University of Southern California Dissertations and Theses

Materials development and characterization of optically-active core-shell nanomaterials

(USC Thesis Other)

Materials development and characterization of optically-active core-shell nanomaterials

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Materials Development and Characterization of Optically-Active Core-Shell Nanomaterials By Rene Zeto A Dissertation Presented to the FACULTY OF THE USC GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulllment of the Requirements for the Degree DOCTOR OF PHILOSOPHY Materials Science August 2021 Copyright 2021 Rene Zeto Acknowledgements I am grateful to have had the opportunity to pursue a PhD, an experience that not many people are privileged enough to attain. Along the way, I was supported by many friends, family, and colleagues. I would like to thank my advisor, Professor Andrea Armani, for teaching me so much about scientic research and for cultivating a unique group of talented individuals that I could work with. The PhD process has its ups and downs, all of which I am thankful for. I took away much from it, and it helped shape my values and enabled me to be the scientist I am today. I would like to thank the rest of my committee: Professor Jayakanth Ravichandran, Professor Wei Wu, and Professor Mark Thompson. Additionally, I am deeply appreciative to the USC Core Center of Excellence in Nano Imaging (USC CNI), with whom I worked for two years during my PhD. I learned a great deal about electron microscopy and materials characterization there, and used its instruments to collect data that is used in this thesis. Special thanks to John Curulli, Carolyn Marks, and Matthew Mecklenberg, as well as my graduate research assistant colleagues Brian Feng and Mythili Surendran. ii Contents Acknowledgements ii List of Tables vi List of Figures vii Abbreviations xii Abstract xiv 1 Introduction 1 1.1 Gold nanoparticles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 ZnO nanostructures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.3 AuZnO core-shell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.4 Core-shell quantum dots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.5 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2 Materials Background 12 2.1 Plasmonic resonance theory . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 2.2 Gold NP synthesis & characterization . . . . . . . . . . . . . . . . . . . . . . 15 2.3 ZnO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2.4 ZnO properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2.5 ZnO characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.6 ZnO defects and dopants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 iii 2.7 Types of ZnO nanostructures . . . . . . . . . . . . . . . . . . . . . . . . . . 25 2.8 Core-shell quantum dots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 2.9 QHMF quantum dots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 2.10 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 2.11 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 3 Core-shell nanoparticles 36 3.1 Background: synthetic challenges . . . . . . . . . . . . . . . . . . . . . . . . 36 3.2 Background: AuZnO core-shell nanoparticles . . . . . . . . . . . . . . . . . . 40 3.3 Background: calcining ZnO . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 3.4 Doped AuZnO: background and motivation . . . . . . . . . . . . . . . . . . 46 3.5 Doped AuZnO: experimental design . . . . . . . . . . . . . . . . . . . . . . . 46 3.6 Doped AuZnO: results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3.7 Doped AuZnO: Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 3.8 Future work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 3.9 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 3.10 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 4 ZAIS Quantum Dots 64 4.1 Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 4.2 Purication and characterization . . . . . . . . . . . . . . . . . . . . . . . . 70 4.3 Experimental goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 4.4 Measuring disorder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 4.5 Results & discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 4.6 Future work: blinking background . . . . . . . . . . . . . . . . . . . . . . . . 78 4.7 Future Work: blinking characterization procedure . . . . . . . . . . . . . . . 81 4.8 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 4.9 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 iv 5 Applied projects & Future works 90 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 5.2 ZnO microlaser project: introduction . . . . . . . . . . . . . . . . . . . . . . 90 5.3 ZnO microlaser: laser background . . . . . . . . . . . . . . . . . . . . . . . . 91 5.4 ZnO microlaser: resonant cavities . . . . . . . . . . . . . . . . . . . . . . . . 91 5.5 ZnO microlaser: whispering gallery mode resonators . . . . . . . . . . . . . . 92 5.6 ZnO microlaser: resonator-material interactions . . . . . . . . . . . . . . . . 95 5.7 ZnO microlaser: random lasing . . . . . . . . . . . . . . . . . . . . . . . . . 98 5.8 ZnO microlaser: challenges & future works . . . . . . . . . . . . . . . . . . . 99 5.9 ZnO microlaser: synthesis and characterization . . . . . . . . . . . . . . . . . 101 5.10 QHMF LED device: introduction . . . . . . . . . . . . . . . . . . . . . . . . 102 5.11 QHMF LED device: emissive layers in LED devices . . . . . . . . . . . . . . 102 5.12 QHMF LED device: device-level improvements . . . . . . . . . . . . . . . . . 104 5.13 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 5.14 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 v List of Tables 3.1 Table of peak identication labels for g. 3.6. . . . . . . . . . . . . . . . . . 45 vi List of Figures 1.1 Left: TEM image of four AuZnO core-shell nanoparticles. Due to the higher density of the core, it appears darker, showing the contrast between the core and the shell layers. The ZnO shell protects the gold core and adds the possibility for tuning via material modication of the shell explored in chapter 3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.2 Progressive quenching of uorescent emission in a type of core-shell quantum dot. This thermal quenching eect is a unique nanoscale behavior to this system, which I explore fully in chapter 4. . . . . . . . . . . . . . . . . . . . 6 2.1 Two examples of the tunable optical parameters for gold nanoparticles. (A) Gold nanorods synthesized with a shorter rod length, approximately 50 nm. (B) Gold nanorods synthesized with a longer rod length, estimated to be about 200-300 nm. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 2.2 Schematic of UV Vis, a common optical characterization technique for nanopar- ticles. UV Vis lters a broadband light source and records transmittance through the sample. This curve is often characteristic of intrinsic material properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2.3 Schematic DLS, another common optical characterization technique for nanopar- ticles. DLS makes assumptions about the particles' shape and motion, and extrapolates size distributions based on the observed laser speckle patterns. . 18 vii 2.4 Left: TEM image of a gold NR showing lattice fringes. This method of imag- ing conveys additional information about the particles that is not present in the SEM image. Right: SEM of mixture of gold nanospheres and nanorods. The geometric contrast is apparent. This manifests itself in very dierent scattering and extinction cross sections, which can be tuned to dierent ap- plications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 2.5 FTIR spectra of ZnO, conrming successful synthesis but also indicating some water adsorption had occurred. Surface chemistry impact such as water ad- sorption can be a critical problem for some applications of ZnO nanowires. . 24 2.6 Wurtzite crystal structure ZnO and its impact on ZnO nanoparticle geometry. The hexagonal symmetry is clearly seen under an SEM image. This structure has in uence over its optical properties. SEM image my own work; crystal structure diagram reproduced from [20] with modication (public domain). . 26 2.7 Schematic of simple LED architecture in which QHMF quantum dots play a role as the active layer, causing emission of the red, green, and blue light nec- essary to form an image on a display. This device leverages the wide spectral tunability of QHMF quantum dots permitted by the exible compositional freedom. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 2.8 Synthesized QHMF quantum dot samples characterized for elemental compo- sition via x-ray photoelectron spectroscopy. The spectra conrm a relatively uniform batch of QHMF quantum dots formed. A diagram of the core-shell structure is shown inset. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 3.1 A diagram illustrating the impact of lattice compatibility at an interface be- tween two dierent crystalline materials. If the lattice is well-matched in the lateral direction, minimal strain occurs. Otherwise, strain and defects such as dislocations can occur. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 viii 3.2 Scanning electron microscope image of AuZnO without sintering (outset), with sintering (inset). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 3.3 TEM image of an AuZnO core-shell nanoparticle. . . . . . . . . . . . . . . . 40 3.4 Strategy for synthesis showing the possible nanoparticle variations, includ- ing doping. Figure created by my lab mate, Mark Veksler (reproduced with permission). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 3.5 UV Vis pre- and post-calcine, demonstrating conversion of the Zn-based mesh shell to ZnO. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 3.6 (A) Pre- and (B) post-calcine XRD scans. The ZnO-associated peaks I, O, and P are only present in the post-calcine scan. . . . . . . . . . . . . . . . . 45 3.7 Scanning electron microscope images of pre-calcine (A-B) and post-calcine (C- D) ZnO particles, showing changes in surface texture. The size distributions are unchanged as the calcining process only converts the shell material into ZnO. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 3.8 Left: partially etched AuZnO core-shell nanoparticles. Right: Totally etched core-shell nanoparticles, leaving only the cores. . . . . . . . . . . . . . . . . . 49 3.9 Sample data providing evidence for successful Ce doping in AuZnO nanopar- ticles. A. TEM image, B. ZnO Spatial EDX, C. Ce Spatial EDX, D. EDX spectrum. The correlation of the ZnO with Ce provides evidence that the Ce intercalated into the ZnO shell. . . . . . . . . . . . . . . . . . . . . . . . . . 50 3.10 DLS spectra of particles synthesized with and without gold cores, highlight- ing the enhanced stability oered by the gold NP core. A. SEM image of the AuZnO particles. B. Narrow DLS distribution of particles with cores despite added dopants. C. SEM image of doped particles without gold cores, show- ing dierent morphology. D. Wide distribution of the corresponding no-core particles. In (B), the second peak between 1 m and 10 m is due to observed contaminants in that particular sample. . . . . . . . . . . . . . . . . . . . . . 54 ix 4.1 Synthesis experimental setup. Reagents are deposited via the access septum, while N 2 is constantly owing through the reaction pot. . . . . . . . . . . . . 66 4.2 Two dierent batches of synthesized ZAIS quantum dots with dierent pho- toluminescence curves. The number of shells grown on the quantum dot in- uence its photoluminescence wavelength. The top row was prepared with one ZnS shell, and the bottom row was prepared with no ZnS shell, highlight- ing the signicant impact shell number has on the luminescence properties of ZAIS quantum dots. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 4.3 Optical absorption spectra for two dierent samples of quantum dots pro- duced. The dashed line represents the theoretical sharp band edge, and the amount that the measured solid line deviates from that may be quantied as the Urbach energy E u . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 4.4 A sharp drop in PL quantum yield associated with an increase in sample Urbach energy at high temperatures, correlating a highly disordered lattice with a loss of luminescence. Quantum yield was unmeasurable beyond 240 C. 76 4.5 UPS data of the annealed samples, indicating a signicant electronic structural change. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 4.6 Quenched quantum yield with increasing temperatures also correlates with a sharp increase in the excited state lifetime for one of the states responsible for luminescence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 4.7 Intensity trace for a quantum dot recorded in a video on a uorescent mi- croscope. This demonstrates proof of concept for measuring ZAIS blinking. The clear on/o behavior is seen at about 50 s and then 100 s. Excitation wavelength: 480 nm, emission wavelength: 560 nm. . . . . . . . . . . . . . . 82 x 5.1 Cartoon schematic of a simple laser cavity. Spatial coherence is achieved through the long optical path length: any ray with a slight angle would be ltered via the many round trips the light takes between the mirrors in the cavity. Temporal coherence is also achieved through the long optical path length: any ray with the wrong wavelength would suer from self interference via the many round trips in the cavity. . . . . . . . . . . . . . . . . . . . . . 93 5.2 Schematic and SEM image of a toroid microresonator. Light circulates in the rim of the suspended toroid, which has a chance to interact with the functional material coating on top of the device. In the SEM image, a ZnO nanowire coating was attempted, but only achieved partial coverage. . . . . . . . . . . 93 5.3 Diagram of the electronic state changes occurring in two photon absorption. Although the same excitation can be achieved via two photon absorption as in normal (single photon) absorption, it is much less likely to occur. . . . . . 98 5.4 Cartoon schematic of the random lasing process. Scattering o of dense ZnO nanowire array structures allows for the optical path length to accumulate, while the amplication comes from a population inversion stimulated in the ZnO nanowires. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 5.5 SEM images of a toroid subject to hydrothermal ZnO nanowire growth. I was able to accomplish uniform coverage, but the process introduced too much byproduct deposition on the surface of the toroid. (A) Zoomed out, (B) close up view of the nanowires on the side of the toroid. . . . . . . . . . . . . . . . 100 xi Abbreviations CTAB cetrimonium bromide. 15, 16, 38, 39, 41, 43, 57 DLS Dynamic Light Scattering. 16, 17, 18, 19 EDX Energy-dispersive X-ray Spectroscopy. 51, 52, 53 EML emission layer. 104, 105 EQE external quantum eciency. 104, 105 FTIR Fourier Transform Infrared Spectroscopy. 16, 19, 23 HMTA hexamethylenetetramine. 41, 52, 101 HOMO highest occupied molecular orbital. 104, 105 HTL hole transport layer. 104, 105 QHMF quaternary heavy-metal-free. 1, 7, 27, 28, 29, 30, 64, 65, 73, 75, 79, 101, 102, 103, 104, 105, 106 SEM Scanning Electron Microscope. 16, 19, 21, 23, 38, 48, 51, 52, 53, 54, 100 TEM Transmission Electron Microscope. 16, 19, 20, 21, 23, 51, 52, 53 TRPL Time-resolved Photoluminescence Decay. 73, 83 UPS Ultraviolet Photoelectron Spectroscopy. 73, 76, 77, 83 xii UV-VIS Ultraviolet-Visible Spectroscopy Instrument. 15, 16, 17, 19, 74 XRD X-ray Diraction. 47, 51, 53, 54 ZAIS AgInS 2 -ZnS. xiv, xv, 6, 64, 65, 66, 70, 71, 72, 74, 75, 80, 81, 83, 102, 103 xiii Abstract Core-shell nanoparticles are a class of nanomaterials comprised of two or more distinct ma- terials composed in a layered, spherical shape. These nanomaterials represent the rst step towards more complicated nanoscale structures and functionality, which has been an ongoing technological goal in the scientic community, leading to advances in and miniaturization of important capabilities such as computing and sensors. In this thesis, I explore two unique core-shell particle platforms. The rst core-shell particle platform is a metal-insulator combination, AuZnO, which has a gold core surrounded by a thick and porous ZnO shell. These materials are synergistic in many ways. The gold core acts as an optically active sensor or receiver. It has a tunable surface plasmon resonance, the frequency of which is sensitive to local dielectric modication, and therefore can act as a nanoscale sensor for changes in the local environment. The ZnO shell provides a porous yet protective layer which prevents uncontrolled modication of the gold core surface. Run in reverse, the gold nanoparticle core is a nanoantenna, receiving optical energy and dispersing it throughout the ZnO shell, which may contain a payload that can be released upon the external optical trigger. I demonstrate the successful modication of this platform to incorporate dopant elements. I also describe experiments which provide insight into the role of the gold core in stabilizing particle morphology, particularly with higher dopant concentrations. My work expands understanding of the AuZnO platform and introduces a new vector for tunability. The second core-shell particle platform is a layered semiconductor quantum dot, AgInS 2 - ZnS (ZAIS). Semiconductor quantum dots are known to be ecient and photostable emitters with applications as bioimaging agents and LED device emission layer materials. However, xiv the state-of-the-art quantum dot materials contain heavy metals. ZAIS is a promising al- ternative quantum dot material which is heavy metal free, but some details about its lumi- nescence mechanism remain unclear. I perform experiments and analysis to shed light on the internal defect-related luminescence mechanism in ZAIS quantum dots, and nd that its thermal sensitivity places limits on the classes of defect responsible for the emission. xv Chapter 1 Introduction Core-shell optical nanomaterials are a class of nanoparticles made up of two or more distinct material regions. The formed particles have a spherical symmetry and are on the scale from 5 nm to 100 nm in size. In this dissertation, I will discuss four main types of nanoparticles that I investigated over the course of my PhD: gold nanospheres and nanorods, ZnO nanos- tructures, AuZnO core-shell structures, and a class of semiconductor quantum dots known as quaternary heavy-metal-free (QHMF) quantum dots. I will also introduce two future works side projects I worked on in chapter 5, which are more on the applied materials engineering side of my thesis work. 1.1 Gold nanoparticles There is much academic discussion about nanoparticles, and so it is easy to miss the forest for the trees and not realize why nanoparticle work is so important and researched. Bulk materials each have their own set of classically intrinsic properties that, to some extent, dene the material. Copper has a conductance of 58 MS/m [1], contributing to its usage in electrical wiring; gold has a sharp re ectivity shoulder at 550 nm [2], causing its characteristic color. However, an intrinsic property is something that does not scale with the amount of the material you have. If you double the amount of copper you have, you do not double the electrical conductivity { and if you double the amount of gold you have, you do not change its color. This is as contrasted with extrinsic properties such as volume or mass. If you double your gold, you double its volume and its mass. 1 This is an idea from classical thermodynamics. It is a good model for the bulk world. However, it does not really make sense in the atomic world. Consider solutions. Take a quantity of salt precursor with gold in it, such as chloroauric acid (HAuCl 4 ). Upon dissolving the chloroauric acid salt, the solution is orange instead of the shiny yellow color of gold. In a solution, you have suspended gold ions drifting around in water { so you are essentially viewing the color of the gold atom itself. If there is a color dierence between atomic gold and bulk gold, then there must be a point where the color transitions. If the color of gold is an intrinsic property of gold, how can that be? This transition region is known as the nanoscale region, and it is the range of material sizes in which the material possesses intrinsic properties intermediate or unique from the bulk and atomic material properties. The nanoscale region can also have some unique phenomena appear, such as plasmonic resonance (detailed in section 2.1). The nanoscale region has some truly counterintuitive behavior. In the above example with gold, I mentioned that no matter how many times you double your quantity of gold, you still end up with the same color of gold { but what about halving it? The remarkable nature of the nanoscale region is that if you repeatedly halve the quantity of your gold, it will eventually turn red. In fact, it will not just turn red, but it will undergo a series of color changes along the way. The color of gold nanoparticles has a well-documented size dependence [3]: large nanoparticles are redder, and small nanoparticles are bluer [4]. In fact, it is not just the size dependence that matters { the shape of the particles also exerts some in uence over the color of the material. There is a full spectrum of color tunability between blue and red for gold nanoparticles of varying size and shape [5]. This concept is actually a recurring theme with nanoparticle systems: an intrinsic property suddenly becomes not so intrinsic at a critical size scale between bulk and atomic, and it exhibits some tunability along the way. But why does this happen in gold? Metallic nanoparticles exhibit a phenomena known as plasmonic resonance in the nanoscale region. Plasmonic resonance is the primary driver of their optical absorbance spectrum, 2 which in turn determines the perceived color of the particles. Normally, bulk metal particles are quite re ective and have no intrinsic absorbance in the visible spectrum { gold has some ability to absorb blue-green wavelengths, which is why its color is yellow, but most metals are re ective of all colors. Plasmonic resonance adds an absorptive energy band to the parti- cles' spectrum, and the position and width of that energy band depends on the size and the shape of the particles in the nanoscale region. Plasmonic resonance theory will be discussed in detail in section 2.1. On the other end of the crystalline materials family, I studied ZnO. Oxides cannot ex- hibit plasmonic resonance due to their insulating nature, but instead have useful emissive properties. 1.2 ZnO nanostructures Zinc oxide is a direct, wide band gap semiconductor with a high exciton binding energy [6]. Consequently, it is a highly desirable material for use for optoelectronic applications in the ultraviolet-blue portion of the spectrum, having absorption and emission peaks centered around 350-400 nm [7]. Another useful property is that the optical emissions are quite stable at room temperature compared to many other wide band gap semiconductors [8{10]. This is due to the band gap emission and absorption being caused by excitonic excitation and recombination, rather than free electron/hole excitation and recombination. This combina- tion of properties makes ZnO a very attractive material for optoelectronic applications in the ultraviolet-blue regime [11]. The main advantage of using ZnO in the scope of my graduate work is that it is a biocompatible [12{14], uorescent ultraviolet-blue emitter. This has applications in many domains, but increasingly in biotechnology, medicine, and communications [12]. For each of these applications, it is useful to have the ZnO in nanoparticle form rather than bulk form, as ZnO nanoparticles have enhanced uorescent properties and unique tunability [15]. Some examples of technologies that utilize ZnO nanoparticles include uorescent dyes [16], cancer 3 therapeutics [17], and optical diodes [18]. The synergy of gold and ZnO nanostructures is something I studied at great length in my PhD, and material backgrounds are given in chapter 2. Growing these two heterogeneous materials together in a stable nanoparticle form has signicant advantages over either of them individually. 1.3 AuZnO core-shell Both gold nanoparticles and ZnO nanoparticles have interesting properties and useful ap- plications, and even some synergies [19, 20]. In the bigger picture of nanomaterials science, creating, understanding, and utilizing homogenous nanoparticles like gold or ZnO is just the rst step towards a larger goal of functional nano systems, or even eventually nanomachines. Modern research is increasingly focused on building more complex nanomaterials than the simple homogeneous particles, and the rst step towards increasing that complexity is a two- material nanoparticle system [21{23]. One way of doing this is creating what is known as a core-shell nanoparticle. Core-shell nanoparticles are spherical nanoscale particles composed of two primary materials; a metal core surrounded by a semiconductor or insulator [24]. Figure 1.1: Left: TEM image of four AuZnO core-shell nanoparticles. Due to the higher density of the core, it appears darker, showing the contrast between the core and the shell layers. The ZnO shell protects the gold core and adds the possibility for tuning via material modication of the shell explored in chapter 3. Often times, the core-shell nanoparticle is greater than the sum of its parts. This is be- 4 cause there can be unique interactions between the two materials, as there is a junction inter- face between the core and the shell. This can lead to similar optoelectronic properties seen in e.g., traditional metal-semiconductor (such as in g. 1.1) or semiconductor-semiconductor interfaces, but with the advantages of being tunable, since it is a nanoparticle-scale junction. In addition to the tunability, these nanoparticles also have a degree of freedom in that they are isolated nanoparticles, meaning that their position and collective eects can be signif- icant. For example, an aggregate of core-shell nanoparticles can have dierent properties than a single core-shell nanoparticle [25, 26]. This enhances the tunability of the material and is an example of how more complex nanostructures can lead to interesting and useful technologies. The AuZnO structure is discussed in detail in chapter 3. 1.4 Core-shell quantum dots At this point, it may seem as though unique nanoscale behavior of materials is limited purely to metallic elements and maybe even metallic compounds. However, this is not true. Unique nanoscale behavior of materials can be ascribed to many types of materials. In the scope of this thesis is also two dierent classes of semiconductor nanomaterials: \zero dimensional" semiconductor quantum dots and \one dimensional" semiconductor nanowires. Semiconductor quantum dots exhibit a nanoscale region behavior in their optoelectronic properties [27]. In bulk, semiconductor materials absorb and emit light, receive a current and emit light, or receive light and emit a current. This is due to charge carrier recombination or generation within the material, and it is a fundamentally quantum process. A full description of carrier recombination and generation is outside the scope of this thesis, so assume for a moment that semiconductors can interact with light. As a simplied model, in bulk semiconductors, the main property that determines the interaction with light is called the band gap. It can be a direct band gap or an indirect band gap, and its value determines the wavelength of light that is absorbed or emitted. In nanoscale semiconductors, the band gap value is in uenced by the size of the parti- 5 Figure 1.2: Progressive quenching of uorescent emission in a type of core-shell quantum dot. This thermal quenching eect is a unique nanoscale behavior to this system, which I explore fully in chapter 4. cle. This is due in part to the in uence of surface states. Surface states have \dangling bonds" which are higher in energy than interior states [27]. When the particle is small, the contribution of these states to the particle's electronic band structure is larger, and so the energy dierence between the valence band and the conduction band of the particle is increased. This means that the particle interacts with higher energy light, which is perceived as emission or absorption of bluer light than for larger particles. The shape of the material also plays a role here for semiconductors, as it does for metallic nanoparticles. There is the idea of the `dimensionality' of nanoparticles: 0d materials are called quantum dots, 1d materials are called nanowires, 2d materials are called thin lms. Of course, no material is truly anything less than 3d. This term is used informally for materials in which all, two, or one of its length scales, respectively, are below some critical size limit. This size limit depends on the context of the phenomena you are trying to observe. In the context of surface plasmonic resonance, that size limit is when the particle's diameter is much less than the incident wavelength of light. As such, the physical mechanisms of emission can be quite complex in semiconductor quantum dots, exhibiting unique behavior such as thermal quenching of emission [28{30] (displayed visually in g. 1.2). Quantum dots are introduced in chapter 2, and then I investigate this phenomenon fully in ZAIS quantum dots in chapter 4, and propose future 6 works relating to applications in LED devices and biological imaging agents in chapter 5. Much work during my PhD was focused on synthesizing metallic nanoparticles, semicon- ductor quantum dots, and semiconductor nanowires for optical applications. In summary, chapter 2 dives into the necessary materials background for the nanosystems that are rele- vant to this these. In chapter 3, I explore the rst of the core-shell nanoparticles I studied, the metal-insulator AuZnO system. In chapter 4, I provide motivation, background, and results on my work on the QHMF quantum dot system I studied. In chapter 5, I introduce two additional side projects I pursued which are more on the applied materials engineering side of my PhD work. 7 References 1 D. R. Lide, CRC handbook of chemistry and physics : a ready-reference book of chemical and physical data, eng (Boca Raton : CRC Press, 2004). 2 K. Shanks and T. Mallick, \Optics for concentrating photovoltaics: Trends, limits and opportunities for materials and design", Renewable and Sustainable Energy Reviews 60, 394{407 (2016). 3 Gold Nanoparticles: Properties and Applications, en. 4 R. Sardar, A. M. Funston, P. Mulvaney, and R. W. Murray, \Gold Nanoparticles: Past, Present, and Future ", en, Langmuir 25, 13840{13851 (2009). 5 J. Perezjuste, I. Pastorizasantos, L. Lizmarzan, and P. Mulvaney, \Gold nanorods: Synthesis, characterization and applications", en, Coordination Chemistry Reviews 249, 1870{1901 (2005). 6 A. Janotti and C. G. V. d. Walle, \Fundamentals of zinc oxide as a semiconductor", en, Reports on Progress in Physics 72, Publisher: IOP Publishing, 126501 (2009). 7 L. Saikia, D. Bhuyan, M. Saikia, B. Malakar, D. Dutta, and P. Sengupta, \Photocatalytic performance of ZnO nanomaterials for self sensitized degradation of Malachite Green dye under solar light", Applied Catalysis A: General 490, 10.1016/j.apcata.2014.10.053 (2014). 8 Z. P. Wei, Y. M. Lu, D. Z. Shen, Z. Z. Zhang, B. Yao, B. H. Li, J. Y. Zhang, D. X. Zhao, X. W. Fan, and Z. K. Tang, \Room temperature p-n ZnO blue-violet light-emitting diodes", Applied Physics Letters 90, Publisher: American Institute of Physics, 042113 (2007). 9 H.-M. Xiong, D.-P. Liu, Y.-Y. Xia, and J.-S. Chen, \Polyether-Grafted ZnO Nanoparticles with Tunable and Stable Photoluminescence at Room Temperature", Chemistry of Materials 17, Publisher: American Chemical Society, 3062{3064 (2005). 8 10 A. Wei, X. W. Sun, C. X. Xu, Z. L. Dong, M. B. Yu, and W. Huang, \Stable eld emission from hydrothermally grown ZnO nanotubes", Applied Physics Letters 88, Publisher: American Institute of Physics, 213102 (2006). 11 L. E. Greene, M. Law, J. Goldberger, F. Kim, J. C. Johnson, Y. Zhang, R. J. Saykally, and P. Yang, \Low-Temperature Wafer-Scale Production of ZnO Nanowire Arrays", Angewandte Chemie 115, eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1002/ange.200351461, 3139{3142 (2003). 12 Z. Li, R. Yang, M. Yu, F. Bai, C. Li, and Z. L. Wang, \Cellular Level Biocompatibility and Biosafety of ZnO Nanowires", The Journal of Physical Chemistry C 112, Publisher: American Chemical Society, 20114{20117 (2008). 13 L.-H. Zhao, R. Zhang, J. Zhang, and S.-Q. Sun, \Synthesis and characterization of biocompatible ZnO nanoparticles", en, CrystEngComm 14, Publisher: Royal Society of Chemistry, 945{950 (2012). 14 C. Dagdeviren, S.-W. Hwang, Y. Su, S. Kim, H. Cheng, O. Gur, R. Haney, F. G. Omenetto, Y. Huang, and J. A. Rogers, \Transient, Biocompatible Electronics and Energy Harvesters Based on ZnO", Small 9, eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1002/smll.201300146, 3398{3404 (2013). 15 O. Lupan, T. Pauport e, B. Viana, and P. Aschehoug, \Electrodeposition of Cu-doped ZnO nanowire arrays and heterojunction formation with p-GaN for color tunable light emitting diode applications", en, Electrochimica Acta, Selected Papers from the 61st ISE Meeting, Nice, France, 2010 56, 10543{10549 (2011). 16 H. Wang, D. Wingett, M. H. Engelhard, K. Feris, K. M. Reddy, P. Turner, J. Layne, C. Hanley, J. Bell, D. Tenne, C. Wang, and A. Punnoose, \Fluorescent dye encapsulated ZnO particles with cell-specic toxicity for potential use in biomedical applications", en, Journal of Materials Science: Materials in Medicine 20, 11 (2008). 9 17 X. Yang, C. Zhang, A. Li, J. Wang, and X. Cai, \Red uorescent ZnO nanoparticle grafted with polyglycerol and conjugated RGD peptide as drug delivery vehicles for ecient target cancer therapy", en, Materials Science and Engineering: C 95, 104{113 (2019). 18 L. Wang, J. Lin, X. Liu, S. Cao, Y. Wang, J. Zhao, and B. Zou, \Mg-Doped ZnO Nanoparticle Films as the Interlayer between the ZnO Electron Transport Layer and InP Quantum Dot Layer for Light-Emitting Diodes", The Journal of Physical Chemistry C 124, Publisher: American Chemical Society, 8758{8765 (2020). 19 H. Liao, W. Wen, G. K. L. Wong, and G. Yang, \Optical nonlinearity of nanocrystalline Au/ZnO composite lms", EN, Optics Letters 28, Publisher: Optical Society of America, 1790{1792 (2003). 20 C. Ma, X. Wang, S. Zhan, X. Li, X. Liu, Y. Chai, R. Xing, and H. Liu, \Photocatalytic Activity of Monosized AuZnO Composite Nanoparticles", en, Applied Sciences 9, Number: 1 Publisher: Multidisciplinary Digital Publishing Institute, 111 (2019). 21 M. Niu, C. Pham-Huy, and H. He, \Core-shell nanoparticles coated with molecularly imprinted polymers: a review", en, Microchimica Acta 183, 2677{2695 (2016). 