Pith. sign in

REVIEW 4 major objections 5 minor 59 references

Quantitative structure determination from experimental four-dimensional scanning transmission electron microscopy via the scattering matrix

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A single 4D-STEM dataset can now reveal a crystal's structure, even with thick samples and multiple scattering.

desk verdict First experimental S-matrix 4D-STEM structure retrieval, with a clean new t+2Δf constraint and honest simulations; the main caveat is an unverified Lorentzian effective source chosen by grid search. read the letter →

arxiv 2506.21004 v1 pith:7ALQABMM submitted 2025-06-26 cond-mat.mtrl-sci physics.ins-det

Emmanuel W.C. Terzoudis-Lumsden , Alireza Sadri , Matthew Weyland , Laure Bourgeois , Stephanie M. Ribet , Georgios Varnavides , Colin Ophus , Timothy C. Petersen
show 1 more author
Scott D. Findlay
This is my paper
classification cond-mat.mtrl-sciphysics.ins-det
keywords 4DSTEMscatteringmatrixstructuredeterminationphaseretrievalpartialspatialcoherencemultipleprojectedpotentialSrTiO3
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

This paper develops the scattering matrix method for quantitative structure determination from four-dimensional scanning transmission electron microscopy (4D STEM) data, targeting samples thick enough that multiple scattering cannot be neglected. Through simulation and experiment, it shows that three modifications — coherence-aware phase retrieval, a symmetry constraint linking thickness to defocus, and inclusion of dark-field data — make the reconstruction work on real experimental data. The paper's central experimental claim is that a single-defocus 4D STEM dataset from an imperfectly coherent microscope suffices to reconstruct the projected electrostatic potential of a periodic crystal with good quantitative agreement. If true, this would simplify a class of atomic-resolution structure determinations that previously relied on through-focal series or more complex experimental setups.

What carries the argument

The scattering matrix S = exp(iπtA/K), the quantum mechanical operator relating incident and exit plane-wave amplitudes through a crystalline sample, together with the structure matrix A whose Fourier coefficients U_g encode the projected electrostatic potential. The reconstruction is a two-step inversion: phase retrieval reconstructs S from 4D STEM intensities, then nonlinear conjugate gradient descent refines A, thickness, defocus, phase factors, and (in the experimental case) mistilt. Three named mechanisms carry the new developments: the Clark–Peele-style amplitude update (a square-root intensity-ratio update that incorporates the effective source convolution into the phase retrieval step); the antidiagonal symmetry constraint S_{h,g} = S_{-g,-h}, whose enforcement yields the thickness–defocus estimate t + 2Δf; and the iteration loop that re-seeds phase retrieval from a simulated S, which supplies phase relations that let dark-field rows enter the cost function.

What would settle it

Re-run the same single-defocus reconstruction pipeline on a well-characterized periodic specimen while independently measuring the effective source distribution (for example, by fitting the probe intensity in vacuum or the source size from a known structure), or by acquiring data with a deliberately measured wider source (HWHM ≥ 0.6 Å). If the reconstructed Fourier coefficients of the potential deviate substantially from the known structure, or if an independently measured non-Lorentzian source of comparable width fails to reproduce the reported reconstruction, the claim that the method is robust to partial spatial coherence on experimental data would be falsified.

Watch

Extended reading notes

Core claim

The paper demonstrates experimental reconstruction of the projected electrostatic potential of a monolithic SrTiO3 crystal from a single-defocus 4D STEM dataset, via the scattering matrix. The reconstruction pipeline first retrieves the scattering matrix S from measured intensities using phase retrieval, then solves for the structure matrix A (whose off-diagonal elements are the Fourier coefficients of the projected potential) by gradient descent. Three developments make this work on imperfect experimental data: (1) a modified phase retrieval amplitude update that accounts for partial spatial coherence as an effective source convolution; (2) use of the antidiagonal symmetry of S to estimate the linear combination t + 2Δf of thickness and defocus when defocus is unknown; and (3) inclusion of dark-field diffraction data after the first iteration re-seeds phase relations in the dark-field rows. The paper claims that combining these advances yields a reconstruction with the Sr, Ti, and weakly-scattering O columns visible at their expected coordinates and Fourier coefficients of the potential in quantitative agreement with the expected structure out to beyond the bright-field disk.

Load-bearing premise

The experimental reconstruction presumes that the true partial spatial coherence is faithfully represented by a Lorentzian effective source with a half-width-at-half-maximum near 0.3 Å — a model chosen by trial-and-error grid search after Gaussian models failed.

