{"id":"dab6673e-63b9-4804-a418-0fdbc97b2520","arxiv_id":"2505.09267","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"A zero-field protocol for 117SnV- color centers keeps the nuclear memory qubit insensitive to optical excitation, demonstrated with high-fidelity microwave control in a photonic integrated circuit.","lead":"A diamond-based quantum device using an SnV color center's nuclear spin is shown to operate as a memory at zero magnetic field, with 97.8% gate fidelity and 2.5 ms coherence. The protocol aims to make quantum network nodes simpler by removing the need for large magnets and dilution fridge temperatures.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Optical insensitivity of the memory is not directly measured; the central ΔωM=0 claim rests on a degeneracy that the experiment itself shows is broken (1.4 MHz Ramsey beat, finite f0 cyclicity), and the quoted 10.4 kHz estimate is conditional on an unverified residual-field assumption.","rationale":"The paper has two logically distinct achievements. First, it demonstrates high-fidelity control (97.8% randomized-benchmarking gate fidelity, 2.5 ms decoupling-enhanced coherence) of the 117SnV- hyperfine ground-state manifold at zero applied field; these numbers are credible as reported and are not the object of this concern. Second, it proposes that the |1B0M>/|1B1M> degeneracy makes the memory invisible to optical excitation, enabling brokered entanglement protocols. This second claim is the one that matters for the paper's stated purpose, and it is not directly measured. The theoretical degeneracy follows from the D3d-symmetric Hamiltonian of Appendix B. The experiment observes a clear violation of that degeneracy in the ground state: a 1.4 MHz Ramsey beat and finite f0 cyclicity fitted by a 223 μT residual field. The paper's own limitation statements (Conclusion and Appendix G) concede that the source of the residual splitting is not identified and that the 10.4 kHz ΔωM estimate is conditional. If the splitting is intrinsic (Jahn-Teller distortion or strain-hyperfine coupling), there is no reason to expect the same splitting in the excited state; the phase kickback could be much larger, and the claimed ~1.5 million excitations before 95% memory fidelity would be invalid. The concrete test above would settle this using the existing apparatus. The reader's conditional verdict is appropriate; no adjustment is needed.","tokens_in":19195,"tokens_out":5092,"duration_ms":48423,"concrete_test":"Directly measure ΔωM: prepare the memory in a superposition on the 0B0M↔0B1M (purple) transition, then interleave N resonant optical π-pulses on the f0 transition (with the repump/readout sequence of Fig. 3a) and record the Ramsey or spin-echo contrast as a function of N. Fit the decay to Eq. (1) with the measured 6 ns lifetime to extract ΔωM. If the extracted value is ≤2π·10 kHz out to N~10^4, the zero-field insensitivity claim is supported; if the contrast decays substantially faster (ΔωM≫10 kHz), the residual-splitting model is wrong and the central networking advantage fails. This uses only the existing setup and avoids any dependence on the unmeasured excited-state Hamiltonian.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that the degeneracy of the |1B0M> and |1B1M> states in both the ground and excited manifolds makes the memory insensitive to optical excitation (ΔωM=0). This degeneracy is exact only within the D3d-symmetric group-theory Hamiltonian of Appendix B. The experiment itself reveals a broken degeneracy: a ~1.4 MHz beat in the Ramsey fringes involving the |1B> states, and finite cyclicity (Λ=132) of the f0 transition. The paper fits these observations to a residual 223 μT magnetic field, but explicitly notes in the Conclusion and Appendix G that Jahn-Teller distortion or strain-dependent hyperfine coupling are alternatives. Appendix G then estimates ΔωM=2π·10.4 kHz only 'under the assumption that the residual splitting is caused by a magnetic field (or that the effective magnetic field due to other effects is the same in the ground and excited state).' That assumption is untested. If the degeneracy is broken by an intrinsic symmetry-lowering term, the same term will generally split the excited |1B> levels differently, so optical excitation can produce a memory phase kickback orders of magnitude larger than 10.4 kHz. The 97.8% gate fidelity and 2.5 ms coherence time are credible demonstrations of ground-state control, but they do not probe the optical-excitation channel, which is the basis for the networking claim. The conclusion that 'optical excitation does not disturb information in the memory qubit' is therefore not yet established for this device.