Editor's synthesis & decision
Minor RevisionRevision requests
- Provide a formal, quantitative derivation for the extinction ratio of the ion-shuttling isolation mechanism to better justify the security parameter epsilon_snd.
- Include a detailed discussion addressing potential side-channel vulnerabilities arising from the shared laser infrastructure, specifically regarding whether common-mode noise could correlate measurement settings or outcomes.
- Clarify the handling of ion loss and recalibration events in the data set; explicitly state whether these periods were included in the calculation of the reported key rates to ensure the distinction between peak and sustained rates is clear.
- Provide a more detailed analysis of the sensitivity of the key rate to finite-size effects, particularly explaining the behavior of the key rate curves at low n values.
Where this could be published
At its current quality
Highly selective broad-interest science journals
- Nature
- Science
- Physical Review X
Then you could realistically aim for
Highly selective broad-interest science journals
- Nature
- Science
- Physical Review X
This manuscript is already at the ceiling of its field and is well-positioned for the most selective venues. The most important improvement required is a more rigorous, transparent defense of the isolation and side-channel assumptions to ensure the security claims are beyond reproach. This guidance is not a guarantee of acceptance, and you should perform a final check of the recent literature to ensure your novelty remains current.
edited by model: gemini-flash-lite-latest
Reviewer comments — reproduced in full, unabridged
Reviewer 1
gemini-flash-lite-latestThis manuscript reports the first experimental realization of device-independent quantum key distribution (DIQKD). By utilizing a trapped-ion platform to generate heralded entanglement and implementing a custom spatially-coupled low-density parity-check (SC-LDPC) error-correction code, the authors successfully extract a secret key with device-independent security guarantees. The work represents a significant technical milestone, demonstrating that the stringent requirements of DIQKD—specifically closing the detection loophole and maintaining node isolation—can be met with current hardware. The experimental design is rigorous, and the theoretical framework, including the use of the entropy accumulation theorem, is well-aligned with the experimental results. This is a foundational advance that validates the feasibility of device-independent cryptography in real-world systems.
Strengths
- First successful experimental demonstration of a complete DIQKD protocol.
- Innovative use of ion shuttling to achieve node isolation, a critical assumption for DIQKD.
- Development of a practical, efficient SC-LDPC error-correction code tailored for finite-size DIQKD statistics.
- Rigorous security proof incorporating finite-size effects and entropy accumulation.
Weaknesses
- The reliance on the isolation assumption is significant and, while physically motivated, remains a potential point of vulnerability that cannot be verified device-independently.
- The key generation rate is low, reflecting the inherent difficulty of the DIQKD protocol, which limits immediate practical utility.
- The manuscript does not fully address the potential for side-channel leakage through the shared laser infrastructure, despite the claim that it can be treated as a power supply.
Detailed comments
The authors state that "the measurement devices can have memory and operate according to the results of previous rounds." This is a crucial feature of the DIQKD security model, and it is well-supported by the use of the entropy accumulation theorem (EAT). However, the manuscript could be improved by explicitly discussing how the specific hardware implementation of the ion-trap control system (ARTIQ) ensures that the memory effects are bounded and do not violate the independence of the rounds beyond what the EAT accounts for. A brief clarification on whether the FPGA-based control logic introduces any unintended correlations between successive rounds would strengthen the argument for the isolation assumption.
The authors claim that "the laser control fields for both nodes were derived from the same set of sources" and argue that this is acceptable because they act as a power supply. While this is a common and often necessary simplification in laboratory settings, it is a potential weakness for a protocol aiming for device-independence. The authors should provide a more detailed justification for why the common-mode noise or potential cross-talk between the laser paths cannot be exploited by an adversary to correlate the measurement settings or outcomes. Specifically, how does the system guarantee that no information about the measurement basis choice $X_i$ or $Y_i$ is encoded in the laser field fluctuations?
