{"id":"e9e411ac-ccf4-46d3-876f-5c99caea1fbe","arxiv_id":"2607.16439","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"STM-based metrology on nominally Si(557) wafers finds a triple-step period of 18b ~ 5.99 nm and identifies the surface as Si(8811), contradicting the previously claimed Si(557) (17b) and Si(7710) (16b) staircases.","lead":"Atomically resolved STM shows that the triple-step staircase formed on nominally Si(557) wafers has a period of 18 atomic rows (5.99 nm), identifying the surface as Si(8811) — a third candidate structure distinct from the Si(557) and Si(7710) assignments in the literature. The result matters because Si(557) wafers are standard templates for atomic-chain and nanowire experiments that assumed a ~5% different step period.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The Si(8811) assignment hinges on a global affine STM correction with unreported residuals; a quantitative LEED periodicity measurement would settle whether 18b is real or a drift artifact.","rationale":"The reader's weakest assumption correctly identifies the Achilles heel: the global affine correction. I agree with that assessment, and the LEED test directly addresses it because it measures the periodicity in reciprocal space without the STM drift uncertainty. The paper has independent strengths—multiple surface areas, both 7×7 and 5×5 calibrations, honest admission of model limitations—but these do not remove the calibration concern. The word 'definitively' in the Introduction is not supported by the reported precision and the observed 17b/19b fluctuations. A CONDITIONAL verdict is appropriate until either the correction residuals are quantified and shown to be small, or an independent method such as LEED confirms the 18b period. My additional emphasis on LEED does not change the reader's verdict, so no adjustment is needed.","tokens_in":16706,"tokens_out":3901,"duration_ms":35108,"concrete_test":"Acquire (or use the existing Fig. 3) LEED pattern at fixed beam energy, measure the spacing between the sharp periodic spots along the 7×7 rows, and convert to a real-space period. The spot spacing on the LEED pattern is set by the step array periodicity independent of STM drift. If the derived period is 18b ≈5.99 nm (within LEED accuracy), the STM assignment is confirmed; if it is 17b or 16b, the STM-based conclusion is a drift artifact. A SPA-LEED spot-profile measurement across the specular beam would give the required precision.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim (Introduction; Figs. 4c, 5c, 8c, 10c) that the triple-step periodicity is L=18b, distinguishing Si(8811) from Si(557) (17b) and Si(7710) (16b), depends entirely on distances measured in D(x,y) maps corrected by a single global affine transformation forced onto the 7×7 or 5×5 adatom lattice. The supplementary figures (S1, S2) state that the ideal lattice coincides with experimental features for only one terrace in the middle of the map; no correction residuals are given for other terraces or step regions. Room-temperature STM drift is time-dependent and not strictly linear, so a global affine transform cannot compensate for spatially varying distortion. The step-region distances (the 'terrace + triple step' width) are the very quantity used to determine L; if the effective scale in the step regions differs from the terrace scale by ≥6%, 18b would be misread from a true 17b period. The paper itself reports 17b and 19b spacings in the same corrected maps, indicating local deviations; without a drift-error budget, the word 'definitively' (Introduction) overstates the case. This is not a fatal flaw, but it makes the Si(8811) assignment conditional on the correction's accuracy.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports STM and LEED studies of periodic triple-step arrays prepared on nominally Si(557) wafers. After applying a global affine correction to differential D(x,y) maps using the 7x7 or 5x5 adatom lattices on Si(111) terraces, the authors find a preferential period of L = 18b = 5.99 nm in projection onto the terrace plane, corresponding to a Si(8811) surface orientation with an 8.93 deg miscut. This is contrasted with earlier assignments of this template to Si(557) (17b) and Si(7710) (16b). The authors further report that the 18b periodicity can be maintained with different terrace widths (half 7x7 or 5x5 unit cells, with or without an additional adatom row) and different triple-step internal structures, and they propose four schematic 'terrace + triple step' models. The same corrected maps also show 17b, 19b, and occasionally 20b spacings, so the Conclusion states L = (18 ± 1)b.","tokens_in":17009,"tokens_out":9949,"duration_ms":98831,"significance":"If