Defects in Functional Materials. Группа авторов

       CHAPTER 2

       Defect Physics in 2D Nanomaterials Explored by STEM/STM

      JINHUA HONG^∗, MAOHAI XIE^†,‡ and CHUANHONG JIN^∗,§

      ^∗State Key Laboratory of Silicon Materials, School of Materials Science and Engineering, Zhejiang University, Hangzhou 310027, China

      ^†Physics Department, The University of Hong Kong, Pokfulam Road, Hong Kong

       ^‡ [email protected]

       ^§ [email protected]

       1. Introduction

      Structural disorders such as point defects (vacancies, anti-sites, interstitials, and dopants), dislocations, and grain boundaries are commonly present in crystalline materials, which play a key role in dominating the mechanical, opto/electronic, and chemical properties of the crystal materials [1–3]. For instance, the success of modern microelectronics is based on the manipulation of the charge carriers in semiconductor channels of field effect transistors through defect engineering. This helps to realize p-to-n conversion and to control the electric and optical performance of semiconductors. In general, structural defects break the translational symmetry of crystals leading to unique electronic states modifying the intrinsic electronic structure, and thereby considerably tailor the electronic, optical, magnetic, as well as the chemical/catalytic properties of crystals, as reflected by optical spectroscopy — such as ultraviolet–visible spectroscopy (UV–Vis), photoluminescence (PL) spectroscopy, X-ray absorption spectroscopy (XAS), alternating current (AC) conductance, or magnetic properties measurements.

      Defect exploration, control, and engineering have always been at the heart of modern materials science and industrial applications such as traditional semiconductor microelectronics, metal refining, and catalyst design. Besides the traditional 3D solids, such a structure–property philosophy of defects remains a common but challenging issue to be explored even in low-dimensional crystal materials, such as nanocrystals, quantum dots, nanotubes, or ribbons and graphene-like two-dimensional (2D) materials. Hence, direct imaging of the defect structure will be of crucial significance to detect the localized defect state and its electronic properties. The whole picture of structure–property correlation in defects may also lead to the discovery of novel nanophysics in nanomaterials such as magnetism/spin crossover and single-atom catalysis.

      In this chapter, five sections will be included in defect characterization: (1) a brief introduction of scanning transmission electron microscopy (STEM) and scanning tunneling microscopy (STM) in direct probing; (2) STEM characterization of point defects (vacancy and antisite) in 2D transition metal dichalcogenides; (3) real-time experimental observation of defects’ migration in monolayer MoS₂; (4) STEM and STM/scanning tunneling spectroscopy (STS) characterization of inversion domain boundaries in molecular beam epitaxy (MBE)-grown MoSe₂; (5) STEM and STS of the domains of bilayer MoSe₂ to reveal the stacking band structure diversity.

       2. Instrumentation and Technique

      Among the available microscopy techniques for direct atom probing in real space, STEM and scanning probe microscopy (SPM) are suitable to visualize the structure of defects at the atomic scale. Both microscopic techniques can enable us in direct atomically resolved imaging of the defect structure in nanomaterials, even with a proper temporal resolution to capture the defect dynamics. The difference lies in the principles: the former STEM utilizes the electron–atom scattering of the fast-incident electron beam (30–300 keV) penetrating the sample to image atomic structures including defects both within a solid and on the surface; the SPM like scanning tunneling microscopy (STM) is based on the quantum tunneling effect of the valence electrons from the surface atoms of a solid sample. In the quantum limit system such as graphene-like 2D materials with only surface and no volume states, all atoms and defects appear as the surface, forming an ideal platform for the defect exploration using both techniques. Tremendous research examples on 2D materials system have demonstrated the versatility of both routes to identify atomic defects and the electronic states induced.

       2.1. Principles of ADF-STEM and EELS in a TEM

      Modern electron microscopy has developed into an era of aberration correction in electron optics. Owing to recent decade’s commercialization and improvement of probe aberration corrector, high-angle annular dark field scanning transmission electron microscopy (HAADF-STEM) has become a standard atomic resolution imaging mode to directly visualize the structure of the crystal samples. In HAADF imaging (Z-contrast imaging), the detector receives only electrons after a large-angle elastic scattering between incident electrons and sample atoms, as shown in Fig. 1(a). In this incoherent imaging, the intensity and contrast will not change drastically with the sample thickness and the defocus compared to high-resolution (HR) TEM and bright-field (BF) STEM, but obey an approximate elastic Rutherford scattering formalism ∝ Z², where Z is the atomic number of the atomic column imaged. Heavy and light atoms will give very contrasting brightness in different columns in the imaging. Thus, HAADF-STEM imaging directly reflects the real atomic structure and resolves the species of the atoms in a compound material, even without any image simulation. A more accurate scaling relation of HAADF-STEM imaging can be utilized for the atom-by-atom structural and chemical analysis of light-element 2D materials. Along with the high-angle scattered signal, low-angle inelastic scattered electrons and transmitted electrons can be collected for annular dark/bright-field (ADF/ABF) imaging simultaneously with the HAADF imaging and spectroscopic analysis (Figs. 1(b)–1(d)). In the ADF-STEM and ABF-STEM imaging, electrons are partially coherent and the yielded contrast in the imaging will change with the sample thickness and defocus. Although the contrast does not directly indicate the atomic structure, they can be interpretable after a quantitative image simulation.

      Figure 1. Electron scattering geometry, imaging and spectroscopy in a