Comments on figures
- Figure 3: The y-axis for 'Wall clock time / h' shows a total duration of 8.5 hours, but the bottom panel indicates significant gaps in data acquisition. The text mentions a 4.4-hour pause due to a laser failure. It would be beneficial to explicitly label the 'vertical steps' in the bottom panel to show exactly when these pauses occurred relative to the CHSH score and QBER stability. This would clarify whether the stability of the CHSH score is maintained immediately after re-locking the laser.
- Figure 4: The key rate is plotted as a function of the number of rounds $n$. The curve for $S=2.64, Q=3.3\%$ shows a significantly lower key rate compared to the other two curves. The authors should clarify if this drop is due to the increased QBER or the specific parameter choices for $\gamma$ and $\omega_{thr}$. A brief discussion on the sensitivity of the key rate to these parameters would provide valuable insight into the protocol's robustness.
Referee report rubric — reproduced exactly as submitted
Reviewer 2
gemini-3.1-flash-liteThis manuscript reports the first experimental realization of device-independent quantum key distribution (DIQKD) using trapped-ion qubits. By establishing heralded entanglement between two remote nodes and implementing a robust, finite-statistics security proof, the authors demonstrate the generation of a secret key with information-theoretic security that does not rely on assumptions about the internal calibration of the quantum devices. The experimental achievement of a detection-loophole-free Bell violation in a cryptographic context is a significant milestone for the field. The paper is technically rigorous, providing a detailed security analysis and a custom error-correction scheme that addresses the practical challenges of finite-size statistics. This work is a substantial advancement that validates the feasibility of DIQKD with current technology.
Strengths
- First experimental demonstration of fully device-independent quantum key distribution.
- Rigorous finite-size security analysis using the entropy accumulation theorem.
- Innovative use of ion shuttling to achieve the necessary isolation of quantum nodes.
- Development of a practical, capacity-approaching SC-LDPC error-correction scheme for DIQKD.
Weaknesses
- The isolation assumption remains the most critical and potentially vulnerable point, as acknowledged by the authors.
- The key rate is relatively low, which is expected for a first-of-its-kind experiment but limits immediate practical utility.
- The reliance on a shared laser source for both nodes, while argued to be acceptable, could be seen as a potential side-channel if not carefully managed.
Detailed comments
The authors state that 'the measurement devices can have memory and operate according to the results of previous rounds'. This is a critical claim for the robustness of their DIQKD protocol. While the entropy accumulation theorem (EAT) is cited as the basis for this, the manuscript could be improved by explicitly clarifying how the memory effects are bounded in the presence of the specific 'shuttling-based link isolation' used. Specifically, how does the authors' model account for potential correlations between the ion's state and the environment that might persist through the shuttling process?
Regarding the isolation assumption, the authors note that 'the information that could possibly leak through the remaining weak coupling is insufficient to compromise our conclusion of a positive key rate'. This is a quantitative argument based on the extinction ratio of the link. It would be beneficial for the authors to provide a more formal derivation or a clearer link between the measured leakage (Table S5) and the security parameter ε_snd. A more explicit discussion of how this leakage is incorporated into the final key rate calculation would strengthen the paper's claim of 'device-independent security'.
Comments on figures
- Figure 3: The 'moving average' for CHSH score and QBER is plotted with a window length of 100,000 rounds. The data show clear vertical steps in the bottom panel corresponding to 'ion loss and recalibration events'. It is unclear how these events are handled in the security proof; specifically, are the rounds immediately preceding and following these events discarded, or are they included in the total n? The authors should clarify if the 'moving average' hides any transient behavior in the CHSH score that might occur during these recalibrations.
- Figure 4: The plot shows the certified key rate as a function of the number of rounds n. The sensitivity to link performance is illustrated by three curves. It would be helpful if the authors could indicate the experimental operating point more clearly on these curves, perhaps with a marker, to demonstrate the margin of safety relative to the abort threshold.