22 W. Sch artl, \Current directions in core{shell nanoparticle design", en, Nanoscale 2, Publisher: Royal Society of Chemistry, 829{843 (2010). 23 S. Thatai, P. Khurana, J. Boken, S. Prasad, and D. Kumar, \Nanoparticles and core{shell nanocomposite based new generation water remediation materials and analytical techniques: A review", en, Microchemical Journal 116, 62{76 (2014). 24 R. Ghosh Chaudhuri and S. Paria, \Core/Shell Nanoparticles: Classes, Properties, Synthesis Mechanisms, Characterization, and Applications", en, Chemical Reviews 112, 2373{2433 (2012). 10 25 S. Anantharaj and S. Kundu, \Enhanced Water Oxidation with Improved Stability by Aggregated RuO2-NaPO3 Core-shell Nanostructures in Acidic Medium", Current Nanoscience 13, 333{341 (2017). 26 C. Vogt, M. S. Toprak, M. Muhammed, S. Laurent, J.-L. Bridot, and R. N. M uller, \High quality and tuneable silica shell{magnetic core nanoparticles", en, Journal of Nanoparticle Research 12, 1137{1147 (2010). 27 H. S. Mansur, \Quantum dots and nanocomposites", en, WIREs Nanomedicine and Nanobiotechnology 2, eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1002/wnan.78, 113{129 (2010). 28 K. Gradkowski, N. Pavarelli, T. J. Ochalski, D. P. Williams, J. Tatebayashi, G. Huyet, E. P. O'Reilly, and D. L. Huaker, \Complex emission dynamics of type-II GaSb/GaAs quantum dots", Applied Physics Letters 95, Publisher: American Institute of Physics, 061102 (2009). 29 Y. Zhao, C. Riemersma, F. Pietra, R. Koole, C. de Mello Doneg a, and A. Meijerink, \High-Temperature Luminescence Quenching of Colloidal Quantum Dots", ACS Nano 6, Publisher: American Chemical Society, 9058{9067 (2012). 30 Y.-h. Wu, K. Arai, and T. Yao, \Temperature dependence of the photoluminescence of ZnSe/ZnS quantum-dot structures", Physical Review B 53, Publisher: American Physical Society, R10485{R10488 (1996). 11 Chapter 2 Materials Background In this chapter, background on ZnO and gold nanoparticles will be discussed. This infor- mation lays the groundwork for later chapters, including Au-ZnO core-shell nanoparticles, and generally core-shell quantum dots as well. The journey into optically active core-shell nanoparticles begins with gold and ZnO. 2.1 Plasmonic resonance theory As mentioned in section 1.1, gold has an interesting property that its color depends on the size of the particle, and something called "plasmonic resonance theory" is responsible for this behavior. What is plasmonic resonance theory? To start, we need to look at plasmonics. Plasmon- ics, in general, is the eld of study associated with the physics of plasmons. A plasmon is the collective movement of charge on the surface of a conductor, such as a metal particle. Because the electric eld inside the volume of a conducting material must be zero, charge on the surface of the conductor must rearrange itself in response to an externally applied electric eld. This means that the electrical charge on the surface will respond dynamically to a changing external electric eld on the surface of a conductor. It takes energy to excite the movement of charge on the surface of a conductor, so this process contributes to the measured absorption loss of the material unless the light is re-radiated by the moving charge at the same frequency and in the same direction. If the conductor is signicantly smaller than the wavelength of light, it will have a unique 12 Figure 2.1: Two examples of the tunable optical parameters for gold nanoparticles. (A) Gold nanorods synthesized with a shorter rod length, approximately 50 nm. (B) Gold nanorods synthesized with a longer rod length, estimated to be about 200-300 nm. and interesting response [1]. Consider a metal particle of radius a. Ifa<<L, where L is the wavelength of light, the particle will experience an eectively spatially-uniform electric eld at any instant of time. This will cause all of the free charge on the surface of the particle to accelerate to one side. Half a period later, the electric eld from the light wave will be pointing in the opposite direction, causing the charge to accelerate to the opposite side of the particle. If the geometry of the metal particle is right, then the charge oscillation will respond strongly to a specic frequency of light - a resonance will appear. This resonance leads to a peak in the absorbance spectrum of the material, changing its color dramatically. Since the resonance position and width depends on the size and shape of the particle, we get the observed color change of metal nanoparticles with respect to their bulk or atomic forms. An experimental example of this phenomena is shown in g. 2.1. The in uence of the geometry on the optical properties of the particles is signicant. In g. 2.1, gold nanorods are shown, rather than just nanospheres. Nanorods can be thought of as having two absorption cross sections due to their unique shape: a long axis absorption, and a short axis absorption. Due to the distinction between long and short axis resonant frequencies, the resulting color 13 of the particles can be manipulated by growing nanorods with dierent aspect ratios (length divided by width). In part (A) of g. 2.1, shorter nanorods have been produced, causing green and red absorption peaks, allowing only blue light to pass through. In part (B), the rods have been grown with a much longer length, pushing the resonance frequency along that axis such that they predominantly absorb in the infrared. The green absorption from the short axis is still present. An analytical model for more complicated geometries such as nanorods is quite dicult to pursue, and usually done numerically or experimentally. But much can be learned from the spherical nanoparticle case. Mathematically, surface plasmonic resonance arises from the Mie model for an electro- magnetic wave interacting with a spherical particle. The Mie model is a solution to Maxwell's equations for the scattering of light o a spherical particle. The solutions have three parame- ters known as the Mie eciencies. These eciencies are related to the extinction, absorption, and scattering rates [2]: Q extinct = 2 x 2 1 X n=0 (2n + 1)<(a n +b n ) (2.1) Q scatter = 2 x 2 1 X n=0 (2n + 1)<(ja n j +jb n j) (2.2) Q absorb =Q extinct Q scatter (2.3) Where a n and b n are coecients calculated used in a solution based on a mathematical expansion that depend on the exact geometry, refractive media, and optical properties of the specic situation [2]. The parameter x is a dimensionless size parameter that is proportional to the radius of the particle: x = 2nR 0 (2.4) Where n is the real refractive index of the medium, 0 is the wavelength of the light in vacuum, and R is the radius of the particle. 14 These cross-section eciencies can be quite complicated to model correctly if the geom- etry of the particle is non-spherical. However, for spherical particles, the rst terms in the sum dominate [2]: a 1 i2x 3 3 1 + 1 (2.5) b 1 i2x 3 3 1 + 1 (2.6) Where is the permittivity of the medium and is the permeability. These coecients can be placed in the sum, dropping all higher order terms, to arrive at a simple expression forQ absorb , which is related to the absorption curve that one would measure in a Ultraviolet- Visible Spectroscopy Instrument (UV-VIS). Again, for more complicated situations, this approximation does not hold and one would have to resort to numerical modeling or empirical knowledge. In complicated geometries, one may nd the peak location to be in dierent locations, or for there to be multiple peaks. 2.2 Gold NP synthesis & characterization In the course of my PhD, I focused on experimental materials science, and so my priority was to synthesize and measure the plasmonic resonance frequency of gold nanorods, rather than model and predict the plasmonic resonance frequency. I will now discuss the experimental side of gold NPs, including synthesis and characterization. Gold nanoparticles were synthesized using the method of Nikoobakht et al [3]. 50 mL of gold nanoparticles were prepared as follows: in a vial kept at 30 deg C, 25 mL of 1 mM HAuCl 4 (99.995%, Millipore Sigma) is added to 25 mL of 200 mM cetrimonium bromide (CTAB) (99%, VWR). 350 l of 78.8 mM ascorbic acid (99%, Millipore Sigma) is added to the growth solution, and the color changes from light yellow to clear. In a separate vial, also kept at 30 C, the seed solution is prepared: 1 mL of .5 mM HAuCl 4 is added to 1 mL of 15 200 mM CTAB. A 10 mM NaBH 4 (99.99%, Millipore Sigma) solution is prepared by adding the appropriate amount of NaBH 4 to ice-cold water. 120 l of the 10 mM NaBH 4 solution is added to the seed solution, and the color of the seed solution changes from light yellow to light brown. Immediately after, 60 l of the seed solution is transferred to the growth vial. The growth solution is thoroughly mixed by shaking and is allowed to sit at 30 C for up to half an hour, until the color changes from clear to pink. A typical example is shown in g. 2.1. The gold nanoparticles are puried via centrifugation. They are split into 2 mL Eppendorf tubes. The goal of the centrifugation process is to create a pellet of gold nanoparticles at the bottom of the Eppendorf tube, allowing for decanting of the supernatant, or excess solvent and reaction byproducts / unreacted precursors. The as-made gold nanoparticles are centrifuged twice at 21; 000g, replacing around 90% of the water each time. Due to the amount of CTAB used in the synthesis process, CTAB will precipitate as a solid if left at room temperature. Since the goal of the centrifugation is to remove excess chemicals like the CTAB, the centrifuge is preheated to 30 C before each use, and all centrifugation is done at 30 C. After centrifugation, the particles are ready for characterization. There are a few main characterization techniques that will provide a lot of important information about the nanoparticles. These techniques are Scanning Electron Microscope (SEM) or Transmis- sion Electron Microscope (TEM) imaging, UV-VIS, Fourier Transform Infrared Spectroscopy (FTIR), and Dynamic Light Scattering (DLS). The main information needed about a sample of gold nanoparticles relates to their size, shape, optical characteristics, and surface chem- istry. These parameters can eectively dene a batch of gold nanoparticles, and allow for use in other applications, such as the core for core-shell nanoparticles as discussed later. Sample preparation for UV-VIS and DLS is simple. The particles just need to be syn- thesized according to the procedure discussed previously, and then centrifuge washed with fresh water. Since UV-VIS spectroscopy and DLS are both aqueous-based characterization 16 techniques, it is more than sucient to stop there. About 500 l of sample volume is needed for UV-VIS, and 60 l is needed for DLS, allowing for sample aliquots from a much larger batch to be used. These volume requirements vary from instrument to instrument, but are related to the beam sizes and required optical path length through the cuvette. In order to get strong signal:noise, UV-VIS measurements require a 1 cm (approx.) path through the sample uid, as most of the optical intensity is lost to ltering (g. 2.2). DLS does not have this constraint as there is no chromatic ltering, and so a smaller sample volume may be used. UV-VIS provides information about the absorbance spectrum of the nanoparticles [4]. Specically, it returns data of intensity absorbed by the sample as a function of wavelength. A schematic of how the instrument is built is shown in g. 2.2 It does this by having a broadband light source, such as a Xe arc lamp, with a monochromator (wavelength-tunable bandpass) lter, allowing for excitation by only one discrete wavelength from the Xe arc lamp at a time. With the lamp intensity as a function of wavelength known, the ltered light is passed through the sample, and the intensity drop is measured to yield the absorbance. If Figure 2.2: Schematic of UV Vis, a common optical characterization technique for nanoparti- cles. UV Vis lters a broadband light source and records transmittance through the sample. This curve is often characteristic of intrinsic material properties. 17 the sample volume is xed, an exact absorbance value can be calculated, with dimensions of 1/length, providing information about how optically dense the sample is. DLS primarily provides quantitative information about the size distribution of scatter- ing site in the sample volume [5]. A schematic of how the instrument is built is shown in g. 2.3. For a solution containing gold nanoparticles, those scattering sites would be the gold nanoparticles themselves. The way DLS instruments can provide this information is through an indirect method involving several assumptions to be made about the sample and the sample medium (e.g. the solvent). DLS instruments shine a monochromatic light source through an aqueous sample, and measure the scattering pattern, or speckle pattern, produced. This speckle pattern is caused by scattering sites in the aqueous medium. Some- time later, the speckle pattern is measured again, and that pattern can be correlated to the rst one to determine how much it has changed. This process is repeated at discrete time intervals, and an autocorrelation function is produced. Knowing some information about the sample, such as the solvent, parameters such as particle size distributions can be extracted from the autocorrelation function. Figure 2.3: Schematic DLS, another common optical characterization technique for nanopar- ticles. DLS makes assumptions about the particles' shape and motion, and extrapolates size distributions based on the observed laser speckle patterns. 18 Sample preparation for SEM/TEM imaging and FTIR analysis is more involved. These techniques are not amenable to liquid samples, and thus a way to extract the particles from the solvent is needed. For our instruments, the best way to do this was to purify the particles via centrifuge-washing, as described above. After this step has been completed, the particles are deposited onto a cleaned silicon/silica wafer, and left to dry on a hot plate at 65 C. For the silicon/silica wafers, we cut them into small, 1 cm by 1 cm pieces and wash with a standard solvent wash procedure { acetone, then methanol, then isopropanol and air gun dry. If necessary, the wafer may also be cleaned by an O 2 plasma cleaner, which produces a light oxygen plasma to gently clean the surface of hydrocarbons. After deposition on the clean wafer chip, the samples are ready to be characterized by SEM and FTIR. SEM imaging provides the most direct information about the nanoparticles { images of them. Although sample preparation and then nding the particles can be dicult, since they are so small, the information obtained from an SEM is quite useful. DLS and UV-VIS can approximately tell you the size distribution of your particles, and some of their optical properties, but it relies on many assumptions about the shapes of the particles, usually assuming that they are spherical. SEM on the other hand, can provide shape information. If enough images are obtained, automated analysis can produce a statistical size distribution that can be compared to the size distribution information obtained from DLS. However, the image locations must be truly random and unbiased, which makes SEM size distributions only semi-quantitative. A sample of TEM and SEM images showing gold nanorods is presented in g. 2.4. TEM imaging provides similar information but can achieve higher magnication. The sample preparation is a little bit dierent for TEM imaging. TEM relies on electrons trans- mitted through the sample, rather than SEM which relies on electrons scattered o the surface. Consequently, the sample must be very thin and electron-transparent, which the aforementioned silicon/silica wafers are not. Standard TEM grids, such as copper grids with a thin carbon lm suspension, are used for TEM analysis in place of the silicon/silica wafer 19 Figure 2.4: Left: TEM image of a gold NR showing lattice fringes. This method of imaging conveys additional information about the particles that is not present in the SEM image. Right: SEM of mixture of gold nanospheres and nanorods. The geometric contrast is ap- parent. This manifests itself in very dierent scattering and extinction cross sections, which can be tuned to dierent applications. chip. Otherwise, all gold nanoparticle preparation is the same. The main advantage of TEM for gold nanoparticle characterization is its higher resolution, which can show information about the crystal lattice of the nanoparticles. Crystal structure details are far too small to be observed with visible light, but become apparent in TEM images. In g. 2.4, if can be seen that there are distinct lines within the nanoparticle, pointing along the short axis. These are characteristic of a particular crystal structure and material, and can help identify what exactly has been synthesized in your samples. Electron microscopy takes advantage of the wave nature of electrons. Optical light also has wave behavior, but cannot image the same scale of features. The smallest resolvable features in optical microscopy are limited by the diraction limit. For a monochromatic optical source, that limit is: d = 2n sin (2.7) Which is known as the Abbe diraction limit [6{8]. In this denition, d is the minimum resolvable distance, is the wavelength of the source,n is the refractive index of the medium, 20 and is the angle of the light converging to the sample. For optical wavelengths traveling through air, this is about 200 to 300 nm. This relationship is also true for electron imaging systems. Considering electrons, the electron's wavelength is related to its momentum by the de Broglie wavelength: = h p (2.8) Whereh is Planck's constant andp is the momentum of the electron. In SEM and TEM, the electrons are accelerated through a "column voltage", picking up speed (momentum) along the way. This decreases the minimum resolvable feature size [9, 10]. To demonstrate this, we can start with an equation for an electron's kinetic energy after traveling through potential V column : V column = p 2 2m (2.9) Substituting eq. (2.8): V column = (h=) 2 2m (2.10) And now solving for : = h p 2mV column (2.11) In TEM, a typical column voltage is 180 kV. Planck's constant is h = 4:136 10 15 eVs, and the mass of an electron is 511 keV/c 2 . Evaluating eq. (2.11), we get = 2:89 pm. Comparing this with an optical wavelength (400-700 nm), it is clear to see how the diraction limit dened in eq. (2.7) can be much, much, lower for an electron imaging system. This is what allows electron microscopy techniques to image nanoparticles and even atomic-scale structures, such as crystal lattice spacings, as seen in g. 2.4. 21 Combined, these characterization techniques can provide a lot of useful information about a prepared batch of gold nanoparticles. This information is critical to many applications, but also specically the next step of the process reported in this work, which is encasing the particles in a shell. 2.3 ZnO The other fundamental nanomaterial I studied over the course of my PhD was ZnO. ZnO and gold are optically synergistic and the combined core-shell nanomaterial, discussed in chapter 3, has some unique properties. But rst some background on ZnO itself is needed. 2.4 ZnO properties As mentioned in section 1.2, the wide band gap makes ZnO (traditionally thought of as an insulator) as an optically active material when considering the ultraviolet-blue portion of the electromagnetic spectrum. This band gap is a direct band gap [11], meaning that the fundamental eciency of ZnO as a semiconductor material will be much higher, as a band gap transition can be activated by a photon of the right energy rather than also needing a change in momentum (provided by temperature). The band gap is 3.7 eV which is approximately 365 nm, right past the edge of the visible range [11]. The width of the band gap also provides opportunities for lower energy transitions related to defects to be in the visible range, which has leveraged for various applications. The unique property of the ZnO as a wide band gap semiconductor is specically its high exciton binding energy (60 meV) [11]. An exciton is the bound state of an electron-hole pair. Similar to quantum models for the hydrogen atom, it has discrete energy levels that are caused by the electric interaction between the electron in the conduction band, and the hole it left behind in the valence band. If enough energy is provided, this interaction is destroyed and thus the recombination is not as likely to occur at the same energy, causing a spread of emission rather than discrete lines in the optical spectrum. Latent energy provided by 22 room temperature is about 20 meV, which is lower than the binding energy of excitons in ZnO, allowing for their stability and consequent recombination. This means that the ZnO emission lines are narrow and exist at room temperature [11]. In nanoparticle form, this emission is tunable based on the quantum connement eect. This eect is not unique to ZnO and is in fact a property of all semiconductor materials synthesized on the nanoscale. The eect is simple but useful. As the material becomes smaller, its band gap will increase. The absorption and emission wavelengths related to band gap transitions become blue shifted, and ultimately the optical properties of the material become tunable if the size is tunable. This is because as the particle becomes smaller, the uncertainty principle broadens the energy state dispersion within the material, similar to the particle in a box model from quantum mechanics. ZnO is susceptible to this eect and the band gap emission/absorption wavelength can be adjusted by controlling the size of the particles synthesized [12]. The question, then, is how to synthesize the particles. 2.5 ZnO characterization ZnO is analyzed in many similar ways as gold nanoparticles, including SEM (g. 2.6) and TEM (chapter 3). But crystal structure isn't everything when it comes to characterization of nanoparticles. Surface chemistry is also important, and one of the best ways to investigate the surface chemistry of some nanoparticles is FTIR. FTIR analysis can provide the needed information about the surface chemistry of nanoparticles. This is particularly important for applications such as gas sensing [13], where the surface chemistry may play a functional role in the ZnO sensor material. FTIR probes the surface chemistry by using infrared light in the range of 400 cm 1 to 4,000 cm 1 . This corresponds to a wavelength range of 2:5 m to 25 m. Phonon modes of chemical bonds are active in this range, such as bending and stretching modes, which have low, but unique energy levels. This means that light with the appropriate wavelength or energy can be used to activate these phonon modes, exciting vibrations in the molecules 23 which will cause absorption peaks in the infrared spectrum. FTIR specically probes the surface chemistry of the material by evanescently coupling infrared light into the material, ensuring that the penetration depth of the light will not be too deep into the sample. An example of freshly characterized ZnO nanospheres I prepared were characterized by FTIR and presented in g. 2.5. Figure 2.5: FTIR spectra of ZnO, conrming successful synthesis but also indicating some water adsorption had occurred. Surface chemistry impact such as water adsorption can be a critical problem for some applications of ZnO nanowires. 2.6 ZnO defects and dopants The optoelectronic properties of ZnO as a semiconductor can be greatly in uenced by defects and dopants in the crystal structure. While band gap recombination is the most common type of emission seen in pure semiconductors, defects in the crystal lattice can introduce new states which electrons and holes can populate in traditionally forbidden regions, such as the band gap. Though there are many types of crystalline defects, the kind that will produce electronic transitions are usually vacancies or dopant atoms being introduced into the struc- ture. This occurs in many types of semiconductors and can be leveraged to manipulate the optoelectronic properties. An example of this is n- or p- type doping of silicon. Silicon is an intrinsic semiconductor, but with enough n- or p- type dopants (like phosphorous or boron, 24 respectively) introduced into the silicon lattice, the primary charge carrier in the material becomes electrons or holes, respectively [14]. Defect control can also be leveraged in ZnO to control its optical properties [15]. Although ZnO's band gap causes it to emit primarily in the ultraviolet, visible range emission has been commonly reported. It is most often green emission, and although the source of this emission is debatable, it has been reported to occur primarily in ZnO samples with a large concentration of crystal defects [16], such as vacancies. This is an intrinsic type of emission that can be achieved when a defect state is introduced into the formerly empty band gap of ZnO. This defect state can allow for transitions of electrons from the valence band to the defect band, and back down, causing emission of a photon at lower energy than the band gap. Similarly, foreign atoms can be introduced into the ZnO lattice with little structural impact. At low concentrations, these foreign atoms are known as dopants. Dopants do not have to be externally introduced; common intrinsic dopants to ZnO include Cu [11] and can be found naturally in unpuried ZnO. Doping is a potential strategy for increasing the tunability of a material, as the properties of ZnO and the dopant atom can synergize to create a new functional material. Doped ZnO nanoparticles have also been reported, and so this strategy can be applied to nanomaterial ZnO as well. 2.7 Types of ZnO nanostructures ZnO nanoparticles can also come in dierent shapes. Dierent shapes and morphologies of ZnO nanoparticles have dierent properties and applications. The primary types of ZnO nanoparticles that have been reported are nanowires, nanotetrapods, and nanospheres [17{ 19]. Nanomaterials are interesting and useful because geometry of the material becomes a variable, or degree of freedom, in determining the functional properties of a given material. For example, given a chunk of bulk ZnO, it does not really matter what shape it is cut into { a cube, or a sphere, the intrinsic material properties like conductivity, uorescence, etc. 25 Figure 2.6: Wurtzite crystal structure ZnO and its impact on ZnO nanoparticle geometry. The hexagonal symmetry is clearly seen under an SEM image. This structure has in uence over its optical properties. SEM image my own work; crystal structure diagram reproduced from [20] with modication (public domain). will all be the same. But for nanomaterial ZnO, the geometry matters quite signicantly { ZnO quantum dots can have radically dierent uorescent emission than ZnO nanowires [18]. Thus, it is important to understand the dierent types of ZnO that can be synthesized, and their applications. The rst type of ZnO nanoparticle to discuss is the ZnO nanowire. ZnO nanowires are often synthesized on a at substrate in a giant array. These nanowires are usually grown to be about 1 m in length, and 50-200 nm in diameter. Due to the wurtzite nature of the ZnO crystal structure, they also have hexagonal symmetry about the vertical axis. This is shown in g. 2.6, which presents a schematic of the wurtzite structure, and also an image of ZnO nanowires. ZnO nanowires have a characteristic hexagonal shape about their vertical axis, which is a consequence of the hexagonal symmetry in the wurtzite crystal structure. ZnO nanowire research was a very active research eld in the last 20 years, with many studies performed to understand synthetic techniques and material properties. Due to their ease of synthesis, they have found many applications in sensing and lasing. One such example is random lasing [20]. Random lasing is a type of lasing based on light scattering through a highly disordered medium, rather than using a re ective cavity of some kind, to create the long optical path length [21{24]. Gain is achieved by pumping a population inversion in the ZnO material itself. Under the right conditions, a ZnO nanowire array can emit 26 Figure 2.7: Schematic of simple LED architecture in which QHMF quantum dots play a role as the active layer, causing emission of the red, green, and blue light necessary to form an image on a display. This device leverages the wide spectral tunability of QHMF quantum dots permitted by the exible compositional freedom. ultraviolet laser light [25, 26], creating a compact, integrated ultraviolet laser system that was a major novelty when it was reported [27, 28]. This would not be possible without the highly disordered medium created when ZnO is synthesized in nanowire form. Attempts at creating a UV random laser were done in my rst year of graduate school, and will be discussed in chapter 5. 2.8 Core-shell quantum dots Another material platform I pursued was a type of semiconductor quantum dot based on a core-shell nanostructure. These are a bit dierent from the metal-insulator combination of gold and ZnO, but the overall structure and challenges in synthesis were quite similar. I specically focused on a class of semiconductor quantum dots known as QHMF quantum dots, which have several interesting properties for practical technologies such as LED display devices (g. 2.7), biomedical imaging agents, or many other applications [29, 30]. The moti- vation to look at semiconductor quantum dots in contrast with the metal-insulator systems I had been studying (chapter 3) were driven by the need for a more emissive nanoparticle core. 27 The traditional quantum dot used as an emissive material is cadmium-based. Usually, these are CdS or CdSe cores, surrounded by a thin ZnS shell layer [31]. Cd-based quantum dots are relatively straightforward to synthesize, and have best-in-class performance for power eciency [32], color saturation [33], and long lifetimes [34]. They have been extensively studied for over 30 years, and at this point, are a quite sophisticated technology with several consumer product applications. However, their mainstream adoption is still quite niche. This is because one of the main hurdles is the usage of cadmium. Cadmium is an extremely toxic metal, and its usage has signicant health and environmental risks. The recyclability of functional materials made using Cd-based quantum dots is poor, and they would require extensive care as a waste product. This poses not only a signicant moral concern for the use of Cd-based quantum dot technologies, but a regulatory concern as well, which hinders the adoption of quantum dot materials and technologies, despite their superior performance [35]. There are a few main competing material technologies. These include perovskites, In- based quantum dots, and QHMF materials [36]. Perovskite materials suer from the same problem as Cd-based quantum dots, as they usually involve Pb [37], which also has a signif- icant toxicity and thus regulatory problem. In-based quantum dots are the group III analog to Cd-based quantum dots, but although signicant work has been poured into optimizing them, they still have poorer optical properties and a more dicult synthesis than Cd-based quantum dots [38]. QHMF quantum dot materials are an exciting new alternative mate- rial with suciently competitive optical properties, and importantly, are free of toxic heavy metals. 2.9 QHMF quantum dots QHMF quantum dot emitters are a QHMF-composition nanoparticle structure alloyed with a fourth element, Zn. Due to the incorporation of four elements, these structures have a much wider spectral tunability, and enhanced emissive properties compared to In-based 28 Figure 2.8: Synthesized QHMF quantum dot samples characterized for elemental composi- tion via x-ray photoelectron spectroscopy. The spectra conrm a relatively uniform batch of QHMF quantum dots formed. A diagram of the core-shell structure is shown inset. quantum dots [35]. Furthermore, this is a class of materials which spans the visible range, and its emissive properties are tunable by both composition and particle size, among other factors, allowing for a high degree of control in theory. Its main drawbacks compared to other quantum dot LED emitter materials are in its broader spectral output, leading to poorer color saturation; its lower quantum yield, although still best in class among heavy metal free quantum dots [35]; and large photon lifetime, causing eciency loss due to competing mech- anisms like Auger recombination [39]. Nonetheless, there is a lot of room for improvement in QHMF quantum dot materials development, both in synthesis, post-processing, and device architecture. The emissive properties of QHMF quantum dots are unique compared to other quantum 29 dot materials, and there is a lot of room for improvement. They exhibit a wide Stokes shift, larger photoluminescence bandwidth, and lower yield than Cd-based quantum dots [35]. There are many factors involved in this, but one that has been studied extensively recently is the recombination mechanism. In traditional II-VI or III-V semiconductor quantum dots, such as CdS or InP, the dominant recombination mechanism is near-band-edge excitonic recombination [40], leading to a fast and predictable emission output. However, the crystal structure and the composition of QHMF quantum dots, especially when doped or alloyed with a fourth element, is much more complicated (as shown in g. 2.8), leading to a plethora of defect states in the middle of the band gap [40]. These defect states allow for more complicated transitions and are enhance the Stokes shift of the emission. However, they also provide a signicant opportunity for control of the emissive properties via defect engineering. This is explored in chapter 5. 2.10 Conclusion Gold nanoparticles are an important metal nanoparticle for their unique optical properties, chemical stability, and ease of synthesis. The main synthetic technique for creating gold nanoparticles was discussed in detail, and the ways to fully characterize a batch of gold nanoparticles including imaging and spectroscopic analysis was discussed. Additionally, a general review of ZnO was given, including basic properties, the importance of defects and dopants in its crystal structure, and the varying types of nanostructures that can be formed, along with their technological relevance. ZnO is an interesting wide band gap semiconductor material, with many favorable properties. It has relatively sophisticated and straightforward synthetic strategies, and many dierent types of ZnO nanostructures can be created. ZnO and its nanostructure forms are crucial to understand and form the basis for other types of core-shell nanoparticles explored in later chapters, such as semiconductor quantum dots and their applications. 