Atom–photon Entanglement with a Single Trapped Cesium Atom
H. Hwang, J. Moon, F. Herzallah, E. Oh, A. Safari, M. Saffman · 2026
Read Atom–photon Entanglement with a Single Trapped Cesium Atom by H. Hwang, J. Moon, F. Herzallah, E. Oh, A. Safari, M. Saffman (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-28968.txt.
Introduction
H. Hwang and colleagues demonstrate entanglement between a single trapped cesium atom and a photon. Resonant excitation plus spontaneous decay links the atom's Zeeman state to the photon's polarization. The photon is gathered by a high-aperture lens and fed into a single-mode fiber for joint measurement. Raw fidelity reaches about 94 percent, rising to about 96 percent after correcting for characterized atom measurement errors, and the setup points toward dual-species rubidium-cesium quantum networking.
Catalog record
| CREATOR | H. Hwang, J. Moon, F. Herzallah, E. Oh, A. Safari, M. Saffman |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43939 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.28968
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Atom–photon Entanglement with a Single Trapped Cesium Atom H. Hwang, J. Moon, F. Herzallah, E. Oh, A. Safari, M. Saffman Abstract We demonstrate atom–photon entanglement using a single cesium atom trapped in an optical tweezer. Entanglement is generated by resonant excitation and subsequent spontaneous decay, which entangles the atomic Zeeman state with photon polarization. The photon is collected with a high numerical aperture objective (NA = 0.55) and coupled into a single-mode fiber, enabling atom photon measurements and measurement of the Bell-state fidelity. We obtain raw entanglement fidelity of ℱ = 0.942 ( 16 ) {\mathcal{F}}=0.942(16) and inferred fidelity of ℱ inf = 0.962 ( 26 ) {\mathcal{F}}_{\rm inf}=0.962(26) after correcting independently characterized atom measurement errors. Compared with related free-space experiments using 87 Rb, the multilevel structure of the relevant excited state in 133 Cs requires the use of a single short excitation pulse in each entanglement attempt in order to suppress unwanted re-excitation. These results establish a free-space Cs atom–photon interface and provide a step toward dual-species Rb–Cs quantum networking. I Introduction Quantum networks aim to interconnect spatially separated quantum processors and sensors using photonic links, thereby providing a route towards modular quantum information processing Kimble 2008 ; Wehner et al. 2018 ; Reiserer and Rempe 2015 ; Covey et al. 2023 . In this architecture, each node combines long-lived stationary qubits for storage and processing with an efficient interface to flying qubits that distribute entanglement across the network Duan et al. 2001 ; Sangouard et al. 2011 . A fundamental resource in a quantum network is atom-photon entanglement which enables remote entanglement of matter qubits, for example, through interference of photons and Bell-state measurements Hong et al. 1987 ; Simon and Irvine 2003 ; Duan and Kimble 2003 ; Browne et al. 2003 ; Feng et al. 2003 . A central performance metric is therefore the rate and fidelity with which qubit–photon entanglement can be generated, collected, and delivered into a well-defined optical mode suitable for interference and Bell-state measurements Volz et al. 2006 ; Wilk et al. 2007 ; van Leent et al. 2020 ; Zhang et al. 2022 ; Hartung et al. 2024 ; Li et al. 2025 ; Safari et al. 2026 . Remote entanglement has been realized on several platforms, including trapped ions Stephenson et al. 2020 ; Krutyanskiy et al. 2023 ; O’Reilly et al. 2024 , color centers Pompili et al. 2021 , quantum dots Stockill et al. 2017 , and neutral atoms Welte et al. 2021 ; van Leent et al. 2022 . The generation rate and fidelity of remote entanglement links are largely determined by the light–matter interface at each node, by the coherence properties of stationary qubits, and by the stability and loss of the optical links. Among these platforms, neutral-atom tweezer arrays provide a scalable and reconfigurable approach with programmable geometries enabling flexible connectivity for large quantum registers Barredo et al. 2016 ; Endres et al. 2016 ; Browaeys and Lahaye 2020 ; Muniz et al. 2025 ; Bluvstein et al. 2026 . Beyond static geometries, dynamic reconfiguration, atom transport, and reservoir-based refilling can enable sustained operation of