Acquisition and Analysis of Scanning Tunneling Spectroscopy Data $-$\n WSe$_2$ Monolayer

R. M. Feenstra, Grayson R. Frazier, Yi Pan, Stefan Fölsch, Yu‐Chuan Lin, Bhakti Jariwala, Kehao Zhang, Joshua A. RobinsonarXiv 2020 · DOI · arXiv

Abstract: Acquisition and analysis is described for scanning tunneling spectroscopy data acquired from a monolayer of $WSe_2$ grown on epitaxial graphene on SiC. Curve fitting of the data is performed, in order to deduce band-edge energies. In addition to describing the details of the theoretical curves used for the fitting, the acquisition and analysis methods are also discussed within the larger context of the historical development of scanning tunneling spectroscopy techniques.

Introduction

This research, led by Randall Feenstra’s group, presents a comprehensive methodology for Scanning Tunneling Spectroscopy (STS) of 2D materials. By integrating innovative experimental techniques with a refined theoretical framework, the authors provide a definitive characterization of the electronic structure of a $WSe_2$ monolayer. The study addresses long-standing challenges in measuring semiconducting band gaps and resolving specific valley positions in the Brillouin zone.

Addressing the Challenges of STS in 2D Materials

Scanning Tunneling Spectroscopy is a powerful tool for measuring the local density of states (LDOS). However, semiconducting 2D materials like $WSe_2$ present significant experimental hurdles. To clearly observe the onsets of valence and conduction bands, a dynamic range of 3–4 orders of magnitude in tunneling current is required. Standard fixed-height measurements often fail: if the tip is far enough to avoid crashing at high voltages, the signal near the band gap is lost in noise; if it is close enough to see gap states, the current at high voltages can damage the tip or sample.

Furthermore, traditional theoretical models often oversimplify the physics of tunneling from specific points in the Brillouin zone, leading to inaccurate fits of experimental data.

Methodology: High-Dynamic-Range Acquisition and Revised Theory

The authors employ two primary strategies to overcome these limitations:

  1. V-shaped Tip-Sample Ramps: Instead of maintaining a fixed height, the tip is continuously moved closer to the surface as the voltage magnitude decreases toward zero. This "amplifies" the signal near the band gap. A normalization procedure is then used to convert this raw data back into a standard constant-separation spectrum.
  2. Revised Bardeen/Tersoff-Hamann Theory: The authors argue that for 2D materials, the parallel wave-vector ($k_{||}$) is a critical factor. States at the $K$-point of the Brillouin zone have a large $k_{||}$ and decay much faster into the vacuum than states at the $\Gamma$-point. Their model explicitly includes this wave-vector-dependent transmission.
  3. Hyperbolic Dispersion: Rather than assuming simple parabolic bands, the researchers utilize a hyperbolic model. This allows bands to be parabolic near the onset while transitioning to linear dispersion at higher energies, more accurately reflecting the real band structure of Transition Metal Dichalcogenides (TMDs).
Scanning tunneling spectra of WSe2 (a-c) and epitaxial graphene (d) showing band features and the effect of different tip-sample separation ramps (Δs).

Key Insights and Results

The $k_{||}$ Filter

A significant finding is the role of the wave-vector in the transmission term. Because tunneling probability decays exponentially with $k_{||}$, certain bands that appear prominent in theoretical density-of-states calculations can appear very faint in STS. The authors posit that a physically accurate model should yield "amplitude" scale factors near unity for all bands; if a fit requires extreme scaling without physical justification, the underlying dispersion model is likely flawed.

Resolving the $K_C$ vs. $Q_C$ Controversy

There has been ongoing debate regarding whether the $K$ or $Q$ valley constitutes the true conduction band minimum in $WSe_2$. By utilizing high-dynamic-range data and their $k_{||}$-aware model, the authors demonstrate that the $K_C$ band lies below the $Q_C$ band. This result is consistent with recent ARPES findings but clarifies previous contradictions in STS interpretations.

Quantitative Findings

  • Band Gap: The electronic band gap of the $WSe_2$ monolayer was determined to be $1.96 \pm 0.02$ eV.
  • Band Identification: The study successfully identified and fit features corresponding to the $\Gamma_V$ and $K_V$ valence bands, as well as the $K_C, Q_C,$ and $\Sigma_C$ conduction bands (as seen in Figure 1).
  • Substrate Interaction: The use of V-shaped ramps allowed for a clear view of the graphene substrate's background conductance even through the $WSe_2$ layer, which was subtracted to isolate faint TMD features.

Limitations and Future Outlook

Despite the improved model, the authors noted an "additional broadening" parameter (approximately 25–38 meV) required to match the experimental steepness of band onsets. This may be attributed to Tip-Induced Band Bending (TIBB), although TIBB is expected to be minimized by the graphene/SiC substrate. Future work will focus on a more rigorous integration of TIBB into the $k_{||}$-dependent tunneling model. Additionally, supplemental data indicates that spectra vary significantly with spatial location, highlighting the sensitivity of local electronic properties to the nanometer-scale environment.

Summary updated

Abstract

Acquisition and analysis is described for scanning tunneling spectroscopy\ndata acquired from a monolayer of WSe$_2$ grown on epitaxial graphene on SiC.\nCurve fitting of the data is performed, in order to deduce band-edge energies.\nIn addition to describing the details of the theoretical curves used for the\nfitting, the acquisition and analysis methods are also discussed within the\nlarger context of the historical development of scanning tunneling spectroscopy\ntechniques.\n
BibTeX
@misc{feenstra2020acquisition,
  author = {R. M. Feenstra and Grayson R. Frazier and Yi Pan and Stefan Fölsch and Yu‐Chuan Lin and Bhakti Jariwala and Kehao Zhang and Joshua A. Robinson},
  title = {{Acquisition and Analysis of Scanning Tunneling Spectroscopy Data $-$\textbackslash\{\}n WSe$\_2$ Monolayer}},
  year = {2020},
  eprint = {2011.14168},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2011.14168},
}

Loading related content...

Figure