30 References 1 T. Wriedt, \Mie Theory: A Review", in The Mie Theory: Basics and Applications, edited by W. Hergert and T. Wriedt (Springer Berlin Heidelberg, Berlin, Heidelberg, 2012), pp. 53{71. 2 J. Duque, J. Bland on, and H. Riascos, \Localized Plasmon resonance in metal nanoparticles using Mie theory", en, Journal of Physics: Conference Series 850, 012017 (2017). 3 B. Nikoobakht and M. A. El-Sayed, \Preparation and Growth Mechanism of Gold Nanorods (NRs) Using Seed-Mediated Growth Method", en, Chemistry of Materials 15, 1957{1962 (2003). 4 W. Haiss, N. T. K. Thanh, J. Aveyard, and D. G. Fernig, \Determination of Size and Concentration of Gold Nanoparticles from UV{Vis Spectra", en, Analytical Chemistry 79, 4215{4221 (2007). 5 T. G. F. Souza, V. S. T. Ciminelli, and N. D. S. Mohallem, \A comparison of TEM and DLS methods to characterize size distribution of ceramic nanoparticles", en, Journal of Physics: Conference Series 733, 012039 (2016). 6 D. R. Mason, M. V. Jouravlev, and K. S. Kim, \Enhanced resolution beyond the Abbe diraction limit with wavelength-scale solid immersion lenses", EN, Optics Letters 35, Publisher: Optical Society of America, 2007{2009 (2010). 7 D. Gatto Monticone, K. Katamadze, P. Traina, E. Moreva, J. Forneris, I. Ruo-Berchera, P. Olivero, I. P. Degiovanni, G. Brida, and M. Genovese, \Beating the Abbe Diraction Limit in Confocal Microscopy via Nonclassical Photon Statistics", Physical Review Letters 113, Publisher: American Physical Society, 143602 (2014). 8 C. W. McCutchen, \Superresolution in Microscopy and the Abbe Resolution Limit", EN, JOSA 57, Publisher: Optical Society of America, 1190{1192 (1967). 31 9 What Is an Electron Microscope (EM) and How Does It Work? - VHA Diagnostic Electron Microscopy Program, en, General Information, Accepted: 20170801. 10 Scanning Electron Microscopy, en-US. 11 A. Janotti and C. G. V. d. Walle, \Fundamentals of zinc oxide as a semiconductor", en, Reports on Progress in Physics 72, Publisher: IOP Publishing, 126501 (2009). 12 X. Liu, X. Xing, Y. Li, N. Chen, I. Djerdj, and Y. Wang, \Controllable synthesis and change of emission color from green to orange of ZnO quantum dots using dierent solvents", en, New Journal of Chemistry 39, Publisher: The Royal Society of Chemistry, 2881{2888 (2015). 13 M. .-. Ahn, K. .-. Park, J. .-. Heo, D. .-. Kim, K. J. Choi, and J. .-. Park, \On-chip fabrication of ZnO-nanowire gas sensor with high gas sensitivity", en, Sensors and Actuators B: Chemical 138, 168{173 (2009). 14 J. Wagner and J. A. del Alamo, \Band-gap narrowing in heavily doped silicon: A comparison of optical and electrical data", en, Journal of Applied Physics 63, 425{429 (1988). 15 C. B. Ong, L. Y. Ng, and A. W. Mohammad, \A review of ZnO nanoparticles as solar photocatalysts: Synthesis, mechanisms and applications", en, Renewable and Sustainable Energy Reviews 81, 536{551 (2018). 16 Y. Gong, T. Andelman, G. F. Neumark, S. O'Brien, and I. L. Kuskovsky, \Origin of defect-related green emission from ZnO nanoparticles: eect of surface modication", en, Nanoscale Research Letters 2, 297 (2007). 17 N. Zhang, R. Yi, R. Shi, G. Gao, G. Chen, and X. Liu, \Novel rose-like ZnO nano owers synthesized by chemical vapor deposition", en, Materials Letters 63, 496{499 (2009). 18 V. M. Diep and A. M. Armani, \Flexible Light-Emitting Nanocomposite Based on ZnO Nanotetrapods", Nano Letters 16, Publisher: American Chemical Society, 7389{7393 (2016). 32 19 Z. Zaidi, S. I. Siddiqui, B. Fatima, and S. A. Chaudhry, \Synthesis of ZnO nanospheres for water treatment through adsorption and photocatalytic degradation: Modelling and process optimization", en, Materials Research Bulletin 120, 110584 (2019). 20 J. Fallert, R. J. B. Dietz, M. Hauser, F. Stelzl, C. Klingshirn, and H. Kalt, \Random lasing in ZnO nanocrystals", en, Journal of Luminescence, Special Issue based on The 15th International Conference on Luminescence and Optical Spectroscopy of Condensed Matter (ICL'08) 129, 1685{1688 (2009). 21 R. C. Polson and Z. V. Vardeny, \Random lasing in human tissues", Applied Physics Letters 85, Publisher: American Institute of Physics, 1289{1291 (2004). 22 X. Wu, W. Fang, A. Yamilov, A. A. Chabanov, A. A. Asatryan, L. C. Botten, and H. Cao, \Random lasing in weakly scattering systems", Physical Review A 74, Publisher: American Physical Society, 053812 (2006). 23 H. Cao, \Review on latest developments in random lasers with coherent feedback", en, Journal of Physics A: Mathematical and General 38, Publisher: IOP Publishing, 10497{10535 (2005). 24 H. Cao, Y. G. Zhao, S. T. Ho, E. W. Seelig, Q. H. Wang, and R. P. H. Chang, \Random Laser Action in Semiconductor Powder", Physical Review Letters 82, Publisher: American Physical Society, 2278{2281 (1999). 25 H.-C. Hsu, C.-Y. Wu, and W.-F. Hsieh, \Stimulated emission and lasing of random-growth oriented ZnO nanowires", Journal of Applied Physics 97, Publisher: American Institute of Physics, 064315 (2005). 26 S. B. Bashar, M. Suja, M. Morshed, F. Gao, and J. Liu, \An Sb-doped p-type ZnO nanowire based random laser diode", en, Nanotechnology 27, Publisher: IOP Publishing, 065204 (2016). 33 27 J. Huang, M. Monzur Morshed, Z. Zuo, and J. Liu, \Distributed Bragg re ector assisted low-threshold ZnO nanowire random laser diode", Applied Physics Letters 104, Publisher: American Institute of Physics, 131107 (2014). 28 L. Miao, S. Tanemura, H. Yang, and S. Lau, \Synthesis and random laser application of ZnO nano-walls: a review", International Journal of Nanotechnology 6, Publisher: Inderscience Publishers, 723{734 (2009). 29 A. Kiraz, M. Atat ure, and A. Imamo glu, \Quantum-dot single-photon sources: Prospects for applications in linear optics quantum-information processing", Physical Review A 69, Publisher: American Physical Society, 032305 (2004). 30 I. L. Medintz and H. Mattoussi, \Quantum dot -based resonance energy transfer and its growing application in biology", en, Physical Chemistry Chemical Physics 11, Publisher: Royal Society of Chemistry, 17{45 (2009). 31 Y. Zhang, C. Xie, H. Su, J. Liu, S. Pickering, Y. Wang, W. W. Yu, J. Wang, Y. Wang, J.-i. Hahm, N. Dellas, S. E. Mohney, and J. Xu, \Employing Heavy Metal-Free Colloidal Quantum Dots in Solution-Processed White Light-Emitting Diodes", en, Nano Letters 11, 329{332 (2011). 32 Y. Shirasaki, G. J. Supran, M. G. Bawendi, and V. Bulovi c, \Emergence of colloidal quantum-dot light-emitting technologies", en, Nature Photonics 7, 13{23 (2013). 33 B. S. Mashford, M. Stevenson, Z. Popovic, C. Hamilton, Z. Zhou, C. Breen, J. Steckel, V. Bulovic, M. Bawendi, S. Coe-Sullivan, and P. T. Kazlas, \High-eciency quantum-dot light-emitting devices with enhanced charge injection", en, Nature Photonics 7, 407{412 (2013). 34 V. Wood and V. Bulovi c, \Colloidal quantum dot light-emitting devices", en, Nano Reviews 1, 5202 (2010). 34 35 B. Chen, N. Pradhan, and H. Zhong, \From Large-Scale Synthesis to Lighting Device Applications of Ternary I{III{VI Semiconductor Nanocrystals: Inspiring Greener Material Emitters", en, The Journal of Physical Chemistry Letters 9, 435{445 (2018). 36 S. Wepfer, J. Frohleiks, A.-R. Hong, H. S. Jang, G. Bacher, and E. Nannen, \Solution-Processed CuInS 2 -Based White QD-LEDs with Mixed Active Layer Architecture", en, ACS Applied Materials & Interfaces 9, 11224{11230 (2017). 37 L. Protesescu, S. Yakunin, M. I. Bodnarchuk, F. Krieg, R. Caputo, C. H. Hendon, R. X. Yang, A. Walsh, and M. V. Kovalenko, \Nanocrystals of Cesium Lead Halide Perovskites (CsPbX3, X = Cl, Br, and I): Novel Optoelectronic Materials Showing Bright Emission with Wide Color Gamut", Nano Letters 15, Publisher: American Chemical Society, 3692{3696 (2015). 38 S. Tamang, C. Lincheneau, Y. Hermans, S. Jeong, and P. Reiss, \Chemistry of InP Nanocrystal Syntheses", en, Chemistry of Materials 28, 2491{2506 (2016). 39 P. O. Anikeeva, C. F. Madigan, J. E. Halpert, M. G. Bawendi, and V. Bulovi c, \Electronic and excitonic processes in light-emitting devices based on organic materials and colloidal quantum dots", Physical Review B 78, Publisher: American Physical Society, 085434 (2008). 40 Y. E. Panl, M. Oded, and U. Banin, \Colloidal Quantum Nanostructures: Emerging Materials for Display Applications", en, Angewandte Chemie International Edition 57, eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1002/anie.201708510, 4274{4295 (2018). 35 Chapter 3 Core-shell nanoparticles Core-shell nanomaterials combine the disparate material properties of two nanomaterials into one functional system. In my PhD, I studied AuZnO nanoparticles extensively, which are gold nanospheres with a thick and porous ZnO shell. The motivation for such particles is that you can combine the plasmonic resonance behavior of gold nanoparticles with the chemical stability of ZnO. The large porous shells also oer some possibility for functional modication via the introduction of chemical dopants, and potentially payload delivery as well. However, there can also be many diculties in creating core-shell nanoparticles compared to homogenous nanoparticles. The system is fundamentally more complicated, and there- fore the synthesis is also. Common diculties faced in synthesizing core-shell nanoparticles include incompatible surface chemistries between the two materials, lattice mismatches caus- ing defect buildup in the shell material [1], and dierences in processing techniques. Our processes are aiming to overcome these diculties in core-shell AuZnO in a unique way, using a new synthesis technique [2]. In this chapter, we will discuss motivation for creating core-shell AuZnO particles, and then synthesis, and characterization. 3.1 Background: synthetic challenges The problems associated with core-shell nanoparticle synthesis are primarily in dierences between the core material and the shell material. If the lattices of each material are sig- nicantly dierent, there can be strain induced at the interface between the two materials. 36 Figure 3.1: A diagram illustrating the impact of lattice compatibility at an interface between two dierent crystalline materials. If the lattice is well-matched in the lateral direction, minimal strain occurs. Otherwise, strain and defects such as dislocations can occur. This is represented by a parameter known as the lattice mismatch, which can be dened as the ratio between key lattice parameters for two dierent crystalline materials. If the lattice parameters for the core and the shell materials are too dierent, then there will be strain at the interface [1]. This causes defects to build up, such as vacancies in either of the lattice, which can mobilize and combine to form dislocations. These dislocations can cause cracks in the shell material, and ultimately destabilize the whole system, or otherwise prevent encapsulation of the core material by the shell material. For a at surface, the lattice mismatch parameter is proportional to the dierence between lateral lattice parameters [3]: f m = 1 2 2 (3.1) Where 1 and 2 are the lattice parameters of materials 1 and 2 in g. 3.1. Depending on the magnitude off m in eq. (3.1), a few dierent types of situations can form at the interface, with varying degrees of impact on the interfacial properties and growth quality. As f m ! 0, defect concentration is minimal and there is a high quality interface between materials 1 and 2. For moderate values of f m , dislocations and other types of mechanical lattice defects can occur, which can impact electronic properties at the interface, and cause structural problems 37 throughout material 2. In the context of core-shell nanoparticles, we are mainly concerned with the ability to grow a thick and uniform layer of material 2 on material 1. Although some adhesion may be possible on material 1 even with signicant lattice mismatch, the layer may not be able to grow very thick or uniformly due to the propagating consequences of the interface. One assumption made here is that material 1 and material 2 have a at interface, as shown in g. 3.1. For the case of nanoparticles with radius 1 nm, we can consider the surface to be locally at, as lattice parameters are typically on the order of several angstroms. The surface chemistries of each material also play a signicant role. Nanoparticles are not purely the nanoparticle material; there is almost always a nal step to functionalize the surface or introduce a new third material as a kind of capping agent. This capping agent can play several roles. These roles include stabilization of the suspension, meaning that the nanoparticles are somewhat repulsed from each other to form a stable suspension in solvent, and sintering protection [4]. If uncapped, many times the particles will sinter together [5]. This is because nanoparticles have a high surface area, and high surface area interfaces have many dangling bonds which are at a higher energy. Gold nanoparticles are usually capped with CTAB [6{9]. But because of this capping, special care needs to be given to the surface chemistry when attempting to deposit a shell layer around the core. The shell material may bind nicely to the unprotected core, but if the core has a capping layer, the shell will not be able to bind with it. I observed and measured this directly in my work with core-shell nanoparticles. Shown in g. 3.2 are two SEM images of attempted AuZnO core-shell nanoparticles (details in section 3.2). The outset of g. 3.2 shows normal, albeit agglomerated, AuZnO core-shell nanoparticles. The outset shows signicantly sintered AuZnO core-shell nanoparticles: the brighter and occasionally rod-shaped gold cores are directly in view. The dierence between these two samples was the choice of surface chemistry - the non-sintered particles had CTAB functionalization. This surface chemistry prevents the interior solid particles from sintering 38 Figure 3.2: Scanning electron microscope image of AuZnO without sintering (outset), with sintering (inset). together. Finally, processing techniques also can be prohibitive factors in core-shell compatibility between two materials. Since the materials need to be synthesized together, it is important that their syntheses not be mutually incompatible. For example, gold nanoparticles are synthesized at 30 C in water. Gold nanoparticles would be incompatible with a processing technique involving deposition of a silica sol gel as the shell material, as silica sol gels require a furnace treatment of 1200 C for several hours. Gold has a melting point of 1064 C, and thus this process would not only destroy the CTAB layer, which decomposes between 200 C and 300 C [10{13], but it would also liquify the gold and prevent formation of a nice core-shell Au-SiO 2 structure. While many nanoparticle materials have a variety of widely dierent processing techniques available to choose from, this can be prohibitive depending on the application. It is important to choose a core-shell synthesis strategy that is amenable to both materials being used. 39 Figure 3.3: TEM image of an AuZnO core-shell nanoparticle. 3.2 Background: AuZnO core-shell nanoparticles The specic core-shell nanoparticle system we are interested in in this section is AuZnO. AuZnO core-shell nanoparticles consist of a gold core, and a ZnO shell, and are about 100- 150 nm in diameter total (30-40 nm core diameter) g. 3.3. This system is able to take advantage of the many unique properties of gold and ZnO nanoparticles. ZnO is known to be biocompatible [14], and when surrounding a gold core, the plasmonic properties of the gold core can be utilized. This is only a possibility since ZnO is transparent in the visible range. If ZnO were not transparent visibly, the entombed gold g. 3.3 would not be optically accessible, limiting the applications of this particle system. AuZnO particles have found applications in photocatalysis [15] and gas sensing [16] and have potential for technological developments in bioimaging [17] and drug delivery [18]. Synthesis of AuZnO particles follows a relatively straightforward procedure, and is based on the work developed by Yang et al [2]. The core-shell nanoparticles were prepared in 20 mL batches. 206 l of 266 mM CTAB, and 220 l of 50 mM ascorbic acid are added to 17.34 mL of DI water. 110 l of 500 mM zinc nitrate hexahydrate is then added. Finally, 1.1 mL 40 of 50 mM hexamethylenetetramine (HMTA) (Millipore Sigma, 99.9%) is added, along with the gold cores. The particles are stable in the fridge at 4 C for several months; up to two months at least. The solution is mixed thoroughly via shaking for 10 seconds and stored at 85 C overnight. After heating overnight, the mixture is a slightly cloudy pink color, just a little bit more turbid than the solution was before heating. The as-made particles are puried and characterized in the same fashion as the gold nanoparticles. It is important to note that the resulting particles are not yet fully ZnO. For these particles, we based our understanding on the model presented in Yang et al [2]. Brie y, it was reported that the Zn + 2 ions could complex with ascorbic acid (AA). This would produce AA- Zn(II). In the presence of HMTA, the pH of the solution is increased when the temperature increases [19], releasing OH to further complex with the Zn ion, forming [AA-Zn(OH) 4 ] 2 . Meanwhile, the CTAB surrounding the gold nanoparticle core can interact with the [AA- Zn(OH) 4 ] 2 complex, which in turn attracts more loose CTAB and forms a meshed net around the gold nanoparticle core. So, while these particles are still core-shell particles in this state, the shell is not yet ZnO, but a Zn-based material involving ligands of ascorbic acid, CTAB, and water. This has important implications on the optical properties of the particle. This precursor Zn-based shell is similar to ZnO but does not behave like ZnO in many crucial ways. For example, a characteristic optical feature of ZnO is its band gap. The presence of the ZnO band gap manifests itself optically by the appearance of a shoulder, or peak centered around 365 nm in the absorbance spectrum of a sample of crystalline ZnO [20{23]. The absorbance spectrum of the Zn-based precursor shell only shows an increasing absorbance in the UV, with no clear peak or any other feature of note around 365 nm. My work in AuZnO core-shell nanoparticles thoroughly investigated this dierence using a multi pronged synthesis approach. A pictorial outline is shown in g. 3.4. There are four main pathways to dierent AuZnO products used in my study: 41 Figure 3.4: Strategy for synthesis showing the possible nanoparticle variations, including doping. Figure created by my lab mate, Mark Veksler (reproduced with permission). A) ZnO shell only, no gold core, no doping B) Doped ZnO shell, no gold core (discussed in section 3.4) C) AuZnO core-shell, no doping D) AuZnO core-shell, doping (discussed in section 3.4) Additionally, between the columns in g. 3.4 labeled 4 and 5, there is the calcination step. I studied the particles before and after calcination, which is where the previously mentioned dierence in the absorption spectrum occurs. The details on this study are discussed in section 3.3. 3.3 Background: calcining ZnO Although the exact composition and nature of the Zn-based precursor shell is unknown, based on the technique described in section 3.2, it is likely to be based on Zn, CTAB, 42 ascorbic acid and water. Therefore, under the right conditions, this shell could possibly be converted into proper ZnO. Based on reports in the literature, a calcining treatment was chosen. Calcining is the process of applying a furnace-based heat treatment to a mineral, in order to dehydrate or modify its crystal structure [24{27]. This is based on the premise that the Zn-based precursor shell could be dehydrated, and the organic products involved in the shell thermally decomposed, leaving only ZnO as the nal product. Special care must be given to not providing too long of a calcine treatment though, as sintering could end up causing signicant particle aggregation, which is undesirable. This is the step between columns labeled 4 and 5 in g. 3.4. The nished, pre-calcine particles were thoroughly centrifuged and decanted to remove any excess reactants: the as-made nanoparticles are centrifuged twice at 21; 000g for 5 minutes, replacing around 90% of the water each time. Due to the amount of CTAB used in the synthesis process, CTAB will precipitate as a solid if left at room temperature. Since the goal of the centrifugation is to remove excess chemicals like the CTAB, the centrifuge is preheated to 30 C before each use, and all centrifugation is done at 30 C. The samples were prepared by drying out the pre-calcine, centrifuged particles on a hot plate, and then transferring the particles to a tube furnace. The calcination process was done at 500 C in air, for 2 hours, with a 15-minute ramp up from room temperature, and then air cooled. After the calcining treatment, there is a distinct color change; the particles are purple-ish, rather than the typical gold nanosphere pink color. This will be discussed in more detail later but is a sign that the calcination treatment has been successful. The particles are then placed in a sonicator and re-dispersed in DI water. The resulting solution is stable for several weeks. Successful conversion of the particles from the Zn-based precursor shell to real ZnO needed to be veried experimentally. To do this, we were able to take advantage of several aspects of the optical dierences between the Zn-based precursor shell, and ZnO. The rst dierence is in the appearance of the band gap absorption peak at 365 nm, shown in g. 3.5. 43 Figure 3.5: UV Vis pre- and post-calcine, demonstrating conversion of the Zn-based mesh shell to ZnO. This peak is not present in the precursor material but is prominently featured in the ab- sorbance spectra of any real ZnO sample. Thus, we could do a before and after calcination comparison of the absorbance spectra obtained by UV Vis spectroscopy, to determine if the calcination process actually works for converting the Zn-based precursor material into ZnO. Upon the calcination treatment, successful conrmation of ZnO conversion was demon- strated in the core-shell nanoparticles. The post-calcine UV Vis showed a clear peak at 365 nm, relating to the wide band gap absorption of crystalline ZnO. An additional feature of the UV Vis spectra in the post calcine particles is the red shift of the gold plasmon resonance. This makes sense, as the plasmon resonance wavelength of a gold nanoparticle depends on the dielectric constant of the surrounding medium. When fully encased in ZnO, the dielec- tric constant changes, resulting in an increase in the resonance wavelength. Powder x-ray diraction studies also indicated ZnO formation (g. 3.6). Several new peaks appeared in the post-calcination XRD that are associated with crystalline ZnO [28]. 44 Figure 3.6: (A) Pre- and (B) post-calcine XRD scans. The ZnO-associated peaks I, O, and P are only present in the post-calcine scan. Peak Angle (deg) Association B 16.8 Pre-calcine shell D 20.3 Pre-calcine shell E 21.0 Pre-calcine shell G 23.8 Pre-calcine shell I 31.5 ZnO shell J 32.7 Substrate K 34.2 ZnO shell L 36.0 ZnO shell M 38.0 Gold core N 44.1 Gold core O 47.4 ZnO shell P 56.4 ZnO shell Table 3.1: Table of peak identication labels for g. 3.6. 45 3.4 Doped AuZnO: background and motivation The technique discussed in section 3.2 for producing AuZnO nanoparticles is relatively low temperature and scalable. However, there are some drawbacks to this synthetic technique, and one of them is that the particles produced are not easily tunable. Several parameters that are usually easy to control during nanoparticle growth, such as the size, and shape, are not a degree of freedom that can be varied by changing the growth conditions. These growth conditions include parameters like reaction temperature and time. So, while this technique is able to overcome issues relating to core encapsulation and material incompatibilities, it suers for being limited with respect to tunability. With this in mind, we set out to nd ways to introduce a new vector for tunability. These pathways were brie y shown in g. 3.4. The thinking is: if a slight change to the synthetic technique could produce more tunable AuZnO nanoparticles with minimal impact to the nal product, then that would be a substantial improvement. We looked at several strategies, both during synthesis, and post-synthesis, for modifying the nal particles in a minimal way while introducing maximal tunability. There were a few criteria that the tunability strategy needed to satisfy in order to be successful. These included not changing the nal product so much that the core-shell particle no longer forms in a recognizable way, not changing the reaction such that any processing incompatibilities arose, and not changing the overall manufacturability of the particles by increasing the costs too much. 3.5 Doped AuZnO: experimental design The most direct way to introduce tunability into the core-shell AuZnO system was to alter the thickness of the ZnO shell. Normally when the particles are synthesized, the core-shell system has a very xed geometry and size distribution. The usual values are 30-40 nm for the gold nanoparticle core diameter, and then 40-60 nm for the Zn-based precursor shell thickness. The gold nanoparticle diameter is easily tunable, as the gold nanoparticles are 46 synthesized in a previous step, and the reaction conditions for gold nanoparticle growth are well understood [29]. That leaves the shell thickness as a variable. Changing the shell thickness could be useful for a couple of reasons. First is the uorescent properties of the particle. The uorescence is related to the band gap, and the defect levels in the shell, both of which can be changed by changing the thickness of the shell. Second is the proximity of the plasmonic resonance to the external environment of the particle. If the radiation due to the plasmonic resonance of the gold core is primarily outside of the particle, the wavelength of the resonance will be shifted, and externally controllable, which could have some useful applications. An intuitive way to decrease the shell thickness would be to simply reduce the reaction time. This is how shell thickness control is achieved in many types of particle systems. If the reaction is quenchable, or if a limiting amount of precursor material is put into the growth solution, the reaction will stop at a controlled time, limiting the ultimate thickness of the shell that nucleates on the gold core. This has been used in other techniques for rough deposition of ZnO on gold cores. However, due to the way that the shell is formed in this technique, this is not as easy to do. Yang et al. [2] included a kinetic study on the ZnO shell formation using this technique. They found that the reaction time mostly aects the degree of cross linking, or interconnection between particles, not the shell thickness. This could be due to the way that the shell is formed. The shell is formed mainly of an ionic network of Zn, CTAB, ascorbic acid, and water, and thus is a kind of molecular framework surrounding the gold core. Upon calcination, this is converted into ZnO, and such the ZnO thickness is predetermined by the Zn-based precursor shell thickness; the similarities between pre-calcine and post-calcine shell sizes can be seen in g. 3.7. Changing the reaction time clearly was not a successful approach for controlling the shell thickness. This was further veried by X-ray Diraction (XRD) scans which showed the formation of strong ZnO peaks associated with the crystalline shell (g. 3.6). An alternative strategy was to add an additional post-processing technique, after calci- 47 Figure 3.7: Scanning electron microscope images of pre-calcine (A-B) and post-calcine (C-D) ZnO particles, showing changes in surface texture. The size distributions are unchanged as the calcining process only converts the shell material into ZnO. nation of the particles. The thought was that if the particles are fully ZnO after calcination, they should be amenable to chemical etching by one of the many methods available in the literature [30{32]. It has been shown that planar ZnO lms are easily etched by many acids, including HCl [33]. As a simple test, we prepared several solutions of ZnO nanoparticles (post-calcine) and introduced varying amounts of HCl to determine the eectiveness of etch- ing. Under strong stirring, and ice-cold conditions, 0.1 mL of 1 mM HCl was introduced dropwise. SEM imaging was used to verify the impact of HCl etching on the nanoparti- cles before, and after the etching process. It was found that etching was not a controllable strategy for reduction of shell thickness, as some particles would be etched completely, but others would be un-etched g. 3.8. This was veried by observing increasing amounts of loose gold nanoparticles on the post-etch SEM images. Additionally, some particles that were half-etched were observed, indicating poor etching uniformity. An alternative vector for achieving tunability needed to be found. 48 Figure 3.8: Left: partially etched AuZnO core-shell nanoparticles. Right: Totally etched core-shell nanoparticles, leaving only the cores. 3.6 Doped AuZnO: results If the geometry of the core-shell nanoparticles cannot be easily changed, another vector for introducing tunability is in tuning the material composition. This is a well-studied technique that has been broadly applied in technologies involving semiconductor materials. Doping is widely used in silicon processing, particularly for transistor-based technologies. Introducing dopants like boron or phosphorous can change the conductivity type of silicon, which is an intrinsic semiconductor when not doped. Specically, doping is dened as the intentional introduction of foreign atoms into the lattice of a host material at low concentrations. In addition to bulk materials, nanoparticles may also be doped, and several reports have been published on this in the past. ZnO nanoparticles are also susceptible to doping [34{36]. This will be the main strategy for this work, as the gold nanoparticle cores are pre-synthesized and likely not to be susceptible to the higher concentrations of doping required, due to their pure crystalline nature. Our strategy was to dope the nanoparticle shells, rather than the core. An example 49 Figure 3.9: Sample data providing evidence for successful Ce doping in AuZnO nanoparticles. A. TEM image, B. ZnO Spatial EDX, C. Ce Spatial EDX, D. EDX spectrum. The correlation of the ZnO with Ce provides evidence that the Ce intercalated into the ZnO shell. demonstration of this is in g. 3.9. This could have many interesting applications. Brie y, one motivation for introducing dopants would be to add uorescence to the particles and make them viable as a bioimaging nanoparticle. This leverages the natural biocompatibility of ZnO nanoparticles, and the uorescence of the dopants used. Additionally, the eld intensity enhancement of the gold nanoparticle could be taken advantage of; if the absorption peak causing uorescence in the dopant atom overlaps with the surface plasmon resonance of the gold core, and the dopant and the core are close enough, then enhanced uorescent emission, even at low concentrations of the dopant, should be observable. Dopants could also signicantly change the electrical or chemical properties of the nanoparticles. No work has been done to study the impact of dopants on core-shell AuZnO nanoparticles made using this unique technique, and that will be the focus of this section. We foresaw some potential problems with doping this particle. First, the synthesis tech- nique is fairly dierent from other techniques. This technique relies upon a complex network of CTAB, ascorbic acid, and Zn ions forming around a gold core, and it was unknown what conditions a foreign atom dopant would need to satisfy in order to embed itself chemically in 50 this network. Second, we did not know if the dopant would have an unexpected interaction with one of the precursor chemicals or nanoparticles, and actually prevent the formation of the shell around the gold core in the rst place. Finally, the doping limits were unknown. It was unclear what concentrations of dopant atom were usable, if any. We designed a few main experiments to gather evidence that a dopant worked and determine the impact of that dopant on the existing particle. These experiments primarily involved DLS, SEM imaging, SEM Energy-dispersive X-ray Spectroscopy (EDX), TEM EDX, and XRD. The dopants chosen to study in this particle system were primarily rare earth elements, and copper. Specically, the list of dopants tested were Cu, Ce, Er, Nd, Tm, and Yb. These dopants were chosen based on their large size and chemical dissimilarity to Zn, making them good stress test dopants for the Zn-based shell; except, in the case of Cu. Cu was chosen specically because it is known to be quite compatible with Zn, and thus it serves as a good control dopant. The main features we wanted to study were rst of all, incorporation of the dopant, namely how much the dopant would intercalate into the Zn shell, if at all. Second, if it was successfully incorporating, we wanted to study the morphological impact of the dopant on the particle, if it has any, with increasing doping percent. Third, we wanted to understand the impact on the crystal structure of the shell pre- and post-calcine and see if any doping concentration was high enough to form new phases. 3.7 Doped AuZnO: Discussion The core-shell nanoparticles were prepared in 20 mL batches according to the method of Yang et al [2]. Next, 206 l of 266 mM CTAB, and 220 l of 50 mM ascorbic acid are added to 17.34 mL of DI water. 110 l of 500 mM Zn nitrate hexahydrate is then added. If a metal dopant is being used, we add a variable amount of 27.5 mM: copper(II) chloride (Millipore Sigma, 99.9%), cerium(III) nitrate hexahydrate (Reacton, 99.99%), erbium(III) nitrate pentahy- drate (Alfa Aesar, 99.9%), neodymium(III) nitrate hydrate (Reacton, 99.9%), thulium(III) nitrate hydrate (Alfa Aesar, 99.9%), or ytterbium(III) nitrate hydrate (Alfa Aesar, 99.99%). 