Session K10: Focus Session: Quantum Memory

Large-bandwidth photonic storage and manipulation with cold and ultracold atoms

In the quest for optical quantum memories that are compatible with common single-photon sources, and which can be integrated into land- and satellite-based optical quantum communications networks, atoms play an important role for providing large-bandwidth and long storage time possibilities, using wavelengths that can be integrated into anticipated standard systems. In our experiments, we use both the conventional "fast" and the Autler-Townes protocols to realize large-bandwidth storage and manipulations of photonic signals, leading to fast and relatively low-noise signals, all of which can be implemented using minimal experimental resources. We demonstrated these protocols using laser-cooled and Bose-condensed atoms as the storage medium, and explored how the performance of each approach could be optimised [in terms of noise, efficiency, and resource requirements (laser power/optical depth)] under realistic parameters. Looking forward, we consider the potential for these protocols to contribute to tasks including wavelength conversion, multimode storage, and quantum information processing.   

Broadband quantum memories for photonic quantum states

Quantum memories that enable storage and retrieval of photonic quantum states are key components of quantum networks and local quantum information processing schemes. As with any processing task, high bandwidth is favored for scalability, but quantum memories that store GHz-THz bandwidth photonic quantum states have suffered from low efficiency. Experimentally and theoretically determining the optimal system and parameters for broadband quantum storage and retrieval is an ongoing area of research.

One of the most common implementations of photonic quantum memories in ensemble systems such as atomic vapors involves a Lambda-type energy level scheme. The photonic quantum state to be stored is mapped to a long-lived excitation via a two-photon Raman transition mediated by a strong control field. In this context we present our experimentally-motivated theoretical work on optimization of quantum memory efficiency, the sensitivity of the memory efficiency to fluctuations of experimental parameters, and our experimental implementation of a THz-bandwidth quantum memory in hot atomic barium vapor.   

Optimizing photon-mediated operations using external control fields

A variety of systems have been successfully implemented as qubits for quantum information processing. Many of these systems offer specific benefits and can be used advantageously for particular functions in a large scale hybrid quantum platform. In this context, optimal light-matter interfaces between different quantum systems is essential. In this talk, we will examine how spectral properties of quantum emitters can be controlled with external field protocols. In particular we will discuss two-photon interference between different quantum emitters that can be hindered by spectral differences between the systems. This problem is exacerbated in solid state systems where fluctuations in the surrounding environment leads to spectral diffusion. We will show that external field protocols can mitigate spectral diffusion for emitters in dynamic environments. We will also show that these protocols can restore photon indistinguishability between otherwise spectrally different quantum emitters and thus improve the efficiency of essential two-photon interference operations.

      

Towards practical quantum repeaters using room-temperature vapor technology

Quantum entanglement sources and memories are critical for quantum networking architecture. To build memory-assisted quantum repeaters, the entanglement source needs to generate photons that are ideal for optical fiber transmission and are simultaneously suitable for storage and buffering in a quantum memory. It is ideal to address these challenges without the need of additional quantum devices such as frequency converters. Most of the available sources, such as spontaneous parametric down conversion sources (SPDC), generate photons that are too spectrally broad for interfacing with atomic quantum devices. The use of optical parametric amplifiers allows for addressing this issue with the cost of significantly increasing the complexity of the devices.   

In this talk we will discuss our progress towards building bichromatic entanglement sources using the process of spontaneous four wave mixing in a warm rubidium vapor. We customize this process to generate entangled photons at 1324nm, suitable for fiber transmission, and 795nm, ideal as an input for room temperature Rb-based quantum memories. Noticeably, the photons' linewidths are typically on the order of a few MHz, and are ultimately limited by the natural atomic processes, while maintaining high spectral brightness. In addition, we will present the results of our elementary quantum repeater node by interfacing these photons with our high fidelity quantum memories.     

Wavelength conversion of ns-long pulses in cold atomic ensemble with extreme optical depth

Interfacing atoms with solid-state emitters such as quantum dots, which emit short (~ns-long) photon pulses, necessitates broadband (~GHz) operation of the atomic medium. Here, we theoretically investigate single photon wavelength conversion using atomic ensembles. In particular, we consider the conversion of photons (~895 nm) emitted by semiconductor quantum dots (InAsP/InAs) to wavelengths suitable for satellite-based QKD (794 nm) or fibre-based (1469 nm) telecommunication using Cs atoms. We show that efficient broad bandwidth photon conversion can be realized with atomic clouds of high optical depths (10³ - 10⁴) which can be created by confining the atoms inside a hollow-core optical fibre. We compare the efficiency of conversion to other platforms and discuss the role of the applied control fields, as well as of the geometry and temperature of the atomic ensemble inside the hollow-core fibre in achieving high conversion efficiencies.   

Reversible tuning of quantum dot emission to caesium D1 line

We describe a new method for reversible and bi-directional in-situ precision tuning of the wavelength of single photons emitted by a quantum dot embedded in a semiconductor nanowire, using the deposition of inert gas. This method, which can be reversed by increasing the temperature or using focused laser to evaporate the deposited gas, has allowed us to match the D₁ transition (∼895 nm) of cesium atom. We report the emitted photon characteristics including second-order correlation function g⁽²⁾(τ) and linewidth, as well as the lifetime of the quantum dot before and after the wavelength tuning. We also discuss our initial experiments studying interactions between the single photons and atomic ensembles.   

Device-Independent Quantum Key Distribution Between Two Ion Trap Nodes

Quantum theory promises that measurements on two entangled systems can yield correlated outcomes that are fundamentally unpredictable to any third party. Following the pioneering work of Ekert [1], we present the experimental realisation of a quantum key distribution (QKD) protocol for cryptography with device-independent security: to prove secrecy, we treat the systems as “black boxes”, relying only on measurement statistics observed during the key generation process. This mitigates many possible attacks on QKD, but requires a great number of observations of a large, detection-loophole-free Bell inequality violation.

We achieve this using two ⁸⁸Sr⁺ ion trap nodes connected by an optical fibre link. A heralded entanglement generation scheme yields about one hundred Bell pairs per second with a fidelity of 96.0(1)%, a new record for optical entanglement of distant matter qubits. We combine this experimental platform with theoretical advances to generate, for the first time, a shared key with device-independent security. Our result [2] demonstrates that provably secure cryptography is possible with real-world devices, and paves the way for further quantum information applications based on the device-independence principle.

[1] PRL 67, 661 (1991)

[2] arXiv:2109.14600