Pith. sign in

REVIEW 2 major objections 3 minor 35 references

Magnetic resonance force microscopy with a one-dimensional resolution of 0.9 nanometers

T0 review · 2 major / 3 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read The paper reports magnetic resonance force microscopy scans with a one-dimensional resolution of 0.9 nm, the first sub-nanometer MRI measurement.

desk verdict Impressive sub-nanometer MRFM scans with a new HSn pulse protocol, but the 0.9 nm headline is edge-localization precision, not a demonstrated two-point spatial resolution. read the letter →

arxiv 1908.04180 v1 pith:LYZFTVOM submitted 2019-08-12 physics.app-ph cond-mat.mes-hall

U. Grob , M. D. Krass , M. Héritier , R. Pachlatko , J. Rhensius , J. Košata , B. A. Moores , H. Takahashi
show 2 more authors
A. Eichler C. L. Degen
This is my paper · ORCID
classification physics.app-phcond-mat.mes-hall
keywords magneticresonanceforcemicroscopynanoscaleimagingspatialresolutionnuclearscanningprobehyperbolicsecantpulsesadiabaticspininversionproton
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 tries to show that magnetic resonance force microscopy (MRFM) can image with sub-nanometer spatial resolution, reporting a one-dimensional resolution of 0.9 ± 0.2 nm and a localization precision of 0.6 ± 0.1 nm. A sympathetic reader would care because this is the first demonstrated sub-nanometer MRI measurement, moving the technique toward three-dimensional imaging of protein and macromolecular structures. The advance comes from a sharper spin-excitation protocol, a magnetic gradient above $10^{6}$ T/m, a well-defined nanorod sample geometry, and instrument drift below 2 nm over 24 hours. The authors argue that at maximum modeled gradient the arrangement supports resolutions down to 0.3 nm given sufficient signal-to-noise ratio.

What carries the argument

The central mechanism is the hyperbolic secant (HSn) pulse protocol, an amplitude- and frequency-modulated radio-frequency excitation that adiabatically inverts nuclear spins in a narrow, well-defined detuning window. The slice function is ξ(Δf) = a₁(Δf) F(Δf) ≈ (4/π) F(Δf), where F is the spin-inversion fidelity over hundreds of cantilever cycles. This sharp slice, combined with a tangential nanorod geometry and a measured lateral gradient of 0.56 × $10^{6}$ T/m, produces a signal step whose maximum slope divided by noise (Eq. 8) gives the sub-nanometer resolution value.

What would settle it

Scan a sample with two parallel 1H-rich edges separated by a known distance around 1 nm: if the measured signal shows one unresolved combined step instead of two distinguishable onsets, then the slope-based 0.9 nm figure would not correspond to two-point imaging resolution. An additional check would be to vary the averaging time and see whether σx,slope improves as expected from noise averaging and whether the fitted onset positions remain stable.

Watch

Extended reading notes

Core claim

The paper's central claim is that MRFM, using optimized hyperbolic secant (HSn) adiabatic pulses to create a sharply defined resonant slice, achieves a one-dimensional spatial resolution of 0.9 ± 0.2 nm and a localization precision of 0.6 ± 0.1 nm, as measured from the slope-to-noise ratio of the signal onset (Eq. 8). This is presented as the first sub-nanometer MRI measurement. The authors find that the imaging slice is sharply bounded—about 30 kHz between full spin inversion and no signal—and that all technical sources of blur are reduced below 1 nm, so resolution is limited only by sensor noise. A model of the tip field indicates that the same arrangement could reach roughly 0.3 nm slice width at maximum gradient.

Load-bearing premise

The 0.9 nm resolution is computed from the steepest slope of the signal step divided by the measurement noise, which assumes that this edge-localization uncertainty equals the smallest separation between two resolvable image features; the paper does not demonstrate two-point resolution.

Editorial extensions

If this is right

  • If the central claim holds, MRFM can localize nuclear spin density features with sub-nanometer precision in one dimension, a prerequisite for meaningful structural imaging of macromolecules.
  • At the maximum modeled gradient of about 6 × 10^6 T/m, the same instrumentation is expected to support slice widths around 0.3 nm, provided the signal-to-noise ratio can be improved.
  • With a more sensitive force transducer, the authors expect three-dimensional images with 1 nm voxel size, corresponding to roughly 100 hydrogen atoms, to become possible.
  • The sharp HSn slice protocol removes the previously limiting slice-edge blur, and the demonstrated drift below 2 nm over 31 hours makes long, undistorted scans feasible.
  • Because MRFM can handle objects larger than 20 nm, sub-nanometer resolution would extend MRI-style imaging to structures that are difficult for other nanoscale MRI techniques.