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proposes and partially demonstrates a zero-field protocol for 117SnV- color centers in which the two |1B> ground states and the corresponding excited states are degenerate by D3d symmetry, so that resonant optical excitation of the broker transition f0 leaves a superposition stored in the 117Sn nuclear memory untouched. The authors integrate a 117SnV- center in a photonic integrated circuit, measure its optical and hyperfine spectra, demonstrate microwave control of all three ground-state transitions, characterize fidelities by randomized benchmarking (97.8% physical gate fidelity for the memory transition, 92.3% for the broker), and extend the memory coherence time to 2.5 ms with two decoupling pulses. They observe two deviations from the standard group-theory model—finite optical cyclicity of f0 (Λ≈132) and a 1.4 MHz residual splitting of the |1B> states—and model both with a 223 µT residual magnetic field. Under the additional assumption that this field is identical in the ground and excited states, they estimate ΔωM = 2π·10.4 kHz and claim the memory could survive roughly 1.5 million optical excitations.","tokens_in":19533,"tokens_out":12612,"duration_ms":121868,"significance":"If the optical-insensitivity claim were directly verified, this would be an important step: it would provide a deterministic nuclear memory in a group-IV color center that is protected during broker readout and entanglement attempts, while operating at zero applied field and 1.3 K. The experimental strengths are real: the integration of 117SnV into a PIC, the clean microwave control data, the high reported gate fidelities, the 2.5 ms memory coherence, and the authors' transparent reporting of deviations from the standard theory and of the assumptions behind their quantitative estimate. The paper also makes two cleanly falsifiable quantitative predictions (ΔωM = 2π·10.4 kHz and an excitation budget of order 1.5 million) that are directly testable. The main significance is currently contingent, however, because the central networking claim rests on a degeneracy that the data show is broken, and the memory response to optical excitation is not directly measured.","major_comments":[{"comment":"The central claim in Sec. I.B that 'ΔωM in equation 1 is identically 0 when no magnetic field is applied' is not supported by the data presented. In Sec. II.C the authors report a 1.4 MHz interference beat in the Ramsey fringes of the transitions involving the |1B> states, which they attribute to an imperfect degeneracy of |1B0M> and |1B1M>. In Appendix G this is modeled by adding a 223 µT residual DC magnetic field, and the estimate ΔωM = 2π·10.4 kHz is obtained only 'under the assumption that the residual splitting is caused by a magnetic field (or that the effective magnetic field due to other effects is the same in the ground and excited state).' This assumption is not tested, and the excited-state hyperfine parameters entering the estimate (A∥^exc = -232 MHz, A⊥^exc = 464 MHz) are themselves inferred from a dipole-dipole form and the zero-strain optical splitting rather than measured in this device. If the degeneracy is broken by an intrinsic symmetry-lowering term such as Jahn-Teller distortion or strain-dependent hyperfine coupling, the excited-state splitting will generally differ from the ground-state splitting, and the optical phase kickback on the memory may be orders of magnitude larger than 10.4 kHz. The statement in the Conclusion that optical excitation does not disturb the memory therefore goes beyond the evidence.","section":"Sec. I.B, Sec. II.C, Appendix G"},{"comment":"The finite cyclicity of the f0 transition is a second, independent threat to the protocol's central claim. In Sec. II.A the authors measure Λ≈132 for f0, whereas the standard group-theory model used in Sec. I.B predicts a perfectly cycling transition. Equation (1) and Appendix A model only the phase kickback ΔωM; they do not account for population leakage out of the |1B> manifold on a non-cycling event. With Λ≈132, after roughly one hundred optical excitations there is a substantial probability that an excitation event takes the system into the |0B> states, which will depolarize or reset the nuclear memory. The statement in Appendix G that the memory 'could be excited nearly 1.5 million times' is therefore misleading unless the cyclicity-induced leakage is included in the fidelity budget; the one-sentence caveat 'although transition cyclicity would limit fidelity well before this point' does not quantify the effect. The manuscript should either incorporate cyclicity into the fidelity model of Eq. (1) or remove the 1.5-million-excitation estimate as a headline number.","section":"Sec. II.A, Appendix G"},{"comment":"The randomized benchmarking and decoupling measurements