This thesis analyzes the chemical vapour deposition (CVD) growth of vertically aligned carbon nanotube (CNT) forests in order to understand how CNT forests grow, why they stop growing, and how to control the properties of the synthesized CNTs. In situ kinetics data of the growth of CNT forests are gathered by in situ optical microscopy. The overall morphology of the forests and the characteristics of the individual CNTs in the forests are investigated using scanning electron microscopy and Raman spectroscopy. The in situ data show that forest growth and termination are activated processes (with activation energies on the order of 1 eV), suggesting a possible chemical origin. The activation energy changes at a critical temperature for ethanol CVD (approximately 870°C). These activation energies and critical temperature are also seen in the temperature dependence of several important characteristics of the CNTs, including the defect density as determined by Raman spectroscopy. This observation is seen across several CVD processes and suggests a mechanism of defect healing. The CNT diameter also depends on the growth temperature. In this thesis, a thermodynamic model is proposed. This model predicts a temperature and pressure dependence of the CNT diameter from the thermodynamics of the synthesis reaction and the effect of strain on the enthalpy of formation of CNTs. The forest morphology suggests significant interaction between the constituent CNTs. These interactions may play a role in termination. The morphology, in particular a microscale rippling feature that is capable of diffracting light, suggest a non-uniform growth rate across the forest. A gas phase diffusion model predicts a non-uniform distribution of the source gas. This gas phase diffusion is suggested as a possible explanation for the non-uniform growth rate. The gas phase diffusion is important because growth by acetylene CVD is found to be very efficient (approximately 30% of the acetylene is converted to CNTs). It is seen that multiple mechanisms are active during CNT growth. The results of this thesis provide insight into both the basic understanding of the microscopic processes involved in CVD growth and how to control the properties of the synthesized CNTs.

<p> Thin silicon based films were produced using low temperature (less than 60° C) electron cyclotron resonance plasma enhanced chemical vapour deposition (ECR PECVD). These films were examined for suitability in flat panel display applications. SiOxNy films were tested for use as insulating films in thin film electroluminescent (TFEL) devices. The ECR PECVD method was found to be suitable when the plasma was created using pure nitrogen (as opposed to argon) in high ratios to the silane precursor.</p> <p> Hydrogenated silicon films were also produced and evaluated for their suitability as semiconductor layers in thin film transistors (TFTs). The silicon films were subject to nickel induced crystallization. The silicon films were found to crystallize at low temperatures, (<950° C) in the presence of nickel. These films were used to produce prototype metal insulator semiconductor (MIS) capacitors and TFTs.</p> / Thesis / Master of Applied Science (MASc)

Metalorganic Chemical Vapour Deposition (MOCVD) is a versatile technique for the deposition of thin films of oxide materials as it offers advantages, such as deposition over large surface area, conformal coverage, selective area deposition, and a high degree of compositional control. The MOCVD process uses metalorganic (MO) complexes, such as β-diketonate and alkoxide-based complexes, as precursors. These complexes are stable and moderately volatile. Because of the direct bond between metal and oxygen, MO complexes are natural precursors for oxide coatings. As the process involves chemical reactions taking place on the substrate surface, growth of thin films by MOCVD depends on various parameters such as the chemical nature and concentration of precursors, reaction pressure, reaction temperature, and the nature of the substrate. Such a large parameter space of the CVD process, when combined with the dynamics (thermodynamics and fluid dynamics) and kinetics, makes it rather complex. This complexity allows one to make thin films of metastable phases, including amorphous materials. One of the important findings of the work is that MOCVD process is capable of making composite coatings of carbonaceous metal oxide. Manganese is multivalent and forms various stable oxides, such as MnO, Mn2O3, Mn3O4 and MnO2. There are various potential applications of manganese oxides. MnO2 is a very well studied material for its electrochemical applications in