the 18b assignment is correct, the result is significant: many studies of Si(557)-derived templates assume a 17b (5.65 nm) or 16b (5.33 nm) periodicity, and a revision to an 18b / Si(8811) staircase would affect the interpretation of nanowire and atomic-chain experiments on this surface. The strongest evidence is the internal cross-check between 7x7- and 5x5-calibrated maps from different surface areas, which both yield 18b. The paper also benefits from the use of DOG-processed maps to resolve features on strongly corrugated step regions and from the explicit admission that multiple 'terrace + triple step' configurations are compatible with the same period. However, the central claim is not yet fully established because the metrological foundation — a single global affine correction with unreported residuals — is not quantified. The manuscript is a careful experimental study, but the periodicity assignment needs additional support before it can be considered definitive.","major_comments":[{"comment":"The load-bearing period assignment L=18b is derived from D(x,y) maps that have been affine-corrected to the 7x7 or 5x5 lattice, but the correction residuals are not reported. The Supplementary captions state that the ideal lattice matches experiment only for one terrace in the middle of the map. Since the measured quantity is the terrace+step width across step regions, a spatially nonuniform scale error of about 6% (the separation between 18b and 17b) cannot be excluded. The presence of 17b and 19b spacings in the same corrected maps (Figs. 4c and 8c) makes this concern concrete. The word 'definitively' in the Introduction is therefore too strong. Please provide the affine parameters, residual maps over the full field of view, and a quantitative error budget; an independent periodicity measurement (e.g., quantitative LEED spot-profile analysis) would be the most direct way to settle the","section":"Results and discussion, Figs. 4c and 8c; Supplementary S1–S2"},{"comment":"The paper simultaneously claims 'definitively corresponds to L=18b' (Introduction) and 'L=(18±1)b' (Conclusion), and the data in Figs. 4c and 8c show individual periods of 17b, 18b, 19b, and 20b. If these are true structural variations, the staircase is not an atomically precise 18b structure but has a statistical preference, and assigning a single Miller index (8 8 11) should be reformulated accordingly. If they are calibration artifacts, the issue in the first major comment becomes decisive. Please provide the period distribution and a statistical analysis from all corrected images, and clarify whether the deviations are attributed to local defects or to measurement uncertainty.","section":"Introduction vs Conclusion"},{"comment":"Reference [33] is cited as 'in preparation (2026)' and is used to support the Fourier-peak suppression confirmation of the periodicity and the comparison with 2D Fourier map calculations. An in-preparation manuscript cannot serve as a verifiable supporting reference. The relevant analysis should be included in the paper or the Supplement, or the citation should be removed and the claim supported by data presented here.","section":"Reference [33]"}],"minor_comments":[{"comment":"Ref. [8] contains a typo: 'Phus.Rev.Lett.' should be 'Phys. Rev. Lett.'. Several figure references in the text use a Cyrillic 'с' (e.g., 'Fig. 4с') instead of the Latin 'c'.","section":"References and text"},{"comment":"The phrase 'schematic model presented in Fig. 11a' appears to be an incorrect cross-reference; the intended model is likely in Fig. 6a.","section":"Fig. S4 caption"},{"comment":"The DOG scale parameters used to construct the D(x,y) maps are not reported. The y-averaged cross-sections of D(x,y) maps are used for quantitative distance determination, so specifying the difference-of-Gaussians parameters (or providing the code) would improve reproducibility. This does not affect the central claim if the reported number is measured in b units from lattice-calibrated maps, but it should still be documented.","section":"Methods / reproducibility"}],"recommendation":"major_revision","confidential_remarks":"This is a potentially valuable experimental paper, but the central periodicity assignment rests on an affine correction whose residuals are not given, and the paper's own 'definitively' (Introduction) conflicts with the reported ±1b spread. I would ask the authors to provide the calibration details, residual statistics, and ideally an independent quantitative LEED or SPA-LEED measurement before publication. The use of an in-preparation reference [33] as supporting evidence should also be corrected."