Quantum Networking at the Speed of Quantum Computation: Kilohertz Entanglement in a Heterogeneous Quantum System

arXiv:2610.10705 · quant-ph · Submitted 2026-10-07 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.

Kai: I'm Kai, and with me are Mira and Lev, guest researcher.

Mira: Today's paper: "Quantum Networking at the Speed of Quantum Computation".

Kai: The gist: This work introduces a hybrid architecture that entangles a trapped ion qubit with a solid-state quantum memory to achieve kilohertz rates,

Mira: First, who's behind it and why it matters.

Title and authors: Kai: We’re starting by looking at the title and the authors of this paper, "Quantum Networking at the Speed of Quantum Computation: Kilohertz Entanglement in a Heterogeneous Quantum System." It immediately sets a very high bar for what this kind of work is aiming to do.

Mira: The title itself points right to the big picture: achieving quantum networking at the speed of quantum computation.

Kai: And the authors are a big team, which tells you it’s coming from a place where they are tackling this problem across different areas of physics and engineering.

Lev: I'm interested in seeing if the scope aligns with what we expect from this kind of paper; is it just a proof of concept, or does it have real implications for scaling?

Kai: It seems like they are aiming for something more substantial than just a basic proof of concept; they’re building an architecture.

Mira: They are combining different physical systems, which suggests the focus isn't just on improving one component in isolation.

Lev: So, I expect to see how they handle the integration challenges between these two very different platforms in their methodology.

Kai: Exactly, and it seems like they are tackling that head-on by designing this specific hybrid architecture.

Mira: They're using a trapped ion qubit and a solid-state quantum memory to create this kind of link, which is the core idea here.

Lev: I wonder how much complexity that interface adds to the overall performance when you are trying to hit these high rates.

Kai: The paper is focused on showing that this combination can produce entanglement at a rate that’s relevant for future quantum computing architectures.

Mira: They’re not just looking for some slow, low-rate link; they are pushing toward kilohertz entanglement.

Lev: Kilohertz is a significant number in the quantum world when you consider the cycle times of these systems.

Kai: Yeah, and that's why they frame it as being comparable to the cycle time of planned trapped-ion processors.

The paper's summary: Kai: Now, let’s look at what the paper actually says in its summary about this work on "Quantum Networking at the Speed of Quantum Computation: Kilohertz Entanglement in a Heterogeneous Quantum System."

Mira: They are summarizing how they entangle a trapped 138Ba+ ion with a silicon-vacancy center in a diamond nanophotonic cavity to achieve this entanglement <ref:2610.10705#pg2>.

Lev: So, what is the main mechanism for that entanglement generation? Is it just one single step, or is it a complex sequence?

Kai: It’s not just one step; they describe a whole process: Ion–photon entanglement, then photon-SiV entanglement, and finally heralded ion–SiV entanglement.

Mira: They detail the specific steps in the process: the ion emits a four hundred ninety-three point five nm photon whose polarization is entangled with its spin state, which they collect via an objective outside the vacuum chamber and couple into a single-mode fiber <ref:2610.10705#pg2,ion emits a 493.5 nm photon whose polarization is entangled with>.

Lev: And then what happens to that photon after it leaves the vacuum environment?

Kai: After coupling, they use difference-frequency generation in a periodically poled lithium niobate crystal pumped at one thousand four hundred ninety-two point three nm to convert the photon to seven hundred thirty-seven point four five nm and preserve its polarization <ref:2610.10705#pg2,difference-frequency generation in a periodically poled lithium niobate>.

Mira: That converted photon then gets reflected from the SiV cavity system, and its detection is what heralds a Bell state between the ion and the quantum memory.

Lev: So that’s how they get that final heralded entanglement you need for quantum networking between modules.

Kai: They also explain how they manage the timing during this process, detailing everything from initialization to readout states in their control flow diagram.

Mira: So it’s a complete picture of the sequence, showing exactly when each subsystem acts on the ion or the memory.

