{"id":"0f2981e0-78f5-4865-a44c-3362010791ce","arxiv_id":"2510.03981","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A single spin-9/2 73Ge nucleus in silicon was hyperfine-coupled to a gate-defined quantum dot electron, with the coupling tunable from 180 to 350 kHz.","lead":"Researchers coupled a single 73Ge nucleus (spin 9/2, a 10-state 'qudit') to an electron in a silicon quantum dot and tuned the coupling between roughly 180 and 350 kHz using gate voltages. This adds a charge-neutral nuclear-spin resource that could support high-dimensional quantum memories and shuttling-based quantum computing in silicon.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Ten-line assignment to a single 73Ge nucleus hinges on untested intensity pattern and lack of control device.","rationale":"The reader's weakest_assumption is exactly the identification of the ten ESR peaks as the signature of a single 73Ge nucleus. My stress test finds this to be the most load-bearing concern because all other elements (hyperfine coupling, tunability) inherit from this assignment. The concern is genuine but not fatal: the evidence in the paper (isotope-selective implantation, 2I+1=10 line count, equal spacing) is strongly suggestive, yet the absence of a control device and the unanalyzed peak amplitudes leave a concrete alternative (nine equivalent 29Si nuclei) that would exactly mimic the line positions. A simple re-analysis of existing data or a control device can settle the question. Thus the verdict should remain CONDITIONAL: the paper is likely correct, but the central claim should be further substantiated before full acceptance.","tokens_in":9755,"tokens_out":12722,"duration_ms":106565,"concrete_test":"Re-analyze the existing Fig. 2(b)-(h) line cuts (data promised on Zenodo): extract the fitted amplitudes of the ten Lorentzian peaks. For a single 73Ge nucleus in thermal equilibrium at T=50 mK and B0=0.3 T, the amplitudes should be equal to within <1%. For a cluster of nine equivalent 29Si spins, they must follow the binomial distribution C(9,k) with central peaks ~7x larger than outer ones. If amplitudes are consistent with binomial statistics, the 73Ge assignment is falsified. As a second check, fabricate and measure a nominally identical device without 73Ge implantation under the same conditions; absence of the 10-line pattern would rule out an intrinsic defect origin.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim rests on identifying the ten equally spaced ESR peaks in Fig. 2 as the hyperfine transitions of a single 73Ge nucleus (I=9/2). The most dangerous alternative is a collective state of nine equivalent 29Si nuclei (I=1/2), which would also produce ten equally spaced lines at frequencies m·A, but with binomial peak amplitudes (1:9:36:84:126:126:84:36:9:1) rather than the near-equal intensities expected for a single thermal spin-9/2. The paper does not report the fitted Lorentzian amplitudes, and no control device without 73Ge implantation is shown to rule out an intrinsic defect or a 29Si-cluster origin. Given the dot contains 800 ppm residual 29Si, and the earlier Hensen et al. result that such dots can couple to two individual 29Si nuclei, the multinuclear possibility is not entirely hypothetical. If the ten lines arise from something other than a single 73Ge nucleus, the central claim fails.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports the observation of hyperfine coupling between a single 73Ge nuclear spin (I=9/2) and an electron spin in a gate-defined SiMOS quantum dot. The ESR spectrum exhibits ten equally spaced peaks, assigned to the ten nuclear spin projections, with a hyperfine constant A ≈ 353 kHz. The authors further demonstrate electrical tuning of A from ~180 to ~350 kHz via confinement gate voltages. The work is positioned as a step toward using 73Ge as a spin-9/2 qudit in silicon quantum dots, exploiting the isoelectronic nature of Ge for electron shuttling.","tokens_in":10033,"tokens_out":9054,"duration_ms":79393,"significance":"If the identification holds, this is the first demonstration of a single 73Ge nuclear spin hyperfine-coupled to an electrostatically defined quantum dot, extending the toolkit of nuclear qudits in silicon to isoelectronic group-IV impurities. The direct extraction of A from the peak spacing is model-independent, and the repeated measurements showing stability (353–357 kHz) are a strength. The gate-voltage tunability is a useful feature for future shuttling-based architectures. However, the evidence for the nuclear species assignment is incomplete, which tempers the significance until the ambiguity is resolved.","major_comments":[{"comment":"The assignment of the ten equally spaced ESR peaks to a single 73Ge nucleus relies solely on the peak count and the implantation dose. The paper does not report the fitted Lorentzian amplitudes, which would discriminate between a single thermal I=9/2 nucleus (near-equal intensities) and an alternative of nine equivalent 29Si nuclei (I=1/2) producing the same number of equally spaced lines with binomial intensities (1:9:36:84:126:126:84:36:9:1). Given the 800 ppm residual 29Si and the fact that Ref. 14 observed hyperfine-coupled 29Si nuclei in a similar device, this alternative is not implausible. The authors should provide the measured peak amplitudes (or a control device without 73Ge) and compare to the expected intensity patterns. Without this, the central claim that the coupled nucleus is 73Ge is not fully established.","section":"Nuclear spin signature, Fig. 