University of Southern California Dissertations and Theses

Optoelectronic, thermoelectric, and photocatalytic properties of low dimensional materials

(USC Thesis Other)

Optoelectronic, thermoelectric, and photocatalytic properties of low dimensional materials

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Optoelectronic, Thermoelectric, and Photocatalytic Properties of Low Dimensional Materials By Jihan Chen A Dissertation Presented to the FACULTY OF THE USC GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Electrical Engineering) December 2019 Copyright 2019 Jihan Chen 1 EPIGRAPH “Somewhere, something incredible is waiting to be known.” -Carl Sagan 2 DEDICATION To my beloved parents 献给我亲爱的父母 3 ACKNOWLEDGEMENTS PhD research is a non-linear long process with both pain and happiness. These fruitful five years is also an important stage for my own evolution and growth. The gratitude that I wish to convey is far more than all the sentiments within these two pages. Therefore, this is a subset of those who deserve credit for helping me going through this important stage of life. First and foremost, I would like to express my sincere gratitude to my academic advisor Prof. Stephen B. Cronin for his invaluable guidance, inspiration and support. His passion for research, dedication to work and insights of science nourishes and inspires me. I could not make the achievement without his help. I would also like to thank, Dr. Han Wang Dr. Aiichiro Nakano, Dr. Wei Wu, Dr. Edward Goo and Dr. Chongwu Zhou for being my dissertation committee and qualifying exam committee members. It is valuable and fortunate to overlap and work with senior Ph.D. students, Dr. Rohan Dhall and Dr. Zhen Li, in Cronin research group. Their professionality, critical thinking and inspiring advice impressed me and set very good examples for my Ph.D. research. I’m lucky to have fellows and friends like Dr. Bingya Hou and Mr. Haotian Shi who shared their experience with me and being good audience, they gave me lots of encouragement and help when the cards all fold. My gratitude also extends to all the fellows in Cronin research lab and other 4 research labs in Eletrophysics: Dr. Shun-Wen Chang, Dr. Adam Bushmaker, Dr. Chun- Chung Chen, Dr. Moh Amer, Dr. Jing Qiu, Dr. Guangtong Zeng, Dr. Shermin Arab, Dr. Nirakar Poudel, Dr. Lang Shen, Mr. Yu Wang, Ms. Sisi Yang, Mr. Bo Wang, Mr. Zhi Cai, Mr. Bofan Zhao, Ms. Indu A A, and et al. Thanks Dr. Kian Kaviani for offering me a position as an EE504L teaching assistant, which improved my communication and teaching skills and to some extent fulfill my life. I’m grateful to all my friends, thanks for their encouragement and support, some of them ever gave me answers when I suspected the world. Finally, I want to thank my beloved parents for their endless love and unlimited support through my life. They are the root of my motivation all the time. 5 TABLE OF CONTENTS EPIGRAPH .................................................................................................................... 1 DEDICATION ............................................................................................................... 2 ACKNOWLEDGEMENTS ........................................................................................... 3 LIST OF FIGURES ....................................................................................................... 8 LIST OF TABLES ....................................................................................................... 12 ABSTRACT ................................................................................................................. 13 Chapter 1 Introduction ................................................................................................... 1 1.1 Nanotechnology ............................................................................................... 1 1.2 Carbon Nanotube ............................................................................................. 2 1.3 Two-Dimensional Materials .......................................................................... 11 1.4 Thermoelectric ............................................................................................... 15 1.5 Photocatalysis ................................................................................................ 20 Chapter 2 Enhanced Photoluminescence in Air-suspended Carbon Nanotubes by Oxygen Doping ............................................................................................................ 26 2.1 Abstract .............................................................................................................. 26 2.2 Introduction ........................................................................................................ 27 2.3 Experimental Details .......................................................................................... 28 2.4 Results and discussion ........................................................................................ 31 2.5 Conclusion .......................................................................................................... 37 Chapter 3 Highly efficient, high speed vertical photodiodes based on few-layer MoS2 ...................................................................................................................................... 39 3.1 Abstract .............................................................................................................. 39 3.2 Introduction ........................................................................................................ 40 6 3.3 Experimental details ........................................................................................... 41 3.4 Results and discussion ........................................................................................ 44 3.5 Conclusion .......................................................................................................... 51 Chapter 4 Enhanced Thermoelectric Efficiency in Topological Insulator Bi2Te3 Nanoplates via Atomic Layer Deposition-based Surface Passivation ......................... 54 4.1 Abstract .............................................................................................................. 54 4.2 Introduction ........................................................................................................ 55 4.3 Experimental details ........................................................................................... 56 4.4 Results and discussions ...................................................................................... 58 4.5 Conclusion .......................................................................................................... 63 Chapter 5 Enhanced Cross-plane Thermoelectric Transport of Rotationally-disordered SnSe2 via Se Vapor Annealing .................................................................................... 65 5.1 Abstract .............................................................................................................. 65 5.2 Introduction ........................................................................................................ 66 5.3 Experimental details ........................................................................................... 67 5.4 Results and discussions ...................................................................................... 70 5.5 Conclusion .......................................................................................................... 77 Chapter 6 Plasmon-Resonant Enhancement of Photocatalysis on Monolayer WSe2 .. 82 6.1 Abstract .............................................................................................................. 82 6.2 Introduction ........................................................................................................ 83 6.3 Experimental details ........................................................................................... 85 6.4 Results and discussions ...................................................................................... 88 6.5 Conclusion .......................................................................................................... 95 Chapter 7 Stacking Independence of WSe2/MoSe2 Heterostructures for Photocatalytic Energy Conversion....................................................................................................... 97 7 7.1 Abstract .............................................................................................................. 97 7.2 Introduction ........................................................................................................ 98 7.3 Experimental Details .......................................................................................... 99 7.4 Results and Discussion ..................................................................................... 103 Chapter 8 Future Work .............................................................................................. 111 8.1 Thermoelectric study of charge density wave transition in TMDC heterostructure ........................................................................................................ 111 Bibliography .............................................................................................................. 115 Appendix: Supplemental Documents ........................................................................ 130 8 LIST OF FIGURES Figure 1.1. Illustrations of (a) a graphene sheet and (b) a single walled carbon nanotube. ................................................................................................................................. 3 Figure 1.2. (a) The unrolled honeycomb lattice of a nanotube. The vectors OA and OB define the chiral vector C h and the translational vector T of the nanotube, respectively. (b) The (4, 2) SWNTs, showing the translation vector T. 5 ................ 4 Figure 1.3. (a) Schematic of the electronic band structure of CNT and “cutting lines”. (b) Schematic of the electronic band structure of metallic and semiconducting CNT, respectively. 8 ........................................................................................................... 6 Figure 1.4. (a, b) Schematics of electronic dispersion and DOS function for metallic and semiconducting CNT. (c) Kataura plot. 9 .......................................................... 7 Figure 1.5. Raman spectra of CNT (a) RBM, D-band, G-band, and G’-band (b) G- and G+ mode of CNT (c) Schematic of RBM mode in CNT (d) Schematic of G- and G+ mode in CNT. 10 ....................................................................................................... 9 Figure 1. 6. PL image (a) and spectra (b) of suspended carbon nanotube .................... 10 Figure 1. 7. Building of Van der Waal heterostructures using 2D materials can be achieved by mechanically-assembled stacks (top) and large-scale growth by CVD or physical epitaxy (bottom). 15 ............................................................................. 11 Figure 1. 8. Atomic structure of single layers of transition metal dichalcogenides (TMDCs) in their trigonal prismatic (2H) phase (left), and distorted octahedral (1T) phase (right). 16 ....................................................................................................... 13 Figure 1. 9. Calculated band structures of (a) bulk MoS2, (b) quadrilayer MoS2, (c) bilayer MoS2, and (d) monolayer MoS2. The solid arrows indicate the lowest energy transitions. 18 .............................................................................................. 14 Figure 1. 10. (a) PL spectra for suspended mono-and bilayer MoS2 samples in the photon energy range from 1.3 to 2.2eV . Inset: PL quantum yield of thin layers for N=1–6. (b) Normalized PL spectra by the intensity of peak A of thin layers of MoS2 for N=1–6. 17 .......................................................................................................... 14 Figure 1. 11. Schematic of Seebeck Effect (a) and Peltier Effect (b). 20 ...................... 16 Figure 1. 12. Power generation/cooling efficiency as a function of average ZT of Seebeck Effect (a) and Peltier Effect (b). 20 ........................................................... 17 Figure 1. 13. Illustration of dependence of Seebeck coefficient, electrical conductivity (top) and thermal conductivity (bottom) as a function of carrier concentration. .. 19 Figure 1. 14. Energy band diagram of various semiconductors plotted with the redox of some chemical reactions. 24 .................................................................................... 22 Figure 1. 15. The band energetics of a semiconductor/liquid contact are shown in three cases: (A) before equilibration between the two phases and (B) after equilibration, but in the dark. 22 .................................................................................................... 23 Figure 1. 16. Electrode potential of water splitting as a function of the pH of the electrolyte, including equilibrium regions for water. 27 ......................................... 25 Figure 2. 1. (a) Optical microscope image of quartz pillar arrays. Photoluminescence 9 image of suspended CNT (b) before and (c) after UV/ozone treatment. (d) SEM image of CNT suspended on top of the pillars ....................................................................... 30 Figure 2. 2. (a) Radial breathing mode (RBM) Raman spectra and (b) PL spectra of CNT before (black) and after (red) exposure to the UV/ozone treatment. ........... 33 Figure 2. 3. D-band and G-band Raman spectra of CNT before (black) and after (red) exposure to the UV/ozone treatment..................................................................... 36 Figure 3. 1. (a) Schematic diagram of the device fabrication process. First, the MoS2 is exfoliated from bulk crystal onto a PDMS substrate mounted on a glass slide. After identifying the desired few-layer flake, the sample is transferred from PDMS onto a Si/SiO2 substrate with pre-patterned metal electrodes. The top electrode is then patterned using electron-beam lithography. (b) Optical microscopy image of the vertical MoS2 device after the fabrication. The dashed line indicates the semi- transparent top electrode. (c) Photoluminescence spectrum of the few-layer MoS2 sample after transfer. ............................................................................................. 44 Figure 3. 2. (a) Photocurrent and (c) photovoltage spectra of the vertical MoS2 device. In addition to the main peak around 1.8–2.1 eV , lower energy peaks at 1.4 and 1.5 eV are also observed due to the indirect transition. (b) Power spectrum of the incident laser plotted together with the EQE of the device. The two peaks at 1.85 and 2.0 eV correspond to the K-point transition. The EQE of the direct transition is about 10 times larger than the indirect transition. (d) Isc–V oc plot showing that the direct band transition (red points) and the indirect band transition (blue points) give different Isc–V oc relations. (e) Extracted electrical power and energy conversion efficiency estimated from the product of the short circuit current and open circuit voltage, assuming a fill factor of 0.25. .................................................................. 47 Figure 3. 3. (a) Photo-I–V characteristics of the vertical MoS2 device under 532 nm illumination. (b) Log plot of the photo-I–V data. ................................................. 48 Figure 3. 4. (a) Photovoltage plotted as a function of time under modulated 680 nm laser illumination. The photovoltage signal is processed after FFT filtering to remove high frequency noise, with the cutoff frequency of the FFT filtering set to 100 kHz. (b) The frequency response of the open circuit photovoltage, which shows a cutoff frequency at 48 kHz. V o is the photovoltage measured with continuous light illumination. .................................................................................................. 50 Figure 4. 1. Schematic diagram of the device fabrication process (a) and optical image of the device before (b) and after (c) Al2O3 deposition. ....................................... 57 Figure 4. 2. Temperature calibration of the right metal RTD of the Bi2Te3 device shown in Figure 1. (a) The resistance of the RTD measured at different temperatures. (b) Normalized resistance (R/R0, R0 is the resistance at 300K) plotted as a function of T. (c) The resistance change of the RTDs under various heating currents. (d) Temperature change of the metal RTDs plotted as a function of heating power. . 58 Figure 4. 3. In-plane Seebeck coefficient of the Bi2Te3 flake (a) before and (b) after Al2O3 deposition. .................................................................................................. 60 10 Figure 4. 4. Hall measurements of the Bi2Te3 flake. (a) Optical microscope image of the device and (b) Hall carrier densities of the Bi2Te3 flake before and after Al2O3 deposition. ............................................................................................................. 61 Figure 5. 1. (a-d) Schematic diagrams of the device fabrication process and (e) optical image of the completed device. ............................................................................ 69 Figure 5. 2. Schematic illustration of a precursor designed to form a layered structure of SnSe when annealed in an inert atmosphere under optimized conditions. Upon heating in Se-vapor for 30 minutes at 300 °C, the layered SnSe converts to SnSe2. Se and Sn atoms are represented in orange and blue, respectively. ...................... 69 Figure 5. 3. Temperature calibration of the bottom RTD of a 50nm SnSe2 film after Se- vapor annealing. (a) The resistance of the RTD measured at different temperatures. (b) Normalized resistance (R/R0, where R0 is the resistance at 300K) plotted as a function of temperature. (c) The resistance changes of the RTD under various heating currents. (d) Temperature change of the metal RTD plotted as a function of heating power. ....................................................................................................... 71 Figure 5. 4. Cross-plane Seebeck coefficient of the 50 nm SnSe film before Se-vapor annealing. .............................................................................................................. 72 Figure 5. 5. Cross-plane Seebeck coefficient of the 50 nm SnSe2 after Se-vapor annealing. .............................................................................................................. 73 Figure 5. 6. Specular X-ray diffraction (XRD) patterns of SnSe films. Both samples were annealed in a nitrogen atmosphere and one was subsequently annealed in a selenium atmosphere (the blue pattern) resulting in a conversation to SnSe2. Miller indices are provided for select reflections. The intensity was plotted on a log scale to enhance weak reflections. ................................................................................. 75 Figure 5. 7. In-plane X-ray diffraction (XRD) patterns for targeted SnSe samples. Both samples were annealed in a nitrogen atmosphere and one was subsequently annealed in a selenium atmosphere (the blue pattern) resulting in a conversation to SnSe2. The reflections are indexed to SnSe and SnSe2 with one substrate peak (marked with the asterisk) in the Se-vapor annealed sample. ............................... 76 Figure 6. 1. (a) Optical microscope image and (b) photoluminescence spectrum of transferred CVD-grown monolayer WSe2 (light grey region) on ITO substrate. . 87 Figure 6. 2. (a) Diagram illustrating the basic sample configuration. (b) Schematic circuit diagram of the three-terminal photoelectrochemical setup with the modulated laser and AC lock-in amplifier. (c) Schematic diagram of the photocatalytic iodine redox process. ..................................................................... 89 Figure 6. 3. DC current, AC current, and AC phase measurements for the bare WSe2 monolayer electrode (i.e., without Au nanoislands) (a) without and (b) with illumination under visible light and (c) bare ITO glass slide without WSe2 under the same illumination conditions in (b). ............................................................... 91 Figure 6. 4. (a) AC photocurrent and (b) AC phase measurements for the monolayer WSe2 electrode with and without Au nanoislands under visible illumination. ..... 92 11 Figure 6. 5. (a) Transmission electron microscope (TEM) image of 5 nm Au nanoisland film (dark grey regions). (b) Enhancement factor of the electric field intensity at the Au nanoislands/monolayer WSe2 interface calculated using the FDTD method. ............................................................................................................................... 94 Figure 7. 1. (a) Energy band diagram of interlayer exciton in a MoSe2/WSe2 heterostructure. (b) Cross-sectional diagram showing an MoSe2-on-WSe2 on ITO substrate. (c) Schematic diagram of the photocatalytic iodine redox process taking place on the TMDC heterostructure. ................................................................... 102 Figure 7. 2. Optical microscope images of (a) MoSe2-on-WSe2 and (c) WSe2-on-MoSe2 heterostructures deposited on ITO substrates. AC photocurrent measurements for the (b) MoSe2-on-WSe2 and (d) WSe2-on-MoSe2 heterostructures under 785 nm laser excitation. (e) Photoluminescence spectrum of the MoSe2-on-WSe2 heterostructures taken with 532 nm wavelength excitation. The photon energy corresponding to 785 nm wavelength light is indicated in the plot. ................... 105 Figure 7. 3. (a) Capacitance-voltage measurements of monolayer WSe2 on ITO electrodes (black) and bare ITO electrode (red). (b) Charge density and reference potential relationship of monolayer WSe2 on ITO substrate extracted from the corresponding curve in (a). ................................................................................. 107 Figure 7. 4. (a) Optical microscope image of MoSe2-on-WSe2 heterostructure deposited on an ITO substrate. (b) AC photocurrent measurements taken in various regions of the MoSe2-on-WSe2 heterostructures on ITO substrate under 532 nm wavelength illumination. ........................................................................................................ 108 Figure 7. 5. (a) Optical microscope image of WSe2-on-MoSe2 heterostructures deposited on an ITO substrate. (b) AC photocurrent measurements taken in various regions of the WSe2-on-MoSe2 heterostructure under 633 nm wavelength illumination. ........................................................................................................ 109 Figure 8. 1. Temperature dependent resistivity and carrier concentration for two different (SnSe)1.15VSe2 samples with CDW transitions at ~100 K measured in the Johnson lab.......................................................................................................... 112 Figure 8. 2. In-plane thermoelectric voltage for a (PbSe)1(VSe2)1 superlattice structure. ............................................................................................................................. 114 12 LIST OF TABLES Table 1. 1. U.S. electricity generation by source, amount, and share of total in 2018 21 Table 2. 1. The PL intensity, linewidth, and center wavelength for 10 suspended CNT samples before and after UV/ozone treatment. ..................................................... 34 Table 2. 2. The D-band intensity and G-band linewidth of 10 suspended CNT samples before and after UV/ozone treatment. ................................................................... 37 Table 5. 1. XRF integrated counts and counts per layer for both SnSe targeted films annealed under different conditions. ..................................................................... 74 13 ABSTRACT This dissertation work presents several research investigations on nanoscale electronic devices based on low-dimensional materials, including one dimensional material--carbon nanotube, and a set of two-dimensional materials, such as MoS2, Bi2Te3, SnSe2, WSe2 and WSe2/MoSe2 heterostructures. These research projects shed light on the fundamental physics and potential optoelectronic, thermoelectric and photocatalytic applications of these nanomaterials and their hetero-structures. It could provide a reference and have a good enlightenment to the future studies. Chapter 1 provides background information that will help to understand this dissertation work. It contains a brief overview of properties of one-dimensional and two-dimensional materials, as well as their electronic band structures, which is important to understand the optoelectronic applications of these low-dimensional materials. A brief introduction of thermoelectric and photocatalysis are also included to help understanding the thermoelectric and photocatalytic applications of these nanomaterials and their heterostructures. Chapter 2 presents an optoelectronic application of air-suspended carbon nanotubes. It reveals photoluminescence imaging and spectroscopy of air-suspended carbon nanotubes can be enhanced by oxygen doping with a direct ultra violet/ozone treatment exposure, which is more efficient than previous methods. This ability to control and engineer defects in carbon nanotubes is important for realizing several optoelectronic applications such as LEDs and single photon sources. 14 Chapter 3 presents an optoelectronic application of the most famous transition metal dichalcogenides, MoS2, in two-dimensional material family. A highly efficient, high speed vertical photodiode can be achieved by using few-layer MoS2. Compared with in-plane devices, it exhibits a substantially larger photocurrent and shorter photo- response, which could pave the way for a more rational device design to boost performance. Chapter 4 and Chapter 5 present thermoelectrical studies of two-dimensional materials, Bi2Te3 and SnSe2, which has been known as good candidates for thermoelectrical applications. Chapter 4 reports a deposition of Al2O3 using atomic layer deposition can enhance the thermoelectric properties of the materials, which offers an appropriate surface passivation method of this high surface-to-volume ratio nanoflake to prevent the degradation of its thermoelectric properties. Chapter 5 reveals the cross-plane thermoelectric transport of rotationally-disordered SnSe2 can be enhanced via Se vapor annealing, which demonstrates a robust method for mitigating unintentional doping in a promising class of materials for thermoelectric applications. Chapter 6 presents photocatalytic application of monolayer WSe2. Plasmon- resonant are used to enhance the photocatalysis performance of monolayer WSe2 and improve the overall photoconversion efficiency. The incident light coupled with plasmonic nanoislands effectively from the far field to the near field in the plane of the monolayer WSe2. The fundamental science and application in these two chapters could be a source of inspiration for increasing the utilization of renewable environment- friendly solar energy. 15 Chapter 7 presents the photocatalytic application of transition metal dichalcogenide (TMDC) heterostructures. We discuss the effect of stacking order of two TMDC materials, WSe2 and MoSe2, in their heterostructures on the photocatalytic performance. Chapter 8 presents the future work of the thermoelectric applications of transition metal dichalcogenide (TMDC) heterostructures. For the next step, we will explore the effect of charge density wave (CDW) transitions on the cross-plane and in- plane thermoelectric transport properties of these heterostructure materials. 1 Chapter 1 Introduction 1.1 Nanotechnology Since the beginning of human civilization, people’s curiosity about the unknown has never stopped. Whether looking up at the sky or overlooking the earth, whether it is the celestial movement in the universe or the cellular structure in the living creature, the research on science often shows a trend of polarization. One is the exploration of the macro world, and the other is the understanding of the micro world. This is why nanotechnology, a method that allow scientist to investigate, manipulate and control matters at nanometer-scale, has attracted the attention of countless people in the past half century. Top-down and bottom-up approaches are commonly used in nanotechnology. The top-down approach is used to create smaller devices by using larger ones to direct their assembly, such as fabricate nano-structures from bulk materials by lithography, etching, and exfoliation etc, which has been widely used in electronic and photonic devices. Bottom-up approach is arranging smaller components into more complex assemblies by chemical synthesis, self-assembly, and deposition, such as molecular beam epitaxy (MBE), chemical vapor deposition (CVD), and electro-deposition etc. Because of the development of nanotechnology, nanomaterials have presented different dimensionalities, especially in low-dimensionalities: zero-dimensional nano- structures, such as nano-particles and quantum dots; one-dimensional nano-structures, such as carbon nanotubes and nanowires; two-dimensional nano-structures, such as graphene and transition metal dichalcogenides. These nanomaterials exhibit unique 2 electrical, thermal, optical, mechanical, and chemical properties and lead to a variety of important applications in modern society. 1.2 Carbon Nanotube Carbon nanotubes (CNTs) have been studied extensively over the last two decades due to their remarkable mechanical, electronic, and thermal properties. It was first discovered by Sumio Ijima in 1991 1 . Their extremely high electron mobility of 100,000 cm 2 /V·s, outstanding thermal conductivity of 6600 W/m·K, as well as their unprecedented high Young’s modulus of 1.8 TPa have been investigated experimentally and theoretically. 2-4 All these simulated or measured properties of the carbon nanotube make it a promising material in nano-electronic devices, thermal management, and nano-mechanical applications. There are two types of CNTs depending on the number of carbon layers: single- walled carbon nanotube (SWNT) and multi-walled carbon nanotube (MWNT). The diameter of a SWNT is typically between 1-2 nm, while that of a MWNT ranges between 5-50 nm. The structure of a SWCNT can be depicted as a rolled-up two-dimensional single sheet of graphite, which consists of honeycomb lattice of carbon atoms extending in two dimensions. Figures 1a and 1b show the structures of a graphene sheet and a carbon nanotube, respectively. 3 Figure 1.1. Illustrations of (a) a graphene sheet and (b) a single walled carbon nanotube. The structure of nanotube can be identified by a combination of two integer indices, (n,m), known as “chiral indices”, which determine the direction in which the sheet of graphene is rolled up. Different ways to “roll up” the graphene sheet result in different chiralities of nanotubes. As shown in Figure 1.2, the hexagonal honeycomb lattice has two basis vectors a1 and a2, which correspond to the so-called zigzag direction and armchair direction, respectively. The rectangle OAB’B defines the unit cell for the nanotube, and the vector going around the circumference of the CNT is called the chiral vector, where C h = n a1 + m a2. Thus, (n,m) has been called as the chirality of the nanotube. The translation vector of a unit cell is 𝑻 = 𝑡 1 𝒂 𝟏 + 𝑡 2 𝒂 𝟐 where t1 and t2 are 𝑡 1 = 2 𝑚 + 𝑛 𝑑 𝑅 , 𝑡 2 = − 𝑚 + 2 𝑛 𝑑 𝑅 , (a) ( b ) (1.1) (1.2) 4 and dR is the greatest common divisor of (2m+n, m+2n). 5 Figure 1.2. (a) The unrolled honeycomb lattice of a nanotube. The vectors OA and OB define the chiral vector C h and the translational vector T of the nanotube, respectively. (b) The (4, 2) SWNTs, showing the translation vector T. 5 The electronic structure of CNTs can also be derived from graphene using the zone-folding technique. The reciprocal lattice of graphene is also hexagonal honeycomb structure. The reciprocal lattice vectors of graphene are 𝒃 𝟏 = ( 2 𝜋 √ 3 𝑎 , 2 𝜋 𝑎 ) , 𝒃 𝟐 = ( 2 𝜋 √ 3 𝑎 , − 2 𝜋 𝑎 ) The electronic dispersion relation of graphene can be calculated analytically from the tight-binding method. 6-7 𝐸 ± ( 𝑘 𝑥 , 𝑘 𝑦 ) = ± 𝛾 √ 3 + 2 cos ( 𝒌 ∙ 𝒂 𝟏 ) + 2 cos ( 𝒌 ∙ 𝒂 𝟐 ) + 2 cos ⁡ ( 𝒌 ∙ ( 𝒂 𝟐 − 𝒂 𝟏 ) ) = ± 𝛾 √ 1 + 4 cos ( √ 3 𝑎 𝑘 𝑦 2 ) cos ( 𝑎 𝑘 𝑥 2 ) + 4 ⁡ cos 2 ( 𝑎 𝑘 𝑥 2 ) (1.3) (1.4) ( b ) ( a ) 5 Where γ is the hopping integral describing the interactions of the two π electrons between nearest neighbors. In the reciprocal lattice of the nanotube, the lattice vectors 𝑲 ⊥ (around the circumference) and 𝑲 ‖ (along the nanotube axis) should satisfy the following relations, 𝐂 ∙ 𝑲 ⊥ = 2 𝜋 , ⁡ ⁡ ⁡ ⁡ ⁡ ⁡ ⁡ 𝑻 ∙ 𝑲 ⊥ = 0 𝐂 ∙ 𝑲 ‖ = 0 , ⁡ ⁡ ⁡ ⁡ ⁡ ⁡ ⁡ ⁡ ⁡ ⁡ 𝑻 ∙ 𝑲 ‖ = 2 𝜋 , They can be rewritten as, 𝑲 ⊥ = ( − 𝑡 2 𝒃 𝟏 + 𝑡 1 𝒃 𝟐 ) 𝑁 , ⁡ ⁡ ⁡ 𝑲 ‖ = ( 𝑚 𝒃 𝟏 − 𝑛 𝒃 𝟐 ) 𝑁 , where N is the number of hexagons per unit cell. 5 Because of the periodic boundary conditions, ⁡ 𝑲 ⊥ of nanotube is quantized and only allowed k values for the electronic wavevector will give rise to possible electronic states in a one-dimensional nanotube, represented by the “cutting lines” as shown in Figure 1.3 (a). Each individual line corresponds to a different “band” within the Brillouin zone unit cell of CNT. If one of the cutting lines goes through the K point, the nanotube is metallic, without a band gap. If none of those cutting lines go through the K point, the nanotube will be semiconducting, as shown in Figure 1.3 (b)&(c). 8 (1.6) (1.5) 6 Figure 1.3. (a) Schematic of the electronic band structure of CNT and “cutting lines”. (b) Schematic of the electronic band structure of metallic and semiconducting CNT, respectively. 8 The density of electronic states (DOS) is inversely proportional to the derivative of E(k), resulting in van Hove singularities (VHSs) in the DOS of CNTs due to their 1D confinement of electrons, as shown in Figure 1.4 (a)&(b), which correspond to metallic CNT and semiconducting CNT, respectively. The electronic energy difference for transitions between singularities (symmetric about zero energy) are often labeled Eii, where i=1 represents the lowest energy transition between the van-Hove singularities closest to zero energy. The relationship between electronic transition energies Eii vs. CNT diameter have been shown in the Kataura plot in Figure 1.4 (c). 9 7 Figure 1.4. (a, b) Schematics of electronic dispersion and DOS function for metallic and semiconducting CNT. (c) Kataura plot. 9 Raman spectroscopy has been used extensively to study the physical and mechanical properties of carbon nanotubes. Due to the unique VHSs in the DOS of carbon nanotubes, Raman scattering of an individual carbon nanotube can be enhanced when the energy of incident photons matches the VHSs in the joint DOS for the valance and conduction bands. Unlike Rayleigh scattering, Raman processes are known to result in inelastic scattering, in which phonons with energy are either emitted or absorbed when the incident photons are scattered by atoms in the material. The two 8 main Raman scattering processes are defined as anti-Stokes and Stokes scattering correspondingly. Despite the large number of phonon modes in carbon nanotubes, only few of them are Raman-active. 