UC Berkeley

There is probably no class of materials more varied, more widely used, or more ubiquitous than metal oxides. Depending on their composition, metal oxides can exhibit almost any number of properties. Of particular interest are the ways in which charge is transported in metal oxides: devices such as displays, touch screens, and smart windows rely on the ability of certain metal oxides to conduct electricity while maintaining visible transparency. Smart windows, fuel cells, and other electrochemical devices additionally rely on efficient transport of ionic charge in and around metal oxides.

Colloidal synthesis has enabled metal oxide nanocrystals to emerge as a relatively new but highly tunable class of materials. Certain metal oxide nanocrystals, particularly highly doped metal oxides, have been enjoying rapid development in the last decade. As in myriad other materials systems, structure dictates the properties of metal oxide nanocrystals, but a full understanding of how nanocrystal synthesis, the processing of nanocrystal-based materials, and the structure of nanocrystals relate to the resulting properties of nanocrystal-based materials is still nascent. Gaining a fundamental understanding of and control over these structure-property relationships is crucial to developing a holistic understanding of metal oxide nanocrystals. The unique ability to tune metal oxide nanocrystals by changing composition through the introduction of dopants or by changing size and shape affords a way to study the interplay between structure, processing, and properties.

This overall goal of this work is to chemically synthesize colloidal metal oxide nanocrystals, process them into useful materials, characterize charge transport in materials based on colloidal metal oxide nanocrystals, and develop ways to manipulate charge transport. In particular, this dissertation characterizes how the charge transport properties of metal oxide nanocrystal-based materials depend on their processing and structure. Charge transport can obviously be taken to mean the conduction of electrons, but it also refers to the motion of ions, such as lithium ions and protons. In many cases, the transport of ions is married to the motion of electrons as well, either through an external electrical circuit, or within the same material in the case of mixed ionic electronic conductors. The collective motion of electrons over short length scales, that is, within single nanocrystals, is also a subject of study as it pertains to plasmonic nanocrystals. Finally, charge transport can also be coupled to or result from the formation of defects in metal oxides. All of these modes of charge transport in metal oxides gain further complexity when considered in nanocrystalline systems, where the introduction of numerous surfaces can change the character of charge transport relative to bulk systems, providing opportunities to exploit new physical phenomena.

Part I of this dissertation explores the combination of electronic and ionic transport in electrochromic devices based on nanocrystals. Colloidal chemistry and solution processing are used to fabricate nanocomposites based on electrochromic tin-doped indium oxide (ITO) nanocrystals. The nanocomposites, which are completely synthesized using solution processing, consist of ITO nanocrystals and lithium bis(trifluoromethylsulfonyl)amide (LiTFSI) salt dispersed in a lithium ion-conducting polymer matrix of either poly(ethylene oxide) (PEO) or poly(methyl methacrylate) (PMMA). ITO nanocrystals are prepared by colloidal synthetic methods and the nanocrystal surface chemistry is modified to achieve favorable nanocrystal-polymer interactions. Homogeneous solutions containing polymer, ITO nanocrystals, and lithium salt are thus prepared and deposited by spin casting. Characterization by DC electronic measurements, microscopy, and x-ray scattering techniques show that the ITO nanocrystals form a complete, connected electrode within a polymer electrolyte matrix, and that the morphology and properties of the nanocomposites can be manipulated by changing the chemical composition of the deposition solution. Careful application of AC impedance spectroscopy techniques and DC measurements are used to show that the nanocomposites exhibit mixed ionic and electronic conductivity, where electronic charge is transported through the ITO nanocrystal phase, and ionic charge is transported through the polymer matrix phase. Additionally, systematic changes in ionic and electronic conductivity with morphology are measured. The synthetic methods developed here and understanding of charge transport ultimately lead to the fabrication of a solid state nanocomposite electrochromic device based on nanocrystals of ITO and cerium oxide.

