{"id":"7e25d761-ad15-41d4-ab8b-1785ff0ed9f4","arxiv_id":"2608.05320","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"AB nitrogen-vacancy centers in lonsdaleite are predicted to have a transverse zero-field splitting of -358 MHz and a Hahn-echo coherence time of about 3.7 ms at zero field, roughly four times longer than computed for cubic diamond.","lead":"The authors predict that a specific arrangement of the nitrogen-vacancy defect in hexagonal diamond, called AB, keeps its quantum spin state about four times longer than the standard defect in cubic diamond when no magnetic field is applied. If experiments confirm the prediction, this form of diamond could become a platform for more sensitive quantum sensors and quantum memory without large magnetic fields.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Zero-field T2 for AB NV is an interpolation estimate, not a direct gCCE result, leaving the headline 3.73 ms and fourfold enhancement without a reported uncertainty.","rationale":"After full-text review, the single most load-bearing concern is the indirect estimation of the zero-field T2 for the AB configuration. The paper reports 3.73 ms as the peak T2, but this value is not a direct gCCE result; it is obtained by averaging linear and quadratic interpolations of T2(B) from finite fields. Since the fourfold-enhancement claim is based on this peak value, any systematic error in the interpolation propagates directly into the headline. The absence of an uncertainty estimate makes the claim impossible to evaluate quantitatively. The ZFS E value is also important—the clock-transition argument depends on E being large—but the paper's strained-cubic control provides some support for the causal role of E, and the interpolation is the point where the central number is least secure. The charge-state stability, taken from Ref. 15, is a reasonable assumption for a computational prediction. The proposed direct calculation at B=0 would settle the concern. We agree with the reader's weakest assumption and keep the conditional verdict.","tokens_in":9370,"tokens_out":9205,"duration_ms":78434,"concrete_test":"Run the gCCE Hahn-echo simulation for the AB configuration at exactly B=0 (and, for robustness, at B=0.1 G and 1 G) using the same 575-atom supercell, bath cutoffs (8 Å dipolar, 40 Å outer radius), and ZFS parameters (D=2.85 GHz, E=-358.27 MHz). Extract T2 from the coherence function and compare to the interpolated 3.73 ms. If the direct T2 is within 20% of 3.73 ms, the interpolation is validated; if it deviates by more than ~30%, the fourfold enhancement claim requires downward revision. Repeat the same protocol for the strained cubic NV (E=-155.80 MHz) to validate the control.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central quantitative claim—a zero-field Hahn-echo T2 of 3.73 ms for the AB NV center, about 4.4 times that of cubic NV—rests on an indirect estimate. The methods section states: 'At zero magnetic field (B), for the AB configuration, we estimated the coherence function using an average of linear and quadratic interpolations.' Figure 4's caption adds that the B=0 data point is estimated by averaging left- and right-biased linear and quadratic interpolations. No direct gCCE simulation at B=0 is reported, and no uncertainty or sensitivity analysis is given for this extrapolation. Because the B=0 value is the maximum of the T2(B) curve and is the basis for the fourfold-enhancement claim, an overestimate in the interpolation would directly inflate the headline. The interpolation scheme is also used for the strained cubic control, so the supporting comparison shares the same potential bias. A direct simulation at exactly B=0 (or at a very small field with controlled extrapolation) is needed to establish that the 3.73 ms value is not an artifact of the averaging procedure.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript uses first-principles calculations (DFT, QDET, PyZFS, and the gCCE method) to characterize negatively charged nitrogen-vacancy (NV) centers in lonsdaleite. It considers two configurations, AA and AB, and finds that the AB configuration has a transverse zero-field splitting E = -358.27 MHz, leading to a clock transition at zero magnetic field and a predicted Hahn-echo T2 of about 3.73 ms, roughly four times the computed cubic-diamond NV value of 0.85 ms. The paper also reports vertical excitation energies, ZPL energies, Huang-Rhys factors, and Debye-Waller factors as spectral fingerprints. The central claim is that symmetry-broken NV centers in lonsdaleite are promising for zero-field quantum sensing.","tokens_in":9589,"tokens_out":6624,"duration_ms":58631,"significance":"If the predicted coherence enhancement is correct, the AB NV center in lonsdaleite would be a genuinely useful