Editorial extensions

If this is right

  • A single-defocus 4D STEM dataset, averaged over a unit cell, can serve as the sole input for quantitative projected-potential reconstruction of a periodic sample, removing the need for through-focal series data with their associated dose and alignment burdens.
  • The t + 2Δf constraint from antidiagonal symmetry gives an experimental route to jointly bracket thickness and defocus before the full structure optimization, shrinking the parameter search space for unknown samples.
  • Including dark-field intensities extends the reliable range of reconstructed Fourier coefficients of the potential, which should help at small probe-forming apertures and for oxygen columns or other weak scatterers.
  • The scattering-matrix formulation provides a sparse, periodic, thickness-independent forward calculation, so for periodic samples it offers a computationally distinct alternative to inverse multislice ptychography, with which the paper demonstrates comparable experimental reconstruction quality.
  • When the effective source width is uncertain, overestimating it yields reliable low-order Fourier coefficients whereas underestimating it fails, so safe practice for this method is to err toward larger assumed source widths.

Reading between the lines

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

  • The same coherence-aware update and dark-field inclusion strategy could be transplanted into multislice ptychography pipelines, since the amplitude update of Eq. (7) is not specific to the scattering-matrix basis.
  • The demonstrated sensitivity to effective source shape suggests that independent, structure-free measurements of the effective source (e.g., from vacuum or amorphous regions of the same dataset) would remove the largest empirical uncertainty of the method and could be tested before applying the reconstruction to an unknown structure.
  • Because the method recovers only the projected potential (with periodicity built in), it is a natural building block for a phased workflow: scattering-matrix reconstruction for a periodic region of interest, followed by inverse multislice ptychography with the reconstructed potential as an initialization for non-periodic parts of the field of view.
  • The reported inability to reconstruct with a 0.6 Å HWHM source at single defocus even when known exactly suggests an approximate coherence cutoff — beyond which no amount of parameter tuning will recover high-order Fourier coefficients — that could be quantified as a practical specification for this technique.
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

4 major / 5 minor

Summary. The manuscript presents three algorithmic modifications to the scattering-matrix (S-matrix) approach for retrieving the projected electrostatic potential from 4D STEM data: a coherence-aware phase retrieval update (Eq. 7), a symmetry-based constraint that estimates the combination t+2Δf (Eqs. 9–14), and inclusion of dark-field detector regions. The methods are tested on simulated data with noise, spatial incoherence, and systematic thickness/defocus grids, then applied to experimental 4D STEM data from monolithic SrTiO3, yielding a reconstruction that the authors describe as in excellent quantitative agreement with the expected structure. The experimental reconstruction is compared with mixed-state inverse multislice ptychography to provide an external consistency check.

Significance. If the experimental result is robust, the paper delivers the first demonstration that a single-defocus 4D STEM dataset from a partially coherent instrument can be inverted to a projected potential under multiple scattering via the S-matrix. The t+2Δf derivation is clean and parameter-free, and the simulation study is unusually thorough, including 100 noise realisations, a dose ladder, and systematic thickness/defocus grids. The significance is tempered by the fact that the experimental demonstration rests on an assumed Lorentzian effective source whose width is chosen by grid search on the same data; the paper's own simulations show strong sensitivity to this parameter, so the central experimental claim is not yet established as robust.