51 At this concentration, 1%-added corresponds to 20 l of the corresponding metal precursor solution. Finally, 1.1 mL of 50 mM HMTA (Millipore Sigma, 99.9%) is added, along with the gold cores. The gold nanoparticle cores were 2x washed via centrifugation at 21; 000g for 5 minutes each. To one 20 mL batch, we add 1 mL of gold nanoparticles that have been 10x concentrated from the as-made concentration. This concentration has been determined to be suitable such that nearly every particle has a gold core, veried by SEM imaging. The nished, pre-calcine particles were thoroughly centrifuged and decanted to remove any excess reactants and ensure that the metal detected in EDX characterization could not be remaining solution precursor. Calcined samples were prepared by drying out the particles on a hot plate, and then transferring the particles to a tube furnace. The calcination process was done at 500 C in air, for 2 hours, with a 15-minute ramp up from room temperature, and then air cooled. The next step was to perform a series of experiments to determine of the doping was actually successful or not. As a preliminary \rst check," we used SEM imaging and SEM- EDX to determine whether or not the doping growth had succeeded. Success was determined by two main conditions. First, that the core-shell nanoparticles had been grown, with no obvious impact to particle morphology or behavior. This was easily veriable by SEM imaging to conrm that the size and shapes of the doped core-shell particles looks similar to those of the non-doped core-shell particles. Second, we used SEM-EDX to try and detect signal from the dopant. This is where the purication process described above is really important { if the particles were not puried, then it is possible that dopant detected in SEM-EDX is just from left over precursor that was deposited onto the substrate along with the particles. If a dopant was successful, we then carried out further characterization to try and learn about how the dopant was intercalating into the ZnO shell. Dopants that passed the rst two tests went on to be subjected to TEM-EDX. In TEM-EDX, the spatial resolution is high enough that individual particles can be analyzed for elemental composition. An example of 52 this is shown in g. 3.9. This led to the rst test to prove dopant intercalation, which was a spatial correlation test. If the dopant signal could be spatially correlated with the Zn signal, it would be evidence that the dopant was actually in the particle shell { not just everywhere on the substrate because it was not puried well enough. This was possible for all dopants tested, except for Cu and Er. Cu could not be resolved, because there was signicant Cu background from the EDX chamber. Er could not be resolved, because its main spectral lines overlap with those of Co, which is present in signicant amounts in the EDX chamber. Nonetheless, Cu and Er were veried in our other tests. We also did SEM-EDX, which has better signal:noise than TEM-EDX (but lower resolu- tion), to learn about the quantitative composition of the doped particles. Due to the lower spatial resolution, particle aggregates, rather than individual particles, had to be scanned. This has the added benet of averaging over many particles, rather than individual parti- cles, giving a more collective picture of the quantitative composition. SEM-EDX has a better background spectrum than TEM-EDX, so in this experiment, Cu and Er were able to be detected without issue. The experimental design is simple. Six samples were prepared for each dopant, with concentrations ranging from 1-6% in 1% increments (relative mol%). The samples were thorough puried, as discussed above, and dried out on a silicon substrate. 10 data points were taken per concentration, for 60 data points in each composition. The data we are measuring in this experience is the ratio of dopant to Zn measured for each concentration, expressed as a percentage. A linear t is performed for each dopant, and an \intercalation eciency", or intake rate of the dopant as the concentration increases, is mea- sured. This measurement is important since a linear t, or constant intercalation eciency, is a good sanity check for whether or not the doping is working, and it tells us about the compatibility between the Zn-based shell and the dopant. Powder XRD was additionally performed for each dopant, at two concentrations (2% and 10%), both pre- and post- calcine. An non-doped core-shell particle sample was also scanned, pre- and post-calcine, as a control. The goal of this experiment was to 1) test 53 Figure 3.10: DLS spectra of particles synthesized with and without gold cores, highlighting the enhanced stability oered by the gold NP core. A. SEM image of the AuZnO particles. B. Narrow DLS distribution of particles with cores despite added dopants. C. SEM image of doped particles without gold cores, showing dierent morphology. D. Wide distribution of the corresponding no-core particles. In (B), the second peak between 1 m and 10 m is due to observed contaminants in that particular sample. the eect of the calcination on the doped particle shells by detecting a change in the XRD pattern, specically as it relates to peaks associated with ZnO, and 2) identify the extent to which the dopants will phase separate as the concentration in the shell is increased. It was expected that if we saw signicant deformation at the high concentration (10%) in SEM imaging, a new dopant phase may form in the post-calcine particle shells. New phases should not show up in the low concentration (2%) samples, as no signicant particle deformation was seen in SEM images at concentrations that low for any dopant. Finally, to study the role of the gold core on the particle formation, core-shell particles were synthesized with and without a core and with a range of dopant concentrations. When the gold core is absent, the shape and size uniformity of the particles changes dramatically, as shown in g. 3.10. This eect is clearly seen in the DLS results quantitatively, as well as the images of the particles qualitatively. We study this eect with copper dopants at increasing concentrations of copper just as a test case, and additionally at xed concentrations but 54 with every other dopant. This set of experiments allows us to 1. verify the successful inter- calation of dopant atoms into the Zn-based shell of the core-shell nanoparticles in multiple independent ways, 2. study the intercalation eciencies of dierent dopants, 3. study the structural eects of dopants and calcining, and 4. study the impact on the size distribution. 3.8 Future work One line of investigation I began to pursue in this project was the inclusion of gold nanopar- ticle cores with varying geometry. Specically, we focused on gold nanorods and ultra-long gold nanorods. The motivation for doing this was simple: gold nanospheres have a tun- able resonance wavelength in the visible regime, but there are many useful applications for nanoparticles in the infrared part of the electromagnetic spectrum. One primary driver for this is intra-tissue imaging. Tissue has an optical transparency window in the near in- frared [37]. Therefore, dyes that are active within this transparency window can be useful as contrast agents and tagging agents for imaging biological structures. It should be noted that this work was largely team eort in collaboration with my under- graduate mentees over the course of my PhD: Omar Garcia, Josh Greenberg, Dan Cummins, Arynn Gallegos and Mike Shao. The work these undergraduates did was invaluable and contributed signicantly to the progress our lab made in biological nanoparticle research. As mentioned elsewhere in section 2.1, the surface plasmon resonance of gold particles have a dependence not only on the size of the particles, but also the shape [38]. If we imagine the gold nanoparticle having an oscillating surface charge density, it will `take longer' for the charge density to complete one period of oscillation for a longer gold nanoparticle. This eect is similar to the particle acting as a real-world antenna. Antennas have tunable response frequencies dependent upon their length. AM radio waves can be hundreds of meters long and thus need an antenna length on that same order of magnitude to be absorbed. In contrast, Wi-Fi signals are only centimeters long and thus can use a much shorter antenna. A similar principle is at play here, but with dierent orders of magnitude involved. 55 Gold nanorods thus can extend the surface plasmon resonance frequency of gold nanopar- ticles into the near infrared [39]. The process for doing this is to extend the nanorod in one dimension while keeping the other dimension constrained. If the gold nanoparticle is ex- tended in both dimensions (becoming a larger sphere), then the nanoparticle approaches bulk gold and loses its plasmon resonance. To extend the gold nanosphere into a gold nanorod, the procedure proceeds as described in the background chapter, but with a minor modication. New to the reaction is silver nitrate (AgNO 3 ). The silver nitrate preferentially associates with dangling gold bonds on a specic exposed crystal plane [40], blocking those sites from further nucleating new gold atoms. This means that the only place for dissolved gold ions to attach to the particle is on the ends of the particle, breaking the symmetry of the new particle and causing one dimension to be much longer than the other. The aspect ratio of the gold nanorods is determined by the amount of silver nitrate added to the solution; typically, 400 uL of 4 mM silver nitrate solution in water was added. The silver nitrate needs to be added to the very last step of the synthesis, after the seed solution has been introduced to the growth solution. Let the particles sit in a water bath at room temperature, and then the color will change over the course of 2 hours. These gold nanorods can be used in the place of the gold nanospheres for AuZnO synthesis with no further modication. As the ZnO shell protects the gold nanorod cores, future work will include studying this gold nanorod core-shell particle as a biological contrast agent. Some preliminary work was done to study the biotoxicity of these AuZnO core-shell particles with and without gold nanorods in collaboration with the Graham lab at USC. This experiment provided a baseline of biotoxicity for the particles that will act as guidance for future work involving AuZnO particles in biological applications. AuZnO particles were dispersed in phosphine-buered saline solution at a distribution of concentrations: 1/10 as- made concentration, 1/100, and 1/1000, along with a control of pure phosphine-buered saline solution. No signicant cell toxicity was found for the adherent cells used in the study, 56 but more rigorous and detailed studies involving dierent cell types should be performed in the future. 3.9 Conclusion In this chapter, we have discussed a motivation for adapting the synthetic technique to introduce more tunability into the particles, as well as many unsuccessful attempts and why doping must be used instead. The procedure for introducing dopants into the shell of the core-shell nanoparticles was discussed, and also potential problems with this strategy, along with the dopants we chose to study. Finally, a full set of characterization experiments were discussed. The goal of these experiments was to both provide evidence that dopant intercalation into the particle shell was successful, as well as understand the interactions between the dopants and the particles, and the eects of calcining. The state of the art for core-shell nanoparticles was brie y discussed, along with the motivation for examining the focus of this work, AuZnO core-shell nanoparticles. Problems facing development of core- shell synthetic techniques were discussed, along with the main strategy used in this work for synthesizing AuZnO core-shell nanoparticles. One caveat of the strategy discussed is that it does not actually produce AuZnO, but gold with a Zn-based precursor shell, consisting of ascorbic acid, CTAB, water, and the Zn ion. Conversion to ZnO is possible, but requires more care and processing, such as a calcining treatment. The calcining treatment was discussed, and optical analysis techniques verifying conversion of the precursor shell to ZnO were presented. This chapter sets up the nanoparticle platform, which will be utilized and discussed in more detail in following chapters. 57 References 1 J. T. L. Gamler, A. Leonardi, X. Sang, K. M. Koczkur, R. R. Unocic, M. Engel, and S. E. Skrabalak, \Eect of lattice mismatch and shell thickness on strain in core@shell nanocrystals", en, Nanoscale Advances 2, 1105{1114 (2020). 2 Y. Yang, S. Han, G. Zhou, L. Zhang, X. Li, C. Zou, and S. Huang, \Ascorbic-acid-assisted growth of high quality M@ZnO: a growth mechanism and kinetics study", en, Nanoscale 5, 11808 (2013). 3 S. C. Jain, A. H. Harker, and R. A. Cowley, \Mist strain and mist dislocations in lattice mismatched epitaxial layers and other systems", en, Philosophical Magazine A 75, 1461{1515 (1997). 4 E. M. Hotze, T. Phenrat, and G. V. Lowry, \Nanoparticle Aggregation: Challenges to Understanding Transport and Reactivity in the Environment", en, Journal of Environmental Quality 39, eprint: https://onlinelibrary.wiley.com/doi/pdf/10.2134/jeq2009.0462, 1909{1924 (2010). 5 T. W. Hansen, A. T. DeLaRiva, S. R. Challa, and A. K. Datye, \Sintering of Catalytic Nanoparticles: Particle Migration or Ostwald Ripening?", en, Accounts of Chemical Research 46, 1720{1730 (2013). 6 J. Rodr guez-Fern andez, J. P erez-Juste, P. Mulvaney, and L. M. Liz-Marz an, \Spatially-Directed Oxidation of Gold Nanoparticles by Au(III){CTAB Complexes", en, The Journal of Physical Chemistry B 109, 14257{14261 (2005). 7 R. Fenger, E. Fertitta, H. Kirmse, A. F. Th unemann, and K. Rademann, \Size dependent catalysis with CTAB-stabilized gold nanoparticles", en, Physical Chemistry Chemical Physics 14, Publisher: Royal Society of Chemistry, 9343{9349 (2012). 58 8 D. K. Smith and B. A. Korgel, \The Importance of the CTAB Surfactant on the Colloidal Seed-Mediated Synthesis of Gold Nanorods", Langmuir 24, Publisher: American Chemical Society, 644{649 (2008). 9 I. Pastoriza-Santos, J. P erez-Juste, and L. M. Liz-Marz an, \Silica-Coating and Hydrophobation of CTAB-Stabilized Gold Nanorods", Chemistry of Materials 18, Publisher: American Chemical Society, 2465{2467 (2006). 10 A. Akgsornpeak, T. Witoon, T. Mungcharoen, and J. Limtrakul, \Development of synthetic CaO sorbents via CTAB-assisted sol{gel method for CO2 capture at high temperature", en, Chemical Engineering Journal 237, 189{198 (2014). 11 H. V. Vasei, S. M. Masoudpanah, M. Adeli, and M. R. Aboutalebi, \Solution combustion synthesis of ZnO powders using CTAB as fuel", en, Ceramics International 44, 7741{7745 (2018). 12 L. Huang, X. Chen, and Q. Li, \Synthesis of microporous molecular sieves by surfactant decompositionElectronic supplementary information (ESI) available: IR spectra and assignments for CTAB and its decomposition product; decomposition of CTAB in the presence of SiO{ or OH{. See http://www.rsc.org.libproxy1.usc.edu/suppdata/jm/b0/b005770n/", en, Journal of Materials Chemistry 11, Publisher: Royal Society of Chemistry, 610{615 (2001). 13 A. Akgsornpeak, T. Witoon, T. Mungcharoen, and J. Limtrakul, \Development of synthetic CaO sorbents via CTAB-assisted sol{gel method for CO2 capture at high temperature", en, Chemical Engineering Journal 237, 189{198 (2014). 14 Z. Li, R. Yang, M. Yu, F. Bai, C. Li, and Z. L. Wang, \Cellular Level Biocompatibility and Biosafety of ZnO Nanowires", The Journal of Physical Chemistry C 112, Publisher: American Chemical Society, 20114{20117 (2008). 59 15 C. Ma, X. Wang, S. Zhan, X. Li, X. Liu, Y. Chai, R. Xing, and H. Liu, \Photocatalytic Activity of Monosized AuZnO Composite Nanoparticles", en, Applied Sciences 9, Number: 1 Publisher: Multidisciplinary Digital Publishing Institute, 111 (2019). 16 X. Li, X. Zhou, H. Guo, C. Wang, J. Liu, P. Sun, F. Liu, and G. Lu, \Design of Au@ZnO Yolk{Shell Nanospheres with Enhanced Gas Sensing Properties", en, ACS Applied Materials & Interfaces 6, 18661{18667 (2014). 17 P. Zhu, Z. Weng, X. Li, X. Liu, S. Wu, K. W. K. Yeung, X. Wang, Z. Cui, X. Yang, and P. K. Chu, \Biomedical Applications of Functionalized ZnO Nanomaterials: from Biosensors to Bioimaging", en, Advanced Materials Interfaces 3, 1500494 (2016). 18 T. Chen, T. Zhao, D. Wei, Y. Wei, Y. Li, and H. Zhang, \Core{shell nanocarriers with ZnO quantum dots-conjugated Au nanoparticle for tumor-targeted drug delivery", en, Carbohydrate Polymers 92, 1124{1132 (2013). 19 K. M. McPeak, T. P. Le, N. G. Britton, Z. S. Nickolov, Y. A. Elabd, and J. B. Baxter, \Chemical Bath Deposition of ZnO Nanowires at Near-Neutral pH Conditions without Hexamethylenetetramine (HMTA): Understanding the Role of HMTA in ZnO Nanowire Growth", en, Langmuir 27, 3672{3677 (2011). 20 J. Jim enez Reinosa, P. Leret, C. M. Alvarez-Docio, A. del Campo, and J. F. Fern andez, \Enhancement of UV absorption behavior in ZnO{TiO2 composites", en, Bolet n de la Sociedad Espa~ nola de Cer amica y Vidrio 55, 55{62 (2016). 21 L. Wu, Y. Wu, and W. L u, \Preparation of ZnO Nanorods and optical characterizations", en, Physica E: Low-dimensional Systems and Nanostructures 28, 76{82 (2005). 22 M. Salavati-Niasari, N. Mir, and F. Davar, \ZnO nanotriangles: Synthesis, characterization and optical properties", en, Journal of Alloys and Compounds 476, 908{912 (2009). 60 23 S. Talam, S. R. Karumuri, and N. Gunnam, \Synthesis, Characterization, and Spectroscopic Properties of ZnO Nanoparticles", en, ISRN Nanotechnology 2012, 1{6 (2012). 24 Y. Hong, C. Tian, B. Jiang, A. Wu, Q. Zhang, G. Tian, and H. Fu, \Facile synthesis of sheet-like ZnO assembly composed of small ZnO particles for highly ecient photocatalysis", en, Journal of Materials Chemistry A 1, Publisher: Royal Society of Chemistry, 5700{5708 (2013). 25 S. Suwanboon, \[No title found]", en, ScienceAsia 34, 031 (2008). 26 C. Wang, B.-Q. Xu, X. Wang, and J. Zhao, \Preparation and photocatalytic activity of ZnO/TiO2/SnO2 mixture", en, Journal of Solid State Chemistry 178, 3500{3506 (2005). 27 M. R. Parra and F. Z. Haque, \Aqueous chemical route synthesis and the eect of calcination temperature on the structural and optical properties of ZnO nanoparticles", en, Journal of Materials Research and Technology 3, 363{369 (2014). 28 R. Zeto, D. Cummins, A. Gallegos, M. Shao, and A. M. Armani, \General strategy for doping rare earth metals into Au{ZnO core{shell nanospheres", en, Journal of Materials Research 34, Publisher: Cambridge University Press, 3877{3886 (2019). 29 B. Nikoobakht and M. A. El-Sayed, \Preparation and Growth Mechanism of Gold Nanorods (NRs) Using Seed-Mediated Growth Method", en, Chemistry of Materials 15, 1957{1962 (2003). 30 J. S. Park, H. J. Park, Y. B. Hahn, G.-C. Yi, and A. Yoshikawa, \Dry etching of ZnO lms and plasma-induced damage to optical properties", Journal of Vacuum Science & Technology B: Microelectronics and Nanometer Structures Processing, Measurement, and Phenomena 21, Publisher: American Institute of Physics, 800{803 (2003). 31 H. Wang, G. Li, L. Jia, G. Wang, and C. Tang, \Controllable Preferential-Etching Synthesis and Photocatalytic Activity of Porous ZnO Nanotubes", The Journal of Physical Chemistry C 112, Publisher: American Chemical Society, 11738{11743 (2008). 61 32 E. A. Meulenkamp, \Size Dependence of the Dissolution of ZnO Nanoparticles", The Journal of Physical Chemistry B 102, Publisher: American Chemical Society, 7764{7769 (1998). 33 H. Maki, T. Ikoma, I. Sakaguchi, N. Ohashi, H. Haneda, J. Tanaka, and N. Ichinose, \Control of surface morphology of ZnO by hydrochloric acid etching", en, Thin Solid Films 411, 91{95 (2002). 34 J.-Z. Kong, A.-D. Li, X.-Y. Li, H.-F. Zhai, W.-Q. Zhang, Y.-P. Gong, H. Li, and D. Wu, \Photo-degradation of methylene blue using Ta-doped ZnO nanoparticle", en, Journal of Solid State Chemistry 183, 1359{1364 (2010). 35 R. Sankar Ganesh, E. Durgadevi, M. Navaneethan, V. Patil, S. Ponnusamy, C. Muthamizhchelvan, S. Kawasaki, P. Patil, and Y. Hayakawa, \Low temperature ammonia gas sensor based on Mn-doped ZnO nanoparticle decorated microspheres", en, Journal of Alloys and Compounds 721, 182{190 (2017). 36 M. Li, S. Pokhrel, X. Jin, L. M adler, R. Damoiseaux, and E. M. V. Hoek, \Stability, Bioavailability, and Bacterial Toxicity of ZnO and Iron-Doped ZnO Nanoparticles in Aquatic Media", en, Environmental Science & Technology 45, 755{761 (2011). 37 C. Li and Q. Wang, \Challenges and Opportunities for Intravital Near-Infrared Fluorescence Imaging Technology in the Second Transparency Window", en, ACS Nano 12, 9654{9659 (2018). 38 J. Cao, T. Sun, and K. T. Grattan, \Gold nanorod-based localized surface plasmon resonance biosensors: A review", en, Sensors and Actuators B: Chemical 195, 332{351 (2014). 39 Y.-N. Wang, W.-T. Wei, C.-W. Yang, and M. H. Huang, \Seed-Mediated Growth of Ultralong Gold Nanorods and Nanowires with a Wide Range of Length Tunability", en, Langmuir 29, 10491{10497 (2013). 62 40 W. Tong, M. J. Walsh, P. Mulvaney, J. Etheridge, and A. M. Funston, \Control of Symmetry Breaking Size and Aspect Ratio in Gold Nanorods: Underlying Role of Silver Nitrate", en, The Journal of Physical Chemistry C 121, 3549{3559 (2017). 63 Chapter 4 ZAIS Quantum Dots In the preceding chapters, metal nanoparticles and a core-shell metal-insulator nanoparticle system were discussed, which made up the early portion of my graduate research. After those projects, I began to investigate a new type of semiconductor core-shell nanomaterial, based on the QHMF quantum dots introduced in section 2.8, ZAIS quantum dots. The motivation for studying these came from the AuZnO project discussed in chapter 3. Although it was a goal to nd some kind of luminescence behavior from the AuZnO particles, it proved dicult as the concentration of dopants had to be quite low. Replacing the metal core with a semiconductor material (AgInS 2 ) would allow for good emissive properties, while keeping the zinc protective outer shell (ZnS in this case of ZAIS quantum dots). Semiconductor quantum dots are a nanoparticle form of traditional bulk semiconductor materials. In this chapter, extensive details on the production of ZAIS quantum dots will be discussed, and key experimental ndings will also be presented. Quantum dots are a photostable [1{6] and ecient light source [7{11]. They may be used in several lighting applications, such as LED displays [12], or biological imaging agents [13]. The latter application is of particular interest to my lab. However, the state of the art in quantum dot technology is currently not ZAIS. The state of the art is founded on Cd-based quantum dot emitters [14], particularly cadmium chalcogenide particles with a protective zinc chalcogenide overcoating. Cadmium, being a heavy metal, has toxicity problems which make its widespread adoption dicult [15]. ZAIS quantum dots may potentially be a viable alternative emitter material to cadmium. 64 ZAIS quantum dots belong more broadly to a material family known as QHMF quantum dots. QHMF quantum dots are a family of materials based on a parent binary chalcogenide structure. The parent structure is a metal dichalcogenide, A x B y S 2 (A = Ag or Cu, B = In or Ga). Remarkable about these dichalcogenides is their ability to tolerate a wide range of chemical compositions; that is, the crystal lattice can still form and retain useful properties with severe deciency or surplus of each constituent element [16]. This introduces a degree of control not present in traditional Cd-based quantum dots. Cd-based quantum dots have emission tunability that is decided primarily by the size of the particle, via the quantum connement eect [17]. Although this eect is also present in QHMF quantum dots, the emission tunability is primarily decided by compositional factors, which can be controlled more precisely than size tunability. Additionally, in QHMF quantum dots, there are four constituent elements, and thus changing the relative concentration of each one of them can in uence many uorescent properties [18]. Therefore, the tunability is not only more precise, but there are more degrees of freedom to the system which allow for material optimization. ZAIS in particular has unique emissive properties. Unlike other quantum dots such as CdS, the source of the luminescence is related to a donor-acceptor pair transition [16]. Past works have investigated the nature of the particular transition that occurs in ZAIS and similar materials in the same family [19] and observed a luminescence quenching eect when the quantum dots are subject to a heat treatment [20]. This eect is worth investigating as only certain types of defects are able to be aected by a thermal annealing process. Atomic interstitials and vacancies can be aected by re- crystallization { changes that aect the composition of the particle, such as foreign impurity atoms or a shift in the relative concentrations, are not viable candidate defects. 4.1 Synthesis In order to perform experiments on ZAIS quantum dots, I rst had to synthesize them. I followed an established technique from the literature. Specically, the hot injection method 65 Figure 4.1: Synthesis experimental setup. Reagents are deposited via the access septum, while N 2 is constantly owing through the reaction pot. of Yoon et al. [21] was used. The precursor materials included Indium(III) acetylacetonate (In(acac) 3 , 99.99%, Aldrich), Silver nitrate (AgNO 3 , Ward's science), Sulfur (S, 99.98%, Aldrich), and Zinc stearate (10 to 12% Zn basis, Aldrich) for form the elemental units of the ZAIS quantum dot. Since a hot injection method was used to create the ZAIS quantum dots, I also need high boiling point solvents: these include 1-octadecene (ODE, 90%, Aldrich), 1-dodecanethiol (DDT, 98%, Aldrich), Oleylamine (OLA, 70%, Aldrich), and Trioctylphosphine (TOP, 90%, Aldrich). The required equipment includes a hot plate with an oil bath, a three necked ask, and a nitrogen purge line. Placing a stir bar in the oil bath and setting it to its max temperature is recommended, as the fume hood air ow tends to cool down the oil bath. A schematic is shown in g. 4.1. The exact amounts of precursor and solvent depend on the composition and volume that is desired. ZAIS quantum dots are characteristically exible in the allowed compositions and this is something that can be experimented with. A typical synthesis of ZAIS quantum dots proceeds as follows. 0.1 mmol of AgNO 3 and 0.4 mmol of In(acac) 3 are added to a the three-necked ask at room temperature. Importantly, do not place the three necked ask into the heated oil bath 66 just yet. The three necked ask needs to remain dry and at room temperature, preferably sitting in some kind of cork ring. Obtain the required amounts of each silver and indium precursor, and deposit them into the dry three necked ask. The required amount depends on the exact concentration that is being targeted, and can vary from concentration to con- centration. Use stoichiometrically appropriate amounts; doubling the silver will double the concentration of silver in the nal product. Be particularly careful with these two precur- sors: silver nitrate causes discolorations and burns on the skin, and indium acetylacetonate is dangerous to inhale or touch. As a side note: since I was focused on studying the photophysics mechanisms of the quantum dots, I focused specically on optical and electronic structure changes. I held the composition xed at 3.6% Ag, 14.5% In, 23.6% S, and 58.2% Zn. There is nothing particularly special about this concentration, but it minimized reagent impact and produced stable quantum dots. After obtaining the silver and indium precursors, pull 1-octadecene out and measure out 25 mmol into the three necked ask. This can be done volumetrically using a pipettor, which is preferred since it makes it easy to be done in the fume hood. 1-octadecene is dangerous and light sensitive, so turn the lights o in the fume hood and do not inhale its vapors. Return the 1-octadecene from whence it came and pull a bottle of oleic acid out of a 4 C fridge. Oleic acid is a fatty acid that solidies in the fridge but melts at room temperature. It is best to keep it cold, since it will spoil quickly at room temperature in its liquid form. Do not thaw the entire bottle. Bring the cold bottle into the fume hood and scrape a bunch of it out into a small, graduated ask. Let that melt (do not use a hot plate) and then pipette 1.5 mmol of it into the reaction pot, the three necked ask. Place a small, round magnetic stir bar into the reaction pot and then hook it up to the nitrogen purge line. Place septums on the necks that are not connected to the N 2 line. Use a small syringe needle in one of the septums to act as an exhaust port. If you do not do this, 67 then there may be too much nitrogen ow into the reaction pot, which will cause something to pop o eventually. It is very dangerous to neglect this, since the reaction is taking place at 200 C. There should not be reactants at 200 C going into a dirty oil bath below the reaction pot. Take care to test the ow rate and use the exhaust needle in one of the septums. Degas the reaction pot for 20 minutes before submerging it into the hot oil bath at 200 C. In parallel, use a pre-dissolved mixture of 0.65 mmol Sulfur in 4 mmol of oleylamine. It takes a long time for this to dissolve, so it is best to do it the day before and stir/heat it up to 70 C when it is needed. The sulfur-oleylamine mixture can be kept at room temperature in the fume hood safely. Oleylamine is quite corrosive and toxic. Take great care when working with oleylamine. Purge the mixture of oleylamine and sulfur with nitrogen gas. After the reaction pot has been purged, ramp the temperature to 90 C. Let the com- pounds in the reaction pot react for 30 minutes. After 30 minutes, transfer the oleylamine- sulfur mixture into the reaction pot and increase the temperature to 120 C for three minutes. What has currently formed in the reaction pot is the core of the quantum dot, silver indium sulde (AgInS 2 ). The core of these particles is itself a luminescent quantum dot (g. 4.2), and the process can stop here. However, there are many advantages to continuing with the reaction by adding on a shell layer, forming the full core-shell quantum dot. Adding shells increases the energy of the emission, which shifts the color to bluer emission. So, adding shells is a way to control the emission color. The second reason is that the quantum dots are not shelf stable without a protective shell layer. The core AIS quantum dot will react to contaminants in the environment and lose its luminescence over time. Therefore, a ZnS protective shell provides stability and longevity to the nanoparticle, which is important for any bio-imaging agent. An example of 1-shell and no-shell particles are shown in g. 4.2. To begin the process for adding the shell, a new solvent, trioctylphosphine, is needed. Trioctylphosphine serves two roles in this reaction: 1) it dissolves the zinc stearate mixture, and 2) it serves as a surface ligand that stabilizes the particles in a suspension in nonpolar solvents. This last part is important as the quantum dots will be immediately prepared in a 68 Figure 4.2: Two dierent batches of synthesized ZAIS quantum dots with dierent photolu- minescence curves. The number of shells grown on the quantum dot in uence its photolu- minescence wavelength. The top row was prepared with one ZnS shell, and the bottom row was prepared with no ZnS shell, highlighting the signicant impact shell number has on the luminescence properties of ZAIS quantum dots. 69 crude mixture of all the organic solvents, and later transferred to methanol. Trioctylphos- phine is pyrophoric, so great caution is needed to not expose it to air. Dissolve 0.4 mmol of Zn stearate and 0.4 mmol of S into 0.4 mmol of trioctylphosphine. This was done in a round bottom ask being purged with nitrogen. The solution was transferred to the reaction pot using a needle. Afterward, the reaction pot's oil bath was raised to 180 C. When the temperature hit 180 C, the reaction pot was left for two hours. Afterwards, the sample may be taken out, or this process repeated to add more shell layers. For stability, one layer is enough. Only add more layers if a bluer emission wavelength is desired. 4.2 Purication and characterization Sample preparation is a nuanced topic when it comes to ZAIS quantum dots. Although they are quite stable and luminescent in their crude form (i.e., in the reaction pot they were made in), there is a lot of other material in the reaction pot. Thus, to proceed with any type of measurement, the sample must rst be puried and separated from all the excess reactant and reactant products. To purify the ZAIS quantum dots from the crude liquid, take an aliquot of 500 l from the reaction pot and place it into an Eppendorf centrifuge tube (1.5 mL). Add 500 l of ethanol, and then ensure the particles are dispersed by sonicating them using an ultrasonic bath. This only needs to be done for a minute or two, until there is no visible sedimentation in the centrifuge tube. Place the tube in the centrifuge with a counterweight of 1.5 mL ethanol, and centrifuge at 21000g for 5 minutes. There should be a clear stratication of the particles and the crude organic compounds that were left over from the reaction pot. Repeat this process three times, and in the third step, use methanol instead of ethanol. Finally, repeat one more time, replacing all of the methanol with chloroform. Transfer to a glass vial { the quantum dots are now suspended in chloroform and shelf stable for at least two weeks. 70 Many standard sample analysis techniques used by materials scientists are not amenable to wet samples. These include things like powder x-ray diraction, x-ray photoelectron spectroscopy, scanning electron microscopy, energy dispersive x-ray analysis, transmission electron microscopy, atomic force microscopy and more. All of these techniques use high energy beams, whether they be electron beams or photon beams, which require operation in vacuum to avoid scattering and distortion caused by the atmosphere or solvent. Clearly, an approach for drying out the ZAIS quantum dots is needed, since all of these techniques are potentially useful for experiments to characterize them, even if they will ultimately be rehydrated for a bio-imaging application. A standard process was developed for drying the particles out. After the purication into chloroform process had been completed, clean glass coverslips were obtained, cleaned with a quick wash in acetone, methanol, and isopropanol, and heated to 60 C. It may also be necessary to resize the glass coverslips if they do not t in the measurement apparatus. To resize the glass coverslips, I used a razor blade and a rubber hammer { the thin glass will fracture cleanly with one swift blow. Pipette 5 l droplets of the ZAIS quantum dots in chloroform onto the heated slide. This may take some time since the slide is below the boiling point of the chloroform. However, this is necessary to avoid bubbles and spattering which would not produce a uniform layer of quantum dots for measurement. Repeat the pipetting until a desired coating opacity has been achieved { 750 l (half a centrifuge tube) was enough to produce optically dense samples that had good signal to noise for measurements. These dried lms on glass coverslips could be used for any optical measurements, such as UV Vis or uorometry. They were also good for powder x-ray diraction. Care needed to be taken when using electron beam measurements, as the glass coverslip will charge up and either get damaged or produce bad data, such as unrecognizable images or spectra with fake peaks and other artifacts. Depending on the measurement desired, sputter coating with some metal may be a viable technique. Energy dispersive x-ray spectroscopy signal can still be collected, as the metal coating layer does not block emission of the x-rays coming from 71 the quantum dots. For transmission electron microscopy, use a similar process, except deposit only 5 l of the chloroform sample onto a copper grid, instead of the glass coverslip. 