in Sec. II.C are ground-state-only experiments. In the RB sequence the microwave gates are applied to the optical ground states and the final readout is optical, but no resonant f0 optical pulse is applied while the memory is in a superposition. Likewise, the coherence measurements in Fig. 4b do not interleave optical excitation with the XY decoupling sequence. As a result, the key requirement 1 of Sec. I—that the broker can be read out or repeatedly excited without disturbing the memory—is not experimentally demonstrated. A direct test would be to prepare the memory in a superposition, apply N resonant f0 excitation pulses, and measure the Ramsey contrast as a function of N; this would directly determine the combination of phase kickback and leakage that enters Eq. (1). Without such a measurement, the conclusion that the zero-field protocol 'lays the groundwork for building quantum network nodes' should be limited to the ground-state control result, with optical insensitivity presented as a prediction rather than an observed property.","section":"Sec. II.C, Conclusion"}],"minor_comments":[{"comment":"The relationship between the reported physical gate fidelities (97.8% memory, 92.3% broker) and the quoted Clifford fidelities (98.6% and 95.2%) is unclear; including ideal Z gates in the averaged Clifford set would not normally increase the fidelity, so the definition and conversion formula should be stated explicitly.","section":"Sec. II.C"},{"comment":"The text states that the fitted 223 µT residual field is 'an order of magnitude stronger' than Earth's magnetic field (~55 µT), but the ratio is about a factor of four; please correct this comparison.","section":"Sec. II.A, Conclusion"},{"comment":"The fit errors reported in Table II (relative errors of 10^-4 or smaller) appear to be statistical only; please state whether systematic uncertainties from the fixed parameters λ = 830 GHz, q = 0.171, and the assumed excited-state hyperfine form were propagated into the quoted ΔωM value.","section":"Appendix G, Table II"},{"comment":"In the cyclicity definition Λ = τ_pol/(2τ), please define the branching ratio into dark states and explain how saturation of the transition is handled in extracting τ_pol; this matters for the comparison with theory in Fig. 7.","section":"Sec. II.A, Eq. (3)"},{"comment":"Given that a 223 µT residual field is used to explain the data, the phrase 'zero-field regime' should be replaced by 'zero applied field' wherever it appears, and the residual-field caveat should be stated at the first use.","section":"Sec. I.B, Fig. 1f"},{"comment":"The notation for strain is inconsistent: Eq. (B2) uses α Egx/Egy, Eq. (B3) uses α Egx and then α^2_Egy, and Table I lists α_gnd; please unify the symbols and define the relation between α and the Egx/Egy components.","section":"Appendix B"}],"recommendation":"major_revision","confidential_remarks":"The ground-state work is solid and publishable; the question is whether the manuscript is allowed to claim the optical-insensitivity property without direct evidence. I would ask the authors either to add an interleaved optical-excitation/Ramsey measurement or to re-scope the title, abstract, and conclusion so that the zero-field protocol is presented as a proposal supported by ground-state control, with optical insensitivity as a clearly identified prediction. I lean major_revision rather than rejection because the required changes are within the scope of a revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Alex,\n\nThe genuinely new thing here is the zero-field protocol: use the large hyperfine splitting of 117SnV- to make the optical transitions resolvable without a magnetic field, and exploit the symmetry of the J=1 ground and excited states so that optical excitation of the broker spin leaves the nuclear memory untouched. That is a clever idea, and unlike the 29Si and 13C approaches in the literature, it is deterministic and does not need a bias field.\n\nThe experimental work is strong. They integrate 117SnV- into a photonic integrated circuit, drive all three ground-state hyperfine transitions, measure 97.8% physical gate fidelity on the best transition (98.6% Clifford for the memory), and push coherence to 2.5 ms with only two decoupling pulses. That is a real step beyond bare electron SnV- coherence. The paper is also honest: it reports two deviations from the standard theory and devotes an appendix to fitting them.