Referee report rubric — reproduced exactly as submitted
Reviewer 3
gemini-flash-lite-latestThis manuscript reports the first experimental realization of device-independent quantum key distribution (DIQKD), a landmark achievement in the field of quantum information. By utilizing a trapped-ion platform to establish high-fidelity remote entanglement and implementing a custom error-correction code based on spatially-coupled low-density parity-check (SC-LDPC) codes, the authors successfully generate a secret key with device-independent security. The experimental setup is impressive, achieving a detection-loophole-free Bell inequality violation that is essential for the protocol's security guarantees. The theoretical framework, including the use of the entropy accumulation theorem, is rigorous and well-integrated with the experimental results. This work represents a significant milestone, moving DIQKD from a theoretical possibility to a practical reality, and it will be of substantial interest to the quantum information and cryptography communities.
Strengths
- First experimental demonstration of a fully device-independent quantum key distribution protocol.
- Innovative use of ion shuttling to achieve necessary link isolation while maintaining entanglement.
- Implementation of a custom SC-LDPC error-correction scheme that is efficient enough for practical finite-statistics DIQKD.
- Rigorous security proof incorporating finite-size effects and the entropy accumulation theorem.
Weaknesses
- The isolation assumption, while necessary, relies on technical measures (shuttling ions) that are not fully characterized in terms of their potential for information leakage.
- The experimental data acquisition rate is relatively low, necessitating long run times and frequent recalibrations.
- The reliance on a shared laser source for both nodes, while argued to be acceptable, introduces a potential point of common-mode failure that requires careful justification in the security model.
Detailed comments
The authors state that 'By design, the trapped-ion quantum bits are well-coupled to the optical fibre linking the nodes during the entanglement generation phase. To meet the isolation assumption, the ions in each node are shuttled 3 μm from the focus of the imaging system after a successful herald by modifying the trapping potential.' While this is a clever solution, the manuscript does not provide a rigorous, independent quantification of the leakage during this 'disconnected' state. Given that the security of DIQKD is entirely predicated on the isolation assumption, the authors should provide more explicit data or a more detailed model of the extinction ratio of this 'hatch' mechanism. A simple statement that 'The information that could possibly leak... is insufficient to compromise our conclusion' is not fully satisfying without a more detailed supporting analysis of the coupling suppression.
Regarding the shared laser infrastructure, the authors note that 'The laser control fields for both nodes were derived from the same set of sources.' While they argue that this is akin to a power supply and thus acceptable, it is a non-trivial assumption in a device-independent context. The authors should explicitly discuss whether any potential side-channel could exist where the laser intensity or phase fluctuations (which are common to both nodes) could be exploited by an adversary to correlate the measurement outcomes in a way that bypasses the Bell-test certification. A more robust defense of why this common-mode noise cannot be used to 'spoof' the Bell violation would strengthen the paper's claims.
Comments on figures
- Figure 3: The bottom panel shows the acquisition timestamp for each Bell pair. The text mentions 'The vertical steps, where time passed without heralds, correspond to ion loss and recalibration events.' It would be beneficial for the authors to clarify if these downtime periods were included in the calculation of the final key rate or if they were excluded. If excluded, the 'avg. rate: 63 s^-1' is a peak rate rather than a sustained rate; the manuscript should be precise about this distinction to avoid overestimating the protocol's performance.
- Figure 4: The plot shows the 'Certified key rate as a function of the number of rounds'. The curves for different S and Q parameters show a clear sensitivity. However, the manuscript does not explain the origin of the 'kink' or the specific behavior at low n. Could the authors provide a more detailed explanation of why the key rate drops so sharply at low n, and whether this is purely a finite-size effect from the EAT or if there are other contributing factors?
Referee report rubric — reproduced exactly as submitted
This document was generated by PsiPi, an AI-assisted manuscript review tool. It is not a product of real peer review, was not issued by any journal, and creates no publication record. The reviews and editorial synthesis above were written by large language models and may contain errors, omissions, or hallucinated claims — treat them as a starting point for revision, not as an authoritative expert assessment.