Pith. sign in

REVIEW 3 major objections 5 minor 69 references

GARFIELD, a toolkit for interpreting ultrafast electron diffraction data of imperfect quasi-single crystals

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read The paper claims that a software toolkit called GARFIELD can interpret ultrafast electron diffraction patterns that defeat geometry-based auto-indexers, using reflection intensities as well as positions and explicitly modeling the…

desk verdict Useful, honest UED indexing toolkit with a real niche, but the intensity-decomposition claim is under-validated at the high mosaicity it targets. read the letter →

arxiv 2412.04197 v1 pith:VV7P34EW submitted 2024-12-05 cond-mat.mtrl-sci cond-mat.other

Alexander Marx , Sascha W. Epp This is my paper · ORCID
classification cond-mat.mtrl-scicond-mat.other
keywords ultrafastelectrondiffractioncrystalindexingmosaicityBraggspotdecompositionkinematicalmodelgraphicaluserinterfacenonlinearleastsquares
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 claims that a software toolkit called GARFIELD can interpret ultrafast electron diffraction patterns that defeat geometry-based auto-indexers: patterns from low-symmetry molecular crystals whose spots are blurred, elongated, and overlapping because the sample is a mosaic of slightly misaligned crystalline domains. The proposed remedy is to use the intensities of the Bragg spots, not only their positions, and to model the physical sources of fuzziness explicitly. If the claim holds, a UED time series only needs one representative image indexed, and merged Bragg spots can be resolved into the individual $hkl$ reflection contributions that pump–probe dynamics analysis actually needs. The paper further claims that an interactive workflow with visual comparison of simulated and recorded diffraction helps prevent wrong indexings from being accepted.

What carries the argument

The load-bearing object is a probabilistic kinematical diffraction model in which each reciprocal lattice point is expanded into a smooth Gaussian distribution by combining four independent effects: finite coherent domain size, orientation spread (mosaicity), beam divergence, and energy bandwidth. For mosaicity, the model uses the infinitesimal-rotation tangent-plane approximation: the rotation distribution is marginalized over rotations about the reciprocal lattice vector, turning a sphere-cap distribution into a flat bivariate normal in the tangent plane. The expected intensity of a reflection is then obtained by evaluating this normal distribution along the trace of the nearly planar Ewald sphere, with geometric corrections for larger rotations. Around this model sit two search tools: GRID SCAN, which tests up to 17,694,720 orientations on a uniformly sampled sphere to produce candidate orientations ranked by a figure of merit, and GEOFIT, which refines up to 22 parameters by nonlinear least squares against a cost function that combines intensity residuals, optional position residuals, and spot “ambits” that assign predicted reflections to observed spots.

What would settle it

Simulate a UED diffraction image from a known crystal structure with a deliberately anisotropic mosaic spread of $\sigma_{\mathrm{mos}}=10^\circ$ to $20^\circ$, run GRID SCAN and GEOFIT on the synthetic reduced data, and compare the recovered orientation and per-reflection intensity decomposition with the ground truth; a substantial disagreement at large mosaic spread, while a small-spread control recovers correctly, would show that the tangent-plane assumption is the limiting step.

Watch

Extended reading notes

Core claim

The paper presents GARFIELD, an interactive toolkit that finds the Laue indices of Bragg reflections in ultrafast electron diffraction images of imperfect quasi-single crystals and quantifies how much each reflection contributes when several overlap into one Bragg spot. Its central claim is that such indexing can succeed where position-only tools fail because the analysis exploits reflection intensities as a filter and because the prediction model accounts for the blurring effects that dominate UED data: domain orientation spread (mosaicity), finite coherent domain size, beam divergence, and energy bandwidth. With a known crystal structure supplied as input, GARFIELD predicts spot positions and intensities in a kinematical approximation, assigns predicted reflections to observed spots through circular “ambits,” and ranks candidate orientations by a cost function that combines intensity residuals with optional position residuals. The user explores parameter sets interactively, starting from a uniform grid search over the full rotation space and refining up to 22 parameters by nonlinear least-squares fitting.

Load-bearing premise

The intensity estimates that discriminate between candidate orientations rest on assuming that the spread of domain orientations is small and Gaussian enough that rotating a reciprocal-lattice point can be approximated by moving it in a flat tangent plane while averaging away rotations about the lattice vector; if the mosaic spread is large or strongly anisotropic, the predicted intensities may be too inaccurate to discriminate solutions.

Editorial extensions

If this is right

  • Because each frame in a UED time series is nearly identical, indexing one representative image per series indexes the whole series, reducing the job to a handful of images.
  • For high-mosaicity samples, many observed spots are superpositions; GARFIELD returns up to five contributing $hkl$ reflections per spot with estimated intensities, so apparent intensity changes in a pump–probe experiment can be traced to actual structural reflections.
  • Using intensities as a filter makes the orientation search robust to moderately inaccurate intensity predictions: errors reduce the selectivity of the filter rather than shifting the solution to a wrong indexing.
  • The model's predictions are detailed enough that weak unexpected spots can be singled out for further study, as in the rubrene examples where unexplained reflections point to a possible mixture of polymorphs.