Reading between the lines

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

  • Beyond the paper's claim, the 0.9 nm figure is an edge-localization uncertainty from a single sharp signal step, not a demonstrated two-point resolution; a two-point phantom measurement would be needed to equate it with the smallest resolvable separation between two features.
  • Since the detected signal arises from statistical rather than thermal spin polarization, the effective resolution likely improves with averaging time and signal-to-noise; this could be tested systematically by measuring σx,slope at different integration times.
  • The sharp edge produced by HSn pulses may be transferable to other MRFM and MRI configurations, potentially improving slice definition wherever adiabatic inversion is used.
  • If the tangential-slice geometry is pushed to maximum gradient, a 0.3 nm resolution would bring MRFM close to imaging secondary protein structure directly, though three-dimensional scanning remains an unsolved step.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 3 minor

Summary. The manuscript reports MRFM lateral scans of a hydrogen-containing adsorbate layer on a silicon nanorod, claiming a one-dimensional spatial resolution of 0.9 ± 0.2 nm and a localization precision of 0.6 ± 0.1 nm. The authors introduce hyperbolic secant (HS^n) adiabatic pulses for sharp slice definition, characterize the slice profile experimentally and by simulation, and demonstrate long-term instrument stability. The resolution metric is obtained from the maximum slope-to-noise ratio of the signal onset (Eq. 8). The paper also derives a slice width of 0.7 nm from the tip model and predicts a possible 0.3 nm resolution with improved SNR. The authors claim the first sub-nanometer MRI and a milestone toward 3D macromolecular imaging.

Significance. The experimental work is of high quality: the HSn pulse protocol is a clear improvement over conventional trapezoidal modulation, the drift of 1.8 nm over 31 h is excellent, and the gradient calibration appears careful. The central quantitative claim, however, rests on the interpretation of the slope-to-noise ratio as 'resolution.' If the claim is reframed as sub-nanometer edge-localization precision, the result is still remarkable and of interest to the MRFM community. The paper also provides a reasonable model-based extrapolation to 0.3 nm, which is clearly labeled as an expectation. No computational or derivational circularity was apparent.

major comments (2)
  1. [§4 (Results and Discussion), Eq. (8)] The quantity σ_x,slope defined in Eq. (8) is the uncertainty in estimating the position of a single step edge from its slope and the noise level; it is a localization precision, not a two-point resolution. The paper's own fits yield an edge width w ≈ 10 nm (Eq. (7)), and the scans in Fig. 4 are single-step onsets. Two features separated by 0.9 nm would produce a merged step whose position could still be estimated to sub-nanometer precision, so the scans do not demonstrate a resolution of 0.9 nm. The abstract and conclusions should either use the term 'edge localization precision' or add a two-point experimental demonstration (e.g., two adsorbate edges at a known separation) to support the resolution claim.
  2. [Conclusion] The statement 'minimum measured slice width of 0.7 nm' in the conclusion is not directly measured; the value is computed from the simulated HSn slice function and the tip-model gradient G_x ≈ 2.3 × 10^6 T/m. Since the slice width is a key figure for the sub-nanometer claim, the text should explicitly label it as a model-derived estimate, or better, provide a direct measurement by scanning the slice across a sharp feature.
minor comments (3)
  1. [Abstract] The word 'suffcient' in the abstract is a typo for 'sufficient'.
  2. [Introduction] The citation 'gradient generation,12,12–16' should read '12–16'; the duplicate '12' appears to be a numbering error.
  3. [§4 (Results and Discussion)] The inset of Fig. 4b shows a confidence interval for one scan; it would be helpful to state how the error bars of σ_x,slope (0.2 nm) were computed across the 11 scans (e.g., standard deviation or standard error).