5 Figure 1.5a shows typical Raman spectra for a semiconducting and a metallic carbon nanotube. There are 4 basic Raman peaks generally used to characterize carbon nanotubes, which are the radial breathing mode (RBM), D band, G band, and G’ band, as indicated in Figure 1.5a. The RBM corresponds to the radial motion of carbon atoms, as depicted in Figure 1.5c. Therefore, nanotube diameter can be calculated by empirical relation: 𝜔 𝑅𝐵𝑀 ~ 248 𝑑 𝑐𝑚 − 1 , ⁡ where d is the diameter of the CNT. The Raman shifts for RBM in my study are usually between 150 to 250 cm -1 range. The D band is due to defects in the CNTs, and the Raman shift for D band is around 1350 cm -1 . The G band has two components: G+ and G- bands, as shown in Figure 1.5b. Tangential vibrations of carbon atoms along the axial and circumferential directions result in two intense peaks around 1580 cm -1 , which correspond to longitudinal and transverse optical phonon modes, labeled as G+ and G-, as shown schematically in Figure 1.5d. 10 G’-band also called 2D-band, which is the double phonon mode of the D band. (1.7) 9 Figure 1.5. Raman spectra of CNT (a) RBM, D-band, G-band, and G’-band (b) G- and G+ mode of CNT (c) Schematic of RBM mode in CNT (d) Schematic of G- and G+ mode in CNT. 10 Photoluminescence (PL) is also an effective method to characterize CNT. It is the emission of photons due to inter-band transitions of electrons from the conduction to valence bands. The basic principle for photoluminescence process in CNTs is when a photon with energy equal to the E22 transition energy absorbed by CNT, it can excite an electron-hole pair into the E22 state, both electron and hole rapidly relax to E11 states in about 20-200 picoseconds, and then recombine into a photon with energy equal to E11 transition. 11 Due to low-lying dark exciton states, the nanotube photoluminescence (c) (d) 10 is intrinsically inefficient and is dependent on chiralities. Photoluminescence may be quenched by applying electric field along and perpendicular on CNTs and is very sensitive to the environmental effect. 12-13 Therefore, in order to see photoluminescence in CNTs, CNTs have to either be suspended from substrate or wrapped in surfactant. As shown in Figure 1.6, the “bright” nanotube has been suspended on quartz pillars and photo-illuminated at 1220nm. Figure 1. 6. PL image (a) and spectra (b) of suspended carbon nanotube (a) (b) 11 1.3 Two-Dimensional Materials The family of two-dimensional (2D) materials have attracted great research interest in the past decade, since the first isolation of graphene. 14 The appearance of each new material brings opportunities and challenges, as the properties of these materials are usually very different from those of their 3D counterparts. 2D materials also offer great flexibility in terms of tuning their electronic properties. Most importantly, 2D materials are useful building blocks that can be stacked mechanically or grown chemically as shown in Figure 1.7. 15 This is a unique benefit of layered 2D materials, while the nanotubes or nanowires does not have. Figure 1. 7. Building of Van der Waal heterostructures using 2D materials can be achieved by mechanically-assembled stacks (top) and large-scale growth by CVD or physical epitaxy (bottom). 15 12 However, the utility of graphene has been limited, due to its lack of an electronic bandgap, which has stimulated the search for 2D materials with semiconducting character. Transition metal dichalcogenides (TMDCs), which are semiconductors with the formula MX2, where M is a transition metal (such as Mo or W) and X is a chalcogen (such as S or Se), provide a promising alternative. TMDCs exhibit a unique combination of atomic-scale thickness, direct /indirect bandgap, and beneficial electronic and mechanical properties, which make them highly attractive for fundamental studies of novel physical phenomena and for a variety of applications in nanoelectronics and nanophotonics. 16 All TMDCs have a hexagonal structure, with each monolayer comprising three stacked layers (X-M-X). Different coordination spheres of the transition metal atoms result in different structural phases. The two most common structural phases are characterized by either trigonal prismatic (2H) (e.g., MoS2 andWS2) or octahedral (1T) (e.g., TiS2) coordination of metal atoms. 15 These structural phases can also be viewed in the way of different stacking orders of the three atomic planes (X-M-X) forming the individual layers of these materials. The 2H phases correspond to an ABA stacking in which chalcogen atoms in different atomic planes occupy the same position A and are located on top of each other in the direction perpendicular to the layer. On the other hand, the 1T phases are characterized by an ABC stacking order, as shown in Figure 1.8. 16 13 Figure 1. 8. Atomic structure of single layers of transition metal dichalcogenides (TMDCs) in their trigonal prismatic (2H) phase (left), and distorted octahedral (1T) phase (right). 16 Because of its robustness, MoS2 is the most studied material in this family. While bulk MoS2 is an indirect band gap semiconductor with a band gap of 1.3eV , quantum confinement converts monolayer MoS2 into a direct band gap material with a band gap of 1.85eV . 17 The calculated evolution of electronic band structures of 2H- MoS2 with the decrease of its thickness from bulk to monolayer has been shown in Figure 1.9. The positions of the valence and conduction band edges change upon reducing thickness, and the indirect bandgap semiconductor bulk material transits into a direct bandgap semiconductor monolayer material. 18 This band structure transition has been confirmed by photoluminescence spectroscopy, as shown in Figure 1.10. 17 14 Figure 1. 9. Calculated band structures of (a) bulk MoS2, (b) quadrilayer MoS2, (c) bilayer MoS2, and (d) monolayer MoS2. The solid arrows indicate the lowest energy transitions. 18 Figure 1. 10. (a) PL spectra for suspended mono-and bilayer MoS2 samples in the photon energy range from 1.3 to 2.2eV . Inset: PL quantum yield of thin layers for N=1– 6. (b) Normalized PL spectra by the intensity of peak A of thin layers of MoS2 for N=1– 6. 17 15 1.4 Thermoelectric Thermal management and energy crisis have been two major problems in this 21st century. The thermoelectric concept is seen as a perfect solution for the both issues since it’s a promising technique to convert the waste thermal energies into useful power. 19 Thermoelectricity incorporates two most common phenomena – the Seebeck effect, and the Peltier effect. German physicist Thomas Johann Seebeck observed direct conversion of heat into electricity in a junction of two different wires in 1821. This phenomenon is called the Seebeck effect, which is shown in Fig. 1.11(a), where an applied temperature difference drives charge carriers to diffuse from hot side to cold side, resulting in a current flow through the circuit. In 1834, French physicist Jean Charles Athanase Peltier discovered Peltier effect which is the presence of temperature difference at an electrified junction of two different conductors, as shown in Figure 1.11(b). 20 After these early discoveries in 1800s, many research projects were carried out in application of thermoelectricity. Started from early 1990s, thermoelectric research has been currently going through a period of renaissance, which stimulated a huge quest for improving its figure of merits (ZT) to cross 3 in order to make it commercially viable. The progress in nanotechnology and the use of semiconductor materials for thermoelectric applications has helped in pushing the science and engineering of thermoelectricity. Electronic band structure engineering and phonon mean free path engineering have helped push the value of ZT to even higher value in recent years. In 1995, Glen Slack introduced the concept of “phonon glass and an electron single crystal” 16 for design of optimal thermoelectric material. 21 This philosophy has driven research in the field to engineer high ZT materials and systems. Figure 1. 11. Schematic of Seebeck Effect (a) and Peltier Effect (b). 20 The most important parameter in thermoelectricity is the concept of Figure of merit, ZT, which is given by the following equation: 𝑍𝑇 = 𝑆 2 𝜎𝑇 𝐾 𝑒 + 𝐾 𝑙 , where 𝑇 is the temperature, 𝑆 is the Seebeck coefficient of the material, 𝜎 is the electrical conductivity, Ke and K 𝑙 represent electronic and lattice contributions to thermal conductivity, respectively. In metals, Ke dominates the thermal conduction, while in semiconductors and insulators, K 𝑙 dominates heat flow. 𝑆 2 𝜎 is also referred to as Power Factor (PF). The concept of ZT is analogous to Carnot efficiency 𝜂 for a （a ）（b ） (1.8) 17 conventional heat engine, which measures the amount of work a heat engine does per heat consumed. The power generation efficiency as a function of average ZT of both Seebeck effect and Peltier effect is shown in Figure 1.12. One can see that if ZTave =3.0 and ΔT = 400 K the power generation efficiency can reach 25%, comparable to that of traditional heat engines. 20 Figure 1. 12. Power generation/cooling efficiency as a function of average ZT of Seebeck Effect (a) and Peltier Effect (b). 20 In reality, optimizing ZT is a much harder problem because of intrinsic trade- offs that arise from the interrelationships between various parameters hidden underneath the simplicity of the equation. Based on the equation of ZT, it can be increased by improving the PF, while suppressing the thermal conductivity. To increase the 𝑃𝐹 = 𝑆 2 𝜎 , we have to improve conductivity and Seebeck coefficient simultaneously. However, there is a fundamental tradeoff between the Seebeck coefficient and the electrical conductivity, as shown in Figure 1.13. For a simple analysis, we can take （b ）（a ） 18 Seebeck coefficient as: 𝑆 𝑛 ( 𝑇 ) = − ( 𝑘 𝐵 𝑞 ) ( 𝐸 𝐶 − 𝐸 𝐹 ) 𝑘 𝐵 𝑇 ⁡,⁡ where 𝐸 C is the energy level in the conduction band, and EF is the Fermi energy. To have large Seebeck coefficient, we want 𝐸 𝐹 ≪ 𝐸 𝐶 , which is more reflective of a non- degenerate semiconductor, since 𝐸 𝐹 of degenerate semiconductor is much closer to 𝐸 𝐶 . In the meanwhile, electrical conductivity (e.g. n-type semiconductor) is given by: 𝜎 𝑛 = − ( 2 𝑞 2 ℎ ) 〈 𝑀 〉 〈 〈 𝜆 〉 〉 = 𝑛 0 𝑞𝜇 𝑛 ,⁡ we prefer the Fermi Level to be as close to conduction band as possible to acquire high electrical conductivity. Therefore, there exists an optimal distance between the fermi level and the conduction band. As the Figure 1.13 illustrates, the Seebeck coefficient is decreasing and the conductivity starts increasing as a function of the increase of carrier concentration, resulting in a maximum value of power factor at a certain carrier concentration. (1.9) (1.10) 19 Figure 1. 13. Illustration of dependence of Seebeck coefficient, electrical conductivity (top) and thermal conductivity (bottom) as a function of carrier concentration. 20 1.5 Photocatalysis Solar energy is an inexhaustible natural resource, with the magnitude of the available solar power striking the earth’s surface equal to 130 million 500 MW power plants. 22 Thus, harvesting energy from the sunlight directly, converting it to clean, transportable forms and the storage of energy has been one of the ultimate goals for scientists in the field of energy generation. Nowadays, solar electricity and solar-derived fuel from biomass become the main means to exploit the solar resource. While the latter provides the primary energy source for over a billion people, solar electricity is only 1.6% percent of the total electricity generated in the United States in 2018, as can be seen from Table 1-1. 23 The huge gap between the enormous demand in renewable energy and the present utilization of solar energy brings a great opportunity and challenge in developing cost-effective and efficient processes to harvest solar energy. As an effective usage of solar energy, artificial photosynthesis is a good replenish to bridge the gap. Water splitting and carbon dioxide reduction are good examples of artificial photosynthesis. Splitting water into hydrogen produces a clean fuel with the only by-product, water, which is environment friendly. Carbon dioxide reduction provides not only hydrocarbon fuels, but also a complete loop to recycle the greenhouse gas from hydrocarbon fuels including fossil fuels. Both routes are promising in renewable and environment friendly energy generation, but still require further development of efficiency and economically viability. 21 Table 1. 1. U.S. electricity generation by source, amount, and share of total in 2018 Energy Source Billion kWh Share of total Total-all sources 4178 Fossil fuels(total) 2651 63.5% Nuclear 807 19.3% Renewables (total) 713 17.1% Hydropower 292 7.0% Wind 275 6.6% Biomass (total) 63 1.5% Wood 41 1.0% Landfill gas 11 0.3% Municipal solid waste (biogenic) 7 0.2% Other biomass waste 3 0.1% Solar (total) 67 1.6% Photovoltaic 63 1.5% Solar thermal 4 0.1% Geothermal 17 0.4% Pumped storage hydropower -6 -0.1% Other sources 13 0.3% 22 Semiconductor is widely used for photo electrolysis such as solar water splitting and carbon dioxide reduction. The photon generated electron-hole pairs in the semiconductor can transfer their charger and energy to the reactants in the electrolyte and stimulate the photo-assisted decomposition or synthesis of chemicals. Figure 1.14 shows various redox potentials compared to the band gap and band edge positions of some semiconductors. The standard potentials for these redox couples indicate the thermodynamic limitations for the photoreactions that can be carried out. For reduction reaction, the conduction band edge of the semiconductor needs to be higher than the relevant reduction level. Opposite to reduction, the valence band edge needs to be lower than the oxidation potential of the species to perform the oxidation reaction. If the energy conditions we mentioned are not satisfied, the photo-excited carriers do not have enough energy to drive the reaction. Thus, an external bias voltage needs to be applied to help the reaction to be performed, which is commonly referred to as applying an overpotential. 24 Figure 1. 14. Energy band diagram of various semiconductors plotted with the redox of some chemical reactions. 24 23 Photocatalytic water splitting has arose great interest since early 1970s, after discovered by Fujishima and Honda under ultraviolet radiation. 25 As shown in Figure 1.15, when a semiconductor electrode is in contact with a solution containing a redox couple, the bending of both conduction band and valence band are caused by electrons flowing between the semiconductor and the solution. The charge transfer process stops when the Fermi level of the semiconductor and the solution reach the same level, establishing a thermodynamic equilibrium and a space charge layer is formed inside the semiconductor electrode. 22 When the semiconductor electrode is irradiated by light with energy higher than the band gap, electron-hole pairs are generated and separated in the space charge layer. The photo-generated electrons flow through the external circuit and hydrogen is produced at the surface of the working electrode, while the photo-generated holes move toward the interface of the counter electrode and the electrolyte and oxygen is formed. 26 Figure 1. 15. The band energetics of a semiconductor/liquid contact are shown in three cases: (A) before equilibration between the two phases and (B) after equilibration, but in the dark. 22 24 The standard reduction potentials (Er) of hydrogen evolution and the standard oxidation potentials (Eo) of oxygen evolution are as follows: Oxidation: 2H2O O2(g) + 4H + + 4e - Eo= 1.23V , pH=0 Reduction: 2H2O + 2e - H2(g) + 2OH - Er= -0.83V , pH=14 Standard reduction-oxidation potentials are measured under standard conditions: 25℃, partial pressure of 1 atm for each gas, 1M concentration for each ion in the reaction, assuming an anode chamber with 1M H + and a cathode chamber with 1M OH - , and metals in pure state. Based on the definition of standard potential, these two equations are at different pH, indicating pH of solution will affect the actual reduction potentials, which can be calculated by the following equation: E=Er +0.059pH, E=Eo -0.059pH, where E is the actual reduction-oxidation potential, Er is the standard reduction potential, Eo is the standard oxidation potential, “+” is used for the reduction half reaction and “-” is used for oxidation half reaction. As shown in Figure 1.16, the actual reduction potential of hydrogen evolution is -0.41V vs. Normal Hydrogen Electrode (NHE), which is potential of a platinum electrode in 1 M acid solution and the oxidation potential of oxygen evolution is 0.82V vs. NHE in pure water, where [H + ] = [OH - ] = 10 -7 M. At pH=0, the reduction potential needed to drive the reduction of hydrogen from water is 0.0V vs. NHE, and oxidation potential of oxygen from water is 1.23V vs. NHE. At pH=14, the reduction potential needed for hydrogen generation is -0.83V vs. NHE and for oxygen formation is 0.4V vs. NHE. 27 (1.11) (1.12) 25 Figure 1. 16. Electrode potential of water splitting as a function of the pH of the electrolyte, including equilibrium regions for water. 27 26 Chapter 2 Enhanced Photoluminescence in Air-suspended Carbon Nanotubes by Oxygen Doping This chapter is similar to Chen et al. 28 , published in Applied Physics Letters. 2.1 Abstract We report photoluminescence (PL) imaging and spectroscopy of air-suspended carbon nanotubes (CNTs) before and after exposure to a brief (20 second) UV/ozone treatment. These spectra show enhanced PL intensities in 10 out of 11 nanotubes that were measured, by as much as 5-fold. This enhancement in the luminescence efficiency is caused by oxygen defects which trap excitons. We also observe an average 3-fold increase in the D-band Raman intensity further indicating the creation of defects. Previous demonstrations of oxygen doping have been carried out on surfactant-coated carbon nanotubes dissolved in solution, thus requiring substantial longer ozone/UV exposure times (~15 hours). Here, the ozone treatment is more efficient because of the surface exposure of the air-suspended CNTs. In addition to enhanced PL intensities, we observe narrowing of the emission linewidth by 3-10nm. This ability to control and engineer defects in CNTs is important for realizing several optoelectronic applications such as LEDs and single photon sources. 27 2.2 Introduction Over the past few years, oxygen doping of carbon nanotubes (CNTs) through ozonolysis has been shown to produce localized exciton states that exhibit enhanced photoluminescence intensities (~20X), long photoluminescence lifetimes (>1nsec), and promising photon antibunching signatures for single photon emission, even at room temperature. 29-35 In the work of Htoon et al., 29 surfactant-coated carbon nanotubes suspensions were mixed with ozone saturated water and exposed to UV light for 17 hours. In their experiments, not only deep trap emission peak for oxygen-doped nanotubes in the 1.06-1.15 eV PL spectral range were observed, but also multiple sharp asymmetric emission peaks were reported at energies 50-300 meV redshifted from the E11 bright exciton peak. In the work of Weisman et al., 31 surfactant-coated carbon nanotubes were exposed to ozone and white light for up to 16 hours. They found that the oxygen-doped single-walled carbon nanotubes (SWCNTs) are much easier to detect than pristine SWCNTs because they give stronger near-infrared emission. This work also showed distinct near-infrared fluorescence at wavelengths 10 to 15% longer than displayed in pristine semiconducting SWCNTs. In the work of Vialla et al., the PL spectra of nanotubes show the emergence of sharp peaks at low temperatures corresponding to localized excitons coupling to phonons. 32 The work of Hofmann et al. showed that, at low temperature, excitons in suspended nanotubes become localized around defect sites and exhibit long-lived luminescence emission, greater than 3nsec. 33 Similar results can also be found in the work of Sarpkaya et al., who reported PL lifetimes up to 18nsec from localized excitons. 34 In the work of Ma et al., oxygen dopant 28 states were controllably introduced, leading to extend lifetimes and single photon emission even at room temperature. 35 While these previous results are exciting and demonstrate the ability to control defects in CNTs for improved light emission, the surfactant-coated geometry is not ideal for device fabrication. The emission wavelength of these previous studies remained close to or below 1000nm. In the work presented here, we study enhanced emission out to the 1550nm wavelength range, which is compatible with fiber optic telecommunications. 2.3 Experimental Details Here, we apply a UV/ozone treatment to air-suspended carbon nanotubes. Because of the exposed surface of the CNTs, sufficent O-doping for enhanced PL is achieved in just 20 seconds, 3 orders of magnitude more rapidly than previous attempts on surfactant-coated CNTs dissolved in solution. 36-42 Since ozone can damage carbon nanotubes, 43-44 care was taken to minimize the exposure required to produce localized excitons. Raman spectra are also collected before and after the UV/ozone treatment in order to further characterize the defect creation process. 40, 45-46 In this work, 2µm × 2µm pillars are etched 8µm deep in a quartz substrate using a Cl-based reactive ion etch plasma, as shown in Figure 2.1. 1nm of Fe is deposited on the pillars to serve as a catalyst for the nanotube growth. Carbon nanotubes are grown by chemical vapor deposition (CVD) at 825 o C using ethanol as the carbon feedstock, 47- 48 resulting in CNTs suspended across adjacent pillars, as show in Figure 2.1d. PL 29 images were taken using a defocused 785nm laser in conjunction with a thermoelectrically cooled InGaAs camera (Xenics, Inc.) with a 5 second integration time. The ozone treatment was carried out in a UV/ozone cleaner (ProCleaner, BioForce, Inc.) using ambient air at 1atm for 20 seconds. This UV/ozone cleaner uses a 10.15W mercury, non-phosphorus coated quartz lamp ( ～4.6mW/cm2 at the sample surface). The lamp spectrum is mostly concentrated in the 180nm-440nm range peaking at 254nm. Raman spectra and PL spectra were collected using a Renihsaw InVia micro- spectrometer equipped with both silicon and InGaAs detector arrays. Scanning electron microscope (SEM) images (Figure 2.1d) were taken after all optical characterization, in order to avoid quenching the CNT photoluminescence. An optical microscope image of a 4 × 5 array of quartz pillars is shown in Figure 2.1a. Figure 2.1b shows a PL image taken from the same region of this sample before ozone treatment. A faint line can be seen connecting two adjacent pillars, corresponding to PL emission from a suspended CNT. Figure 2.1c shows a PL image of the same sample after ozone treatment, exhibiting a much brighter line, indicating substantial enhancement in the PL emission. In addition to the bright line in this image, several new bright spots appear originating from other CNTs that were previously too dim to see. Figure 2.1d shows an SEM image of the suspended CNT connecting two adjacent quartz pillars. 30 Figure 2. 1. (a) Optical microscope image of quartz pillar arrays. Photoluminescence image of suspended CNT (b) before and (c) after UV/ozone treatment. (d) SEM image of CNT suspended on top of the pillars 31 2.4 Results and discussion Figure 2.2 shows the Raman and PL spectra of the suspended CNT shown in Figure 2.1. A normalized PL spectrum is also included in Supplemental Material 49 as Figure S2.1. A sharp radial breathing mode (RBM) can be seen at 150cm-1 (FWHM=10cm-1). While the frequency of this mode remains essentially unchanged after the ozone treatment, we observe a substantial linewidth broadening after UV/ozone treatment, most likely due to sample inhomogeneity. The PL spectra exhibit a sharp peak centered at 1530nm (FWHM=28nm). The linewidth was fitted without including the shoulder at 1510 and 1560nm, which are possibly due to phonon sidebands. In previous literature, similar spectral wings have been attributed to phonon sidebands emerging at low temperature, but were not observed at room temperature. 32, 50 This peak can be seen to increase in intensity by a factor of 5.1X, in addition to a linewidth narrowing of 10nm after UV/ozone treatment. The PL spectra peak position remains nearly unchanged after UV/ozone treatment. We believe the PL intensity enhancement shown in Fig. 2.2b originates from the introduction of oxygen atoms as dopants in our suspended CNTs. Those oxygen atoms create perturbed regions in the CNT that trap excitons and result in more efficient radiative recombination. Thus, the PL intensity enhancement and linewidth narrowing can be observed after UV/ozone treatment. Unlike the previous reports in the literature, there are no new red-shifted features in the PL spectra. Enhanced photoluminescence emission was observed in 10 out of a total of 11 individual air-suspended carbon nanotubes after a 20-second UV/ozone treatment. These results are summarized in Table 2.1. Here, an average 32 increase in PL intensity of 3X is observed as well as a linewidth narrowing from an average of 18.6nm to 13.6nm after UV/ozone exposure. While most of the nanotubes show linewidth narrowing, samples 7 and 10 show a linewidth broadening, which could be due to the presence of many defects causing inhomogeneous broadening. The photoluminescence emission from the nanotubes in this study lie in the wavelength range from 1265-1530nm. This range is largely determined by the finite diameter distribution and the fixed excitation laser wavelength (785nm), which samples only a small portion of the Kataura space. In order to extend these wavelengths to 1550nm, different diameters should be selected, and a longer excitation wavelength should be used. Typically, trapped excitons are lower in energy by several kBT with respect to free exciton states. However, here, we do not observe any systematic redshift in our spectra. One possible explanation for this is that the excitons are getting imprisoning between two reflecting O-defects. Another possible explanation is that the O-plasma treatment is simply cleaning the surface of these nanotubes (i.e., removing non-radiative recombination centers), thus not producing a redshift. 33 Figure 2. 2. (a) Radial breathing mode (RBM) Raman spectra and (b) PL spectra of CNT before (black) and after (red) exposure to the UV/ozone treatment. 100 120 140 160 180 200 0 5000 10000 15000 20000 25000 Raman Intensity Raman Shift (cm -1 ) RBM After Before 1460 1480 1500 1520 1540 1560 1580 1600 0 1000 2000 3000 Before After PL Intensity Wavelength (nm) before ozone treatment after ozone treatment 5X 34 Table 2. 1. The PL intensity, linewidth, and center wavelength for 10 suspended CNT samples before and after UV/ozone treatment. Sample RBM (cm -1 ) PL Intensity Peak Height PL Linewidth (nm) PL Center Wavelength (nm) before after before after before After 1 150 3413 8663 23.1 14.1 1508 1504 2 151 1123 3631 21.4 12.7 1424 1422 3 187 628 3211 28.8 18.2 1530 1530 4 132 449 2189 28.6 18.7 1528 1530 5 148 1553 6735 16.6 11.8 1476 1473 6 238 3619 17148 14.0 10.2 1287 1284 7 227 2606 8180 8.5 11.7 1265 1268 8 201 3737 8776 19.0 13.5 1476 1472 9 203 3747 7458 15.2 11.2 1352 1352 10 202 3634 8448 11.1 14.5 1420 1420 Average 179 2451 7444 18.6 13.6 1427 1425 35 The D- and G-band Raman spectra taken before and after UV/ozone treatment are plotted in Figure 2.3. Here, we see a 3-fold increase in the D-band Raman intensity, corresponding to the introduction of defects, while the intensity of the G-band does not change after UV/ozone treatment. Table 2.2 lists a summary of changes observed in the D-band Raman intensity and G-band linewidth after UV/ozone treatment. While substantial changes in the G-band linewidth (both broadening and narrowing) are observed, no symmetric shift is observed in the ten samples measured in this study. All ten of these nanotubes, however, showed a substantial increase in the D-band Raman intensity. We also performed longer UV/ozone exposures to the photoluminescent pristine CNT for 2 minutes. After the 2-minute UV/ozone exposure (Figure S2.2 of the Supplemental Material 49 ), all of the bright spots disappear. Previous demonstrations of oxygen doping have been carried out on surfactant-coated carbon nanotubes dissolved in solution, thus requiring substantial longer ozone/UV exposure times (~15 hours). Here, the ozone treatment is more efficient because the surface is exposed in these air- suspended CNTs. Therefore, there is an inherent limitation in the UV/ozone treatment of these air-suspended nanotubes. Once a substantial amount of defects are formed, these delicate hanging structures simply collapse, which quenches their photoluminescence. As such, it is unlikely that the high densities of O-defects previously obtained in surfactant coated nanotubes can be achieved using this approach. 36 Figure 2. 3. D-band and G-band Raman spectra of CNT before (black) and after (red) exposure to the UV/ozone treatment. 1200 1300 1400 1500 1600 1700 1000 2000 3000 4000 5000 6000 7000 8000 Raman Intensity Raman Shift (cm -1 ) before UV/ozone treatment after UV/ozone treatment D-Band G-Band 37 Table 2. 2. The D-band intensity and G-band linewidth of 10 suspended CNT samples before and after UV/ozone treatment. Sample D-band Intensity G-band Linewidth (cm -1 ) before after before after 1 1399 5911 8.4 12.5 2 334 975 10.5 14.7 3 3572 7118 8 10.4 4 3045 7018 9.8 7.5 5 732 2361 9.6 10.5 6 1167 8255 25.7 13 7 576 1288 10.8 13.2 8 854 2658 11 14.5 9 1307 3577 10.4 13.6 10 750 2479 12.3 15 Average 1374 4164 11.7 12 2.5 Conclusion In summary, we demonstrate a reliable and scalable method using a UV/ozone treatment to enhance the photoluminescence intensity of air-suspended carbon nanotubes. Here, a 20-second UV/ozone exposure gives enhanced PL intensities by up to a factor of 5X. This is considerably more rapid than oxygen doping on surfactant- 38 coated carbon nanotubes dissolved in solution reported in previous literature. The enhanced photoluminescence originates from oxygen defects that locally trap excitons, increasing their radiative recombination efficiency. This UV/ozone treatment also results in a 3-fold increase in the D-to-G band Raman intensity ratio, further corroborating the creation of defects. The ability to control and engineer defects in CNTs is important for realizing several optoelectronic applications such as LEDs and single photon sources. This research was supported by the Department of Energy DOE Award No. DE- FG02-07ER46376 (J.C.), NSF Award No. 1402906 (S.Y .) and Northrop Grumman- Institute of Optical Nanomaterials and Nanophotonics (NG-ION2) (B.W). We also gratefully acknowledge the support of Karen Trentelman and Catherine Patterson at The Getty Center for use of their InGaAs detector and Dr. Brutchey at USC for the UV/ozone generator. 39 Chapter 3 Highly efficient, high speed vertical photodiodes based on few-layer MoS 2 This chapter is similar to Chen et al 51 , published in 2D Materials 3.1 Abstract Layered transition metal dichalcogenides, such as MoS2, have recently emerged as a promising material system for electronic and optoelectronic applications. The two- dimensional nature of these materials enables facile integration for vertical device design with novel properties. Here, we report highly efficient photocurrent generation from vertical MoS2 devices fabricated using asymmetric metal contacts, exhibiting an external quantum efficiency of up to 7%. Compared to in-plane MoS2 devices, the vertical design of these devices has a much larger junction area, which is essential for achieving highly efficient photovoltaic devices. Photocurrent and photovoltage spectra are measured over the photon energy range from 1.25 to 2.5 eV , covering both the 1.8 eV direct K-point optical transition and the 1.3 eV Σ-point indirect transition in MoS2. Photocurrent peaks corresponding to both direct and indirect transitions are observed in the photocurrent spectra and exhibit different photovoltage–current characteristics. Compared to previous in-plane devices, a substantially shorter photoresponse time of 7.3 μs is achieved due to fast carrier sweeping in the vertical devices, which exhibit a −3 dB cutoff frequency of 48 kHz. 40 3.2 Introduction 2D layered materials, including graphene, transition metal dichalcogenides (TMDCs), and black phosphorous, have attracted great research interest in the past decade. Unlike graphene, TMDCs such as MoS2 exhibit finite bandgaps in the visible wavelength range. 17-18, 52-53 Furthermore, the band structures of MoS2 and other TMDCs depend on the number of layers of material. 17 While most research on MoS2 has focused on the indirect-to-direct band gap transition occurring in monolayer MoS2, this has an inherently small optical density that limits its potential application in practical optoelectronic devices. Few-layer MoS2 films, on the other hand, provide substantially larger optical densities than monolayers and can withstand substantially higher injection currents, which are advantageous for solar energy conversion and light emitting diode applications. Various MoS2 devices have been fabricated and studied to improve the performance of MoS2, including graphene/MoS2, 54 monolayer TMDC stacks, 55-57 and MoS2/metal junctions. 58-60 Because of the absence of dangling bonds at the interfaces of two-dimensional materials, the Schottky junction between metals and MoS2 more ideal’ than 3D semiconductors like Si or III–V compounds. 61-62 This is an essential advantage of TMDCs, which enables the possibility of building vertical MoS2 devices with atomically clean and sharp interfaces, leading to unique physics, such as split excitons, 63 interlayer charge transfer, 56 and plasmonic-exciton interactions. 58-60 In previous studies of in-plane MoS2 photodetectors, 64-67 the charge separation region of the device is defined by the Schottky junction between the metal and the MoS2. As a result, only incident light shining within the depletion width, which is only 41 approximately 100 nm, can generate a photocurrent. To improve this, comb-shaped source and drain metal electrodes have been introduced to increase the effective charge separation region. 67 Nevertheless, the effective area is still a small fraction (less than 1%) of the device footprint. In addition, the efficiency of monolayer in-plane devices is further limited by the small optical density of monolayer MoS2. Lastly, a gate bias and large drain-source bias voltages (10 V) are essential for obtaining high sensitivity in these in-plane devices, which gives rise to high power consumption and may eventually degrade the device. 68 3.3 Experimental details In the work presented here, we use few-layer instead of monolayer MoS2, with a vertical device structure. By utilizing few-layer MoS2 in a cross-plane geometry, the optical density is enhanced, and the effective charge separation region is increased to the entire volume of the MoS2 flake. As a result, an external quantum efficiency (EQE) of up to 7% is achieved, which is more than 100 times larger than previous reports based on monolayer MoS2 flakes. 