Reading between the lines

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

  • The same intensity-plus-blurring strategy could plausibly be applied to other snapshot electron diffraction geometries, such as serial electron diffraction of nanocrystals, where blurred and overlapping spots also defeat geometry-based indexing.
  • The per-reflection intensity decomposition could be tested quantitatively on tilt-series or precession data in which the same overlapped reflections are measured separated, giving an external check the paper does not perform.
  • If the tangent-plane approximation is the bottleneck, a straightforward extension would replace the marginalization over rotations about the lattice vector with a numerical integration over the full rotation distribution, at the cost of the speed that the current Gaussian algebra provides.
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

3 major / 5 minor

Summary. The manuscript introduces GARFIELD, an interactive R-based software toolkit for indexing ultrafast electron diffraction (UED) patterns of imperfect quasi-single crystals. The program combines position and intensity information in a kinematical diffraction model with Gaussian models for mosaicity, beam divergence, bandwidth, and finite-size broadening. It provides two core tools: GRID SCAN, a global search over crystal orientations using a figure of merit that includes intensities, and GEOFIT, a nonlinear least-squares fitting tool that optimizes up to 22 parameters including orientation, beam center, mosaicity, and spot intensities. The central claims are (A) that GARFIELD can index blurred and overlapping UED reflections in cases where conventional geometry-only indexing tools fail, and (B) that it can decompose overlapping Bragg spots into individual reflection contributions useful for time-resolved structural dynamics. The paper includes a detailed model description, cost-function definitions, and application examples on three compounds (κ-(BEDT-TTF)2Cu[N(CN)2]Br, TBAI3, and rubrene), with reported R-factors of 14.9%, 26%, and 16% respectively.

Significance. If the claims hold, GARFIELD fills a genuine niche: UED data from thin, mosaic, low-symmetry crystals are often too blurred for standard auto-indexing tools, and the interactive, intensity-aware, mosaicity-aware approach is a plausible and potentially valuable alternative. The paper's strengths include a clear statement of design principles (use all available information, model fuzziness), an explicit and mostly well-documented mathematical model, a freely available open-source implementation (GPLv3) with a graphical user interface, and application to real experimental data from multiple compounds. The self-identified limitations—kinematical approximation, Gaussian profile assumptions, fixed σ3 values in some fits, and unexplained rubrene spots—are openly discussed. However, the validation is largely in-sample and the intensity-decomposition capability, which is central to claim B, is not independently validated in the high-mosaicity regime where GARFIELD is intended to operate.

major comments (3)
  1. [§IV C 2, Eqs. (12)–(14) and Fig. 5] The intensity predictions rest on the tangent-plane approximation obtained by marginalizing the mosaicity distribution over rotations about each reciprocal-lattice vector. The paper itself states that this is 'the most critical step in estimating diffraction intensities' and that σ_mos up to about 5° are acceptable, with values above 10° tolerable only under moderate anisotropy. Yet the rubrene applications (Supplementary §2.3.3) use fitted σ_mos values of 12.7° and 13.3° for slice 1 and 7.4° and 9.0° for slice 2, with σ3 fixed. No exact or numerically integrated calculation of the marginal/conditional intensity profile is reported for these high-mosaicity, anisotropic cases. Because claim B (decomposition of overlapping spots into individual reflection contributions) depends directly on these predicted intensities, the accuracy of the decomposition in the tool's intended operating regime is unverified. I recommend adding a synthetic-data validation: generate diffraction patterns with known orientation distributions (including anisotropic and high-σ_mos cases), then compare GARFIELD's per-reflection intensity decomposition against direct numerical integration of the full mosaicity distribution (without the tangent-plane approximation). This would establish the error bounds of the decomposition as a function of σ_mos and anisotropy.
  2. [Section V and Supplementary applications] The evaluation is in-sample. All application examples fit model parameters to the same observed diffraction patterns that are then used for the reported R-factors and visual comparisons; the simulated images are predictions from the fitted parameters, so the agreement is a fit-quality metric rather than an independent test. The paper's appeal to visual comparison as a 'special kind of cross-validation' (Section II) is informal. This matters for claim B because overfitting could produce a misleading decomposition of overlapping spots. I suggest adding an out-of-sample test: for the TBAI3 tilt series, fit parameters to one or two images and compare predictions (positions and intensities) to the remaining recorded images; alternatively, use synthetic data with known ground truth for the individual reflection contributions.
  3. [Introduction and Conclusion] The position that GARFIELD is needed because 'other indexing tools are ineffective' is not substantiated by any quantitative comparison. The paper does not report attempts to index the example patterns with established tools such as MOSFLM, DIALS, XDS, or serial-crystallography indexers, nor does it quantify how many spots those tools would fail to index. Since the stated contribution is explicitly relative to existing tools, a head-to-head benchmark (or at least a documented report that such tools fail on these data) is needed to support the significance claim. This is not a correctness issue with the model, but it is load-bearing for the paper's advertised advantage.
minor comments (5)
  1. [General] There are several typographical errors: 'If the the unit cell parameters' (Section II), 'there are reasons to belief' (Section IV C), 'posssible' (Section V), and 'correspont' (Section IV C 2). These should be corrected.
  2. [§IV C 2] The notation for covariance matrices is confusing: Σ_ω^(xyz), Σ_ω^(123), and Σ_ω^(12) appear in quick succession. A table or more explicit subscript convention would help the reader track which coordinate system is meant.
  3. [§IV C 2, Fig. 5] The caption of Figure 5 says 'A couple of rotations with axes perpendicular to an RLP ... are marked by a series of points' but the figure shows many more than a couple; rephrase to 'A set of rotations'.
  4. [§IV C 2] The paragraph introducing P_h states that it 'occupies a two-dimensional surface given by a sphere of radius h'; since the distribution is on a sphere in three-dimensional reciprocal space, it may be clearer to call it a two-dimensional manifold embedded in three dimensions, to avoid confusion with a planar surface.
  5. [Supplementary §2.3] The supplementary text refers to 'CmpdA' in Figure S1 and S2 legends, while the main text discusses Me4P[Pt(dmit)2]2 and other compounds; please ensure the naming is consistent or explained.