2"},{"comment":"The manuscript states that the data supporting this work are available in a Zenodo repository, but gives no URL, DOI, or accession code. Because the central claims rest on fitting ten-peak spectra and the tuning curve, the absence of a retrievable dataset prevents independent verification of the peak amplitudes and the extracted A values. Please provide the repository link in the final version.","section":"Data Availability (Appendix)"}],"minor_comments":[{"comment":"There is a numerical inconsistency between the text and the figure: the text says the linear fit slope is m = 473 kHz/V, while the equation in the figure is A(VCB) = 437 kHz/V · VCB + 227 kHz. One of these is a typo and should be corrected.","section":"Fig. 3(b)"},{"comment":"The Hamiltonian in Eq. (2) uses γe and γGe as positive constants, while the ESR frequency is written as f_ESR = |γe|B0 + m_I A. Please clarify the sign convention for the electron gyromagnetic ratio and define the ordering of m_s levels.","section":"Eq. (2) and f_ESR"},{"comment":"The sentence 'In Eq. 3, we assume...' refers to the Fermi-contact expression labeled (3). Ensure the equation numbering is clear in the final typeset version, as the current text might be ambiguous.","section":"Eq. (3) reference"}],"recommendation":"major_revision","confidential_remarks":"The paper reports a potentially significant result, but the identification of the ten-line spectrum as 73Ge needs to be tightened. The lack of amplitude analysis or a control device leaves a plausible 29Si-cluster alternative on the table. I am not asking for a new device necessarily; a careful intensity analysis and an explicit discussion of the 29Si scenario could suffice. Also, the missing Zenodo link is a serious reproducibility issue that should be fixed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nWorth your time: this is the first demonstration of a hyperfine-coupled 73Ge nuclear spin (I=9/2) in a gate-defined silicon quantum dot. The spectroscopy is clean. In Fig. 2, the authors see ten equally spaced ESR peaks with A=353(1) kHz, and the spacing is stable across repetitions (353-357 kHz). The voltage-tuning data in Fig. 3 show A changing from 179 kHz to 357 kHz, roughly a factor of two, with the simulated wavefunction shift providing a plausible mechanism. That is a solid, well-executed extension of the Hensen et al. approach to a spin-9/2 isotope, and it opens a path toward isoelectronic qudits that are shuttle-compatible.\n\nThe central claim holds up. The ten-line pattern is essentially conclusive for an I=9/2 system; a 29Si cluster alternative would require nine equivalent spin-1/2 nuclei with identical coupling, which contradicts the known hyperfine landscape in these dots and the prior observation of only two 29Si nuclei in a similar device. The stress-test worry about binomial intensities is not supported by the physics here. Still, the paper would be stronger with a 73Ge-free control device or a clear statement that the earlier device in Ref. 14 serves that role.\n\nSoft spots, in order of importance:\n\n1. Data availability: the text says \"available in a Zenodo repository\" but gives no DOI or link. For a spectroscopy paper, the raw traces should be findable. Fixable.\n2. \"Readout\" in the abstract overstates what is shown. They observe the nuclear state via the electron spin resonance condition, which is a form of projective measurement of the nuclear spin, but they haven't demonstrated single-shot nuclear readout or repeated weak measurement. The conclusion actually admits this. Adjust the wording.\n3. Minor inconsistency: the text gives the linear fit slope as 473 kHz/V, while the Fig. 3 caption says 437 kHz/V. Same intercept, different slope. One of them is wrong. Check.\n4. The paper says \"Eq. 3\" but the equations are unnumbered; trivial.\n\nNot fatal: the absence of ENDOR initialization and the weakly effective J-gate are acknowledged as limitations, so the lack of coherent nuclear control is not hidden. The simulation in Fig. 3 is qualitative but adequate for the claim of tunability.\n\nThis deserves a serious referee. It is not a breakthrough outside the silicon spin-qubit community, but within that community it is the first 73Ge qudit resource and will be cited. Send it to review; the authors will need a revision to fix the data link and the slope inconsistency, but the physics is believable.","headline":"First clean ESR spectrum of a single 73Ge nucleus in a SiMOS dot; worthy of refereeing, with two fixable gaps and a small overclaim.","tokens_in":10595,"tokens_out":2752,"would_cite":true,"duration_ms":24958,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A single 73Ge nuclear spin couples to a silicon quantum dot electron, with hyperfine coupling tunable from 180 to 350 kHz via gate voltages.","keywords":["73Ge nuclear spin","silicon quantum dot","hyperfine interaction","spin-9/2 qudit","electron spin resonance","ion implantation","SiMOS","nuclear spin readout"],"falsifier":"Run the identical experiment on a device fabricated without any 73Ge implantation; if the same ten-line pattern appears, or if a double-resonance measurement shows a nuclear Larmor frequency different from 73Ge's expected gyromagnetic ratio of 1.49 MHz per tesla, the central claim is wrong.","tokens_in":9676,"feed_emoji":"⚛️","tokens_out":12046,"duration_ms":146847,"temperature":0.7,"pith_summary":"The paper sets out to show that a single 73Ge nucleus — a spin-9/2 system with ten accessible states — can be hyperfine-coupled