69 Also, the trapped charge and surface states in the in-plane devices result in large time delays, which limit the cutoff frequency to less than 200 Hz. 65 In the vertical few-layer devices presented here, the TMDC is insulated from moisture and other kinds of dopants in air by the top electrode. This reduces the delay time because there is a smaller amount of surface charge and unintentional dopants. Also, the shorter channel lengths in the vertical devices make it easier for the remaining 42 carriers to be collected by the electrodes when the light is off, further reducing the response time of the device. A substantially lower delay is obtained using the AC lock- in technique, which exhibits a −3 dB cutoff frequency of 48 kHz, corresponding to time response of 7.3 μs (tr = 0.35/ fcut-off). The devices measured in this work are fabricated using a dry transfer technique, as shown in Figures 3.1(a) and (b). 70 First, the MoS2 flake is exfoliated onto a piece of PDMS mounted on a glass slide. Target MoS2 flakes are identified using an optical microscope. The MoS2 flake is then transferred from the PDMS stamp onto a Si/SiO2 (300 nm) substrate with pre-patterned metal electrodes (1 nm/30 nm Ti/Au) using a home-built contact aligner. After the transfer, a semi-transparent 10 nm Pd film is deposited as the top electrode. After fabrication, photoluminescence (PL) spectra are measured from the few-layer flake, as shown in Figure 3.1(c). In addition to the PL peaks at 1.8 eV and 2.0 eV corresponding to the direct band transitions at the K-point of the Brillouin zone, 17-18 there is another weaker peak around 1.3 eV , which corresponds to the indirect transition at the Σ-point. In this fabrication scheme, different metals (with different work functions) are used for the top and bottom electrodes of the MoS2 flake. Au is used for the bottom electrode while top electrode is Pd. According to previous studies on graphene and MoS2, 71-72 an asymmetric contact for the source and drain will break the mirror symmetry of the internal electric fields in the channel and improve the photovoltaic performance of the device. Given that the work function of Pd is 5.6 eV , and the electron affinity of MoS2 is about 4.4 eV , 73 the Pd serves as a hole- doping contact for MoS2. On the other hand, Au is electron-doping, so the two Schottky 43 junctions formed at these contacts produce an additive photovoltaic effect. 54, 72 Based on the optical properties of Pd films in the visible range, 74-75 the absorption coefficient of Pd at visible range is about 50–80 μm −1 . As a result, only about 50% of the incident illumination will penetrate the 10 nm Pd top electrode film. We expect there to be a tradeoff between optical density (which increases with thickness) and series conductance (which decreases with thickness). Another constraint here is that the thickness of the top Pd electrode must be thicker than the MoS2 flake in order to be provide a continuous electrode at the step edge (i.e., boundary) of the MoS2 flake. This means that if we further increase the thickness of the MoS2 flake, we need to deposit a thicker top metal electrode, which will very quickly become opaque. In general, MoS2 flakes around 10 nm are best for our device structures. 44 Figure 3. 1. (a) Schematic diagram of the device fabrication process. First, the MoS2 is exfoliated from bulk crystal onto a PDMS substrate mounted on a glass slide. After identifying the desired few-layer flake, the sample is transferred from PDMS onto a Si/SiO2 substrate with pre-patterned metal electrodes. The top electrode is then patterned using electron-beam lithography. (b) Optical microscopy image of the vertical MoS2 device after the fabrication. The dashed line indicates the semi- transparent top electrode. (c) Photoluminescence spectrum of the few-layer MoS2 sample after transfer. 3.4 Results and discussion Figure 3.2(a) shows photocurrent spectra measured from the few-layer MoS2 vertical device. As in the PL spectra shown in Figure 3.1(c), the dominant photocurrent peaks start around 1.8 eV and extend to 2.2 eV , corresponding to the direct transitions at the K-point in the Brillouin zone. Interestingly, a few lower energy peaks are 45 observed around 1.2–1.4 eV , likely due to indirect transitions at the Σ-point, which have not been reported previously. The power spectrum of the incident laser light is plotted in Figure 3.2(b). Given an effective illumination of 50%, considering the light absorption in the top Pd electrode, the EQE is calculated and plotted together with the power spectrum in Figure 3.2(b). As a comparison, previous photocurrent spectra measured from monolayer MoS2 exhibited EQEs of about 2.5×10 −4 , 69 mainly due to low optical density of the monolayer and also the small area of the in-plane charge separation region. From the vertical few-layer MoS2 device with asymmetric metal contacts, an EQE of up to 7% is achieved. In addition to the short circuit current, we also measured the open circuit voltage from the device, as plotted in Figure 3.2(c). The open circuit voltage of the device is relatively small, less than 80 mV . It is known that a Schottky photodiode usually suffers from low output voltage, because of the large dark saturation current. 76-77 In a pn-junction photodiode, the open circuit voltage obeys the following relation: 𝑉 𝑜𝑐 = 𝑘𝑇 𝑞 𝑙𝑛 ( 1 + 𝐼 𝑠𝑐 𝐼 𝑜 ) ≈ ⁡ 𝑘𝑇 𝑞 ⁡𝑙𝑛 𝐼 𝑠𝑐 𝐼 𝑜 ,⁡ where Isc is the short circuit current, Io is the dark saturation current, and Io<<Isc. In our vertical MoS2 device, the thickness of the few-layer MoS2 is only 10 nm (see Figure S3.1), which is well below the depletion width of the corresponding bulk semiconductor device. In this case, the leakage current caused by the drift of photoexcited carriers is not a constant but related to the incident photon number. As a result, the tunneling effect makes this device very leaky, and the V oc is small, only a few kT. The V oc–Isc relation is plotted in Figure 3.2(d). Instead of a logarithmic relationship, the V oc–Isc relation is 46 rather linear. Also, we observe two distinct behaviors in the relation between V oc and Isc, which varies with different incident photon energy. The data points associated with incident photon energies between 1.75 and 2.25 eV (K-point transition) have a different slope than those obtained with photon energies between 1.25 and 1.4 eV (Σ-point transition). This difference can be explained by considering the different band gap energies of these two optical transitions. Since the band gap at the K-point is approximately 0.5 eV higher than the Σ-point band gap, we expect the open circuit voltage to be higher in the case of the K-point excitation. Lastly, we estimate the electrical output power and energy conversion efficiency of the device (see Figure 3.2(e)) based on both the short circuit current and the open circuit voltage, and by assuming a fill factor of 0.25. With 7 μW total incident optical power, 4 nW of electrical power can be extracted from the device. It is not surprising that the energy conversion efficiency of the device is relatively low due to the small thickness of the device and a large leakage current. 47 Figure 3. 2. (a) Photocurrent and (c) photovoltage spectra of the vertical MoS2 device. In addition to the main peak around 1.8–2.1 eV, lower energy peaks at 1.4 and 1.5 eV are also observed due to the indirect transition. (b) Power spectrum of the incident laser plotted together with the EQE of the device. The two peaks at 1.85 and 2.0 eV correspond to the K-point transition. The EQE of the direct transition is about 10 times larger than the indirect transition. (d) Isc–Voc plot showing that the direct band transition (red points) and the indirect band transition (blue points) give different Isc– Voc relations. (e) Extracted electrical power and energy conversion efficiency estimated from the product of the short circuit current and open circuit voltage, assuming a fill factor of 0.25. 48 As we mentioned above, the thickness of few-layer MoS2 is much smaller than the ‘depletion width’ and the minority carrier diffusion length. Because of this, the I–V characteristics measured from this device are exhibit a linear dependence, as shown in Figure 3.3, instead of rectifying behavior. This linear behavior also limits the energy conversion efficiency of the device by restricting the fill factor to just 0.25. Figure 3.3 shows the photo-I–V characteristics of a device under different incident laser powers (532 nm wavelength, power 0.92, 12, and 108 μW). With higher incident optical powers, the Isc is increased from 1.3 to 400 nA, and V oc is increased from 1 to 100 mV . Figure 3. 3. (a) Photo-I–V characteristics of the vertical MoS2 device under 532 nm illumination. (b) Log plot of the photo-I–V data. 49 Compared to an in-plane MoS2 device, the very short ‘channel length’ of the vertical device, which is smaller than the depletion width, limits the device’s performance as a diode. On the other hand, the short channel length also increased the conductance of these vertical few-layer devices by more than 1000 times, compared to in-plane MoS2 photodetectors. 65-67 As such, our vertical device can operate at a lower voltage and does not require the application of an externally applied gate voltage, both of which reduce the power consumption significantly A photoresponsivity of 10 mAW −1 is measured at a bias voltage of 0.1 V (under an optical power density of 1 μW μm –2 ), which is more than 10X larger than the maximum value measured from an in-plane MoS2 device under the same bias voltage. 64, 67 The very short channel length and built-in electric field in the depletion region help eliminate the photon generated carriers and trapped charges when the illumination is off. This makes the vertical device response much faster than a corresponding in- plane device. The time-resolved photoresponse of the device is shown in Figure 3.4(a). Here, a 680 nm diode laser (Communications Grade VCSEL) is used as the light source, modulated by a function generator (Agilent EE4432B) at 1 kHz. The output open circuit photovoltage and its rise and fall time are measured using an oscilloscope (Agilent Infiniium MSO8104A), as plotted in Figure 3.4(a). To further investigate the frequency response of the vertical MoS2 device, an AC measurement is performed using a lock-in amplifier. The frequency of the light modulation is varied from 100Hz to 100 kHz, which shows a cutoff frequency of 48 kHz, showing Figure 3.4(b). The peak in the frequency response of this device around 20 kHz is related to the RLC resonance of the 50 circuit, as described in more detail in the Supplemental Document. Here, our device is operated without any applied gate or bias voltage and exhibits an open circuit voltage cutoff frequency of 48 kHz. The cutoff frequency measured from the lock-in shows that the response time of our device is 7.3 μs, which is over 10 times faster than the best result from corresponding in-plane device. 65, 67 Figure 3. 4. (a) Photovoltage plotted as a function of time under modulated 680 nm laser illumination. The photovoltage signal is processed after FFT filtering to remove high frequency noise, with the cutoff frequency of the FFT filtering set to 100 kHz. (b) The frequency response of the open circuit photovoltage, which shows a cutoff frequency at 48 kHz. Vo is the photovoltage measured with continuous light illumination. 51 3.5 Conclusion In summary, we have shown that vertical metal/MoS2/metal devices with asymmetric metal contacts can be used for highly efficient photocurrent generation and photodetection, with an EQE of up to 7%. In contrast to typical values of in-plane devices which span a wide range from 0.02% to 0.1% under the same illumination power density. 69, 78 For the first time, the photocurrent spectra are measured from a few- layer MoS2 device exhibiting lower energy photocurrent peaks around 1.2–1.4 eV , corresponding to the Σ-point indirect transition. A photoresponsivity of 10 mAW−1 is measured at a bias voltage of 0.1 V . The vertical device can be run at zero bias voltage without any gating, which reduces the power consumption significantly compared to the corresponding in-plane MoS2 devices. Also, we performed the first frequency response measurement on vertical MoS2 photodetectors, which show a bandwidth of 48 kHz. The cutoff frequency may be further increased by optimizing the device design to reduce the parasitic capacitance. The I–V characteristics of the device show a linear behavior because the thickness of the few-layer MoS2 is smaller than the ‘depletion width’ and the minority carrier diffusion length. Increasing the thickness of the MoS2 flakes will reduce the leakage current and result in more rectifying behavior. The metal/MoS2/metal vertical structure presents a more effective design layout with a larger photovoltaic area than corresponding in-plane devices and can readily be integrated with silicon microelectronics. 52 Methods Fabrication of vertical few-layer MoS 2 device: Few-layer MoS2 films were exfoliated from bulk MoS2 (SPI Supplies) onto PDMS based gel (Gel-Pack®, #PF-3-X4) mounted on a glass slide with pre-patterned alignment markers using the ‘Scotch tape’ method. 14 After identifying the target MoS2 flake, the glass slide with the film is then mounted in a X–Y–Z micromanipulator. With the alignment markers on the glass slide, the MoS2 can be readily located using an optical microscope and then the transferred onto a Si/SiO2 (300 nm) substrate with pre- patterned metal electrodes (1 nm Ti and 30 nm gold), as illustrated in Figure 3.1(a). The top electrode is fabricated using electron-beam lithography followed by 10 nm Pd deposition, as shown in Figure 3.1(b). Photoluminescence spectroscopy and photo-I–V measurement: Photoluminescence spectra and photo-I–V characteristics are taken using a Renishaw InVia spectrometer with a 532 nm laser focused through a 100× objective lens. PL spectra are collected at room temperature, under ambient conditions. The I–V characteristics are recorded using an HP 4145B semiconductor parameter analyzer. Photocurrent and photovoltage spectra: Photocurrent spectra are collected using a Fianium supercontinuum white light laser source in conjunction with a Princeton Instruments double grating monochromator to provide monochromatic light over the 450–1000 nm wavelength range. A Keithley 53 2401 SourceMeter® is used for the electrical measurement. The laser power is measured using a ThorLabs PM100D power and energy meter. Time response measurement and frequency response measurement: A 680 nm diode laser (Communications Grade VCSEL) is used for the measurement. The turn on voltage of the diode laser is 2.2 V . A function generator (Agilent EE4432B) is used as the power supply of the diode laser. The output signal of the function generator is set as a 1 kHz square wave. The output photovoltage is measured using an oscilloscope (Agilent Infiniium MSO8104A). In the frequency response measurement, a lock-in amplifier (Stanford Research Systems, SR830) is used as the voltage supply for the diode laser, while the frequency is swept from 100 Hz to 100 kHz. This research was supported by Department of Energy (DOE) Award No. DE-FG02– 07ER46376 (ZL) and NSF Award No. 1402906 (JC). 54 Chapter 4 Enhanced Thermoelectric Efficiency in Topological Insulator Bi 2 Te 3 Nanoplates via Atomic Layer Deposition-based Surface Passivation This chapter is similar to Chen et al 79 , published Applied Physics Letter 4.1 Abstract We report in-plane thermoelectric measurements of Bi2Te3 nanoplates, a typical topological insulator (TI) with Dirac-like metallic surface states, grown by chemical vapor deposition (CVD). The as-grown flakes exposed to ambient conditions exhibit relatively small thermopowers around -34µV/K due to unintentional surface doping (e.g., gas adsorption and surface oxidation). After removal of the unintentional surface doping and surface passivation by deposition of 30nm of Al2O3 using atomic layer deposition (ALD), the Seebeck coefficient of these flakes increases by a factor of 5X to -169µV/K. Here, we believe that the ALD-based surface passivation can prevent the degradation of the thermoelectric properties caused by gas adsorption and surface oxidation processes, thus, reducing the unintentional doping in the Bi2Te3 and increasing the Seebeck coefficient. The high surface-to-volume ratio of these thin (~10nm thick) nanoplates make them especially sensitive to surface doping, which is a common problem among nanomaterials in general. An increase in the sample resistance is also observed after the ALD process, which is consistent with the decrease in doping. 55 4.2 Introduction The potential use of nanoscale materials to enhance the thermoelectric properties of energy conversion dates back to the theoretical predictions of Hicks and Dresselhaus in 1994. 80-81 Mechanisms of enhancement include increased density of electronic states in low dimensions, 82 increased phonon scattering 83 , and quantum confinement-induced band gap engineering. 84 However, in practice, reducing materials and devices to low dimensions (or nanoscale dimensions) often presents deleterious effects associated with poor electrical contacts, 85 surface depletion, 86-87 and unintentional doping, 88 which remain secondary concerns in bulk material systems. For example, the thermoelectric performance of monolayer MoS2 varies more widely than bulk MoS2. 89-91 In fact, nanoscale forms of Bi2Te3 typically have lower ZT values that their bulk counterparts due to these issues. 92-94 Oxygen plasma treatment of 2D materials has been shown to improve the materials properties, through the remediation of unintentional doping. Dhall et al. shows a 16-fold increase in the luminescence efficiency of MoS2 after exposure to a brief oxygen plasma and a shift in the threshold voltage of as much as 18V in MoS2- based field effect transistors. 57, 88 Javey’s group reported an air-stable, solution-based chemical treatment using an organic non-oxidizing superacid (bis(trifluoromethane) sulfonimide (TFSI)), which uniformly enhances the photoluminescence and minority carrier lifetime of MoS2 monolayers by more than two orders of magnitude. 95 Pettes et al. explored the effects of surface band bending and scattering on thermoelectric transport in suspended bismuth telluride nanoplates using microfabricated heaters and 56 thermometers. 96 They saw the direct effect of surface scattering and surface doping as a function of Bi2Te3 nanoplate thickness. They also reported that both chemical alloying and surface potential modification can play important roles in optimizing the thermoelectric properties in ultrathin (Bi1−xSbx)2Te3 nanoplates as well. 97 In the work presented here, we report a facile method to improve the thermoelectric efficiency of Bi2Te3 nanoplates through remediation of unintentional surface doping. 4.3 Experimental details In this work, two dimensional (2D) single crystal nanoplates (NPs) of bismuth telluride were synthesized by a catalyst-free vapor-solid method following an approach similar to that of Kong et al. 98 Bulk Bi2Te3 was placed in the center of a 1 inch fused quartz tube in a hot wall furnace, and silicon substrates coated with 280 nm thermally grown SiO2 were placed approximately 10-14 inches downstream. The quartz tube was evacuated with a mechanical pump, and then the temperature was slowly ramped (1- 2 o C min -1 ) to 450-480 o C under flowing Ar. The pressure was maintained in the range of 20-70 Torr for the duration of the synthesis before slowly cooling back to room temperature. Typical resultant 2D crystals are hexagonal or triangular in shape with sizes on the order of 2-50 μm and uniform thicknesses of 3-50 nm. Flakes are then transferred to an oxidized silicon wafer using a home-built transfer setup consisting of a sharp tungsten tip to transfer the nanoplate from the growth substrate to the device substrate. Top electrodes, heaters, and 4-probe resistance thermometers (RTDs) are then 57 patterned using electron beam lithography followed by metal deposition of Au, as shown in Figure 4.1. Optical microscope images of the sample before and after Al2O3 deposition are shown in Figure 4.1. Before 30nm Al2O3 ALD deposition, fabricated devices are immersed into buffered hydrochloric acid (HCl) to remove the native oxide and surface adsorbates. For 10:1 HCl, the typical etching rate of oxide is 23nm/min. We expect that our 4-second HCl dip is just enough to etch away the 1-2 nm native oxide, while having a minimal impact on the thickness of the Bi2Te3 flake. This acid dip is followed immediately by an evaporation of 1nm Al passivation, which improves the nucleation of the ALD layer. After the ALD process, a clear contrast can be seen in the Bi2Te3 nanoplate after the ALD process. Interestingly, there is no change observed in the Raman spectra, as shown in Figure S4.1 of the Supporting Information, indicating that there is no significant change in the lattice structure after the ALD process. Figure 4. 1. Schematic diagram of the device fabrication process (a) and optical image of the device before (b) and after (c) Al2O3 deposition. Heater Bi 2 Te 3 Flake RTD (a) (b) (c) Al 2 O 3 Deposition 20μm 20μm 58 4.4 Results and discussions The resistance vs. temperature relation of each 4-probe RTD are calibrated in a temperature-controlled (Linkam) stage before and after depositing Al2O3, as shown in Figure 4.2. The RTD resistance (40Ω) is two orders magnitude smaller than the in-plane resistance of the Bi2Te3 nanoplate (8kΩ), ensuring that there is negligible current leakage in and out of the Bi2Te3 nanoplates from the metal thermometer line. Figure 4. 2. Temperature calibration of the right metal RTD of the Bi2Te3 device shown in Figure 1. (a) The resistance of the RTD measured at different temperatures. (b) Normalized resistance (R/R0, R0 is the resistance at 300K) plotted as a function of T. (c) The resistance change of the RTDs under various heating currents. (d) Temperature change of the metal RTDs plotted as a function of heating power. 300 305 310 315 320 325 330 44.0 44.5 45.0 45.5 46.0 46.5 Temperature (K) R (Ohm) Right RTD 300 305 310 315 320 325 330 1.00 1.01 1.02 1.03 1.04 1.05 Right RTD Fitting R/R 0 Temperature (K) 0 2 4 6 8 44.15 44.20 44.25 44.30 44.35 44.40 44.45 Heater Current (mA) R (Ohm) Right RTD 0 5 10 15 20 25 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Right RTD Fitting T (K) Heater Power (mW) (a) (c) (d) (b) 59 Figure 4.3 shows the thermoelectric voltage plotted as a function of the temperature gradient measured from the Bi2Te3 nanoplate shown in Figure 4.1. Before Al2O3 deposition, the Seebeck coefficient of this flake was -34.3µV/K. Here, the two datasets plotted in the Figure correspond to positive and negative voltages applied to the heater. After ALD deposition, the Seebeck coefficient increases to - 168.9µV/K, which is a factor of 5X increase. We also observe a factor of 13X increase in the 4-probe resistance of the nanoplate, which is consistent with a decrease in the unintentional doping of this material. Together, this corresponds to a 2-fold increase in the power factor (i.e., S 2 ) of the Bi2Te3 nanoplates with a value of 0.434 mW/m·K 2 . The value is approximately one order of magnitude below bulk values for Bi2Te3 alloys. 99 60 Figure 4. 3. In-plane Seebeck coefficient of the Bi2Te3 flake (a) before and (b) after Al2O3 deposition. 0.0 0.2 0.4 0.6 0.8 1.0 -40 -35 -30 -25 -20 -15 -10 -5 0 T (K) V (uV) DC pos DC neg Fitting S= - 34.3 µV/K (a) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 -250 -200 -150 -100 -50 0 T (K) V (uV) DC pos DC neg Fitting S= - 168.9µV/K (b) 61 Figure 4.4 shows the results of Hall effect measurements plotted as a function of temperature of the Bi2Te3 nanoplate before and after Al2O3 deposition. An optical microscope image of the device is shown in Figure 4.4a. Hall carrier densities plotted as a function of temperature before and after ALD deposition are shown in Figure 4.4b. In comparison, after the ALD-based surface passivation, the carrier concentration is approximately 40% lower than before, which indicates that the difference of the carrier concentrations arises from the unintentional n-type surface doping caused by the exposure to ambient conditions. Figure 4. 4. Hall measurements of the Bi2Te3 flake. (a) Optical microscope image of the device and (b) Hall carrier densities of the Bi2Te3 flake before and after Al2O3 deposition. Figure S4.6 of the Supplement Document shows the results of similar thermoelectric measurements performed on a different Bi2Te3 nanoplate sample. Here, we observe a similar change in the optical contrast of the material after the ALD process, as evident in Figure S4.2. Here, the Seebeck coefficient of the as-grown Bi2Te3 nanoplate is -30.1µV/K, which increases by 4.8-fold to -145.7µV/K after Al2O3 (a) (b) 0 50 100 150 200 250 300 350 10 13 10 14 10 15 Before Al 2 O 3 After Al 2 O 3 Carrier Density (cm -2 ) Temperature (K) 20μm 62 deposition. These results confirm that the effects observed in Figure 4.3 are roust and repeatable. Again, we observe a 11-fold increase in the 4-probe resistance of the samples after the ALD process, corresponding to a net increase in the power factor of 2-fold. In our previous work, we have used a two-channel model to analyze the carrier concentration dependence of the Seebeck coefficient and electrical conductivity. Based on this model, we estimate a change in the Fermi energy of approximately 50-100meV due to the atomic layer deposition process. As a prototypical topological insulator, Bi2Te3 has the same unconventional band structures as other TIs that have an insulating bulk band gap and metallic surface state, which consists of a single Dirac cone. Thus, the topological surface state (TSS) plays an important role in transport measurements of Bi2Te3 nanoplates and the modification of the TSS have been reported during the past decade. Chen et al. have shown a strong n-type doping of the surface state of Bi2Te3 after an exposure to air at room temperature. 100 The formation of two-dimensional quantum well states have also been observed near the exposed surface of the Bi2Te3. In the work of Benia et al., topological insulator Bi2Se3 was exposed to water vapor and results in a band bending that shifts the Dirac point deep into the occupied states and quantum well states were created as well. 101 The work of Pettes et al., reported that pronounced n-type surface band bending can yield a suppressed and even negative Seebeck coefficient in unintentionally p-type doped Bi2Te3 nanoplates. 102 Ye et al. have shown that the proper selection of ALD precursors play a critical role in device performance of Bi2Te3 based TI field-effect transistors. As precursors, H2O creates much weaker surface damage 63 than ozone during the first several cycles of ALD growth since ozone is a stronger oxidant than H2O. 103 To minimize the surface damage from precursor, we use TMA/H2O as precursors for our ALD process. In the work presented here, before the surface passivation, our Bi2Te3 nanoplates exposed to ambient air have a native oxide and gas adsorption on the surface, which leads to n-type surface band bending, which is consistent with the negative Seebeck coefficient (-34.3µV/K). We believe that after the removal of surface doping, ALD-based surface passivation provides an effective way to preserve the TSS from environmental degradation, presenting a possible mechanism for “TSS protection”, thus inhibiting the effects of unintentional doping in the nanomaterial. This accounts for both the marked increase Seebeck coefficient and increase in sample resistance. 4.5 Conclusion In summary, we report in-plane thermoelectric measurements of Bi2Te3 nanoplates grown by chemical vapor deposition (CVD). The as-grown flakes exhibit relatively small thermopowers around -34µV/K, due to unintentional surface doping, likely due to gas adsorption and surface oxidation. After removal of the surface doping in a buffered HCl solution, and 1nm Al passivation followed by a deposition of 30nm of Al2O3 using atomic layer deposition (ALD), the Seebeck coefficient of these flakes increases by a factor of 5X to -169µV/K. Here, we believe that the ALD-based surface passivation can effectively reveal the topological surface states of Bi2Te3 nanoplates by 64 inhibiting n-type surface band bending, which is caused by environmental exposure, thus preventing unintentional doping in the Bi2Te3 and thereby increasing the Seebeck coefficient. The high surface-to-volume ratio of these thin (~10nm thick) nanoplates make them especially sensitive to surface doping and nanomaterials in general. An increase in the sample resistance is also observed after the ALD process, which is consistent with the decrease in doping. See supplementary material for the details of the Raman spectroscopy of Bi2Te3 sample before and after ALD passivation, Seebeck measurement for the second Bi2Te3 sample and the Transmission Electron Microscope (TEM) Energy Dispersive Spectroscopy (EDS) analysis. Acknowledgements: This research was supported by the Department of Energy (DOE) Award Nos. DE-FG02-07ER46376 (J.C.), DE-FG02-07ER46377 (J.K. and L.S.), and NSF Award No. 1402906. (N.P.). 65 Chapter 5 Enhanced Cross-plane Thermoelectric Transport of Rotationally-disordered SnSe 2 via Se Vapor Annealing This chapter is similar to Chen et al 104 , published in Nano Letters 5.1 Abstract We report cross-plane thermoelectric measurements of SnSe and SnSe2 films grown by the modulated element reactant (MER) approach. These materials exhibit ultra-low cross-plane thermal conductivities, which are advantageous for thermoelectric energy conversion. The initially grown SnSe films have relatively low cross-plane Seebeck coefficients (-38.6 V/K) due to significant unintentional doping originating from Se vacancies when annealed in nitrogen, as a result of the relatively high vapor pressure of Se. By performing post-growth annealing at a fixed Se partial pressure (300°C for 30 minutes using SnSe2 as the Se source in a sealed tube), a transition from SnSe-to-SnSe2 is induced, which is evidenced by clear changes in the X-ray diffraction patterns of the films. This results in a 16-fold increase in the cross- plane Seebeck coefficient (from -38.6 to -631V/K) after Se annealing due to both the SnSe to SnSe2 transition and the mitigation of unintentional doping by Se vacancies. We also observe a corresponding 6-fold drop in the electrical conductivity (from 3S/m to 0.5S/m) after Se annealing, which is consistent with both a drop in the carrier concentration and an increase in band gap. The power factor S 2  increased by 44X (from 4.5 nW/m·K 2 to 0.2 W/m·K 2 ) after Se annealing. We believe that these results demonstrate a robust method for mitigating unintentional doping in a promising class of materials for thermoelectric applications. 66 5.2 Introduction Over the past two decades, significant improvements in thermoelectric efficiencies have been achieved through a reduction of lattice thermal conductivity while maintaining good electrical conductivity. 105 Highly anisotropic materials with weak van der Waals bonding across incoherent interfaces gives rise to exceptionally low cross-plane thermal conductance. Cahill and Johnson have shown that WSe2 “disordered layered crystals" (i.e., solids that combine order and disorder in the random stacking of two-dimensional crystalline sheets) have a thermal conductivity that is only a factor of 2 larger than air. 106 Subsequent investigations have shown that ultralow thermal conductivity is a general feature of disordered layered crystals, (MSe)m(TSe2)n, (T'Se2)m(TSe2)n, (where M = Sn, Pb, Bi, La,…; T' and T = Ti, V , Cr, Nb, Mo, Ta, W and Sn) and related materials with rotational disorder between the layers. 107-110 The cause of this low thermal conductivity is explained by the large anisotropy in elastic constants that suppresses the density of phonon modes propagating along the soft direction. Over the past several years, the Johnson group has developed a controlled synthesis route that enables the preparation of interdigitated layers of two or more constituents that do not have an epitaxial relationship between their structures. 111 Within the planes, the constituent layers are crystalline. From plane to plane, the layers are randomly misregistered in x and y and rotated relative to each other. To date, the constituent 2D layers include transition metal dichalcogenides, rock salt structured layers (e.g., SnSe, PbSe, BiSe, LaSe), Bi2Se3 and related compounds, more exotic layered structures including Vn+1Se2n+2, and alloys of these constituents. 112-117 These 67 materials are intermediate between crystalline and amorphous and have been called ferecrystals (from Latin fere, meaning almost). They are closely related to misfit layered structures, which contain two constituent layers with an epitaxial relationship along one of the in-plane axes and no systematic rotational order. 118 These disordered layered crystals have been found to have extremely low cross-plane thermal conductivity, with total thermal conductivity values less than 0.10 Wm -1 K -1 for a large number of different constituents (MoSe2, WSe2, PbSe, SnSe, and SnSe2) in a variety of different configurations. 110 While cross-plane thermal conductivities have been measured, the cross-plane electrical and thermoelectric properties have not been reported for any of these compounds. 5.3 Experimental details In the work presented here, cross-plane thermoelectric devices based on SnSe and SnSe2 are fabricated on a Si wafer with a 300nm insulating oxide. One set of samples was annealed in an open system (i.e., flowing N2 gas environment), resulting in the formation of SnSe with some SnSe2 present (referred to as “SnSe” in the following description). This first sample likely has some Se vacancies due to the annealing in an open N2 atmosphere. A second set of samples was annealed in a fixed partial pressure of Se vapor after the initial annealing in an open system, resulting in the formation of SnSe2 (referred to as “SnSe2” henceforth). To fabricate the cross-plane thermoelectric device, a bottom metal 4-probe 68 resistance temperature detector (RTD) is patterned using electron beam lithography followed by metal deposition of Ti/Au, as illustrated in Figure 5.1. Following this, a layer of PMMA is spin-coated on the substrate, and an 12μm×12μm window is opened on the bottom RTD. The SnSe material is then deposited on the substrate by physical vapor deposition followed by a lift-off process and annealing in an inert N2 environment at 350C for 30 min to form the crystalline layered structure, as illustrated in Figure 5.2. After annealing in a N2 environment, SnSe2 is obtained by annealing in Se vapor at a fixed Se partial pressure, which transitions the SnSe into SnSe2 (see Methods section). The top metal RTD is patterned in the same fashion as the bottom metal RTD. The samples are then capped with a 50nm insulating film of Al2O3 deposited by atomic layer deposition (ALD) at 200°C using trimethylaluminum (TMAl) and water as precursors. Lastly, a serpentine metal heater with 5 nm Ti and 35 nm Pd is patterned on top of the Al2O3 layer. The heater also contains four probes in order to measure the heating power precisely. The device fabrication process and an optical microscope image of a completed device are shown in Figure 5.1. 69 Figure 5. 1. (a-d) Schematic diagrams of the device fabrication process and (e) optical image of the completed device. Figure 5. 