dry cells, lithium-ion batteries, and in supercapacitors. Hence, it becomes pertinent to explore the properties of thin films of manganese oxides prepared by MOCVD for various electrochemical and other applications. The thesis work is divided into two parts. Part 1 describes the synthesis of manganese complexes, their characterization, and their application to the CVD of coatings, especially those of carbonaceous MnO. Part 2 is devoted to a detailed study of electrochemical aspects of the carbonaceous MnO coatings, followed by a report on their unusual transport properties. Chapter 1 begins with a brief introduction to thin film deposition processes. In particular, the CVD process is described with reference to various parameters such as carrier gas flow, pressure, temperature and most importantly, the CVD precursor. The chapter ends with a description of the scope of the work undertaken for the present thesis. Chapter 2 deals with “Synthesis and Characterization of MO complexes”. It begins with a description of the classification of CVD precursors with the description of MO complexes such as β-diketonates, which are generally subliming crystalline solids. Manganese β-diketonate complexes are discussed in detail. Due to the multivalent nature of Mn, there are two possible complexes namely Mn(acac)2(H2O)2 and Mn(acac)3. These complexes have been synthesised and characterized (confirmed) by various techniques, such as elemental analysis (CHN), X-ray diffraction (XRD), FTIR spectroscopy, and mass spectroscopy. Thermal analysis of the complexes shows that they are suitable as MOCVD precursors. We have used Mn(acac)2(H2O)2 as a precursor in the present work. Metalorganic complexes, where metal ion is directly bonded with both nitrogen and oxygen, can be potential candidates for the precursor for oxynitrides coatings. We have therefore studied solid crystalline anthranilate complexes of various metal ions, such as Mn2+, Co2+, Cu2+ and Zn2+ and confirmed their formation. Thermal analysis shows that anthranilate complexes are fairly volatile below 250oC and decompose below 500oC. These complexes were pyrolysed in open air and in sealed tube at different temperatures, and the resulting powder product examined by XRD, SEM, EDAX and FTIR. This preliminary study shows that anthranilate complexes yield different oxides of Mn, Co and Cu under different pyrolysis conditions, with very interesting morphological features. Pyrolysis of Zn(aa)2 in a sealed tube leads to the formation of a nanocomposite of carbon and zinc oxide (wuerzite), rich in carbon, with potential for applications in catalysis. On the other hand, the pyrolysis of Zn(aa)2 in air at the same temperature leads to leads to crystalline, nanostructured zinc oxide (wuerzite). However, no attempt has been made to use these anthranilates as CVD precursors. Chapter 3 deals with “MOCVD of Manganese Oxides and their Characterization”. It begins with a brief review of various manganese oxides and their properties. This is followed by description of the CVD reactor used for the present work, together with the conditions employed for the deposition of MnOx films. Depositions have been carried out on different substrates such as SS-316, ceramic alumina and Si (111), while varying various deposition parameters, viz., substrate, reactor pressure, carrier gas (argon) flow rate, and the duration of deposition. Significantly, depositions are divided into two categories: one, carried out in argon ambient, in the absence of a supply of oxygen (or any other oxidant) and the second one, under oxygen flow, using argon as carrier gas. The films deposited in the absence of oxygen flow are thick, black in colour, and electrically conducting, indicating the presence of carbon. The growth rate follows a typical thermal pattern, with activation energy of ~ 1.7 eV. Detailed characterization by XRD, TEM/ED, Raman, FTIR and XPS (X-ray photoelectron spectroscopy) shows that these films are composed of MnO in a carbon-rich amorphous matrix. High-resolution SEM (fig. 1) reveals a fractal pattern of cauliflower morphology, comprising very fine particles (4 – 10 nm), characteristic of very large specific surface area of the film, which is confirmed by volumetric BET measurement (~2000 m2/g). We conclude that growth in argon ambient leads to a homogenous nanocomposite film of hydrated MnO in carbon-rich matrix. Thus, our study reveals that MOCVD is a novel one-step chemical method to produce homogenous composite thin films, wherein all components of the nanocomposite film emerge from the same chemical precursor. Carbon incorporation is generally avoided by empirical process design, as it is viewed as