4.3 Experimental goals Past works have provided strong evidence that the mechanism responsible for photolumi- nescence in ZAIS quantum dots is fundamentally related to a defect-assisted transition [22]. Defect states provide near band edge traps for carriers, and then those carriers can later recombine to relax and emit a photon [19]. However, the specic type of defects involved re- mains a matter of debate, complicated by the large compositional degree of freedom present in ZAIS quantum dots. Furthermore, past works have demonstrated this thermal photolu- minescence quenching eect [20]. It was thought that, since the luminescence is primarily a defect-assisted process, thermal annealing causes a signicant recrystallization of the quan- tum dots, which eliminates the necessary defect centers [23]. This work provides new evidence that the defect centers assisting luminescence must be a type of thermally controlled defect, such as an interstitial atom or a lattice vacancy. Our strategy was multi-pronged. First, a quick-cool annealing treatment was used to produce the photoluminescence quenching. The motivation for this is that in those past works, it was suggested that the loss of photoluminescence is due to an overall increase in lattice order after the thermal treatment, possibly due to the treatment allowing for the loss of defect states at high temperature, perhaps due to recrystallization [19, 23]. A recrystallization process would allow for ions in the lattice to move around to thermodynamically more favorable states, which could plausibly reduce the defect concentration, and thus the likelihood that carriers radiatively recombine [23]. A quick-cool annealing treatment sidesteps this and locks in the metastable thermal state, which should contain be highly disordered. The quick-cool annealing process involved a temperature ramp to a variable set point at a xed rate (5 C per 10 minutes), followed by a temperature hold at the variable set 72 point temperature. When that was nished, the samples were quickly removed from the furnace and allowed to cool in air, with the goal of locking in the excited thermal state of the material. Again, this is in contrast with explicitly recrystallization-focused experiments, where the slow cooling allowed for optimization of the crystal lattice in the quantum dots. Additionally, I probed the electronic structure of the band gap directly using Ultraviolet Photoelectron Spectroscopy (UPS). UPS is a ne-resolution technique for measuring the binding energies of outer shale (e.g., valence) electrons on the surface of a material. As such, UPS measurements can provide a sense of the density of states at band-gap energies, which are normally too narrow of a band to measure with x-ray photons in traditional XPS measurements (XPS linewidths several eV vs UV linewidths .01 eV [24]). From this, I can demonstrate signicant structural electronic change to the material, despite the fact that its optical interactions have been quenched by the quick cool treatment. To the extent that they were not, I measured excited state lifetimes using Time-resolved Photoluminescence Decay (TRPL). 4.4 Measuring disorder Disorder, in the context of QHMF quantum dots, means that there is a large concentration of defect states near the band edges which can \blur" the band gap. This presents itself in the absorption spectra of the particles. Ideally, semiconductor quantum dots will exhibit a sharp edge in the absorption spectrum at the band gap energy E g . This is due to the fact that photons with an energy below E g do not have the energy to excite an electron from the valence band to the conduction band, and thus will pass through or scatter o of the crystal, rather than being absorbed. This transition is sharp in highly ordered crystals, but not in disordered crystals [25]. By measuring the absorption of the particles and looking at the "sharpness" of that transition, the amount of disorder can be quantied as an Urbach energy, E u [26]. 73 Figure 4.3: Optical absorption spectra for two dierent samples of quantum dots produced. The dashed line represents the theoretical sharp band edge, and the amount that the mea- sured solid line deviates from that may be quantied as the Urbach energy E u . Samples in g. 4.3 demonstrate this calculation for two test samples of ZAIS without the ZnS shell ("AIS"). These two samples were made under slightly dierent conditions, and thus have slightly dierent band gap energies. The dashed lines represent the theoretical sharp transition of a typical semiconductor absorption spectra, but the large defect concentration present in ZAIS samples with no shell create a signicant softening of that sharp transition (the measured solid line). MeasuringE u for sample of quantum dots can be done by suspending the quantum dots in solution and taking an absorption measurement with UV-VIS. Absorption measurements for semiconductor quantum dots typically have a at region with little to no absorbance for photon energies less than the band gap (E < E g ), and then a monotonically increasing region for photon energies greater than the band gap (E >E g ). Tauc coordinates [27] are a useful way to model the absorbance, as shown in eq. (4.1). When E >E g : ()h = 0 p hE g (4.1) where() is the absorption of the particles, h is the incident photon energy, and 0 is 74 a proportionality constant. This allows for determination of the band gap from absorption data. When the semiconductor is highly disordered, as in the case of ZAIS quantum dots, the long Urbach tail can be modeled by eq. (4.2). For E <E g [27]: () = 0 exp h E u (4.2) Plotting ln() vs h, E u can thus be extracted from the slope of the linear region. This value E u parameterizes the disorder of the crystal lattice in this system [27]. It can be manipulated by various treatments and synthesis modications, as reported in past studies [28]. In this study, samples of QHMF ZAIS quantum dots were produced with dierent E u values via the quick-cool annealing process. 4.5 Results & discussion Our quick-cool treatment (described in section 4.3) succeeded in producing thermally quenched quantum yields, similar to what was seen in past works. Unique to this work is that I also performed measurements of the crystalline lattice disorder, as parameterized by the samples' Urbach energies. I found that with increasing temperatures, the samples had larger Urbach energies, indicating a higher degree of lattice disorder. If this is the case, then there should be a larger concentration of defects, and the photoluminescence quantum yield should be generally unaected, or at least not completely quenched. Nonetheless, the visible emission was completely gone after a thermal treatment, with quantum yield values approaching zero at higher temperatures (g. 4.4). 75 Figure 4.4: A sharp drop in PL quantum yield associated with an increase in sample Urbach energy at high temperatures, correlating a highly disordered lattice with a loss of lumines- cence. Quantum yield was unmeasurable beyond 240 C. It is important to emphasize that where past works focused on recrystallization as a vehi- cle for manipulating the lattice disorder, this experiment quickly cooled the samples, locking in potentially more disordered congurations that were generated by the high temperatures. With increasing temperatures, the Urbach energy was generally higher, indicating a more disordered crystal lattice. The increase in disorder was caused by the quick air-cooling times of the samples. Results from the quick-cool annealing treatment showed that the annealing process was successful at manipulating the Urbach energies of the samples (see g. 4.4), and showed the surprising result that more disordered lattices still had the thermally-quenched photoluminescence. Electronic structure measurements obtained via UPS showed a distinct change in the shape of the valence band edge at higher temperatures (g. 4.5). This provided strong evidence that an intrinsic electronic change had occurred in the samples annealed at high temperatures, which is directly correlated with the photoluminescence drop and higher Ur- bach energies. In particular, the shoulder that appears in the spectrum of the samples annealed higher than 200 C seems to suggest the appearance or consolidation of states in a band, rather than the loosely smeared out edge expected in amorphous semiconductors. 76 Figure 4.5: UPS data of the annealed samples, indicating a signicant electronic structural change. Specically, these UPS measurements showed a sharper band edge appearing around 2 eV binding energy. Samples annealed at a lower temperature presented an exponentially decaying band tail, which is consistent with the optical Urbach energies measured. This result provides evidence that a signicant change in the electronic structure of the quantum dots occurred in samples annealed at temperatures above 200 C, coincident with the drop in quantum yield. This rules out the possibility of chemical or other non-electronic eects quenching the photoluminescence. The photoluminescence decay data provided insight as to what may be occurring. At the hottest temperature, a signicantly longer excited state lifetime for one of the two states responsible for the transition was measured (g. 4.6). This provides more evidence for fundamental electronic structure change as a result of the thermal annealing treatment and suggests a possible reason for the photoluminescence quenching itself. Since the second excited state lifetime has changed, it is plausible that another, less radiative recombination pathway could have been created, which facilitated the drop in quantum yield measured at 77 high temperatures despite the high degree of crystalline disorder. Figure 4.6: Quenched quantum yield with increasing temperatures also correlates with a sharp increase in the excited state lifetime for one of the states responsible for luminescence. Two exponentials were found to be the most appropriate t for the recorded decay curves, indicating two excited states predominantly involved in the emission process. This agrees with the donor-acceptor recombination hypothesis suggested in the literature [16, 29, 30]. Signicantly, samples that were annealed above 200 C had a longer excited state lifetime for one of the two states, from 550 ns to 600 ns as shown in g. 4.6. This result suggests that the one of the excited states is signicantly perturbed by the annealing process, indicating that the responsible defect state is greatly aected by thermal processes (e.g., vacancy or interstitial state). 4.6 Future work: blinking background It has been well-documented that a quantum dot's uorescence is intermittent [31] { that is, even under steady-state excitation, a quantum dot's uorescence will randomly turn o for some timet off . On a practical level, this means that with a certain probability for some timet off , quantum dot unable to be turned on. What complicates the matter is that recent studies [31] have shown that the probability distribution which governs t off for quantum 78 dots appears to obey a power law relationship, meaning that it is not possible to predict when or if the quantum dot will turn on again. This behavior is known as blinking. A key future work for this line of investigation is a blinking study. This blinking study is underway in collaboration with my lab mate, Yasaman Moradi, who is specically inves- tigating it for biomedical imaging agent applications. In this section, I will present some background on quantum dot blinking, and why it is a problem for practical applications such as imaging. Blinking is a general property of quantum dots and thus is not specic to QHMF quan- tum dots. However, the particular causes of blinking are complex, and situational [32{35]. Early experiments showed blinking to be associated with a charging event occurring within a single quantum dot [36]. The idea is as follows: when a photon with the appropriate energy is absorbed by the quantum dot, an electron-hole pair is generated, which later may recombine radiatively to produce uorescence if the quantum dot is neutral. If the quantum dot possesses an excess charge, there will be an imbalance of carriers (electrons or holes, depending on the sign of the charge) within the particle. In the presence of a carrier imbal- ance, another recombination process becomes favorable: Auger recombination [37]. Auger recombination may be thought of as a non-radiative process in which an electron-hole pair recombine and transfer their energy into a exciting a third carrier, rather than emitting a photon. Auger recombination is much faster than radiative recombination [31], leading to quenching of uorescence in the favor of non-radiative Auger recombination. For example, if the quantum dot has a negative charge, there will be a surplus of electrons within the quan- tum dot. When a photoexcited electron in the negatively charged quantum dot recombines with a hole, Auger recombination will quickly transfer that energy into exciting a second electron to a higher state before radiative recombination can occur, causing no photon to be emitted. This begs the question: if the quantum dot is electrically neutral to begin with, how does it acquire a charge? The answer is that it actually does not need to acquire a charge; a 79 spatial charge separation within the particle is enough [36, 38{41]. It is believed that when a neutral quantum dot absorbs a photon, one of the photoexcited carriers may nd its way to a trap state near the surface of the particle [31]. The means by which it can do this is an active research topic, with some of the leading theories being quantum tunneling and thermal assistance [31]. Regardless of how the electron-hole pair becomes separated, if one of them resides in a trap state near the surface, large electric elds get turned on within the particle. This is enough to suppress radiative recombination in favor of Auger recombination, turning the quantum dot uorescence o for some time. The diculty with blinking for using this material in an applied technology lies in its unpredictability. After the discovery of quantum dot blinking, it was suggested that the amount of time a quantum dot spends turned o (t off ) should obey a single exponential probability density function; that is, longer times spent o should be exponentially less likely to occur than shorter times. Experiments showed that the actual behavior was much more complex and problematic, as the probability density function P(t off ) obeyed a power law distribution, shown in eq. (4.3): P (t off ) 1 t off (4.3) where is reported to range from 1.4 to 1.8 [31]. The signicance of this is that for < 2, the integral to calculate an average o-time for the quantum dots diverges. Therefore, when a quantum dot turns o, it is impossible to say with some certainty when it could be expected to turn back on. This is dierent than if the probability density function were a single exponential decay function, where it would be quite simple to say with certainty that most quantum dots will blink for a time less than some multiple of the average o-time,ht off i, allowing the issue to be mitigated in other ways. This problem is dicult but not intractable, as in ZAIS quantum dots, it is fundamen- tally related to the disordered nature of the crystal structure. As demonstrated by Scher et al.[42], the power law distribution function, mathematically, can arise from a sampling of 80 exponentially-distributed rate constants for processes which are in fact governed by a single exponential distribution. Physically, the source of the exponential distribution of rate con- stants can arise from many sources, such as an exponentially-distributed set of trap state depths [31]. Experiments probing into the disordered nature of the crystal lattice could po- tentially shed light on this, as decreasing the dispersion of trap state energies and positions could lead to a more single exponential blinking behavior, increasing the reliability of the uorescence. Future work in this topic would pursue that line of experiments with extended blinking studies on ZAIS quantum dot samples. 4.7 Future Work: blinking characterization procedure I worked on an experimental process to measure and characterize the blinking behavior for ZAIS quantum dots. The goal of this approach is to produce a gure that describes the blinking state of a sample of ZAIS quantum dots. The most direct form of that data would be a P (t off ) vs t off probability density plot. This type of gure is quite informative as it describes, for an ensemble of ZAIS quantum dots, the relative probability that a quantum dot might turn o for some xed duration of time. This approach forms the basis for experiments design to optimize ZAIS quantum dot behavior in the future. Begin with some puried ZAIS quantum dots as prepared according to the protocol in section 4.2. Clean two glass coverslips in a methanol wash, and then dry them out on a clean hotplate surface at 55 C. Take the quantum dots (which should be in chloroform) and pipette an entire 1.5 mL sample onto the heated, cleaned coverslip, 100 l at a time. The set point of 55 C is important as it is just below the chloroform boiling point of 61:2 C. Therefore, the quantum dot solvent should evaporate quickly, but not violently, and a smooth, uniform lm of quantum dots is produced on the glass coverslip. Take the second glass coverslip and attach it to the rst one at the sides with some adhesive. The glass coverslip samples are then ready for a measurement. A uorescent microscope with a laser or LED source at the excitation wavelength of the quantum dots is needed. I 81 Figure 4.7: Intensity trace for a quantum dot recorded in a video on a uorescent microscope. This demonstrates proof of concept for measuring ZAIS blinking. The clear on/o behavior is seen at about 50 s and then 100 s. Excitation wavelength: 480 nm, emission wavelength: 560 nm. used a GE DeltaVision OMX 4 with a 480 nm laser source, and I set the emission lter to 560 nm, based on the excitation and emission curves I recorded for that particular sample (an example of which is shown in g. 4.2). I then acquired video data at dierent frame times - typically 50 ms to 500 ms, with the gain and laser power settings adjusted depending on the sample. Since each sample is dierent, the gain and laser power settings need to be adapted to the specic sample under test in order to optimize signal to noise. This takes some practice. With the videos acquired, some processing needs to be done to identify the pixels which contain blinking quantum dots. This is easy to do by eye, but dicult to do in an automated way, especially for noisy videos. It is tantamount to the problem of machine peak recognition. I wrote a manual tool in the Python programming language to identify the peaks, but future researchers may nd it advantageous to apply simple computer vision techniques to these datasets to automate quantum dot location nding. Once the quantum dot has been identied, its intensity vs time curve can be produced 82 (g. 4.7) by proceeding through all of the frames in the video in order, and measuring the pixel intensity value for the pixel that has been associated with the quantum dot. This has been done in g. 4.7 for one particular quantum dot that was identied manually. To produce the P (t off ) curve for an ensemble of quantum dots, this process will need to be repeated over every quantum dot that can be identied in the sample. Future works will focus on the applications of this process, such as by using it to optimize ZAIS quantum dots for biological imaging agents. 4.8 Conclusion In closing, this chapter presented background on a novel type of quantum dot material I and my lab mates studied in the later half of my tenure at USC. I performed a study to investigate the photophysics behind luminescence in ZAIS quantum dots, specically. This study produced and characterized ZAIS quantum dots in order to probe the origin of their photoluminescence experimentally. A quick-cooled thermal annealing treatment was used to manipulate the degree of lattice disorder present in the samples. To parameterize this order, the Urbach energy was measured by recording the tail of the absorption spectrum band edge, and it was found that higher temperature-treated samples actually had higher degrees of disorder, despite the recrystallization hypothesis in past works. Photoluminescence quenching was still observed, indicating that the origin of the defect must be thermal in nature (e.g., not foreign impurities). To conrm that this quenching eect was electronic in nature, UPS measurements were performed, showing the appearance of a band shoulder for samples treated at higher temperatures. Finally, TRPL decay measurements suggest the possibility of an alternate recombination pathway which is nonradiative, providing an alternative mechanism for quenching than simple defect concentration loss. Further work will be continued by my lab mates to investigate and optimize the blinking behavior of this useful class of core shell nanomaterials. 83 References 1 S. Jun, J. Lee, and E. Jang, \Highly Luminescent and Photostable Quantum Dot{Silica Monolith and Its Application to Light-Emitting Diodes", ACS Nano 7, Publisher: American Chemical Society, 1472{1477 (2013). 2 X. Zhang, D. Jia, C. H agglund, V. A. Oberg, J. Du, J. Liu, and E. M. J. Johansson, \Highly photostable and ecient semitransparent quantum dot solar cells by using solution-phase ligand exchange", en, Nano Energy 53, 373{382 (2018). 3 L. Y. Lee, S. L. Ong, J. Y. Hu, W. J. Ng, Y. Feng, X. Tan, and S. W. Wong, \Use of Semiconductor Quantum Dots for Photostable Immuno uorescence Labeling of Cryptosporidium parvum", en, Applied and Environmental Microbiology 70, Publisher: American Society for Microbiology Section: METHODS, 5732{5736 (2004). 4 M. Samadpour, A. Irajizad, N. Taghavinia, and M. Molaei, \A new structure to increase the photostability of CdTe quantum dot sensitized solar cells", en, Journal of Physics D: Applied Physics 44, Publisher: IOP Publishing, 045103 (2011). 5 S. Cho, S. Jung, S. Jeong, J. Bang, J. Park, Y. Park, and S. Kim, \Strategy for Synthesizing Quantum Dot-Layered Double Hydroxide Nanocomposites and Their Enhanced Photoluminescence and Photostability", Langmuir 29, Publisher: American Chemical Society, 441{447 (2013). 6 Y. Cao, A. Stavrinadis, T. Lasanta, D. So, and G. Konstantatos, \The role of surface passivation for ecient and photostable PbS quantum dot solar cells", en, Nature Energy 1, Number: 4 Publisher: Nature Publishing Group, 1{6 (2016). 7 C. V. V. M. Gopi, M. Venkata-Haritha, Y.-S. Lee, and H.-J. Kim, \ZnO nanorods decorated with metal suldes as stable and ecient counter-electrode materials for high-eciency quantum dot-sensitized solar cells", en, Journal of Materials Chemistry A 4, Publisher: The Royal Society of Chemistry, 8161{8171 (2016). 84 8 V.-D. Dao, Y. Choi, K. Yong, L. L. Larina, and H.-S. Choi, \Graphene-based nanohybrid materials as the counter electrode for highly ecient quantum-dot-sensitized solar cells", en, Carbon 84, 383{389 (2015). 9 M. Pelton, C. Santori, J. Vuckovi c, B. Zhang, G. S. Solomon, J. Plant, and Y. Yamamoto, \Ecient Source of Single Photons: A Single Quantum Dot in a Micropost Microcavity", Physical Review Letters 89, Publisher: American Physical Society, 233602 (2002). 10 J. Claudon, J. Bleuse, N. S. Malik, M. Bazin, P. Jarennou, N. Gregersen, C. Sauvan, P. Lalanne, and J.-M. G erard, \A highly ecient single-photon source based on a quantum dot in a photonic nanowire", en, Nature Photonics 4, Number: 3 Publisher: Nature Publishing Group, 174{177 (2010). 11 Y. Kim, E. Yassitepe, O. Voznyy, R. Comin, G. Walters, X. Gong, P. Kanjanaboos, A. F. Nogueira, and E. H. Sargent, \Ecient Luminescence from Perovskite Quantum Dot Solids", ACS Applied Materials & Interfaces 7, Publisher: American Chemical Society, 25007{25013 (2015). 12 S. Coe-Sullivan, \Quantum dot developments", en, Nature Photonics 3, Number: 6 Publisher: Nature Publishing Group, 315{316 (2009). 13 I. L. Medintz, H. T. Uyeda, E. R. Goldman, and H. Mattoussi, \Quantum dot bioconjugates for imaging, labelling and sensing", en, Nature Materials 4, Number: 6 Publisher: Nature Publishing Group, 435{446 (2005). 14 D. Mo, L. Hu, G. Zeng, G. Chen, J. Wan, Z. Yu, Z. Huang, K. He, C. Zhang, and M. Cheng, \Cadmium-containing quantum dots: properties, applications, and toxicity", en, Applied Microbiology and Biotechnology 101, 2713{2733 (2017). 15 B. A. Rzigalinski and J. S. Strobl, \Cadmium-containing nanoparticles: Perspectives on pharmacology and toxicology of quantum dots", en, Toxicology and Applied Pharmacology, New Insights into the Mechanisms of Cadmium Toxicity 238, 280{288 (2009). 85 16 C. Coughlan, M. Ib a~ nez, O. Dobrozhan, A. Singh, A. Cabot, and K. M. Ryan, \Compound Copper Chalcogenide Nanocrystals", Chemical Reviews 117, Publisher: American Chemical Society, 5865{6109 (2017). 17 H. Yu, J. Li, R. A. Loomis, P. C. Gibbons, L.-W. Wang, and W. E. Buhro, \Cadmium Selenide Quantum Wires and the Transition from 3D to 2D Connement", en, 2. 18 M. D. Regulacio, K. Y. Win, S. L. Lo, S.-Y. Zhang, X. Zhang, S. Wang, M.-Y. Han, and Y. Zheng, \Aqueous synthesis of highly luminescent AgInS2{ZnS quantum dots and their biological applications", en, Nanoscale 5, 2322 (2013). 19 A. S. Fuhr, H. J. Yun, N. S. Makarov, H. Li, H. McDaniel, and V. I. Klimov, \Light Emission Mechanisms in CuInS 2 Quantum Dots Evaluated by Spectral Electrochemistry", en, ACS Photonics 4, 2425{2435 (2017). 20 Y. Matsuda, T. Torimoto, T. Kameya, T. Kameyama, S. Kuwabata, H. Yamaguchi, and T. Niimi, \ZnS{AgInS2 nanoparticles as a temperature sensor", en, Sensors and Actuators B: Chemical 176, 505{508 (2013). 21 H. C. Yoon, J. H. Oh, M. Ko, H. Yoo, and Y. R. Do, \Synthesis and Characterization of Green Zn{Ag{In{S and Red Zn{Cu{In{S Quantum Dots for Ultrahigh Color Quality of Down-Converted White LEDs", ACS Applied Materials & Interfaces 7, Publisher: American Chemical Society, 7342{7350 (2015). 22 F. L. N. Sousa, D. V. Freitas, R. R. Silva, S. E. Silva, A. C. Jesus, H. S. Mansur, W. M. Azevedo, and M. Navarro, \Tunable emission of AgIn5S8 and ZnAgIn5S8 nanocrystals: electrosynthesis, characterization and optical application", en, Materials Today Chemistry 16, 100238 (2020). 23 T. Torimoto, S. Ogawa, T. Adachi, T. Kameyama, K.-i. Okazaki, T. Shibayama, A. Kudo, and S. Kuwabata, \Remarkable photoluminescence enhancement of ZnS{AgInS 2 solid solution nanoparticles by post-synthesis treatment", en, Chemical Communications 46, Publisher: Royal Society of Chemistry, 2082{2084 (2010). 86 24 S. H ufner, Photoelectron Spectroscopy: Principles and Applications, en, Google-Books-ID: f6nvCAAAQBAJ (Springer Science & Business Media, Mar. 2013). 25 A. Raevskaya, O. Rozovik, A. Novikova, O. Selyshchev, O. Stroyuk, V. Dzhagan, I. Goryacheva, N. Gaponik, D. R. T. Zahn, and A. Eychm uller, \Luminescence and photoelectrochemical properties of size-selected aqueous copper-doped Ag{In{S quantum dots", en, RSC Advances 8, Publisher: Royal Society of Chemistry, 7550{7557 (2018). 26 G. Cody, \Urbach edge of crystalline and amorphous silicon: a personal review", en, Journal of Non-Crystalline Solids 141, 3{15 (1992). 27 N. Ghobadi, \Band gap determination using absorption spectrum tting procedure", en, International Nano Letters 3, 2 (2013). 28 T. Chevallier, A. Benayad, G. L. Blevennec, and F. Chandezon, \Method to determine radiative and non-radiative defects applied to AgInS 2 {ZnS luminescent nanocrystals", en, Physical Chemistry Chemical Physics 19, Publisher: Royal Society of Chemistry, 2359{2363 (2017). 29 H. Zang, H. Li, N. S. Makarov, K. A. Velizhanin, K. Wu, Y.-S. Park, and V. I. Klimov, \Thick-Shell CuInS 2 /ZnS Quantum Dots with Suppressed \Blinking" and Narrow Single-Particle Emission Line Widths", en, Nano Letters 17, 1787{1795 (2017). 30 O. Stroyuk, F. Weigert, A. Raevskaya, F. Spranger, C. W urth, U. Resch-Genger, N. Gaponik, and D. R. T. Zahn, \Inherently Broadband Photoluminescence in Ag{In{S/ZnS Quantum Dots Observed in Ensemble and Single-Particle Studies", The Journal of Physical Chemistry C 123, Publisher: American Chemical Society, 2632{2641 (2019). 31 A. L. Efros and D. J. Nesbitt, \Origin and control of blinking in quantum dots", en, Nature Nanotechnology 11, Number: 8 Publisher: Nature Publishing Group, 661{671 (2016). 87 32 B. Mahler, P. Spinicelli, S. Buil, X. Quelin, J.-P. Hermier, and B. Dubertret, \Towards non-blinking colloidal quantum dots", en, Nature Materials 7, Number: 8 Publisher: Nature Publishing Group, 659{664 (2008). 33 J. Tang and R. A. Marcus, \Mechanisms of uorescence blinking in semiconductor nanocrystal quantum dots", The Journal of Chemical Physics 123, Publisher: American Institute of Physics, 054704 (2005). 34 B. Ji, E. Giovanelli, B. Habert, P. Spinicelli, M. Nasilowski, X. Xu, N. Lequeux, J.-P. Hugonin, F. Marquier, J.-J. Greet, and B. Dubertret, \Non-blinking quantum dot with a plasmonic nanoshell resonator", en, Nature Nanotechnology 10, Number: 2 Publisher: Nature Publishing Group, 170{175 (2015). 35 G. Yuan, D. E. G omez, N. Kirkwood, K. Boldt, and P. Mulvaney, \Two Mechanisms Determine Quantum Dot Blinking", ACS Nano 12, Publisher: American Chemical Society, 3397{3405 (2018). 36 T. D. Krauss and J. J. Peterson, \A charge for blinking", en, Nature Materials 11, 14{16 (2012). 37 C. Galland, Y. Ghosh, A. Steinbr uck, J. A. Hollingsworth, H. Htoon, and V. I. Klimov, \Lifetime blinking in nonblinking nanocrystal quantum dots", en, Nature Communications 3, Number: 1 Publisher: Nature Publishing Group, 908 (2012). 38 A. Boulesbaa, A. Issac, D. Stockwell, Z. Huang, J. Huang, J. Guo, and T. Lian, \Ultrafast Charge Separation at CdS Quantum Dot/Rhodamine B Molecule Interface", Journal of the American Chemical Society 129, Publisher: American Chemical Society, 15132{15133 (2007). 39 J. Huang, K. L. Mulfort, P. Du, and L. X. Chen, \Photodriven Charge Separation Dynamics in CdSe/ZnS Core/Shell Quantum Dot/Cobaloxime Hybrid for Ecient Hydrogen Production", Journal of the American Chemical Society 134, Publisher: American Chemical Society, 16472{16475 (2012). 88 40 Y. Wang, X. Liu, J. Liu, B. Han, X. Hu, F. Yang, Z. Xu, Y. Li, S. Jia, Z. Li, and Y. Zhao, \Carbon Quantum Dot Implanted Graphite Carbon Nitride Nanotubes: Excellent Charge Separation and Enhanced Photocatalytic Hydrogen Evolution", en, Angewandte Chemie 130, eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1002/ange.201802014, 5867{5873 (2018). 41 E. Zillner, S. Fengler, P. Niyamakom, F. Rauscher, K. K ohler, and T. Dittrich, \Role of Ligand Exchange at CdSe Quantum Dot Layers for Charge Separation", The Journal of Physical Chemistry C 116, Publisher: American Chemical Society, 16747{16754 (2012). 42 H. Scher and E. W. Montroll, \Anomalous transit-time dispersion in amorphous solids", en, Physical Review B 12, 2455{2477 (1975). 89 Chapter 5 Applied projects & Future works 5.1 Introduction In this chapter, I will discuss two of the side projects I researched, which were applied device-level technologies rather than fundamental materials science projects. Roadblocks encountered and potential for future works will be discussed. 5.2 ZnO microlaser project: introduction Early on in my PhD, I focused on the synthesis and fabrication of devices using ZnO nanowires. Background on ZnO was given in section 2.3, but ZnO nanowires themselves are particularly unique. ZnO nanowires are a \1d" nanomaterial form of bulk ZnO, which are able to be grown chemically on dierent types of substrates. The lower dimensionality of this nanoparticle oers several advantages over using other forms, such as the directionality of the output emission [1{3], the electronic constraints in the two transverse directions [4, 5], and the resulting crystalline quality [6, 7]. The motivation for growing ZnO nanowires in particular was to develop a new type of ultraviolet laser. As described in section 2.3, ZnO is known for its UV-blue emission wavelength [8], attributable to the deep bond energy of the Zn-O in the crystal structure. The state of the art is that ultraviolet lasers exist but are dicult to miniaturize for several reasons [9]. Most ultraviolet lasers are large and powerful pulsed lasers that are used for engraving or machining of ceramics and metal [10]. Yet there are still industrial applications for small and low power ultraviolet lasers in the elds of biotechnology and communications 90 [11, 12]. I pursued this topic as my lab has signicant expertise in the eld of optical microresonators, and there might be some synergistic overlap between the directional UV emissive properties of ZnO nanowires and my lab's optical microresonators. Our resonators in particular were quite amenable to surface functionalization, as demonstrated in my PI's early works [13, 14]. The ability to drop in dierent kinds of materials and achieve synergistic eects with the resonator is a unique platform. 5.3 ZnO microlaser: laser background The central concept of a laser is the long optical path length that the light emitted by the laser has taken [15]. Similar to the light emitted by stars, the light emitted by a laser has taken a long trip to be able to each you. This long optical path length enables the output light to have useful properties, such as temporal coherence and beam directionality. But unlike a star, the light emitted by a laser has been traveling through a gain medium the entire trip, and so it's also intense. Any new laser platform will need this fundamental long path length component, and there are a few ways of achieving that with the resonator-ZnO system. 