system for zero-field quantum sensing, with the mechanism clearly explained by Eq. (4). The work is methodologically strong and transparent: the ZFS tensor is computed from the ground-state wavefunction rather than fitted, the gCCE simulations are converged with respect to cutoffs and bath size, and the data and codes are stated to be openly available. The comparison with strained cubic diamond and with earlier dislocation work provides plausible physical grounding. However, the headline zero-field T2 is obtained by an interpolation estimate rather than by a direct simulation, and the magnitude of E is computed with a single exchange-correlation functional. These issues make the quantitative fourfold-enhancement claim conditional rather than fully established.","major_comments":[{"comment":"The zero-field T2 for the AB configuration is not computed directly. The text states that at zero magnetic field the coherence function was \"estimated using an average of linear and quadratic interpolations,\" and the Fig. 4 caption confirms that the B=0 point is obtained by averaging left- and right-biased linear and quadratic interpolations. Because this point is the maximum of the T2(B) curve and is the basis for the fourfold-enhancement claim, the headline result depends on an ad hoc interpolation formula with no reported uncertainty or sensitivity analysis. I request a direct gCCE calculation at B=0, or at a few very small fields with a controlled extrapolation, together with an estimate of the interpolation error. The same caveat applies to the strained-cubic control, whose B=0 T2 of 2.71 ms is produced by the same averaging procedure.","section":"Coherence-times section and Fig. 4"},{"comment":"The clock-transition mechanism makes the zero-field T2 controlled by the magnitude of E, but E is obtained from PBE ground-state wavefunctions with no functional sensitivity analysis. The manuscript's own strained-cubic comparison illustrates the scale of possible uncertainty: PyZFS gives E = -155.80 MHz, while the literature strain-spin relation gives E = -130.68 MHz, a difference of about 25 MHz or roughly 20%. A similar relative uncertainty in the AB value of -358 MHz could change the predicted T2 substantially. Please compute E with a hybrid functional such as DDH, or at least propagate a plausible range of E through the coherence calculation, before the quantitative enhancement is taken as established.","section":"Table I and Eq. (4)"},{"comment":"All predictions are for the negatively charged NV- center, and the paper's experimental relevance claim depends on the stability of this charge state in bulk lonsdaleite. The manuscript relies on formation-energy calculations from Ref. 15, which were performed for nanoscale lonsdaleite with specific surface terminations and a different functional, rather than computing the formation energies in the present bulk supercells. This is an external assumption that does not affect the internal AA/AB/cubic comparison but is load-bearing for the statement that AB centers in lonsdaleite are experimentally accessible. Please either compute the charge-state stability here or explicitly state that the -1 charge state is assumed.","section":"Charge-state assumption"}],"minor_comments":[{"comment":"The space-group symbol \"P63/mmc\" should be typeset as \"P6_3/mmc\", with the sixfold screw axis indicated by a subscript.","section":"Fig. 1 caption"},{"comment":"The statement says the data and codes are openly available on Qresp, but no link, DOI, or accession identifier is provided; please add a resolvable reference for reproducibility.","section":"Data Availability Statement"},{"comment":"The table mixes computed values, literature values, and experimental values without explicitly separating them; a column or footnote indicating which entries are from the present calculations would improve clarity.","section":"Table I"},{"comment":"The sentence explaining the difference from Ref. 15 for the AA ZFS is helpful, but a brief note on why the AA D=4.56 GHz value differs so strongly from both cubic and AB values would help readers assess the discrepancy.","section":"Comparison with Ref. 15"}],"recommendation":"major_revision","confidential_remarks":"The paper's central derivation is sound, and I see no circularity problem: E is computed from the wavefunction via PyZFS rather than fitted, and the gCCE simulation then follows mechanistically. The main issue is that both the headline zero-field T2 and the supporting strained-cubic value rely on an interpolation estimate rather than a direct calculation, and the magnitude of E has no functional-sensitivity assessment. These are fixable within the scope of a revision by adding direct simulations or controlled extrapolations with uncertainties. The heavy self-citation to the group's own methods is appropriate given that the methods were developed in these papers."