Circularity Check

0 steps flagged · score 0.0 of 10

This paper shows no significant circularity: GARFIELD's indexing and intensity estimates are model-based optimizations, not predictions that reduce to their fitted inputs by construction.

full rationale

The derivation chain is not circular. Section IV C builds the diffraction prediction from kinematic theory and explicit Gaussian models for finite size, mosaicity, divergence, and bandwidth; the predicted positions and intensities then enter the cost function S (Eqs. 1, 7, 8) together with the observed reduced data, and GEOFIT/GRID SCAN minimize that cost function. The reflection indices and per-reflection contribution estimates (claims A and B) are outputs of that optimization, not quantities defined to equal the observed spot intensities. The validation examples are in-sample fits, e.g. Supplementary Section 2.3.3 fits sigma_mos = 12.7 degrees and 13.3 degrees for rubrene slice 1 and then compares the simulated pattern with the same recorded image; this demonstrates fit quality rather than out-of-sample prediction, but the paper does not relabel those fits as independent predictions, so no fitted parameter is being renamed as a prediction. The only self-referential validation is the TBAI3 statement that 'the orientation analysis was repeated with GARFIELD and fully confirms the orientation and indexing found in the original analysis with a precursor of GARFIELD' (Supplementary Section 2.2.1), citing the authors' own preprint; this is a minor self-reference and is not load-bearing for the model derivation. The paper itself flags the main limitation in Section IV C 2: the marginalization over omega_3 and identification with the tangent-plane distribution is called 'the most critical step in estimating diffraction intensities', with sigma_mos up to 5 degrees acceptable and >10 degrees tolerable only under moderate anisotropy, while the rubrene fits reach 12.7 degrees and 13.3 degrees. That is a correctness and validation risk in the claimed operating regime, but it is an approximation-accuracy concern, not circularity: an inaccurate approximation would make the intensity estimates worse without making them equal to the data by definition.

Assumptions & free parameters 10 free parameters · 5 assumptions · 0 invented entities

The central indexing method depends on many fitted model parameters (up to 22 per image) and on several modeling assumptions: kinematical diffraction, Gaussian broadening, and a known unperturbed crystal structure. No new physical entities, particles, or forces are introduced.