an impurity. The potential advantages of carbon incorporation are thus not examined and the composite nature of carbonaceous films not recognized in the literature. Carbonaceous nanocomposite film can be significant as an electrode in supercapacitors, as discussed in part 2 of the thesis. Chapter 3 describes films deposited under oxygen flow, which are no longer black and are highly resistive, indicating the absence of carbon in the film, as confirmed by Raman spectroscopy. XRD, FTIR and Raman spectroscopy reveal that the films obtained under oxygen flow are more crystalline than the ones obtained in the absence of oxygen flow, and that the films are generally nanocrystalline composites of two manganese oxides, such as MnO and Mn3O4. Given the context of the carbonaceous MnO films described above, chapter 4 begins with a review of electrochemical capacitors (also called supercapacitors or ultracapacitors), which are emerging as important energy storage devices. Until now, in the Mn-O system, hydrated MnO2 has been well-studied as an electrode material due to its low cost and environmental compatibility, but the low electrical conductivity of MnO2, together with irreversible redox reactions, reduces its performance. In electrochemical capacitor applications, metal-oxide/carbon composites are finding importance. Chapter 4 deals with “MnO/C Nanocomposite Coatings as Electrodes for Electrochemical Capacitor”. In this chapter, we have examined the novel EM, i.e., the hydrated MnO/C nanocomposite coating prepared by the MOCVD process on a conducting substrate (current collector) such as SS-316 as an electrode. Electrochemical measurements have been carried out for both the 3-electrode assembly (for basic aqueous electrolyte) and 2-electrode assembly (for gel polymer electrolyte) using cyclic voltammetry (CV), AC impedance and charge-discharge techniques. The studies lead to a maximum specific capacitance of 230 – 270 F/g at 1 mA/cm2 discharge current density for the MnO/C nanocomposite coating grown at 680oC. The Bode plot shows a maximum phase angle of around 74 – 82o, indicating capacitive behaviour. The MnO/C nanocomposite film shows a very small time constant (0.5 – 3 msec), which is good for high frequency applications. The pulse power figure of merit is found to be 650 – 2000 W/g. Capacitance determined for a large number of charge-discharge cycles (~20000), and at large current densities (50 mA/cm2) show promising results. The energy density (5 - 32 Wh/kg) and power density (2 – 4 kW/kg) estimated from charge-discharge data at 1 mA/cm2 shows the potential of the nanocomposite MnO/C as electrode for superior capacitor devices. Gel polymer electrolytes (GPE) offer the advantage of large electrochemical potential window due to its structural and chemical stability. Studies have been carried out to show that the MnO/C nanocomposite film is compatible with gel polymer electrolytes based on poly(methyl methacrylate) (PMMA) and poly(acrylonitrile) (PAN) with salts of magnesium triflate and magnesium perchlorate, respectively) and plasticizers (ethylene carbonate (EC) + propylene carbonate (PC)), in a 2-electrode assembly. Chapter 5 deals with “Magnetoconductance in MnO/C Nanocomposite Coatings on Alumina”. Amorphous systems, such as MnO/C composites wherein carbon is amorphous and MnO is nearly so, are highly symmetric condensed phases, which do not possess long range translational or orientational order. Disorder in the system creates Anderson localized states just above the valence band, which lead to reduced electrical conductivity. Amorphous systems show either a small negative magnetoresistance (~ 5%) or a small positive magnetoconductance (~ 7%) at very low temperatures (~ 10 K). As such, the transport properties of the MnO/C nanocomposite film have been investigated, and are reported in chapter 5. Transport and magnetotransport measurements have been made on the MnO/C nanocomposite film grown on alumina. It is found that the MnO/C nanocomposite coating exhibits a giant negative MR (22.3%) at a temperature as high as 100 K, which is unusual because pure MnO is anti-ferromagnetic and does not ordinarily show any magnetoresistance (MR), while amorphous carbon is known to show a small MR at very low temperatures (~7 K), due to weak-localization. The present results mean that a mechanism other than weak-localization plays a role in this nanocomposite material. Further study of this material is called for, which can perhaps lead to giant magnetoresistance (GMR) at room temperature in a metal-oxide/carbon nanocomposite. A summary of the work and an outlook for further research are given in the concluding chapter 6.