5.4 ZnO microlaser: resonant cavities The most straightforward way to create long optical path lengths in a laser is to include an optical resonator cavity. The resonator cavity is essentially like a box with mirrors on the inside (g. 5.1). A fundamental property of the laser is its coherence, both in a temporal sense and a spatial sense. A resonator with long optical path length helps to achieve both. Temporal coherence, also known as chromatic coherence, is the degree to which the light emitted by the laser is monochromatic. Lasers are unlike any other tool in that they emit intense monochromatic light. Before the laser, the best way of achieving temporally-coherent light was by ltering out all the undesirable wavelengths, which throws away a lot of intensity. Lasers get around this by selectively amplifying the single wavelength that is permitted to 91 resonate in the resonator cavity. The cavity geometry is the main factor in determining which wavelengths can resonate. Consider a cavity of length L. If there are at mirrors on each side of the cavity, then the phase accumulated due to the optical path length of one trip through the cavity is = nL (5.1) Where n is the index of the cavity, and is the wavelength of the light. If L = m 2 n , where m is an integer, then the re ected waves will constructively add only for waves of wavelength . For many round trips in the cavity, non- waves will be suppressed, leading to temporal coherence [15]. Spatial coherence is the degree to which two rays on some given wavefront position have a xed phase relationship. Similar to the technique of wavelength ltering to achieve temporal coherence, spatial coherence can be achieved by sending the light emitted by some optical source through a very long distance. The light from stars has to travel a very long distance to reach us on Earth, and therefore the light we receive is extremely uniform: all rays sent o at dierent angles from the star failed to reach Earth. A resonant laser cavity mimics this by using two mirrors to create the large optical path length. Any rays that have even the slightest angle will have that angle amplied by the mirrors, sending them on a collision course with the absorptive sidewalls. This is detailed schematically in g. 5.1. Since no mirror is perfect, some of the light will emit with each re ection o of the mirrors. Part of that emission is used as the output of the laser [15]. 5.5 ZnO microlaser: whispering gallery mode resonators My lab specialized in a type of optical microresonator known as a whispering gallery mode microresonator. These devices usually have a three-dimensional spherical or toroidal geom- etry, shown in g. 5.2. 92 Figure 5.1: Cartoon schematic of a simple laser cavity. Spatial coherence is achieved through the long optical path length: any ray with a slight angle would be ltered via the many round trips the light takes between the mirrors in the cavity. Temporal coherence is also achieved through the long optical path length: any ray with the wrong wavelength would suer from self interference via the many round trips in the cavity. Figure 5.2: Schematic and SEM image of a toroid microresonator. Light circulates in the rim of the suspended toroid, which has a chance to interact with the functional material coating on top of the device. In the SEM image, a ZnO nanowire coating was attempted, but only achieved partial coverage. 93 Although it looks quite dierent from the \box with mirrors" resonator design, the work- ing principle is still similar. Instead of using high-re ective coated glass for mirrors, the whispering gallery mode resonator relies on the principle of total internal re ection [16, 17]. Total internal re ection is the automatic re ection of oblique rays of light o of a high index to low index dielectric boundary. The natural curvature of the sphere or toroid resonator ensures that the light traveling within the resonator is always incident at an oblique angle, and thus total internal re ection keeps the light contained within the resonator. The con- dition of light interference still applies, and this forms the basis for optical modes that are allowed to circulate in the cavity [16, 17]. The exact frequencies of these optical modes are a function of the cavity geometry, which, for most systems, must be solved numerically or computationally. The performance of a resonator can be described quantitatively by a gure of merit called the quality factorQ. The quality factor is a general property of resonator systems and is not unique to whispering gallery mode resonators. However, the denition varies from system to system. For whispering gallery mode resonators, we dene the quality factor Q as: Q =! (5.2) where ! is the angular frequency of the circulating light, and is the cavity ringdown time [18]. The total quality factor is a function of several loss mechanisms [19], but the most relevant one in this context is the loss due to surface scattering sites: Q 1 Q 1 scat (5.3) with all other loss factors held constant [19]. This can be measured experimentally in a couple of dierent ways, which are beyond the scope of this thesis (for details, see [18]). The main purpose of bringing this up is that Q is dependent on several resonator properties, one of them being the number of surface 94 scattering sites. If the surface has many scattering sites,Q 1 scat in eq. (5.3) drops signicantly, and soQ drops signicantly, which by eq. (5.2) implies that light is circulating for less time in the resonator, and therefore the intensity in the resonator is lower. This is a signicant problem in attempting to build a ZnO nanowire microlaser on this platform, discussed in section 5.6. 5.6 ZnO microlaser: resonator-material interactions The unique aspect of the whispering gallery mode microresonators are that they are easily amenable to surface material functionalization (g. 5.2). Because the \mirror" containing the light is the surface boundary of the resonator itself, the light circulating in the resonator can be easily interacted with by changing the surface conditions of the resonator. Total internal re ection is not so total; when a light wave undergoes total internal re ection o the surface of a dielectric boundary, there is always an exponentially decaying optical eld very near to the surface. This eld transmits no power out of the cavity unless another dielectric material is brought close to it. When that's the case, a small amount of optical power may \tunnel" through the decaying region and into the second material [19{22]. To understand evanescent elds, consider a monochromatic plane wave: U(~ r;t) = ~ E exp(j ~ k~ r) exp(j(!t + 0 )) (5.4) Where ~ E is the electric eld amplitude and polarization vector, ~ k is the wave vector, ! is the angular frequency of the light, and 0 is an arbitrary phase oset. The expression in eq. (5.4) is a solution to the optical wave equation [cite] if the following holds: k z = q k 2 k 2 x k 2 y (5.5) where 95 k = 2n (5.6) and k x = ^ x ~ k (5.7) k y = ^ y ~ k (5.8) For a wave propagating in the ^ z direction, where ^ x and ^ y represent a pair of transverse direction vectors. In cases where k 2 < (k 2 x +k 2 y ), k z becomes non-real and thus the wave solution in eq. (5.4) picks up an exponentially-decaying factor, since ~ k~ r is no longer real: U(~ r;t) = ~ E exp(jjj ~ kj( ^ k~ r)) exp(j(!t + 0 )) (5.9) U(~ r;t) = ~ E exp(j ~ kj( ^ k~ r)) exp(j(!t + 0 )) (5.10) The expression in eq. (5.10) is an evanescent wave. The decay constantj ~ kj( ^ k~ r) determines the length of the evanescent tail. Shorter wavelengths have longer evanescent tails sincej ~ kj increases. Thus for the case of a ZnO microlaser, there is the tradeo between material interaction and ideal operating wavelength. The optical circulating intensity in the whispering gallery mode resonators can be quite high, on the order of 1 GW/cm 2 [18]. Therefore, if such a high-quality resonator is coated in a material that can interact with the type of light circulating in the cavity, interactions only seen at high intensities of light may be achieved [19]. Such material interactions are called nonlinear optical eects, as their intensity scales nonlinearly with the intensity of the source light and are nonexistent at lower optical source intensities. This is an entire eld of study and so the scope of this section will be specically on one nonlinear optical phenomena that 96 is relevant to our system. That phenomenon is called two photon absorption [23, 24]. The main diculty with using whispering gallery mode microresonators as opposed to the traditional \mirrors in a box" resonator is that the whispering gallery mode cavity is made of some high index material, such as silica. This has its advantages and disadvantages. The main disadvantage it poses is that the cavity itself constrains the type of light that may circulate in it. Silica is transparent in some regions of the optical spectrum, but its transparency drops o hard below the visible range, meaning that it is not viable to have a high-Q resonator for shorter wavelengths, especially those below the visible range. The material loss of silica becomes too signicant and the light will not propagate for long inside the cavity. There are also numerous other diculties involved, such as injecting the light into the toroid, and the fact that the toroid becomes dicult to fabricate [18]. With these constraints in mind, there must be a way to create the type of ultraviolet pump light that ZnO requires. The ZnO itself can be leveraged for this purpose. One of the unique features of ZnO nanowires in particular is that they have a high two photon absorption coecient. Two photon absorption, simply put, is the ability to excite a transition normally caused by one photon, but with two photons instead (g. 5.3). This only happens at very high optical intensities, because the two photons have to be absorbed at the right time and position. The two-photon transition is overall very unlikely to happen, but it is signicantly more likely in ZnO nanowires than in other material systems [25, 26]. Silica microtoroid resonators produced by my lab have been demonstrated to eciently circulate 750 nm light [cite]. ZnO's peak band gap absorption (single photon) is at 375 nm [8, 27]. Coincidentally, 375 nm is half of 750 nm, and so two 750 nm photons have the same energy as one 375 nm photon. Combining the high two photon absorption coecient of the ZnO with the high circulating intensity of the 750 nm light in the microtoroid resonator, it is theoretically possible that the circulating 750 nm light can transfer some signicant power into a ZnO nanowire array grown on the surface of the toroid, triggering two photon emission events and therefore ultraviolet emission. Therefore, in the quest to produce an ultraviolet 97 Figure 5.3: Diagram of the electronic state changes occurring in two photon absorption. Although the same excitation can be achieved via two photon absorption as in normal (single photon) absorption, it is much less likely to occur. microlaser, some of the fundamental components needed are there: the microtoroid circulat- ing 750 nm light serves as the optical pump source for the laser, and the ZnO nanowires on the surface do the initial conversion to 375 nm. 5.7 ZnO microlaser: random lasing The remaining piece of this is the gain medium. That single 375 nm photon produced at the two-photon absorption rate is not a laser, it needs amplication through a long optical path length. Random lasing is a phenomenon that has been leveraged to create ZnO nanowire laser arrays in the past. The fundamental principle of random lasing is that the optical path length of the pump light (the converted 375 nm photon) becomes extremely long due to scattering inside of a dense, disordered medium, as illustrated in g. 5.4. If the ZnO nanowire array can be grown suciently dense on the surface of the microtoroid resonator, it can trap the light, forcing the light to interact with many ZnO nanowires before nally being emitted. 98 Figure 5.4: Cartoon schematic of the random lasing process. Scattering o of dense ZnO nanowire array structures allows for the optical path length to accumulate, while the ampli- cation comes from a population inversion stimulated in the ZnO nanowires. This has been demonstrated for at arrays that are directly pumped via other lasers [28{30]. A few things need to go right for random lasing to cause amplication of the 375 nm photons. The rst is that there needs to be a population inversion achieved in the ZnO nanowires. A population inversion is the state of a semiconducting material where a sucient quantity of the carriers is in the proper excited state [31]. For ZnO, these carriers trigger the emission of a 375 nm photon upon relaxation into their ground state, and that event can be catalyzed by the scattering of another 375 nm photon, causing an avalanche emission event which produces a lot of very similar 375 nm photons. For this to happen, enough two photon absorption needs to have occurred to put the ZnO electrons in their excited state prior to the beginning of emission. There is some study on this in the eld [28{30] and ZnO lasing has been achieved via more direct pump processes. 5.8 ZnO microlaser: challenges & future works The main challenge I encountered in this work was the process of fabricating a high quality microresonator with ZnO deposited on the surface. The quality factor of the resulting optical microresonator is highly sensitive to the surface conditions. Any deviations from a perfectly smooth and curved resonator surface produces scattering sites for the photons circulating in 99 Figure 5.5: SEM images of a toroid subject to hydrothermal ZnO nanowire growth. I was able to accomplish uniform coverage, but the process introduced too much byproduct deposition on the surface of the toroid. (A) Zoomed out, (B) close up view of the nanowires on the side of the toroid. the pump, producing scattering loss, and decreasing the circulating optical intensity inside the whispering gallery mode resonator. This is seemingly in direct contrast with the goal of having a dense ZnO nanowire array as required for the random lasing in section 5.7. I was able to achieve uniformly coated ZnO nanowire toroid resonators (g. 5.5). However, experimentally, I found that most of the problem with the scattering loss in microtoroid resonators coated with ZnO came from the synthesis byproducts, not the ZnO nanowires themselves. I observed this through direct resonator testing and extensive SEM imaging. I had a method to synthesize ZnO nanowire arrays that ranged from dense to very sparse and found that the resulting quality factor of the resonator did not seem to scale with the ZnO nanowire array density. The real damage to the resonator came from the excess chemical byproducts coating the surface of the resonator, as the procedure I used (described in section 5.9) was a wet hydrothermal method. Improvements to this project may have to come from using an alternative synthetic method for producing ZnO nanowires on microtoroids. Possible ideas include molecular beam epitaxial growth of the ZnO nanowires [32] and low temperature dry thermal syntheses 100 [33]. 5.9 ZnO microlaser: synthesis and characterization This method was not successful at producing a viable ZnO nanowire toroid, but it was successful at producing ZnO nanowire arrays (even on curved surfaces), and so I will describe it here for posterity. Hydrothermal synthesis of ZnO nanowires following seed-based protocols from the literature [34, 35]. First, a "seed" solution was prepared, to provide nucleation sites for the ZnO nanowires on a substrate. Substrate preparation: clean a 1 cm by 1 cm silicon-SiO 2 substrate with a simple solvent wash: rinse the substrate in acetone, methanol, and isopropanol. Proceed to dry the sample out with compressed air. Using a pipette, wet the substrate with a drop or two of zinc acetate solution in ethanol (1.18 g zinc acetate dihydrate per 100 g absolute ethanol). Rinse the wet substrate with clean ethanol after 10 seconds, and then blow dry with compressed air. Repeat this process ve times for uniform substrate coverage with the ZnO nucleation sites. The growth solution was prepared by mixing 2.8 g of HMTA (Millipore Sigma, 99.9%) and 5.94 g of Zn nitrate hexahydrate in 200 mL of deionized water. This solution must be mixed for three hours at room temperature. The seeded substrates were then placed in the growth solutions, and put into a 90 C oven overnight. After the growth period, the silicon substrates were pulled out of the growth vial, and gently rinsed with deionized water. After drying out, they were ready for charac- terization. If a growth was being attempted on a toroid array, the toroid chip was used in place of the blank silicon chip described earlier. This concludes my work the ZnO microlaser project, which may be picked up by others. The other side project I proposed as a future work was relating to my work on QHMF quantum dots. 101 5.10 QHMF LED device: introduction One of the applications to my work in chapter 4 is applying the ZAIS quantum dots in a real LED device. In chapter 4, I discussed in detail the materials science behind QHMF emitters and a particular subclass of them ZAIS. I would now like to shift the focus and propose in detail a series of projects focused on engineering and application of ZAIS, and more generally, QHMF quantum dots. QHMF quantum dots have an important practical application in LED technology. In this section, I will discuss a project for future works based on an application of QHMF technology that utilizes the extremely wide spectral tunability window of these quantum dots. This combination of heavy-metal-free materials and the wide spectral tunability makes it ideal for LED devices, having high technological impact. 5.11 QHMF LED device: emissive layers in LED devices I would like to improve the emissive qualities of QHMF quantum dots using three main strategies. The rst strategy is to improve the color saturation of the QHMF quantum dots by decreasing the ensemble linewidth. Recent studies have shown that although an ensemble of QHMF quantum dots has a signicantly worse PL linewidth than competing quantum dot materials, this is not a fundamental property of QHMF quantum dots. Experiments on single QHMF quantum dot emission has shown that the majority of the PL linewidth can be attributed to poor ensemble uniformity [36], as single-particle emission is much narrower. This is thought to be caused by poor size uniformity as produced by the various synthetic techniques, such as the hot injection or heat-up methods [37], a wide distribution of Cu- related defect states in the particles, and a split between the oxidation states of Cu in dierent QHMF quantum dots [38, 39]. Defect states are an important vector for emissive control in QHMF quantum dots as established in chapter 4. The second strategy is to experiment with various treatments to control defect levels in the synthesized quantum dots. In particular, in CuInS 2 quantum dots, 102 the channel by which radiative recombination occurs has been determined to be related to the recombination of an electron in the conduction band with a localized hole trapped on a Cu-related defect in the middle of the band gap [39]. A similar mechanism is responsible for emission in the ZAIS quantum dots described in chapter 4. Although the exact photophysics behind recombination in QHMF quantum dots like this are still greatly debated, various groups have demonstrated control over the luminescence of these materials by applying certain treatments, such as electrochemical redox treatments, growth under Cu-deprived conditions, and doping [36{39]. Specically, I would like to investigate applying the recent advances in theoretical understanding of the QHMF quantum dot photophysics to optimize emissive properties of QHMF quantum dots in an LED device. I have already done some work to probe the mechanism responsible for this eect (chapter 4) and found it to be a thermal- related defect, yet more work remains to optimize the emission for tangible behaviors like color purity. Other advances have been made in the sophisticated eld of Cd-based LED technology, and may be modied to advance QHMF quantum dot LED devices. Two such examples are in optimization of the passivating inorganic shell, rather than optimization of the quantum dot core. Traditionally, CdS and CdSe are grown with a thin (less than 10 monolayers) shell of ZnS or ZnSe, which serves to chemically protect the emissive core, and passivate surface states caused by dangling bonds o of the Cd-based core [40]. This is important, as photochemical eects can severely impact the lifetime of the Cd-based core without the protective layer, and surface states can create new recombination channels that will degrade the color saturation and eciency of the emission. However, this also introduces new problems. Lattice mismatch between the core and shell materials can induce strain-related defects into the core, lowering the quality of the emission. Furthermore, the electronic structure and thickness of the shell needs to be considered, as it acts as a potential transport layer or barrier to carriers in a charge-injection device. The shell can also improve the eciency of the quantum dot emission by preventing FRET between parasitic transport materials in the LED device, 103 and the emission material [41]. Thus the third strategy for improving QHMF quantum dot emission in this project direction is to investigate potential improvements to be made in the shell. In particular, I will explore the growth of an inorganic buer layer to enhance lattice matching between the shell and the core, reducing strain, and thus reducing the number of strain-related defects in the core material. I will also seek to optimize the shell thickness to minimize FRET between neighboring quantum dot emitters, and neighboring layers. 5.12 QHMF LED device: device-level improvements Several improvements can also be made at the device level. The typical LED device archi- tecture consists of a sandwich of heterogeneous layers designed to eciently inject charge carriers into the device, and transport them to the radiative emission layer (EML), where light is generated and coupled out. Advances in device architecture have been primarily made for Cd-based quantum dots as the EML material, and there is opportunity in modify- ing these advancements to suit LEDs with QHMF quantum dots as the EML material. First, I would like to investigate the thickness and positioning of the EML in a QHMF quantum dot-based device. These properties dramatically impact the external quantum e- ciency (EQE) of the device [41, 42], which is due to a combination of issues, such as charging and parasitic FRET. Charging of the EML happens due to a band misalignment introduced by the hole transport layer (HTL). Brie y, due to the high position of the highest occupied molecular orbital (HOMO) level in typical HTL materials, with respect to the valence band of the EML, holes must overcome an energy barrier in order to reach the active EML re- gion. Electrons do not pay the same price, leading to an unbalanced charge transport into the EML. When there are more electrons than holes (or vice versa), Auger recombination becomes likely [40, 43]. Auger recombination is a type of nonradiative recombination involv- ing two electrons and one hole, in which the energy of recombination between one of the electrons and one of the holes gets transferred into exciting a hole deeper into the valence band, rather than transferred into generation of a photon. Furthermore, parasitic FRET has 104 been reported to occur between quantum dots, and between the EML and other layers of the device [44]. FRET is a distance-dependent energy transfer mechanism caused by overlap between the absorbance and uorescence of two materials in proximity. I will optimize the thickness and positioning of the EML in order to minimize charge buildup on the EML, and minimize parasitic FRET between quantum dots and adjacent charge transfer layers. Second, I would like to investigate opportunities to improve the materials selection for the HTL. As mentioned previously, the high HOMO levels of HTL materials typically used in quantum dot LED devices is problematic, as it introduces an energy barrier for transfer of holes into the EML. In contrast, the best ETL material (ZnO quantum dots) has no such penalty for injecting electrons into the EML, leading to a charge imbalance, and thus reduced photon generation eciency due to the competing Auger recombination. As the nal layer deposited is the HTL, the standard today is an organic HTL, due to its delicate deposition technique. Inorganic materials with better band alignment have been used in the past, but required harsh sputtering deposition processes, which destroyed the quality of the layers underneath [40]. I would like to investigate solution-processable quantum dot-based HTL materials, and design a fabrication procedure that does not damage the layers underneath. This would improve band alignment and thus lead to a device specically designed for QHMF quantum dots, with a higher EQE than reported before. 5.13 Conclusion In closing, future works include a series of projects to investigate improving the eciency and color saturation of heavy-metal-free QHMF quantum dot LEDs. This work leverages my expertise in core-shell nanoparticle materials science, demonstrated in Chapters 1-5. My strategy to accomplish these goals involves taking advantage of recent signicant ad- vancements in fundamental understanding of the synthesis and recombination mechanisms in QHMF quantum dots. Furthermore, several improvements in device architecture can be made for QHMF quantum dot LEDs, to improve device eciency. If successful, these 105 projects will allow for better displays and biomedical diagnostics devices to be developed, thanks to the characteristic extremely wide spectral window of QHMF quantum dot LEDs. 106 References 1 L. K. van Vugt, S. R uhle, and D. Vanmaekelbergh, \Phase-Correlated Nondirectional Laser Emission from the End Facets of a ZnO Nanowire", Nano Letters 6, Publisher: American Chemical Society, 2707{2711 (2006). 2 J. C. Johnson, H. Yan, P. Yang, and R. J. Saykally, \Optical Cavity Eects in ZnO Nanowire Lasers and Waveguides", The Journal of Physical Chemistry B 107, Publisher: American Chemical Society, 8816{8828 (2003). 3 D. Vanmaekelbergh and L. K. van Vugt, \ZnO nanowire lasers", en, Nanoscale 3, 2783 (2011). 4 L. K. van Vugt, S. R uhle, P. Ravindran, H. C. Gerritsen, L. Kuipers, and D. Vanmaekelbergh, \Exciton Polaritons Conned in a ZnO Nanowire Cavity", en, Physical Review Letters 97, 147401 (2006). 5 J. Goldberger, D. J. Sirbuly, M. Law, and P. Yang, \ZnO Nanowire Transistors", en, The Journal of Physical Chemistry B 109, 9{14 (2005). 6 M.-C. Jeong, B.-Y. Oh, W. Lee, and J.-M. Myoung, \Comparative study on the growth characteristics of ZnO nanowires and thin lms by metalorganic chemical vapor deposition (MOCVD)", en, Journal of Crystal Growth 268, 149{154 (2004). 7 H. Y. Chao, J. H. Cheng, J. Y. Lu, Y. H. Chang, C. L. Cheng, and Y. F. Chen, \Growth and characterization of type-II ZnO/ZnTe core-shell nanowire arrays for solar cell applications", en, Superlattices and Microstructures, Proceedings of the 9th International Conference on Physics of Light-Matter Coupling in Nanostructures, PLMCN 2009 (Lecce - Italy) 47, 160{164 (2010). 8 A. Janotti and C. G. V. d. Walle, \Fundamentals of zinc oxide as a semiconductor", en, Reports on Progress in Physics 72, Publisher: IOP Publishing, 126501 (2009). 107 9 E. A. Seddon, J. A. Clarke, D. J. Dunning, C. Masciovecchio, C. J. Milne, F. Parmigiani, D. Rugg, J. C. H. Spence, N. R. Thompson, K. Ueda, S. M. Vinko, J. S. Wark, and W. Wurth, \Short-wavelength free-electron laser sources and science: a review", en, Reports on Progress in Physics 80, Publisher: IOP Publishing, 115901 (2017). 10 H. Bravo, B. T. Szapiro, P. W. Wachulak, M. C. Marconi, W. Chao, E. H. Anderson, C. S. Menoni, and J. J. Rocca, \Demonstration of Nanomachining With Focused Extreme Ultraviolet Laser Beams", IEEE Journal of Selected Topics in Quantum Electronics 18, Conference Name: IEEE Journal of Selected Topics in Quantum Electronics, 443{448 (2012). 11 N. Ponelies, J. Scheef, A. Harim, G. Leitz, and K. Greulich, \Laser micromanipulators for biotechnology and genome research", en, Journal of Biotechnology 35, 109{120 (1994). 12 P. R. Herman, R. S. Marjoribanks, A. Oettl, K. Chen, I. Konovalov, and S. Ness, \Laser shaping of photonic materials: deep-ultraviolet and ultrafast lasers", en, Applied Surface Science 154-155, 577{586 (2000). 13 A. M. Armani, R. P. Kulkarni, S. E. Fraser, R. C. Flagan, and K. J. Vahala, \Label-Free, Single-Molecule Detection with Optical Microcavities", en, Science 317, Publisher: American Association for the Advancement of Science Section: Report, 783{787 (2007). 14 A. M. Armani, A. Srinivasan, and K. J. Vahala, \Soft Lithographic Fabrication of High Q Polymer Microcavity Arrays", Nano Letters 7, Publisher: American Chemical Society, 1823{1826 (2007). 15 W. T. Silfvast, Laser Fundamentals, en, Google-Books-ID: x3VB2iwSaxsC (Cambridge University Press, Jan. 2004). 16 A. B. Matsko, A. A. Savchenkov, D. Strekalov, V. S. Ilchenko, and L. Maleki, \Review of Applications of Whispering-Gallery Mode Resonators in Photonics and Nonlinear Optics", en, 51. 108 17 A. Bozzola, S. Perotto, and F. De Angelis, \Hybrid plasmonic{photonic whispering gallery mode resonators for sensing: a critical review", en, The Analyst 142, 883{898 (2017). 18 D. Chen, Optical simulation and development of novel whispering gallery mode microresonators. 19 A. Kovach, D. Chen, J. He, H. Choi, A. H. Dogan, M. Ghasemkhani, H. Taheri, and A. M. Armani, \Emerging material systems for integrated optical Kerr frequency combs", EN, Advances in Optics and Photonics 12, Publisher: Optical Society of America, 135{222 (2020). 20 S. Soltani, V. M. Diep, R. Zeto, and A. M. Armani, \Stimulated Anti-Stokes Raman Emission Generated by Gold Nanorod Coated Optical Resonators", en, ACS Photonics 5, 3550{3556 (2018). 21 X. Shen, H. Choi, D. Chen, W. Zhao, and A. M. Armani, \Raman laser from an optical resonator with a grafted single-molecule monolayer", en, Nature Photonics 14, Number: 2 Publisher: Nature Publishing Group, 95{101 (2020). 22 J. Bures and R. Ghosh, \Power density of the evanescent eld in the vicinity of a tapered ber", EN, JOSA A 16, Publisher: Optical Society of America, 1992{1996 (1999). 23 M. Pawlicki, H. A. Collins, R. G. Denning, and H. L. Anderson, \Two-Photon Absorption and the Design of Two-Photon Dyes", en, Angewandte Chemie International Edition 48, eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1002/anie.200805257, 3244{3266 (2009). 24 J. Gea-Banacloche, \Two-photon absorption of nonclassical light", Physical Review Letters 62, Publisher: American Physical Society, 1603{1606 (1989). 25 C. Zhang, F. Zhang, T. Xia, N. Kumar, J.-i. Hahm, J. Liu, Z. L. Wang, and J. Xu, \Low-threshold two-photon pumped ZnO nanowire lasers", EN, Optics Express 17, Publisher: Optical Society of America, 7893{7900 (2009). 109 26 X. Wang, X. Wang, D. Yu, D. Yu, S. Xu, and S. Xu, \Determination of absorption coecients and Urbach tail depth of ZnO below the bandgap with two-photon photoluminescence", EN, Optics Express 28, Publisher: Optical Society of America, 13817{13825 (2020). 27 B. E. Sernelius, K.-F. Berggren, Z.-C. Jin, I. Hamberg, and C. G. Granqvist, \Band-gap tailoring of ZnO by means of heavy Al doping", en, Physical Review B 37, 10244{10248 (1988). 28 J. Fallert, R. J. B. Dietz, M. Hauser, F. Stelzl, C. Klingshirn, and H. Kalt, \Random lasing in ZnO nanocrystals", en, Journal of Luminescence, Special Issue based on The 15th International Conference on Luminescence and Optical Spectroscopy of Condensed Matter (ICL'08) 129, 1685{1688 (2009). 29 H. Y. Yang, S. P. Lau, S. F. Yu, A. P. Abiyasa, M. Tanemura, T. Okita, and H. Hatano, \High-temperature random lasing in ZnO nanoneedles", Applied Physics Letters 89, Publisher: American Institute of Physics, 011103 (2006). 30 X. Ma, J. Pan, P. Chen, D. Li, H. Zhang, Y. Yang, and D. Yang, \Room temperature electrically pumped ultraviolet random lasing from ZnO nanorod arrays on Si", EN, Optics Express 17, Publisher: Optical Society of America, 14426{14433 (2009). 31 J. Hecht, \The laser guidebook", English, Publisher: McGraw-Hill,New York, NY (1986). 32 Y. W. Heo, L. C. Tien, Y. Kwon, D. P. Norton, S. J. Pearton, B. S. Kang, and F. Ren, \Depletion-mode ZnO nanowire eld-eect transistor", Applied Physics Letters 85, Publisher: American Institute of Physics, 2274{2276 (2004). 33 S. C. Lyu, Y. Zhang, H. Ruh, H.-J. Lee, H.-W. Shim, E.-K. Suh, and C. J. Lee, \Low temperature growth and photoluminescence of well-aligned zinc oxide nanowires", en, Chemical Physics Letters 363, 134{138 (2002). 110 34 L. E. Greene, M. Law, J. Goldberger, F. Kim, J. C. Johnson, Y. Zhang, R. J. Saykally, and P. Yang, \Low-Temperature Wafer-Scale Production of ZnO Nanowire Arrays", Angewandte Chemie 115, eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1002/ange.200351461, 3139{3142 (2003). 35 C. P. Tsangarides, H. Ma, and A. Nathan, \ZnO nanowire array growth on precisely controlled patterns of inkjet-printed zinc acetate at low-temperatures", en, Nanoscale 8, 11760{11765 (2016). 36 H. Zang, H. Li, N. S. Makarov, K. A. Velizhanin, K. Wu, Y.-S. Park, and V. I. Klimov, \Thick-Shell CuInS 2 /ZnS Quantum Dots with Suppressed \Blinking" and Narrow Single-Particle Emission Line Widths", en, Nano Letters 17, 1787{1795 (2017). 37 B. Chen, N. Pradhan, and H. Zhong, \From Large-Scale Synthesis to Lighting Device Applications of Ternary I{III{VI Semiconductor Nanocrystals: Inspiring Greener Material Emitters", en, The Journal of Physical Chemistry Letters 9, 435{445 (2018). 38 A. S. Fuhr, H. J. Yun, N. S. Makarov, H. Li, H. McDaniel, and V. I. Klimov, \Light Emission Mechanisms in CuInS 2 Quantum Dots Evaluated by Spectral Electrochemistry", en, ACS Photonics 4, 2425{2435 (2017). 39 W. van der Stam, M. de Graaf, S. Gudjonsdottir, J. J. Geuchies, J. J. Dijkema, N. Kirkwood, W. H. Evers, A. Longo, and A. J. Houtepen, \Tuning and Probing the Distribution of Cu+ and Cu2+ Trap States Responsible for Broad-Band Photoluminescence in CuInS2 Nanocrystals", ACS Nano 12, Publisher: American Chemical Society, 11244{11253 (2018). 40 Y. E. Panl, M. Oded, and U. Banin, \Colloidal Quantum Nanostructures: Emerging Materials for Display Applications", en, Angewandte Chemie International Edition 57, eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1002/anie.201708510, 4274{4295 (2018). 111 41 P. O. Anikeeva, C. F. Madigan, J. E. Halpert, M. G. Bawendi, and V. Bulovi c, \Electronic and excitonic processes in light-emitting devices based on organic materials and colloidal quantum dots", Physical Review B 78, Publisher: American Physical Society, 085434 (2008). 42 Z. Tan, B. Hedrick, F. Zhang, T. Zhu, J. Xu, R. H. Henderson, J. Ruzyllo, and A. Y. Wang, \Stable Binary Complementary White Light-Emitting Diodes Based on Quantum-Dot/Polymer-Bilayer Structures", IEEE Photonics Technology Letters 20, Conference Name: IEEE Photonics Technology Letters, 1998{2000 (2008). 43 V. Wood and V. Bulovi c, \Colloidal quantum dot light-emitting devices", en, Nano Reviews 1, 5202 (2010). 44 S. Eustis and M. A. El-Sayed, \Why gold nanoparticles are more precious than pretty gold: Noble metal surface plasmon resonance and its enhancement of the radiative and nonradiative properties of nanocrystals of dierent shapes", en, Chem. Soc. Rev. 35, 209{217 (2006). 112