free parameters (10)
  • Crystal lattice orientation (Theta, Phi, Psi) = Example fits: kappa 54.5 deg, 66.4 deg, 49.5 deg; rubrene from GridScan
    Orientation of the crystal lattice relative to the beam and detector; fitted by grid search and NLS; central to indexing.
  • Mosaicity (isotropic sigma_omega) = kappa 2.4 deg; TBAI3 1.793 deg
    Orientation spread of coherent domains, assumed isotropic in standard modeling; fitted per sample.
  • Mosaicity covariance matrix (ANISO model) = rubrene sigma_1/sigma_2 fitted (12.7/13.3 deg for cut 1, 7.4/9.0 deg for cut 2); sigma_3 fixed at 3 deg or 2.5 deg
    Six parameters (three standard deviations and three Euler angles) describe anisotropic orientation spread; used in some examples.
  • Shape transform width and relrod tilt (sigma_shp) = not quoted per sample; kappa effective thickness 36 nm
    Gaussian approximation to finite-size broadening, including the direction and width of reciprocal lattice rods.
  • Effective thickness t = kappa 36 nm
    Fitted but acknowledged to be an effective, biased value, not the geometric thickness; the paper states it cannot be used to fit true thickness.
  • Beam center (X, Y) = not quoted
    Detector coordinates of the direct beam; fitted because UED images often lack a precise beam center.
  • Magnification / camera length = not quoted
    Distance scaling between reciprocal space and detector pixels; fitted in GeoFit.
  • Image distortion parameters = not quoted
    Elliptical correction (two parameters) for detector or lens distortion; can be fitted.
  • Global intensity scale factor = not quoted
    Overall scaling between observed and calculated intensities; fitted.
  • Overall B factor = not quoted
    Debye-Waller-like correction applied to calculated intensities; fitted.
assumptions (5)
  • domain assumption Kinematical diffraction theory is adequate for describing UED patterns
    Invoked in Section IV C; the paper argues thin, mosaic samples justify neglecting dynamical scattering, but multiple scattering may still contribute.
  • domain assumption All four broadening effects (finite size, mosaicity, divergence, bandwidth) are independent normal distributions
    Section IV C: 'each individual effect is described by a single Gaussian distribution, and the distributions are mutually independent'; this enables analytic convolutions but is an approximation.
  • domain assumption Tangent plane and infinitesimal rotations approximation for mosaicity, including marginalization of omega_3, is accurate enough for intensity prediction
    Section IV C 2: 'the most critical step in estimating diffraction intensities is the infinitesimal rotations approximation'; errors grow with anisotropy and with mosaicity beyond about 5 to 10 degrees.
  • domain assumption The crystal structure of the unperturbed sample is known a priori
    Section III B: 'the crystal structure ... must be known a priori'; GARFIELD cannot index unknown structures or determine unit cells from scratch.
  • domain assumption Observed spot intensities are positively correlated with kinematically calculated intensities
    Section IV: GARFIELD relies on 'the correlation between estimated and observed intensities is positive' for the intensity filter to favor the correct solution.

how reviews work

0 comments
Cite this review

Pith. "Pith review of GARFIELD, a toolkit for interpreting ultrafast electron diffraction data of imperfect quasi-single crystals." pith.science (2026). https://pith.science/paper/VV7P34EW

@misc{pith2026241204197,
  author       = {Pith},
  title        = {Pith review of: GARFIELD, a toolkit for interpreting ultrafast electron diffraction data of imperfect quasi-single crystals},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VV7P34EW}},
  note         = {Machine review of arXiv:2412.04197}
}
read the original abstract

The analysis of ultrafast electron diffraction (UED) data from low-symmetry single crystals of small molecules is often challenged by the difficulty of assigning unique Laue indices to the observed Bragg reflections. For a variety of technical and physical reasons, UED diffraction images are typically of lower quality when viewed from the perspective of structure determination by single-crystal X-ray or electron diffraction. Nevertheless, time series of UED images can provide valuable insight into structural dynamics, provided that an adequate interpretation of the diffraction patterns can be achieved. GARFIELD is a collection of tools with a graphical user interface designed to facilitate the interpretation of diffraction patterns and to index Bragg reflections in challenging cases where other indexing tools are ineffective. To this end, GARFIELD enables the user to interactively create, explore, and optimize sets of parameters that define the diffraction geometry and characteristic properties of the sample.

Figures

Figures reproduced from arXiv: 2412.04197 by the authors.

Figure 1
Figure 1. FIG. 1. Types of diffraction images and patterns available from G [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. The G [PITH_FULL_IMAGE:figures/full_fig_p013_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. Observed spot intensities and the contributions of individual Bragg reflections according to G [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: FIG. 4. Autocorrelation function (ACF) and the intensity profile ( [PITH_FULL_IMAGE:figures/full_fig_p022_4.png]
Figure 5
Figure 5. Figure 5: FIG. 5. Error due to marginalization of [PITH_FULL_IMAGE:figures/full_fig_p026_5.png]
Figure 6
Figure 6. Figure 6: FIG. 6. Ewald sphere construction to predict the position and intensity of a reflection belonging to an RLP [PITH_FULL_IMAGE:figures/full_fig_p030_6.png]
Figure 7
Figure 7. Figure 7: FIG. 7. Importance of the orientation spread (mosaicity). [PITH_FULL_IMAGE:figures/full_fig_p031_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

69 extracted references · 67 canonical work pages

  1. [1]

    author author W. Kabsch ,\ title title Automatic indexing of rotation diffraction patterns , \ 10.1107/S0021889887009737 journal journal Journal of Applied Crystallography \ volume 21 ,\ pages 67--72 ( year 1988 ) NoStop

    work page doi:10.1107/s0021889887009737 1988

  2. [2]

    author author A. J. M. \ Duisenberg ,\ title title Indexing in single-crystal diffractometry with an obstinate list of reflections , \ 10.1107/S0021889891010634 journal journal Journal of Applied Crystallography \ volume 25 ,\ pages 92--96 ( year 1992 ) NoStop

    work page doi:10.1107/s0021889891010634

  3. [3]

    author author W. Kabsch ,\ title title Automatic processing of rotation diffraction data from crystals of initially unknown symmetry and cell constants , \ 10.1107/S0021889893005588 journal journal Journal of Applied Crystallography \ volume 26 ,\ pages 795--800 ( year 1993 ) NoStop

