Geng Integrity Review: Vapour–liquid–solid–solid growth of two-dimensional non-layered β-Bi₂O₃ crystals with high hole mobility

Academic fraud report · Geng Detector

Summary

This Geng-style integrity report flags the Nature Materials paper (DOI: 10.1038/s41563-025-02141-w, Xiong et al., Vol. 24, May 2025) as highly suspicious, based on text-only forensic review. Verdict: 🟠 highly suspicious — not a definitive fraud ruling. Key issues include (1) arithmetic inconsistency in an EDS composition claim (40% O + 35% Bi + 15% Cl + 5% Na = 95%, not 100%); (2) a stated 'fourfold' mobility increase at 80 K that mathematically corresponds to ~2.66× when comparing to the room-temperature value of 136.6 cm² V⁻¹ s⁻¹; (3) a dimensional-analysis error in the driving-force expression F = ∫(σ₁−σ₂) dA, where σ·A has units of J·m, not N; (4) an unusually long 30-month review timeline and a 2024 grant acknowledgement cited on a 2022-submitted paper; and (5) a typographical anomaly 'on/oϕ' in the abstract. No image-based checks were possible. Confidence is moderate; findings require verification with raw data.

Verdict

🟠 Highly suspicious. Multiple internal inconsistencies in numerical claims and dimensional reasoning, plus an anomalous publication timeline. Definitive fraud determination requires access to raw data, original figures, and an institutional investigation.

Key findings

Finding 1 (🔴 Arithmetic / data self-consistency): Page 689 — VLSS growth mechanism section quotes EDS droplet composition as "40% O, 35% Bi, 15% Cl and 5% Na". Sum = 95%, not 100%. Unaccounted 5% is unexplained.
Finding 2 (🟠 Arithmetic / text–figure mismatch): Page 694 — text claims "a fourfold increase" in hole mobility at 80 K with value 364.6 ± 35.4 cm² V⁻¹ s⁻¹, versus room-temperature maximum of 136.6 cm² V⁻¹ s⁻¹. Actual ratio = 364.6 / 136.6 ≈ 2.67×, not 4×.
Finding 3 (🟠 Dimensional / methodological): Page 692, Fig. 2i — driving-force expression F = ∫(σ₁−σ₂) dA. With σ in J/m² (interface energy) and dA in m², the integrand has units of J·m, not N. Force should be expressed as F = −dG/dx or σ·L, not as energy×area.
Finding 4 (🟡 Timeline / provenance anomaly): Page 688 — submitted 2 July 2022; accepted 15 January 2025 (≈30 months review). Page 696 — Acknowledgements cite grant 2024YFB3612400, a 2024-awarded funding supporting a 2022-submitted manuscript, consistent with late-stage revision additions.
Finding 5 (🟡 Typographical anomaly): Page 688 Abstract — "on/oϕ current ratio" appears in place of the standard "on/off ratio"; ϕ is a Greek letter, likely a font/formula-editor artifact.
Finding 6 (ℹ️ Image evidence not assessable): Text-only report; HAADF-STEM (Fig. 1), optical micrographs (Fig. 2), Raman mapping in Supplementary, and AFM files were not available. Independent ImageJ/clone-tool inspection is recommended.

Evidence highlights

EDS composition sum: 40 + 35 + 15 + 5 = 95 (Page 689).
Mobility values cited: 136.6 cm² V⁻¹ s⁻¹ (room-temperature maximum, Page 694 / Fig. 4c caption) vs. 364.6 ± 35.4 cm² V⁻¹ s⁻¹ at 80 K (Page 694). Computed ratio ≈ 2.67.
Formula unit mismatch: F = ∫(σ₁−σ₂) dA, with σ in J/m² → integrand in J·m ≠ N (Page 692, Fig. 2i).
Dates: Received 2 July 2022; Accepted 15 January 2025 (Page 688).
Grant acknowledgement: National Key R&D Program of China, no. 2024YFB3612400 (Page 696).
DOI: https://doi.org/10.1038/s41563-025-02141-w

Notes

All findings are derived from text content only; no pixel-level or statistical image forensics were performed.
The EDS sum discrepancy, the 'fourfold' vs. ~2.67× mismatch, and the σ·dA dimensional error could in principle stem from copyediting, LaTeX transcription, or genuine confusion rather than deliberate fabrication; they nonetheless indicate insufficient numerical and dimensional checking by authors and reviewers.
A 30-month review window at a top-tier journal is unusual but can occur for legitimate reasons (e.g., extensive revision, reviewer turnover, pandemic-era backlog).
The 'on/oϕ' token is most plausibly a rendering artifact, but its presence in the abstract suggests inadequate proofing.
Recommended next steps: request raw EDS spectra and unprocessed mobility dataset from the authors; raise these specific arithmetic and dimensional points on PubPeer; if accessible, perform ImageJ-based duplication/noise analysis on Supplementary HAADF-STEM, AFM, and Raman maps.
This report is AI-assisted and intended for academic discussion only; final determination of misconduct requires institutional investigation.