Abstract (if available)

Abstract Core-shell nanoparticles are a class of nanomaterials comprised of two or more distinct materials composed in a layered, spherical shape. These nanomaterials represent the first step towards more complicated nanoscale structures and functionality, which has been an ongoing technological goal in the scientific community, leading to advances in and miniaturization of important capabilities such as computing and sensors. In this thesis, I explore two unique core-shell particle platforms. ❧ The first core-shell particle platform is a metal-insulator combination, AuZnO, which has a gold core surrounded by a thick and porous ZnO shell. These materials are synergistic in many ways. The gold core acts as an optically active sensor or receiver. It has a tunable surface plasmon resonance, the frequency of which is sensitive to local dielectric modification, and therefore can act as a nanoscale sensor for changes in the local environment. The ZnO shell provides a porous yet protective layer which prevents uncontrolled modification of the gold core surface. Run in reverse, the gold nanoparticle core is a nanoantenna, receiving optical energy and dispersing it throughout the ZnO shell, which may contain a payload that can be released upon the external optical trigger. I demonstrate the successful modification of this platform to incorporate dopant elements. I also describe experiments which provide insight into the role of the gold core in stabilizing particle morphology, particularly with higher dopant concentrations. My work expands understanding of the AuZnO platform and introduces a new vector for tunability. ❧ The second core-shell particle platform is a layered semiconductor quantum dot, AgInS₂-ZnS (ZAIS). Semiconductor quantum dots are known to be efficient and photostable emitters with applications as bioimaging agents and LED device emission layer materials. However, the state-of-the-art quantum dot materials contain heavy metals. ZAIS is a promising alternative quantum dot material which is heavy metal free, but some details about its luminescence mechanism remain unclear. I perform experiments and analysis to shed light on the internal defect-related luminescence mechanism in ZAIS quantum dots, and find that its thermal sensitivity places limits on the classes of defect responsible for the emission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Nonlinear optical nanomaterials in integrated photonic devices