Lev: That level of detail is what you need to know if you want to try and build something similar with your own hardware.

The paper's improvements: Kai: The paper points out a few ways they can improve their results, focusing on making the rate higher and the fidelity better. They suggest increasing ion-to-fiber coupling and making the signal chain more efficient overall.

Mira: They also have a specific suggestion about the gate itself: replacing that amplitude-based spin–photon gate of the SiV with a phase gate.

Lev: That’s interesting because changing the type of interaction fundamentally changes how you build the entanglement, so it has to be done carefully.

Kai: If they can do that, they think it would raise the entanglement rate to over five kHz, which is significantly higher than their current one point zero three(one) kHz result <ref:2610.10705#pg2>.

Mira: So if they can achieve that five-kilohertz rate while maintaining good fidelity, that would make a huge difference for scaling up the technology.

Lev: Five thousand Hertz is definitely where you start talking about practical applicability for continuous logical operations in a modular setting.

Kai: They also mentioned using mid-circuit reset techniques to keep coherence at ninety-one percent by applying a sequence of operations to correct for errors caused by photons that get entangled with the SiV but don't register a herald at the SNSPDs due to losses <ref:2610.10705#pg2>.

Mira: That shows they are thinking about how to actively manage those losses in real-time rather than just hoping they happen infrequently.

Lev: It’s smart to have a method for active error management when you’re pushing performance limits on hardware like this.

Conclusion: Kai: To wrap up, the paper, "Quantum Networking at the Speed of Quantum Computation: Kilohertz Entanglement in a Heterogeneous Quantum System," shows they achieved an average rate of one point zero three(one) kHz with an average fidelity of eighty-seven point nine(seven) percent through their ion-SiV system <ref:2610.10705#pg2>.

Mira: It establishes that this hybrid approach is viable for creating entanglement links at a rate that’s relevant for the ten to one hundred kHz requirements in distributed quantum computation, even though they noted there's some error budget to manage.

Lev: I think the key takeaway is demonstrating how these different platforms can actually work together effectively, which is a step toward building scalable quantum infrastructure.

Kai: It shows that this heterogeneous system can act as a photonic front-end for ion processors for applications like PCA, or even for blind quantum computing where SiVs receive weak photonic signals.

Mira: Overall, it’s a solid demonstration of how to use different physical systems together to move towards faster entanglement rates in quantum networks.

Lev: It's a necessary piece of the puzzle if we are going to build those larger processors that require that level of throughput.

IonQ, Inc. · Duke Quantum Center Departments of Electrical and Computer Engineering and Physics, Duke University

quant-ph

Submitted: 2026-10-07

Updated: 2026-10-09

Code: https://github.com/ionq-publications/khz-heterogeneous-entanglement

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 86/100

The gist: The gist: This work introduces a hybrid architecture that entangles a trapped ion qubit with a solid-state quantum memory to achieve kilohertz rates, which are comparable to the cycle time of planned

Key concepts

Hybrid Architecture
The system combines two different quantum platforms: a trapped ion qubit and a solid-state quantum memory (SiV−). This hybrid approach links the atomic qubit's photon emission with the high efficiency of the solid-state memory, overcoming challenges in interfacing these distinct physical systems.
Entanglement Mechanism
Entanglement is achieved by having an ion emit a photon whose polarization is linked to its spin state. This photon is collected, converted into a usable signal via difference-frequency generation in a PPLN crystal, and then detected by the SiV− center. A single successful detection acts as a 'herald,' confirming the entanglement between the ion and the memory qubit.
Kilohertz Entanglement Rate
The system generates ion-SiV Bell pairs at an average rate of 1.03 kHz. This speed is significant because it is four times faster than current photonic links between trapped ions and matches the processing cycle time required for continuous logical operations in a modular quantum computer.

