Pith. sign in

REVIEW 6 minor 1 cited by

Quantum sensing in the presence of pulse errors and qubit leakage

T0 review · 0 major / 6 minor · reviewed 2026-08-04 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read This paper shows that a dynamical decoupling sequence's ability to preserve coherence under pulse errors does not predict its performance when actually sensing, and that leakage makes the ranking flip between protocols.

desk verdict The central claim holds up: sensing robustness and coherence-time robustness decouple, and this paper shows it with clean simulations and an experiment that actually confirms the ordering. read the letter →

arxiv 2509.09874 v1 pith:WDRCYU35 submitted 2025-09-11 quant-ph physics.atom-ph

David M. Lancaster , Muhammad Ali Shahbaz , Hamed Goli Yousefabad , Sanway Chatterjee , Eegan Ram , Jonathan D. Weinstein This is my paper
classification quant-phphysics.atom-ph
keywords quantumsensingdynamicaldecouplingpulseerrorsqubitleakageCPMGXY16MLEV32Ycoherencetime
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper asks which pulsed sensing protocol stays accurate when pulses are imperfect and the qubit can leak into extra levels. It shows that CPMG and APCP keep coherence well off-resonance but lose sensitivity and split their spectral line when actually sensing, while XY16 and MLEV32Y tolerate much larger pulse errors during sensing but suffer from leakage. In a rubidium-in-neon experiment, MLEV32Y sensed neighboring 21Ne nuclear spins with an order-of-magnitude shorter sensing time and linewidth than APCP or CPMG. The central lesson is that coherence time by itself is the wrong figure of merit for choosing a sensing protocol.

What carries the argument

The central object is the family of Carr-Purcell-derived pulse protocols — CPMG, APCP, XY16, MLEV32Y — which differ only in the phases of their π pulses. Simulations use a single spin-1/2 sensor coupled to a spin-1/2 target, with pulse errors represented by a rotation fraction and a scaled detuning, and leakage treated as a third level coupled off-resonantly with assumed equal transition strengths. The mechanism is phase interference: CPMG/APCP keep the initial state protected from rotation errors off-resonance, but once the target interaction moves the state off the protected axis, the same errors accumulate; the phase patterns of XY16/MLEV32Y distribute the error so sensing survives, at th

What would settle it

Measure the on-resonance CPMG sensing signal while sweeping rotation fraction from 0.9 to 1.1 in a system with negligible leakage; the paper's simulation predicts the signal collapses to near zero outside a tiny window while MLEV32Y stays near full contrast. If CPMG maintains high contrast across that range, the claimed pulse-error sensitivity of CPMG sensing is falsified.

Watch

Extended reading notes

Core claim

Using simulations of a spin-1/2 sensor dipole-coupled to a target spin, plus experiments on 85Rb atoms in a neon matrix at 3 K, the paper shows that the robustness of dynamical decoupling sequences to pulse errors depends sharply on whether the sequence is preserving a protected state or evolving to sense. CPMG and APCP are nearly immune to rotation and detuning errors when the interaction is off, but on resonance their signal splits and broadens even for small rotation errors; XY16 and MLEV32Y tolerate a much wider error range while sensing. Off-resonant coupling to other Zeeman levels reverses the ranking: APCP/CPMG keep their coherence under leakage, while XY16/MLEV32Y lose an order of ma

Load-bearing premise

The leakage-simulation ordering relies on the assumption that the |1⟩↔|2⟩ and |2⟩↔|3⟩ transition strengths are exactly equal; real dipoles differ, so the predicted T2 ranking could shift.

Editorial extensions

If this is right

  • Coherence-time ranking and sensing-performance ranking are nearly opposite for the four tested protocols.
  • Under simultaneous pulse errors and leakage, MLEV32Y is the best available compromise among the four protocols, giving order-of-magnitude faster sensing with narrower linewidth than APCP/CPMG in the rubidium experiment.
  • Shaped (tapered) pulses reduce leakage only modestly (≲50%), so choice of phase pattern matters more than pulse shaping in this system.
  • For systems dominated by leakage with minimal pulse errors, APCP/CPMG should outperform XY/MLEV; for systems with small leakage and significant pulse errors, the opposite holds.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The near-opposite ranking suggests a combined metric — e.g., error tolerance weighted by leakage resistance — could replace coherence time as a design target for sensing sequences.
  • The line-splitting prediction for CPMG/APCP under rotation errors offers a clean single-spin experimental test that separates pulse-error mechanisms from environmental decoherence.
  • With unequal real dipole moments, the quantitative leakage ordering likely shifts; the paper's assumption that the two transition strengths are equal is the main caveat to extrapolating the T2 ranking to other atomic species.
  • Hybrid sequences that start with APCP/CPMG for coherence preservation and switch to MLEV/XY only during the sensing window might inherit benefits of both, a direction the paper does not explore.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