\n\nThe catch is that the central claim is not directly measured. The observed 1.4 MHz Ramsey beat and the finite cyclicity of the f0 transition show that the J=1 degeneracy is broken. The authors fit a residual magnetic field of 223 µT—about ten times Earth's field—and then, only under the assumption that the same effective field acts in the excited state, they estimate ΔωM = 2π·10.4 kHz. That assumption is untested. If the degeneracy is broken by a strain-dependent hyperfine term or Jahn-Teller distortion, the excited-state splitting will generally differ, and the optical kickback could be far larger. So the protocol idea and the ground-state control are solid, but the networking claim—repeated optical excitation does not disturb the memory—is not yet established for this device.\n\nThat said, the paper deserves a serious referee. It is a well-executed experiment with a novel proposal, and the authors clearly identify the open question. The right outcome is probably publication with a request for a direct measurement of memory coherence under optical driving, or at least a measurement that constrains the excited-state perturbation. Anyone working on group-IV color centers or quantum memories should read this.\n\nMy recommendation: send it to review.","headline":"Strong experimental paper on a clever zero-field SnV- memory protocol, but the central optical-insensitivity claim rests on an untested degeneracy and a residual-field fit.","tokens_in":20156,"tokens_out":4224,"would_cite":true,"duration_ms":35340,"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":"This paper claims that 117SnV– diamond color centers can hold a nuclear memory qubit insensitive to optical readout, by exploiting a zero-field degeneracy, and demonstrates 97.8% gate fidelity with 2.5 ms coherence.","keywords":["tin-vacancy color center","117SnV–","zero-field quantum protocol","nuclear spin memory","spin-photon interface","quantum networking","diamond photonic integrated circuit","hyperfine coupling"],"falsifier":"Take a set of 117SnV– devices spanning different strain values, null the ambient field with a vector magnet, and measure the $|1B\\rangle$ Ramsey beat and the $f_0$ cyclicity at each setting. If some nulling field removes the beat and restores the theory-predicted cycling of $f_0$, the residual-field explanation and the exact $\\Delta\\omega_M = 0$ claim survive; if the beat persists at all field settings and changes with strain, the degeneracy is intrinsically broken by the hyperfine interaction and the zero-field insensitivity must be revised.","tokens_in":18967,"feed_emoji":"💎","tokens_out":15429,"duration_ms":132124,"temperature":0.7,"pith_summary":"The paper proposes and tests a zero-magnetic-field operating protocol for the 117SnV– diamond color center, in which the electronic spin acts as a network 'broker' qubit while the 117Sn nuclear spin stores quantum information as a local memory. Its central claim is that at zero field the two $|1B\\rangle$ hyperfine states remain exactly degenerate in both the ground and excited electronic manifolds, so the memory's precession frequency does not change during optical excitation ($\\Delta\\omega_M = 0$) and readout or entanglement attempts leave the stored state undisturbed. Unlike 29SiV– or weakly coupled 13C memories, which need a field bias and still suffer excitation-induced phase kickback, 117SnV– has a hyperfine optical splitting larger than the linewidth even without any field. The paper demonstrates the protocol in a diamond emitter integrated on a photonic circuit at 1.3 K: 98.4% optical polarization, 97.8% physical gate fidelity on the memory transition, and 2.5 ms memory coherence with two decoupling pulses. If the claim holds, group-IV color centers gain a built-in, deterministic memory that survives the many excitation attempts required for entanglement distillation and repeater protocols.","feed_headline":"Zero-field scheme lets a diamond nuclear memory ignore readout light","feed_subtitle":"By exploiting an exact zero-field degeneracy, 117SnV– shows 97.8% gate fidelity and a 2.5 ms memory coherence time.","key_machinery":"The load-bearing object is the zero-field degeneracy of the $|1B0_M\\rangle$ and $|1B1_M\\rangle$ states: the $J = 1$, $m_J = \\pm 1$ joint electron-nuclear states in which the broker spin and 117Sn memory spin are aligned or anti-aligned. In the effective Hamiltonian built from spin-orbit coupling, transverse strain, and hyperfine terms $A_\\parallel, A_\\perp$, this pair sits at energy $A_\\parallel/4$ with no dependence on strain, while the $|0B\\rangle$ pair splits with strain; because the same pairing recurs in the excited manifold, the memory transition frequency is unchanged by optical excitation. The protocol also requires the hyperfine optical splitting to exceed the linewidth without a bias field, and the microwave scheme (three allowed transitions among the four ground states) to provide independent broker control, controlled-memory rotations, and a SWAP for full two-qubit control. A second-order expansion of the same Hamiltonian shows the memory