major comments (4)
  1. [Experimental reconstruction] The central experimental claim rests on the assumption of a purely Lorentzian effective source with HWHM 0.3 Å, selected by grid search over reconstructions of the same experimental data (Fig. 7). The paper's own simulations (Fig. 3a,d) show that underestimating the HWHM by 50% causes the reconstruction to fail, and Fig. 3e shows failure for a 0.6 Å HWHM source even when known exactly. Because no independent measurement of the effective source is provided, and because Gaussian source models failed (requiring the Lorentzian form), the experimental success is not yet established as robust. I request either an independent source characterization, a demonstration that the final potential is stable under plausible variation of the source model and width, or a quantitative statement of the resulting uncertainty in the reconstructed potential.
  2. [Experimental reconstruction] The final parameter selection uses qualitative appearance of the potential and Pearson correlation between experimental and simulated 4D STEM intensities (Fig. 7). The manuscript itself notes (Fig. 3 and accompanying text) that extending the apodisation range improves Pearson correlation even when the reconstruction is degrading due to overfitting to noise. Thus Pearson correlation alone is not a reliable accuracy metric. The authors should report how the final potential changes for neighboring grid points, particularly the parameter combinations that achieved similar Pearson correlations in Fig. 7b, and provide a quantitative accuracy metric against the known SrTiO3 structure.
  3. [Experimental reconstruction] The abstract and conclusion claim 'excellent quantitative agreement' with the expected structure, but the only quantitative evidence is the Fourier-coefficient visualisation in Fig. 8(a), which is not accompanied by numerical values or error bars. The reconstructed absorptive potential is acknowledged to disagree with simulation, which shows that model mismatch can be present without invalidating the visual fit. To support the quantitative claim, the authors should provide, for example, a table of reconstructed versus expected low-order structure factors with uncertainties.
  4. [Spatial incoherence] The trial-and-error failure of Gaussian effective source models and success of Lorentzian models is surprising and unexplained; the physical effective source is more typically Gaussian- or Voigt-like. The Lorentzian may be absorbing other model errors, such as residual aberrations, scan noise, or sample mistilt. Please discuss or test whether the Lorentzian width is compensating for other parameter errors, for example by repeating the fine grid search with a Voigt profile or with a source model plus residual aberration parameters.
minor comments (5)
  1. [Experimental reconstruction] There is a typo in the Experimental reconstruction section: 'descibed' should be 'described'.
  2. [Fig. 7 caption] The caption says 'The starting guess for sample thickness is set via the antidiagonal symmetry constraint from the initial S reconstruction and the assumed defocus'; this should be made explicit, since the antidiagonal constraint only gives t+2Δf, so the thickness is obtained as t = (t+2Δf) − 2Δf.
  3. [Introduction and methods] The symbol Λ(R) for the effective source distribution is visually very close to λ for wavelength; please use a distinct notation or font to avoid confusion.
  4. [Experimental reconstruction] The Pearson correlation values (e.g., 0.94 in Fig. 7b) are cited without stating the number of data points or the reciprocal-space region used; please define the metric precisely.
  5. [Fig. 8(b)] The ptychographic reconstruction is initialised with the same defocus and thickness as the S-matrix reconstruction; because these parameters are not independently determined, the comparison should be described explicitly as a consistency check rather than a validation of the parameters.

Circularity Check

0 steps flagged · score 0.0 of 10

The paper's derivation chain is self-contained: the t+2Δf symmetry result is derived from the stated model equations, and the experimental reconstruction is benchmarked against an independent ptychographic reconstruction rather than being forced by a fit.

full rationale

No load-bearing circular step is present. The forward model (Eqs. 1–4), the phase-retrieval update (Eq. 7), the antidiagonal-symmetry derivation (Eqs. 9–14), and the dark-field inclusion are all derived from stated assumptions and equations within the paper, with no fitted quantity entering as an equivalent of the claimed output. The experimental parameters (defocus, thickness, source HWHM, mistilt) are selected by grid search against the same 4D STEM data, but these are nuisance parameters of an inverse problem, not the structural prediction itself; the central claim is the reconstructed potential, which is cross-checked against mixed-state inverse multislice ptychography (Fig. 8b) and against the expected SrTiO3 structure. The Lorentzian effective-source model is an empirical assumption whose sensitivity is openly demonstrated in the paper's own simulations (Fig. 3), so it is a correctness risk rather than a circularity. Self-citations to Findlay et al. (2021) and Sadri and Findlay (2023) supply prior methodology, but the load-bearing derivations here are re-derived from first principles in the text, so the self-citations are not load-bearing. The acknowledged limitations — Gaussian source models failing, absorptive potential disagreement with simulation — are stated plainly and do not indicate that any equation reduces to its own input by construction.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

No new physical entities are postulated; the reconstruction rests on the Bloch-wave scattering model, a set of experimentally fitted parameters (defocus, thickness, source HWHM, mistilt) obtained by grid search, and the assumption that compact-support phase retrieval recovers S up to row phase factors. The main accounting concern is that four fitted quantities are chosen to optimise agreement with the very data used to demonstrate success, though the ptychography comparison provides external anchoring.