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the 0.9 nm resolution claim is an experimental slope-to-noise measurement, not a derived prediction.

full rationale

The paper's central claim is an experimental measurement of the signal onset in lateral MRFM scans, analyzed through Eq. 8, which compares the maximum signal slope to the noise variance. This is a standard edge-localization statistic computed from measured slopes and noise; it is not a fitted parameter disguised as a prediction. The slice function is obtained from independent simulations of HSn adiabatic pulses, and the tip field/gradient model is calibrated using AFM topography and separate MRFM calibration scans, not from the same onset positions used to claim resolution. The expected 0.3 nm resolution is an extrapolation from the modeled slice width and maximum gradient, clearly labeled as an expectation rather than a measured result. The only substantive concern is semantic: single-edge localization precision is equated with 'spatial resolution' without a two-point resolvability demonstration. That is a validity/interpretation issue, not circularity, because no derivation step reduces to its own inputs. Therefore the paper is self-contained against its own experimental data and merits a circularity score of 0.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central measurement (0.9 nm resolution) is derived directly from the measured signal slope and noise, so it rests on few free parameters. The main auxiliary claims (slice width 0.7 nm, projected 0.3 nm resolution) rely on a calibrated tip model whose parameters are fitted to calibration scans. The paper's assumptions about the sample and spin physics are standard for MRFM and are cited to prior literature.

free parameters (2)
  • Lateral magnetic gradient Gx at scan position = 0.56 x 10^6 T/m
    Extracted from the tip field model (calibrated by AFM topography and MRFM calibration scans, see SI). Used to map slice frequency offsets to lateral positions in the scan analysis and to estimate slice widths.
  • Maximum tip gradient |G| = ~6 x 10^6 T/m at z=10 nm
    From the numerical tip model. Used for the extrapolated 0.3 nm slice width projection in the results section.
assumptions (4)
  • domain assumption The magnetic tip field map is accurately modeled by a combination of AFM topography and MRFM calibration scans.
    Used in the Experimental Setup and Fig. 1d-e to convert Larmor frequency offsets to spatial positions and to estimate gradients and slice widths. The calibration is described as being in the SI.
  • domain assumption The 1H NMR signal originates solely from a uniform approximately 1 nm thick adsorbate layer on the silicon nanorod.
    Stated in Results and Discussion. The analysis of the signal onset assumes this thin layer enters the resonance slice over a few nanometers, producing the sharp edge used for the resolution estimate.
  • domain assumption The detected signal is proportional to the variance of the spin force because only statistical spin polarization is measured.
    Invoked after Eq. 1, citing Refs. 24 and 25. This justifies the use of F^2_spin and the noise statistics in the resolution estimate.
  • standard math HSn pulse inversion fidelity can be accurately simulated using a piece-wise constant Hamiltonian density-matrix evolution.
    Used to compute the slice function and spin reversal fidelity (Eqs. 2-6). The simulation is standard quantum spin dynamics and is consistent with the measured dependence on modulation depth in Fig. 2e.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Magnetic resonance force microscopy with a one-dimensional resolution of 0.9 nanometers." pith.science (2026). https://pith.science/paper/LYZFTVOM

@misc{pith2026190804180,
  author       = {Pith},
  title        = {Pith review of: Magnetic resonance force microscopy with a one-dimensional resolution of 0.9 nanometers},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LYZFTVOM}},
  note         = {Machine review of arXiv:1908.04180}
}
read the original abstract

Magnetic resonance force microscopy (MRFM) is a scanning probe technique capable of detecting MRI signals from nanoscale sample volumes, providing a paradigm-changing potential for structural biology and medical research. Thus far, however, experiments have not reached suffcient spatial resolution for retrieving meaningful structural information from samples. In this work, we report MRFM imaging scans demonstrating a resolution of 0.9 nm and a localization precision of 0.6 nm in one dimension. Our progress is enabled by an improved spin excitation protocol furnishing us with sharp spatial control on the MRFM imaging slice, combined with overall advances in instrument stability. From a modeling of the slice function, we expect that our arrangement supports spatial resolutions down to 0.3 nm given suffcient signal-to-noise ratio. Our experiment demonstrates the feasibility of sub-nanometer MRI and realizes an important milestone towards the three-dimensional imaging of macromolecular structures.

Figures

Figures reproduced from arXiv: 1908.04180 by the authors.