to the electron of a gate-defined quantum dot in silicon. Using isotope-selective implantation of 73Ge near the Si/SiO2 interface, the authors observe ten equally spaced electron spin resonance lines, the fingerprint of a single spin-9/2 nucleus, with a spacing A = 353(1) kHz. They then show that A is electrically tunable from about 180 to 350 kHz by changing the lateral gate voltages, which moves and re-confines the electron wavefunction. The paper presents this as a viable ten-level qudit for silicon quantum computing, with the advantage that, being isoelectronic, it adds no charge and the electron can be shuttled away without destroying the nuclear state.","feed_headline":"Single 73Ge nucleus couples to a silicon quantum dot electron","feed_subtitle":"Hyperfine coupling tunes from 180 to 350 kHz by gate voltages, opening a ten-state nuclear qudit.","key_machinery":"The Fermi-contact hyperfine interaction, A = (2/3)ℏ² μ0 γe γGe |Ψ(0)|², is the central coupling. It links the electron spin to the 73Ge nuclear spin (I=9/2) and splits each electron spin resonance into 2I+1 = 10 equally spaced lines separated by A. The electron wavefunction density at the nucleus, |Ψ(0)|², is what the gate electrodes alter; moving and confining the electron changes |Ψ(0)|² and hence A, yielding the observed tunability from 180 to 350 kHz. The ten-line spectrum is the direct experimental signature of a single spin-9/2 nucleus.","core_discovery":"In a SiMOS device, a single 73Ge nuclear spin is coupled to a quantum-dot electron through the Fermi-contact hyperfine interaction. The electron spin resonance spectrum shows ten equally spaced peaks with spacing A = 353(1) kHz, matching a spin-9/2 nucleus. The coupling strength is tuned between 179(1) and 357(2) kHz by adjusting the confinement gate voltage, which shifts and squeezes the electron wavefunction, changing its density at the nucleus. This is the first demonstration of a 73Ge nucleus coupled to a gate-defined quantum dot, extending the family of isoelectronic nuclear spins in silicon beyond 29Si to a high-spin qudit.","pith_inferences":["If the identification holds, the same fabrication approach could place multiple 73Ge nuclei at known positions, each serving as a ten-level qudit, with electron shuttling providing a scalable coupling mechanism between them — a step the paper leaves implicit.","The demonstrated range of A (roughly 180–350 kHz) is small compared with donor hyperfine couplings (~100 MHz), implying that nuclear readout via the electron will be slower; quantifying this trade-off experimentally would be a natural next step.","Because no control device without 73Ge is shown, a direct comparison with an unimplanted but otherwise identical device would test whether the ten-line pattern is specific to 73Ge, ruling out alternative spin-9/2 defects.","The linear dependence of A on the simulated wavefunction shift suggests that A could serve as a sensitive, in-situ probe of the electron wavefunction position and confinement in such devices."],"forward_implications":["The ten-dimensional Hilbert space of a single 73Ge nucleus is now addressable in a gate-defined dot, enabling future qudit control and readout experiments.","Because 73Ge is isoelectronic, the quantum-dot electron can be shuttled away and back without destroying the nuclear spin coherence, a key requirement for scalable architectures that move electrons between dots.","The gate-voltage tunability of the hyperfine coupling provides a control knob for adjusting the electron-nuclear interaction strength during device operation.","With improved electrostatic control (e.g., a more effective J-gate), deterministic initialization via ENDOR and coherent nuclear spin manipulation should become possible, as the authors note.","In the longer term, this could support entanglement distribution between distant nuclear spins and repeated weak measurements of a nuclear spin, as the authors suggest."],"fun_headline_variants":["First 73Ge nuclear spin coupled to a gate-defined quantum dot","Tunable hyperfine coupling of a 73Ge nucleus from 180 to 350 kHz","Spin-9/2 73Ge qudit tuned by quantum dot voltages","73Ge nuclear spin readout and tuning in a SiMOS dot"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The claim assumes that the ten equally spaced lines in the electron spin resonance spectrum come from a single 73Ge nucleus in the quantum dot and not from some other impurity or a set of nuclei.","fun_headline_variants_meta":{"raw":{"variants":["First 73Ge nuclear spin coupled to a gate-defined quantum dot","Tunable hyperfine coupling of a 73Ge nucleus from 180 to 350 kHz","Spin-9/2 73Ge qudit tuned by quantum dot voltages","73Ge nuclear spin readout and tuning in a SiMOS dot"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000225,"raw_usage":{"total_tokens":1308,"prompt_tokens":758,"completion_tokens":550,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":502,"completion_tokens_details":{"reasoning_tokens":467}},"tokens_in":502,"tokens_out":550,"duration_ms":9648,"temperature":1.0,"reasoning_tokens":467,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T11:32:23.640460+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the identical experiment on a device fabricated without any 73Ge implantation; if the same ten-line pattern appears, or if a double-resonance measurement shows a nuclear Larmor frequency different from 73Ge's expected gyromagnetic ratio of 1.49 MHz per tesla, the central claim is wrong.","supporting_citations":[],"review_version":1}