Terminology

Summary

The gist: This work introduces a hybrid architecture that entangles a trapped ion qubit with a solid-state quantum memory to achieve kilohertz rates, which are comparable to the cycle time of planned trapped-ion processors.

System Architecture and Entanglement Mechanism

The core innovation is the combination of free-space photon collection from an atomic qubit with the near-perfect efficiency from a solid-state memory in a cavity Here we entangle a trapped 138Ba+ ion with a silicon-vacancy (SiV−) center in a diamond nanophotonic cavity This heterogeneous link has been considered difficult because two different platforms must operate together through interfaces such as quantum frequency conversion The ion emits a photon whose polarization is entangled with the ion spin state, which is collected via an objective outside the vacuum chamber and coupled into a single-mode fiber This photon is then converted in wavelength and encoding using difference-frequency generation in a periodically poled lithium niobate (PPLN) crystal The resulting photon is reflected from the SiV−–cavity system and detected, so that a single detected photon heralds entanglement

Performance Metrics and Rate Comparison

The generated ion–SiV− Bell pairs achieve an average rate of 1.03(1) kHz with a fidelity of 87.9(7)% This rate is four times the rate of the fastest photonic link between two trapped ions and corresponds to one Bell pair per millisecond, matching the cycle time of planned trapped-ion processors Estimates for entanglement rates needed to support continuous logical operations between modules range from 10 to 100 kHz

Error Budget and Fidelity Analysis

The fidelity is determined by the product of subsystem efficiencies, where Rent = frep ηion ηsig ηmem ηprot The predicted heralded entanglement rate based on calibrated subsystem efficiencies is 1.21(20) kHz, which lies 14 % below the measured rate of 1.032(11) kHz The fidelity error budget shows that the ion contributes most with a reduction of ∆F = 4.3 %, mainly from optical-pumping light that leaks through the electro-optic modulators after a herald The memory contributes 3.9 %, about 3 % of which arises from the finite spin-dependent reflection contrast of the cavity-coupled SiV− and from nuclear spin flips during the sequence

Control Sequence and Real-Time Corrections

The experimental sequence involves initialization, entanglement attempt, reset, and readout states During the readout step, two real-time corrections are applied to compensate for variable photon emission time and which detector clicked These corrections can be applied after the measurement by relabeling the outcomes according to which detector clicked and when The system uses an FPGA-based sequencer that executes pulse sequences and can branch in real time

Future Potential and Applications

The authors expect that a higher ion-to-fiber coupling, a more efficient signal chain, and replacing the amplitude-based spin–photon gate of the SiV− by a phase gate will raise the entanglement rate to > 5 kHz This heterogeneous system can serve as a photonic front-end that receives photons for ion processors for algorithms such as quantum principal component analysis (PCA) Similarly, SiVs can receive weak photonic signals over fiber in quantum communication or blind quantum computing

Experimental Setup Details

The setup utilizes a dilution refrigerator operating at a base temperature of 100 mK for the SiV− center The SiV− nanophotonic device is cooled in a dilution refrigerator (Bluefors LD400) to a base temperature of 100 mK The cavity QED parameters show that the cooperativity C = 4g 2/ (κγ) = 25.8(9) The overall single-pass efficiency of the basis converter is 59(2)%

State Reconstruction and Tomography

The density matrix of the generated ion-SiV− entangled state is reconstructed using nine measurement settings A, B ∈ X, Y, Z The reconstruction uncertainty is estimated by generating 1000 synthetic datasets from the fitted density matrix The final fidelity to the target Bell state Φ+⟩ is obtained as F = ⟨Φ+ρˆΦ+⟩

Control Hardware and State Machine

The experiment is controlled by a Quantum Machines OPX1000, which generates all time-varying control signals, executes the real-time logic and flow control, and serves as the primary time base for the experiment The state machine consists of Initialization, Entanglement Attempt, Reset, and Readout states The readout step involves a 1762 nm π pulse first shelving ↑ion⟩ into D⟩ (see Sec. S1 for ion state definitions) The final state delivered is therefore Φ+⟩ = ↑ion↓SiV⟩ + ↓ion↑SiV = 00⟩ + 11⟩