Part II of this dissertation considers electron transport within individual metal oxide nanocrystals themselves. It primarily examines relationships between synthetic chemistry, doping mechanisms in metal oxides, and the accompanying physics of free carrier scattering within the interior of highly doped metal oxide nanocrystals, with particular mind paid to ITO nanocrystals. Additionally, synthetic methods as well as metal oxide defect chemistry influences the balance between activation and compensation of dopants, which limits the nanocrystals' free carrier concentration. Furthermore, because of ionized impurity scattering of the oscillating electrons by dopant ions, scattering must be treated in a fundamentally different way in semiconductor metal oxide materials when compared with conventional metals. Though these effects are well understood in bulk metal oxides, further study is needed to understand their manifestation in nanocrystals and corresponding impact on plasmonic properties, and to develop materials that surpass current limitations in free carrier concentration and mobilities. In particular, efforts to address these limitations by developing new nanocrystal materials (with careful consideration of structure-property relationships) are described. Synthetic control of nanocrystal shape is also explored. Each of these topics have implications in determining the properties of localized surface plasmon resonances (LSPRs) in these nanocrystals.

Part II culminates as the defect chemistry of metal oxides is identified as a major factor influencing LSPR and charge transport in doped metal oxide nanocrystals. Aliovalent dopants and oxygen vacancies act as centers for ionized impurity scattering of electrons, and this electronic damping leads to lossy, broadband LSPR with low quality factors, limiting applications that require near field concentration of light. However, the appropriate dopant can mitigate ionized impurity scattering. Herein, the synthesis and characterization of a novel doped metal oxide nanocrystal material, cerium-doped indium oxide (Ce:In2O3) is described. Ce:In2O3 nanocrystals display tunable mid-infrared LSPR with exceptionally narrow line widths and the highest quality factors observed for nanocrystals in this spectral region. Drude model fits to the spectra indicate that a drastic reduction in ionized impurity scattering is responsible for the enhanced quality factors, and high electronic mobilities reaching 33 cm2/Vs are measured optically, well above the optical mobility for ITO nanocrystals. The microscopic mechanisms underlying this enhanced mobility are investigated with density functional theory calculations, which suggest that scattering is reduced because cerium orbitals do not hybridize with the In orbitals that dominate the bottom of the conduction band. Ce doping may also reduce the equilibrium oxygen vacancy concentration, further enhancing mobility. Absorption spectra of single Ce:In2O3 nanocrystals are used to determine the dielectric function, and simulations predict strong near field enhancement of mid-IR light, especially around the vertices of Ce:In2O3 nanocubes.

Part III examines how the defect chemistry of metal oxides can be used to manipulate not only electronic transport, but also ionic transport in materials that are relevant for high temperature electrochemistry. Over the past few years, the observation of unexpected but significant proton conductivity in porous, nanocrystalline ceramics has generated substantial scientific interest mirroring the excitement surrounding ionic conduction in other nanostructured or porous materials. Numerous studies, to varying degrees of success, have attempted to describe or control the mechanisms that enable proton motion in nanocrystalline ceramics. Here, colloidally synthesized ceramic nanocrystals of cerium oxide (CeO\textsubscript{2}) and titanium oxide (TiO\textsubscript{2}) are utilized to systematically study how grain size, microporosity, and composition influence proton conduction. By measuring the temperature-dependent impedance of porous thin films of these nanocrystals under dry and wet atmospheres, it was found that both CeO2 and TiO2 display significant proton conductivity at intermediate temperatures between 100C and 350C. Furthermore, oxygen activity strongly impacts proton transport; using oxygen as a carrier gas drastically reduced the proton conductivity by up to 60 times. Together, these results suggest that the most likely source of mobile protons in these systems is dissociative adsorption of water at surface oxygen vacancies, with composition, nanocrystal size, and oxide defect equilibria influencing the surface activity toward this reaction and hence the proton conductivity.