2. Schematic illustration of a precursor designed to form a layered structure of SnSe when annealed in an inert atmosphere under optimized conditions. Upon heating in Se-vapor for 30 minutes at 300 ° C, the layered SnSe converts to SnSe2. Se and Sn atoms are represented in orange and blue, respectively. SnSe deposition and lift-off ALD deposition and heater fabrication Top RTD fabrication (a) (d) (b) (c) (e) 20μm 70 5.4 Results and discussions Once the thermoelectric device is fabricated, the top and bottom RTDs are calibrated in a vacuum, temperature-controlled stage, as shown in Figure 5.3. First, the resistance of the RTD is measured from 300 to 330 K in increments of 5 K. To anneal out any strain induced in the samples due to thermal expansion, several thermal cycles are performed until the resistance becomes stabilized. The resistance is normalized with respect to the room temperature resistance (i.e., R/R0, where R0 is the resistance at 300 K) and fitted to a linear function of the temperature (see Figure 5.3b). The RTDs’ resistance measurement is performed as a function of the heater current (see Figure 5.3c). Based on this relation and the data in Figure 5.3b, we establish the relation between the temperature change and applied heater power, as shown in Figure 5.3d. A similar temperature calibration is carried out for the top RTD, as shown in Figure S5.1 – S5.3 of the Supplemental Document. 71 Figure 5. 3. Temperature calibration of the bottom RTD of a 50nm SnSe2 film after Se- vapor annealing. (a) The resistance of the RTD measured at different temperatures. (b) Normalized resistance (R/R0, where R0 is the resistance at 300K) plotted as a function of temperature. (c) The resistance changes of the RTD under various heating currents. (d) Temperature change of the metal RTD plotted as a function of heating power. The thermoelectric voltage (V) is then measured as a function of the temperature difference between the top and bottom RTDs (T), and the Seebeck coefficient is obtained from the slope of this data. Figure 5.4 shows the thermoelectric voltage plotted as a function of the temperature difference across the as-grown SnSe device (without Se-vapor annealing). The thermoelectric voltage measurement was performed with both positive and negative heating voltages, which provide nearly the same result, indicating that the voltage drop across the device is in fact a thermoelectric effect rather than a potential difference induced by the heater voltage. Also, the leakage 300 305 310 315 320 325 330 252 254 256 258 260 262 264 Bottom RTD Temperature (K) R (Ohm) 300 305 310 315 320 325 330 1.00 1.01 1.02 1.03 1.04 Bottom RTD Fitting Temperature (K) R/R 0 0.0 0.5 1.0 1.5 2.0 249 250 251 252 253 Bottom RTD Heater Current (mA) R (Ohm) 0 1 2 3 4 5 6 7 0 2 4 6 8 10 12 Bottom RTD Fitting Heater Power (mW) T (K) (a) (b) (c) (d) 72 currents between the Pd heater and the top RTD is less than 50 pA for applied bias voltages up to 5 V (>GΩ), as shown in Figure S5.4 of the Supplemental Document. All data sets were fit to linear functions with the slope corresponding to the Seebeck coefficient (S), as indicated in the Figure. Here, the cross-plane Seebeck coefficient of the SnSe film without Se-vapor annealing is -38.6 μV/K and the electrical conductivity is 3S/m. Figure 5. 4. Cross-plane Seebeck coefficient of the 50 nm SnSe film before Se-vapor annealing. Figure 5.5 shows the cross-plane Seebeck coefficient of the SnSe2 film after Se- vapor annealing at a fixed Se partial pressure at 300°C for 30 minutes. The cross-plane Seebeck coefficient (Figure 5.5) of the SnSe2 device after Se-vapor annealing is -630.8 μV/K, and the electrical conductivity is 0.5S/m. This corresponds to a factor of 16X 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 0 10 20 30 40 50 60 70 DC pos DC neg Fitting V(V) T(K) S= - 38.6 μV/K 73 improvement in the Seebeck coefficient, reflecting the transition from SnSe to SnSe2 and the mitigation of unintentional doping caused by Se vacancies. Figure 5. 5. Cross-plane Seebeck coefficient of the 50 nm SnSe2 after Se-vapor annealing. X-ray fluorescence (XRF) measurements were collected to monitor the amount of each elemental species (including oxygen) present in each sample under the two annealing conditions (see Table 5.1). Here, the Sn counts are roughly the same for both annealing conditions, as expected. The oxygen counts are negligible, indicating that the films are not degraded upon annealing, and annealing under different conditions does not cause excessive oxidation. After annealing in Se vapor, we observe an approximate doubling in the total Se counts measured. This data further suggests that Se vapor annealing results in a conversion from SnSe to SnSe2. 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 0 500 1000 1500 2000 2500 DC pos DC neg Fitting V(V) T(K) S=- 630.8 μV/K 74 Table 5. 1. XRF integrated counts and counts per layer for both SnSe targeted films annealed under different conditions. Specular XRD patterns for the nitrogen-annealed and Se vapor-annealed samples are shown in Figure 5.6. In the N2 annealed sample, only (00l) reflections are observed, which correspond to the c axis lattice constant of the SnSe structure. This results from an alignment of the material with the substrate in the c axis direction while remaining randomly oriented in the ab-plane. 119 After Se annealing, the reflections shift to lower 2 angles, and can be indexed to the SnSe2 structure. This change is consistent with the doubling of Se counts measured with XRF (see Table 5.1). Reflections that do not correspond to the (00l) were also observed for the Se-annealed film, which is indicative of a lower degree of alignment with the substrate. Sn counts Se counts O counts Sn counts/layer Se counts/layer N 2 anneal 1.975 2.912 0.007 0.23 0.035 N 2 & Se anneal 1.946 5.908 0.006 0.23 0.071 75 Figure 5. 6. Specular X-ray diffraction (XRD) patterns of SnSe films. Both samples were annealed in a nitrogen atmosphere and one was subsequently annealed in a selenium atmosphere (the blue pattern) resulting in a conversation to SnSe2. Miller indices are provided for select reflections. The intensity was plotted on a log scale to enhance weak reflections. In-plane XRD measurements were collected for films annealed under both conditions, as shown in Figure 5.7, in order to study the structure of the compound in the ab-plane. Here, the reflections for both patterns are indexed to either SnSe or SnSe2, with the underlying substrate peak identified with an asterisk. For the SnSe film annealed in a N2 atmosphere, all the reflections are indexed to (hk0) reflections of either SnSe or SnSe2. Here, reflections for both constituents have nearly equal intensities, indicating that there are nearly equal amounts of both crystallinities present. The lack of (00l) reflections for the SnSe2 film is likely the result of the preferred alignment of the SnSe constituent obscuring the SnSe2 constituent from specular diffraction. 119 The presence of SnSe2 in the N2-annealed sample indicates that there is extra Se present and that longer annealing times are required to form only the monoselenide. The film 76 annealed in Se vapor shows only reflections corresponding to SnSe2, but also contains non-(hk0) reflections. The presence of these extra reflections along with the 00l axis supports the fact that the sample loses a considerable degree of alignment upon converting from the monoselenide to the diselenide. Figure 5. 7. In-plane X-ray diffraction (XRD) patterns for targeted SnSe samples. Both samples were annealed in a nitrogen atmosphere and one was subsequently annealed in a selenium atmosphere (the blue pattern) resulting in a conversation to SnSe2. The reflections are indexed to SnSe and SnSe2 with one substrate peak (marked with the asterisk) in the Se-vapor annealed sample. One potential concern in the measurement of these extremely thin films (~50nm), which are on the same order as the thickness of the metal RTDs (30nm), is that the temperature and voltage drops at the contacts would significantly affect the measurement, resulting in substantially underestimated values of the Seebeck coefficient. In order to verify the validity of this measurement technique, we measured 77 samples with different thicknesses. Figure S5.5 of the Supplemental Document shows the Seebeck measurements of 50nm-thick and 100nm-thick SnSe films grown in the Johnson lab. Other than the thickness, these films were prepared under identical conditions. Both samples show nearly the same Seebeck coefficient (differing by <3%), which verifies the validity of the measurement and indicates that the effect of the contacts is negligible for this material system. That is, the voltage and temperature drop across the contacts seem to have a negligible effect on the measurement. This is an important result, which indicates that the relatively low Seebeck coefficients observed in Figures 5.4 and S5.5 are not simply due to the measurement technique and instead reflect the true nature of the material composition. Another important consideration in this general measurement approach is the relative resistance of the RTDs and the heterostructure structure itself. If the resistance of the heterostructure is smaller than that of the RTDs, there will be electrical shorting of the RTD through the thin film of interest material, rendering the RTDs ineffective. Typically, the RTD resistance is approximately 4 and the cross-plane resistance of these samples is around 120, which is well within the reliable range of operation. We estimate that reliable results can be obtained below a ratio RRTD/Rfilm of approximately 10%. 5.5 Conclusion In conclusion, we report an enhancement in the cross-plane thermoelectric properties of SnSe films due to Se vapor annealing, which induces a SnSe-to-SnSe2 78 (i.e., monoselenide-to-diselenide) transition and mitigates the effects of unintentional doping. This results in an extremely high Seebeck coefficient (-631V/K), and increased power factor (0.2 W/m·K 2 ). Our XRF measurements show a doubling in the total Se counts, which is consistent with a transition from SnSe to SnSe2, stoichiometrically. This SnSe-to-SnSe2 transition is corroborated by specular and in- plane XRD measurements. After the Se vapor annealing, the diffraction peaks can be indexed to the SnSe2 structure. By conducting post-growth Se annealing at a fixed Se partial pressure, the compound changes to SnSe2 and alleviates unintentional doping due to Se-vacancies that resulted in the relatively low Seebeck coefficients observed in our previous work on disordered layered SnSe-based materials. As a result, we observe a 16-fold increase in the cross-plane Seebeck coefficient (from -38.6 to -631V/K), and a 44-fold increase in power factor (from 4.5 nW/m·K 2 to 0.2 W/m·K 2 ). A corresponding 6-fold drop in the electrical conductivity is observed, which is consistent with the drop in the free carrier concentration. 79 Methods: Preparation of Sn-Se containing thin films. The targeted SnSe films were crystalized by heating designed precursors prepared by high vacuum physical vapor deposition. The films were prepared using a modified method of that described by Fister et al. 120 Sn was deposited with an electron beam gun and Se was deposited with a Knudson effusion cell. Elemental layers were deposited sequentially to obtain a precursor with compositional modulation that mimics the desired final product. A quartz crystal microbalance was used to monitor the deposition rates and pneumatic shutters positioned above the sources controlled the flux of material to the substrate. The deposition parameters were calibrated using X-ray fluorescence (XRF), X-ray reflectivity (XRR), and X-ray diffraction (XRD) to ensure that the amount of material deposited in each layer was correct for crystallization of the desired product. The films were deposited on Si with a native SiO2 layer for structure and composition characterization, as well as patterned substrates for cross-plane transport measurements as well as fused silica for in-plane transport measurements. After deposition, all films were annealed on a hotplate at 350 ºC for 30 minutes in a nitrogen environment to facilitate crystallization of the desired materials. Some films were subsequently annealed in a sealed quartz tube with a Se vapor pressure provided by powdered SnSe2. 121 The tubes were heated in a single zone Carbolite tube furnace at 300 ºC for 3 hours. 80 X-ray diffraction characterization of Sn-Se containing films. X-Ray fluorescence spectra were collected using a Rigaku ZSX Primus II with a rhodium source. Counts were determined by integrating the area under the intensity line at the º range where a fluorescence peak is expected for each element in question. The integrated area is proportional to the atom/area for the element in question. X-ray reflectivity and specular X-ray diffraction patterns were collected using a Bruker D-8 Discover diffractometer in a locked-coupled -2 geometry. In-plane X-ray diffraction spectra were collected with a Rigaku Smartlab diffractometer using an in-plane gracing incidence geometry. All diffraction experiments were conducted using Cu K radiation. The Supporting Information is available free of charge on the ACS Publications website at DOI: Temperature calibration of metal RTDs of all the devices, cross-plane Seebeck coefficient and electrical resistance of 50nm-thick and 100nm-thick SnSe film without Se annealing, Seebeck coefficient as a function of Fermi energy for SnSe and SnSe2, and calculated range of errors that correspond to the expected range of interface conductance. Acknowledgements: This research was supported by the Department of Energy (DOE) Award Nos. DE-FG02-07ER46376 (J.C.), DE-FG02-07ER46377 (L.Shen.), and NSF 81 Award No. 1402906. (N.P.). Authors D.M.H. and D.C.J. acknowledge support from the National Science Foundation under Grant DMR1710214. This material is based upon work supported by the National Science Foundation Graduate Research Fellowship Program under Grant No. 1309047. 82 Chapter 6 Plasmon-Resonant Enhancement of Photocatalysis on Monolayer WSe 2 This chapter is similar to Chen et al 122 , published in ACS Photonics 6.1 Abstract We report plasmonic enhancement of photocatalysis by depositing 5 nm Au nanoislands onto tungsten diselenide (WSe2) monolayer films. Under 532 nm wavelength illumination, the bare WSe2 film produces a relatively small photocurrent (20 nA). With the addition of Au nanoparticles, we observe enhancements of up to 7 × (0.14 µA) in the measured photocurrent. Despite these relatively small photocurrents, it is remarkable that adequate charge separating fields are generated over just 7.3 Å of material. Here, the improvement in the photocatalytic performance is caused by the local electric field enhancement produced in the monolayer WSe2 by the plasmonic Au nanoislands, as verified by electromagnetic simulations using the finite different time domain (FDTD) method. The near-field optical enhancement increases the electron- hole pair generation rate at the surface of WSe2, thus increasing the amount of photogenerated charge contributing to photoelectrochemical reactions. Despite reducing the effective surface area of WSe2 in contact with the electrolytic solution by 70%, the plasmonic nanoislands couple the incident light very effectively from the far field to the near field in the plane of the monolayer WSe2, thereby improving the overall photoconversion efficiency from 3.5% to 24.7%. 83 6.2 Introduction The production of hydrogen and other fuels from photocatalysis has attracted considerable attention as a potential means of generating and storing energy from sunlight. This approach can provide renewable energy in a clean and environmentally friendly manner without utilizing fossil fuels or emitting carbon dioxide. 22, 123-126 Over one hundred semiconductor materials, including ZnO, GaN, WO3, and BiVO4, have already been identified as suitable candidates for solar energy conversion after the first demonstration of photocatalytic splitting of water under ultraviolet radiation on TiO2 electrodes by Fujishima and Honda in 1972. 25, 127-132 However, many challenges remain in the improvement of quantum efficiencies, promotion of charge transfer at interfaces, increasing activation of catalysts, and suppression of backward reactions involving shuttle redox mediators. 133-136 The transition metal dichalcogenide (TMDC) family of two-dimensional (2D) materials have high surface-to-volume ratios and exhibit a wide range of interesting electronic and optical properties that stand in contrast to most known bulk materials. These properties include large exciton binding energies, which are stable at room temperature and directly affect a material’s ability to function as a photocatalyst. Their high surface-to-volume ratios provide large available surface area for photocatalytic activities and reduces the distance from photogenerated carriers to the solid/water interface. This large surface-to-volume ratio, on the other hand, can increase non- radiative recombination substantially, which is detrimental to photocatalytic performance, presenting a trade-off in the overall photocatalytic performance that is yet 84 to be fully explored. The TMDC materials also offer band gaps spanning a wide range from the near infrared through the visible and ultraviolet wavelength ranges to make full use of the entire solar spectrum. 136-137 Utilization of 2D materials as photocatalysts has been shown to exceed the photocatalytic performance compared to their bulk material counterparts. Sun et al. reported an all-surface-atomic SnS sheet-based photoelectrode exhibiting a 67.1% incident photon-to-current conversion efficiency at 490 nm, which is strikingly higher than that of the bulk counterpart which is only 1.66%. Under visible wavelength illumination, the photocurrent density of the all-surface-atomic SnS sheet-based photoelectrode could reach a value up to 5.27 mA ⋅cm −2 , which is two orders of magnitude higher than that of the corresponding bulk material. 138 Looking strictly at the electrocatalytic properties of WS2 (rather than the photocatalytic properties), V oiry et al. has shown that monolayer nanosheets of chemically exfoliated WS2 are efficient catalysts for hydrogen evolution under very low overpotentials. Here, the enhanced electrocatalytic activity of WS2 was attributed to the high concentration of the strained metallic octahedral phase in the as-exfoliated nanosheets. 139 Liang et al. reported 3 nm thick free-floating SnO sheets (~6 layers thick) synthesized via a liquid exfoliation strategy showing an incident photon-to-current conversion efficiency of up to 20.1% at 300 nm, remarkably higher than 10.7% and 4.2% for the 5.4 nm thick SnO (~11 layers thick) sheet-based and bulk SnO-based photoelectrodes. 140 85 6.3 Experimental details In this work, we demonstrate enhanced photocatalysis using high-quality monolayer WSe2 grown by chemical vapor deposition (CVD), as shown in the optical microscope image of Figure 6.1 (a). This monolayer material exhibits strong photoluminescence (PL) with direct band gap emission around 1.63 eV , as shown in Figure 6.1 (b). In addition, we integrate plasmon resonant nanostructures with this photocatalytic semiconductor, which is an approach that has gained considerable attention in the fields of photocatalytic water splitting, 141-142 water purification, 143-144 and CO2 reduction, 145 as well as dye sensitized solar cells. 146-147 Previous studies have shown plasmonic enhancement of bulk material photocatalysts. 148-149 For example, Liu et al. showed that the addition of the plasmonic Au nanoparticles can enhance the performance of photocatalytic water splitting of bulk TiO2 material by 66 × times under visible wavelength illumination. 141 Christopher et al. showed that plasmonic nanostructures of silver can concurrently use low-intensity visible light and thermal energy to drive catalytic oxidation reactions at lower temperatures than their conventional counterparts that use only thermal stimulus. 150 Tian et al. reported photocatalytic oxidation of ethanol and methanol in nanoporous TiO2 films exhibiting enhancement by loading with plasmonic gold and silver nanoparticles. 151 Kowalska et al. showed that gold-modified titania powders have enhanced photocatalytic activity under visible illumination, due to greater light absorption resulting from transverse and longitudinal local surface plasmon resonance of rod-like gold particles. 152 86 In the work presented here, monolayer WSe2 was grown on 90 nm of SiO2/Si via CVD with solid precursors in a similar process to that previously shown for MoSe2 monolayer growth. Approximately 100 mg of solid Se pellets were put into an alumina boat and placed into the first zone of a two-zone, 2-inch diameter quartz furnace. Another alumina boat containing 25 mg of WO3 powder was placed 25 cm downstream from the Se source in the second, higher temperature zone of the furnace. The SiO2/Si growth substrate was treated with hexamethyldisilazane (HMDS) and perylene- 3,4,9,10 tetracarboxylic acid tetrapotassium salt (PTAS) before being placed face-down about 5 mm above the WO3 powder. After purging with 1000 sccm of Ar for 5 minutes, the Se zone was heated to 500 °C and main growth zone to 900 °C. For the growth phase, a gas flow of 25 sccm Ar and 5 sccm H2 were maintained for 30 minutes, after which the furnace was ramped down under inert Ar flow until cooled. Monolayer WSe2 was then transferred from the SiO2 growth substrate onto an indium tin oxide (ITO)- coated glass slide using a drop-casted PMMA support layer and dilute 2% HF to aid in delamination. We then evaporate a gold film with a nominal thickness of 5 nm on half of the surface of the WSe2 monolayer using a shadow mask. This thin gold film is known to form island-like growth that is strongly plasmonic and serves as a good substrate for surface enhanced Raman spectroscopy (SERS) as well as plasmon- enhanced photochemistry. 141 An alternative method of monolayer WSe2 growth have been added to the Supplemental Document. 153-155 87 Figure 6. 1. (a) Optical microscope image and (b) photoluminescence spectrum of transferred CVD-grown monolayer WSe2 (light grey region) on ITO substrate. 1.3 1.4 1.5 1.6 1.7 1.8 0 2000 4000 6000 8000 Intensity (a.u.) Energy (eV) 200 μm 88 6.4 Results and discussions Figure 6.2a shows an illustration of the sample geometry, where an insulated copper wire is attached to the WSe2/ITO using silver paint and the whole sample, excluding the top surface, is encased in epoxy to insulate it from the electrolytic solution. The photoelectrochemical reaction rate obtained with this WSe2 monolayer electrode both with and without the Au nanoislands was measured in a pH=7, 1 M NaI solution using a three-terminal potentiostat (Gamry, Inc.) with the monolayer WSe2 monolayer, an Ag/AgCl electrode, and a Pt wire functioning as the working, reference, and counter electrodes, respectively. Here, the iodide/iodine redox system is chosen because most redox systems oxidize the surface of WSe2 and MoSe2 to an extent that increases gradually with increasing positive redox potential. The I - /I2 redox couple counteracts the surface changes, suppressing the potential drop in the surface, which offer higher solar-to-electrical conversion efficiency and better stability for WSe2 and MoSe2. 156-158 An AC lock-in technique is used to detect the very small photocurrents (nA) generated by the WSe2 films, as illustrated in Figure 6.2b. The incident laser is chopped at 150Hz by a chopper wheel (Stanford Research Systems, Model SR540) connected to the “REF IN” terminal of the lock-in amplifier (Stanford Research Systems, Model SRS830 DSP), as shown in Figure S6.1 of the Supplemental Document. This enables the lock-in amplifier to detect responses only at the specific frequency of the modulated light, providing a very sensitive measure of the photoresponse of these photocatalytic surfaces, while ignoring the DC electrochemical current. The current output of the Gamry potentiostat is connected to the input channel of the lock-in amplifier to measure 89 the AC photoresponse from the sample at the same frequency as the chopped laser. Figure 6. 2. (a) Diagram illustrating the basic sample configuration. (b) Schematic circuit diagram of the three-terminal photoelectrochemical setup with the modulated laser and AC lock-in amplifier. (c) Schematic diagram of the photocatalytic iodine redox process. WSe 2 ITO Incident Light I 2 I - (c) Au Nanoislands Gamry Potentiostat Pt Counter Electrode Monolayer WSe 2 ITO/Glass Slide Substrate Ag/AgCl Reference Electrode 1 M NaI solution Lock-in Amplifier 5nm Au Chopper Incident Laser (a) (b) 90 The DC current, AC current, and AC phase for the bare monolayer WSe2 monolayer electrode (i.e., without Au nanoislands) without and with illumination under visible light (0.21 mW/cm 2 , 532 nm) are shown in Figures 6.3a and 6.3b, respectively. Without illumination (Figure 6.3a), the WSe2 exhibits only noise in the AC current, as expected. Under illumination (Figure 6.3b), however, the bare WSe2 shows approximately an AC photocurrent of 45 nA, which is approximately one thousandth of the DC current, demonstrating that our AC lock-in technique provides an extremely sensitive measure of the photoresponse from the WSe2 photocatalytic surfaces while ignoring the purely DC electrochemical signal. Since we are irradiating monolayer WSe2, the photocurrent is limited by the inherently small optical density of the monolayer material. As a control experiment, Figure 6.3c shows the same AC photocurrent measurements for bare ITO glass slide without any WSe2. Under the same illumination conditions, this bare ITO electrode shows no photocurrent (<1 nA), verifying that the photoresponse we observed from the monolayer WSe2 electrode is not simply originating from the underlying ITO substrate. 91 Figure 6. 3. DC current, AC current, and AC phase measurements for the bare WSe2 monolayer electrode (i.e., without Au nanoislands) (a) without and (b) with illumination under visible light and (c) bare ITO glass slide without WSe2 under the same illumination conditions in (b). Figure 6.4 shows the photocatalytic enhancement due to the addition of the plasmonic Au nanoislands observed under 532nm wavelength illumination (0.21 mW/cm 2 ). Figure 6.4a shows a direct comparison of the AC photocurrent taken for WSe2 both with and without Au nanoislands. For bare WSe2 without Au nanoparticles, a small photocurrent around 20 nA is observed. A significant enhancement in the photocurrent (7 ×) is evident for the WSe2 with plasmonic Au nanoparticles, resulting in a photocurrent of 0.14 µA. The AC phase for WSe2 both with and without Au 92 nanoparticles are nearly identical during the measurement, as shown in Figure 6.4b, verifying that our signal is “locked” to the specific frequency of the modulated light. Figure 6. 4. (a) AC photocurrent and (b) AC phase measurements for the monolayer WSe2 electrode with and without Au nanoislands under visible illumination. 93 Figure 6.5 shows the electromagnetic response of the Au nanoislands/monolayer WSe2 composite film calculated using the finite difference time domain (FDTD) method. Figure 6.5a shows a TEM image of the gold nanoislands deposited on top of monolayer WSe2. The dark grey regions correspond to the gold nanoislands and the light regions represent the space between. The electric field intensity distribution of this film when irradiated on resonance is shown in Figure 6.5b, and is dominated by localized “hot spots” which occur between nearly touching Au nanoislands. 159-162 Here, the “hot spot” regions can be seen clearly between nearly touching Au nanoislands. In these local hot spot regions, the electric field intensity at the monolayer WSe2 surface can be as much as 890× times higher than the incident electric field intensity. This means that the photon absorption rate and electron-hole pair generation rate are three orders of magnitude higher than that of the incident electromagnetic radiation in the brightest spots (i.e., without the Au nanoislands). It is also evident from Figure 6.5b that a vast majority of the photocatalytic surface is not significantly enhanced (i.e., away from the “hot spots”). We can estimate the overall plasmonic enhancement factor from these electromagnetic simulations by evaluating the following integral: ⁡EF = ∫ | E | 2 ⅆx ⅆy ∫ | E 0 | 2 ⅆx ⅆy where the integral is taken over the area of the simulation window in x and y (200 μm × 200 μm). Evaluating this integral, based on the simulation results shown in Figure 6.5, yields an enhancement factor of 3.9 ×, which is consistent with our experimental observations. Figure S6.3 shows an average measured enhancement factor of 5.0 at 532 94 nm and 7.2 at 633 nm, which indicates that the plasmonic nanoislands couple 633 nm light more efficiently in the plane of the WSe2 monolayer than 532 nm. Figure S6.4 in the Supplemental Document shows a calculated enhancement factor of 8.26 at 633 nm. Based on these results, we feel that we have obtained a detailed understanding of the mechanism of enhancement observed in our experimental work. In essence, these FDTD simulations show that the plasmonic nanoislands couple light very efficiently from the far field to near-field in the plane of the monolayer WSe2 monolayer, thus improving the photoconversion efficiency. Figure 6. 5. (a) Transmission electron microscope (TEM) image of 5 nm Au nanoisland film (dark grey regions). (b) Enhancement factor of the electric field intensity at the Au nanoislands/monolayer WSe2 interface calculated using the FDTD method. 95 6.5 Conclusion In conclusion, we demonstrated plasmonic enhancement of monolayer WSe2 photocatalysis in the visible wavelength range by exploiting the surface plasmon resonance of 5 nm gold nanoislands. The addition of Au nanoparticles leads to a 7 × increase in the photocurrent (from 20 nA to 0.14 µA) at a wavelength of 532 nm and improves the incident photon-to-current conversion efficiency from 3.5% to 24.7%. FDTD simulation of this process suggests that the local electric fields produced by the surface plasmons couple light efficiently to the surface of monolayer WSe2, particularly in the local hot spot regions, thereby increasing the electron-hole pair generation rate at the surface of the WSe2 through near-field optical enhancement. This increases the amount of photogenerated charge contributing to these photoelectrochemical reactions and presents a possible route towards improving direct solar fuel production. By integrating the electric field intensity over the entire area of our simulation window, we obtain overall enhancement factor of 3.9×, which agrees well with our experimental findings. Supporting Information: Schematic of AC lock-in technique, photocurrent measurements for controlled experiment samples, wavelength dependent and intensity dependent of photocurrent measurement, and FDTD simulation for 633 nm. 96 Acknowledgements: This research was supported by NSF Award No. 1512505 (J.C.) and 1708581 (H.S.), Army Research Office ARO Award No. W911NF-14-1-0228 (L.S.), Air Force Office of Scientific Research (AFOSR) Grant No. FA9550-15-1-0184 (B.H.) and U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Award DE-SC0019322 (Y . W.). The work at Stanford was supported in part by AFOSR Grant FA9550-14-1-0251 and by the NSF EFRI 2-DARE Grant 1542883. C.S.B. acknowledges support from the NSF Graduate Research Fellowship under Grant No. DGE-114747. 97 Chapter 7 Stacking Independence of WSe 2 /MoSe 2 Heterostructures for Photocatalytic Energy Conversion This chapter is similar to Chen et al, accepted by ACS Applied Nano Materials 7.1 Abstract We report a comparison of the photocatalytic performance of WSe2-on-MoSe2 and MoSe2-on-WSe2 heterostructures. While built-in electric fields exist in these heterostructures on the order of 100 kV/cm due to band offsets between these two materials, the photocatalytic performance (i.e., photocurrent) is independent of the stacking order of the two materials. Solving Poisson’s equation under these conditions, we find that the built-in electric field produced in the heterostructure is at least one order of magnitude smaller than that produced in the electrochemical double layer (i.e., Helmholtz layer). Mott-Schottky measurements indicate that transition metal dichalcogenides (TMDCs) on ITO electrodes have similar capacitance to that of bare ITO, providing further evidence that the interfacial electric fields produced in the solid state heterostructure are negligible compared to the fields generated by the ions in solution. The photocatalytic performance of these heterostructures provided the largest relative enhancement in the heterojunction region under 785 nm irradiation, compared with 532 nm and 633 nm wavelength excitation. These results indicate that there is a significant difference between ground state excitation of excitons at 785 nm (1.58 eV) and free particle excitation at higher photon energies. 98 7.2 Introduction Since the discovery that monolayer graphene and transition metal dichalcogenides (TMDCs) can be easily exfoliated from bulk materials, scientists and engineers have marveled at the exciting possibilities and often surprisingly high performance of atomically thin materials and devices. 163-165 Over the past few years, several groups have demonstrated photocatalytic energy conversion in monolayer, bi- layer, and few-layer TMDCs. 122, 166-170 These initial studies revealed several interesting phenomena and open questions. For example, Todt et al. reported “dark” MoSe2 flakes (>50%), which turned out to be a result of electrically insulating residue deposited during mechanical exfoliation (i.e., scotch tape) process. 167 Later, the authors found that bi-layer material had much higher efficiencies than monolayer or tri-layer material. However, the mechanism underlying this difference is still not clearly understood. More recent studies using photocurrent imaging of bulk WSe2 and MoSe2 crystals by Velazquez et al. 171 and Todt et al. 167 have shown that edge sites can act as both “recombination sites” and “active catalytic sites”. Thus, defects at or near edge sites influence the photocurrent response in a more complex way than simply acting as recombination centers. Charge separation in TMDC materials can be improved by physically stacking materials with different electronic properties. MoS2/WSe2 heterostructures, for example, form a Type II heterojunction, in which the conduction and valence band energy alignments favor electron flow from WSe2 to MoS2. 172-173 In solid state photovoltaic cells, MoS2/WSe2 heterostructures have achieved only ~1% power conversion 99 efficiencies, despite the fact that they absorb ~10% of the incident solar irradiation. 174- 176 In these solid state heterojunctions, the charge carriers are all generated at the heterostructure interface, however, the efficiency is limited by lateral charge transport arising from the adjacent electrical contacts in the device, which require charge carriers to traverse several microns to the electrical contacts. 7.3 Experimental Details Interlayer excitons can also exist at the interface between different TMDC materials, as illustrated in Figure 7.1a for monolayer MoSe2/WSe2 heterostructures. Because of the band offsets in the two materials, it is energetically favorable for the electron to prefer to be in the MoSe2 layer and the hole in the WSe2 layer in the MoSe2/WSe2 heterostructure, forming an interlayer exciton. These excitons are more stable and have longer lifetimes than intralayer excitons. Jauregui et al. recently reported interlayer excitons with lifetimes of 225 ns, which is 10,000 times longer than intralayer excitons. 