{"id":"dd3c0f38-d167-43a6-8ef5-587f5557e9ae","arxiv_id":"2412.21005","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Electron ptychography is the most robust phase retrieval method for nanoscale magnetic structures, while automatic differentiation struggles on an irregular nanowire.","lead":"Three electron-microscopy tricks for recovering the magnetic phase of nanoscale magnets are compared, and electron ptychography from scanning transmission electron microscopy comes out most consistent on a nickel-iron nanowire. The other methods work in some cases but are noisier or fail on irregular samples.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"ePIE 'most robust' claim rests on an estimated 11 µm defocus with no probe validation; a defocus error propagates into the phase gradient and the fitted Ms.","rationale":"The reader's weakest_assumption identifies the same load-bearing point: ePIE assumes a known probe at an estimated defocus of 11 µm, and an error in that estimate propagates into the total phase, the magnetic phase gradient, and the extracted saturation magnetization. My independent read of the strongest claim confirms that this is the least secure link in the paper's central argument. The simulations provide some internal support for ePIE's qualitative robustness, and the open data and code commitments are helpful, but the experimental comparison that drives the concluding ranking is not calibrated against any independent measure of probe or defocus. The missing sensitivity analysis and the unexplained sign reversal of the magnetization direction in ePIE further weaken the quantitative magnetic claims. None of this overturns the qualitative ranking, so the reader's CONDITIONAL verdict remains appropriate. A single controlled defocus sweep test would settle whether the concern actually lands, because it directly probes how much of the reported phase accuracy and Ms value depends on the estimated defocus parameter.","tokens_in":14149,"tokens_out":6789,"duration_ms":77008,"concrete_test":"Re-run py4DSTEM ePIE on the same experimental 4D-STEM dataset with defocus values of 5, 8, 11, 14, and 20 µm, keeping all other parameters fixed, and, if available, reconstruct the probe from a vacuum region. Measure the total phase gradient across the nanowire and the magnetic phase difference outside the nanowire for each defocus. If the magnetic phase difference varies by more than the claimed 0.106 rad agreement margin, or if the fitted Ms changes by more than a few percent, then the ePIE 'phase accuracy' claim is not robust to the stated 11 µm estimate. Report the resulting Ms for each defocus value.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central ranking that ePIE is 'most robust in retrieving the object's phase with the highest spatial resolution and phase accuracy, especially as it pertains to magnetic contributions' is not secure because the experimental ePIE reconstruction uses an estimated, not measured, defocus of 11 µm (Methods, Section 2.2.2) and a single-slice multiplicative object model. In ptychography, an incorrect probe defocus can be partially absorbed into the reconstructed object as a low-order phase error, and the magnetic signal is extracted from the gradient of the total phase outside the nanowire. A small probe-phase error can therefore shift the stray-field distribution, alter the reported phase differences (ePIE: 2.096 rad versus simulation: 1.990 rad), and change the fitted saturation magnetization (478.8×10^3 A/m, Discussion). The paper reports no sensitivity analysis, no independent probe calibration, and no uncertainty on this Ms value. The comparison with OAH is further complicated by the reported reversal of the magnetization direction in ePIE, which is attributed to sample handling but is not quantitatively reconciled. Thus the headline claim about ePIE's magnetic phase accuracy is contingent on an unverified reconstruction parameter. This is a load-bearing weakness, not merely a missing error bar.