In Situ Exfoliation Method of Large-Area 2D Materials

Abstract

2D materials provide a rich platform to study novel physical phenomena arising from quantum confinement of charge carriers. Many of these phenomena are discovered by surface sensitive techniques, such as photoemission spectroscopy, that work in ultra-high vacuum (UHV). Success in experimental studies of 2D materials, however, inherently relies on producing adsorbate-free, large-area, high-quality samples. The method that yields 2D materials of highest quality is mechanical exfoliation from bulk-grown samples. However, as this technique is traditionally performed in a dedicated environment, the transfer of samples into vacuum requires surface cleaning that might diminish the quality of the samples. In this article, a simple method for in situ exfoliation directly in UHV is reported, which yields large-area, single-layered films. Multiple metallic and semiconducting transition metal dichalcogenides are exfoliated in situ onto Au, Ag, and Ge. The exfoliated flakes are found to be of sub-millimeter size with excellent crystallinity and purity, as supported by angle-resolved photoemission spectroscopy, atomic force microscopy, and low-energy electron diffraction. The approach is well-suited for air-sensitive 2D materials, enabling the study of a new suite of electronic properties. In addition, the exfoliation of surface alloys and the possibility of controlling the substrate-2D material twist angle is demonstrated.

Keywords: 2D materials; angle-resolved photoemission spectroscopy; band structure; exfoliation; transition metal dichalcogenides.

Figure 1

Sketch of the KISS exfoliation procedure. In the step 1, the sample surface is cleaved in UHV to expose an adsorbate‐free surface and the single‐crystal metal substrate is sputtered and annealed to generate an atomically‐clean and flat surface. In the step 2, the two surfaces are brought into contact. In the step 3, the sample and substrate are gently separated, resulting in the in situ exfoliation of the 2D material onto the substrate. A high‐quality, large‐area 2D material is left on the substrate, as seen in the optical microscopy image for the case of SL WSe₂ on Au(111). The SL thickness of the sample is demonstrated by its characteristic height of 0.7 nm as measured by AFM data in the region marked with a red dashed square in the optical image.

Figure 2

Silver‐assisted exfoliation of large‐area SL WSe₂. a) The optical microscopy image shows a flake of SL WSe₂ exfoliated by the KISS method onto Ag(111); the two major dimensions indicated by arrows are 292 µm and 246 µm. The color balance was adjusted to make the single layer more visible. The dark region near the top is a small multi‐layered region. b) LEED image of the WSe₂ flake; the angle between the Bragg peaks of the sample and the substrate is ≈15°, confirming that the exfoliation angle can be arbitrarily chosen and does not rely on coherent epitaxy. c) ARPES data showing WSe₂ bands around the $\overline{\mathrm{\Gamma }}$ (left) and $\overline{\mathrm{K}}$ points (right). The energy axis is referenced to the valence band maximum (VBM) that is located at the $\overline{\mathrm{K}}$ point. The green arrow indicates where the hybridization between the WSe₂ and Ag(111) bands appear. The higher energy of the bands at the $\overline{\mathrm{K}}$ point is characteristic of the SL nature of the flake.

Figure 3

Universality of the KISS method: Exfoliation of different TMDC materials onto multiple substrates. Optical microscopy images are shown in the top row and ARPES data acquired in situ in the bottom row for three different systems. The insets illustrate the cuts in momentum space that are plotted in the respective ARPES spectra. a) Flakes of WS₂/Au(111) are over 100 µm in size and predominantly SL as demonstrated by the photoemission data, where the valence band maximum is at the $\overline{\mathrm{K}}$ point. Above the $\overline{\mathrm{\Gamma }}$ point, an Au(111) surface state is visible. b) Exfoliated flakes of WS₂/Ge(001) demonstrate that exfoliation of 2D materials directly onto semiconducting substrates is also possible. The ARPES data is of high quality, but the flakes are generally smaller and thicker than on metallic substrates. c) 2D WTe₂ has been exfoliated onto Ag(111), yielding flakes of hundreds of microns in size. The KISS exfoliation method is particularly useful for air‐sensitive samples, such as WTe₂, which can be exfoliated and measured in situ. The ARPES data is of remarkable quality and is shown from both a multilayered (3D) area of the flake (left) and from a 2D area (right). The spectra are markedly different, especially close to the Fermi level, highlighting the vast topological difference between the two cases.

Figure 4

Formation of AgTe alloy on Ag(111). a) Model of the crystal structure of the AgTe alloy on Ag(111) (top) and sketch of the corresponding surface Brillouin zones (bottom). b) AgTe band structure along the $\overline{\mathrm{\Gamma }}$–$\overline{\mathrm{K}}$ direction of the substrate. $\overline{\mathrm{\Gamma }}$, $\overline{\mathrm{M}}$, and ${\overline{\mathrm{\Gamma }}}_2$ mark the high‐symmetry points of AgTe. The dashed red line indicates where the constant energy contour in (c) is taken. c) Constant energy contour taken at E_Bin = 1.4 eV. Bulk and surface states (SS) of Ag(111) are visible, together with the AgTe bands. d,e) The AgTe band structure taken at the $\overline{\mathrm{\Gamma }}$ point in the first surface Brillouin zone and the ${\overline{\mathrm{\Gamma }}}_2$ point in the second surface Brillouin zone, respectively. The directions of the cuts directions are indicated in the schematics next to the data. Blue and red arrows indicate Ag and AgTe bands, respectively.