Session Y10: Quantum Networks

Entanglement of trapped-ion qubits separated by 230 meters

Entanglement-based quantum networks hold out the promise of new capabilities for secure communication, distributed quantum computing, and interconnected quantum sensors. However, only a handful of elementary quantum networks have been realized to date. I will present results from our prototype network, in which two calcium ions are entangled with one another over a distance of 230 m, via a 520 m optical fiber channel linking two buildings [1]. The ion-ion entanglement is based on ion-photon entanglement mediated by coherent Raman processes in optical cavities. Fidelities of up to (88.0+2.2−4.7)% are achieved with respect to a maximally entangled Bell state, with a success probability of 4×10⁻⁵. We will examine the strengths of trapped ions in a quantum-network setting and the role that cavities can play as quantum interfaces at network nodes. In ongoing work, we aim to extend this result to metropolitan-scale distances, enabled by quantum-frequency conversion to telecom wavelengths [2].

[1] V. Krutyanskiy, M. Galli et al., Phys. Rev. Lett. 130, 050803 (2023)

[2] V. Krutyanskiy et al., Phys. Rev. Lett. 130, 213601 (2023)   

Entangling quantum memories at channel capacity

Generating entanglement between quantum memories, mediated by optical-frequency or microwave channels, at high rates and fidelities is key for linking qubits over short to long ranges. In this talk, I will discuss a few architectural alternatives for how to generate heralded entanglement among two quantum memory registers, and their use in repeater and satellite-assisted quantum links. We will discuss some well-known protocols---of both the midpoint-swap and midpoint-source flavors---which encode up to one qubit per optical mode, hence entangling one pair of memory qubits per transmitted mode over the channel, with probability equaling the channel's transmissivity. The rate, measured in ideal Bell states (or, ebits) per mode, is thus proportional to this transmissivity, which is the optimal scaling, in terms of the quantum capacity, at high loss. However, the quantum channel capacity shoots up toward infinity for low loss of the channel connecting the memory registers, making the known schemes highly rate-suboptimal for shorter ranges, viz., intra-processor, data center and even local-area quantum network links. I will discuss how a cavity-assisted memory-photon interface can be used to entangle matter memories with Gottesman-Kitaev-Preskill (GKP) photonic qudits, which along with dual-homodyne entanglement swaps that retain and process the analog information, enables capacity-approaching entanglement rates at low loss. This scheme benefits from the loss resilience of GKP qudits, and their ability to encode multiple qubits in one bosonic mode. Further, the memory-photon interface supports the preparation of needed ancilla GKP qudits. I will end the talk with an overview of research being performed as part of the NSF-funded Center for Quantum Networks, a consortium of multiple universities and companies in the US, working on the full stack development of fault-tolerant quantum networking.   

Hybrid atom — rare-earth ion quantum interfaces and network nodes

Future global quantum networks will merge heterogeneous quantum systems to simultaneously perform multiple tasks; namely, reliably store, process, and transmit quantum information as well as distribute entanglement. The functionality of such networks relies crucially on coherent interfaces between disparate, distant nodes, and telecom photons which connect them through low-loss optical fibers. We propose a modular hybrid network architecture composed of a pair of nodes with matching telecom wavelengths to circumvent loss due to quantum frequency conversion processes. On one side, atom-based nodes, including a warm atomic ensemble and an atom-array – nanophotonic cavity system, serve as our photon source node and processor node respectively, compatible with each other and capable of generating atom-photon entanglement. On the other side, a rare-earth ion-doped crystal serves as our memory node, where inhomogeneous broadening allows broadband and multiplexed storage for time-bin photonic qubits, creating entanglement between the hybrid nodes.

In this talk, we will present our results identifying mode-matching conditions between a rubidium vapor, our source, and an erbium-doped crystal, our memory. We will show experimental progress towards a fully functioning module: the former outputs a heralded telecom single photon via phase-matched four-wave mixing, and the latter coherently stores the photon via the atomic frequency comb protocol, preserving the non-classical correlation between the photon and its heralding pair. We will further discuss the outlook on integrating both elements with nanophotonics for distributed quantum computing, and deterministic entanglement distribution across a metropolitan area.   

Entanglement of Nanophotonic Quantum Memory Nodes in a Metropolitan Telecom Network

Long-range quantum networks have the potential to enable provably secure communications and distributed quantum computing, but a key obstacle thus far has been the difficulty in extending lab-scale entanglement to deployed city-scale fiber networks. Here, we present a two-node quantum network based on the silicon-vacancy center (SiV) in diamond in nanophotonic cavities, with each node containing an electronic spin communication qubit and a nuclear spin memory qubit. Cavity-enhanced interactions enable heralded entangling gates between the spin qubits and a time-bin photonic qubit, allowing us to generate remote entanglement between two nodes located in separate cryostats, while utilizing the long-lived nuclear spins to achieve second-long entanglement storage with integrated error detection. By integrating bi-directional quantum frequency conversion to the low-loss telecom band at 1350 nm, we demonstrate compatible operation of our quantum network with existing commercial fibers and entangle two nuclear spin qubits through a 35 km long deployed fiber loop in the Boston metropolitan area. This represents a significant step towards the practical deployment of quantum networks and enables applications in blind quantum computing and nonlocal sensing.   

High-rate entanglement generation between ions in a multispecies ion trap using photonic interconnects

Long chains of trapped ions are a leading platform for quantum information processing, but as chains grow longer their control suffers from spectral crowding and excess motional heating. One method to increase the computational power available is the use of photonic interconnects to entangle ions in separate traps. Toward this goal, we have built an ion trap system with two high numerical aperture (NA=0.8) in-vacuum objectives, achieving the highest free-space collection efficiency of photons from an ion. We trap two ¹³⁸Ba⁺ ions and generate entanglement between them using photons, demonstrating the efficacy of its photonic interconnects. To increase the entanglement generation rate, we co-trap ¹⁷¹Yb⁺ to sympathetically cool the barium ions. We achieve the highest entanglement generation rate between two qubits entangled via photonic interconnects.    

Remote entanglement of trapped ions using time bin encoded photons

Trapped ions are one of the leading platforms for quantum computing and quantum communication. They naturally emit photons, thus providing an interface with spatially separated ion trap quantum nodes. Various degrees of freedom of photons can be used to generate entanglement, such as polarization, time bins, angular momentum, frequency etc. Polarization-based encoding has been used in previous works for high-rate entanglement generation between ion traps, but fidelity has always been limited by polarization errors. Photonic time bin qubits are a good candidate for networking because of their intrinsic insensitivity to polarization errors, especially when travelling through fiber. Here, we report our progress towards entangling two remote trapped ion qubits via time encoded photons. We analyze the fidelity of our resulting entangled state and compare it with the fidelity from our polarization qubit entanglement counterparts.   

Networking Ion Traps with Telecom-Compatible 88Sr+ Photons

Generating entanglement between distant trapped ions is necessary for realizing quantum networks and distributed computation on this platform. Current methods require sending ion-entangled photons down optical fibers between trapped ion nodes. It remains a technical challenge to do so with photons that exhibit low loss in fibers. Most ion transitions realizing the scheme emit at high-loss wavelengths, necessitating frequency conversion techniques to achieve appreciable networking distances, which can add noise and technical complexity. As an alternative, we use the P1/2 → D3/2 transition in 88Sr+. The transition emits at 1092 nm, a wavelength that has relatively low loss in standard telecom fibers. This makes the transition efficient for medium-distance quantum networking (~<10 km). We report on experimental results from our first network node coupling ion emissions into telecom fiber, characterizing loss, and measuring ion-photon entangled state fidelity.