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript compares three phase-retrieval methods—TIE, RMAD, and ePIE—for Lorentz TEM and Lorentz 4D-STEM imaging of magnetic nanostructures. It first benchmarks the methods on simulated images of Permalloy nano-islands using SSIM and line profiles, then applies OAH, RMAD, and ePIE to experimental data from a 36 nm diameter NiFe nanowire, comparing the reconstructed total phase and phase gradient against a micromagnetic simulation. From this comparison, the authors report a saturation magnetization of 478.8 × 10^3 A/m for the nanowire and conclude that ePIE is the most robust method in terms of spatial resolution and phase accuracy, especially for magnetic contributions.","tokens_in":14325,"tokens_out":3022,"duration_ms":30213,"significance":"If the conclusions are supported, the paper would provide a useful practical comparison of widely used phase-retrieval tools for nanoscale magnetic imaging, with potential impact on how researchers choose between LTEM and Ltz-4D-STEM methods. The manuscript benefits from using open-source software (PyLorentz, ADLTEM, py4DSTEM), publicly archived raw data, and a clear statement of experimental parameters. However, the significance is currently limited by unresolved quantitative issues: the key experimental ePIE reconstruction uses an estimated defocus rather than a measured one, the saturation magnetization is a fitted simulation parameter without an uncertainty estimate, and the central claim that ePIE is 'most robust' is not fully supported by the paper's own simulation metrics.","major_comments":[{"comment":"The ePIE reconstruction of the NiFe nanowire relies on a defocus value that is only estimated ('Defocus of the dataset is estimated to be 11 µm'). In single-slice ptychography, an incorrect probe defocus can be partially absorbed into the reconstructed object as low-order phase errors, which then propagate directly into the total phase gradient and the stray-field analysis. Since the paper's headline claim about ePIE's magnetic phase accuracy depends on this unmeasured parameter, the authors should provide an independent probe calibration, a defocus sensitivity analysis, or a quantitative demonstration that the extracted phase differences (2.096 rad versus 1.990 rad from simulation) are stable over a plausible range of defocus values.","section":"§2.2.2, Methods"},{"comment":"The reported saturation magnetization of 478.8 × 10^3 A/m is presented as an experimental finding, but according to the text it is obtained by running micromagnetic simulations and matching their phase to the experimental phase. This is a fitted input parameter, not an independent prediction, and the fitting procedure, the parameter ranges explored, and any uncertainty estimate are not reported. As a fitted value, it cannot serve as corroboration of the ePIE reconstruction. The authors should either clearly label this as a model-dependent fit with error bars derived from the reconstruction uncertainties, or remove it from the list of validated findings.","section":"§3, Discussion"},{"comment":"The claim that ePIE 'proves to be the most robust' in the simulated comparison is not supported by the manuscript's own quantitative metric: RMAD yields a higher SSIM than ePIE for both the total phase (94.87% vs 93.36%) and the phase gradient (98.995% vs 98.24%). The text justifies the preference for ePIE using the line profiles in Fig. 2 and the weak-phase variations, but that argument is not reconciled with the global SSIM scores. The authors should either present a consistent quantitative ranking or weaken the claim to say that ePIE and RMAD are comparably accurate in simulations, with ePIE showing specific advantages in certain line-profile features.","section":"§2.1, Fig. 1 and §3, Discussion"},{"comment":"Using the OAH reconstruction as the 'experimental ground truth basis' is problematic for a comparison whose goal is to assess phase accuracy, because OAH itself contains noise and distortions that the authors acknowledge. In addition, the ePIE reconstruction shows a reversed magnetization direction relative to both OAH and the simulation, which is attributed to sample handling but is not quantitatively reconciled. If the ePIE phase is sign-flipped for comparison, the reported phase differences need to be recomputed with the sign convention stated; if not, the quoted agreement between ePIE and the simulation (difference of 0.106 rad) is not a direct comparison. Please clarify the sign handling and report the sensitivity of the quantitative comparisons to the OAH noise level.","section":"§2.2.3, Fig. 5"}],"minor_comments":[{"comment":"There are multiple typographical errors and duplicated words, e.g., 'Morevover' in the Introduction, 'we explored we explored' in §2.2, 'betwen' in §2.2.3, and inconsistent use of 'SFIG' versus 'Fig.' for supplementary figures.","section":"Throughout"},{"comment":"The caption references subfigures (d), (e), (f) as total phase