0 major / 6 minor

Summary. The manuscript investigates how pulse imperfections (rotation errors and detuning) and qubit leakage affect dynamical-decoupling sequences used for quantum sensing. Using a spin-1/2 sensor coupled to a spin-1/2 target, the authors simulate CP, CPMG, APCP, XY16, and MLEV32Y with delta-function and finite-width pulses, and experimentally compare APCP, CPMG, XY16, and MLEV32Y on 85Rb atoms in a neon matrix sensing unpolarized 21Ne nuclei. The main finding is that long coherence times off-resonance do not guarantee robust sensing on-resonance: CPMG/APCP are protected from rotation errors only when the sensor remains in a special state, while XY16/MLEV are sensing-robust but suffer from leakage to other levels. MLEV32Y offers a compromise, and experimentally enables sensing roughly an order of magnitude faster and with narrower NMR linewidth than APCP/CPMG.

Significance. The paper makes a useful and non-obvious point: a sequence's coherence time in the absence of a target is not a reliable proxy for its sensing performance under realistic pulse errors. The simulations are not fitted to the data and make falsifiable predictions (e.g., Rabi-frequency dependence of T2, rotation-error tolerance, on/off-resonance ordering) that are confirmed in Table II and Figs. 9-11. The experimental demonstration of an order-of-magnitude improvement in sensing time/linewidth with MLEV32Y is direct and relevant to the broader quantum-sensing community, e.g., NV centers and donors. The paper is clearly written and the supporting figures are informative.

minor comments (6)
  1. [Section IV] The three-level leakage model assumes equal |1⟩↔|2⟩ and |2⟩↔|3⟩ transition strengths. For the experimental 85Rb F=3 system the relevant leak paths have different dipole matrix elements, and the experiment actually involves two leak paths (from |m_F=-1⟩ and |m_F=0⟩). The qualitative ordering APCP/CPMG≫MLEV32Y>XY16 is, however, corroborated by the Rabi-frequency dependence in Table II. Please add a sentence justifying the simplification or a brief discussion of robustness to unequal couplings.
  2. [Sections III-V] The simulations are described but no code or detailed parameter set is made available. Given that the quantitative curves in Figs. 3, 5, 7, and 8 are central to the argument, please add a data/code availability statement or include the relevant parameters in the text or an appendix.
  3. [Section V D / Fig. 10] The NMR linewidth comparison uses different Rabi frequencies and sequence durations for each protocol (3.3/0.87/6.6 μs and 30/10/3 ms for APCP/CPMG/MLEV32Y). This is a reasonable best-performance comparison, but it should be stated more explicitly in the main text so the reader does not mistake it for a head-to-head measurement at identical settings.
  4. [Fig. 7 caption] The sentence "to express it in terms of the Rabi frequency used, the x-axis should be scaled by a factor of 0.5" is confusing. Please define the units of δ (e.g., δ/ω, δ/(2π/τ)) unambiguously.
  5. [Table II] Typo: "correponds" should be "corresponds" in the table footnote.
  6. [Section III vs. Section V] The simulation target is a spin-1/2 particle, while the experimental target is 21Ne with I=3/2. A one-sentence justification that the qualitative conclusions are insensitive to the target spin magnitude would help bridge the simulation and experiment.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central sensing/leakage results are produced by parameter-free simulations and confirmed by independent experiment; self-citations are only apparatus context.

full rationale

The paper's derivation chain is not circular. Section III simulates the sensing sequences under rotation and detuning errors from the Schrödinger/von Neumann equation with no free parameters fitted to the target data; the predictions (e.g., CPMG/APCP are coherence-robust but sensing-fragile, XY/MLEV are sensing-robust) are concrete and falsifiable. Section IV simulates leakage in a three-level model and predicts the protocol ordering APCP/CPMG ≫ MLEV32Y > XY16 and the Rabi-frequency dependence, which is later checked against experiment in Table II and Fig. 8. The experimental T2 values are extracted from data using a fitting convention (flat distribution of decay rates, following ref. [8]), but those fits are not used as the predictions; they are measurements compared with the simulations. Self-citations (refs. [7,32,33]) describe the neon-matrix apparatus and LIF readout, which are experimental context, not proof of the central physics. No fitted parameter is renamed as a prediction, no uniqueness claim is imported, and no ansatz is smuggled in via citation. The simplified assumption of equal |1⟩↔|2⟩ and |2⟩↔|3⟩ coupling in Section IV is an idealization that could affect quantitative accuracy, but it is not a circular step: the predicted ordering is a derived consequence of the pulse-phase patterns, not an input. Overall the paper is self-contained against external benchmarks and its central claim is supported by independent simulation and experiment.