    work page doi:10.1107/s0021889893005588

  4. [4]

    author author H. R. \ Powell ,\ title title The rossmann fourier autoindexing algorithm in mosflm. \ @noop journal journal Acta Crystallogr D Biol Crystallogr \ volume 55 ,\ pages 1690--1695 ( year 1999 ) NoStop

    work page 1999

  5. [5]

    author author E. M. \ Lauridsen , author S. Schmidt , author R. M. \ Suter , \ and\ author H. F. \ Poulsen ,\ title title Tracking: a method for structural characterization of grains in powders or polycrystals , \ 10.1107/S0021889801014170 journal journal Journal of Applied Crystallography \ volume 34 ,\ pages 744--750 ( year 2001 ) NoStop

    work page doi:10.1107/s0021889801014170

  6. [6]

    author author N. K. \ Sauter , author R. W. \ Grosse-Kunstleve , \ and\ author P. D. \ Adams ,\ title title Robust indexing for automatic data collection , \ 10.1107/S0021889804005874 journal journal Journal of Applied Crystallography \ volume 37 ,\ pages 399--409 ( year 2004 ) NoStop

    work page doi:10.1107/s0021889804005874

  7. [7]

    Kabsch ,\ title title Xds , \ 10.1107/S0907444909047337 journal journal Acta Crystallographica Section D \ volume 66 ,\ pages 125--132 ( year 2010 ) NoStop

    author author W. Kabsch ,\ title title Xds , \ 10.1107/S0907444909047337 journal journal Acta Crystallographica Section D \ volume 66 ,\ pages 125--132 ( year 2010 ) NoStop

    work page doi:10.1107/s0907444909047337 2010

  8. [8]

    author author H. R. \ Powell , author O. Johnson , \ and\ author A. G. W. \ Leslie ,\ title title Autoindexing diffraction images with imosflm , \ 10.1107/S0907444912048524 journal journal Acta Crystallographica Section D \ volume 69 ,\ pages 1195--1203 ( year 2013 ) NoStop

    work page doi:10.1107/s0907444912048524

Show all 69 references
  1. [9]

    author author R. J. \ Gildea , author D. G. \ Waterman , author J. M. \ Parkhurst , author D. Axford , author G. Sutton , author D. I. \ Stuart , author N. K. \ Sauter , author G. Evans , \ and\ author G. Winter ,\ title title New methods for indexing multi-lattice diffraction...

    2014

  2. [10]

    author author N. K. \ Sauter , author J. Hattne , author A. S. \ Brewster , author N. Echols , author P. H. \ Zwart , \ and\ author P. D. \ Adams ,\ title title Improved crystal orientation and physical properties from single-shot xfel stills , \ 10.1107/S1399004714024134 jour...

    doi

  3. [11]

    author author S. Schmidt ,\ title title GrainSpotter : a fast and robust polycrystalline indexing algorithm , \ @noop journal journal Journal of Applied Crystallography \ volume 47 ,\ pages 276--284 ( year 2014 ) NoStop

    2014

  4. [12]

    author author H. M. \ Ginn , author P. Roedig , author A. Kuo , author G. Evans , author N. K. \ Sauter , author O. Ernst , author A. Meents , author H. Mueller-Werkmeister , author R. J. D. \ Miller , \ and\ author D. I. \ Stuart ,\ title title TakeTwo : an indexing algorithm...

    doi

  5. [13]

    author author K. R. \ Beyerlein , author T. A. \ White , author O. Yefanov , author C. Gati , author I. G. \ Kazantsev , author N. F.-G. \ Nielsen , author P. M. \ Larsen , author H. N. \ Chapman , \ and\ author S. Schmidt ,\ title title Felix: an algorithm for indexing multip...

    doi

  6. [14]

    Gevorkov , author O

    author author Y. Gevorkov , author O. Yefanov , author A. Barty , author T. A. \ White , author V. Mariani , author W. Brehm , author A. Tolstikova , author R.-R. \ Grigat , \ and\ author H. N. \ Chapman ,\ title title XGANDALF -- extended gradient descent algorithm for lattic...

    doi

  7. [15]

    Li , author X

    author author C. Li , author X. Li , author R. Kirian , author J. C. H. \ Spence , author H. Liu , \ and\ author N. A. \ Zatsepin ,\ title title SPIND : a reference-based auto-indexing algorithm for sparse serial crystallography data , \ 10.1107/S2052252518014951 journal journ...

    doi

  8. [16]

    Gevorkov , author A

    author author Y. Gevorkov , author A. Barty , author W. Brehm , author T. A. \ White , author A. Tolstikova , author M. O. \ Wiedorn , author A. Meents , author R.-R. \ Grigat , author H. N. \ Chapman , \ and\ author O. Yefanov ,\ title title pinkIndexer -- a universal indexer...

    doi

  9. [17]

    author author T. A. \ White , author R. A. \ Kirian , author A. V. \ Martin , author A. Aquila , author K. Nass , author A. Barty , \ and\ author H. N. \ Chapman ,\ title title Crystfel: a software suite for snapshot serial crystallography , \ 10.1107/S0021889812002312 journal...

    doi

  10. [18]

    Hattne , author N

    author author J. Hattne , author N. Echols , author R. Tran , author J. Kern , author R. J. \ Gildea , author A. S. \ Brewster , author R. Alonso-Mori , author C. Glockner , author J. Hellmich , author H. Laksmono , author R. G. \ Sierra , author B. Lassalle-Kaiser , author A....

