Academic Commons

Theses Doctoral

Van der Waals Layered Materials: Surface Morphology, Interlayer Interaction, and Electronic Structure

Yeh, Po-Chun

Over the past decades, new materials have formed the backbone in shaping the landscape of technology. From the Si-based transistors in our smart devices to the carbon fibers that have redefined air-transportation, the pursuit for a stronger, lighter, and cost-effective material has never ceased, as well as the attempt to fully understand their physics and material properties. Moore's law just turned 50th this year. Moore's law seems harder and harder to hold as the industry has reached a point where the dimensions of those Si-based transistors are getting too small and thin to proceed quickly and without incurring substantial additional cost. Also, the transistor dimensions have been getting closer and closer to the physical limitation of the Si. Today, the most advanced node is merely "7 nm" – less than 15 layers of Si, silicon oxides, or other metal oxides. At this point, every layers of atoms counts. The search for new ultrathin materials as the "new silicon" has begun.
In this regard, graphene, which is a single sheet of carbon atoms arranged in a hexagonal honeycomb lattice, has led to a huge interest in the science and technology communities due to its exotic physics that arises from low-dimensional confinement. This interest soon extended to two-dimensional (2D) electronic materials systems, especially semiconducting van der Waals layered-materials such as MoS₂ that, unlike graphene, has a direct bandgap material in its monolayer form. There are also many other promising candidates such as transition metal dichalcogenides (TMDCs), black phosphorene, and perovskites on this rapid growing "2D materials" family tree. From condensed matter physics' point of view, studying the electronic behavior of these 2D systems can provide insight into a variety of phenomena, including epitaxial growth, interfacial charge transfer, energy-momentum relation, and carrier mobility, that leads to advanced device fabrication and engineering. In this dissertation, I examine (1) the surface structure, including the growth, the crystal quality, and thin film surface corrugation of a monolayer sample and a few layers of MoS₂ and WSe₂, and (2) their electronic structure. The characteristics of these electronic systems depend intimately on the morphology of the surfaces they inhabit, and their interactions with the substrate or within layers. These physical properties will be addressed in each chapter.
This thesis has dedicated to the characterization of mono- and a few layers of MoS₂ and WSe₂ that uses surface-sensitive probes such as low-energy electron microscopy and diffraction (LEEM and LEED). Prior to our studies, the characterization of monolayer MoS₂ and WSe₂ has been generally limited to optical and transport probes. Furthermore, the heavy use of thick silicon oxide layer as the supporting substrate has been important in order to allow optical microscopic characterization of the 2D material. Hence, to the best of our knowledge, this has prohibited studies of this material on other surfaces, and it has precluded the discovery of potentially rich interface interactions that may exist between MoS₂ and its supporting substrate. Thus, in our study, we use a so-called SPELEEM system (Spectroscopic Photo-Emission and Low Energy Electron Microscopy) to address these imaging modalities: (1) real-space microscopy, which would allow locating of monolayer MoS₂ samples, (2) spatially-resolved low-energy diffraction which would allow confirmation of the crystalline quality and domain orientation of MoS₂ samples, and, (3) spatially-resolved spectroscopy, which would allow electronic structure mapping of MoS₂ samples. Moreover, we have developed a preparation procedure for samples that yield, a surface-probe ready, ultra-clean, and can be transferred on an arbitrary substrate.
In this thesis, to fully understand the physics in MoS₂ such as direct-to-indirect band gap transition, hole mobility, strain, or large spin-orbit splitting, we investigate our sample using micro-probe angle-resolved photoemission (µ-ARPES), which is a powerful tool to directly measure the electronic structure. We find that the valence bands of monolayer MoS₂, particularly the low-binding-energy bands, are distinctly different from those of bulk MoS₂ in that the valence band maximum (VBM) of a monolayer is located at Κ ̅ of the first Brillouin zone (BZ), rather than at Γ ̅, as is the case in bilayer and thicker MoS₂ crystals. This result serves as a direct evidence, if complemented with the photoluminescence studies of conduction bands, which shows the direct-to-indirect transition from mono- to milti-layer MoS₂. We also confirmed this same effect in WSe₂ in our later studies. Also, by carefully studying the uppermost valence band (UVB) of both exfoliated and CVD-grown monolayer MoS₂, we found a compression in energy in comparison with the calculated band, an effect, which were also observed in suspended sample with minimum-to-none substrate interaction. We tentatively attribute it to an intrinsic effect of monolayer MoS₂ owning to lattice relaxation. The degree of compression in CVD-grown MoS₂ is larger than that in exfoliated monolayer MoS₂, likely due to defects, doping, or stress. Furthermore, we find that the uppermost valence band near Κ ̅ of monolayer MoS₂ is less dispersive than that of the bulk, which leads to a striking increase in the hole effective-mass and, hence, the reduced carrier mobility of the monolayer compared to bulk MoS₂.
Beyond monolayer MoS₂, we have studied the evolution of bandgap as a function of interlayer twist angles in a bilayer MoS₂ system. Our µ-ARPES measurements over the whole surface-Brillouin zone reveal the Γ ̅ state is, indeed, the highest lying occupied state for all twist angles, affirming the indirect bandgap designation for bilayer MoS₂, irrespective of twist angle. We directly quantify the energy separation between the high symmetry points Γ ̅ and K ̅ of the highest occupied states; this energy separation is predicted to be directly proportional to the interlayer separation, which is a function of the twist angle. We also confirm that this trend is a result of the energy shifting of the top-most occupied state at Γ ̅, which is predicted by DFT calculations. Finally, we also report on the variation of the hole effective mass at Γ ̅ and K ̅ with respect to twist angle and compare it with theory. Our study provides a direct measurement and serves as an example for how the interlayer coupling can affect the band structure and electron transitions, which is crucial in designing TMDs devices.
To the end of this thesis, I briefly sum up our angle-resolve two-photon photoemission (2PPE) studies on self-assembly molecules, organic molecules, and graphene on highly-crystalline metal systems, and our investigation of their interfacial charge transfer/trapping, image potential states, and coverage-dependent dipole moments, as well as their work functions by using a tunable ultra-fast femtosecond laser.


  • thumnail for Yeh_columbia_0054D_12945.pdf Yeh_columbia_0054D_12945.pdf binary/octet-stream 7.41 MB Download File

More About This Work

Academic Units
Electrical Engineering
Thesis Advisors
Osgood Jr., Richard M.
Ph.D., Columbia University
Published Here
September 17, 2015
Academic Commons provides global access to research and scholarship produced at Columbia University, Barnard College, Teachers College, Union Theological Seminary and Jewish Theological Seminary. Academic Commons is managed by the Columbia University Libraries.