transition has $\\lambda_M = 0$, meaning no coupling to the phonon-coupled orbital branch, which explains the memory's 'nuclear-like' coherence.","core_discovery":"The paper's central claim is that 117SnV– has hyperfine optical transitions separated by more than the optical linewidth at zero magnetic field, and that its joint electron-nuclear level structure then admits a memory encoding that is exactly insensitive to optical excitation. In the ground manifold the energies of the states $|1B0_M\\rangle$ and $|1B1_M\\rangle$ are both $A_\\parallel/4$, independent of strain and of the transverse hyperfine component $A_\\perp$; the same $|1B\\rangle$ pairing recurs in the excited manifold, so the memory transition frequency is identical in ground and excited states and $\\Delta\\omega_M$ in the fidelity formula is identically zero at zero field. Optical pulses on the $f_0$ transition therefore imprint no random phase on a stored memory superposition, replicating for a group-IV center what the $m_S = 0$ state provides in NV–. The authors back this with experiments in a photonic integrated circuit: 98.4% ground-state polarization, all three ground microwave transitions driven at meghertz Rabi rates, Clifford fidelities of 98.6% (memory) and 95.2% (broker), and 2.5 ms coherence for the memory. They also report that the $|1B\\rangle$ degeneracy is imperfect in their device — a ~1.4 MHz Ramsey beat and finite cyclicity of $f_0$ — which they fit to a residual ~223 µT field and note could instead come from Jahn-Teller distortion or strain-dependent hyperfine coupling.","pith_inferences":["A testable screening rule follows from the paper's numbers: any group-IV (or other) defect whose ground-versus-excited hyperfine difference exceeds its optical linewidth, and whose $m_J = \\pm 1$ pairing is preserved in the excited state, should admit the same zero-field no-kickback encoding; the paper's Table I parameters give the explicit threshold to apply.","The observed ~1.4 MHz degeneracy-breaking offers a clean way to discriminate the two proposed explanations: if the beat frequency and the $f_0$ cyclicity track strain across devices and persist under active magnetic-field nulling, the hyperfine interaction itself breaks the degeneracy, and the protocol would need a corrected (non-zero) $\\Delta\\omega_M$ or field compensation.","The 5.4 percentage-point gap between the memory and broker physical gate fidelities is plausibly the Bloch-Siegert shift of the 31 MHz transition at high drive power, which the paper avoided by running at low power; shaped or composite pulses that pre-compensate that shift should close the gap without slowing the gates."],"forward_implications":["A 117SnV– node could survive orders of magnitude more excitation attempts than existing group-IV memories: the paper estimates ~1.5 million optical excitations before memory fidelity drops below 95%, against ~2 for the 29SiV– bias scheme, making distillation and repeater protocols practical.","Network nodes could operate without superconducting magnets or dilution refrigerators, since the demonstrated control runs at 1.3 K with zero applied field.","The memory qubit is predicted to be immune to phonon-mediated dephasing ($\\lambda_M = 0$), giving 'nuclear-like' coherence of 2.5 ms with only two decoupling pulses — an order of magnitude beyond bare-electron SnV– under comparable decoupling.","The same zero-field level structure is predicted to transfer to other heavy group-IV centers such as 73GeV–, 207PbV–, and 61NiV–, extending the protocol beyond tin.","Combined with demonstrated high-cooperativity SnV– cavities, the protocol should enable single-shot readout and heralded spin-spin entanglement with a protected local memory in place."],"supporting_citations":[{"why":"The NV– brokered-entanglement demonstration whose mS = 0 no-kickback scheme the zero-field protocol seeks to replicate for group-IV centers.","marker":"[1]"},{"why":"The 29SiV– network node with stored nuclear memory states, whose ~35 MHz optical phase kickback is the limitation the zero-field protocol removes; also the comparison baseline for memory fidelity per excitation.","marker":"[4]"},{"why":"The heralded entanglement demonstration between nanostructure-integrated group-IV color centers that motivates adding a protected local memory.","marker":"[12]"},{"why":"Supplies the measured group-IV hyperfine trend and the 117SnV– Hamiltonian whose level structure the zero-field protocol exploits.","marker":"[22]"},{"why":"Prior demonstration of information storage in 117SnV– hyperfine levels; supplies the zero-strain optical hyperfine splitting (453 MHz) used to fix excited-state parameters.","marker":"[23]"},{"why":"The group-theory Hamiltonian and transition matrix elements used to