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a careful experimental re-identification of the 'Si(557)' triple-step staircase as Si(8811) with an 18b period. I think the case is plausible and the observation of multiple iso-period terrace-plus-triple-step units is new and worth knowing. The paper is not the last word: the 'definitively' in the Introduction is not earned given the missing drift-error budget and period statistics.\n\nWhat is genuinely good: the STM data are atomically resolved, from several independent areas, and the DOG maps are a sensible tool for a surface where step and terrace facets lie at large angles. The internal cross-check—7x7-corrected and 5x5-corrected maps both giving 18b—is real evidence, not a fitted parameter. The paper is also honest about its own limits: the models are schematic, bias dependence is flagged, and other 18b configurations are admitted. The preparation was careful and the LEED patterns show a regular array, though the spot positions are not quantified.\n\nThe soft spot is exactly where the stress-test focuses. The period is read from distances between terraces in images corrected by a single global affine transform to the 7x7 or 5x5 lattice. The supplementary only shows overlay agreement for one terrace in the middle of each map, not residuals for other terraces or the step regions. Since room-temperature STM drift is time-dependent and not strictly linear, a global affine correction cannot be assumed to kill local distortion in the step regions, and that is where an 18b vs 17b distinction (a 6% difference) happens. The fact that the same maps contain 17b and 19b spacings tells you there is local variability; without a histogram and a drift budget, 'definitively' overstates the case. The Fourier evidence is in preparation by the same group, so it doesn't count yet.\n\nBottom line: this deserves a serious referee. It is a genuine contribution to a two-decade dispute about a widely used template. What I'd ask for in revision: per-image distributions of measured periods, overlay residuals across the whole map, and ideally a quantitative LEED spot-profile analysis. If those come back consistent, the Si(8811) assignment will be solid.","headline":"Plausible but over-claimed STM re-identification of the Si(557) template as Si(8811); the 18b period needs a drift-error budget and period statistics before 'definitively' is earned.","tokens_in":17660,"tokens_out":3735,"would_cite":true,"duration_ms":34935,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["68.37.Ef","68.35.Bd"],"model":"deepseek-v4-flash","headline":"Atomically resolved STM data re-assign the periodic triple-step staircase on nominally Si(557) wafers to a Si(8811) orientation with a period of 18b ≈ 5.99 nm, and show that several distinct step/terrace configurations produce the same peri","keywords":["vicinal silicon surfaces","triple-step staircase","Si(8811)","scanning tunneling microscopy","difference-of-Gaussians","surface reconstruction","step periodicity"],"falsifier":"Measure the staircase with an independent, drift-free probe such as grazing-incidence X-ray diffraction or spot-profile-analysis LEED: a step period of 5.65 nm (17b) or 5.33 nm (16b) rather than 5.99 nm (18b) would refute the Si(8811) assignment.","tokens_in":16498,"feed_emoji":"🔬","tokens_out":6825,"duration_ms":55610,"temperature":0.7,"pith_summary":"On nominally Si(557) wafers, the authors prepared a periodic triple-step staircase and measured its period using atomically resolved scanning tunneling microscopy with affine correction to the ideal terrace lattices. The corrected differential maps place the step-to-step distance at 18b ≈ 5.99 nm in the terrace-plane projection, where b = 0.333 nm is the Si(111)1×1 atomic-row spacing — an orientation that corresponds to Si(8811), not the previously assigned Si(557) (17b) or Si(7710) (16b). The same 18b period persists across several distinct atomic configurations: terraces carrying 7×7, 5×5, or 9×9 reconstruction fragments, and triple steps built from a monatomic step, a narrow mini-terrace, and a double step. The paper concludes that step regularity on these templates is not fixed by a unique reconstruction, and that defects and step-step interactions likely stabilize the array.","feed_headline":"Silicon staircase measured at 18-row period, not 17","feed_subtitle":"Lattice-corrected STM maps re-assign the vicinal surface to Si(8811) and show multiple step structures share one period.","key_machinery":"The key measurement device is the difference-of-Gaussians (DOG) differential map, which enhances atomic features on steeply sloped facets, combined with an affine warp that forces the terrace adatom lattice