Assumptions & free parameters 2 free parameters · 8 assumptions · 0 invented entities

The central claims rest on a set of stated modeling simplifications: weak dipole-dipole interaction treated perturbatively, equal S_zS_z and S_zS_x couplings, perfect matching of interaction and sequence duration, no decoherence in simulations, and an idealized three-level leakage model with equal couplings. These are reasonable first-order choices but are not all independently verified. The experimental T2 values and NMR linewidths are extracted via fits, which are the only fitted numbers in the paper.

free parameters (2)
  • T2 decay fit maximum rate = Varied; inferred T2 from 0.18 ms to 207 ms across conditions (Table II)
    Section V B: fit LIF decay assuming a flat distribution of exponential decay rates from 0 to a maximum rate (following ref [8]); T2 is the inverse of the average rate. This is a data-analysis fit, not a physical model parameter.
  • Lorentzian fit parameters for NMR spectra = FWHM 25, 14, 1.3 kHz for APCP, CPMG, MLEV32Y (Fig 10)
    Section V D: fits to extract linewidths for comparing spectral resolution across protocols.
assumptions (8)
  • domain assumption The dipole-dipole interaction is weak relative to the sensor-bias-field interaction and is included via first-order perturbation theory.
    Section III, simulation model for sensing. If this breaks, the signal model changes.
  • domain assumption The S_zS_z and S_zS_x interaction terms are of equal magnitude.
    Section III: 'we assume the spatial separation ... such that the S_zS_z and S_zS_x interaction terms ... are of equal magnitude.' This is a special geometry, not generally true.
  • domain assumption The interaction strength and duration of the sequence are perfectly matched, so a target spin flips the sensor conditionally for perfect pulses.
    Section III. This idealization defines the sensing signal; real devices have a distribution of interaction strengths.
  • domain assumption The pulse repetition rate is on resonance with the Larmor precession of the target spin.
    Section III. Necessary for the sensing condition; experiments set 1/2τ to the 21Ne precession frequency.
  • domain assumption No decoherence in simulations.
    Sections III-IV: 'we propagate the system in time via the von Neumann equation with the assumption of no decoherence.' This isolates pulse errors and leakage.
  • domain assumption The strengths of the |1>↔|2> and |2>↔|3> transitions are equal.
    Section IV leakage model. In real Rb the dipole matrix elements differ.
  • domain assumption Thermal mixed state (infinite temperature) for the target spin.
    Section III: target spin in a mixed state with equal probability up/down. Simplifies sensing to no phase matching.
  • standard math Von Neumann equation and Rabi formula as the evolution rules.
    Sections III-IV, standard quantum mechanics.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Quantum sensing in the presence of pulse errors and qubit leakage." pith.science (2026). https://pith.science/paper/WDRCYU35

@misc{pith2026250909874,
  author       = {Pith},
  title        = {Pith review of: Quantum sensing in the presence of pulse errors and qubit leakage},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WDRCYU35}},
  note         = {Machine review of arXiv:2509.09874}
}
read the original abstract

Using both simulation and experiment, we investigate the robustness of dynamical decoupling sequences to pulse errors: rotation errors and detuning errors. Whereas prior work examined the effect of errors on coherence times, here we show that quantum sensing can be affected by pulse errors in dramatically different ways than coherence times alone. We also explore the effects of qubit leakage: off-resonant coupling to other quantum levels. We find order-of-magnitude differences between commonly-used dynamical decoupling sequences in both their sensitivity to pulse errors and leakage.

Figures

Figures reproduced from arXiv: 2509.09874 by the authors.