reconstructions and (g), (h), (i) as phase gradients, but the main text refers to 'Fig. 1(d)' as a line plot and 'Fig. 1(e)' as the TIE reconstruction, which is confusing. Please renumber the panels or correct the cross-references.","section":"Fig. 1 caption"},{"comment":"The statement that 'we note that these will vary based on individual datasets' after listing ePIE reconstruction parameters is vague; it would be more useful to state explicitly which parameters are robust and which are dataset-dependent.","section":"Methods, §5"},{"comment":"Several references are incomplete or malformed, for example Ref. 18 ends with 'zhou (2021)' instead of the full citation, and Ref. 21 lacks author and venue information.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses a timely methodological comparison and provides open data, but the central claims currently rest on an unverified defocus estimate and a circularly defined saturation magnetization. These are fixable with additional analysis, but as written they are load-bearing and would need to be addressed before publication. I would not reject the manuscript outright, as the qualitative ranking is plausible and the experimental dataset is valuable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know two things about this paper before deciding what to do with it. First, it is a genuinely useful comparative study: it puts TIE, RMAD, and ePIE through the same paces on simulated Permalloy islands with known ground truth, then applies the same methods plus OAH to a real NiFe nanowire. The simulation section is clean, the SSIM-based ranking is straightforward, and the conclusion that ePIE is the most robust for these samples is plausible and reasonably supported. Second, the paper's most quantitative experimental claim—a saturation magnetization of 478.8 × 10^3 A/m for the nanowire—is a fitted simulation parameter, not an independent measurement, and the paper never describes how the fit was done or attaches an uncertainty to it. That is the softest spot, and it is a real one.\n\nWhat is new and good: the head-to-head comparison on nanoscale magnetic structures is, as far as I know, not in the literature; the authors use open-source tools throughout, post raw data on Zenodo, and say they will share processing code; and the stray-field gradient visualization around the nanowire tip is a nice demonstration. The line profiles and SSIM numbers in the simulated section give the ePIE ranking some quantitative legs.\n\nThe soft spots, in rough order of importance. The Ms value is fit to make simulated phase match experimental phase, so it is circular as a \"finding\"—it needs a described forward model, a sensitivity scan over defocus and other parameters, and an error bar. Relatedly, the experimental ePIE reconstruction uses an estimated 11 µm defocus with no probe calibration or sensitivity analysis; in ptychography, probe-phase errors can leak into the reconstructed object phase and shift the gradient you then interpret as magnetic. The single-slice weak-phase model for a 36 nm NiFe nanowire is also optimistic, and the paper does not test it. I do not think these issues overturn the ePIE ranking—the simulated section, where ground truth is known, already points that way—but they do undercut the quantitative phase-accuracy claims. Also, using OAH as the experimental ground truth is a bit shaky given its own noise artifacts, and the magnetization reversal seen in ePIE is attributed to sample handling but not quantitatively reconciled. The RMAD defocus-subset selection looks post hoc, though that is minor since the main RMAD limitation is visible in the images.\n\nWho is this for: electron microscopists working on magnetic nanostructures who want a practical map of which phase retrieval method to choose. It deserves a serious referee, but the referee should push for a real Ms extraction protocol, uncertainty analysis, and ideally a defocus sensitivity check. I would not desk-reject it.","headline":"A useful head-to-head benchmark of phase retrieval for magnetic nanostructures, with a plausible ePIE ranking but a fitted and unquantified saturation magnetization that needs work.","tokens_in":14936,"tokens_out":1857,"would_cite":true,"duration_ms":21041,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Electron ptychography retrieves the total phase of magnetic nanostructures more accurately and at higher resolution than TIE, RMAD, or off-axis holography in this comparison.","keywords":["Lorentz