free parameters (5)
  • Effective source HWHM (Lorentzian) = 0.3 Å
    Chosen by grid search over 0.2, 0.4, 0.6 Å after a Gaussian source model produced failed reconstructions; enters the phase retrieval update (Eq. 7).
  • Probe defocus Δf = 70 Å overfocus
    Grid-searched from -90 to +90 Å; the final value sets the thickness via the t+2Δf constraint and the pair is refined jointly.
  • Sample thickness t = 110 Å
    Initialised from the t+2Δf estimate at the chosen defocus, then refined in a finer grid search; thickness is an optimised variable in the A-from-S step.
  • Sample mistilt = (-2, 2) mrad
    Selected from three trial values (0,0), (-2,2), (-4,4) mrad on the basis of reconstruction quality and Pearson correlation.
  • Apodisation envelope parameters = q_max/2 for 3 iterations, expanding to about 2x q_max
    Hand-chosen learning-rate envelope controlling which Fourier coefficients are updated; stabilises convergence but is not derived from data.
assumptions (6)
  • domain assumption Antidiagonal symmetry A_h,g = A_-g,-h holds for the experimental sample (Eq. 8).
    Used to resolve row phase factors and to derive the t+2Δf constraint (Eq. 14); requires the projected potential approximation and periodicity along the beam direction.
  • domain assumption arg[S_h,h] is dominated by free-space propagation, arg[S_h,h] ≈ -π t h²/K (used to derive Eq. 9).
    The diagonal phase assumption underlies the t+2Δf estimate; the paper supports it phenomenologically via simulation rather than by direct experimental verification.
  • domain assumption Spatial incoherence is correctly modelled by convolving with a known effective source distribution and the modified amplitude update (Eqs. 6 and 7).
    The experimental source distribution is not measured; a Lorentzian is assumed after a Gaussian model failed, and reconstruction quality is sensitive to the assumed width (Fig. 3).
  • domain assumption The only significant unknown aberration is defocus.
    Relies on aberration correction being effective; residual astigmatism is visible in the ptychographic probe modes (Fig. 9) and is ignored in the S-matrix model.
  • domain assumption Mean unit cell averaging makes the sample exactly periodic over the averaged region.
    Averaging the 45 Å x 45 Å scan assumes identical unit cells; residual drift, thickness variation, and defects become unmodelled model mismatch, which the authors acknowledge.
  • domain assumption The Bloch-wave scattering model with absorptive potential (Eqs. 1-2) adequately describes the 110 Å SrTiO3 sample.
    Thermal scattering is treated phenomenologically; the authors show this is adequate for the elastic potential, but the reconstructed absorptive potential is not in good quantitative agreement with simulation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Quantitative structure determination from experimental four-dimensional scanning transmission electron microscopy via the scattering matrix." pith.science (2026). https://pith.science/paper/7ALQABMM

@misc{pith2026250621004,
  author       = {Pith},
  title        = {Pith review of: Quantitative structure determination from experimental four-dimensional scanning transmission electron microscopy via the scattering matrix},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7ALQABMM}},
  note         = {Machine review of arXiv:2506.21004}
}
abstract

Considerable inroads have recently been made on algorithms to determine the sample potential from four-dimensional scanning transmission electron microscopy data from thick samples where multiple scattering cannot be neglected. This paper further develops the scattering matrix approach to such structure determination. Through simulation, we demonstrate how this approach can be modified to better handle partial spatial coherence, unknown probe defocus, and information from the dark field region. By combining these developments we reconstruct the electrostatic potential of a monolithic SrTiO$_3$ crystal showing good quantitative agreement with the expected structure.

Figures

Figures reproduced from arXiv: 2506.21004 by the authors.

Figure 1
Figure 1. Top row: Schematic outlining the forward problem. A known structure is encoded into the structure matrix A from which the scattering matrix S and then the 4D STEM intensities I4D can be calculated. Bottom row: Schematic outlining the inverse problem. From measured 4D STEM data I4D, phase retrieval allows S to be partially reconstructed (some portions are inaccessible due to the probe-forming aperture). From the reco… view at source ↗
Figure 2
Figure 2. (a) Comparison of simulated CBED patterns between [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. From 4D STEM data incorporating a Gaussian effective source distribution with a small (0.3 [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: Box plots summarising the distribution of thickness values estimated by enforcing antidiagonal symmetry on [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: (a,b) Simulated CBED patterns at 300 keV with the probe atop a 200 [PITH_FULL_IMAGE:figures/full_fig_p013_5.png]
Figure 6
Figure 6. Figure 6: (a) Virtual ABF image from raw data, with the mean unit cell inset. (b) 3 [PITH_FULL_IMAGE:figures/full_fig_p015_6.png]
Figure 7
Figure 7. Figure 7: (a) Visualisation of the reconstructed potential from experimental 4D STEM SrTiO [PITH_FULL_IMAGE:figures/full_fig_p017_7.png]
Figure 8
Figure 8. Figure 8: (a) Reconstructed elastic projected potential from experimental 4D STEM data via the [PITH_FULL_IMAGE:figures/full_fig_p018_8.png]
Figure 9
Figure 9. Figure 9: Reconstructed probe modes from mixed-state inverse [PITH_FULL_IMAGE:figures/full_fig_p021_9.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