Figure 1
Figure 1. FIG. 1. Schematic of the pulse sequence. After an initial [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. The probability that the sensor spin ends the [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. Simulation of the outcome of the sensing se [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: FIG. 4. Probability of the sensor spin being up after [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 6
Figure 6. Figure 6: FIG. 6. Schematic of the 3-level model [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]
Figure 7
Figure 7. Figure 7: FIG. 7. The probability of each state at the end of a sensing sequence, plotted as a function of the detuning [PITH_FULL_IMAGE:figures/full_fig_p006_7.png]
Figure 8
Figure 8. Figure 8: FIG. 8. The probability of being in the correct state at [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]
Figure 9
Figure 9. Figure 9: FIG. 9. The LIF signal (as described in section V A) [PITH_FULL_IMAGE:figures/full_fig_p008_9.png]
Figure 10
Figure 10. Figure 10: FIG. 10. NMR spectra for three different pulse proto [PITH_FULL_IMAGE:figures/full_fig_p010_10.png]
Figure 11
Figure 11. Figure 11: FIG. 11. The signal (as defined in Section V A) as a func [PITH_FULL_IMAGE:figures/full_fig_p010_11.png]

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Suppressing Detuning-Induced Bias in Ramsey Magnetometry with Composite Pulses

    quant-ph 2026-07 conditional novelty 5.0 of 10

    A three-pulse composite sequence cancels first-order detuning bias in single-qubit Ramsey magnetometry, removing the error floor that limits the single-pulse protocol.

Reference graph

Works this paper leans on

37 extracted references · cited by 1 Pith paper

  1. [1]

    envelope

    However, even if the probability of a leak from a single pulse is low, it may not remain low over large numbers of pulses. For a multi-pulse sequence, significant additional complexity arises due to interference between succes- sive pulses. This interference will depend sensitively on the phase evolution of the system, which is deter- mined by the detunin...

  2. [2]

    Using tapered pulses can reduce the power at frequencies far from the car- rier frequency, which would be expected to reduce leaks to off-resonance levels

    Shaped pulses We explored the use of shaped pulses by measur- ing coherence times as a function of varying the ta- per of a Tukey window [35]. Using tapered pulses can reduce the power at frequencies far from the car- rier frequency, which would be expected to reduce leaks to off-resonance levels. Indeed, by changing from a rectangular pulse to a tapered ...

  3. [3]

    Ezzell, B

    N. Ezzell, B. Pokharel, L. Tewala, G. Quiroz, and D. A. Lidar, Physical Review Applied20, 064027 (2023)

    2023

  4. [4]

    C. L. Degen, F. Reinhard, and P. Cappellaro, Rev. Mod. Phys.89, 035002 (2017)

    2017

  5. [5]

    H. Y. Carr and E. M. Purcell, Physical review94, 630 (1954)

    1954

  6. [6]

    Kolkowitz, Q

    S. Kolkowitz, Q. P. Unterreithmeier, S. D. Ben- nett, and M. D. Lukin, Physical review letters109, 137601 (2012)

    2012

  7. [7]

    M¨ uller, X

    C. M¨ uller, X. Kong, J.-M. Cai, K. Melentijevi´ c, A. Stacey, M. Markham, D. Twitchen, J. Isoya, S. Pezzagna, J. Meijer,et al., Nature communica- tions5, 4703 (2014)

    2014

  8. [8]

    M. H. Abobeih, J. Cramer, M. A. Bakker, N. Kalb, M. Markham, D. J. Twitchen, and T. H. Taminiau, Nature communications9, 2552 (2018)

    2018

Show all 37 references
  1. [9]

    Dargyte, D

    U. Dargyte, D. M. Lancaster, and J. D. Weinstein, Phys. Rev. A104, 032611 (2021)

    2021

  2. [10]

    Upadhyay, U

    S. Upadhyay, U. Dargyte, D. Patterson, and J. D. Weinstein, Phys. Rev. Lett.125, 043601 (2020)

    2020

  3. [11]

    Cywi´ nski, R

    L. Cywi´ nski, R. M. Lutchyn, C. P. Nave, and S. D. Sarma, Physical Review B77, 174509 (2008)

    2008

  4. [12]

    C. A. Ryan, J. S. Hodges, and D. G. Cory, Physical Review Letters105, 200402 (2010)

    2010

  5. [13]

    Naydenov, F

    B. Naydenov, F. Dolde, L. T. Hall, C. Shin, H. Fed- der, L. C. Hollenberg, F. Jelezko, and J. Wrachtrup, Physical Review B—Condensed Matter and Mate- rials Physics83, 081201 (2011)

    2011

  6. [14]

    Hirose, C

    M. Hirose, C. D. Aiello, and P. Cappellaro, Physical Review A—Atomic, Molecular, and Optical Physics 86, 062320 (2012)

    2012

  7. [15]

    A. M. Souza, G. A. ´Alvarez, and D. Suter, Philo- sophical Transactions of the Royal Society A: Math- ematical, Physical and Engineering Sciences370, 4748 (2012)

    2012

  8. [16]

    J. Cai, B. Naydenov, R. Pfeiffer, L. P. McGuinness, K. D. Jahnke, F. Jelezko, M. B. Plenio, and A. Ret- zker, New Journal of Physics14, 113023 (2012)