    2014

  11. [19]

    author author W. Kabsch ,\ title title Processing of x-ray snapshots from crystals in random orientations , \ 10.1107/S1399004714013534 journal journal Acta Crystallographica Section D \ volume 70 ,\ pages 2204--2216 ( year 2014 ) NoStop

    doi

  12. [20]

    Winter , author D

    author author G. Winter , author D. G. \ Waterman , author J. M. \ Parkhurst , author A. S. \ Brewster , author R. J. \ Gildea , author M. Gerstel , author L. Fuentes-Montero , author M. Vollmar , author T. Michels-Clark , author I. D. \ Young , author N. K. \ Sauter , \ and\ ...

    doi

  13. [21]

    Gemmi , author E

    author author M. Gemmi , author E. Mugnaioli , author T. E. \ Gorelik , author U. Kolb , author L. Palatinus , author P. Boullay , author S. Hovmöller , \ and\ author J. P. \ Abrahams ,\ title title 3d electron diffraction: The nanocrystallography revolution , \ 10.1021/acscen...

    doi

  14. [22]

    Shi , author B

    author author D. Shi , author B. L. \ Nannenga , author M. G. \ Iadanza , \ and\ author T. Gonen ,\ title title Three-dimensional electron crystallography of protein microcrystals , \ 10.7554/eLife.01345 journal journal eLife \ volume 2 ( year 2013 ),\ 10.7554/eLife.01345 NoStop

    doi

  15. [23]

    Palatinus , author V

    author author L. Palatinus , author V. Pet r \' i c ek , \ and\ author C. A. \ Corr \^ e a ,\ title title Structure refinement using precession electron diffraction tomography and dynamical diffraction: theory and implementation , \ 10.1107/S2053273315001266 journal journal Ac...

    doi

  16. [24]

    Smeets , author X

    author author S. Smeets , author X. Zou , \ and\ author W. Wan ,\ title title Serial electron crystallography for structure determination and phase analysis of nanocrystalline materials , \ 10.1107/S1600576718009500 journal journal Journal of Applied Crystallography \ volume 5...

    doi

  17. [25]

    Bücker , author P

    author author R. Bücker , author P. Hogan-Lamarre , \ and\ author R. J. D. \ Miller ,\ title title Serial electron diffraction data processing with diffractem and crystfel , \ 10.3389/fmolb.2021.624264 journal journal Frontiers in Molecular Biosciences \ volume 8 ,\ pages 415 ...

    2021

  18. [26]

    author author B. J. \ Siwick , author J. R. \ Dwyer , author R. E. \ Jordan , \ and\ author R. J. D. \ Miller ,\ title title An atomic-level view of melting using femtosecond electron diffraction , \ 10.1126/science.1090052 journal journal Science \ volume 302 ,\ pages 1382--1...

    doi

  19. [27]

    \ Ruan , author F

    author author C.-Y. \ Ruan , author F. Vigliotti , author V. A. \ Lobastov , author S. Chen , \ and\ author A. H. \ Zewail ,\ title title Ultrafast electron crystallography: Transient structures of molecules, surfaces, and phase transitions , \ 10.1073/pnas.0307302101 journal ...

    doi

  20. [28]

    author author A. H. \ Zewail ,\ title title 4d ultrafast electron diffraction, crystallography, and microscopy , \ https://doi.org/10.1146/annurev.physchem.57.032905.104748 journal journal Annual Review of Physical Chemistry \ volume 57 ,\ pages 65--103 ( year 2006 ) NoStop

  21. [29]

    Sciaini \ and\ author R

    author author G. Sciaini \ and\ author R. J. D. \ Miller ,\ title title Femtosecond electron diffraction: heralding the era of atomically resolved dynamics , \ @noop journal journal Reports on Progress in Physics \ volume 74 ,\ pages 096101 ( year 2011 ) NoStop

    2011

  22. [30]

    author author R. J. D. \ Miller ,\ title title Femtosecond crystallography with ultrabright electrons and x-rays: Capturing chemistry in action , \ 10.1126/science.1248488 journal journal Science \ volume 343 ,\ pages 1108--1116 ( year 2014 ) NoStop

    doi

  23. [31]

    author author A. A. \ Ischenko , author P. M. \ Weber , \ and\ author R. J. D. \ Miller ,\ title title Capturing chemistry in action with electrons: Realization of atomically resolved reaction dynamics , \ 10.1021/acs.chemrev.6b00770 journal journal Chemical Reviews \ volume 1...

    doi

  24. [32]

    author author L. P. R. \ de Cotret , author M. R. \ Otto , author M. J. \ Stern , \ and\ author B. J. \ Siwick ,\ title title An open-source software ecosystem for the interactive exploration of ultrafast electron scattering data , \ @noop journal journal Advanced Structural a...