La découverte du graphène a récemment ouvert de nouvelles opportunités en termes de fonctionnalités et de performances pour les dispositifs de spintronique. Ce travail comporte deux études sur l’utilisation du graphène en spintronique.C’est en premier lieu pour ses propriétés de transport de spin que le graphène a suscité un fort intérêt en spintronique. En effet, de par sa forte mobilité et son faible couplage spin-orbite, il est rapidement apparu comme ayant un fort potentiel pour le transport de l’information de spin avec des longueurs de diffusion de spin attendues de l’ordre de la centaine de microns.Dans une première étude, je m'intéresse au graphène en tant que plateforme pour propager un courant polarisé en spin. Je décris tout d'abord les principales techniques de mesure de vannes de spin latérales, en insistant sur l'importance de la barrière tunnel pour se placer dans les conditions appropriées à la mesure des propriétés intrinsèques au graphène. Je présente ensuite les résultats que j’ai obtenus. Je commence par ceux sur graphène épitaxié sur SiC dans lequel nous avons réussi à injecter, propager et détecter un courant polarisé en spin créé soit grâce à un injecteur ferromagnétique (Co/Al2O3), soit par effet Hall de spin (à partir du platine). Je présente ensuite les résultats obtenus sur un autre type de graphène grande surface, le graphène CVD monocouche, pour lequel j'ai pu expérimenter une nouvelle barrière tunnel: le nitrure de bore hexagonal.Par-delà ses propriétés de transport latéral, le graphène pourrait avoir un autre intérêt pour la spintronique, par exemple dans le cadre de la passivation des couches ferromagnétiques dans les jonctions tunnel magnétiques.Dans une seconde étude, je m'intéresse au graphène comme membrane pour protéger une électrode ferromagnétique de l'oxydation tout en autorisant l’extraction d’un courant polarisé en spin. Aujourd’hui, dû à la propension naturelle des matériaux ferromagnétiques à s’oxyder, les procédés humides/oxydants sont souvent exclus de la fabrication de dispositifs de spintronique. Après avoir introduit les enjeux, je présente mes résultats expérimentaux. Je montre tout d'abord qu’une monocouche de graphène suffit à empêcher l'oxydation d'une électrode de nickel et qu’un filtrage de spin intéressant apparaît à l'interface Ni/Graphène. Je valide ensuite l'ensemble de ce potentiel en montrant qu'on peut utiliser une technique oxydative de dépôt tel que l'Atomic Layer Deposition (ALD) sans endommager les propriétés de l'électrode ferromagnétique Ni+Graphène. Le procédé d’ALD, bien qu'utilisé partout en électronique (cette technique sert aujourd’hui à réaliser les grilles des transistors d’Intel), était jusqu’ici proscrit car il met en jeu des molécules telles que l'ozone ou l'eau et est donc par nature oxydant. Enfin, je montre que le filtrage de spin à l’interface Ni/Graphène aboutit alors à une inversion quasi-totale de la polarisation en spin du Ni.Ce travail de thèse montre que le graphène peut être utilisé comme canal de transport d’un courant polarisé en spin, comme membrane protectrice imperméable à l’oxydation ou encore comme filtre à spin. L’ensemble de ces travaux illustre la richesse des applications du graphène pour la spintronique. / Graphene discovery has opened new opportunities in terms of functionality and performance for spintronics devices. This work presents two examples of what graphene can bring to the spintronics field.Graphene first aroused interest amongst the community because of its excellent properties for transporting spin information. Indeed, thanks to its high reported mobilities and its weak spin-orbit coupling, graphene quickly became a high-potential candidate to transport spin information with expected spin diffusion length in the hundreds of microns range.In the first part of this thesis, I study graphene as a platform to propagate a spin polarized current. I first describe the main techniques to measure lateral spin valves, emphasizing the importance of the tunnel barrier being under the right conditions to permit measurement of the intrinsic properties of graphene. I then present my results. I begin with the results obtained on epitaxial graphene on SiC, in which I was able to inject, propagate and detect a spin current created either by a ferromagnetic injector (Co/Al2O3), or through the spin Hall effect (from Pt). Then, I present the results obtained on another large-area graphene, a single layer of graphene grown by CVD on which I tested a new unnel barrier : hexagonal boron-nitrideBeyond its potential as a platform to transport spin information, other opportunities for graphene in spintronics exist, for example its use in the passivation of ferromagnetic layers in magnetic tunnel junctions.In the second part of this thesis, I am interested in graphene’s potential as a membrane that could protect ferromagnets from oxidation while simultaneously allowing the extraction of a spin current. Indeed, because of the natural propensity of the ferromagnetic material to be oxidized, humid and oxidative processes are excluded from the fabrication of spintronic devices. After introducing the background motivation, I present my experimental results. I first show that a single layer of grapheneis enough to prevent the oxidation of a Ni electrode and that an interesting spin filtering effect happens at the interface Ni/Graphene. I then confirm this by showing that it is possible to use an oxidative technique like Atomic Layer Deposition (ALD) without damaging the properties of the ferromagnetic electrode Ni+Graphene. ALD is widely used in electronics (Intel uses it to make its transistor gates) but was up to now prohibited in spintronics because it involves oxidative molecules like water or ozone. Finally, I show that the spin-filtering effect at the interface Ni/Graphene leads to a quasi-total reversal of the spin polarisation of the Ni.This thesis shows that graphene can be used as a channel to transport spin information, as a protective membrane to protect against oxidation, or as a spin filter. All this work illustrates the richness of graphene applications for spintronics.