59 extracted references · 58 canonical work pages

  1. [1]

    L. J. Allen and M. P. Oxley. Structure determination at the atomic level from dynamical electron diffraction data under systematic row conditions. Ultramicroscopy, 88 0 (3): 0 195--209, 2001

    work page 2001

  2. [2]

    L. J. Allen, H. Leeb, and A. E. C. Spargo. Retrieval of the projected potential by inversion from the scattering matrix in electron--crystal scattering. Acta Crystallographica Section A: Foundations of Crystallography, 55 0 (2): 0 105--111, 1999

    work page 1999

  3. [3]

    L. J. Allen, H. M. L. Faulkner, and H. Leeb. Inversion of dynamical electron diffraction data including absorption. Acta Crystallographica Section A: Foundations of Crystallography, 56 0 (2): 0 119--126, 2000

    work page 2000

  4. [4]

    Bird and Q

    D. Bird and Q. King. Absorptive form factors for high-energy electron diffraction. Acta Crystallographica Section A: Foundations of Crystallography, 46: 0 202--208, 1990

    work page 1990

  5. [5]

    H. G. Brown. py\_multislice . https://github.com/HamishGBrown/py_multislice, 2024. [Online; accessed 15-July-2024]

    work page 2024

  6. [6]

    H. G. Brown, Z. Chen, M. Weyland, C. Ophus, J. Ciston, L. J. Allen, and S. D. Findlay. Structure retrieval at atomic resolution in the presence of multiple scattering of the electron probe. Physical Review Letters, 121 0 (26): 0 266102, 2018

    work page 2018

  7. [7]

    H. G. Brown, P. M. Pelz, S.-L. Hsu, Z. Zhang, R. Ramesh, K. Inzani, E. Sheridan, S. M. Griffin, M. Schloz, T. C. Pekin, et al. A three-dimensional reconstruction algorithm for scanning transmission electron microscopy data from a single sample orientation. Microscopy and Microanalysis, 28 0 (5): 0 1632--1640, 2022

    work page 2022

  8. [8]

    M. C. Cao, Z. Chen, Y. Jiang, and Y. Han. Automatic parameter selection for electron ptychography via B ayesian optimization. Scientific Reports, 12: 0 12284, 2022

    work page 2022

Show all 59 references
  1. [9]

    Z. Chen, Y. Jiang, Y.-T. Shao, M. E. Holtz, M. Odstr c il, M. Guizar-Sicairos, I. Hanke, S. Ganschow, D. G. Schlom, and D. A. Muller. Electron ptychography achieves atomic-resolution limits set by lattice vibrations. Science, 372 0 (6544): 0 826--831, 2021

    2021

  2. [10]

    J. N. Clark and A. G. Peele. Simultaneous sample and spatial coherence characterisation using diffractive imaging. Applied Physics Letters, 99: 0 154103, 2011

    2011

  3. [11]

    Clark and P

    L. Clark and P. D. Nellist. Electron ptychography. arXiv preprint arXiv:2503.10917, 2025

    2025 arXiv

  4. [12]

    Diederichs, Z

    B. Diederichs, Z. Herdegen, A. Strauch, F. Filbir, and K. M \"u ller-Caspary. Exact inversion of partially coherent dynamical electron scattering for picometric structure retrieval. Nature Communications, 15: 0 101, 2024

    2024

  5. [13]

    J. J. Donatelli and J. C. H. Spence. Inversion of many-beam B ragg intensities for phasing by iterated projections: R emoval of multiple scattering artifacts from diffraction data. Physical Review Letters, 125 0 (6): 0 065502, 2020

    2020

  6. [14]

    Dwivedi, A

    P. Dwivedi, A. Konijnenberg, S. Pereira, and H. Urbach. Lateral position correction in ptychography using the gradient of intensity patterns. Ultramicroscopy, 192: 0 29--36, 2018

    2018

  7. [15]