    2012

  9. [17]

    Farfurnik, A

    D. Farfurnik, A. Jarmola, L. M. Pham, Z.-H. Wang, V. V. Dobrovitski, R. L. Walsworth, D. Budker, and N. Bar-Gill, Physical Review B92, 060301 (2015)

    2015

  10. [18]

    Farfurnik, A

    D. Farfurnik, A. Jarmola, D. Budker, and N. Bar- Gill, Journal of Optics20, 024008 (2018)

    2018

  11. [19]

    Rebentrost and F

    P. Rebentrost and F. K. Wilhelm, Phys. Rev. B79, 060507 (2009)

    2009

  12. [20]

    J. M. Gambetta, F. Motzoi, S. T. Merkel, and F. K. Wilhelm, Phys. Rev. A83, 012308 (2011)

    2011

  13. [21]

    Z. Chen, J. Kelly, C. Quintana, R. Barends, B. Campbell, Y. Chen, B. Chiaro, A. Dunsworth, A. G. Fowler, E. Lucero, E. Jeffrey, A. Megrant, J. Mutus, M. Neeley, C. Neill, P. J. J. O’Malley, P. Roushan, D. Sank, A. Vainsencher, J. Wenner, T. C. White, A. N. Korotkov, and J. M. ...

    2016

  14. [22]

    Hyypp¨ a, A

    E. Hyypp¨ a, A. Veps¨ al¨ ainen, M. Papiˇ c, C. F. Chan, S. Inel, A. Landra, W. Liu, J. Luus, F. Marxer, C. Ockeloen-Korppi, S. Orbell, B. Tarasinski, and J. Heinsoo, PRX Quantum5, 030353 (2024)

    2024

  15. [23]

    Meiboom and D

    S. Meiboom and D. Gill, Review of scientific instru- ments29, 688 (1958)

    1958

  16. [24]

    C. P. Slichter,Principles of magnetic resonance, Vol. 1 (Springer Science & Business Media, 2013)

    2013

  17. [25]

    Maudsley, Journal of Magnetic Resonance (1969) 69, 488 (1986)

    A. Maudsley, Journal of Magnetic Resonance (1969) 69, 488 (1986)

    1969

  18. [26]

    Gullion, D

    T. Gullion, D. B. Baker, and M. S. Conradi, Journal of Magnetic Resonance (1969)89, 479 (1990)

    1969

  19. [27]

    M. H. Levitt and R. Freeman, Journal of Magnetic Resonance (1969)43, 502 (1981)

    1969

  20. [28]

    M. H. Levitt, R. Freeman, and T. Frenkiel, Journal of Magnetic Resonance (1969)47, 328 (1982)

    1969

  21. [29]

    M. H. Levitt, R. Freeman, and T. Frenkiel, Journal of Magnetic Resonance (1969)50, 157 (1982)

    1969

  22. [30]

    Abobeih, J

    M. Abobeih, J. Randall, C. Bradley, H. Bartling, M. Bakker, M. Degen, M. Markham, D. Twitchen, and T. Taminiau, Nature576, 411 (2019)

    2019

  23. [31]

    Cohen-Tannoudji, B

    C. Cohen-Tannoudji, B. Diu, and F. Laloe,Quan- tum Mechanics(John Wiley & Sons, Inc., 1977)

    1977

  24. [32]

    C. E. Bradley, J. Randall, M. H. Abobeih, R. C. Berrevoets, M. J. Degen, M. A. Bakker, M. Markham, D. J. Twitchen, and T. H. Taminiau, Physical Review X9, 031045 (2019)

    2019

  25. [33]

    I. I. Rabi, N. Ramsey, and J. Schwinger, Reviews of Modern Physics26, 167 (1954). 12

    1954

  26. [34]

    D. M. Lancaster, U. Dargyte, S. Upadhyay, and J. D. Weinstein, Phys. Rev. A103, 052614 (2021)

    2021

  27. [35]

    D. M. Lancaster, U. Dargyte, and J. D. Weinstein, Phys. Rev. Res.6, L012048 (2024)

    2024

  28. [36]

    J. E. Sansonetti, W. C. Martin, and S. L. Young,Handbook of Basic Atomic Spectro- scopic Data (version 1.1.3)(NIST, 2013) http://physics.nist.gov/PhysRefData/Handbook/

    2013

  29. [37]

    M. A. Bernstein, K. F. King, and X. J. Zhou,Hand- book of MRI pulse sequences(Elsevier, 2004)

    2004

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.