Thesis (PhD (Chemistry and Polymer Science))--Stellenbosch University, 2008. / The successful preparation and structural characterization of a number of N,N-dialkyl-N’-benzoyl(thio)selenourea ligands is described; where the intermolecular interactions are characterized by the presence of Resonance Assisted Hydrogen Bonding (RAHB), π- π interactions between neighbouring benzene residues only being evident amongst the longer alkyl chain derivatives. The first structural characterization of an asymmetrically substituted N,N-dialkyl- N’-benzoylselenourea ligand reveals an increased stability of the Z isomer in the solid state, this being reflected by the sulfur analogue. Attempts to synthesise N,N-dicyclohexyl-N’-benzoylselenourea led to the isolation and structural characterization of a novel 1,3,5-oxaselenazine salt and dicyclohexylaminobenzoate. The first structural characterization of a “bipodal” N,N-dialkyl-N’-benzoylselenourea ligand, 3,3,3’,3’-tetrabutyl-1,1’- isophthaloylbis(selenourea), reveals RAHB in the crystal lattice similar to that exhibited by the “monopodal” analogue, N,N-dibutyl-N’-benzoylselenourea. The successful complexation of the N,N-dialkyl-N’-benzoyl(thio)selenourea ligands to a number of different transition metal ions is reported allowing the preparation of several potential single source precursors. Coordination through the O and Se/S donor atoms to Pd(II) results in the formation of square planar metal complexes, with a cis conformation, several of which could be structurally characterized. In particular, the first structural elucidation of an asymmetrically substituted N,N-dialkyl-N’-benzoylselenourea metal complex, cis-bis(N-benzyl-N-methyl-N’- benzoylselenoureato)palladium(II) indicates the increased stability of the EZ isomer in the solid state. Structural elucidation of the novel (N,N-diphenyl-N’-benzoylselenoureato)cadmium(II) reveals a bimetallic complex in the solid state, where the expected 2:1 ligand : metal ratio is maintained, and the two Cd(II) centres are 5 and 6 coordinated, with O and Se donor atoms. Multinuclear Nuclear Magnetic Resonance (NMR) Spectroscopy has been employed in the thorough characterisation of the potential single source precursors, 77Se NMR spectroscopy indicating a decreased shielding of the 77Se nucleus as the “hardness” of the central metal ion increases i.e. Pd(II) > Zn(II) > Cd(II). Use of 113Cd NMR spectroscopy indicates the preferential binding of N,N-diethyl-N’- benzoylselenourea to Cd(II) over that of its sulfur analogue, and initial studies suggest a form of chelate metathesis taking place in solution. 31P NMR spectroscopy is used to gain insight into the formation of cis-bis(N,N-diethyl-N’- benzoylselenoureato)Pt(II). Thermolysis of (N,N-diethyl-N’-benzoylselenoureato)cadmium(II) and its sulfur analogue led to the successful synthesis of CdSe and CdS quantum dots respectively, where thermolysis over a range of temperatures allows a degree of size control over the resulting nanoparticles. The effect of precursor alkyl chain length on nanoparticle morphology was investigated for both the N,N-dialkyl-N’-benzoylthio- and –selenoureas. A correlation between the two for the (N,N-dialkyl-N’-benzoylselenoureato)Cd(II) complexes is described and possible growth mechanisms are discussed. Preliminary investigations into the use of other N,N-dialkyl-N’-benzoyl(thio)selenourea metal complexes as single source precursors reveal that both (N,N-diethyl-N’-benzoylselenoureato)Zn(II) and its sulfur analogue show potential as single source precursors for the formation of ZnO and ZnS nanoparticles respectively. Initial studies into the use of N,N-dialkyl-N’-benzoyl(thio)selenourea metal complexes as single source precursors for the synthesis of core-shell nanoparticles is briefly described. The Aerosol Assisted Chemical Vapour Deposition (AACVD) of several N,N-dialkyl-N’-benzoyl(thio)selenourea metal complexes is