Figure 1
Figure 1. (a) Configuration of the MRFM experiment. (b) Sketch of nanorod apex, [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Signal encoding and generation by hyperbolic secant ( [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Series of 1H NMR spectra as a function of vertical approach distance d = 10 − 150 nm. The schematics show the configuration of nanorod and nanomagnet; the adsorbate layer is color-coded with the gradient Gx (scale of Fig. 1e). (a) Spectra taken with the tip positioned over the edge of the nanomagnet (x = −200 nm, where x is the center-to￾center distance between cantilever and nanomagnet). The white star marks the po… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Lateral imaging scans showing 0.9 nm spatial resolution. (a) Two consecutive line scans taken within 31 hours. Averaging time is 4 min per point. (b) Series of x scans for excitation frequencies frf = 252.3−252.8 MHz. The sharp signal rise around x = 20−50 nm indicates…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 35 canonical work pages

  1. [1]

    Sidles, J. A. Applied Physics Letters 1991, 58, 2854

    work page 1991

  2. [2]

    A.; Garbini, J

    Sidles, J. A.; Garbini, J. L.; Bruland, K. J.; Rugar, D.; Z\"uger, O.; Hoen, S.; Yannoni, C. S. Reviews of Modern Physics 1995, 67, 249

    work page 1995

  3. [3]

    L.; Poggio, M.; Mamin, H

    Degen, C. L.; Poggio, M.; Mamin, H. J.; Rettner, C. T.; Rugar, D. Proceedings of the National Academy of Sciences of the United States of America 2009, 106, 1313

    work page 2009

  4. [4]

    Poggio, M.; Degen, C. L. Nanotechnology 2010, 21, 342001

    work page 2010

  5. [5]

    M.; Naibert, T

    Nichol, J. M.; Naibert, T. R.; Hemesath, E. R.; Lauhon, L. J.; Budakian, R. Physical Review X 2013, 3, 031016

    work page 2013

  6. [6]

    J.; Rugar, D

    Mamin, H. J.; Rugar, D. Applied Physics Letters 2001, 79, 3358

    work page 2001

  7. [7]

    J.; Liu, D

    Moser, J.; Guttinger, J.; Eichler, A.; Esplandiu, M. J.; Liu, D. E.; Dykman, M. I.; Bachtold, A. Nature Nanotechnology 2013, 8, 493--496

    work page 2013

  8. [8]

    M.; Moores, B

    Tao, Y.; Boss, J. M.; Moores, B. A.; Degen, C. L. Nature Communications 2014, 5, 3638

    work page 2014

Show all 35 references
  1. [9]

    R.; Cadeddu, D.; Vasyukov, D.; Tutuncuoglu, G.; i Morral, A

    Rossi, N.; Braakman, F. R.; Cadeddu, D.; Vasyukov, D.; Tutuncuoglu, G.; i Morral, A. F.; Poggio, M. Nature Nanotechnology 2017, 12, 150

    2017

  2. [10]

    M.; Pigeau, B.; Besga, B.; Vincent, P.; Poncharal, P.; Arcizet, O

    de Lepinay, L. M.; Pigeau, B.; Besga, B.; Vincent, P.; Poncharal, P.; Arcizet, O. Nature Nanotechnology 2017, 12, 156

    2017

  3. [11]

    D.; Tao, Y.; Degen, C

    Heritier, M.; Eichler, A.; Pan, Y.; Grob, U.; Shorubalko, I.; Krass, M. D.; Tao, Y.; Degen, C. L. Nano Letters 2018, 18, 1814--1818

    2018

  4. [12]

    J.; Rettner, C

    Mamin, H. J.; Rettner, C. T.; Sherwood, M. H.; Gao, L.; Rugar, D. Applied Physics Letters 2012, 100, 013102

    2012

  5. [13]

    M.; Hemesath, E

    Nichol, J. M.; Hemesath, E. R.; Lauhon, L. J.; Budakian, R. Physical Review B 2012, 85, 054414

    2012

  6. [14]

    G.; Mamin, H

    Longenecker, J. G.; Mamin, H. J.; Senko, A. W.; Chen, L.; Rettner, C. T.; Rugar, D.; Marohn, J. A. ACS Nano 2012, 6, 9637--9645

    2012

  7. [15]

    Tao, Y.; Eichler, A.; Holzherr, T.; Degen, C. L. Nature Communications 2016, 7, 12714

    2016

  8. [16]

    Physical Review Applied 2017, 7, 024019

    Wagenaar, J.; den Haan, A.; Donkersloot, R.; Marsman, F.; de Wit, M.; Bossoni, L.; Oosterkamp, T. Physical Review Applied 2017, 7, 024019

    2017

  9. [17]