TEM","Lorentz STEM","phase retrieval","magnetic materials","transport-of-intensity equation","automatic differentiation","off-axis holography","electron ptychography"],"falsifier":"Reconstruct the same 36 nm NiFe nanowire from the published 4D-STEM dataset with a multislice ptychographic model that accounts for the wire's thickness and compare the magnetic phase shift outside the wire with the single-slice ePIE result; a change larger than about 0.1 radians would show that the single-slice weak-phase assumption, not the algorithm, controls the reported agreement.","tokens_in":13907,"feed_emoji":"🧲","tokens_out":12651,"duration_ms":109502,"temperature":0.7,"pith_summary":"This paper compares four phase-retrieval routes for magnetic nanostructures—transport-of-intensity (TIE), reverse-mode automatic differentiation (RMAD), off-axis holography (OAH), and the extended ptychographic iterative engine (ePIE)—and asks which one gives the most faithful reconstruction of the electron phase shift. On simulated Permalloy island arrays and on an experimental 36 nm NiFe nanowire, ePIE retrieves the total phase with the highest spatial resolution and phase accuracy, especially in the magnetic contribution, while TIE blurs at large defocus and RMAD produces nonphysical stray fields for the irregular nanowire. OAH remains accurate but is noisier and requires a reference vacuum region. The recovered phase-gradient maps expose stray-field and inter-island proximity effects, and yield a nanowire saturation magnetization of $478.8\\times10^3$ A/m, about 57% of bulk Permalloy.","feed_headline":"Electron ptychography recovers magnetic phase defocus methods blur","feed_subtitle":"On a 36 nm NiFe nanowire, its stray-field map matches simulation within 0.1 radians while defocus methods distort it.","key_machinery":"The load-bearing quantity is the total electron phase shift $\\phi_t(\\mathbf{r}_\\perp)=\\phi_e+\\phi_m$, related by the standard Aharonov-Bohm relation to the projected electrostatic potential and the magnetic vector potential; the gradient $\\nabla\\phi_t$ maps the projected magnetic induction, so the stray fields and inter-island interactions appear as faint phase signals outside the sample edges. The four algorithms constitute the machinery: TIE solves the transport-of-intensity equation from through-focal intensities; RMAD fits the same focal series by reverse-mode automatic differentiation, iteratively updating a phase model to match measured intensities; OAH reconstructs the phase from a biprism interference pattern; and ePIE alternates updates of the complex object and illuminating probe using overlapping diffraction patterns recorded in Lorentz-mode 4D-STEM. ePIE with batch size one is described as a limiting case of stochastic gradient descent, and it is the reconstructed phase gradient, not the raw phase, that serves as the quantitative magnetic observable.","core_discovery":"The central claim is that electron ptychography, specifically the extended ptychographic iterative engine (ePIE) implemented as stochastic gradient descent with a batch size of one, is the most robust of the compared methods for quantitative magnetic phase retrieval. In simulations of Permalloy nanoislands, ePIE's reconstructions track the ground truth closely, with structural similarity indices of 93.36% for the total phase and 98.24% for the phase gradient, while TIE loses low-spatial-frequency signal at $\\pm100\\,\\mu$m defocus. In the experimental NiFe nanowire, the ePIE phase gradient matches the micromagnetic simulation in orientation and intensity distribution, with the magnetic phase shift outside the wire differing from simulation by 0.106 radians; OAH differs by 0.278 radians and RMAD's stray field is nonphysical. The paper further reports a measured saturation magnetization of $478.8\\times10^3$ A/m for the 36 nm nanowire and shows that phase-gradient maps can visualize proximity effects between neighboring magnetic islands.","pith_inferences":["An implicit extension of the ePIE result is that the same single-dataset workflow could map magnetic textures in continuous films and devices where OAH cannot be used, and in principle be combined with tilt series for three-dimensional magnetic induction retrieval.","A testable consequence the paper does not draw: if RMAD's failure on the nanowire comes from the single-slice approximation, a multislice forward model applied to the same published through-focal dataset should remove the nonphysical upward stray-field