Data Availability and AI Usage

The data that support the findings of this study, with the source data and analysis code for all figures, are available at https://github.com/ionq-publications/khz-heterogeneous-entanglement During the preparation of this work, we used AI tools from Anthropic and OpenAI to improve readability and language, and to generate code for data visualization The authors take full responsibility for the content of the publication

Acknowledgments

We thank Ashish Kalakuntla, Brendan Pankovich, Isabella Goetting, Jesse AmatoGrill, John Hannegan, Mayer Feldman and Sergey Dushenko for helpful discussions The data and code availability section confirms that the data are available at https://github.com/ionq-publications/khz-heterogeneous-entanglement The paper includes a detailed error budget for the ion–SiV− Bell-state fidelity, showing total error probability of 9.36% This analysis is crucial for understanding the limitations of current entanglement generation protocols in modular quantum computing. The results establish a step toward the 10–100 kHz rates required for distributed quantum computation.

Improvements for AI systems

  1. Bold header: Enhanced Modular Quantum Computing Architecture

Modular quantum computers can consistently generate large numbers of high-fidelity entangled qubit pairs between modules with an estimated rate of 10 to 100 kHz, enabling fault-tolerant operations that scale beyond local two-qubit operations.

  1. Bold header: Kilohertz Entanglement Rate for Networking

The system achieves an entanglement rate of 1.03(1) kHz, which is explicitly stated as four times the rate of the fastest photonic link between two trapped ions and one Bell pair per millisecond, matching the cycle time of planned trapped-ion processors.

  1. Bold header: Heterogeneous Quantum Interface Development

The architecture successfully entangles a trapped 138Ba+ ion with a silicon-vacancy (SiV−) center in a diamond nanophotonic cavity, demonstrating a hybrid architecture that combines the free-space photon collection from an atomic qubit with the near-perfect efficiency from a solid-state memory.

  1. Bold header: High Fidelity Entanglement Protocol

The protocol achieves an average fidelity of 87.9(7) % for the ion–SiV− Bell state, which is comparable to local operations between ions, providing a step toward the 10–100 kHz rates required for distributed quantum computation.

  1. Bold header: Real-Time Error Correction and State Preservation

The system implements real-time corrections during readout, such as a phase correction on the ion compensates for the variable photon emission time, ensuring that every herald deliver[s] the same state Φ+⟩ = 00⟩ + 11⟩.

  1. Bold header: Robust Quantum Memory Coherence

The quantum memory preserves coherence against photon loss, with a Mid-circuit reset technique that keeps the coherence at 91% by applying a sequence of operations to correct for errors caused by photons that become entangled with the SiV− but do not register a herald at the SNSPDs due to losses.

  1. Bold header: Scalable Photonic Front-End Integration

The SiV centers can serve as a photonic front-end that receives photons for ion processors, allowing them to receive sparse photons from a weak optical sensing target and teleport the photon state into an ion processor for algorithms such as quantum principal component analysis (PCA).

Abstract

Distributed quantum computing requires rapid, high-fidelity entanglement between qubits in separate processor modules. Photonic links between single-atom qubits provide switchable, long-range connections, but photon loss limits their entanglement rate. Here we entangle a trapped 138 Ba+ ion with a silicon-vacancy (SiV-) center in a diamond nanophotonic cavity. A photon emitted by the ion is converted in wavelength and encoding, reflected from the SiV- --cavity system and detected, so that a single detected photon heralds entanglement. Including all conversion losses, we generate ion--SiV- Bell pairs at an average rate of 1.03(1),kHz with a fidelity of 87.9(7)%. This is four times the rate of the fastest photonic link between two trapped ions and corresponds to one Bell pair per millisecond, matching the cycle time of planned trapped-ion processors.

Sources

Related papers