177 Because the two lattices are misoriented by an angle θ, there is k-space suppression of interlayer tunneling and recombination, which results in the long lifetime of the interlayer excitons. While there have been many studies of the optoelectronic properties of solid state TMDC heterostructures, there have been no reports of their photocatalytic properties. In the work presented here, we explore photocatalysis on TMDC heterostructures by performing photoelectrochemical measurements under various electrochemical potentials and wavelengths of illumination, as well as capacitance- 100 voltage (i.e., Mott-Schottky) measurements, in order to show that the built-in electric fields within the solid-state heterojunction are, in fact, negligible. A simple electrostatic model is used to calculate the electric field strengths of the solid-state heterojunction system and that of the electrochemical double layer generated by the ions in solution. Here, monolayer MoSe2 and WSe2 were grown separately on thermally oxidized silicon chips (i.e., Si/SiO2) via chemical vapor deposition (CVD) using solid precursors in processes previously reported. 122, 178 In short, approximately 100 mg of solid Se pellets were placed into an alumina boat and loaded into the first zone of a two-zone, 2 inch diameter quartz furnace. Another alumina boat containing either 25 mg of WO3 powder (for WSe2) or ~ 0.1 mg MoO3 powder (for MoSe2) was placed 25 cm downstream from the Se source in the second, higher temperature zone of the furnace. The Si/SiO2 growth substrate was first treated with hexamethyldisilazane (HMDS) and perylene-3,4,9,10 tetracarboxylic acid tetrapotassium salt (PTAS) before being placed face-down about 5 mm above the transition metal oxide powder. After purging with Ar flowing at 1000 sccm for 5 minutes, the Se zone was heated to 500 °C and the main growth zone to either 900 °C or 850 °C for WSe2 and MoSe2, respectively. For the growth phase, a gas flow of 25 sccm Ar and 5 sccm H2 was maintained for 30 minutes, after which the furnace temperature was ramped down under inert Ar flow until cooled to room temperaure. The monolayer WSe2 and MoSe2 were then transferred onto an indium tin oxide (ITO)-coated glass slide using a drop-casted PMMA support layer and dilute 2% HF to aid in the delamination. After drying the sample and dissolving the PMMA in acetone, the transfer process was repeated with the other TMDC monolayer, 101 resulting in a stacked heterostructure. The different order of transfers resulted in two unique material stacks, either MoSe2-on-WSe2 or WSe2-on-MoSe2. It should be noted that, although the CVD growth does not result in continuous monolayer growth, nucleation density and crystal sizes are large enough that two transfer processes result in a significant amount of randomly-oriented heterostructures. Cross-sectional diagram showing an MoSe2-on-WSe2 on ITO substrate has been shown in Figure 7.1b and Figure 7.1c depicts the photocatalytic iodine redox process (2I - + 2e + → I2) taking place on the TMDC heterostructure. In the photoelectrochemical measurements, we use a water immersion lens to identify the TMDC heterostructures and an AC lock-in technique to detect the very small photocurrents (nA) generated by the TMDC heterostructures, as illustrated in Figure S7.1a in the Supplemental Document. The incident laser is chopped at 100 Hz by a chopper wheel (Stanford Research Systems, Model SR540) connected to the “REF IN” terminal of the lock-in amplifier (Stanford Research Systems, Model SRS830 DSP), as illustrated in Figure S7.1b. The current output of the Gamry potentiostat is connected to the input channel of the lock-in amplifier to measure the AC photoresponse from the sample at the same frequency as the chopped laser. 122 In this configuration, the lock-in amplifier can detect responses only at the specific frequency of the modulated laser, providing a very sensitive measure of the photoresponse of these photocatalytic interfaces, while being completely insensitive to the DC component of the electrochemical current. 102 Figure 7. 1. (a) Energy band diagram of interlayer exciton in a MoSe2/WSe2 heterostructure. (b) Cross-sectional diagram showing an MoSe2-on-WSe2 on ITO substrate. (c) Schematic diagram of the photocatalytic iodine redox process taking place on the TMDC heterostructure. (a) (b) (c) 1.57eV 1.65eV WSe 2 MoSe 2 ITO 6.5Å 6.5Å I - + + + + + + I 2 E + - I - I - I - I - I - I - WSe 2 ITO Incident Laser MoSe 2 I 2 I - 103 7.4 Results and Discussion Figures 7.2a and 7.2c shows optical microscope images of MoSe2-on-WSe2 and WSe2-on-MoSe2 heterostructures, respectively. The corresponding photocurrents produced with 785 nm wavelength excitation are plotted in Figures 7.2b and 7.2d. Here, the photocurrent produced in the monolayer regions of WSe2 and MoSe2 are roughly equal in magnitude (≈44 nA), and the photocurrent in the heterostructure region (labeled “both”) is approximately twice that observed in the monolayer regions. Figure 7.2e shows the photoluminescence spectrum taken with 532 nm wavelength excitation. The photon energy corresponding to the 785 nm wavelength light used in the photoelectrochemical measurements is indicated on this plot, illustrating that this excitation is resonant with the ground state exciton of the system. The photocurrent profiles are roughly the same for MoSe2-on-WSe2 and the WSe2-on-MoSe2 heterostructures, indicating that the effect of the built-in cross-plane electric fields, which dominate the behavior of solid-state photovoltaic devices, 51, 179-181 is negligible here. We can estimate the built-in electric field produced in the heterostructure by integrating over the electrostatic charge densities in the two TMDC monolayers, in accordance with Poisson’s equation. These are the fields that are responsible for creating photocurrent in solid-state photovoltaic devices. 182-183 Typical defect concentrations for TMDCs are =10 12 cm -2 (per layer, 6.5 Å) 184-185 or 1.54 x 10 19 cm -3 . Assuming that each layer of the heterostructure is fully depleted, we can estimate the built-in electric field in the center of the heterostructure by integrating Poisson’s 104 equation: 𝐸 = ∫ ( 𝑒𝜌 𝜀 ) 𝑑𝑧 6 . 5Å 0 = 82.2 kV/cm where e is the elementary charge on the electron and  is the dielectric constant of WSe2 ( =   ). 186-187 This is the maximum electric field in the junction, and it is at least one order of magnitude smaller than that produced by the electrochemical ions in solution, as discussed below. We can estimate the electrochemically-induced electric field at the electrode/electrolyte interface by considering the electric field of a charged infinite plane using the equation 𝐸 = 𝜎 2 𝜀 where  is the 2D charge density. 𝜎 = 𝑄 𝐴𝑟𝑒 𝑎 = ⁡ 7 . 8 ⁡ × 10 − 4 ⁡ 𝐶 / 𝑐𝑚 2 𝐸 = 𝜎 2 𝜀 = 2 ⁡ 𝑀𝑉 / 𝑐𝑚 105 Figure 7. 2. Optical microscope images of (a) MoSe2-on-WSe2 and (c) WSe2-on-MoSe2 heterostructures deposited on ITO substrates. AC photocurrent measurements for the (b) MoSe2-on-WSe2 and (d) WSe2-on-MoSe2 heterostructures under 785 nm laser excitation. (e) Photoluminescence spectrum of the MoSe2-on-WSe2 heterostructures taken with 532 nm wavelength excitation. The photon energy corresponding to 785 nm wavelength light is indicated in the plot. MoSe 2 WSe 2 Both Light off 785nm 1.58eV 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 0 5000 10000 15000 20000 25000 30000 35000 PL Intensity (a.u.) Energy (eV) MoSe 2 WSe 2 20 μm (a) (b) (d) (e) 0 30 60 90 120 150 180 0 20 40 60 80 100 Photocurrent (nA) Time (s) MoSe 2 WSe 2 Both Light off 0 30 60 90 120 150 180 0 20 40 60 80 100 Photocurrent (nA) Time (s) 20 μm (c) MoSe 2 WSe 2 106 In order to show that the built-in electric fields are, in fact, negligible compared with the electric fields produced by the electrochemical ions in solution, we performed capacitance-voltage measurements of monolayer TMDC on ITO electrodes. Figure 7.3 shows a comparison of the monolayer WSe2 on ITO electrode and bare ITO electrodes. Here, the two datasets are nearly identical proving that the capacitance and, hence, electric fields created in these heterostructures in electrolyte are dominated by the electrochemical ions in solution. In our previous work, Shi et al. measured the local surface fields using Stark shift spectroscopy to be >1 MV/cm, which is much larger that the built-in fields produced in the solid state heterostructure system, as described above. 188 Figure 7.4 shows an optical microscope image and photocurrent measurements of a MoSe2-on-WSe2 heterostructure taken with 532 nm wavelength light. Again, we find that the photocurrents produced in the monolayer regions of the WSe2 and MoSe2 are roughly equal. However, the photocurrent produced in the heterostructure (i.e., “both”) is only slightly larger than that produced in the monolayer regions, despite there being twice the amount of material in the optical pathlength. This is understood on the basis of exciton vs. free-carrier excitation. The 785 nm wavelength illumination resonantly excites the ground state exciton, while the 532 nm wavelength illumination produces free carriers which can decay faster than bound excitons. 107 Figure 7. 3. (a) Capacitance-voltage measurements of monolayer WSe2 on ITO electrodes (black) and bare ITO electrode (red). (b) Charge density and reference potential relationship of monolayer WSe2 on ITO substrate extracted from the corresponding curve in (a). (a) (b) 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0 2 4 6 8 10 12 With WSe 2 Without WSe 2 1/C 2 x10 10 (C -2 ) Reference Potential (V) 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0 2 4 6 8 With WSe 2 Reference Potential (V) Charge Density (x10 -4 C/cm 2 ) 108 Figure 7. 4. (a) Optical microscope image of MoSe2-on-WSe2 heterostructure deposited on an ITO substrate. (b) AC photocurrent measurements taken in various regions of the MoSe2-on-WSe2 heterostructures on ITO substrate under 532 nm wavelength illumination. Figure 7.5 shows an optical microscope image and photocurrent measurements of a WSe2-on-MoSe2 heterostructure taken with 633 nm wavelength light. The photocurrent measurement is similar to the results shown in Figure 4 with 532 nm wavelength laser. The photocurrent produced in the heterostructure (i.e., “both”) is larger than that produced in the monolayer regions, and the photocurrent produced in 0 30 60 90 120 150 180 0 30 60 90 120 150 180 Photocurrent (nA) Time(s) 20 μm MoSe 2 WSe 2 MoSe 2 WSe 2 Both Light off (a) (b) 109 the monolayer region of the MoSe2 and WSe2 are almost the same. Figure 7. 5. (a) Optical microscope image of WSe2-on-MoSe2 heterostructures deposited on an ITO substrate. (b) AC photocurrent measurements taken in various regions of the WSe2-on-MoSe2 heterostructure under 633 nm wavelength illumination. In conclusion, we have shown that the photocatalytic performance of WSe2- on-MoSe2 and MoSe2-on-WSe2 heterostructures are approximately equivalent and do not depend on the order in which these materials are stacked. As such, the photoelectrochemical behavior of these heterostructures are fundamentally different 0 30 60 90 120 150 180 0 30 60 90 120 150 Time(s) Photocurrent (nA) MoSe 2 WSe 2 Both Light off 20 μm (a) (b) MoSe 2 WSe 2 110 from their solid-state counterparts in which the built-in fields play a crucial role in the charge separation process. A simple electrostatic model is used to substantiate this claim that the built-in fields are much smaller than the electrochemically-induced fields. Capacitance-voltage measurements provide further evidence that these fields are negligible. A comparison of photocatalytic performance obtained with 532 nm, 633 nm, and 785 nm wavelength illumination show different responses, reflecting the difference between exciton-driven and free carrier-driven photocatalysis. This result opens up the possibility of widely tunable material systems that can be used for both photoelectrochemical oxidation and reduction, if their structural/chemical stability can be maintained. Acknowledgements: This research was supported by NSF Award No. 1512505 (J.C.) and 1708581 (H.S.), Army Research Office ARO Award No. W911NF-14-1-0228 (Y .W.), Air Force Office of Scientific Research (AFOSR) Grant No. FA9550-15-1-0184 (B.W.) and U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Award DE-SC0019322 (Z.C.). The work at Stanford was supported in part by AFOSR Grant FA9550-14-1-0251 and by the NSF EFRI 2-DARE Grant 1542883. C.S.B. acknowledges support from the NSF Graduate Research Fellowship under Grant No. DGE-114747. 111 Chapter 8 Future Work 8.1 Thermoelectric study of charge density wave transition in TMDC heterostructure Several TMDCs exhibit a charge density wave (CDW) below the Peierl’s transition temperature, Tp. A CDW is a modulation of the conduction electron density accompanied by a modulation of the atomic positions within the crystal structure. Below Tp, the atomic lattice forms a periodic modulation creating a condensed ground state of electrons and a gap in the Fermi surface. 189-190 Consequently, the electrical resistivity increases abruptly below Tp. Under a sufficiently large applied gate electric field, the CDW can be unpinned from impurities and slide relative to the lattice. The sliding CDW can contribute to the electrical conduction, in addition to the contribution of single-particle conductance. The collective conductance caused by this sliding motion can be modulated by the applied gate voltage by as much as two orders of magnitude larger than that of single-particle conductance. 191 The temperature and field effect modulation of the electrical conductance has received interest for applications for novel switching devices. In TiSe2 and VSe2, the resistivity increases due to the CDW transition and the transition temperature has been found to depend on both the applied pressure and layer thickness, 192-195 which tune the interlayer interaction. VSe2 is composed of hexagonally stacked Se-V-Se sheets with interlayer van der Waals bonding, 194 and has a CDW transition temperature near 100 K. 196 Recent work by the Johnson group has shown that 112 the transition temperature can be tuned by changing the SnSe separation thickness (m) or the VSe2 layer thickness (n) in [(SnSe)1.15]m(VSe2)n heterostructures. 197-198 The in- plane resistivity increase below Tp is caused by a reduction in carrier concentration as shown in Figure 8.1 for [(SnSe)1.15]1(VSe2)1. The in-plane resistivity increase below Tp in [(SnSe)1.15]m(VSe2)1 structures is much more pronounced than observed in bulk VSe2. A similar CDW is observed in [(PbSe)1.11]m(VSe2)1. 199 Figure 8. 1. Temperature dependent resistivity and carrier concentration for two different (SnSe)1.15VSe2 samples with CDW transitions at ~100 K measured in the Johnson lab. Previous measurements of the thermal conductivity across the CDW transition in bulk CDW compounds have shown an anomalous peak in the thermal conductivity at the CDW transition. This peak has been explained by an additional contribution from low frequency phasons, 200-201 which are the energy quantum of a fluctuating CDW. The cross-plane thermal conductivity of semiconducting layered heterostructures produced 113 by the Johnson group has been found to be ultralow. 202 The in-plane lattice thermal conductivity (l) of these materials has also been found to be quite low according to the measurements by Li Shi’s group with the use of suspended micro-devices. 202-203 While the electronic contribution to the thermal conductivity (e) is negligible in semiconducting heterostructures, e can be significant in [(SnSe)1.15]m(VSe2)n in the normal metallic state above Tp, and can drop considerably below Tp when a gap forms in the Fermi surface. The e change due to the CDW transition can be even larger than that of l, based on the Wiedemann-Franz law and the measured electrical resistivity. 197- 198 This abrupt change in e due to the CDW transition can provide a novel mechanism to realize a thermal switch. However, the Wiedemann-Franz law does not necessarily hold true for CDW systems, 204 due to the collective conductance associated with the CDW sliding motion. 205-206 The cross plane properties of [(MSe)1 +x]m(VSe2)1 compounds have not been measured, and can provide new insights into the e and phason contribution in CDW systems because the parasitic lattice thermal conductivity is extremely low in these layered heterostructures. 207 For the next step, we will explore the effect of charge density wave (CDW) transitions on the cross-plane and in-plane thermoelectric transport properties of these heterostructure materials. Figure 8.2 shows the in-plane thermoelectric voltage plotted as a function of temperature for a (PbSe)1(VSe2)1 superlattice structure created by the Oregon group. Here, we see a sudden change in the thermovoltage around 120K, which corresponds to the charge density wave transition in the VSe2 material. It should be noted that the temperature gradient was not known in these preliminary measurements 114 and, as such, we report the thermovoltage measured rather than the Sebeeck coefficient. In the proposed work, we will measure the ratio of the Seebeck coefficient/Hall coefficient, which gives a measure of the entropy per carrier. Our hypothesis is that the entropy per carrier changes dramatically at these phase transitions. It will be interesting to see how the cross-plane thermoelectric transport changes as a result of these CDW transitions. We will also measure temperature-dependent Raman spectroscopy to further characterize the CDW transitions, as reported by Goli et al. 192 Figure 8. 2. In-plane thermoelectric voltage for a (PbSe)1(VSe2)1 superlattice structure. 60 80 100 120 140 160 180 200 220 240 -14 -12 -10 -8 -6 Thermovoltage (V) Temperature (K) Sudden Change 115 Bibliography 1. Iijima, S., Helical microtubules of graphitic carbon. Nature 1991, 354 (6348), 56. 2. Dürkop, T.; Getty, S.; Cobas, E.; Fuhrer, M., Extraordinary mobility in semiconducting carbon nanotubes. Nano Letters 2004, 4 (1), 35-39. 3. Berber, S.; Kwon, Y . K.; Tománek, D., Unusually high thermal conductivity of carbon nanotubes. Physical Review Letters 2000, 84 (20), 4613. 4. Yu, M. F.; Files, B. S.; Arepalli, S.; Ruoff, R. S., Tensile loading of ropes of single wall carbon nanotubes and their mechanical properties. Physical Review Letters 2000, 84 (24), 5552. 5. Dresselhaus, M. S.; Dresselhaus, G.; Saito, R.; Jorio, A., Raman spectroscopy of carbon nanotubes. Physics Reports 2005, 409 (2), 47-99. 6. Reich, S.; Maultzsch, J.; Thomsen, C.; Ordejon, P., Tight-binding description of graphene. Physical Review B 2002, 66 (3), 035412. 7. Wallace, P. R., The band theory of graphite. Physical Review 1947, 71 (9), 622. 8. Javey, A.; Kong, J., Carbon Nanotube Electronics. Springer Science & Business Media: 2009. 9. Liu, Z., Electrical, thermal, catalytic and magnetic properties of nano-structured materials and their applications. 2011. 10. Jorio, A.; Pimenta, M.; Souza Filho, A.; Saito, R.; Dresselhaus, G.; Dresselhaus, M., Characterizing carbon nanotube samples with resonance Raman scattering. New Journal of Physics 2003, 5 (1), 139. 11. Hagen, A.; Steiner, M.; Raschke, M. B.; Lienau, C.; Hertel, T.; Qian, H.; Meixner, A. J.; Hartschuh, A., Exponential decay lifetimes of excitons in individual single- walled carbon nanotubes. Physical Review Letters 2005, 95 (19), 197401. 12. Freitag, M.; Steiner, M.; Naumov, A.; Small, J. P.; Bol, A. A.; Perebeinos, V .; Avouris, P., Carbon nanotube photo-and electroluminescence in longitudinal electric fields. ACS Nano 2009, 3 (11), 3744-3748. 13. Steiner, M.; Freitag, M.; Perebeinos, V .; Naumov, A.; Small, J. P.; Bol, A. A.; Avouris, P., Gate-variable light absorption and emission in a semiconducting carbon nanotube. Nano Letters 2009, 9 (10), 3477-3481. 14. Novoselov, K. S.; Geim, A. K.; Morozov, S. V .; Jiang, D.; Zhang, Y .; Dubonos, S. V .; Grigorieva, I. V .; Firsov, A. A., Electric field effect in atomically thin carbon films. Science 2004, 306 (5696), 666-669. 15. Novoselov, K.; Mishchenko, A.; Carvalho, A.; Neto, A. C., 2D materials and van der Waals heterostructures. Science 2016, 353 (6298), aac9439. 16. Manzeli, S.; Ovchinnikov, D.; Pasquier, D.; Yazyev, O. V .; Kis, A., 2D transition metal dichalcogenides. Nature Reviews Materials 2017, 2 (8), 17033. 17. Mak, K. F.; Lee, C.; Hone, J.; Shan, J.; Heinz, T. F., Atomically thin MoS2: a new direct-gap semiconductor. Physical Review Letters 2010, 105 (13), 136805. 18. Splendiani, A.; Sun, L.; Zhang, Y .; Li, T.; Kim, J.; Chim, C. Y .; Galli, G.; Wang, F., Emerging photoluminescence in monolayer MoS2. Nano Letters 2010, 10 (4), 1271-1275. 19. Alam, H.; Ramakrishna, S., A review on the enhancement of figure of merit from 116 bulk to nano-thermoelectric materials. Nano Energy 2013, 2 (2), 190-212. 20. Zhang, X.; Zhao, L. D., Thermoelectric materials: Energy conversion between heat and electricity. Journal of Materiomics 2015, 1 (2), 92-105. 21. Nolas, G. S.; Sharp, J.; Goldsmid, J., Thermoelectrics: basic principles and new materials developments. Springer Science & Business Media: 2013; V ol. 45. 22. Walter, M. G.; Warren, E. L.; McKone, J. R.; Boettcher, S. W.; Mi, Q.; Santori, E. A.; Lewis, N. S., Solar water splitting cells. Chemical Reviews 2010, 110 (11), 6446-6473. 23. Administration, U. S. E. I., U.S. electricity generation by source, amount, and share of total. 2018. 24. Qiu, J. Enhanced Photocatalysis on Titanium Oxide Passivated III-V Semiconductors. University of Southern California, 2015. 25. Fujishima, A.; Honda, K., Electrochemical photolysis of water at a semiconductor electrode. Nature 1972, 238 (5358), 37. 26. Fujishima, A.; Rao, T. N.; Tryk, D. A., Titanium dioxide photocatalysis. Journal of Photochemistry and Photobiology C: Photochemistry reviews 2000, 1 (1), 1- 21. 27. Wikipedia, Electrolysis of water-Pourbaix diagram for water. 2017. 28. Chen, J.; Dhall, R.; Hou, B.; Yang, S.; Wang, B.; Kang, D.; Cronin, S. B., Enhanced photoluminescence in air-suspended carbon nanotubes by oxygen doping. Applied Physics Letters 2016, 109 (15), 153109. 29. Ma, X.; Adamska, L.; Yamaguchi, H.; Yalcin, S. E.; Tretiak, S.; Doorn, S. K.; Htoon, H., Electronic structure and chemical nature of oxygen dopant states in carbon nanotubes. ACS Nano 2014, 8 (10), 10782-10789. 30. Miyauchi, Y .; Iwamura, M.; Mouri, S.; Kawazoe, T.; Ohtsu, M.; Matsuda, K., Brightening of excitons in carbon nanotubes on dimensionality modification. Nature Photonics 2013, 7 (9), 715. 31. Ghosh, S.; Bachilo, S. M.; Simonette, R. A.; Beckingham, K. M.; Weisman, R. B., Oxygen doping modifies near-infrared band gaps in fluorescent single-walled carbon nanotubes. Science 2010, 330 (6011), 1656-1659. 32. Vialla, F.; Chassagneux, Y .; Ferreira, R.; Roquelet, C.; Diederichs, C.; Cassabois, G.; Roussignol, P.; Lauret, J. S.; V oisin, C., Unifying the low-temperature photoluminescence spectra of carbon nanotubes: The role of acoustic phonon confinement. Physical Review Letters 2014, 113 (5), 057402. 33. Hofmann, M. S.; Glückert, J. T.; Noé, J.; Bourjau, C.; Dehmel, R.; Högele, A., Bright, long-lived and coherent excitons in carbon nanotube quantum dots. Nature Nanotechnology 2013, 8 (7), 502. 34. Sarpkaya, I.; Zhang, Z.; Walden-Newman, W.; Wang, X.; Hone, J.; Wong, C. W.; Strauf, S., Prolonged spontaneous emission and dephasing of localized excitons in air-bridged carbon nanotubes. Nature Communications 2013, 4, 2152. 35. Ma, X.; Hartmann, N. F.; Baldwin, J. K.; Doorn, S. K.; Htoon, H., Room- temperature single-photon generation from solitary dopants of carbon nanotubes. Nature Nanotechnology 2015, 10 (8), 671. 36. Iakoubovskii, K.; Minami, N.; Kim, Y .; Miyashita, K.; Kazaoui, S.; Nalini, B., 117 Midgap luminescence centers in single-wall carbon nanotubes created by ultraviolet illumination. Applied Physics Letters 2006, 89 (17), 173108. 37. Cai, L.; Bahr, J. L.; Yao, Y .; Tour, J. M., Ozonation of single-walled carbon nanotubes and their assemblies on rigid self-assembled monolayers. Chemistry of Materials 2002, 14 (10), 4235-4241. 38. Banerjee, S.; Wong, S. S., Rational sidewall functionalization and purification of single-walled carbon nanotubes by solution-phase ozonolysis. The Journal of Physical Chemistry B 2002, 106 (47), 12144-12151. 39. Li, M.; Boggs, M.; Beebe, T. P.; Huang, C., Oxidation of single-walled carbon nanotubes in dilute aqueous solutions by ozone as affected by ultrasound. Carbon 2008, 46 (3), 466-475. 40. Ogrin, D.; Chattopadhyay, J.; Sadana, A. K.; Billups, W. E.; Barron, A. R., Epoxidation and deoxygenation of single-walled carbon nanotubes: Quantification of epoxide defects. Journal of the American Chemical Society 2006, 128 (35), 11322-11323. 41. Sham, M. L.; Kim, J. K., Surface functionalities of multi-wall carbon nanotubes after UV/Ozone and TETA treatments. Carbon 2006, 44 (4), 768-777. 42. Aria, A. I.; Gharib, M., Reversible tuning of the wettability of carbon nanotube arrays: the effect of ultraviolet/ozone and vacuum pyrolysis treatments. Langmuir 2011, 27 (14), 9005-9011. 43. Chen, Z.; Ziegler, K. J.; Shaver, J.; Hauge, R. H.; Smalley, R. E., Cutting of single- walled carbon nanotubes by ozonolysis. The Journal of Physical Chemistry B 2006, 110 (24), 11624-11627. 44. Mawhinney, D. B.; Naumenko, V .; Kuznetsova, A.; Yates, J. T.; Liu, J.; Smalley, R., Infrared spectral evidence for the etching of carbon nanotubes: ozone oxidation at 298 K. Journal of the American Chemical Society 2000, 122 (10), 2383-2384. 45. Rao, A. M.; Eklund, P.; Bandow, S.; Thess, A.; Smalley, R. E., Evidence for charge transfer in doped carbon nanotube bundles from Raman scattering. Nature 1997, 388 (6639), 257. 46. Simmons, J.; Nichols, B.; Baker, S.; Marcus, M. S.; Castellini, O.; Lee, C. S.; Hamers, R.; Eriksson, M., Effect of ozone oxidation on single-walled carbon nanotubes. The Journal of Physical Chemistry B 2006, 110 (14), 7113-7118. 47. Amer, M. R.; Chang, S. W.; Cronin, S. B., Competing Photocurrent Mechanisms in Quasi-Metallic Carbon Nanotube pn Devices. Small 2015, 11 (26), 3119-3123. 48. Chang, S. W.; Theiss, J.; Hazra, J.; Aykol, M.; Kapadia, R.; Cronin, S. B., Photocurrent spectroscopy of exciton and free particle optical transitions in suspended carbon nanotube pn-junctions. Applied Physics Letters 2015, 107 (5), 053107. 49. See supplementary material at for normalized PL spectra of carbon nanotubes before and after 20s UV/ozone exposures and PL images of carbon nanotubes before and after 2 minutes UV/ozone exposures. 50. Sarpkaya, I.; Ahmadi, E. D.; Shepard, G. D.; Mistry, K. S.; Blackburn, J. L.; Strauf, S., Strong acoustic phonon localization in copolymer-wrapped carbon 118 nanotubes. ACS Nano 2015, 9 (6), 6383-6393. 51. Li, Z.; Chen, J.; Dhall, R.; Cronin, S. B., Highly efficient, high speed vertical photodiodes based on few-layer MoS2. 2D Materials 2016, 4 (1), 015004. 52. Wang, Q. H.; Kalantar-Zadeh, K.; Kis, A.; Coleman, J. N.; Strano, M. S., Electronics and optoelectronics of two-dimensional transition metal dichalcogenides. Nature Nanotechnology 2012, 7 (11), 699. 53. Chhowalla, M.; Shin, H. S.; Eda, G.; Li, L. J.; Loh, K. P.; Zhang, H., The chemistry of two-dimensional layered transition metal dichalcogenide nanosheets. Nature Chemistry 2013, 5 (4), 263. 54. Yu, W. J.; Liu, Y .; Zhou, H.; Yin, A.; Li, Z.; Huang, Y .; Duan, X., Highly efficient gate-tunable photocurrent generation in vertical heterostructures of layered materials. Nature Nanotechnology 2013, 8 (12), 952. 55. Gong, Y .; Lin, J.; Wang, X.; Shi, G.; Lei, S.; Lin, Z.; Zou, X.; Ye, G.; Vajtai, R.; Yakobson, B. I., Vertical and in-plane heterostructures from WS2/MoS2 monolayers. Nature Materials 2014, 13 (12), 1135. 56. Lee, C. H.; Lee, G. H.; Van Der Zande, A. M.; Chen, W.; Li, Y .; Han, M.; Cui, X.; Arefe, G.; Nuckolls, C.; Heinz, T. F., Atomically thin p–n junctions with van der Waals heterointerfaces. Nature Nanotechnology 2014, 9 (9), 676. 57. Dhall, R.; Neupane, M. R.; Wickramaratne, D.; Mecklenburg, M.; Li, Z.; Moore, C.; Lake, R. K.; Cronin, S., Direct Bandgap Transition in Many-Layer MoS2 by Plasma-Induced Layer Decoupling. Advanced Materials 2015, 27 (9), 1573. 58. Buscema, M.; Steele, G. A.; van der Zant, H. S.; Castellanos-Gomez, A., The effect of the substrate on the Raman and photoluminescence emission of single- layer MoS2. Nano Research 2014, 7 (4), 561-571. 59. Bhanu, U.; Islam, M. R.; Tetard, L.; Khondaker, S. I., Photoluminescence quenching in gold-MoS2 hybrid nanoflakes. Scientific Reports 2014, 4, 5575. 60. Li, Z.; Ezhilarasu, G.; Chatzakis, I.; Dhall, R.; Chen, C. C.; Cronin, S. B., Indirect band gap emission by hot electron injection in metal/MoS2 and metal/WSe2 heterojunctions. Nano Letters 2015, 15 (6), 3977-3982. 61. Tung, R. T., Chemical bonding and Fermi level pinning at metal-semiconductor interfaces. Physical Review Letters 2000, 84 (26), 6078. 62. Lince, J. R.; Carré, D. J.; Fleischauer, P. D., Schottky-barrier formation on a covalent semiconductor without Fermi-level pinning: The metal-MoS2 (0001) interface. Physical Review B 1987, 36 (3), 1647. 63. Huang, C.; Wu, S.; Sanchez, A. M.; Peters, J. J.; Beanland, R.; Ross, J. S.; Rivera, P.; Yao, W.; Cobden, D. H.; Xu, X., Lateral heterojunctions within monolayer MoSe2-WSe2 semiconductors. Nature Materials 2014, 13 (12), 1096. 64. Yin, Z.; Li, H.; Li, H.; Jiang, L.; Shi, Y .; Sun, Y .; Lu, G.; Zhang, Q.; Chen, X.; Zhang, H., Single-layer MoS2 phototransistors. ACS Nano 2011, 6 (1), 74-80. 65. Lopez-Sanchez, O.; Lembke, D.; Kayci, M.; Radenovic, A.; Kis, A., Ultrasensitive photodetectors based on monolayer MoS2. Nature Nanotechnology 2013, 8 (7), 497. 66. Choi, W.; Cho, M. Y .; Konar, A.; Lee, J. H.; Cha, G. B.; Hong, S. C.; Kim, S.; Kim, J.; Jena, D.; Joo, J., High-detectivity multilayer MoS2 phototransistors with 119 spectral response from ultraviolet to infrared. Advanced Materials 2012, 24 (43), 5832-5836. 67. Tsai, D. S.; Liu, K. K.; Lien, D. H.; Tsai, M. L.; Kang, C. F.; Lin, C. A.; Li, L. J.; He, J. H., Few-layer MoS2 with high broadband photogain and fast optical switching for use in harsh environments. ACS Nano 2013, 7 (5), 3905-3911. 68. Ionescu, A. M.; Riel, H., Tunnel field-effect transistors as energy-efficient electronic switches. Nature 2011, 479 (7373), 329. 69. Li, Z.; Chang, S. W.; Chen, C. C.; Cronin, S. B., Enhanced photocurrent and photoluminescence spectra in MoS2 under ionic liquid gating. Nano Research 2014, 7 (7), 973-980. 70. Castellanos-Gomez, A.; Buscema, M.; Molenaar, R.; Singh, V .; Janssen, L.; Van Der Zant, H. S.; Steele, G. A., Deterministic transfer of two-dimensional materials by all-dry viscoelastic stamping. 2D Materials 2014, 1 (1), 011002. 71. Mueller, T.; Xia, F.; Avouris, P., Graphene photodetectors for high-speed optical communications. Nature Photonics 2010, 4 (5), 297. 72. Fontana, M.; Deppe, T.; Boyd, A. K.; Rinzan, M.; Liu, A. Y .; Paranjape, M.; Barbara, P., Electron-hole transport and photovoltaic effect in gated MoS2 Schottky junctions. Scientific Reports 2013, 3, 1634. 73. Schlaf, R.; Lang, O.; Pettenkofer, C.; Jaegermann, W., Band lineup of layered semiconductor heterointerfaces prepared by van der Waals epitaxy: Charge transfer correction term for the electron affinity rule. Journal of Applied Physics 1999, 85 (5), 2732-2753. 74. Sullivan, B. T., Optical properties of palladium in the visible and near UV spectral regions. Applied Optics 1990, 29 (13), 1964-1970. 75. Heavens, O. S., Optical properties of thin solid films. Courier Corporation: 1991. 76. Peckerar, M.; Lin, H.; Kocher, R. In Open circuit voltage of MIS Schottky diode solar cells, 1975 International Electron Devices Meeting, IEEE: 1975; pp 213- 216. 77. Ponpon, J.; Siffert, P., Open-circuit voltage of MIS silicon solar cells. Journal of Applied Physics 1976, 47 (7), 3248-3251. 78. Zhang, W.; Chuu, C. P.; Huang, J. K.; Chen, C. H.; Tsai, M. L.; Chang, Y . H.; Liang, C. T.; Chen, Y . Z.; Chueh, Y . L.; He, J. H., Ultrahigh-gain photodetectors based on atomically thin graphene-MoS2 heterostructures. Scientific Reports 2014, 4, 3826. 79. Chen, J.; Kim, J.; Poudel, N.; Hou, B.; Shen, L.; Shi, H.; Shi, L.; Cronin, S., Enhanced thermoelectric efficiency in topological insulator Bi2Te3 nanoplates via atomic layer deposition-based surface passivation. Applied Physics Letters 2018, 113 (8), 083904. 80. Hicks, L. D.; Dresselhaus, M. S., Effect of quantum-well structures on the thermoelectric figure of merit. Physical Review B 1993, 47 (19), 12727-12731. 81. Hicks, L. D.; Dresselhaus, M. S., Thermoelectric figure of merit of a one- dimensional conductor. Physical Review B 1993, 47 (24), 16631-16634. 82. Dresselhaus, M. S.; Dresselhaus, G.; Sun, X.; Zhang, Z.; Cronin, S. B.; Koga, T., Low-dimensional thermoelectric materials. Physics of the Solid State 1999, 41 120 (5), 679-682. 83. Cahill, D. G.; Ford, W. K.; Goodson, K. E.; Mahan, G. D.; Majumdar, A.; Maris, H. J.; Merlin, R.; Phillpot, Sr., Nanoscale thermal transport. Journal of Applied Physics 2003, 93 (2), 793-818. 84. Koga, T.; Sun, X.; Cronin, S. B.; Dresselhaus, M. S., Carrier pocket engineering to design superior thermoelectric materials using GaAs/AlAs superlattices. Applied Physics Letters 1998, 73 (20), 2950-2952. 85. Cronin, S. B.; Lin, Y . M.; Rabin, O.; Black, M. R.; Ying, J. Y .; Dresselhaus, M. S.; Gai, P. L.; Minet, J. P.; Issi, J. P., Making electrical contacts to nanowires with a thick oxide coating. Nanotechnology 2002, 13 (5), 653-658. 86. Arab, S.; Yao, M.; Anderson, P.; Povinelli, M.; Zhou, C.; Dapkus, P. D.; Cronin, S. B., Enhanced Fabry-Perot Resonance in GaAs Nanowires through Surface Passivation and Local Field Enhancement. Nano Research 2014, submitted. 87. Arab, S.; Chi, C.; Shi, T.; Wang, Y .; Dapkus, D. P.; Jackson, H. E.; Smith, L. M.; Cronin, S. B., Effects of Surface Passivation on Twin-Free GaAs Nanosheets. ACS Nano 2015, 10.1021/nn505227q. 88. Dhall, R.; Li, Z.; Kosmowska, E.; Cronin, S. B., Charge neutral MoS2 field effect transistors through oxygen plasma treatment. Journal of Applied Physics 2016, 120 (19), 195702. 89. Bhattacharyya, S.; Pandey, T.; Singh, A. K., Effect of strain on electronic and thermoelectric properties of few layers to bulk MoS2. Nanotechnology 2014, 25 (46), 465701. 90. Kumar, S.; Schwingenschlögl, U., Thermoelectric Response of Bulk and Monolayer MoSe2 and WSe2. Chemistry of Materials 2015, 27 (4), 1278-1284. 91. Kayyalha, M.; Maassen, J.; Lundstrom, M.; Shi, L.; Chen, Y . P., Gate-tunable and thickness-dependent electronic and thermoelectric transport in few-layer MoS2. Journal of Applied Physics 2016, 120 (13), 134305. 92. Zhou, J.; Jin, C.; Seol, J. H.; Li, X.; Shi, L., Thermoelectric properties of individual electrodeposited bismuth telluride nanowires. Applied Physics Letters 2005, 87 (13), 133109. 93. Mavrokefalos, A.; Moore, A. L.; Pettes, M. T.; Shi, L.; Wang, W.; Li, X., Thermoelectric and structural characterizations of individual electrodeposited bismuth telluride nanowires. Journal of Applied Physics 2009, 105 (10), 104318. 94. Hamdou, B.; Kimling, J.; Dorn, A.; Pippel, E.; Rostek, R.; Woias, P.; Nielsch, K., Thermoelectric characterization of bismuth telluride nanowires, synthesized via catalytic growth and post-annealing. Advanced Materials 2013, 25 (2), 239-44. 