Surface-dipole contamination of the Schottky-Mott slope: why the Fermi-level pinning factor does not transfer to MXene contacts

Abstract The pinning factor S = dΦSB/dW is the standard measure of Fermi-level pinning at metal–semiconductor junctions and is increasingly applied to contacts between two-dimensional semiconductors and MXenes. We show that this transfer is not valid. In a metal series W is varied by changing the element; in an MXene series it is varied by changing the surface termination, whose dipole is also the dominant source of deviation from the Schottky–Mott limit. Abscissa and deviation share one origin, and the fitted slope is not a pinning parameter. For the published MoS2/Ti2CT2 data of Gan et al the apparent slope is Sapp = 0.19, conventionally read as strong pinning, although the Ti2CF2 interface obeys the Schottky–Mott rule to within 0.03 eV. The decomposition Sapp = γ + dμIS/dWT, with μIS the interface dipole, resolves the contradiction: the potential discontinuity of 2.50 eV reported by those authors gives a dipole term of −0.79. To measure γ we compute a fixed-termination series of eight metallic F-terminated MXenes on MoS2, which gives γ = 0.87 (95% interval [0.77, 0.96]); the decomposition then returns 0.08 against the observed 0.19, the dipole term accounting for the departure of Sapp from γ to within 20%. An O-terminated series confirms that the dipole stays small at fixed termination. We also re-derive S for every published metal–MoS2 data set from raw barrier heights and document two hazards: at least three distinct quantities are reported as the slope and compared as though interchangeable, and a widely used compilation merges vertical and lateral barriers. A bootstrap test shows that the difference between the frequently quoted experimental S ≈ 0.11 and the first-principles cluster at S ≈ 0.34 is not established at the 95% level.

Hali tarjima qilinmagan