reported, where both (N,N-diethyl-N’-benzoylselenoureato)Cd(II) and its sulfur analogue allow the deposition of crystalline CdSe and CdS respectively. The AACVD of (N,N-diethyl-N’- benzoylselenoureato)Zn(II) leads to the deposition of crystalline ZnSe, ZnS being deposited by (N,N-diethyl-N’-benzoylthioureato)Zn(II). The deposition of heazelwoodite (Ni3S2) with varying morphologies results from the AACVD of cis-bis(N,N-diethyl-N’-benzoylthioureato)Ni(II). Thermal annealing of the amorphous material deposited by the AACVD of cis-bis(N,N-diethyl-N’-benzoylthioureato)Pd(II), allows the formation of highly crystalline palladium. The deposition of metallic platinum using cis-bis(N,N-diethyl-N’-benzoylthioureato)Pt(II) is described as well as the deposition of crystalline Pd17Se15 from cis-bis(N,N-diethyl-N’-benzoylselenoureato)Pd(II). This, to the best of our knowledge, is the first time that AACVD has been performed, using the N,N-dialkyl-N’- benzoyl(thio)selenourea metal complexes as single source precursors, in addition, we believe it to be the first time that palladium selenide has been deposited using the AACVD technique.

A series of ruthenium complexes of the general type Ru(CO)2(P(n-Bu)3)2(O2CR)2 (4a, R = Me; 4b, R = Et; 4c, R = i-Pr; 4d, R = t-Bu; 4e, R = CH2OCH3; 4f, R = CF3; 4g, R = CF2CF3) was synthesized by a single-step reaction of Ru3(CO)12 with P(n-Bu)3 and the respective carboxylic acid. The molecular structures of 4b, 4c and 4e–g in the solid state are discussed. All ruthenium complexes are stable against air and moisture and possess low melting points. The physical properties including the vapor pressure can be adjusted by modification of the carboxylate ligands. The chemical vapor deposition of ruthenium precursors 4a–f was carried out in a vertical cold-wall CVD reactor at substrate temperatures between 350 and 400 °C in a nitrogen atmosphere. These experiments show that all precursors are well suited for the deposition of phosphorus-doped ruthenium layers without addition of any reactive gas or an additional phosphorus source. In the films, phosphorus contents between 11 and 16 mol% were determined by XPS analysis. The obtained layers possess thicknesses between 25 and 65 nm and are highly conformal and dense as proven by SEM and AFM studies. / Dieser Beitrag ist aufgrund einer (DFG-geförderten) Allianz- bzw. Nationallizenz frei zugänglich.

The doping of rhenium into molybdenum disulfide was achieved by Aerosol Assisted Chemical Vapour Deposition (AACVD) from single source precursors. Rhenium can be studied as a model for immobilization of radioactive technetium-99 (99Tc) in MoS2. The metals Mo(IV), Re(IV), and Tc(IV) have similar ionic radii 0.65, 0.63 and 0.65 Å respectively, and their Shannon-Prewitt crystal radii 0.79, 0.77 and 0.79 Å Hence demonstrating the potential storage of nuclear waste in geologic like formations in of groundwater may be possible. The interaction between the nuclear waste forms and groundwater, which could lead to release and transport low concentrations or vapour of radionuclides to the near field, as a result, decomposition of engineered barriers. The molecular precursors [Mo(S2CNEt2)4], [Re3(μ-SiPr)3(SiPr)6], [Re(S2CC6H5)(S3CC6H5)2], and [Re2(μ-S)2(S2CNEt2)4] have been used to deposit Re-doped MoS2 thin films. Mo-doped ReS2 alloyed, polycrystalline thin films were synthesised using [Re(S2CC6H5)(S3CC6H5)2], [Mo(S2CNEt2)4] via AACVD, adding with a low concentration of Mo source for the first time . We reported as well a new way for production of ultrathin ReS2 nanosheets by coupling bottom up processing AACVD with top-down LPE. This is important in synthetic pathways for the production of rare transition dichalcogenide, also, our processing methodology is potentially scalable and thus could be a way to commercial exploitation. Characterisation of produced materials performed by pXRD, SEM, TEM, STEM, EDX, ICP and Raman spectroscopy.