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThis paper is worth taking seriously. The central prediction—that the AB NV in lonsdaleite has a transverse zero-field splitting of about -358 MHz and therefore a zero-field Hahn-echo T2 near 3.7 ms, roughly four times the cubic value—is genuinely new. Nobody has computed E or the coherence time for this configuration before, and the mechanism is physically sound. Equation 4 is the right clock-transition argument: a larger E makes the transition frequency second-order insensitive to magnetic noise.\n\nThe computational work is solid. The ZFS parameters are computed from the ground-state wavefunction, not fitted. The QDET excited-state results, the PL spectra, and the Debye-Waller factors give spectral fingerprints that should help experimental identification. The strained-cubic comparison against Udvarhelyi et al. is a reasonable sanity check and puts the lonsdaleite E in context.\n\nThe soft spot is exactly where the stress test points. The headline zero-field T2 for AB is not a direct gCCE result; it is the average of linear and quadratic interpolations of the B-field dependence. No uncertainty is attached, and the same interpolation is used for the strained cubic control. Because Eq. 4 says the second-order sensitivity goes as 1/E, a modest error in E directly inflates or deflates the claimed enhancement. To be concrete: if E were actually -250 MHz instead of -358 MHz, the zero-field protection would be weaker and the fourfold claim shrinks. So the number 3.73 ms should be treated as an estimate with unknown error bars, not a converged prediction.\n\nA smaller concern: the ZFS is computed at the PBE level. The authors defend this by saying they are comparing trends, which is fair, but the absolute E matters here. A hybrid-functional benchmark for E would be cheap and would sharpen the headline. The -1 charge state stability is borrowed from Ref. 15; that is acceptable but is an inherited assumption.\n\nNone of this undermines the qualitative conclusion. A symmetry-broken NV with a finite E in lonsdaleite is a plausible route to zero-field qubits, and the paper says so without overclaiming—except for the unqualified 3.73 ms in the abstract and Table I. The authors are transparent about the interpolation in the text, which is to their credit.\n\nThis paper deserves a serious referee. It is a computational prediction with real experimental implications, the methods are state of the art, and the flaw is fixable. I would send it out with a request for a direct B = 0 calculation or a sensitivity analysis, and a hybrid-functional check of E. After that it would be a well-supported prediction.","headline":"A genuine new prediction—AB NV in lonsdaleite has a large transverse ZFS and a plausible zero-field coherence advantage—but the headline T2 is an interpolation estimate, not a direct simulation.","tokens_in":10143,"tokens_out":2416,"would_cite":true,"duration_ms":21910,"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":"The AB nitrogen-vacancy center in lonsdaleite reaches a zero-field coherence time of about 3.73 ms, four times the cubic-diamond value, via a clock transition from a finite transverse zero-field splitting.","keywords":["nitrogen-vacancy center","lonsdaleite","hexagonal diamond","zero-field splitting","clock transition","spin coherence","quantum sensing","first-principles calculation"],"falsifier":"A direct second-order gCCE simulation at exactly B = 0 for the AB configuration would settle the main claim: if the true zero-field T2 lies substantially below 3.73 ms, the fourfold enhancement is an artifact of the interpolation. On the experimental side, a zero-field ODMR measurement looking for the E ≈ 358 MHz splitting and a Hahn-echo decay at B = 0 in lonsdaleite NV ensembles would test the same prediction.","tokens_in":9159,"feed_emoji":"💎","tokens_out":9082,"duration_ms":70741,"temperature":0.7,"pith_summary":"This paper uses first-principles calculations to argue that negatively charged nitrogen-vacancy (NV) centers in lonsdaleite, the hexagonal form of diamond, can preserve quantum information far longer than their cubic-diamond counterparts. The central player is the \"AB\" defect configuration, whose reduced symmetry produces a finite transverse zero-field splitting E = -358.27 MHz. That splitting creates a clock transition at zero magnetic field, making the spin transition frequency largely immune to magnetic noise; the authors estimate a Hahn-echo coherence time T2 of about 3.73 ms, roughly 4.4 times the computed cubic-diamond value of 0.85 ms. The paper also provides photoluminescence and zero-phonon-line fingerprints to distinguish the AB configuration