to its ideal 7×7 (or 5×5) geometry. This calibration lets the distance between equivalent terrace features be read directly in units of b = 0.333 nm, the spacing between Si(111)1×1 zigzag atomic rows; the distinction between 18b, 17b, and 16b — a 6–11% spread — is made on the y-averaged cross-sections of the corrected maps.","core_discovery":"The atomically resolved STM data, after affine correction to the ideal 7×7 and 5×5 lattices on the terraces, show that the periodic triple-step staircase on nominally Si(557) wafers has a period of 18b ≈ 5.99 nm in projection onto the terrace plane, corresponding to the Si(8811) orientation with an 8.93° miscut angle. This contradicts the earlier Si(557) (17b) and Si(7710) (16b) assignments. The step period is maintained even though the steps are not a single crystallographic facet: each triple step is a monatomic step, a narrow Si(111) mini-terrace, and a double step, and the terraces can carry 7×7, 5×5, or 9×9 reconstruction fragments with different widths. Four schematic 'terrace + triple","pith_inferences":["Because the 18b vs 17b/16b distinction is a ~6–11% length difference, a drift-free structural probe (e.g., grazing-incidence X-ray diffraction or spot-profile LEED) applied to the same wafer could settle the orientation assignment independently; if it returns 5.65 nm or 5.33 nm, the corrected-map analysis is over-correcting the step regions.","The paper's support for defect-stabilized step ordering suggests a testable lever: controlled dosing of sub-monolayer adsorbates (or doping) on a clean vicinal wafer might tune the step periodicity continuously or switch between 16b, 17b, and 18b staircases, extending the template toolkit.","If multiple step configurations truly share the same period, then their energy differences are small; the paper's schematic models could be ranked by a total-energy calculation that relaxes the full 'terrace+step' unit including step-edge dimers, which would predict which configuration is most abundant."],"forward_implications":["If the assignment holds, the triple-step template repeatedly used as 'Si(557)' is actually a Si(8811) staircase for this preparation; studies that interpreted electronic, transport, or growth results on this template should check whether their conclusions depend on the now-replaced facet index.","The same periodicity can be realized by multiple distinct atomic step configurations, meaning atomically precise periodicity is not evidence for a single atomic structure; structural models of vicinal Si templates must be constrained by more than the measured period.","The local orientation of the staircase (8.93° miscut) deviates from the wafer's nominal 9.45° miscut, so the global miscut angle alone does not determine which Si(hhm) facet forms; preparation history and step-defect densities select among competing periodicities.","The observed occasional ±b deviations (17b, 19b, rarely 20b) around the dominant 18b period give a quantitative measure of staircase regularity on the micron scale, useful as a length-standard tolerance."],"fun_headline_variants":["Silicon staircase is Si(8811), not Si(557)","18-row period redefines silicon vicinal surface","Triple steps on silicon: period 18b, not 17","STM corrects silicon surface assignment to 8811","Why Si(557) is really Si(8811): 18-row steps"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The step-period measurement assumes that a single affine correction, calibrated on the terrace adatom lattice, also removes distortion in the step regions where the periodicity is measured; if scanner drift varies across the image, the 18b reading could shift to 17b or 16b.","fun_headline_variants_meta":{"raw":{"variants":["Silicon staircase is Si(8811), not Si(557)","18-row period redefines silicon vicinal surface","Triple steps on silicon: period 18b, not 17","STM corrects silicon surface assignment to 8811","Why Si(557) is really Si(8811): 18-row steps"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000361,"raw_usage":{"total_tokens":1782,"prompt_tokens":733,"completion_tokens":1049,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":477,"completion_tokens_details":{"reasoning_tokens":960}},"tokens_in":477,"tokens_out":1049,"duration_ms":7777,"temperature":1.0,"reasoning_tokens":960,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T20:57:33.269164+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the staircase with an independent, drift-free probe such as grazing-incidence X-ray diffraction or spot-profile-analysis LEED: a step period of 5.65 nm (17b) or 5.33 nm (16b) rather than 5.99 nm (18b) would refute the Si(8811) assignment.","supporting_citations":[],"review_version":1}