95. Amani, M.; Lien, D.-H.; Kiriya, D.; Xiao, J.; Azcatl, A.; Noh, J.; Madhvapathy, S. R.; Addou, R.; KC, S.; Dubey, M.; Cho, K.; Wallace, R. M.; Lee, S.-C.; He, J.- H.; III, J. W. A.; Zhang, X.; Yablonovitch, E.; Javey, A., Near-unity photoluminescence quantum yield in MoS2. Science 2015, 350, 1065. 96. Pettes, M. T.; Maassen, J.; Jo, I.; Lundstrom, M. S.; Shi, L., Effects of Surface Band Bending and Scattering on Thermoelectric Transport in Suspended Bismuth Telluride Nanoplates. Nano Letters 2013, 13 (11), 5316-5322. 97. Pettes, M. T.; Kim, J.; Wu, W.; Bustillo, K. C.; Shi, L., Thermoelectric transport 121 in surface- and antimony-doped bismuth telluride nanoplates. APL Materials 2016, 4 (10), 104810. 98. Kong, D.; Dang, W.; Cha, J. J.; Li, H.; Meister, S.; Peng, H.; Liu, Z.; Cui, Y ., Few- layer nanoplates of Bi2Se3 and Bi2Te3 with highly tunable chemical potential. Nano Letters 2010, 10 (6), 2245-50. 99. Li, D.; Sun, R.; Qin, X., Improving thermoelectric properties of p-type Bi2Te3- based alloys by spark plasma sintering. Progress in Natural Science: Materials International 2011, 21 (4), 336-340. 100. Chen, C.; He, S.; Weng, H.; Zhang, W.; Zhao, L.; Liu, H.; Jia, X.; Mou, D.; Liu, S.; He, J.; Peng, Y .; Feng, Y .; Xie, Z.; Liu, G.; Dong, X.; Zhang, J.; Wang, X.; Peng, Q.; Wang, Z.; Zhang, S.; Yang, F.; Chen, C.; Xu, Z.; Dai, X.; Fang, Z.; Zhou, X. J., Robustness of topological order and formation of quantum well states in topological insulators exposed to ambient environment. Proc Natl Acad Sci U S A 2012, 109 (10), 3694-8. 101. Benia, H. M.; Lin, C.; Kern, K.; Ast, C. R., Reactive chemical doping of the Bi2Se3 topological insulator. Phys Rev Lett 2011, 107 (17), 177602. 102. Pettes, M. T.; Maassen, J.; Jo, I.; Lundstrom, M. S.; Shi, L., Effects of surface band bending and scattering on thermoelectric transport in suspended bismuth telluride nanoplates. Nano Letters 2013, 13 (11), 5316-22. 103. Liu, H.; Ye, P. D., Atomic-layer-deposited Al2O3 on Bi2Te3 for topological insulator field-effect transistors. Applied Physics Letters 2011, 99 (5), 052108. 104. Chen, J.; Hamann, D. M.; Choi, D. S.; Poudel, N.; Shen, L.; Shi, L.; Johnson, D. C.; Cronin, S. B., Enhanced Cross-plane Thermoelectric Transport of Rotationally-disordered SnSe2 via Se Vapor Annealing. Nano Letters 2018. 105. Shakouri, A., Recent Developments in Semiconductor Thermoelectric Physics and Materials. In Annual Review of Materials Research, Vol 41, Clarke, D. R.; Fratzl, P., Eds. Annual Reviews: Palo Alto, 2011; V ol. 41, pp 399-431. 106. Cahill, D. G., Extremes of heat conduction-Pushing the boundaries of the thermal conductivity of materials. Mrs Bulletin 2012, 37 (9), 855-863. 107. Gunning, N. S.; Feser, J.; Beekman, M.; Cahill, D. G.; Johnson, D. C., Synthesis and Thermal Properties of Solid-State Structural Isomers: Ordered lntergrowths of SnSe and MoSe2. Journal of the American Chemical Society 2015, 137 (27), 8803-8809. 108. Mortensen, C.; Beekman, M.; Johnson, D. C., Probing the effects of alloying, grain size, and turbostratic disorder on thermal conductivity. Science of Advanced Materials 2011, 3 (4), 639-645. 109. Chiritescu, C.; Mortensen, C.; Cahill, D. G.; Johnson, D.; Zschack, P., Lower limit to the lattice thermal conductivity of nanostructured Bi2Te3-based materials. Journal of Applied Physics 2009, 106 (7), 073503. 110. Chiritescu, C.; Cahill, D. G.; Heideman, C.; Lin, Q.; Mortensen, C.; Nguyen, N. T.; Johnson, D.; Rostek, R.; Böttner, H., Low thermal conductivity in nanoscale layered materials synthesized by the method of modulated elemental reactants. Journal of Applied Physics 2008, 104 (3), 033533. 111. Johnson, D. C., Controlled synthesis of new compounds using modulated 122 elemental reactants. Current Opinion in Solid State and Materials Science 1998, 3 (2), 159-167. 112. Wood, S. R.; Merrill, D. R.; Mitchson, G.; Lygo, A. C.; Bauers, S. R.; Hamann, D. M.; Sutherland, D. R.; Ditto, J.; Johnson, D. C., Modulation doping in metastable heterostructures via kinetically controlled substitution. Chemistry of Materials 2016, 29 (2), 773-779. 113. Mitchson, G.; Hadland, E.; Go ̈ hler, F.; Wanke, M.; Esters, M.; Ditto, J.; Bigwood, E.; Ta, K.; Hennig, R. G.; Seyller, T., Structural changes in 2D BiSe bilayers as N increases in (BiSe)1+ δ(NbSe2)N (N= 1–4) heterostructures. ACS Nano 2016, 10 (10), 9489-9499. 114. Bauers, S.; Ditto, J.; Moore, D.; Johnson, D., Structure–property relationships in non-epitaxial chalcogenide heterostructures: the role of interface density on charge exchange. Nanoscale 2016, 8 (30), 14665-14672. 115. Hamann, D. M.; Merrill, D. R.; Bauers, S. R.; Mitchson, G.; Ditto, J.; Rudin, S. P.; Johnson, D. C., Long-Range Order in [(SnSe)1.2][TiSe2] Prepared from Designed Precursors. Inorganic Chemistry 2017, 56 (6), 3499-3505. 116. Alemayehu, M. B.; Falmbigl, M.; Ta, K.; Ditto, J.; Medlin, D. L.; Johnson, D. C., Designed synthesis of van der Waals heterostructures: The power of kinetic control. Angewandte Chemie International Edition 2015, 54 (51), 15468-15472. 117. Gunning, N. S.; Dankwort, T.; Falmbigl, M.; Ross, U.; Mitchson, G.; Hamann, D. M.; Lotnyk, A.; Kienle, L.; Johnson, D. C., Expanding the Concept of van der Waals Heterostructures to Interwoven 3D Structures. Chemistry of Materials 2017, 29 (19), 8292-8298. 118. Wiegers, G., Misfit layer compounds: Structures and physical properties. Progress in Solid State Chemistry 1996, 24 (1-2), 1-139. 119. Hamann, D. M.; Hadland, E. C.; Johnson, D. C., Heterostructures containing dichalcogenides-new materials with predictable nanoarchitectures and novel emergent properties. Semiconductor Science and Technology 2017, 32 (9), 093004. 120. Fister, L.; Li, X. M.; McConnell, J.; Novet, T.; Johnson, D. C., Deposition system for the synthesis of modulated, ultrathin-film composites. Journal of Vacuum Science & Technology A: Vacuum, Surfaces, and Films 1993, 11 (6), 3014-3019. 121. Lin, Q.; Tepfer, S.; Heideman, C.; Mortensen, C.; Nguyen, N.; Zschack, P.; Beekman, M.; Johnson, D. C., Influence of selenium vapor postannealing on the electrical transport properties of PbSe–WSe2 Nanolaminates. Journal of Materials Research 2011, 26 (15), 1866-1871. 122. Chen, J.; Bailey, C. S.; Hong, Y .; Wang, L.; Cai, Z.; Shen, L.; Hou, B.; Wang, Y .; Shi, H.; Sambur, J.; Pop, E.; Cronin, S. B., Plasmon-Resonant Enhancement of Photocatalysis on Monolayer WSe2. ACS Photonics 2019, 6 (3), 787-792. 123. Maeda, K.; Domen, K., New non-oxide photocatalysts designed for overall water splitting under visible light. The Journal of Physical Chemistry C 2007, 111 (22), 7851-7861. 124. Lee, J. S., Photocatalytic water splitting under visible light with particulate semiconductor catalysts. Catalysis Surveys from Asia 2005, 9 (4), 217-227. 123 125. Kudo, A.; Miseki, Y ., Heterogeneous photocatalyst materials for water splitting. Chemical Society Reviews 2009, 38 (1), 253-278. 126. Grätzel, M., Photoelectrochemical cells. Nature 2001, 414 (6861), 338. 127. Sayama, K.; Mukasa, K.; Abe, R.; Abe, Y .; Arakawa, H., Stoichiometric water splitting into H2 and O2 using a mixture of two different photocatalysts and an IO 3− /I − shuttle redox mediator under visible light irradiation. Chemical Communications 2001, (23), 2416-2417. 128. Kato, H.; Hori, M.; Konta, R.; Shimodaira, Y .; Kudo, A., Construction of Z- scheme type heterogeneous photocatalysis systems for water splitting into H2 and O2 under visible light irradiation. Chemistry Letters 2004, 33 (10), 1348-1349. 129. Jaramillo, T. F.; Baeck, S. H.; Kleiman, S. A.; McFarland, E. W., Combinatorial electrochemical synthesis and screening of mesoporous ZnO for photocatalysis. Macromolecular Rapid Communications 2004, 25 (1), 297-301. 130. Zhao, Z. G.; Miyauchi, M., Nanoporous-Walled Tungsten Oxide Nanotubes as Highly Active Visible-Light-Driven Photocatalysts. Angewandte Chemie International Edition 2008, 47 (37), 7051-7055. 131. Abe, R.; Takata, T.; Sugihara, H.; Domen, K., Photocatalytic overall water splitting under visible light by TaON and WO3 with an IO 3− /I − shuttle redox mediator. Chemical Communications 2005, (30), 3829-3831. 132. Higashi, M.; Abe, R.; Teramura, K.; Takata, T.; Ohtani, B.; Domen, K., Two step water splitting into H2 and O2 under visible light by ATaO2N (A= Ca, Sr, Ba) and WO3 with IO 3- /I - shuttle redox mediator. Chemical Physics Letters 2008, 452 (1- 3), 120-123. 133. Fujishima, A.; Zhang, X.; Tryk, D. A., Heterogeneous photocatalysis: from water photolysis to applications in environmental cleanup. International Journal of Hydrogen Energy 2007, 32 (14), 2664-2672. 134. Bak, T.; Nowotny, J.; Rekas, M.; Sorrell, C., Photo-electrochemical hydrogen generation from water using solar energy. Materials-related aspects. International Journal of Hydrogen Energy 2002, 27 (10), 991-1022. 135. Liu, R.; Lin, Y .; Chou, L. Y .; Sheehan, S. W.; He, W.; Zhang, F.; Hou, H. J.; Wang, D., Water Splitting by Tungsten Oxide Prepared by Atomic Layer Deposition and Decorated with an Oxygen-Evolving Catalyst. Angewandte Chemie 2011, 123 (2), 519-522. 136. Ni, M.; Leung, M. K.; Leung, D. Y .; Sumathy, K., A review and recent developments in photocatalytic water-splitting using TiO2 for hydrogen production. Renewable and Sustainable Energy Reviews 2007, 11 (3), 401-425. 137. Singh, A. K.; Mathew, K.; Zhuang, H. L.; Hennig, R. G., Computational Screening of 2D Materials for Photocatalysis. J Phys Chem Lett 2015, 6 (6), 1087- 98. 138. Sun, Y .; Sun, Z.; Gao, S.; Cheng, H.; Liu, Q.; Lei, F.; Wei, S.; Xie, Y ., All-Surface- Atomic-Metal Chalcogenide Sheets for High-Efficiency Visible-Light Photoelectrochemical Water Splitting. Advanced Energy Materials 2014, 4 (1), 1300611. 139. V oiry, D.; Yamaguchi, H.; Li, J.; Silva, R.; Alves, D. C.; Fujita, T.; Chen, M.; 124 Asefa, T.; Shenoy, V . B.; Eda, G.; Chhowalla, M., Enhanced catalytic activity in strained chemically exfoliated WS2 nanosheets for hydrogen evolution. Nature Materials 2013, 12 (9), 850-5. 140. Liang, L.; Sun, Y .; Lei, F.; Gao, S.; Xie, Y ., Free-floating ultrathin tin monoxide sheets for solar-driven photoelectrochemical water splitting. Journal of Materials Chemistry A 2014, 2 (27), 10647. 141. Liu, Z.; Hou, W.; Pavaskar, P.; Aykol, M.; Cronin, S. B., Plasmon resonant enhancement of photocatalytic water splitting under visible illumination. Nano Letters 2011, 11 (3), 1111-6. 142. Qiu, J.; Zeng, G.; Pavaskar, P.; Li, Z.; Cronin, S. B., Plasmon-enhanced water splitting on TiO2-passivated GaP photocatalysts. Physical Chemistry Chemical Physics 2014, 16 (7), 3115-3121. 143. Hou, W.; Liu, Z.; Pavaskar, P.; Hung, W. H.; Cronin, S. B., Plasmonic enhancement of photocatalytic decomposition of methyl orange under visible light. Journal of Catalysis 2011, 277 (2), 149-153. 144. Gomez, L.; Sebastian, V .; Arruebo, M.; Santamaria, J.; Cronin, S. B., Plasmon- enhanced photocatalytic water purification. Physical Chemistry Chemical Physics 2014, 16 (29), 15111-15116. 145. Hou, W.; Hung, W. H.; Pavaskar, P.; Goeppert, A.; Aykol, M.; Cronin, S. B., Photocatalytic conversion of CO2 to hydrocarbon fuels via plasmon-enhanced absorption and metallic interband transitions. ACS Catalysis 2011, 1 (8), 929-936. 146. Chien, T. M.; Pavaskar, P.; Hung, W. H.; Cronin, S.; Chiu, S. H.; Lai, S. N., Study of the plasmon energy transfer processes in dye sensitized solar cells. Journal of Nanomaterials 2015, 2015, 2. 147. Hou, W.; Pavaskar, P.; Liu, Z.; Theiss, J.; Aykol, M.; Cronin, S. B., Plasmon resonant enhancement of dye sensitized solar cells. Energy & Environmental Science 2011, 4 (11), 4650-4655. 148. Wu, N., Plasmonic metal–semiconductor photocatalysts and photoelectrochemical cells: a review. Nanoscale 2018, 10 (6), 2679-2696. 149. DuChene, J. S.; Sweeny, B. C.; Johnston, P. A. C.; Su, D.; Stach, E. A.; Wei, W. D., Prolonged hot electron dynamics in plasmonic-metal/semiconductor heterostructures with implications for solar photocatalysis. Angewandte Chemie International Edition 2014, 53 (30), 7887-7891. 150. Christopher, P.; Xin, H.; Linic, S., Visible-light-enhanced catalytic oxidation reactions on plasmonic silver nanostructures. Nature Chemistry 2011, 3 (6), 467. 151. Tian, Y .; Tatsuma, T., Plasmon-induced photoelectrochemistry at metal nanoparticles supported on nanoporous TiO2. Chem Commun (Camb) 2004, (16), 1810-1. 152. Kowalska, E.; Mahaney, O. O.; Abe, R.; Ohtani, B., Visible-light-induced photocatalysis through surface plasmon excitation of gold on titania surfaces. Phys Chem Chem Phys 2010, 12 (10), 2344-55. 153. Gao, L.; Ren, W.; Xu, H.; Jin, L.; Wang, Z.; Ma, T.; Ma, L. P.; Zhang, Z.; Fu, Q.; Peng, L. M., Repeated growth and bubbling transfer of graphene with millimetre- size single-crystal grains using platinum. Nature Communications 2012, 3, 699. 125 154. Gao, Y .; Liu, Z.; Sun, D. M.; Huang, L.; Ma, L. P.; Yin, L. C.; Ma, T.; Zhang, Z.; Ma, X. L.; Peng, L. M., Large-area synthesis of high-quality and uniform monolayer WS2 on reusable Au foils. Nature Communications 2015, 6, 8569. 155. Gao, Y .; Hong, Y . L.; Yin, L. C.; Wu, Z.; Yang, Z.; Chen, M. L.; Liu, Z.; Ma, T.; Sun, D. M.; Ni, Z.; Ma, X. L.; Cheng, H. M.; Ren, W., Ultrafast Growth of High- Quality Monolayer WSe2 on Au. Advanced Materials 2017, 29 (29). 156. Tributsch, H., Electrochemical solar cells based on layer-type transition metal compounds: Performance of electrode material. Solar Energy Materials 1979, 1 (3-4), 257-269. 157. Kline, G.; Kam, K.; Canfield, D.; Parkinson, B., Efficient and stable photoelectrochemical cells constructed with WSe2 and MoSe2 photoanodes. Solar Energy Materials 1981, 4 (3), 301-308. 158. Kam, K.; Parkinson, B., Detailed photocurrent spectroscopy of the semiconducting group VIB transition metal dichalcogenides. The Journal of Physical Chemistry 1982, 86 (4), 463-467. 159. Pavaskar, P.; Cronin, S. B., Iterative optimization of plasmon resonant nanostructures. Applied Physics Letters 2009, 94 (25), 253102. 160. Pavaskar, P.; Hsu, I. K.; Theiss, J.; Hsuan Hung, W.; Cronin, S. B., A microscopic study of strongly plasmonic Au and Ag island thin films. Journal of Applied Physics 2013, 113 (3), 034302. 161. Theiss, J.; Pavaskar, P.; Echternach, P. M.; Muller, R. E.; Cronin, S. B., Plasmonic nanoparticle arrays with nanometer separation for high-performance SERS substrates. Nano Letters 2010, 10 (8), 2749-2754. 162. Theiss, J.; Aykol, M.; Pavaskar, P.; Cronin, S. B., Plasmonic mode mixing in nanoparticle dimers with nm-separations via substrate-mediated coupling. Nano Research 2014, 7 (9), 1344-1354. 163. Mak, K. F.; Lee, C.; Hone, J.; Shan, J.; Heinz, T. F., Atomically thin MoS2: a new direct-gap semiconductor. Physical Review Letters 2010, 105 (13), 136805. 164. Mak, K. F.; He, K.; Shan, J.; Heinz, T. F., Control of valley polarization in monolayer MoS2 by optical helicity. Nature Nanotechnology 2012, 7 (8), 494. 165. Mak, K. F.; He, K.; Lee, C.; Lee, G. H.; Hone, J.; Heinz, T. F.; Shan, J., Tightly bound trions in monolayer MoS2. Nature Materials 2013, 12 (3), 207. 166. Isenberg, A. E.; Todt, M. A.; Wang, L.; Sambur, J. B., Role of Photogenerated Iodine on the Energy-Conversion Properties of MoSe2 Nanoflake Liquid Junction Photovoltaics. ACS Applied Materials & Interfaces 2018, 10 (33), 27780-27786. 167. Todt, M. A.; Isenberg, A. E.; Nanayakkara, S. U.; Miller, E. M.; Sambur, J. B., Single-Nanoflake Photo-Electrochemistry Reveals Champion and Spectator Flakes in Exfoliated MoSe2 Films. The Journal of Physical Chemistry C 2018, 122 (12), 6539-6545. 168. Wang, L.; Sambur, J. B., Efficient Ultrathin Liquid Junction Photovoltaics Based on Transition Metal Dichalcogenides. Nano Letters 2019, 19 (5), 2960-2967. 169. Francis, S. A.; Velazquez, J. M.; Ferrer, I. M.; Torelli, D. A.; Guevarra, D.; McDowell, M. T.; Sun, K.; Zhou, X.; Saadi, F. H.; John, J., Reduction of aqueous CO2 to 1-propanol at MoS2 electrodes. Chemistry of Materials 2018, 30 (15), 126 4902-4908. 170. Wang, L.; Schmid, M.; Nilsson, Z. N.; Tahir, M.; Chen, H.; Sambur, J. B., Laser Annealing Improves the Photoelectrochemical Activity of Ultrathin MoSe2 Photoelectrodes. ACS Applied Materials & Interfaces 2019, 11 (21), 19207- 19217. 171. Velazquez, J. M.; John, J.; Esposito, D. V .; Pieterick, A.; Pala, R.; Sun, G.; Zhou, X.; Huang, Z.; Ardo, S.; Soriaga, M. P., A scanning probe investigation of the role of surface motifs in the behavior of p-WSe2 photocathodes. Energy & Environmental Science 2016, 9 (1), 164-175. 172. Kang, J.; Tongay, S.; Zhou, J.; Li, J.; Wu, J., Band offsets and heterostructures of two-dimensional semiconductors. Applied Physics Letters 2013, 102 (1), 012111. 173. Chiu, M. H.; Zhang, C.; Shiu, H. W.; Chuu, C. P.; Chen, C. H.; Chang, C. Y . S.; Chen, C. H.; Chou, M. Y .; Shih, C. K.; Li, L. J., Determination of band alignment in the single-layer MoS2/WSe2 heterojunction. Nature Communications 2015, 6, 7666. 174. Furchi, M. M.; Höller, F.; Dobusch, L.; Polyushkin, D. K.; Schuler, S.; Mueller, T., Device physics of van der Waals heterojunction solar cells. npj 2D Materials and Applications 2018, 2 (1), 3. 175. Akama, T.; Okita, W.; Nagai, R.; Li, C.; Kaneko, T.; Kato, T., Schottky solar cell using few-layered transition metal dichalcogenides toward large-scale fabrication of semitransparent and flexible power generator. Scientific Reports 2017, 7 (1), 11967. 176. Wong, J.; Jariwala, D.; Tagliabue, G.; Tat, K.; Davoyan, A. R.; Sherrott, M. C.; Atwater, H. A., High Photovoltaic Quantum Efficiency in Ultrathin van der Waals Heterostructures. ACS Nano 2017, 11 (7), 7230-7240. 177. Jauregui, L. A.; Joe, A. Y .; Pistunova, K.; Wild, D. S.; High, A. A.; Zhou, Y .; Scuri, G.; De Greve, K.; Sushko, A.; Yu, C. H.; Taniguchi, T.; Watanabe, K.; Needleman, D. J.; Lukin, M. D.; Park, H.; Kim, P., Electrical control of interlayer exciton dynamics in atomically thin heterostructures. arXiv preprint arXiv:1812.08691 2018. 178. Smithe, K. K.; Krayev, A. V .; Bailey, C. S.; Lee, H. R.; Yalon, E.; Aslan, O. z. r. B.; Mun ̃ oz Rojo, M.; Krylyuk, S.; Taheri, P.; Davydov, A. V ., Nanoscale heterogeneities in monolayer MoSe2 revealed by correlated scanning probe microscopy and tip-enhanced Raman spectroscopy. ACS Applied Nano Materials 2018, 1 (2), 572-579. 179. Schrier, J.; Demchenko, D. O.; Alivisatos, A. P., Optical properties of ZnO/ZnS and ZnO/ZnTe heterostructures for photovoltaic applications. Nano Letters 2007, 7 (8), 2377-2382. 180. Perera, V .; Pitigala, P.; Jayaweera, P.; Bandaranayake, K.; Tennakone, K., Dye- sensitized solid-state photovoltaic cells based on dye multilayer-semiconductor nanostructures. The Journal of Physical Chemistry B 2003, 107 (50), 13758- 13761. 181. Peumans, P.; Bulović, V .; Forrest, S. R., Efficient photon harvesting at high optical intensities in ultrathin organic double-heterostructure photovoltaic diodes. 127 Applied Physics Letters 2000, 76 (19), 2650-2652. 182. Cheng, R.; Li, D.; Zhou, H.; Wang, C.; Yin, A.; Jiang, S.; Liu, Y .; Chen, Y .; Huang, Y .; Duan, X., Electroluminescence and photocurrent generation from atomically sharp WSe2/MoS2 heterojunction p–n diodes. Nano Letters 2014, 14 (10), 5590- 5597. 183. Pesci, F. M.; Sokolikova, M. S.; Grotta, C.; Sherrell, P. C.; Reale, F.; Sharda, K.; Ni, N.; Palczynski, P.; Mattevi, C., MoS2/WS2 heterojunction for photoelectrochemical water oxidation. ACS Catalysis 2017, 7 (8), 4990-4998. 184. Campbell, P. M.; Tarasov, A.; Joiner, C. A.; Tsai, M. Y .; Pavlidis, G.; Graham, S.; Ready, W. J.; V ogel, E. M., Field-effect transistors based on wafer-scale, highly uniform few-layer p-type WSe2. Nanoscale 2016, 8 (4), 2268-2276. 185. Docherty, C. J.; Parkinson, P.; Joyce, H. J.; Chiu, M. H.; Chen, C. H.; Lee, M. Y .; Li, L. J.; Herz, L. M.; Johnston, M. B., Ultrafast transient terahertz conductivity of monolayer MoS2 and WSe2 grown by chemical vapor deposition. ACS Nano 2014, 8 (11), 11147-11153. 186. Li, Y .; Chernikov, A.; Zhang, X.; Rigosi, A.; Hill, H. M.; van der Zande, A. M.; Chenet, D. A.; Shih, E. M.; Hone, J.; Heinz, T. F., Measurement of the optical dielectric function of monolayer transition-metal dichalcogenides: MoS2, MoSe2, WS2, and WSe2. Physical Review B 2014, 90 (20), 205422. 187. Ramasubramaniam, A., Large excitonic effects in monolayers of molybdenum and tungsten dichalcogenides. Physical Review B 2012, 86 (11), 115409. 188. Shi, H.; Cai, Z.; Patrow, J.; Zhao, B.; Wang, Y .; Wang, Y .; Benderskii, A.; Dawlaty, J.; Cronin, S. B., Monitoring Local Electric Fields at Electrode Surfaces Using Surface Enhanced Raman Scattering-Based Stark-Shift Spectroscopy during Hydrogen Evolution Reactions. ACS Applied Materials & Interfaces 2018, 10 (39), 33678-33683. 189. Wilson, J. A.; Di Salvo, F. J.; Mahajan, S., Charge-Density Waves in Metallic, Layered, Transition-Metal Dichalcogenides. Physical Review Letters 1974, 32 (16), 882-885. 190. Rossnagel, K., On the origin of charge-density waves in select layered transition- metal dichalcogenides. Journal of Physics: Condensed Matter 2011, 23 (21), 213001. 191. Adelman, T. L.; Zaitsev-Zotov, S. V .; Thorne, R. E., Field-Effect Modulation of Charge-Density-Wave Transport in NbSe3 and TaS3. Physical Review Letters 1995, 74 (26), 5264-5267. 192. Goli, P.; Khan, J.; Wickramaratne, D.; Lake, R. K.; Balandin, A. A., Charge Density Waves in Exfoliated Films of van der Waals Materials: Evolution of Raman Spectrum in TiSe2. Nano Letters 2012, 12 (11), 5941-5945. 193. Kusmartseva, A. F.; Sipos, B.; Berger, H.; Forró, L.; Tutiš, E., Pressure Induced Superconductivity in Pristine 1T-TiSe2. Physical Review Letters 2009, 103 (23), 236401. 194. Yang, J.; Wang, W.; Liu, Y .; Du, H.; Ning, W.; Zheng, G.; Jin, C.; Han, Y .; Wang, N.; Yang, Z.; Tian, M.; Zhang, Y ., Thickness dependence of the charge-density- wave transition temperature in VSe2. Applied Physics Letters 2014, 105 (6), 128 063109. 195. Friend, R. H.; Jérome, D.; Schleich, D. M.; Molinié, P., Pressure enhancement of charge density wave formation in VSe2; The role of coulomb correlations. Solid State Communications 1978, 27 (2), 169-173. 196. Eaglesham, D. J.; Withers, R. L.; Bird, D. M., Charge-density-wave transitions in 1T-VSe2. Journal of Physics C: Solid State Physics 1986, 19 (3), 359. 197. Westover, R. D.; Atkins, R. A.; Ditto, J. J.; Johnson, D. C., Synthesis of [(SnSe)1.16–1.09]1[(NbxMo1–x) Se2]1 Ferecrystal Alloys. Chemistry of Materials 2014, 26 (11), 3443-3449. 198. Falmbigl, M.; Fiedler, A.; Atkins, R. E.; Fischer, S. F.; Johnson, D. C., Suppressing a Charge Density Wave by Changing Dimensionality in the Ferecrystalline Compounds ([SnSe]1.15)1(VSe2)n with n = 1, 2, 3, 4. Nano Letters 2015, 15 (2), 943-948. 199. Hite, O. K.; Falmbigl, M.; Alemayehu, M. B.; Esters, M.; Wood, S. R.; Johnson, D. C., Charge density wave transition in (PbSe)1+ δ(VSe2)n compounds with n= 1, 2, and 3. Chemistry of Materials 2017, 29 (13), 5646-5653. 200. Harper, J. M. E.; Geballe, T. H.; DiSalvo, F. J., Thermal properties of layered transition-metal dichalcogenides at charge-density-wave transitions. Physical Review B 1977, 15 (6), 2943-2951. 201. Smontara, A.; Biljaković, K.; Artemenko, S. N., Contribution of charge-density- wave phase excitations to thermal conductivity below the Peierls transition. Physical Review B 1993, 48 (7), 4329-4334. 202. Mavrokefalos, A.; Lin, Q.; Beekman, M.; Seol, J. H.; Lee, Y . J.; Kong, H.; Pettes, M. T.; Johnson, D. C.; Shi, L., In-plane thermal and thermoelectric properties of misfit-layered [(PbSe)0.99]x(WSe2)x superlattice thin films. Applied Physics Letters 2010, 96 (18), 181908. 203. Mavrokefalos, A.; Nguyen, N. T.; Pettes, M. T.; Johnson, D. C.; Shi, L., In-plane thermal conductivity of disordered layered WSe2 and (W)x(WSe2)y superlattice films. Applied Physics Letters 2007, 91 (17), 171912. 204. Mahajan, R.; Barkeshli, M.; Hartnoll, S. A., Non-Fermi liquids and the Wiedemann-Franz law. Physical Review B 2013, 88 (12), 125107. 205. Bayard, M.; Sienko, M., Anomalous electrical and magnetic properties of vanadium diselenide. Journal of Solid State Chemistry 1976, 19 (4), 325-329. 206. Xu, K.; Chen, P.; Li, X.; Wu, C.; Guo, Y .; Zhao, J.; Wu, X.; Xie, Y ., Ultrathin Nanosheets of Vanadium Diselenide: A Metallic Two-Dimensional Material with Ferromagnetic Charge-Density-Wave Behavior. Angewandte Chemie International Edition 2013, 52 (40), 10477-10481. 207. Pérez, N.; Chirkova, A.; Skokov, K. P.; Woodcock, T. G.; Gutfleisch, O.; Baranov, N. V .; Nielsch, K.; Schierning, G., Thermoelectric determination of electronic entropy change in Ni-doped FeRh. arXiv preprint arXiv:1808.04183 2018. 129 130 Appendix: Supplemental Documents Supplemental Document for Chapter 2 Figure S2.1. PL image before (a)and after (b) a 2-minute UV/ozone treatment. 131 Supplemental Document for Chapter 3 AFM image of the MoS2 flake: To check the thickness for the MoS2 flakes, we performed AFM measurements of the samples after the transfer, which are shown below in Figure S3.1. The top part of the flake is on top the metal electrode, while the bottom part is on Si/SiO2 substrate. Figure S3.1. AFM images (left) and height profiles extracted from the image (right) of few layer MoS2 films. The scale bar at the bottom-left corner of the image is 2μm. The black line in image indicates where the profile is selected. 132 Supplemental Document for Chapter 4 Figure S4.1. Normalized Raman spectra of the Bi2Te3 flake before and after Al2O3 deposition. Figure S4.2. Schematic diagram of the device fabrication process (a) and optical image of the device before (b) and after (c) Al2O3 deposition. 0 50 100 150 200 250 300 350 400 450 0.0 0.2 0.4 0.6 0.8 1.0 Before Al 2 O 3 deposition After Al 2 O 3 deposition Normalized Raman Intensity Raman Shift (cm -1 ) Heater Bi 2 Te 3 Flake RTD Al 2 O 3 Deposition (a) (b) (c) 20μm 20μm 133 Figure S4.3. Temperature calibration of the left metal RTD of the Bi2Te3 device shown in Figure 4.1. (e) The resistance of the RTD measured at different temperatures. (f) Normalized resistance (R/R0, R0 is the resistance at 300K) plotted as a function of T. (g) The resistance change of the RTDs under various heating currents. (h) Temperature change of the metal RTD plotted as a function of heating power. 134 Figure S4.4. Temperature calibration of the right metal RTD of the Bi2Te3 device shown in Figure S4.2. (a) The resistance of the RTD measured at different temperatures. (b) Normalized resistance (R/R0, R0 is the resistance at 300K) plotted as a function of T. (c) The resistance change of the RTDs under various heating currents. (d) Temperature change of the metal RTD plotted as a function of heating power. 135 Figure S4.5. Temperature calibration of the left metal RTD of the Bi2Te3 device shown in Figure S4.2. (e) The resistance of the RTD measured at different temperatures. (f) Normalized resistance (R/R0, R0 is the resistance at 300K) plotted as a function of T. (g) The resistance change of the RTDs under various heating currents. (h) Temperature change of the metal RTD plotted as a function of heating power. 136 Figure S4.6. In-plane Seebeck coefficient of the Bi2Te3 flake before (a) and after (b) Al2O3 deposition. 137 Figure S4.7. Transmission Electron Microscope (TEM) Energy Dispersive Spectroscopy (EDS) analysis for Focused Ion Beam (FIB) lamella of Bi2Te3 with (a) and without (b) ALD-passivation. (a) (b) 138 Supplemental Document for Chapter 5 Figure S5.1. Temperature calibration of the top resistive temperature detector (RTD) of 50nm SnSe film without Se-vapor annealing. (a) The resistance of the RTD measured at different temperatures. (b) Normalized resistance (R/R0, where R0 is the resistance at 300K) plotted as a function of T. (c) The resistance change of the RTD under various heating currents. (d) Temperature change of the metal RTD plotted as a function of heating power. 139 Figure S5.2. Temperature calibration of the bottom RTD of 50nm SnSe film without Se-vapor annealing. (a) The resistance of the RTD measured at different temperatures. (b) Normalized resistance (R/R0, where R0 is the resistance at 300K) plotted as a function of T. (c) The resistance change of the RTD under various heating currents. (d) Temperature change of the metal RTD plotted as a function of heating power. 300 305 310 315 320 325 330 6.50 6.55 6.60 6.65 6.70 6.75 6.80 Temperature (K) R (Ohm) Bottom RTD 300 305 310 315 320 325 330 1.00 1.01 1.02 1.03 1.04 1.05 Bottom RTD Fitting R/R 0 Temperature (K) 0.0 0.5 1.0 1.5 2.0 6.46 6.48 6.50 6.52 6.54 6.56 Heater Current (mA) R (Ohm) Bottom RTD 0 1 2 3 4 5 6 7 0 2 4 6 8 10 12 Heater Power (mW) T (K) Bottom RTD Fitting (a) (b) (c) (d) 140 Figure S5.3. Temperature calibration of the top RTD of 50nm SnSe2 film after Se-vapor annealing. (a) The resistance of the RTD measured at different temperatures. (b) Normalized resistance (R/R0, where R0 is the resistance at 300K) plotted as a function of T. (c) The resistance change of the RTD under various heating currents. (d) Temperature change of the metal RTD plotted as a function of heating power. -3 -2 -1 0 1 2 3 -20 -10 0 10 Leak through ALD Leakage Current (pA) Bias Voltage (V) Figure S5.4. Leakage current measurement through the atomic layer deposition layer. 300 305 310 315 320 325 330 4.40 4.44 4.48 4.52 4.56 Top RTD Temperature (K) R (Ohm) 300 305 310 315 320 325 330 1.00 1.01 1.02 1.03 1.04 Top RTD Fitting Temperature (K) R/R 0 0.0 0.5 1.0 1.5 2.0 4.375 4.400 4.425 4.450 Top RTD Heater Current (mA) R (Ohm) 0 1 2 3 4 5 6 7 0 2 4 6 8 10 12 14 16 Top RTD Fitting Heater Power (mW) T (K) (a) (b) (c) (d) 141 Figure S5.5. Cross-plane Seebeck coefficient of a (a) 50nm-thick and (b)100nm-thick SnSe film (without Se annealing). The cross-plane electrical resistances of the 50nm and 100nm samples were 116Ω and 197Ω, respectively. The thermal resistance of the 50nm and 100nm samples were 261 K/W and 455 K/W, respectively. Based on these values, we can estimate that the electrical contact resistance is 35Ω, the intrinsic cross-plane resistivity of SnSe2 is 1.89 Ω-m, the thermal contact resistance is 67 K/W, and the intrinsic cross-plane thermal conductivity of SnSe2 is 0.89 W/m·K. Using these values, we estimate that the intrinsic figure of merit for the film in Figure 5 of main manuscript to be approximately ZT7.1×10 -5 after removing the effects of contact resistance. 0.0 0.4 0.8 1.2 1.6 0 20 40 60 DC pos DC neg Fitting T (K) V (V) S= - 38.6 V/K (a) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 0 10 20 30 40 DC pos DC neg Fitting T (K) V (V) S= - 37.5 V/K (b) t=100nm t=50nm 142 Figure S5.6. Cross-plane electrical resistances of (a) 50nm-thick and (b)100nm-thick SnSe film (without Se annealing), respectively. Figure S5.7. Seebeck coefficient as a function of Fermi energy for SnSe and SnSe2, respectively. 0.0 0.2 0.4 0.6 0.8 1.0 0 2 4 6 8 Current (mA) Voltage (V) 50nm SnSe Fitting 0.0 0.2 0.4 0.6 0.8 1.0 0 1 2 3 4 5 100nm SnSe Fitting Current (mA) Voltage (V) (a) (b) E F (ev) Seebeck Coefficient (μV/K) SnSe 2 SnSe 143 Figure S5.8. Calculated range of errors that correspond to the expected range of interface conductance values found in the literature. The in-plane resistivities of samples annealed in nitrogen and Se were 38.64 mΩ-m and 23.33 mΩ-m, respectively. Here, we observe a 40% reduction in the resistivity due to the mitigation of Se vacancies. This is compared to the 6.6-fold increase in the cross- plane resistance observed after Se annealing. The in-plane Seebeck coefficients of samples annealed in nitrogen and Se were 592V/K and -342V/K, respectively. Here, we observe a dramatic change in sign of the thermopower due to the mitigation of Se vacancies. This is compared to the 16-fold increase in the cross-plane thermopower observed after Se annealing. Here, the large discrepancy between in-plane and cross- plane transport behavior further demonstrates the highly anisotropic nature of these rotationally-disordered layered materials. In these anisotropic layered materials, in- plane transport is dominated by the more conducting layer, while the cross-plane transport is dominated by the more insulating layer. 10 8 10 9 Interface Conductance [W /m 2 K] 0.7 0.75 0.8 0.85 0.9 0.95 T [K] T M eausured T Actual 10 8 10 9 Interface Conductance [W /m 2 K] 0 2 4 6 8 10 12 14 Error in T [% ] (a) (b) 10 8 10 9 Interface Conductance [W /m 2 K] 0.7 0.75 0.8 0.85 0.9 0.95 T [K] T M eausured T Actual 10 8 10 9 Interface Conductance [W /m 2 K] 0.7 0.75 0.8 0.85 0.9 0.95 T [K] T M eausured T Actual 144 Supplemental Document for Chapter 6 Figure S6.1. Schematic circuit diagram of the AC lock-in technique, in which the incident light is modulated by a chopper wheel and chopper controller, which provides a reference signal for the lock-in amplifier. This enables the lock-in amplifier to detect only voltages at the specific frequency of the modulated light, providing a very sensitive measure of the photo-response of these photocatalytic surfaces. Gamry Working Electrode Terminal Signal Reference Frequency 145 Figure S6.2. (a) DC current, (b) AC current, and (c) AC phase measurements for the ITO electrode with only 5nm Au nanoislands without WSe2. 0 20 40 60 80 100 120 0 2 4 6 8 AC Current Current (nA) Time(s) 0 20 40 60 80 100 120 -200 -100 0 100 200 AC Phase Time(s) Phase (Degree) 0 40 80 120 160 200 0 10 20 30 40 Current (A) DC Current Time (s) (a) (b) (c) 146 Figure S6.3. (a) Intensity dependent of photocurrent measurement for 532 nm wavelength (b) Intensity dependent of photocurrent measurement for 633 nm wavelength. 0.09 0.25 1 0 50 100 150 200 250 300 633nm without 5nm Au 633nm with 5nm Au Photocurrent (nA) Power Intensity (mW) 0.28 1 3.7 0 200 400 600 800 532nm without 5nm Au 532nm with 5nm Au Photocurrent (nA) Power Intensity (mW) (a) (b) 147 Figure S6.4. Enhancement factor of the electric field intensity at the Au-WSe 2 interface for 633 nm wavelength calculated using the FDTD method. An alternative growth method of monolayer WSe2: WSe2 was grown on gold foil by ambient-pressure CVD with solid precursors 155 , which is similar to that previously shown for WS2 monolayer growth. 154-155 A piece of polycrystalline Au foil was first finely polished and annealed, and then placed in a small quartz boat together with 100 mg WO3 powder upstream, loaded at the reaction zone of a horizontal CVD furnace equipped with a 1-inch-diameter quartz tube. Another small quartz boat containing 1000 mg Se pellets was placed upstream at the low temperature zone of the furnace. After the reaction zone was heated to 900 °C in an argon atmosphere (100 s.c.c.m.), a small flow rate of H2 (0.5 s.c.c.m.) was introduced to initiate WSe2 growth. After several minutes, we rapidly stopped the growth by quickly moving the two boats to low-temperature zone. Monolayer WSe2 was then transferred from the gold foil onto an indium tin oxide (ITO)-coated glass slide by electrochemical bubbling transfer method using a drop-casted PMMA support layer. 153, 155 50nm 50 nm 10 3 10 2 10 1 10 0 10 -1 148 Supplemental Document for Chapter 7 Figure S7.1. (a) Schematic diagram of the three-terminal photoelectrochemical setup with the modulated laser and AC lock-in amplifier. (b) Schematic circuit diagram of the AC lock-in measurement, in which the incident light is modulated with a chopper wheel and controller, which provides a reference signal for synchronization of the lock- in amplifier. (a) (b) Gamry Potentiostat Pt Counter Electrode WSe 2 ITO/Glass Slide Substrate Ag/AgCl Reference Electrode 1 M NaI Solution Lock-in Amplifier Chopper Incident Laser MoSe 2 Gamry Working Electrode Terminal Signal Reference Frequency