from the AA configuration, which behaves essentially like a cubic NV. If the estimate holds, symmetry-broken NVs in lonsdaleite become a concrete route to zero-field quantum sensing and spin-based quantum information.","feed_headline":"Hexagonal-diamond NV centers reach 3.73 ms zero-field coherence","feed_subtitle":"A symmetry-broken nitrogen-vacancy configuration gains a clock transition, beating cubic diamond's 0.85 ms by fourfold.","key_machinery":"The load-bearing mechanism is the transverse zero-field splitting parameter E in the ground-state spin Hamiltonian H = D $S_z^{2}$ + E($S_x^{2}$ - $S_y^{2}$) + gamma_e B_z S_z. For E ≠ 0, the |±1> states hybridize into |±> = (|-1> ± |+1>)/$\\sqrt$(2), and the transition frequencies to |0> become omega_{0,±} = D ± $\\sqrt$((gamma_e B_z)^2 + $E^{2}$). The first derivative with respect to B_z vanishes at B_z = 0, making the transition a first-order clock transition, while the second-order sensitivity is ± $gamma_e^{2}$ / E, which shrinks as E grows. In the AB lonsdaleite configuration, the computed E = -358.27 MHz is roughly two hundred times the cubic value, and this is what suppresses magnetic-noise dephasing in the gCCE calculations. The supporting computations use density functional theory for geometries, quantum defect embedding theory (QDET) for many-body excited states, a generating-function approach for vibronic spectra, and second-order gCCE for coherence times.","core_discovery":"The central claim is that the AB configuration of the NV- center in lonsdaleite possesses a transverse zero-field splitting factor E = -358.27 MHz (with D = 2.85 GHz), and that this finite E converts the |±1> manifold into symmetric and antisymmetric superpositions whose transition frequencies have zero first-order magnetic-field dependence at B = 0. Because the residual second-order sensitivity scales as $gamma_e^{2}$ / E, a larger E suppresses dephasing from magnetic noise. From generalized cluster-correlation expansion (gCCE) simulations, the authors obtain a Hahn-echo coherence time of about 3.73 ms at zero magnetic field for the AB configuration, about 4.4 times the computed cubic-NV value of 0.85 ms; the same simulations give 0.9 ms for the AA configuration, which preserves C3v symmetry and closely resembles cubic NV centers. The authors further calculate many-body excitation energies, zero-phonon-line energies, Huang-Rhys and Debye-Waller factors, and photoluminescence spectra for both configurations, offering spectral fingerprints for experimental identification. The mechanism itself has precedent in symmetry-broken NV centers near dislocations in cubic diamond, but here it arises intrinsically from the hexagonal host lattice.","pith_inferences":["Editorial inference: if the zero-field interpolation is later confirmed by a direct B = 0 simulation, the same symmetry-breaking mechanism should apply to any NV variant with E of order 100 MHz; other polytypes, stacking faults, or engineered strains may yield even larger zero-field T2 enhancements.","Editorial inference: because the second-order sensitivity scales as gamma_e^2 / E, one could deliberately tune E by strain to trade coherence against residual magnetic-field curvature, and the clock-transition physics may also suppress noise from other fields that couple through the same magnetic channel.","Editorial inference: the coexistence of AA and AB centers in one sample could support a two-species quantum memory or built-in calibration scheme, in which the AA center serves as a magnetically sensitive reference and the AB center serves as a magnetically insensitive clock."],"forward_implications":["At zero magnetic field, the AB NV in lonsdaleite should preserve a Hahn-echo signal for about 3.73 ms, roughly 4.4 times the computed 0.85 ms of cubic-diamond NV centers.","The finite E ≈ -358 MHz makes the |0> ↔ |±> transition a first-order clock transition, so sensor operation at zero field should be far less sensitive to ambient magnetic noise than conventional cubic NVs.","The AA configuration, with E ≈ 0, behaves essentially like a cubic NV (T2 ≈ 0.9 ms), meaning a single lonsdaleite sample can host both standard and clock-transition-protected NV species.","The computed zero-phonon-line energies (1.48 eV for AB, 1.73 eV for AA) and Debye-Waller factors (3.65% and 4.42%) provide experimentally observable fingerprints to identify which configuration is present.","A 3% tensile strain along an N-C bond in cubic diamond produces E ≈ -156 MHz and T2 ≈ 2.71 ms, so strain engineering can partially reproduce the enhancement, though not to the same level as the AB lonsdaleite configuration."],"supporting_citations":[{"why":"Supplies the formation-energy calculations establishing that AA and AB NVs in lonsdaleite can be stabilized in the -1 charge