This study reports on the synthesis of N-doped carbon spheres (N-CSs) by a simple synthetic procedure. A horizontal CVD type reactor was used to synthesize N-CSs from pyridine. Depending on the dilution of the pyridine with toluene, a nitrogen content of 0.13-5 mol % was obtained. The use of a vertical CVD reactor gave N-CSs with a N-content of 0.19-3 mol % when an ammonium solution and acetylene were used as reactants. The diameters of carbon spheres were found to be in the range of 40 nm to 1000 nm for both CVD reactors. The diameter can be controlled by varying the flow rate, temperature, time, concentration and the reactor type. The samples were characterized by TEM, HRTEM, elemental analysis, Raman spectroscopy, TGA, PXRD and ESR. We have demonstrated that unsubstituted thiophene can be polymerized by Fe3+-catalyzed oxidative polymerization. The average particle size was about 50 nm, within a narrow particlesize distribution. The undoped carbon spheres (CSs) were reacted with thiophene to give polymer/carbon composites containing polythiophene and carbon nanospheres via chemical oxidative polymerization reaction. Polythiophene molecules were either chemically bonded or physically adsorbed to the surface of carbon spheres. The microstructure and properties of the two types of composites were compared. The thermogravimetric analysis data confirmed that the presence of CSs in the polymer\carbon composites is responsible for the higher thermal stability of the composite material in comparison with pristine polythiophene. The FTIR analysis showed that covalent functionalized nanocomposites exhibit a high intensity of a C-S bond This study reports on the synthesis of N-doped carbon spheres (N-CSs) by a simple synthetic procedure. A horizontal CVD type reactor was used to synthesize N-CSs from pyridine. Depending on the dilution of the pyridine with toluene, a nitrogen content of 0.13-5 mol % was obtained. The use of a vertical CVD reactor gave N-CSs with a N-content of 0.19-3 mol % when an ammonium solution and acetylene were used as reactants. The diameters of carbon spheres were found to be in the range of 40 nm to 1000 nm for both CVD reactors. The diameter can be controlled by varying the flow rate, temperature, time, concentration and the reactor type. The samples were characterized by TEM, HRTEM, elemental analysis, Raman spectroscopy, TGA, PXRD and ESR. We have demonstrated that unsubstituted thiophene can be polymerized by Fe3+-catalyzed oxidative polymerization. The average particle size was about 50 nm, within a narrow particlesize distribution. The undoped carbon spheres (CSs) were reacted with thiophene to give polymer/carbon composites containing polythiophene and carbon nanospheres via chemical oxidative polymerization reaction. Polythiophene molecules were either chemically bonded or physically adsorbed to the surface of carbon spheres. The microstructure and properties of the two types of composites were compared. The thermogravimetric analysis data confirmed that the presence of CSs in the polymer\carbon composites is responsible for the higher thermal stability of the composite material in comparison with pristine polythiophene. The FTIR analysis showed that covalent functionalized nanocomposites exhibit a high intensity of a C-S bondThis study reports on the synthesis of N-doped carbon spheres (N-CSs) by a simple synthetic procedure. A horizontal CVD type reactor was used to synthesize N-CSs from pyridine. Depending on the dilution of the pyridine with toluene, a nitrogen content of 0.13-5 mol % was obtained. The use of a vertical CVD reactor gave N-CSs with a N-content of 0.19-3 mol % when an ammonium solution and acetylene were used as reactants. The diameters of carbon spheres were found to be in the range of 40 nm to 1000 nm for both CVD reactors. The diameter can be controlled by varying the flow rate, temperature, time, concentration and the reactor type. The samples were characterized by TEM, HRTEM, elemental analysis, Raman spectroscopy, TGA, PXRD and ESR. We have demonstrated that unsubstituted thiophene can be polymerized by Fe3+-catalyzed oxidative polymerization. The average particle size was about 50 nm, within a narrow particlesize distribution. The undoped carbon spheres (CSs) were reacted with thiophene to give polymer/carbon composites containing polythiophene and carbon nanospheres via chemical oxidative polymerization reaction. Polythiophene molecules were either chemically bonded or physically adsorbed to the surface of carbon spheres. The microstructure and properties of the two types of composites were compared. The thermogravimetric analysis data confirmed that the presence of CSs in the polymer\carbon composites is responsible for the higher thermal stability of the composite material in comparison with pristine polythiophene. The FTIR analysis showed that covalent functionalized nanocomposites exhibit a high intensity of a C-S bond at 695 cm-1 , which is not observed in the noncovalent functionalized nanocomposites