    2018

  25. [33]

    Yang , author K

    author author J. Yang , author K. Gen , author N. Naruse , author S. Sakakihara , \ and\ author Y. Yoshida ,\ title title A compact ultrafast electron diffractometer with relativistic femtosecond electron pulses , \ 10.3390/qubs4010004 journal journal Quantum Beam Science \ vo...

    doi

  26. [34]

    Lee , author K

    author author Y. Lee , author K. Y. \ Oang , author D. Kim , \ and\ author H. Ihee ,\ title title A comparative review of time-resolved x-ray and electron scattering to probe structural dynamics , \ 10.1063/4.0000249 journal journal Structural Dynamics \ volume 11 ,\ pages 031...

    doi

  27. [35]

    author author R. P. \ Chatelain , author V. Morrison , author C. Godbout , author B. van der Geer , author M. de Loos , \ and\ author B. J. \ Siwick ,\ title title Space-charge effects in ultrafast electron diffraction patterns from single crystals , \ @noop journal journal Ul...

    2012

  28. [36]

    Ishikawa , author S

    author author T. Ishikawa , author S. A. \ Hayes , author S. Keskin , author G. Corthey , author M. Hada , author K. Pichugin , author A. Marx , author J. Hirscht , author K. Shionuma , author K. Onda , author Y. Okimoto , author S.-y. \ Koshihara , author T. Yamamoto , author...

  29. [37]

    author author H. J. \ Bernstein , author J. C. \ Bollinger , author I. D. \ Brown , author S. Gra z ulis , author J. R. \ Hester , author B. McMahon , author N. Spadaccini , author J. D. \ Westbrook , \ and\ author S. P. \ Westrip ,\ title title Specification of the Crystallog...

    doi

  30. [38]

    note A Bragg spot can be a single Bragg reflection or the superposition of several Bragg reflections. Stop

  31. [39]

    author author R Core Team ,\ https://www.R-project.org/ title R: A Language and Environment for Statistical Computing ,\ organization R Foundation for Statistical Computing ,\ address Vienna, Austria ( year 2020 ) NoStop

    2020

  32. [40]

    author author J. Rustom, UnaMesa Association ,\ @noop title TiddlyWiki, a unique non-linear notebook for capturing, organising and sharing complex information , \ howpublished https://tiddlywiki.com ( year TiddlyWiki version 5.2.3, 2022 ) NoStop

    2022

  33. [41]

    note At the beginning of a project, when the orientation of the crystal is not yet established, overestimating the mosaicity can actually be beneficial for NLS fitting. Stop

  34. [42]

    Thus, potential complications due to improper rotations (which change the handedness) are excluded

    note GridScan does by design not exploit crystal symmetry to minimize the region of three-dimensional rotation space to be scanned. Thus, potential complications due to improper rotations (which change the handedness) are excluded. On the other hand, symmetry-equivalent orient...

  35. [43]

    note Note that ( , , ) are not Euler angles. Stop

  36. [44]

    author author K. M. \ Gorski , author E. Hivon , author A. J. \ Banday , author B. D. \ Wandelt , author F. K. \ Hansen , author M. Reinecke , \ and\ author M. Bartelmann ,\ title title HEALPix: A Framework for high-resolution discretization and fast analysis of data distribut...

    2005

  37. [45]

    note HEALPix sampling of the unit sphere is most efficient in the sense that sampling with a set of n grid points is the most uniform sampling possible with n points provided that the same applies for sampling with any larger set obtained by adding more grid points. Stop

  38. [46]

    note A Cartesian coordinate system with basis vectors e _x, e _y, e _z fixed in the crystal lattice is defined by a e _x and c ^ e _z . Stop

  39. [47]

    note The detector is assumed to be perpendicular to the incident beam, which travels along the z axis. Stop

  40. [48]

    Normally the default value should not need to be changed

    note Note that the ambit radius r_ a is the same for all observations. Normally the default value should not need to be changed. If the default radius is larger than necessary, it can easily be decreased. It can likewise also be increased, with the consequence that spots with ...

  41. [49]

    note The resolution range and the possibility to exclude certain areas of the detector plane by applying a mask will not be mentioned further below in order not to overload the formal description. Stop

  42. [50]

    contains all observations that are theoretically possible

    note Weight w_2 should not be set to a positive value unless the list of reduced data is complete, i.e. contains all observations that are theoretically possible. If this is not the case, e.g. because reflections are hidden behind the beamstop, a suitable mask should be defined. Stop

  43. [51]

    author author J. M. \ Cowley , author P. Goodman , author B. K. \ Vainshtein , author B. B. \ Zvagin , \ and\ author D. Dorset ,\ title Electron diffraction and electron microscopy in structure determination , \ \ ( publisher International Tables for Crystallography ,\ year 20...