state, and gives prior DFT-based ZFS estimates the authors compare against.","marker":"15"},{"why":"Introduces the quantum defect embedding theory (QDET) approach and provides the reference cubic-NV many-body states used for comparison.","marker":"16"},{"why":"Provides the quantum defect embedding formulation used to compute many-body excitation energies of both lonsdaleite configurations.","marker":"17"},{"why":"Defines the generalized cluster-correlation expansion method used to simulate spin coherence times.","marker":"19"},{"why":"Shows that symmetry-broken NVs near dislocations in cubic diamond acquire finite E and longer low-field T2, the trend the lonsdaleite result extends.","marker":"36"},{"why":"Provides the gCCE implementation used to obtain the reported T2 values.","marker":"50"},{"why":"Gives the strain-spin relationship used to benchmark the strained cubic NV case with E ≈ -156 MHz.","marker":"51"}],"fun_headline_variants":["Lonsdaleite NV centers reach 3.73 ms coherence, 4x cubic","Hexagonal diamond NV qubits: zero-field T2 hits 3.73 ms","AB-config NV centers in lonsdaleite outdo cubic diamond","Symmetry-broken NV centers in hexagonal diamond extend coherence","Zero-field NV centers in lonsdaleite show 4x longer T2"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The zero-field coherence time of 3.73 ms is not obtained from a direct calculation at B = 0; it comes from averaging a linear and a quadratic interpolation between neighboring field values, so the fourfold-enhancement claim stands or falls on that hand-chosen interpolation, and on the adopted assumption that the -1 charge state is stable.","fun_headline_variants_meta":{"raw":{"variants":["Lonsdaleite NV centers reach 3.73 ms coherence, 4x cubic","Hexagonal diamond NV qubits: zero-field T2 hits 3.73 ms","AB-config NV centers in lonsdaleite outdo cubic diamond","Symmetry-broken NV centers in hexagonal diamond extend coherence","Zero-field NV centers in lonsdaleite show 4x longer T2"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000932,"raw_usage":{"total_tokens":4002,"prompt_tokens":971,"completion_tokens":3031,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":587,"completion_tokens_details":{"reasoning_tokens":2929}},"tokens_in":587,"tokens_out":3031,"duration_ms":17617,"temperature":1.0,"reasoning_tokens":2929,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-08T15:23:49.000150+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct second-order gCCE simulation at exactly B = 0 for the AB configuration would settle the main claim: if the true zero-field T2 lies substantially below 3.73 ms, the fourfold enhancement is an artifact of the interpolation. On the experimental side, a zero-field ODMR measurement looking for the E ≈ 358 MHz splitting and a Hahn-echo decay at B = 0 in lonsdaleite NV ensembles would test the same prediction.","supporting_citations":[{"cited_title":"Extracting phonon coupling parameters from multi color center photoluminescence","cited_arxiv_id":null,"evidence_quote":"Supplies the formation-energy calculations establishing that AA and AB NVs in lonsdaleite can be stabilized in the -1 charge state, and gives prior DFT-based ZFS estimates the authors compare against."},{"cited_title":"Luminescence lineshapes of nitrogen vacancy center in lonsdaleite and dual structure of diamond/lonsdaleite: a DFT study","cited_arxiv_id":null,"evidence_quote":"Introduces the quantum defect embedding theory (QDET) approach and provides the reference cubic-NV many-body states used for comparison."},{"cited_title":"Nitrogen-vacancy centres in lonsdaleite: a novel nanoscale sensor?","cited_arxiv_id":null,"evidence_quote":"Provides the quantum defect embedding formulation used to compute many-body excitation energies of both lonsdaleite configurations."},{"cited_title":"Generalized gradient approximation made simple","cited_arxiv_id":null,"evidence_quote":"Defines the generalized cluster-correlation expansion method used to simulate spin coherence times."},{"cited_title":"Spin-strain interaction in nitrogen-vacancy centers in diamond","cited_arxiv_id":null,"evidence_quote":"Shows that symmetry-broken NVs near dislocations in cubic diamond acquire finite E and longer low-field T2, the trend the lonsdaleite result extends."},{"cited_title":"Spin dynamics in the optical cycle of single nitrogen-vacancy centres in diamond","cited_arxiv_id":null,"evidence_quote":"Provides the gCCE implementation used to obtain the reported T2 values."},{"cited_title":"Excited state properties of point defects in semiconductors and insulators investigated with time-dependent density functional theory","cited_arxiv_id":null,"evidence_quote":"Gives the strain-spin relationship used to benchmark the strained cubic NV case with E ≈ -156 MHz."}],"review_version":1}