Geng Integrity Report: Tetraphenylsilane-Engineered MR-TADF Emitters for Pure-Green OLEDs (Adv. Mater., DOI: 10.1002/adma.72684)

Academic fraud report · Geng Detector

Summary

Verdict: Highly suspicious. The photophysical kinetic data reported in Table 1 are mathematically self-contradictory: the prompt fluorescence lifetimes (τPF) reported for p-DBFSi, m-DBFSi, and m-DBFSi-d (6.7, 6.3, and 5.8 ns, respectively) are inconsistent with the radiative and non-radiative rate constants (kr,S and knr,S) the authors also report. Using the relation τPF = 1/(kr,S + knr,S), the implied lifetimes are 43.7 ns, 27.5 ns, and 26.1 ns, i.e., 4–6× larger than the stated values. In addition, the reverse intersystem crossing rate constants contradict the reported delayed fluorescence lifetimes in a counter-intuitive manner, and the device current efficiency quoted in the text (146.4 cd A−1) disagrees with Table 2 (146.7 cd A−1). Together these findings strongly suggest the kinetic table was constructed or copied without verification. Pixel-level image analysis was not possible from the text-only OCR. Final determination requires an institutional investigation and access to original decay traces.

Verdict

🟠 Highly suspicious. Multiple internal mathematical inconsistencies in the reported photophysical kinetics, together with text–table data mismatches, indicate that the key data table is likely fabricated or copied without re-checking. No image-based manipulation evidence is available from the supplied text.

Key findings

Severe mathematical inconsistency between τPF and kr,S/knr,S (Finding 1, 🔴). Using τPF = 1/(kr,S + knr,S) with the authors' own numbers gives lifetimes far longer than those reported.
Counter-intuitive kRISC vs. τDF ordering (Finding 2, 🟠). The compound with the highest kRISC shows the longest τDF, contrary to expected TADF kinetics.
Text–table numerical mismatch (Finding 3, 🟡). Body text states CEmax of 143.2/146.4 cd A−1, while Table 2 lists 146.7 cd A−1 for m-DBFSi-d.
Image analysis not assessable (Finding 4, ⚠️). Pixel-level checks of Figures 1–5 require the original high-resolution files.

Evidence highlights

p-DBFSi: Reported τPF = 6.7 ns; reported kr,S = 2.2 × 10⁷ s⁻¹, knr,S = 8.9 × 10⁵ s⁻¹ → 1/(kr,S + knr,S) ≈ 43.7 ns (≈6.5× the stated value).
m-DBFSi: Reported τPF = 6.3 ns; reported kr,S = 3.55 × 10⁷ s⁻¹, knr,S ≈ 8.8 × 10⁵ s⁻¹ → 1/(kr,S + knr,S) ≈ 27.5 ns (≈4.4× the stated value).
m-DBFSi-d: Reported τPF = 5.8 ns; reported kr,S = 3.75 × 10⁷ s⁻¹, knr,S ≈ 8.7 × 10⁵ s⁻¹ → 1/(kr,S + knr,S) ≈ 26.1 ns (≈4.5× the stated value).
kRISC ordering vs. τDF: kRISC values stated as 1.6 × 10⁵, 1.1 × 10⁵, and 1.2 × 10⁵ s⁻¹ for p-DBFSi, m-DBFSi, and m-DBFSi-d, respectively; yet τDF in Table 1 is longest for p-DBFSi (42.1 µs), inconsistent with the expectation that higher kRISC yields shorter τDF.
Numerical mismatch: Section 2.4 quotes CEmax of 143.2/146.4 cd A⁻¹, while Table 2 records 146.7 cd A⁻¹ for m-DBFSi-d.
DOI: 10.1002/adma.72684.

Notes

The arithmetic checks in Finding 1 are robust within rounding error and do not depend on measurement uncertainty, since they use the authors' own reported rate constants.
Finding 2 is corroborative rather than independently decisive; TADF kinetics are influenced by additional factors, but the discrepancy amplifies the concern raised by Finding 1.
Finding 3 is a low-level but damaging consistency error typical of incomplete data revisions.
Original transient decay traces, fitting scripts, and high-resolution figures (especially Figure 3) are required to reach a definitive conclusion. PubPeer posting and an editorial inquiry are appropriate next steps.