    2006

  44. [52]

    author author D. L. \ Dorset ,\ title title Direct methods and refinement in electron and x-ray crystallography – diketopiperazine revisited , \ doi:10.1524/zkri.2010.1198 journal journal Zeitschrift für Kristallographie \ volume 225 ,\ pages 86--93 ( year 2010 ) NoStop

    2010

  45. [53]

    \ Peng , author G

    author author L.-M. \ Peng , author G. Ren , author S. L. \ Dudarev , \ and\ author M. J. \ Whelan ,\ title title Robust parameterization of elastic and absorptive electron atomic scattering factors , \ @noop journal journal Acta Crystallographica Section A \ volume 52 ,\ page...

    1996

  46. [54]

    author author L. M. \ Peng ,\ title title Electron scattering factors of ions and their parameterization , \ @noop journal journal Acta Crystallographica Section A \ volume 54 ,\ pages 481--485 ( year 1998 ) NoStop

    1998

  47. [55]

    Darwin ,\ title title XCII

    author author C. Darwin ,\ title title XCII. The reflexion of X-rays from imperfect crystals , \ 10.1080/14786442208633940 journal journal The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science \ volume 43 ,\ pages 800--829 ( year 1922 ) NoStop

    doi

  48. [56]

    Brehm , author T

    author author W. Brehm , author T. White , \ and\ author H. N. \ Chapman ,\ title title Crystal diffraction prediction and partiality estimation using Gaussian basis functions , \ 10.1107/S2053273323000682 journal journal Acta Crystallographica Section A \ volume 79 ,\ pages 1...

    doi

  49. [57]

    Variation in cell constants are not taken into account

    note Here, coherently diffracting domains or mosaic blocks are considered to be small but otherwise perfect crystals. Variation in cell constants are not taken into account. This is in line with the original treatment of mosaicity by Darwin (1922) Darwin1922 . Stop

    1922

  50. [58]

    note A rotation vector is an axial vector with length equal to the rotation angle (in radians), and direction parallel to the axis of rotation NoStop

  51. [59]

    note Differences in domain size and other features of the domains that affect their diffracting power are ignored. Stop

  52. [60]

    note To avoid cluttering of symbols, indices distinguishing rotated RLPs from the mean RLP are suppressed. Stop

  53. [61]

    The context and the names of the variables used should remove any ambiguity

    note Distributions and their density functions in various parameterizations are denoted by the same symbol. The context and the names of the variables used should remove any ambiguity. Stop

  54. [62]

    note Rotation angles > 90° need not be taken into account. The limitations of the correction lie in the approximation of the trace of the Ewald sphere in ( _1, _2) space by a straight line within the region of significant density P_ h ( _1, _2) . This approximation is excellen...

  55. [63]

    This is a general result that holds for multivariate and degenerate normal distributions

    note The convolution of two normal distributions with covariance matrices _1 and _2 is a normal distribution with covariance matrix _1 + _2 . This is a general result that holds for multivariate and degenerate normal distributions. If the dimensions of _1 and _2 do not match, ...

  56. [64]

    They are not strictly necessary for indexing with Garfield

    note The fact that some effects of beam divergence and energy bandwidth can be neglected in parameter fitting, but not in simulation of diffraction images, has a corollary: some of the available model parameters can hardly be fitted because the cost function does not strongly ...

  57. [65]

    This seems to be the most natural assumption for samples like thin foils that are curved or corrugated

    note Here, it is assumed that the orientation of the effective shape function, P_ shp follows that of the crystallites. This seems to be the most natural assumption for samples like thin foils that are curved or corrugated. However, it is difficult to make a general statement ...

  58. [66]

    First, the shape function must have a marked anisotropy

    note A clear difference in predicted reflection intensities and positions can only be expected under certain conditions. First, the shape function must have a marked anisotropy. (This could be the standard case for electron diffraction of samples in thin plate-like geometry.) ...

  59. [67]

    This is usually not the case at the beginning of a project

    note Standard modeling is the only method used in the GridScan tool and should always be preferred in GeoFit unless there are clear indications that anisotropy in mosaicity needs to be considered. This is usually not the case at the beginning of a project. By contrast, ANISO m...

  60. [68]

    If inclined, the tilt is defined by two parameters (polar and azimuthal angle), which can be fitted

    note In GeoFit , the user can choose whether the direction of the relrods should always be fixed and parallel to the Z axis, or possibly inclined to the Z axis. If inclined, the tilt is defined by two parameters (polar and azimuthal angle), which can be fitted. Stop

  61. [69]

    note The Garfield software is provided with no guarantee as to the accuracy or the correctness of the results. Stop

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.