed architectures require quantum nodes that provide an efficient quantum light-matter interface for entanglement generation, serve as a quantum memory allowing long-lived storage of quantum states, and have the possibility to connect to telecommunication fiber-based photonic channels. This thesis presents the development and experimental realization of a quantum network node based on a single Rb-87 atom. A single atom is confined in a tightly focused optical dipole trap, and entanglement is generated between the polarization of an emitted photon and the Zeeman state of the atom. To enable long-lived quantum storage, we coherently map the atomic qubit to a basis less sensitive to magnetic fields using Zeeman-state-selective Raman transfer, reducing magnetic field sensitivity by a factor of 500. Such hyperfine qubit control is implemented via a two-photon Raman process driven by a laser field that is amplitude modulated at the qubit splitting using a fiber-based Mach-Zehnder interferometer. With this approach, we achieve a 1/e storage lifetime of 10 ms, extending coherence by a factor of 30 compared with the initial qubit encoding. Residual dephasing arises primarily from position-dependent effective magnetic fields caused by the tightly focused optical dipole trap beam together with atomic motion, and the differential AC-Stark shifts come from the trap detuning. While the first can be fully canceled via a standing-wave dipole trap, which needs to change the trap setup. We mitigate the second effects with a spin-echo pulse, extending coherence by a factor of two. Nonetheless, the significantly longer coherence time already allows high-fidelity entanglement distribution over fiber links with a length of a few hundred kilometers. To reduce fiber attenuation for long-distance distribution, we convert the emitted 780 nm single photons to the low-loss telecom S band using difference-frequency generation in a periodically poled lithium niobate waveguide, pumped at 1607 nm to produce 1517 nm photons. By combining long-lived quantum memory with a telecom interface, we demonstrated long-distance atom-photon entanglement distribution through 50 km and 101 km of spooled fiber with fidelities of 85% and 71%, respectively, establishing a key component for scalable quantum communication. These results establish neutral atom nodes as promising building blocks for metropolitan-scale quantum networking and outline a scalable path toward repeater protocols based on multiplexing memories and Rydberg-mediated interactions.
Interaction of quantum radiation with multi-particle ensembles has sparked off intense research efforts during the past decade. Emblematic of this field is the quantum memory scheme, where a quantum state of light is mapped onto an ensemble of atoms and then recovered in its original shape. While opening new access to the basics of light–atom interaction, quantum memory also appears as a key element for information processing applications, such as linear optics quantum computation and long-distance quantum communication via quantum repeaters. Not surprisingly, it is far from trivial to practically recover a stored quantum state of light and, although impressive progress has already been accomplished, researchers are still struggling to reach this ambitious objective. This special issue provides an account of the state-of-the-art in a fast-moving research area that makes physicists, engineers and chemists work together at the forefront of their discipline, involving quantum fields and atoms in different media, magnetic resonance techniques and material science. Various strategies have been considered to store and retrieve quantum light. The explored designs belong to three main—while still overlapping—classes. In architectures derived from photon echo, information is mapped over the spectral components of inhomogeneously broadened absorption bands, such as those encountered in rare earth ion doped crystals and atomic gases in external gradient magnetic field. Protocols based on electromagnetic induced transparency also rely on resonant excitation and are ideally suited to the homogeneous absorption lines offered by laser cooled atomic clouds or ion Coulomb crystals. Finally off-resonance approaches are illustrated by Faraday and Raman processes. Coupling with an optical cavity may enhance the storage process, even for negligibly small atom number. Multiple scattering is also proposed as a way to enlarge the quantum interaction distance of light with matter. The quest for higher efficiency, better fidelity, broader bandwidth, multimode capacity and longer storage lifetime is pursued in all those approaches, as shown in this special issue. The improvement of quantum memory operation specifically requires in-depth study and control of numerous physical processes leading to atomic decoherence. The present issue reflects the development of rare earth ion doped matrices offering long lifetime superposition states, either as bulk crystals or as optical waveguides. The need for quantum sources and high efficiency detectors at the single photon level is also illustrated. Several papers address the networking of quantum memories either in long-haul cryptography or in the prospect of quantum processing. In this context, much attention has been paid recently to interfacing quantum light with superconducting qubits and with nitrogen-vacancy centers in diamond. Finally, the quantum interfacing of light with matter raises questions on entanglement. The last two papers are devoted to the generation of entanglement by dissipative processes. It is shown that long lifetime entanglement may be built in this way. We hope this special issue will help readers to become familiar with the exciting field of ensemble-based quantum memories and will stimulate them to bring deeper insights and new ideas to this area.
Quantum storage of flying optical qubits, namely coherently mapping photonic states into and out of an optically controlled memory on demand, constitutes an essential component in the optical quantum information processing science, with applications to long-distance optical communication [1]. In this context, practical protocols require sufficiently large storage efficiency to achieve a performance better than direct transmission. Theoretically, the storage and retrieval efficiency can only be improved with the increase of the optical depth (OD). Recent progresses have shown that high OD media enable to reach high-efficiency electromagnetically induced transparency (EIT)-based optical storage, albeit without the demonstration of qubit storage [2, 3]. Here we experimentally implement an optical quantum memory with single-photon-level probes, showing a EIT-based storage efficiency around 70%. Furthermore, thanks to the spatial multiplexing of the dense atomic ensemble, highly efficient storage of polarization qubit is performed.
Quantum memory for flying optical qubits is a key enabler for a wide range of applications in quantum information. A critical figure of merit is the overall storage and retrieval efficiency. So far, despite the recent achievements of efficient memories for light pulses, the storage of qubits has suffered from limited efficiency. Here we report on a quantum memory for polarization qubits that combines an average conditional fidelity above 99% and efficiency around 68%, thereby demonstrating a reversible qubit mapping where more information is retrieved than lost. The qubits are encoded with weak coherent states at the single-photon level and the memory is based on electromagnetically-induced transparency in an elongated laser-cooled ensemble of cesium atoms, spatially multiplexed for dual-rail storage. This implementation preserves high optical depth on both rails, without compromise between multiplexing and storage efficiency. Our work provides an efficient node for future tests of quantum network functionalities and advanced photonic circuits.