    Dwyer, C

    C. Dwyer, C. Maunders, C. L. Zheng, M. Weyland, P. C. Tiemeijer, and J. Etheridge. Sub-0.1 nm-resolution quantitative scanning transmission electron microscopy without adjustable parameters. Applied P hysics L etters , 100 0 (19), 2012

    2012

  8. [16]

    Findlay, N

    S. Findlay, N. Shibata, H. Sawada, E. Okunishi, Y. Kondo, and Y. Ikuhara. Dynamics of annular bright field imaging in scanning transmission electron microscopy. Ultramicroscopy, 110 0 (7): 0 903--923, 2010

    2010

  9. [17]

    S. D. Findlay. Quantitative structure retrieval using scanning transmission electron microscopy. Acta Crystallographica Section A: Foundations of Crystallography, 61 0 (4): 0 397--404, 2005

    2005

  10. [18]

    S. D. Findlay, H. G. Brown, P. M. Pelz, C. Ophus, J. Ciston, and L. J. Allen. Scattering matrix determination in crystalline materials from 4D scanning transmission electron microscopy at a single defocus value. Microscopy and Microanalysis, 27 0 (4): 0 744--757, 2021

    2021

  11. [19]

    C. Gao, C. Hofer, D. Jannis, A. B \'e ch \'e , J. Verbeeck, and T. J. Pennycook. Overcoming contrast reversals in focused probe ptychography of thick materials: an optimal pipeline for efficiently determining local atomic structure in materials science. Applied Physics Letters...

    2022

  12. [20]

    Gilgenbach, X

    C. Gilgenbach, X. Chen, and J. M. LeBeau. A methodology for robust multislice ptychography. Microscopy and Microanalysis, 30 0 (4): 0 703--711, 2024

    2024

  13. [21]

    Gladyshev, B

    A. Gladyshev, B. Haas, T. M. Boland, P. Rez, and C. T. Koch. Reconstructing lattice vibrations of crystals with electron ptychography. arXiv preprint arXiv:2309.12017, 2023

    2023 arXiv

  14. [22]

    S. P. Gleason, A. Rakowski, S. M. Ribet, S. E. Zeltmann, B. H. Savitzky, M. Henderson, J. Ciston, and C. Ophus. Random forest prediction of crystal structure from electron diffraction patterns incorporating multiple scattering. arXiv preprint arXiv:2406.16310, 2024

    2024 arXiv

  15. [23]

    Herdegen, B

    Z. Herdegen, B. Diederichs, and K. M \"u ller-Caspary. Thermal vibrations in the inversion of dynamical electron scattering. Physical Review B, 110 0 (6): 0 064102, 2024

    2024

  16. [24]

    C. J. Humphreys. The scattering of fast electrons by crystals. Reports on Progress in Physics, 42: 0 1825, 1979

    1979

  17. [25]

    Jiang, Z

    Y. Jiang, Z. Chen, Y. Han, P. Deb, H. Gao, S. Xie, P. Purohit, M. W. Tate, J. Park, S. M. Gruner, V. Elser, and D. A. Muller. Electron ptychography of 2 D materials to deep sub- ngstr \"o m resolution. Nature, 559 0 (7714): 0 343--349, 2018

    2018

  18. [26]

    Kaipio and E

    J. Kaipio and E. Somersalo. Statistical inverse problems: discretization, model reduction and inverse crimes. Journal of computational and applied mathematics, 198: 0 493--504, 2007

    2007

  19. [27]

    S. V. Kalinin, C. Ophus, P. M. Voyles, R. Erni, D. Kepaptsoglou, V. Grillo, A. R. Lupini, M. P. Oxley, E. Schwenker, M. K. Chan, et al. Machine learning in scanning transmission electron microscopy. Nature Reviews Methods Primers, 2 0 (1): 0 11, 2022

    2022

  20. [28]

    S. V. Kalinin, D. Mukherjee, K. Roccapriore, B. J. Blaiszik, A. Ghosh, M. A. Ziatdinov, A. Al-Najjar, C. Doty, S. Akers, N. S. Rao, et al. Machine learning for automated experimentation in scanning transmission electron microscopy. npj Computational Materials, 9 0 (1): 0 227, 2023

    2023

  21. [29]

    X. Li, O. E. Dyck, M. P. Oxley, A. R. Lupini, L. McInnes, J. Healy, S. Jesse, and S. V. Kalinin. Manifold learning of four-dimensional scanning transmission electron microscopy. npj Computational Materials, 5 0 (1): 0 5, 2019