trajectory.","The low saturation magnetization invites a dedicated check: spatially resolved core-loss spectroscopy across the same nanowire would test whether the reduction is a surface oxide shell or an intrinsic size effect.","The reversed magnetization direction between ePIE and OAH, attributed to specimen handling, suggests a controlled before-and-after transfer experiment could quantify how much mechanical handling perturbs the magnetic state of such nanowires."],"forward_implications":["A single 4D-STEM dataset should be sufficient for quantitative magnetic phase maps, removing the reference vacuum region that off-axis holography needs and opening extended or embedded samples to measurement.","RMAD phase retrieval should not be relied on for inhomogeneous, strongly scattering nanostructures unless the forward model accounts for multiple scattering or geometric irregularity.","TIE remains a fast screening tool for magnetic induction, but its resolution penalty at high defocus makes it unsuitable for few-nanometer quantitative work.","The reported $478.8\\times10^3$ A/m saturation magnetization implies that Lorentz phase measurements can detect property changes—oxidation, size effects, or contamination—in individual nanowires.","Phase-gradient maps can serve as a practical way to visualize stray-field coupling in arrays of coupled magnetic islands, relevant to artificial spin ice and related systems."],"supporting_citations":[{"why":"It provides the Aharonov-Bohm identity connecting the retrieved total phase to the electrostatic and magnetic vector potentials.","marker":"[19]"},{"why":"It supplies the simulation and TIE implementation used to generate ground-truth phases and through-focal image stacks.","marker":"[18]"},{"why":"It introduces the reverse-mode automatic differentiation method that the paper benchmarks and finds lacking on inhomogeneous nanowires.","marker":"[25]"},{"why":"It defines the extended ptychographic iterative engine, the algorithm the paper identifies as most robust.","marker":"[29]"},{"why":"It provides the 4D-STEM analysis suite used for ePIE and tilt-corrected bright-field reconstructions.","marker":"[32]"},{"why":"It supplies the iterative phase-retrieval framework and parallax implementation used in the supplementary comparisons.","marker":"[36]"},{"why":"It supplies the micromagnetic simulation engine that generates the magnetization states used as ground truth.","marker":"[37]"},{"why":"It defines the structural-similarity metric used to score reconstruction fidelity.","marker":"[38]"}],"fun_headline_variants":["Ptychography reveals magnetic stray fields defocus methods miss","ePIE ptychography outperforms TIE and holography for magnetic phase","Stray-field maps from ptychography beat defocus methods by 0.1 rad","For magnetic nanowires, ePIE beats TIE and holography in phase retrieval","Ptychography matches simulation within 0.1 rad on 36 nm wire"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The numbers assume the nanowire is thin enough that the electron wave changes in one flat slice and that the microscope's probe and focus are known exactly; a strongly scattering, irregular 36 nm wire can break that assumption.","fun_headline_variants_meta":{"raw":{"variants":["Ptychography reveals magnetic stray fields defocus methods miss","ePIE ptychography outperforms TIE and holography for magnetic phase","Stray-field maps from ptychography beat defocus methods by 0.1 rad","For magnetic nanowires, ePIE beats TIE and holography in phase retrieval","Ptychography matches simulation within 0.1 rad on 36 nm wire"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000828,"raw_usage":{"total_tokens":3651,"prompt_tokens":1013,"completion_tokens":2638,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":629,"completion_tokens_details":{"reasoning_tokens":2532}},"tokens_in":629,"tokens_out":2638,"duration_ms":17885,"temperature":1.0,"reasoning_tokens":2532,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T23:04:19.399666+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Reconstruct the same 36 nm NiFe nanowire from the published 4D-STEM dataset with a multislice ptychographic model that accounts for the wire's thickness and compare the magnetic phase shift outside the wire with the single-slice ePIE result; a change larger than about 0.1 radians would show that the single-slice weak-phase assumption, not the algorithm, controls the reported agreement.","supporting_citations":[],"review_version":1}