PDF

Integration of cost effective luminescent materials into organic light emitting diodes (OLEDs) and molecular approaches to improve the device efficiency

PDF

Optical simulation and development of novel whispering gallery mode microresonators

PDF

Molecular design strategies for blue organic light emitting diodes

PDF

Optically triggered smart polymers for environmental monitoring

PDF

Phase change heterostructures for electronic and photonic applications

PDF

Engineering solutions for biomaterials: self-assembly and surface-modification of polymers for clinical applications

PDF

Surface-enhanced on-chip whispering gallery mode resonators for improved nonlinear optics

PDF

Development of hybrid optical microcavities for Plasmonic laser and improving biodetection

PDF

Transport studies of phase transitions in a quasi-1D hexagonal chalcogenide

PDF

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

PDF

Fabrication and characterization of yoroidal resonators for optical process improvement

PDF

Perovskite chalcogenides: emerging semiconductors for visible to infrared optoelectronics

PDF

Development of novel optical materials for whispering gallery mode resonator for nonlinear optics

PDF

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

PDF

Nanomaterials for bio-imaging and therapeutics

PDF

Printed and flexible carbon nanotube macroelectronics

PDF

Expanding the synthesis space of 3D nano- and micro-architected lattice materials

PDF

Developing improved silica materials and devices for integrated optics applications

PDF

Single-wall carbon nanotubes separation and their device study

Asset Metadata

Creator Zeto, Rene Wallace (author)

Core Title Materials development and characterization of optically-active core-shell nanomaterials

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Materials Science

Degree Conferral Date 2021-08

Publication Date 07/17/2021

Defense Date 04/27/2021

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

Tag fluorescent,nanoparticle,OAI-PMH Harvest

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Armani, Andrea (committee chair), Ravichandran, Jayakanth (committee member), Thompson, Mark (committee member), Wu, Wei (committee member)

Creator Email zeto@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC15595905

Unique identifier UC15595905

Legacy Identifier etd-ZetoReneWa-9760

Document Type Dissertation

Format application/pdf (imt)

Rights Zeto, Rene Wallace

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 author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

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

Repository Email cisadmin@lib.usc.edu