    2019

  22. [30]

    A. M. Maiden, M. J. Humphry, and J. M. Rodenburg. Ptychographic transmission microscopy in three dimensions using a multi-slice approach. Journal of the Optical Society of America A, 29: 0 1606--1614, 2012

    2012

  23. [31]

    Maunders, C

    C. Maunders, C. Dwyer, P. C. Tiemeijer, and J. Etheridge. Practical methods for the measurement of spatial coherence—a comparative study. Ultramicroscopy, 111 0 (8): 0 1437--1446, 2011

    2011

  24. [32]

    Mukherjee, J

    D. Mukherjee, J. T. Gamler, S. E. Skrabalak, and R. R. Unocic. Lattice strain measurement of core@shell electrocatalysts with 4D scanning transmission electron microscopy nanobeam electron diffraction. ACS Catalysis, 10 0 (10): 0 5529--5541, 2020

    2020

  25. [33]

    Munshi, A

    J. Munshi, A. Rakowski, B. H. Savitzky, S. E. Zeltmann, J. Ciston, M. Henderson, S. Cholia, A. M. Minor, M. K. Chan, and C. Ophus. Disentangling multiple scattering with deep learning: application to strain mapping from electron diffraction patterns. npj Computational Material...

    2022

  26. [34]

    P. D. Nellist and J. M. Rodenburg. Beyond the conventional information limit: the relevant coherence function. Ultramicroscopy, 54: 0 61--74, 1994

    1994

  27. [35]

    K. Ooe, A. Sadri, and S. D. Findlay. Optimising structure factors from four-dimensional scanning transmission electron microscopy via the scattering matrix, 2025. Submitted

    2025

  28. [36]

    C. Ophus. Four-dimensional scanning transmission electron microscopy ( 4D-STEM ): F rom scanning nanodiffraction to ptychography and beyond. Microscopy and Microanalysis, 25 0 (3): 0 563--582, 2019

    2019

  29. [37]

    Ophus, J

    C. Ophus, J. Ciston, and C. T. Nelson. Correcting nonlinear drift distortion of scanning probe and scanning transmission electron microscopies from image pairs with orthogonal scan directions. Ultramicroscopy, 162: 0 1--9, 2016

    2016

  30. [38]

    Ophus, T

    C. Ophus, T. R. Harvey, F. S. Yasin, H. G. Brown, P. M. Pelz, B. H. Savitzky, J. Ciston, and B. J. McMorran. Advanced phase reconstruction methods enabled by four-dimensional scanning transmission electron microscopy. Microscopy and Microanalysis, 25 0 (S2): 0 10--11, 2019

    2019

  31. [39]

    Ophus, S

    C. Ophus, S. E. Zeltmann, A. Bruefach, A. Rakowski, B. H. Savitzky, A. M. Minor, and M. C. Scott. Automated crystal orientation mapping in py4DSTEM using sparse correlation matching. Microscopy and Microanalysis, 28 0 (2): 0 390--403, 2022

    2022

  32. [40]

    P. M. Pelz, H. G. Brown, S. Stonemeyer, S. D. Findlay, A. Zettl, P. Ercius, Y. Zhang, J. Ciston, M. Scott, and C. Ophus. Phase-contrast imaging of multiply-scattering extended objects at atomic resolution by reconstruction of the scattering matrix. Physical Review Research, 3 ...

    2021

  33. [41]

    R. S. Pennington, W. Van den Broek, and C. T. Koch. Third-dimension information retrieval from a single convergent-beam transmission electron diffraction pattern using an artificial neural network. Physical Review B, 89: 0 205409, 2014

    2014

  34. [42]

    D. Ren, C. Ophus, M. Chen, and L. Waller. A multiple scattering algorithm for three dimensional phase contrast atomic electron tomography. Ultramicroscopy, 208: 0 112860, 2020

    2020

  35. [43]

    S. M. Ribet, G. Varnavides, C. Pedroso, B. E. Cohen, P. Ercius, M. C. Scott, and C. Ophus. Uncovering the three-dimensional structure of upconverting core--shell nanoparticles with multislice electron ptychography. Applied Physics Letters, 124: 0 240601, 2024

    2024

  36. [44]

    K. M. Roccapriore, O. Dyck, M. P. Oxley, M. Ziatdinov, and S. V. Kalinin. Automated experiment in 4D-STEM : exploring emergent physics and structural behaviors. ACS Nano, 16 0 (5): 0 7605--7614, 2022

    2022

  37. [45]

    Sadri and S

    A. Sadri and S. D. Findlay. Determining the projected crystal structure from four-dimensional scanning transmission electron microscopy via the scattering matrix. Microscopy and Microanalysis, 29: 0 967--982, 2023

    2023

  38. [46]

    B. H. Savitzky, S. E. Zeltmann, L. A. Hughes, H. G. Brown, S. Zhao, P. M. Pelz, T. C. Pekin, E. S. Barnard, J. Donohue, L. R. DaCosta, et al. py4DSTEM : A software package for four-dimensional scanning transmission electron microscopy data analysis. Microscopy and Microanalysi...