    J.; Oosterkamp, T

    Mamin, H. J.; Oosterkamp, T. H.; Poggio, M.; Degen, C. L.; Rettner, C. T.; Rugar, D. Nano Letters 2009, 9, 3020

    2009

  10. [18]

    Q.; Jeon, N.; Lauhon, L

    Rose, W.; Haas, H.; Chen, A. Q.; Jeon, N.; Lauhon, L. J.; Cory, D. G.; Budakian, R. Physical Review X 2018, 8, 011030

    2018

  11. [19]

    J.; Kim, M.; Sherwood, M

    Mamin, H. J.; Kim, M.; Sherwood, M. H.; Rettner, C. T.; Ohno, K.; Awschalom, D. D.; Rugar, D. Science 2013, 339, 557--560

    2013

  12. [20]

    A.; Reinhard, F.; Wrachtrup, J

    Staudacher, T.; Shi, F.; Pezzagna, S.; Meijer, J.; Du, J.; Meriles, C. A.; Reinhard, F.; Wrachtrup, J. Science 2013, 339, 561--563

    2013

  13. [21]

    A.; Eichler, A.; Tao, Y.; Takahashi, H.; Navaretti, P.; Degen, C

    Moores, B. A.; Eichler, A.; Tao, Y.; Takahashi, H.; Navaretti, P.; Degen, C. L. Applied Physics Letters 2015, 106, 213101

    2015

  14. [22]

    L.; Mamin, H

    Poggio, M.; Degen, C. L.; Mamin, H. J.; Rugar, D. Physical Review Letters 2007, 99, 017201

    2007

  15. [23]

    M.; Garbini, J

    Chao, S.; Dougherty, W. M.; Garbini, J. L.; Sidles, J. A. Review of Scientific Instruments 2004, 75, 1175

    2004

  16. [24]

    L.; Poggio, M.; Mamin, H

    Degen, C. L.; Poggio, M.; Mamin, H. J.; Rugar, D. Physical Review Letters 2007, 99, 250601

    2007

  17. [25]

    E.; Cadeddu, D.; Xue, F.; Peddibhotla, P.; Poggio, M

    Herzog, B. E.; Cadeddu, D.; Xue, F.; Peddibhotla, P.; Poggio, M. Applied Physics Letters 2014, 105, 043112

    2014

  18. [26]

    S.; Joseph, R

    Silver, M. S.; Joseph, R. I.; Hoult, D. I. Physical Review A 1985, 31, 2753--2755

    1985

  19. [27]

    Journal of Magnetic Resonance

    Tannus, A.; Garwood, M. Journal of Magnetic Resonance. Series A 1996, 120, 133--137

    1996

  20. [28]

    T.; van Beek, J

    Tomka, I. T.; van Beek, J. D.; Joss, R.; Meier, B. H. Physical Chemistry Chemical Physics 2013, 15, 3438--3441

    2013

  21. [29]

    Journal of Magnetic Resonance

    Kupce, E.; Freeman, R. Journal of Magnetic Resonance. Series A 1995, 115, 273 -- 276

    1995

  22. [30]

    R.; Pegg, D

    Bendall, M. R.; Pegg, D. T. Journal of Magnetic Resonance 1986, 67, 376 -- 381

    1986

  23. [31]

    Loretz, M.; Pezzagna, S.; Meijer, J.; Degen, C. L. Applied Physics Letters 2014, 104, 033102

    2014

  24. [32]

    C.; den Haan, A

    Overweg, H. C.; den Haan, A. M. J.; Eerkens, H. J.; Alkemade, P. F. A.; la Rooij, A. L.; Spreeuw, R. J. C.; Bossoni, L.; Oosterkamp, T. H. Applied Physics Letters 2015, 107, 072402

    2015

  25. [33]

    S.; Schliesser, A

    Tsaturyan, Y.; Barg, A.; Polzik, E. S.; Schliesser, A. Nature Nanotechnology 2017, 12

    2017

  26. [34]

    H.; Fedorov, S

    Ghadimi, A. H.; Fedorov, S. A.; Engelsen, N. J.; Bereyhi, M. J.; Schilling, R.; Wilson, D. J.; Kippenberg, T. J. Science 2018, 360, 764--768

    2018

  27. [35]

    Usenko, O.; Vinante, A.; Wijts, G.; Oosterkamp, T. H. Applied Physics Letters 2011, 98, 133105 mcitethebibliography document

    2011

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.