large ordered arrays by replenishing lost atoms and maintaining a usable register for repeated communication attempts Pause et al. 2023 ; Norcia et al. 2024 ; Gyger et al. 2024 ; Chiu et al. 2025 . Heterogeneous dual-species neutral-atom systems offer additional capabilities for species-selective trapping and control and separate roles such as long-lived memory qubits and optically active communication qubits Young et al. 2022 . This dual-species architecture enables mid-circuit readout with negligible optical crosstalk Singh et al. 2023 ; Miles et al. 2026 which is valuable for continuous operation of a network architecture and will enable entanglement distillation protocols to be incorporated into network nodes. Single neutral atom – photon entanglement has so far been demonstrated with 87 Rb in several free-space and cavity-based implementations Volz et al. 2006 ; Wilk et al. 2007 ; van Leent et al. 2020 ; Zhang et al. 2022 ; Hartung et al. 2024 ; Safari et al. 2026 and with 171 Yb atoms Li et al. 2025 in free space. While entanglement has been demonstrated between 133 Cs ensembles containing many atoms and photons Chou et al. 2005 , we demonstrate here for the first time a 133 Cs atom–photon entanglement interface using a single atom in an optical tweezer. The atomic qubit is encoded in long lived hyperfine states and the optical qubit is encoded in the polarization basis. Entanglement is demonstrated using the approach introduced in Volz et al. 2006 for 87 Rb, but applied to the level structure of 133 Cs. In addition we demonstrate mapping of the atomic qubit to internal states that extend the coherence time by two orders of magnitude compared to the magnetically sensitive states that are natively prepared in the entanglement generation process. We quantify the Bell-state fidelity and analyze the dominant error mechanisms, thereby establishing 133 Cs as a viable atomic element for future Rb–Cs dual-species quantum networks and modular quantum processors. II Single photon and entangled state generation Figure 1: Optical setup and experimental sequence. (a) Experimental setup used for 133 Cs atom trapping, single-photon characterization, and atom–photon entanglement verification. For the g^{(2)} measurement, a 50:50 non-polarizing beamsplitter is used, yielding g^{(2)}(0)=0.096(66) . For atom–photon entanglement, the beamsplitter is replaced by a polarizing beamsplitter (PBS), enabling polarization-basis analysis with single-photon counting modules (SPCMs). (b) Cesium atom–photon decay channels and histogram of photon arrival times. (c) Experimental sequence for entanglement generation and verification. In stage (i), a single atom is loaded using a MOT and polarization-gradient cooling. After confirming single-atom loading, in stage (ii) the atom is optically pumped to \ket{f=3,m_{f}=0} and excited to \ket{f^{\prime}=2,m_{f}=0} with a 12-ns pulse. Emitted photons with \sigma_{\pm} polarization are collected, coupled into a single-mode fiber, and detected on an SPCM with total collection and detection efficiency \eta=0.6\% . Conditioned on a photon detection event, the atomic state is analyzed by applying basis rotations and measuring X -, Y - and Z -basis parity oscillations. If no photon is detected, we repeat the cooling–optical pumping–excitation sequence. After every three attempts of excitation, we check whether the atom is still trapped. If the atom remains, we continue until a photon is detected, otherwise a new atom is loaded. Our experiment uses a high-NA objective lens both to trap a single atom and to collect photons emitted from it in free space. As shown in Fig. 1 (a), 1064 nm trapping light is focused by an objective lens (Jenoptik) with numerical aperture \mathrm{NA}=0.55 to form a single optical tweezer. Fluorescence from the trapped atom is collected by two identical objectives placed on opposite sides of the vacuum cell, one of which is also used to form the optical tweezer. Photons collected by one objective are coupled into a single mode optical fiber leading to single-photon counting modules (SPCM-AQRH-15-FC) for single-photon and atom–photon correlation measurements, whereas photons collected by the second objective are imaged onto an EMCCD camera to identify the spatial position of the trapped atom. The trap beam waist ( 1/e^{2} intensity radius) is 1.10(6) \mu m, and the maximum photon collection efficiency of the optical