    2021

  39. [47]

    Schloz, T

    M. Schloz, T. C. Pekin, H. G. Brown, D. O. Byrne, B. D. Esser, E. Terzoudis-Lumsden, T. Taniguchi, K. Watanabe, S. D. Findlay, B. Haas, et al. Improved three-dimensional reconstructions in electron ptychography through defocus series measurements. Microscopy and Microanalysis,...

    2024

  40. [48]

    H. Sha, J. Cui, and R. Yu. Deep sub-angstrom resolution imaging by electron ptychography with misorientation correction. Science Advances, 8: 0 eabn2275, 2022

    2022

  41. [49]

    J. C. H. Spence, B. Calef, and J. M. Zuo. Dynamic inversion by the method of generalized projections. Acta Crystallographica Section A: Foundations of Crystallography, 55 0 (2): 0 112--118, 1999

    1999

  42. [50]

    M. Sun, C. M. Azumaya, E. Tse, D. P. Bulkley, M. B. Harrington, G. Gilbert, A. Frost, D. Southworth, K. A. Verba, Y. Cheng, and D. A. Agard. Practical considerations for using K3 cameras in CDS mode for high-resolution and high-throughput single particle cryo- EM . Journal of ...

    2021

  43. [51]

    E. W. C. Terzoudis-Lumsden, T. C. Petersen, H. G. Brown, P. M. Pelz, C. Ophus, and S. D. Findlay. Resolution of virtual depth sectioning from four-dimensional scanning transmission electron microscopy. Microscopy and Microanalysis, 29 0 (4): 0 1409--1421, 2023

    2023

  44. [52]

    Thibault and A

    P. Thibault and A. Menzel. Reconstructing state mixtures from diffraction measurements. Nature, 494 0 (7435): 0 68--71, 2013

    2013

  45. [53]

    Van den Broek and C

    W. Van den Broek and C. T. Koch. Method for retrieval of the three-dimensional object potential by inversion of dynamical electron scattering. Physical Review Letters, 109: 0 245502, 2012

    2012

  46. [54]

    Van den Broek and C

    W. Van den Broek and C. T. Koch. General framework for quantitative three-dimensional reconstruction from arbitrary detection geometries in TEM . Physical Review B, 87: 0 184108, 2013

    2013

  47. [55]

    Varnavides, S

    G. Varnavides, S. M. Ribet, S. E. Zeltmann, Y. Yu, B. H. Savitzky, V. P. Dravid, M. C. Scott, and C. Ophus. Iterative phase retrieval algorithms for scanning transmission electron microscopy. arXiv preprint arXiv:2309.05250, 2023

    2023 arXiv

  48. [56]

    F. Wang, R. S. Pennington, and C. T. Koch. Inversion of dynamical scattering from large-angle rocking-beam electron diffraction patterns. Physical Review Letters, 117 0 (1): 0 015501, 2016

    2016

  49. [57]

    W. Yang, H. Sha, J. Cui, L. Mao, and R. Yu. Local-orbital ptychography for ultrahigh-resolution imaging. Nature Nanotechnology, 19 0 (5): 0 612--617, 2024

    2024

  50. [58]

    S. E. Zeltmann, S.-L. Hsu, H. G. Brown, S. Susarla, R. Ramesh, A. M. Minor, and C. Ophus. Uncovering polar vortex structures by inversion of multiple scattering with a stacked bloch wave model. Ultramicroscopy, 250: 0 113732, 2023

    2023

  51. [59]

    M. Zhu, M. Xu, Y. Yun, L. Wu, O. Shafir, C. Gilgenbach, L. W. Martin, I. Grinberg, J. E. Spanier, and J. M. LeBeau. Insights into chemical and structural order at planar defects in Pb _2 MgWO _6 using multislice electron ptychography. ACS nano, 2025

    2025

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.