<p>In this dissertation core level electron spectroscopy has been employed to study various aspects of metal oxide films grown under ultra-high vacuum conditions. </p><p>Studies on <i>in situ</i> ion insertion of lithium into thin TiO₂ systems were performed. The electronic and geometric properties are investigated in detail, along with an estimation of charge transfer from Li to Ti. </p><p>A detailed study of chemical vapour deposition of ZrO₂ on Si(100)-(2x1) was performed. ZrO₂ is found to be an insulator, i.e. its electronic levels are decoupled from the substrate and the Zr levels are best referenced to the local vacuum level. The alignment of the valence and conduction band has been determined. </p><p>Combinatorial chemical vapour deposition of TiO₂ and ZrO₂ on Si(100)-(2x1) was realized. A film with graded stoichiometry consisting of pure TiO₂ and ZrO₂ on the opposing ends and mixed composition of both oxides in the middle was obtained. A detailed study of the electronic levels revealed that ZrO₂ remains an insulator in the monolayer regime and that modification of ZrO₂ with a small amount of TiO₂ leads to a more symmetric alignment of the bands relative to Si. </p><p>The influence of a core hole on the O 1s x-ray absorption spectrum in TiO₂ and ZrO₂ is elucidated. Supported by O 1s photoemission measurements and <i>ab initio</i> calculations it is concluded that the static final state picture as well as dynamical threshold effects must be considered in order to determine the location of the conduction band minimum within the XAS framework. </p><p>Finally a Co modified Co:ZnO film was shown to display ferromagnetic properties. It could be evidenced that Co with oxygen as nearest neighbours was responsible for the magnetism and not metallic Co.</p>

For an electronic device well-designed interfaces are critical for the performance. Studies of interfaces down to an atomic level are thus highly motivated both from a fundamental and technological point of view. In this thesis, a surface science approach has been employed to study the formation of interfaces in systems relevant for transistor and solar cell applications. Surface science methodology entails ultra high vacuum environment, single crystalline surfaces, submonolayer control of deposited material, surface sensitive spectroscopy and atomic resolution microscopy. The primary experimental method for characterization is electron spectroscopy. This is a family of very powerful experimental techniques capable of giving information on the atomic level. Additionally, studies have been performed using scanning tunnelling microscopy. Combined these two methods can provide an atomic level characterisation of the geometric and electronic properties of the surface. The emphasis of this work is placed on ultra thin TiO2 and ZrO2 films grown on silicon substrates by means of ultra-high vacuum metal-organic chemical vapour deposition. ZrO2 has also been grown on SiC and FeCrAl. Deposition has been performed with different process parameters. The interface region of each film has been characterised. The band alignment, a most important issue with regard to the development of new transistor devices, for the ZrO2/Si(100) system has been explored. Decomposition pathways of the metal organic precursors have been studied in detail. Changing process parameters is shown to alter both the precursor decomposition pathway and the nature of the interface region, thus opening the possibility to tailor the material function. The titanium dioxide films grown in situ have shown to be excellent models of nanostructured electrode materials. In this spirit, interfaces of model systems for the solid-state dye-sensitized solar cell have been studied. Links between device performance and interface structure have been elucidated.