Abstract (if available)

Abstract This dissertation work presents several research investigations on nanoscale electronic devices based on low-dimensional materials, including one-dimensional material--carbon nanotube, and a set of two-dimensional materials, such as MoS₂, Bi₂Te₃, SnSe₂, WSe₂ and WSe₂/MoSe₂ heterostructures. These research projects shed light on the fundamental physics and potential optoelectronic, thermoelectric and photocatalytic applications of these nanomaterials and their hetero-structures. It could provide a reference and have good enlightenment for future studies. ❧ Chapter 1 provides background information that will help to understand this dissertation work. It contains a brief overview of the properties of one-dimensional and two-dimensional materials, as well as their electronic band structures, which is important to understand the optoelectronic applications of these low-dimensional materials. A brief introduction of thermoelectric and photocatalysis are also included to help understanding the thermoelectric and photocatalytic applications of these nanomaterials and their heterostructures. ❧ Chapter 2 presents an optoelectronic application of air-suspended carbon nanotubes. It reveals photoluminescence imaging and spectroscopy of air-suspended carbon nanotubes can be enhanced by oxygen doping with a direct ultraviolet/ozone treatment exposure, which is more efficient than previous methods. This ability to control and engineer defects in carbon nanotubes is important for realizing several optoelectronic applications such as LEDs and single photon sources. ❧ Chapter 3 presents an optoelectronic application of the most famous transition metal dichalcogenides, MoS₂, in two-dimensional material family. A highly efficient, high speed vertical photodiode can be achieved by using few-layer MoS₂. Compared with in-plane devices, it exhibits a substantially larger photocurrent and shorter photo-response, which could pave the way for a more rational device design to boost performance. ❧ Chapter 4 and Chapter 5 present thermoelectrical studies of two-dimensional materials, Bi₂Te₃ and SnSe₂, which has been known as good candidates for thermoelectrical applications. Chapter 4 reports a deposition of Al₂O₃ using atomic layer deposition can enhance the thermoelectric properties of the materials, which offers an appropriate surface passivation method of this high surface-to-volume ratio nanoflake to prevent the degradation of its thermoelectric properties. Chapter 5 reveals the cross-plane thermoelectric transport of rotationally-disordered SnSe₂ can be enhanced via Se vapor annealing, which demonstrates a robust method for mitigating unintentional doping in a promising class of materials for thermoelectric applications. ❧ Chapter 6 presents the photocatalytic application of monolayer WSe₂. Plasmon-resonant is used to enhance the photocatalysis performance of monolayer WSe₂ and improve the overall photoconversion efficiency. The incident light coupled with plasmonic nanoislands effectively from the far-field to the near field in the plane of the monolayer WSe₂. The fundamental science and application in these two chapters could be a source of inspiration for increasing the utilization of renewable environment-friendly solar energy. ❧ Chapter 7 presents the photocatalytic application of transition metal dichalcogenide (TMDC) heterostructures. We discuss the effect of the stacking order of two TMDC materials, WSe₂ and MoSe₂, in their heterostructures on the photocatalytic performance. ❧ Chapter 8 presents the future work of the thermoelectric applications of transition metal dichalcogenide (TMDC) heterostructures. For the next step, we will explore the effect of charge density wave (CDW) transitions on the cross-plane and in-plane thermoelectric transport properties of these heterostructure materials.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Light Emission from Carbon Nanotubes and Two-Dimensional Materials