predict the level structure, cyclicity, and the f0 anomaly.","marker":"[25]"},{"why":"Supplies the cyclicity relation Λ = τpol/2τ used to quantify the finite cyclicity of f0.","marker":"[27]"},{"why":"The bare-electron SnV– coherence measurement (100–400 µs) that provides the decoupling-pulse comparison baseline and fixed fit parameters (λ = 830 GHz, q = 0.171).","marker":"[28]"},{"why":"The phonon-decoherence theory for group-IV centers that predicts λM = 0 for the memory subspace, explaining the 2.5 ms coherence.","marker":"[31]"}],"fun_headline_variants":["Zero-field diamond nuclear memory ignores readout light","117SnV memory ignores light at zero field with 97.8% fidelity","Exact zero-field degeneracy shields diamond nuclear spin from light","Zero-field 117SnV memory shows 97.8% gate fidelity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole zero-field no-kickback guarantee rests on one premise: that the $|1B0_M\\rangle$ and $|1B1_M\\rangle$ states are exactly degenerate at zero magnetic field in both the ground and excited manifolds — and the paper's own device shows a residual ~1.4 MHz splitting, so if that splitting is intrinsic to the hyperfine interaction rather than a removable stray field, the $\\Delta\\omega_M = 0$ claim is only approximate.","fun_headline_variants_meta":{"raw":{"variants":["Zero-field diamond nuclear memory ignores readout light","117SnV memory ignores light at zero field with 97.8% fidelity","Exact zero-field degeneracy shields diamond nuclear spin from light","Zero-field 117SnV memory shows 97.8% gate fidelity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001049,"raw_usage":{"total_tokens":4465,"prompt_tokens":1064,"completion_tokens":3401,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":680,"completion_tokens_details":{"reasoning_tokens":3326}},"tokens_in":680,"tokens_out":3401,"duration_ms":25824,"temperature":1.0,"reasoning_tokens":3326,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T21:35:32.230762+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a set of 117SnV– devices spanning different strain values, null the ambient field with a vector magnet, and measure the $|1B\\rangle$ Ramsey beat and the $f_0$ cyclicity at each setting. If some nulling field removes the beat and restores the theory-predicted cycling of $f_0$, the residual-field explanation and the exact $\\Delta\\omega_M = 0$ claim survive; if the beat persists at all field settings and changes with strain, the degeneracy is intrinsically broken by the hyperfine interaction and the zero-field insensitivity must be revised.","supporting_citations":[{"cited_title":"II A) and attempt heralded entan- glement on the broker qubit without disturbing the information stored in the memory","cited_arxiv_id":null,"evidence_quote":"The NV– brokered-entanglement demonstration whose mS = 0 no-kickback scheme the zero-field protocol seeks to replicate for group-IV centers."},{"cited_title":"Robust multi-qubit quantum network node with integrated error detection","cited_arxiv_id":"2207.13128","evidence_quote":"The 29SiV– network node with stored nuclear memory states, whose ~35 MHz optical phase kickback is the limitation the zero-field protocol removes; also the comparison baseline for memory fidelity per excitation."},{"cited_title":"Measurement based entanglement under conditions of extreme photon loss","cited_arxiv_id":"0710.4352","evidence_quote":"Supplies the measured group-IV hyperfine trend and the 117SnV– Hamiltonian whose level structure the zero-field protocol exploits."},{"cited_title":"Childress, J","cited_arxiv_id":null,"evidence_quote":"Prior demonstration of information storage in 117SnV– hyperfine levels; supplies the zero-strain optical hyperfine splitting (453 MHz) used to fix excited-state parameters."},{"cited_title":"Freely Scalable Quantum Technologies using Cells of 5-to-50 Qubits with Very Lossy and Noisy Photonic Links","cited_arxiv_id":"1406.0880","evidence_quote":"The group-theory Hamiltonian and transition matrix elements used to predict the level structure, cyclicity, and the f0 anomaly."},{"cited_title":"Microwave-based quantum control and coherence protection of tin-vacancy spin qubits in a strain-tuned diamond membrane heterostructure","cited_arxiv_id":"2307.11916","evidence_quote":"The bare-electron SnV– coherence measurement (100–400 µs) that provides the decoupling-pulse comparison baseline and fixed fit parameters (λ = 830 GHz, q = 0.171)."},{"cited_title":"A diamond nanophotonic interface with an optically accessible deterministic electronuclear spin register","cited_arxiv_id":"2305.18923","evidence_quote":"The phonon-decoherence theory for group-IV centers that predicts λM = 0 for the memory subspace, explaining the 2.5 ms coherence."}],"review_version":1}