Atomically Resolved Defects Modulate Electronic Structure in Plasma-Assisted 2D Janus MoSSe Monolayers

Abstract

Janus transition metal dichalcogenides, such as MoSSe, are potential materials for advanced electronics, yet their real-world device performance often fails to meet theoretical expectations. The origin of this discrepancy, rooted in atomic-scale imperfections, has remained critically unexplored. Here, using scanning tunneling microscopy and spectroscopy, this work provides atomic-scale insights into the complex electronic structures of monolayer Janus MoSSe, revealing distinct defect species that govern device performance. The residual sulfur dopants are found to introduce a broad band (≈0.5 eV) of shallow in-gap states near the valence band with spatially inhomogeneous distribution. Moreover, this work unveils two distinct native charge defects with spatially electronic influence extending ≈2.5 nm: conductive charge traps that reduce the local effective bandgap by more than half and insulating scattering centers that impede carrier transport. This microscopic understanding of defect-induced electronic modifications explains how atomic-scale imperfections influence macroscopic device limitations, providing fundamental design criteria for the engineering of Janus devices.

Keywords: 2D materials; Janus; STM; defect; electronic device; electronic structure; scanning tunneling microscopy.

1

Synthesis and characterization of monolayer Janus MoSSe. (a) Schematic of the plasma-assisted conversion process from monolayer MoS₂ to Janus MoSSe. Further details can be found in the Methods section. (b) Raman spectra of pristine MoS₂ (green) and Janus MoSSe (orange), highlighting the characteristic shifts from the E_2g and A_1g vibrational modes of MoS₂ to the A₁ ¹ and E² vibrational modes of Janus MoSSe. (c) Photoluminescence (PL) spectra showing the emission peak redshift from MoS₂ (green) to Janus MoSSe (orange). (d) Cross-sectional transmission electron microscopy (TEM) image confirming the monolayer structure. (e) Atomic force microscopy (AFM) topography of a typical MoSSe flake; the inset shows a height profile measured along the dotted line, indicating a monolayer thickness of ≈0.6 nm.

2

Atomic and electronic structures of monolayer Janus MoSSe. (a) Atomic-resolution STM topography images (6.5 nm × 6.5 nm) of the Janus MoSSe surface (V _sample = −1.4 V, I _t = 500 pA, T = 13 K). (b) Height profile along the dashed line in (a), yielding a lattice constant of 3.25 Å. (c) Fast Fourier transform (FFT) of the image in (a), confirming the hexagonal lattice symmetry. (d) Spatially averaged dI/dV spectra comparing Janus MoSSe (orange) and pristine MoS₂ (green), revealing an upward shift of the valence band maximum (VBM) and the emergence of prominent in-gap states from −0.8 to −1.3 eV in MoSSe. The positions of the VBM and CBM, indicated by the inverted triangles, are determined from the intersection of linear fits (dashed lines) of both band edge onsets and the bandgap region. Further detail of band edge extraction can be found in previous literature. , (e) XPS spectra of the Mo 3d/S 2s/Se 3s, S 2p/Se 3p, and Se 3d core levels, used for compositional analysis. The deconvoluted peaks for Mo (blue), S (green), and Se (orange) are shown.

3

Spatially resolved electronic inhomogeneity due to sulfur dopants. (a) High-resolution STM topography showing electronically distinct bright (Region B) and dark (Region D) domains (V _sample = −1.4 V, I _t = 500 pA, T = 13 K). (b) Corresponding current imaging tunneling spectroscopy (CITS) map of the same area taken at −1.15 eV, confirming that the contrast originates from variations in the local density of states (LDOS). The dashed lines indicate the representative areas used for the spectral averaging and statistical analysis shown in (c). (c) Spectral analysis comparing the two regions. The main panel shows averaged STS curves for Regions B (magenta) and D (green). Superimposed are a box plot and scatter plot of individual in-gap state peak measurements (intensity vs energy), extracted from each STS curve, and their projected histograms, which quantitatively visualize the higher intensity and shallower energy level of the states in Region B.

4

DFT-calculated density of states for MoSSe with varying sulfur concentration. Calculated total density of states (black) is shown overlaid with the projected density of states (PDOS) for Mo (red), S (blue), and Se (purple) atoms. Two models are compared. Left panel: a “sulfur-deficient” surface with 30% S concentration on the top Se-layer (MoSSe (30% S)), and right panel: a “sulfur-rich” surface with 70% S concentration on the top Se-layer (MoSSe (70% S)). The PDOS contribution from Mo atoms has been scaled to account for the tunneling distance in STS measurements, allowing for a more direct comparison with experimental spectra (see Supplementary Note 5 for details). The calculations demonstrate that a higher sulfur concentration leads to a more prominent in-gap state feature near the valence band maximum. The orange shaded regions highlight this area of the in-gap states.

5

Bias-dependent imaging and electronic signatures of native charge defects. (a, b) STM topography of the same region imaged at (a) positive (+1.5 V) and (b) negative (−1.5 V) sample bias, revealing the emergence of bright, localized charge defects at negative bias. (c, d) CITS maps of a selected area with multiple defects at (c) +0.88 V (conduction band states) and (d) −1.24 V (in-gap states), V _sample = −0.8 V, I _t = 300 pA, and T = 13 K. The distinct appearance at negative bias allows for the classification of defects into Type A (bright protrusion) and Type B (dark depression). (e) Averaged dI/dV spectra, extracted from the defect areas demarcated in (c) and (d), compare the distinct electronic fingerprints of a Type A (yellow curve) and a Type B (green curve) defect.

6

Spatially resolved spectroscopy of individual charge defects. (a, b) CITS maps of a representative Type A and Type B defect, selected from the overview maps in Figure c,d at +0.88 and −1.24 V, respectively. The colored markers indicate the trajectory of the line-profile STS measurements. (c, d) Waterfall plots showing representative STS spectra extracted from different regions along the line profiles in (a) and (b), respectively. The spectra are color-coded to represent the surrounding normal region (yellow), the defect shell (purple for Type A), and the defect core (green). These plots provide an excellent way to visualize the evolution of the LDOS across the (c) Type A and (d) Type B defects.

7

Electronic band alignment across individual charge defects. (a, b) Position-dependent dI/dV maps constructed from the complete line-profile STS data set for a (a) Type A and (b) Type B defect. The map for Type A reveals a deep potential well, where the highest intensity of the in-gap states (the onset of in-gap states is indicated by a green dashed line) is localized in a conductive shell surrounding a weaker core. In contrast, the map for Type B can be characterized by a uniform reduction of state density and a distinct downward bending of the VBM. The letters C, S, and N on the bars at the top denote the core, shell, and normal regions, respectively. The black dashed lines are visual guides for the VBM, CBM, and Fermi level (E _F).