Integrity Assessment Report: Vapour–liquid–solid–solid growth of two-dimensional non-layered β-Bi2O3 crystals with high hole mobility (Nature Materials, 2025)

Academic fraud report · Geng Detector

Summary

Verdict: Clear (no substantive fraud indicators identified). This Geng-style integrity report examines a 2025 Nature Materials paper (DOI: 10.1038/s41563-025-02141-w) reporting VLSS growth of 2D non-layered β-Bi2O3 crystals with hole mobility up to 136.6 cm2 V−1 s−1 at 2.4 nm thickness. The authors are Xiong, Xu, Zou, Xu, Yan, Wu, Qian, Song, Qu, Zhao, Gao, Yang, Zhang, Zhang, Wang, Chen & Zeng. Key findings: (1) unusually long review timeline — submitted 2 July 2022, accepted 15 January 2025, published online 7 March 2025 (~2.5 years); flagged as abnormal but not evidence of misconduct; (2) mobility data (60.4 ± 4.8, 130.4 ± 6.2 cm2 V−1 s−1) show physically plausible thickness dependence with non-uniform standard deviations (4.0–6.2), inconsistent with low-effort fabrication; (3) equipment (FEI Titan3 G2 60-300, Keysight B1500A, Bruker Multimode 8) and reagents are conventional; (4) image-based forensic checks limited because only text was extracted — image duplication/splicing not assessable. Overall verdict: clear, with moderate confidence limited by absence of raw image inspection.

Verdict

Clear. Based on the available textual evidence, no substantive indicators of systematic fraud were identified. Confidence is moderate because pixel-level forensic analysis of figures was not possible (only text extracted from the PDF).

Key findings

Abnormally long review timeline: Submission date 2 July 2022; acceptance date 15 January 2025; online publication 7 March 2025. Total review span ~2.5 years, far above the typical 3–9 months for Nature Materials. Unusual, but consistent with extensive revisions or supplementary experiments; not itself evidence of misconduct.
Plausible mobility data distribution: Reported peak field-effect hole mobility of 136.6 cm2 V−1 s−1 at 2.4 nm thickness, with thickness-dependent means of 60.4 ± 4.8 and 130.4 ± 6.2 cm2 V−1 s−1. Standard deviations are non-uniform (range 4.0–6.2) and show a physical trend with thickness, inconsistent with low-level numerical fabrication.
No temporal–equipment inconsistencies: Instruments cited (FEI Titan3 G2 60-300 double spherical-aberration-corrected TEM, Keysight B1500A semiconductor device analyzer, Bruker Multimode 8 AFM) are mature, pre-2022 commercial models. Reagents (acetone, isopropanol from Alfa; Bi2O3 from Sigma-Aldrich) are standard commercial products with no anomalous catalog numbers.
Reference currency is internally consistent: Citations include 2023–2024 references (e.g., ref 12, 14, 46), consistent with updates added during the extended revision period.
Image forensics not performed: Figures 1–4 (including HAADF-STEM images in Fig. 1d/e and FET transfer curves in Fig. 4c) could not be checked for cloning, splicing, or duplication because raw pixel data was unavailable. This remains a residual unassessed risk.

Evidence highlights

DOI: 10.1038/s41563-025-02141-w
Title: *Vapour–liquid–solid–solid growth of two-dimensional non-layered β-Bi2O3 crystals with high hole mobility*
Authors: Yunhai Xiong, Duo Xu, Yousheng Zou, Lili Xu, Yujie Yan, Jianghua Wu, Chen Qian, Xiufeng Song, Kairui Qu, Tong Zhao, Jie Gao, Jialin Yang, Kai Zhang, Shengli Zhang, Peng Wang, Xiang Chen & Haibo Zeng
Journal: Nature Materials
Received: 2 July 2022; Accepted: 15 January 2025; Published online: 7 March 2025
Reported peak mobility: 136.6 cm2 V−1 s−1 (2.4 nm thickness)
Thickness-dependent mobility statistics: 60.4 ± 4.8; 130.4 ± 6.2 cm2 V−1 s−1 (Fig. 4d)

Notes

This assessment relies solely on text extracted from the PDF; pixel-level forensic analysis (e.g., ImageJ/Forensic plug-ins, Error Level Analysis) of HAADF-STEM and FET figures was not possible.
The 2.5-year review timeline is exceptional but does not, by itself, indicate misconduct; it may reflect major revisions, additional DFT calculations, or detailed VLSS kinetic studies.
Numerical values (e.g., 136.6, 55.8) are within the range of plausible experimental scatter and are not statistically uniform — a positive (though not conclusive) signal.
No further action is recommended on the basis of current evidence. If high-resolution original images become available, a follow-up blind image-forensic pass on Figs. 1d/e and 4c is advised.