PDF

In plane a & cross-plane thermoelectric characterizations of van der Waals heterostructures

PDF

GaAs nanowire optoelectronic and carbon nanotube electronic device applications

PDF

Synthesis, characterization, and device application of two-dimensional materials beyond graphene

PDF

Carbon nanotube nanoelectronics and macroelectronics

PDF

Hot carriers in bare metals and photocatalytically active defect sites in dielectric/metal structures

PDF

Printed electronics based on carbon nanotubes and two-dimensional transition metal dichalcogenides

PDF

In-situ characterization of nanoscale opto-electronic devices through optical spectroscopy and electron microscopy

PDF

Electronic and optoelectronic devices based on quasi-metallic carbon nanotubes

PDF

Thermoelectric and transport studies of low dimensional materials & hot electron-driven photocatalysis on plasmon-resonant grating nanostructures

PDF

Design and characterization of metal and semiconducting nanostructures and nanodevices

PDF

Transient pulsed plasma for pollution remediation and energy conversion applications

PDF

Carbon nanotube macroelectronics

PDF

Optical and electrical characterization of one-dimensional (1D) and two-dimensional (2D) nanostructures

PDF

Raman spectroscopy of carbon nanotubes under axial strain and surface-enhanced Raman spectroscopy of individual carbon nanotubes

PDF

Low-dimensional material based devices for neuromorphic computing and other applications

PDF

Optoelectronic properties and device physics of individual suspended carbon nanotubes

PDF

Plasmonic enhancement of catalysis and solar energy conversion

PDF

Carbon material-based nanoelectronics

PDF

Nanomaterials for macroelectronics and energy storage device

Asset Metadata

Creator Chen, Jihan (author)

Core Title Optoelectronic, thermoelectric, and photocatalytic properties of low dimensional materials

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Electrical Engineering

Publication Date 12/12/2019

Defense Date 05/07/2019

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag carbon nanotubes,OAI-PMH Harvest,optoelectronic,photocatalytic,thermoelectric,two-dimensional materials

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Cronin, Stephen Burke (committee chair), Nakano, Aiichiro (committee member), Wang, Han (committee member)

Creator Email chenjihan17@gmail.com,jihanche@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-252175

Unique identifier UC11674767

Identifier etd-ChenJihan-8051.pdf (filename),usctheses-c89-252175 (legacy record id)

Legacy Identifier etd-ChenJihan-8051.pdf

Dmrecord 252175

Document Type Dissertation

Rights Chen, Jihan

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA