The high electric and thermal conductivity of copper has made it to a prime candidate as interconnect material in future integrated circuits. In this thesis, Atomic Layer Epitaxy has been used to deposit thin copper films on a variety of substrates, using both CuCl and Cu(II)2,2,6,6-tetramethyl-3,5-heptanedionate, Cu(thd)2, as precursors and hydrogen as reducing agent. Besides the experimental work, this thesis also comprises a large theoretical investigation where calculations based on Density Functional Theory has been performed in order to elucidate the deposition mechanisms in the CuCl/H2 process.  Experiments showed that highly pure copper with low resistivity could be selectively deposited in the Cu(thd)2 process on Pt/Pd seeded substrates at temperatures below 300 °C. At higher temperatures, a selectivity was lost due to a thermal decomposition of the precursor, The selectivity was explained by the high catalytic activity of the seedlayer.  Copper deposition by means of the CuCl/H2 process was kinetically controlled with an activation energy of approximately 85 kJmol-1 in the reduction step. Calculations showed that the rate determining step was the surface reaction between hydrogen and CuCl.  Place, publisher, year, edition, pages Uppsala: Acta Universitatis Upsaliensis , 1999. , p. [8], 45 Series Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology, ISSN 1104-232X ; 421 Keywords [en] Chemistry, Atomic Layer Epitaxy, ALE, copper, ab initio calculations, DFT, CuCl, Cu(thd)2, selectivity 

Artificially structured dielectric materials — thin films, mixtures of metal oxides and nanolaminates of Ta205, Hf02, Zr02, Nb205 and A1203 — can be successfully grown in atomic layer chemical vapour deposition (AL-CVD) process. Advanced dielectric properties can be obtained in the films grown at temperatures as low as 200-230°C. The dielectric performance of the binary metal oxides is determined by their chemical nature and structure. In certain extent, the residue content, film density and dielectric permittivity can be improved by increasing the deposition temperature or using better precursors. Further improvement in the film quality can be achieved by aiming at the formation of high-permittivity solid solutions by mixing different oxides like Ta205 and Nb205. The leakage currents can also be decreased by making use of the nanocrystalline nature of ultrathin Zr02 or Hf02 layers between amorphous Ta205 or (Nb^Ta^Os constraints in the multi layer structures. The charge storage capability of such advanced dielectric struc tures may be increased by ten times compared to the binary materials. It is obvious that, besides inherent thickness control, AL-CVD technique allows the precise tuning of the film composition. Moreover, the control over size-dependent dielectric properties is achieved. These advantages can be used for creating high-quality structured materials applicable in (large area) micro electronics such as integrated circuit (IC) processing or TFEL devices.

Bismuth titanate Bi4Ti3O12, is one of the bismuth based layered ferroelectric materials that is a candidate for replacing the lead based ferroelectric materials in for instance non-volatile ferroelectric random access memories (FRAM). This is due to the fact that the bismuth based ferroelectrics consists of pseudo perovskite units sandwiched in between bismuth oxide layers, which gives them a better fatigue nature. In this thesis thin films of Bi4Ti3O12 have been deposited by chemical vapour deposition (CVD) using the metal iodides, BiI3 and TiI4 as precursors. Films grown on MgO(001) substrates were found to grow epitaxially. The electrical properties were determined for films grown on Pt-coated silicon and good properties such as a high dielectric constant (ε) of 200, low tan δ of 0.018, a remnant polarisation (Pr ) of 5.3 μC/cm2 and coercive field (Ec) as high as 150 kV/cm were obtained. Thin films in the Bi-Ti-O system were also deposited by atomic layer deposition (ALD) using metalorganic precursors. In addition to the ternary bismuth titanates, films in the binary oxide systems i.e. bismuth oxides and titanium oxides were deposited. Epitaxial TiO2 films were deposited both by CVD and ALD using TiI4 as precursor. The rutile films deposited by ALD were found to grow epitaxially down to a temperature of at least 375 °C on α–Al2O3(0 1 2) substrates. The TiO2 ALD process was also studied in-situ by QCM. Different bismuth oxides were deposited by halide-CVD using BiI3 as precursor on MgO(0 0 1) and SrTiO3(0 0 1) substrates and the results were summarised in an experimental CVD stability diagram. The Bi2O2.33 phase was found to grow epitaxially on both substrates. Keywords: Halide-CVD, ALD, Bismuth titanate, Bi4Ti3O12, Titanium oxide, TiO2, Bismuth oxide, Bi2O2.33, Epitaxy, QCM. 

State-of-the-art SrS based blue thin film electroluminescent (TFEL) phosphors, namely, SrS:Ce, SrS:Cu, and SrS:Ag,Cu,Ga, were characterized by combined ion beam analysis techniques and photoluminescence (PL) and electroluminescence (EL) measurements. A selection of different elements were ion implanted into SrS:Ce and SrS:Cu thin films and their effects on the luminescence properties of the phosphor materials were examined. Impurities in thin films of SrS:Ce made by Atomic Layer Epitaxy (ALE) and SrS:Ce,Mn,Cl made by reactive evaporation were analyzed by various ion beam techniques, viz. Rutherford backscattering spectroscopy, elastic recoil detection analysis (ERDA), timeof- flight (TOF)-ERDA, nuclear resonance broadening, proton induced x-ray emission, particle induced γ-ray emission, and deuteron induced reactions. All samples were of high purity, with Sr/S or (Sr+Mn)/S ratio close to unity. The major impurities in the thin film bulk were H, C, and O. In ALE SrS:Ce, good EL performance correlated with an overall low impurity content, in particular low C content. In intentionally codoped SrS:Ce,Na samples, Na was found to concentrate at the phosphor−insulator interface. The EL performances of the corresponding TFEL devices were poor. For the reactively evaporated SrS:Ce,Mn,Cl samples, the EL performances were better than the ALE SrS:Ce devices despite their higher levels of H, C, and O impurities. Ion implantation of ALE SrS:Ce thin films with Na, K, Ag, P, Ga, F, and Cl showed that positive ions may be more favorable as codopants than negative ions. Implantation of F resulted in about 10 nm blue shift of the emission, but annealing above 500 °C quenched the PL intensity. K implantation enhanced the PL intensity by a factor of two when annealed at 800 °C, and even greater enhancement was achieved with Ag implantation under the same annealing conditions. Blue shift of about 10 nm was also present in these samples as a result of high temperature annealing. Implanted SrS:Ce,Ag exhibited the best decay value (SN=22 ns) ever reported for ALE SrS:Ce thin films and the EL results are encouraging. Implantation of Cl and Na did not improve the PL of SrS:Ce thin films, while P and Ga quenched the luminescence. Negative ions seem to be more favorable as codopants for SrS:Cu, perhaps as a result of the presence of S vacancies. Implantation of F, Cl, and O enhanced the PL emission. SrS:Cu,Cl also exhibited a pronounced blue shift of the green emission band. Implantation of B also improved the emission intensity, but the oxidation state of B is yet undetermined. Implantation of Ag, Al, and Ga did not improve the PL intensity. Nevertheless, a few SrS:Cu,Ag films showed the blue band due to Ag emission. It was verified that both blue and green emission may be observed in SrS:Cu at room temperature. The color gamut is determined by the intensity ratio of two emission bands located at 460 (H band) and 520 nm (L band). The L band is attributed to the emission of isolated Cu+ ion substitutes octahedral coordinated Sr but at off−center position and the H band to Cu+ at a different site symmetry. Luminescence of SrS:Cu is likely to follow a three−level mechanism as characterized by the increased decay time with decreasing temperature. The completely green luminescence of SrS:Cu with a single band located above 520 nm possibly originates from Cu pairs and aggregated Cu centers. The blue luminescence of SrS:Ag,Cu,Ga may be related to Ag+−Ag+ pairs or Ag+−(Cu+) associated centers, or both. 

In vapour phase deposition, the knowledge and control of homogeneous and heterogeneous reactions in connection to precursor design may lead to the deposition of the desired material; structure or phase. This thesis is a document attempting to increase the knowledge of film growth in the B-N system. In the present work, surface processes like adsorption, abstraction, migration and nucleation have been modelled on an atomic scale using density functional theory (DFT). The systems studied are mainly cubic and hexagonal boron nitride surfaces ((c-BN) vs. (h-BN)), but also the α-boron (001) surface. It has been shown that DFT and a cluster approach is a reliable tool in modelling boron nitride surfaces and surface processes, provided that certain functionals, basis sets and geometrical constraints are used. By using surface stabilisers such as H species in an electron- or radical-rich environment, it has been shown that i) the structure of cubic boron nitride surfaces can be sustained, and ii) c-BN may nucleate on the h-BN (001) basal plane. Furthermore, the nucleation of c-BN from arbitrary and experimental growth species is energetically preferable over a continuous growth of h-BN on the h-BN (001) edges. An atomic layer deposition (ALD) process for boron nitride was developed. It resulted in turbostratic (t-BN), transparent, well-adherent and almost atomically smooth BN films. However, with the cubic phase of boron nitride absent in the ALD films, more effort needs to be put into both the theoretical and the experimental branches of this field of science.

The atomic layer epitaxy (ALE) technique was used to grow thin films of binary metal oxides Al2O3, Ga2O3 and La2O3, and ternary metal oxides LaNiO3, LaCoO3, LaAlO3, and LaGaO3. In addition, another type of mixed-oxide, viz. phosphorus-doped Al2O3 was studied. The binary oxides Ga2O3 and La2O3 and all the ternary oxides were deposited by the ALE method for the first time. New ALE processes were developed for the undoped and doped Al2O3 films. The thin films were characterized by a wide range of methods for structural and surface analysis, including XRD, FTIR, XPS, AFM, XRF, RBS, TOF-ERDA, and SIMS. A review of previous work on these trivalent metal oxide thin films is presented by way of background. The Al2O3 and Ga2O3 films deposited from metal beta-diketonates and ozone were of high quality: stoichiometric, uniform, dense, and free of any significant contamination. However, the La2O3 films contained an excess of oxygen, due to the carbonate-type impurity that was detected. All Al2O3 and Ga2O3 films were amorphous, but polycrystalline, cubic La2O3 was formed at temperatures above 300 oC. Hexagonal La2O3 film was obtained by annealing the as-deposited amorphous and cubic La2O3 films. Simultaneously, the carbon content in the films was reduced. In contrast to the stable Al2O3 and Ga2O3 films, the cubic and hexagonal La2O3 films were chemically unstable and reacted with ambient air, transforming to LaO(OH) and La(OH)3, respectively. The perovskite-type oxides LaNiO3, LaCoO3, LaAlO3, and LaGaO3 were deposited using metal beta-diketonates and ozone as precursors. No optimal ALE process could be demonstrated for the LaNiO3 and LaCoO3 films, which were non-uniform in thickness and either consisted of separate oxide layers or contained an excess of the transition metal. The LaAlO3 and LaGaO3 film growth, in contrast, was well-controlled yielding stoichiometric, uniform, and smooth films, demonstrating the potential of the ALE technique for producing the more complex ternary oxide films. All the as-deposited ternary oxide films were amorphous but crystallized with cubic structure when post-annealed ex situ. After annealing high-quality, epitaxial LaAlO3 and LaGaO3 films were obtained on lattice-matched perovskite-type substrates while randomly or slightly oriented films were obtained on non-lattice matched substrates. The films were relatively pure and only small amounts of common impurities, carbon and hydrogen, were detected. Phosphorus-doped Al2O3 films were deposited from AlCl3, P2O5 or trimethylphosphate, and water. The phosphorus content could be controlled by the phosphorus doping ratio: when the P/Al atomic ratio was below 1.0, the films consisted of both Al2O3 and AlPO4, whereas above that they contained predominantly AlPO4. Phosphorus was uniformly distributed when the phosphorus content exceeded 5 at.%, and it was enriched on the film/substrate interface when the content was lower. 

Due to the constantly decreasing dimensions of electronic devices, the conventional dielectric material in transistors and capacitors, SiO2, has to be replaced by a material with higher dielectric constant. Some of the most promising candidates are tantalum oxide,Ta2O5, zirconium oxide, ZrO2 and hafnium oxide, HfO2. This thesis describes new chemical vapour deposition (CVD) and atomic layer deposition (ALD) processes for deposition of Ta2O5, ZrO2 and HfO2 using the metal iodides as starting materials. The layer-by-layer growth in ALD was also studied in real time with a quartz crystal microbalance (QCM) to examine the process characteristics and to find suitable parameters for film deposition. All the processes presented here produced high-purity films at low deposition temperatures. It was also found that films deposited on Pt substrates generally crystallise at lower temperature, or with lower thickness, than on silicon and single-crystalline oxide substrates. Films grown on MgO(001) and a-Al2O3(001) substrates were strongly textured or epitaxial. For example, monoclinic HfO2 deposited on MgO(001) were epitaxial for deposition temperatures of 400-500􀀀C in ALD and 500-600􀀀C in CVD. Electrical characterisation showed that the crystallinity of the films had a strong effect on the dielectric constant, except in cases of very thin films, where the dielectric constant was more dependent on layer thickness. Keywords: CVD, ALD, Dielectric constant, Tantalum oxide, Ta2O5, Zirconium oxide, ZrO2, Hafnium oxide, HfO2, QCM. 

The shrinking feature sizes of basic electronic components, like transistors, memory cells and  metal wires, in integrated circuits sets high demands on both materials and thin film deposition  methods.One of themajor changes in thematerials has been the adoption of copper interconnects.  Due to the tendency of copper to react with silicon and insulators, use of a diffusion barrier is  necessary.With the currently used depositionmethods the future requirements for thin films, for  example, strict conformality and lowdeposition temperature, can not be fulfilled at the same time.  One of the most promising methods to be exploited in the near future is atomic layer deposition  (ALD). In the presentwork theALDmethod was used to deposit copper andmolybdenum films,  and some transition metal nitride films focusing on TiN.  Copper andmolybdenum films were deposited from the corresponding chlorides using zinc as a  reducing agent.The dissolution and outdiffusion of zinc caused problemswith the controllability  of the film growth.The deposition of copperwas also studied from various copper precursors and  reducing agents of different chemical nature. Altogether, the ALD of metals seemed quite  complicated, and although metals could be deposited their properties were only modest.  The main problem with the existing TiN ALD processes is that usually the films have to be  deposited at too high temperatures to obtain films with reasonably good properties. Fairly good  properties can be achieved by using zinc as an additional reducing agent, but it is known to  dissolve into silicon. In most of the previous ALD processes NH3 has been used as a nitrogen  source and but it is not very effective at low temperatures. In order to be able to grow TiN films  at temperatures tolerable in the future (below 400 °C), the deposition of TiN films was studied  from titanium halides by using twodifferent approaches. Nitrogen sourcesmore reactive thanNH3,  namely dimethylhydrazine, tert-butylamine and allylamine,were used.Another approach was to  deposit TiN films by using trimethylaluminium (TMA) as an additional reducing agent withNH3.  Each of the studied approaches produced filmswith better properties than those obtainedwith bare  NH3at lowtemperatures. The lowest resistivity (around 150:Scm)was obtained by usingTMA  as an additional reducing agent, but the carbon contamination was quite high (above 6 at.%).  Somewhat higher resistivities were obtainedwith dimethylhydrazine and amines (200 - 500 :S  cm), but with tert-butylamine the carbon contents were only minor (below 1 at.%).  In order to understand the reactionmechanisms involved inALD, the growth of Al2O3 andTi(Al)N  were studied bymeans of in situmass spectrometry. The results obtained gave information about  the possible surface reactions, and it seems that through the in situ mass spectrometry the ALD  reaction mechanisms can be identified.  

Catalyst supports with novel chemical and physical properties are needed for produc ing new families of catalytic materials. The goals of this work were to demonstrate the preparation of aluminium nitride type materials and evaluate their properties as catalyst supports. To obtain aluminium nitride in high-surface-area form suitable for catalyst applica tions, porous silica and alumina supports were surfaced with aluminium nitride by the atomic layer deposition (ALD) technique by repeating the separate, saturating reactions of gaseous trimethylaluminium (TMA) and ammonia. The reaction temperatures of TMA and ammonia were 150 and 550  C, respectively. Six reaction cycles led to an average growth of 2.4 aluminium atoms per cycle per square nanometre, and, according to low energy ion scattering, 74% coverage. The aluminium nitride species were shown to be evenly distributed on the silica. The aluminium nitride appeared amorphous in X-ray diffraction, but 27Al nuclear magnetic resonance (NMR) spectroscopy confirmed its for mation. Insight into the growth mechanism of aluminium nitride was obtained by investigation of the individual steps leading to the growth. The surface reaction products after the TMA and ammonia reactions were identified by diffuse reflectance Fourier transform infrared spectroscopy and 1H, 13C and 29Si NMR, and they were quantified by elemental analysis and 1H NMR. TMA reacted through ligand exchange with hydrogen atoms present on the surface in hydroxyl groups (OH) and amino groups (NHx), releasing methane, and further through dissociation in siloxane bridges, coordinatively unsaturated aluminium–oxygen pairs and nitrogen bridges. Steric hindrance imposed by the methyl ligands defined the saturation of the surface with adsorbed species; at saturation, there were five to six methyl groups per square nanometre. The ammonia reaction replaced the methyl groups present on TMA-modified surfaces with NHx groups (x = 2, 1 or 0). The results for the TMA reaction agreed quantitatively with the results obtained by others for the ALD growth of aluminium oxide thin films. A model was derived that relates the size and reactivity of the metal reactant to the growth per cycle of the oxide by ALD. The properties of the AlN/oxide materials as catalyst supports were evaluated for cobalt hydroformylation and chromium dehydrogenation catalysts. In the preparation of the catalysts from cobalt(III) and chromium(III) acetylacetonate by ALD, the factor defining the saturation of the reaction was the same as on the respective oxides: the steric hindrance imposed by the acetylacetonate (acac) ligands. Dissociative and associative reactions of the metal acetylacetonate reactants and of the Hacac released in ligand ex change reaction took place on the AlN/oxide supports. This was a disadvantage for the Co/AlN/silica catalysts, as the high acac=Co ratio of the surface complex led to the des orption of cobalt(II) acetylacetonate during catalytic testing in hydroformylation. After removal of the remaining acac ligands with ammonia, the activity of the Cr(III)/alumina and Cr(III)/AlN/alumina catalysts was evaluated in isobutane dehydro genation at 580  C. The aluminium nitride modification of the support decreased the dehydrogenation activity of the chromium catalysts. Pairs of chromium and oxygen ions seem to be required for active chromium catalysts, and replacing the neighbouring oxygen with nitrogen was a disadvantage. Although no improvements were observed in catalytic performance of the prepared catalysts relative to conventional systems, the information obtained will be useful in the future investigations of the growth of aluminium nitride and of other materials by ALD and in the identification of the active sites on dehydrogenation catalysts.

Determination of atomic concentration distributions in thin films is a key problem in materials science. The optimisation process of the thin film growth parameters in particular requires detailed information about the elemental concentrations of the main constituents and undesired impurities. So far, the characterisation methods capable of a depth-sensitive analysis of all elements have remained limited. In the research for this thesis, the concentration distributions of elements in surface layers were studied using heavy ion elastic recoil detection analysis (HI-ERDA). The analysis was expanded to include hydrogen and the heaviest elements. The energy-dependent detection efficiency of the time-of-flight energy telescope was determined for the lightest elements. The reliability of the concentration distributions obtained was seen to be strongly affected by the multiple scatterings of the ions and surface roughness of the sample. Both of them were studied by comparing Monte Carlo simulation results with the experimental ones. The surface topographies used in the simulations were determined with a scanning probe microscope. The analysis procedures developed were applied to characterise novel materials such as atomic layer deposited thin films used in future integrated circuit designs and pulsed vacuum arc deposited thin films, which are candidates for fusion reactor wall materials.

Atomic layer epitaxy (ALE) method was employed for the study of growth of binary and ternary metal oxide thin films. As background for the study, the basic principles of the ALE method are presented together with a review of existing ALE deposition processes and precursors for oxide thin films. 

The suitability of 13-diketonate type precursors (M(thd)3 M=Sc,Y,La; thd = 2,2,6,6-tetramethylheptanedione) and ozone were studied for ALE depositions of Group 3 oxides, namely Sc203, Y203 and La203. All three oxides could be deposited by a. self-limiting ALE process once a suitable deposition temperature was identified. The optimal deposition temperature was found to depend on the position of the self-limiting deposition region, but also on the impurity content, which increases at low deposition temperatures. Deposition rate of Sc203 was considerably higher from organometallic precursor, (C5H5)3Sc, than from 13-cliketonate precursor (0.75 A(cycle)-1 vs. (0.125 A(cycle)-1). 

In a. second set of experiments, the suitability of the ALE processes developed was tested for the deposition of ternary thin films, namely yttria-stabilised zirconia (YSZ) and lanthanum aluminate. Before these processes were applied, study was made of the deposition of ZrO2 from 13-diketonate and organometallic precursors at 200-500 °C. Furthermore, ALE deposited MgO films were tested for their suitability as buffer layers between silicon substrate and LaA103 film. Crystalline YSZ films were obtained regardless of the yttrium to zirconium ratio, whereas the LaA103 films were crystalline only after annealing at 900 °C. 
 

In this thesis, several oxide atomic layer deposition (ALD) processes were studied.  The main focus was on insulating materials to be used for microelectronic devices. At the  moment Al2O3, ZrO2, HfO2, rare earth oxides and their aluminates and silicates are the  most promising materials for metal oxide semiconductor field effect transistors  (MOSFETs). In MOSFETs the gate insulator should be deposited in a way that no more  than one or two monolayers of SiO2 would form during the whole device fabrication.  SrTiO3 and BaxSr1-xTiO3 are the most promising materials for dynamic random access  memories (DRAM).  A new method was developed for depositing oxide thin films on silicon by using  alkoxides as oxygen sources. Several binary and mixed oxides could be deposited. Of these  oxides Al2O3 was examined more closely and found to grow on silicon without a SiO2  interface layer. ZrxTiyOz thin films deposited using this method had a permittivity of 45 -  65 and a leakage current of 10-4 A/cm2 at 0.2 MV/cm.  A detailed understanding of thin film growth mechanism is important. A  quadrupole mass spectrometer (QMS) - quartz crystal microbalance (QCM) in situ  characterization system was developed in this study. QMS monitors the gas phase while  QCM the mass changes on the surface. By combining these two methods reaction  mechanisms can be examined in real time during the ALD growth in flow-type reactor  conditions. The main drawback of QCM is its temperature sensitivity and therefore  methods for compensating this effect were developed.  Reaction mechanisms were studied in various oxide processes such as Al2O3, TiO2,  ZrO2, ZrxTiyOz, and SrTiO3. Different titanium alkoxides were studied as ALD precursors.  The alkoxide group was found to have a strong effect on the ALD reaction mechanism and  the stability of the precursor against thermal decomposition. The oxygen source in all  binary processes studied was D2O and thus the main reaction byproduct was the deuterated  ligand while in the metal alkoxide - metal halide process two reaction paths were found.  In the SrTiO3 process at 250 oC the surface had no effect on the growth rate or  reaction mechanism, while at 325 oC especially the TiO2 surface catalyzed the reactions.  The change in the growth mechanism was attributed to the formation of crystalline SrTiO3  phase at 325 oC instead of a solid mixture of two oxides at 250 oC. In addition, in situ  methods showed a possibility to monitor the metal ratio of ternary films.  Key words: Atomic layer deposition, Oxide, Reaction mechanism, Quadrupole mass  spectrometry, Quartz crystal microbalance  

Smaller cellular telephones and more energy-efficient windows are just two examples of technological advances which call for new materials. Materials chemists seek to develop new materials, both out of pure curiosity to see which combination of elements and structures can be obtained and in efforts to produce materials, with specific properties. The starting materials (in solid, liquid or gaseous form) can then be combined and prepared in various ways. A chemical method that is gaining more attention for thin-film growth is Atomic Layer Deposition (ALD). This is a sophisticated type of vapor deposition in which the precursor gases are introduced separately into the reaction chamber. Silicon carbide (SiC) and cubic boron nitride (c-BN) are extremely hard diamond-like mater ials, both with a high potential for application within the modern microelectronics and tool industry. Hexagonal boron nitride (h-BN), with its graphite-like layered structure, is a promising ceramics material. Deposition of thin SiC and BN films from gaseous precursors was studied by theoretical and experimental methods. The chemical composition and atomic arrangement of a growing surface is important for vapor growth. The surface may be terminated (e.g., by hydrogen atoms) and adopt various geometrical structures. Reconstruction of unterminated SiC(0001) surfaces, as well as H abstraction from the corresponding H-terminated surfaces, were studied using quantum mechanical calculations. Elementary reactions for vapor growth of SiC and BN, and in situ incorporation of dopant and contaminant species into these surfaces were also investigated theoretically. Moreover, thin films of BN were deposited by means of laser-assisted ALD. The general goal has been to predict and/or explain experimental results by investigating growth mechanisms.

Temperature-programmed desorption (TPD), reduction (TPR) and oxidation (TPO) are thermoanalytical techniques for characterising chemical interactions between gaseous reactants and solid substances. The data collected by these techniques are commonly interpreted on a qualitative basis or by utilising simple, approximate kinetic methods. However, temperature-programmed techniques can also be regarded as transient response techniques and the experimental data can be utilised for dynamic modelling. This work comprises case studies on kinetic analysis of TPR, TPD and TPO related to the characterisation of heterogeneous catalysts. The emphasis is on methodological aspects and on assessing the potential of temperature-programmed data as a source of kinetic information. Kinetic analysis was applied to the TPR results for series of alumina-supported chromium oxide and vanadium oxide catalysts. Hydrogen was used as the reducing agent. Different kinetic models were tested against the experimental data and parameters were estimated. The chromium oxide and vanadium oxide contents of the catalysts were clearly reflected in the reduction behaviour and in the best-fit kinetic models and their parameters. The kinetic results suggested that reduction takes place via a topochemical mechanism, as growing domains, on both supported chromium and supported vanadium oxide catalysts with close to monolayer content. The interaction of hydrogen with a commercial nickel catalyst was studied in TPD experiments under continuous flow and ambient pressure. A model to account for the heterogeneity in the chemisorption interaction and for the readsorption was formulated and tested against experimental data. The heterogeneity was described by introducing a sufficient number of different adsorption states. The rapid readsorption occurring during TPD was taken into account by describing the intrinsic dynamics of an adsorption state as a quasi equilibrium adsorption/desorption between the gas phase and the surface. A model with two adsorption states of hydrogen was able to describe the experimental data with physically acceptable parameters in the temperature range of 323–673 K. The regeneration kinetics of a deactivated cracking catalyst was investigated on the basis of the experimental evolution rates of carbon monoxide and carbon dioxide during TPO. Different kinetic models were tested and kinetic parameters were estimated. A power-law kinetic expression with orders unity and 0.6 for coke and oxygen, respectively, was capable of describing the experimental data. In each case study, a phenomenological model was established and the kinetic parameters of the model were determined via nonlinear regression analysis in MATLAB ® environment. The results demonstrate that common catalyst characterisation data on reduction, desorption and oxidation collected in the temperature-programmed mode can fruitfully be subjected to detailed kinetic analysis. Mechanism and parameter identifiability require diversity in the experimental data, which can be achieved, for example, by applying multiple heating rates in experiments. Kinetic analysis extends the interpretability of temperature-programmed reactions in catalyst characterisation and it is potentially useful for the elucidation of fundamental reaction mechanistic information and establishing kinetic models for engineering applications.

Many materials exhibit interesting and novel properties when prepared as thin films. Thin film metal oxides have had an impact on the technological progress of the microelectronics mainly due to their electrical and optical properties. Since the future goes towards the nanometre scale there is an increasing demand for thin film deposition processes that can produce high quality metal oxide films in this scale with high accuracy.
This thesis describes atomic layer deposition of Ta2O5, HfO2 and SnO2 thin films and chemical vapour deposition of SnO2 thin films. The films have been deposited by employing metal iodides and oxygen as precursors. All these processes have been characterised with regards to important processing parameters. The films themselves have been characterised by standard thin film analysing techniques such as x-ray diffraction, scanning electron microscopy, atomic force microscopy and transmission electron microscopy. The chemical and physical properties have been coupled to critical deposition parameters. Furthermore, additional data in the form of electrical and gas sensing properties important to future applications in the field of microelectronics have been examined.
The results from the investigated processes have shown the power of the metal iodide based atomic layer deposition (ALD) and chemical vapour deposition (CVD) processes in producing high quality metal oxide thin films. Generally no precursor contaminations have been observed. In contrast to metal chloride based processes the metal iodide processes produces films with a higher degree of crystalline quality when it comes to phase purity, roughness and epitaxy. The use of oxygen as oxidising precursor allowed depositions at higher temperatures than normally employed in water based ALD processes and hence a higher growth rate for epitaxial growth was possible.
 

There is a current need to provide simple methods to place conformal, pinhole-free, nanoscale-thickness films on fine particles. Such processing can be done using atomic layer deposition (ALD) in a fluidized bed reactor, as will be shown in this thesis. This work is the first application of ALD to coat bulk quantities of fine powders. A fundamental understanding of the fluidization of fine cohesive particles at reduced pressure along with the ALD processing of these particles is the major focus of this thesis. The applications for such nanocoated particles are broad and can be found in many different areas of materials science including microelectronics, defense, biomedical, consumer products, advanced materials, and others. The minimum fluidization velocity (tint) of fine cohesive particles at reduced pressure can be calculated using a balance of forces method. Two additional forces are added to a general force balance on a particle with an upward gas flow under vacuum conditions. The final equation is a quadratic in unif, and can be used to accurately describe umf for a variety of particle sizes, shapes, and densities. Additionally, as fine particles are coated with an alumina film, the cohesive forces between the particles will change. For the particles of interest in this thesis, the change in the cohesive force is small. 

Experiments for alumina deposition on 1.510-4 m (150 itm) diameter nickel particles, several different sized boron nitride (BN) particles (from 510-6 to 1.5.104 m (5 to 150 1.1m) average diameter), and fine (-5.10-6m (5 p.m)) iron particles were conducted using trimethylaluminum (TMA) and water as dosing reagents at 450 K. Successful deposition of alumina films, with thickness controllable at the angstrom level, was observed based upon TEM imaging, ICP-AES, XPS, particle size distributions, surface area analysis, and WDS imaging. Nickel particles are coated quite easily. For BN platelet particles, a small exposure (3.25.102 Pas• (2.5106 L)) of the reagent gases will coat the edge planes only. A larger dose of 1.3.104 Pa•s (1108 L) will coat the entire particle. After 10 ALD cycles, the exposure can be lowered back to 1.3.102 Pa•s (1.106 L) as the film is then growing on alumina and not BN. Improved interfacial adhesion between epoxy and the filler material is noted for alumina-coated BN particles. Nanocoated iron particles show improved oxidation resistance relative to uncoated particles, as long as the film is greater than 2.5.10-9 m (25 A) thick. 
 

Hydrocinnamaldehyde and cinnamyl alcohol are produced in cinnamaldehyde hydrogenation. Both are of great practical importance with wide application in the fine chemicals, pharmaceuticals and perfume industries. In addition, cinnamyl alcohol is an important building block in organic synthesis. In view of the importance of these products, work was undertaken to prepare selective hydrogenation catalysts. Palladium, platinum and ruthenium catalysts supported on alumina and silica were prepared by gas phase deposition in an atomic layer epitaxy (ALE) reactor and by impregnation techniques. For study of the effect of the acidity of the support, Ru/β zeolite and Pt/β zeolite catalysts were prepared solely by impregnation. The materials were characterised by a variety of techniques. The catalytic properties of the catalysts were studied in cinnamaldehyde hydrogenation. Particle sizes were smaller for the ALE-deposited palladium than the corresponding impregnated samples. For the platinum and ruthenium samples, they were essentially the same for the two methods of preparation. Metal particles were small if a ligand exchange reaction occurred between metal precursor and support. In the ALE deposition, ligand exchange reaction and metal formation occurred for Pd(thd)2 and (CH3)3(CH3C5H4)Pt both on alumina and on silica. Ligand exchange and metal formation also took place for impregnated Pt catalysts with (CH3)3(CH3C5H4)Pt on both supports. In impregnation the interaction of Pd(thd)2 and Ru(thd)3 with the supports was associative adsorption. Palladium catalysts were more active than ruthenium and platinum catalysts, and the palladium catalysts prepared by ALE showed the highest initial activity in cinnamaldehyde hydrogenation because of the small particle size of metals obtained by ALE. Ruthenium on β zeolites were more active than platinum on β zeolites. The acidity of β zeolites affected the reduction behaviour of ruthenium and the particle size, which subsequently influenced the activity. As acidity increased, particle size decreased and the activity increased. The adsorption of cinnamaldehyde was preferably via the C=C bond on palladium catalysts, via the C=C and C=O bonds on ruthenium and via the C=O bond on platinum catalysts. Hydrocinnamaldehyde was the main product with all Pd catalysts. Ruthenium catalysts differ in selectivity. Only hydrocinnamaldehyde and 3-phenyl-1-propanol were produced with Ru/SiO2 prepared by ALE. Ruthenium on β zeolites were selective to hydrocinnamaldehyde. The other ruthenium catalysts formed a variety of hydrogenated products. The best choice of catalysts for cinnamyl alcohol formation is the impregnated 1.2 wt-% Pt/SiO2 catalyst with particle size of 4 nm. With use of this catalyst the selectivity toward cinnamyl alcohol was as much as 90% at conversion of 15%. For the formation of hydrocinnamaldehyde, 4.9 wt-% Pd/SiO2 is the best catalyst that was selective only to hydrocinnamaldehyde at conversion below 10%.

The continuous and aggressive dimensional miniaturization of the conventional complementary-metal-oxide semiconductor (CMOS) architecture has been the main impetus for the vast growth of IC industry over the past decades. As the CMOS downscaling approaches the fundamental limits, unconventional materials and novel device architectures are required in order to guarantee the ultimate scaling in device dimensions and maintain the performance gain expected from the scaling. This thesis investigates both unconventional materials for the gate stack and the channel and a novel notched-gate device architecture, with the emphasis on the challenging issues in process integration. High-κ gate dielectrics will become indispensable for CMOS technology beyond the 65-nm technology node in order to achieve a small equivalent oxide thickness (EOT) while maintaining a low gate leakage current. HfO2 and Al2O3 as well as their mixtures are investigated as substitutes for the traditionally used SiO2 in our MOS transistors. These high-κ films are deposited by means of atomic layer deposition (ALD) for an excellent control of film composition, thickness, uniformity and conformality. Surface treatments prior to ALD are found to have a crucial influence on the growth of the high-κ dielectrics and the performance of the resultant transistors. Alternative gate materials such as TiN and poly-SiGe are also studied. The challenging issues encountered in process integration of the TiN or poly-SiGe with the high-κ are further elaborated. Transistors with TiN or poly-SiGe/high-κ gate stack are successfully fabricated and characterized. Furthermore, proof-of-concept strained-SiGe surface-channel pMOSFETs with ALD high-κ dielectrics are demonstrated. The pMOSFETs with a strained SiGe channel exhibit a higher hole mobility than the universal hole mobility in Si. A new procedure for extraction of carrier mobility in the presence of a high density of interface states found in MOSFETs with high-κ dielectrics is developed. A notched-gate architecture aiming at reducing the parasitic capacitance of a MOSFET is studied. The notched gate is usually referred to as a local thickness increase of the gate dielectric at the feet of the gate above the source/drain extensions. Two dimensional simulations are carried out to investigate the influence of the notched gate on the static and dynamic characteristics of MOSFETs. MOSFETs with optimized notch profile exhibit a substantial enhancement in the dynamic characteristics with a negligible effect on the static characteristics. Notched-gate MOSFETs are also experimentally implemented with the integration of a high-κ gate dielectric and a poly-SiGe/TiN bi-layer gate electrode.

Challenges with integrating high-κ gate dielectric, retrograde Si1-xGex channel and silicided contacts in future CMOS technologies are investigated experimentally and theoretically in this thesis. pMOSFETs with either Si or strained Si1-xGex surface-channel and different high-κ gate dielectric are examined. Si1-xGex pMOSFETs with an Al2O3/HfAlOx/Al2O3 nano-laminate gate dielectric prepared by means of Atomic Layer Deposition (ALD) exhibit a great-than-30% increase in current drive and peak transconductance compared to reference Si pMOSFETs with the same gate dielectric. A poor high-κ/Si interface leading to carrier mobility degradation has often been reported in the literature, but this does not seem to be the case for our Si pMOSFETs whose effective mobility coincides with the universal hole mobility curve for Si. For the Si1-xGex pMOSFETs, however, a high density of interface states giving rise to reduced carrier mobility is observed. A method to extract the correct mobility in the presence of high-density traps is presented. Coulomb scattering from the charged traps or trapped charges at the interface is found to play a dominant role in the observed mobility degradation in the Si1-xGex pMOSFETs. Studying contacts with metal silicides constitutes a major part of this thesis. With the conventional device fabrication, the Si1-xGex incorporated for channel applications inevitably extends to the source-drain areas. Measurement and modeling show that the presence of Ge in the source/drain areas positively affects the contact resistivity in such a way that it is decreased by an order of magnitude for the contact of TiW to p-type Si1-xGex/Si when the Ge content is increased from 0 to 30 at. %. Modeling and extraction of contact resistivity are first carried out for the traditional TiSi2-Si contact but with an emphasis on the influence of a Nb interlayer for the silicide formation. A two-dimensional numerical model is employed to account for effects due to current crowding. For more advanced contacts to ultra-shallow junctions, Ni-based metallization scheme is used. NiSi1-xGex is found to form on selectively grown p-type Si1-xGex used as low-resistivity source/drain. Since the formed NiSi1-xGex with a specific resistivity of 20 µΩcm replaces a significant fraction of the shallow junction, a three-dimensional numerical model is employed in order to take the complex interface geometry and morphology into account. The lowest contact resistivity obtained for our NiSi1-xGex/p-type Si1-xGex contacts is 5×10-8 Ωcm2 , which satisfies the requirement for the 45-nm technology node in 2010. When the Si1-xGex channel is incorporated in a MOSFET, it usually forms a retrograde channel with an undoped surface region on a moderately doped substrate. Charge sheet models are used to study the effects of a Si retrograde channel on surface potential, drain current, intrinsic charges and intrinsic capacitances. Closed-form solutions are found for an abrupt retrograde channel and results implicative for circuit designers are obtained. The model can be extended to include a Si1-xGex retrograde channel. Although the analytical model developed in this thesis is one-dimensional for long-channel transistors with the retrograde channel profile varying along the depth of the transistor, it should also be applicable for short-channel transistors provided that the short channel effects are perfectly controlled.

Because ALD growth is based on the surface reactions of precursor molecules with a substrate, characterization of the surface species on porous substrates is very important. TG, DRIFTS/PLS, and 1H MAS NMR were found to be accurate techniques for the characterization of the surface species. With all these methods, not just the surface silanol groups but also the bulk internal silanols on silica are determined. While TG offers a relatively fast and simple method for the quantification of the total number of silanol groups, the numbers of different types of silanols cannot be determined. Combined DRIFTS/PLS method also allows the determination of the total number of silanol groups. Once calibration has been performed DRIFTS/PLS offers a fast analytical method, that can easily be applied to numerous kinds of samples in process analysis, for example. The advantage of 1H MAS NMR is the quantification of both isolated and hydrogen-bonded silanol groups on silica. It is, however, time-consuming and expensive. The present study showed that amino-functionalized silica surfaces can be prepared by ALD without solvent in a simple, conformal, and reproducible manner. Volatile aminopropylalkoxysilanes studied can be used as precursors because they do not decompose during vaporization or subsequent deposition. The reaction temperature affected the surface species on silica so that the use of relatively high reaction temperatures, i.e. 150-300 °C, led to sidereactions between the amino groups of bi- and trifunctional precursors and silanols groups or alkoxy groups of other precursor molecules on silica. Thus, deposition temperatures less than 150 °C (under pressure of 2-5 kPa) should be used to avoid side-reactions. The deposition and characterization of a single surface-saturated molecular layer on the surface allows study of various surface structures on porous substrates. The number of adsorbed precursor molecules (and terminal amino groups) on silica could be controlled between 1.0 and 3.0 molecules (or NH2 groups/nm2) through heat-treatment of silica (200-800 °C), choice of the precursor (APTMS, APTS, AAPS, APDMS, APDMES), and the number of reaction cycles (from 1 to 4) of gas-phase reactions of aminopropylalkoxysilane precursor and water. Such control is not possible with the liquid-phase methods currently applied to the preparation of aminofunctionalized silica surfaces. The highest amino group density was achieved with the bifunctional precursor, APDMS, on silica pretreated at 200 °C when a single surface-saturated molecular layer was deposited. Still higher amino group densities were obtained when sequential reactions of trifunctional APTMS, APTS or AAPS, and water were applied. A high-density aminopropylsiloxane network, which can be considered as a monolayer, could be deposited on silica because the surface was observed to be saturated with the precursor molecules. The results obtained from these experiments on porous substrates can further be applied to planar substrates, and valuable information on the surface chemistry and deposition process on surfaces can be obtained. As shown in this work, ALD enables a controlled deposition of functionalized surfaces, in addition to the oxides, sulfides, and nitrides earlier deposited on porous supports for catalyst applications. It was also shown that even more complex organic layers, such as polyimides, could be deposited on functionalized surfaces. In the present study polyimide structures were deposited on aminosilylated silica using PMDA and DAH as precursors. Low temperatures are especially desired to prevent decomposition of organic precursors. Pressure within the ALD reactor should be as low as possible, so as to provide the lowest vaporization and deposition temperatures. In addition to the present ALD applications, ALD has great potential for industrial applications in the future in completely novel areas where organic layers may be applied to planar or porous substrates. 

Atomic layer deposition was used to grow Al2O3, TiO2 and ZrO2 thin films. The mechanism of film growth was studied with a quadrupole mass spectrometer (QMS) and a quartz crystal microbalance (QCM). A brief literature review on the ALD growth of binary oxides and on in situ studies on selected oxide processes is presented as background. The effect of water dose on the growth of Al2O3, TiO2, ZrO2 and Ta2O5 was studied. The increased water dose increased the growth rate in all cases. According to the analysis data, it did not seem to affect the film properties. Al2O3 and TiO2 films were deposited for corrosion protection studies. Although ALD grown Al2O3 is amorphous and there should not be any pinholes, it did not work as a protective coating against corrosive media. TiO2 films are crystalline, and the corrosive media was able to penetrate through the coating and cause corrosion of the underlying substrate. The good characteristics of both materials were combined as Al2O3 – TiO2 multilayer structures, which resisted corrosion better than the single oxides. The reaction mechanism studies on Me2AlCl–D2O and TiCl4–D2O ALD processes were carried out using a combination of QMS and QCM integrated to the ALD reactor. QMS gives information about the gaseous products formed in the surface reactions involved in the film growth. QCM, in turn, monitors changes in the film mass. ZrO2 films were deposited using water and new alkoxide precursors: Zr(dmae)4, Zr(dmae)2(OtBu)2 and Zr(dmae)2(OiPr)2. The reaction mechanism was studied using QMS–QCM. The precursors seem to decompose, so the growth can not be regarded as ideal ALD. 

Thin films play an important role in science and technology today. By combining different materials, properties for specific applications can be optimised. In this thesis growth of copper, copper(I) oxide and copper(I) nitride on two different substrates, amorphous Sio, and single crystalline a-A1,0, by the so called Atomic Layer Deposition (ALD) techniques has been studied. This technique allows precise control of the growth process at monolayer level on solid substrates. Other characteristic features of ALD are that it produces films with excellent step coverage and good uniformity even as extremely thin films on complicated shaped substrates. Alternative deposition schemes were developed for the materials of interest. It was demonstrated that use of intermediate water pulses affected the deposition pathways considerably. By adding water, the films are thought to grow via formation of an oxide over-layer instead of through a direct reaction between the precursors as in the case without water. For growth of copper(I) nitride from Cu(hfac), and ammonia no film growth occurred without adding water to the growth process. The Cu, N films could be transformed into conducting copper films by post annealing. In copper growth from CuCl and H, the water affected film growth on the alumina substrates considerably more than on the fused silica substrates. The existence of surface -OH and/or -NH groups was often found to play an important role, according to both theoretical calculations and experimental results.

Cu(ln,Ga)Se, (CIGS) thin film solar cells contain a thin layer of CdS. To avoid toxic heavy-metal-containing waste in the module production the development of a cadmium-free buffer layer is desirable. This thesis considers alternative Cd-free buffer materials deposited by Atomic Layer Deposition (ALD). Conditions of the CIGS surface necesvury for ALD growth are investigated and the heterojunction interface is characterized by band alignment studies of ZnO/CIGS and In,S3/CIGS interfaces. The thesis also includes investigations on the surface modification of the CIGS absorber by sulfurization. According to ALD theory the growth process is limited by surface saturated reactions. The ALD growth on CIGS substrates shows nucleation failure and generally suffers from surface contaminations of the CIGS layer. The grade of growth disturbance varies for different ALD precursors. The presence of surface contaminants is related to the substrate age and sodium content. Improved growth belavior is demonstrated by different pretreatment procedures. The alignment of the energy bands in the buffer/absorber interface is an important parameter for minimization of the losses in a solar cell. The valence band and conduction band offsets was determined by in situ X-ray and UV photoelectron spectroscopy during layer by layer formation of buffer material. The conduction band offset (AEA) should be small but positive for optimal solar cell electrical performance according to theory. The conduction band offset was determined for the ALD ZnO/CIGS interface (AE = -0.2 eV) and the ALD In,SICIGS interface (AEA = -0.25 eV). A high temperature process for bandgap grading and a low temperature process for surface passivation by post deposition sulfurization in H,S were investigated. It is concluded that the high temperature sulfurization of Culno.spazSe, leads to phase separation when x>0. The low temperature process did not result in enhanced device performance. 

Atomic layer deposition (ALI)) is gaining attention as a promising method for depositing high quality, conformal, ultra thin films used in the fabrication of advanced microelectronic devices. Atomic layer deposition relies upon self-terminating surface reactions that limit the growth in most cases to no more than one atomic layer at a time. Since the ALE) pmess is very sensitive to surface characteristics, by modifying the substrate, we can achieve patterned deposition of thin films that can be useful for semiconductor processing. The ability to perform area selective ALD would provide a number of benefits such as a Suction in the cost and number of process steps required for pattern-wise deposition of materials, elimination of possible substrate and device damage induced by the traditional etching of thin films, and the ability to directly pattern materials that are difficult to etch. We have explored the possibility of using vapor phased deposited self-assembled monolayers as a blocking layer to develop area selective atomic layer deposition techniques (AS-ALD). 
We have modified Si surfaces with several types of alkoxy- or chlorosilane-based monolayers which have different head groups and backbone chain lengths. We have varied several reaction parameters such as temperature, pressure and trace amount of water to optimize each self-assembled monolayer forming process. After completing the optimization of vapor delivered SAMs to get ordered hydrophobic surfaces, area-selectivity of modified samples towards ALD was investigated by X-ray photoelectron spectroscopy. We found that deposition of H102 and Pt can be blocked effectively, achieved by deactivation of Si02 substrates by attaching octadecyltrichlosilane (ODTS) through vapor phase deposition over 2 day's silylation exposure. Also, we have successfully blocked the ALD process by several other organic molecules adsorbed on the surfaces. To investigate the characteristics of self-assembled monolayers by the vapor delivery method in more detail, several analytical techniques including, ellipsometry, water contact angle analysis, and multiple internal reflection Fourier transform infrared (MIR-FTIR) spectroscopy have been used We found that to block the substrate from the ALD precursors, a well ordered, densely packed structure is necessary.
 

The purpose of this research has been to explore noble metals prepared using thermal atomic layer deposition (ALD) for gate electrode applications in CMOS technology. ALD Ruthenium and Rhodium metal films have been focused due to their high work function, low resistivity of their oxidation forms. Study includes 1) ALD Ru process, 2) ALD Ru nucleation behaviors, 3) area-selective ALD Ru process, 4) ALD Ru work function modification, 5) ALD Rh process and work function. For item 1), ALD Ru films were formed using ruthenocene and oxygen as precursors. ALD window was discovered within 310 to 340°C from ALD Ru growth rate dependence on deposition temperature. Self-limiting reaction behavior was shown from growth rate versus precursor dose time. ALD Ru resistivity was measured to about 20~30µΩ⋅cm and Auger spectroscopy result was consistent with metallic Ru. Foe item 2), ALD Ru was deposited on chemical SiO2, thermal SiO2, and H-terminated Si surfaces. From thickness vs. ALD cycle, growth rates of Ru on those substrates were similar while as initial nucleation periods were different. Contact angle values of initial substrates showed hydrophilicity was related to the incubation time difference between substrates. ALD Ru Nucleation behavior was investigated on H-terminated Si during incubation period and growth model was proposed. For item 3), extending ALD Ru nucleation study, area-selective ALD Ru process was demonstrated. Octadecyltrichlorosilane was used to make surface very hydrophobic inhibiting nucleation. Metal-oxide-semiconductor (MOS) capacitor was fabricated using selective deposition process and spectroscopic (XPS) and electrical (capacitance voltage) measurements of the capacitor confirmed the viability of selective deposition. For item 4), ALD Ru work functions on SiO2 and HfO2 was measured and it turned out that Ru work functions on high-k dielectrics are smaller than on SiO2 possibly due to dipole formation at metal/dielectric interface. Organic self-assembled monolayers were applied on high-k dielectric surfaces prior to ALD Ru deposition to modify the dipole at the interface. ALD Ru work functions increased with amine-terminated self-assembled monolayer and decreased with vinyl-terminated monolayer. For item 5), ALD Rh has rarely been studied even though Rh is a candidate material for PMOS gate electrode. We investigated and developed successful ALD Rh process using Rhodium acetylacetonate and oxygen as precursors. ALD window was found at 280 to 310°C. It was shown that ALD Rh resistivity decreased with deposition temperature having minimum (~10µΩ⋅cm) at 300°C. XPS result was consistent with metallic Rh.

One of the major challenges of the next generations of integrated circuits concerns interconnections. The copper lines connecting the active components will delimit the performance and reliability of future microprocessors. The performance of the interconnects depends on the effective resistivity of the metallic line and the dielectric coefficient of the interline insulation. To increase performance, new low permittivity insulators and reduced thickness diffusion barriers are introduced. In addition, the reduction in dimensions leads to a modification of the resistivity of the copper and an increase in the current density. The objective of this thesis was to analyze this increase in resistivity and to deepen the knowledge of electromigration in the lines of interconnections in which these new materials are integrated. On the one hand, a very detailed study of the increase in resistivity in copper when the minimum dimensions of the lines become of the same order of magnitude as the mean free path of the electron has been carried out. A model based on the work of [Mayadas and Shatzkes, 1970] has been proposed. The increase in resistivity is explained on the one hand by a diffusion of electrons at the grain boundaries of copper, a first parameter (1-Rm) is introduced corresponding to the probability of passage of the electron at the grain boundaries and on the other hand by a diffusion of electrons on the walls of lines. A second parameter p corresponding to the probability of elastic reflection of the electron on the walls accounts for this phenomenon. The electrical characterization of lines with a width of less than 50 nm was carried out using a method of electrical extraction of the resistivity and the surface of copper. The uncertainty on the values obtained was estimated. We have indeed observed an increase in resistivity. The Mayadas model makes it possible to correctly model the experimental data. On the other hand, we have shown that the two electron scattering mechanisms have the same width dependence and we have mentioned the difficulty of differentiating between the two mechanisms. Complementary studies will have to be carried out to characterize the two phenomena more finely: measurement of grain size, variation in the thickness of the lines with more or less long CMPs, measurement of the roughness of the flanks by AFM, etc. On the other hand, a global study of electromigration in the interconnections of the next generations of interconnection has been carried out. The interpretation of the results was accomplished using the modeling of [Korhonen et al., 1993]. This takes into account the confinement of copper in the phenomenon of diffusion of the metal under the effect of a "wind" of electrons via an effective modulus B. This modulus was determined for our geometries of line and for our material parameters thanks to the CASTEM finite element software. Classical electromigration studies (temperature and current stress, post-mortem SEM observation) have enabled us to extract the electromigration behavior of new materials used to increase the electrical performance of interconnect lines. We first observed a decrease in electromigration performance with the use of a porous dielectric. The properties of this material are a low electric permittivity in order to decrease the interline coupling and a low Young's modulus due to the presence of porosity in the dielectric. The simulation of B shows a reduction in the effective modulus with the use of this material. The analytical solutions of the Korhonen model then indicate an increase in lifetime for a lifetime limited by the nucleation of a cavity and a decrease in lifetime in the case of a lifetime limited by the growth of a cavity. Our experimental results indicate an activation energy identical to that obtained with a dense insulator indicating an identical diffusion site in both cases but a lifetime one decade lower in the case of the porous material. The lifespan therefore seems to be limited by the growth of a cavity. The interpolations to the operating conditions indicate that the porous material does not meet the requirements of the 65 nm node in terms of reliability. The impact of the diffusion barrier (PVD TaN/Ta, ALD TaN, CVD TiN) was determined. An almost similar activation energy is observed for each variant possessing a TaN/Ta PVD barrier (about 0.8 eV). On the other hand, an increase in the dispersion of the results is obtained by reducing the thickness of the line and a reduction in the dispersion by carrying out an H2 plasma before deposition of the barrier. The activation energy obtained for the TiN barrier is lower (0.62 eV) and the dispersion of the results greater than in the case of a TaN/Ta barrier. It is difficult to interpret this activation energy and we believe that this energy reduction may be related to a non-optimized copper deposition (CVD). On the other hand, a lifespan is observed under test conditions that are much longer than the PVD TaN/Ta barrier. This result could be related to the greater thickness of the TiN CVD barrier and therefore a higher copper confinement. The ALD barrier has the thinnest barrier width (3.5 nm) and therefore the lowest effective resistivity. From the electromigration point of view, we observe a multimodal behavior of the fractures which prevents us from correctly extracting an activation energy. By carrying out extrapolations to operating conditions, we show that none of these three barriers makes it possible to obtain the desired performance with a porous dielectric for the 65 nm node. We then showed the impact of the upper corner of copper lines on electromigration using new morphological characterizations: in situ SEM and EBSD. The first makes it possible to follow the evolution in real time of the growth of a cavity while the second makes it possible to correlate the presence of a cavity with the crystalline orientations on either side of the cavity. We have thus shown that cavity nucleation is favored at the intersection of a grain boundary and the diffusion barrier. We could not conclude on the presence of a crystalline orientation favoring electromigration. The reduction in the dimensions of the lines does not lead to a marked drop in lifetimes, but an increase in the dispersion of breaks is observed. New processes have been evaluated with the aim of improving the electromigration behavior of interconnects. The use of copper alloy made it possible to obtain a higher activation energy (1.3 eV), but in return a higher resistivity of the copper was measured. Different self-positioned metallic barrier deposition processes allowing to reduce the interline capacitance have been evaluated. These metal upper barriers have made it possible to significantly improve electromigration performance. An activation energy close to the theoretical value of solid copper was obtained, meaning a complete blockage of the diffusion paths at the interfaces. The phenomenon of electromigration at the operating conditions of the integrated circuits is then almost nil. On the other hand, the presence of a metal barrier is not a sufficient condition to obtain such performance. We also note a possible increase in the resistivity of the copper by diffusion of the elements of the barrier in the copper matrix. These studies were conducted on single level structures. To complete it, it is necessary to qualify double-level structures, one of the particularities of which is the presence of a diffusion barrier at the bottom of the via. In situ SEM characterization can be coupled with EBSD analysis and conducted on narrower line widths. The metal barriers seem very promising and additional studies will have to be carried out with in particular more statistics and characterizations of the post mortem defects.

KELLY, MICHAEL JASON. Reactions of High-k Gate Dielectrics: Studies in Hafnium, Zirconium, Yttrium, and Lanthanum-based Dielectrics and in-situ Infrared Results for Hafnium Dioxide Atomic Layer Deposition. (Under the direction of Dr. Gregory N. Parsons.) According to the International Technology Roadmap for Semiconductors (2004) integrating a high dielectric constant (high-k) material into the gate stack will be necessary within the next two years (i.e., by 2007) to maintain the rate of scaling that has come to characterize the microelectronics industry. This work presents results for Y-, Zr-, Hf-, and La-based high-k gate dielectrics prepared by ex-situ oxidation of sputtered thin metal films and for HfO2 prepared by atomic layer deposition (ALD). The kinetics of substrate consumption during formation of yttrium silicate thin films were studied. We find results consistent with high-k dielectric formation by a two-step process in which yttrium metal reacts with the silicon substrate to form a metal silicide which is then oxidized to form the yttrium silicate dielectric. In other experiments, we show flatband voltage shifts of -0.2 and -0.95V in devices containing Zr-based dielectrics formed by oxidation of 8Å of Zr metal on Si at 600° C in N2O for 15 and 300s, respectively. Silicon oxidized in the same environment does not show this shift. The fixed charge scales with EOT for these films and is consistent with charge generation due to disruption of the SiO2 network by metal ions. Zr-based dielectrics exhibit this effect more strongly than Hf-based dielectrics. We show that La-based dielectrics absorb atmospheric H2O and CO2, and that reactions between these materials and deposited silicon electrodes are accelerated when H2O or other OH species are present at the interface. We show that the electrical properties of gate stacks having Ru and RuO2 electrodes in contact with PVD Y-silicate are more stable during thermal anneal than similar gate stacks having PVD ZrO2 or CVD Al2O3 dielectrics. For this work, we configured a Fourier transform infrared spectrometer for in-situ attenuated total reflection measurements and investigated ALD deposition of HfO2. We report the direct reaction of tetrakis(diethylamino) hafnium (TDEAHf) with SiH groups on HF-last Si. Island growth of HfO2 occurs, and SiH features are still present and shrinking after 200 cycles. To the best of our knowledge, these are the first in-situ FTIR results presented for atomic layer deposition using TDEAHf/H2O chemistry.

Light alkenes, such as propene and butenes, are important intermediates in the manufacture of fuel components and chemicals. The direct catalytic dehydrogenation of the corresponding alkanes is a selective way to produce these alkenes and is frequently carried out using chromia/alumina catalysts. The aim of this work was to obtain structure–activity information, which could be utilised in the optimisation of this catalytic system. The properties of chromia/alumina catalysts were investigated by advanced in situ and ex situ spectroscopic methods, and the activities were measured in the dehydrogenation of isobutane. The dehydrogenation activity of chromia/alumina was attributed to coordinatively unsaturated redox and non-redox Cr3+ ions at all chromium loadings. In addition, the oxygen ions in the catalyst appeared to participate in the reaction. The reduction of chromia/alumina resulted in formation of adsorbed surface species: hydroxyl groups bonded to chromia and alumina were formed in reduction by hydrogen and alkanes, and carbon-containing species in reduction by carbon monoxide and alkanes. Prereduction with hydrogen or carbon monoxide decreased the dehydrogenation activity. The effect by hydrogen was suggested to be related to the amount of OH/H species on the reduced surface affecting the number of coordinatively unsaturated chromium sites, and the effect by carbon monoxide to the formation of unselective chromium sites and carboncontaining species. The chromia/alumina catalysts were deactivated with time on stream and in cycles of (pre)reduction–dehydrogenation–regeneration. The deactivation with time on stream was caused mainly by coke formation. The nature of the coke species changed during dehydrogenation. Carboxylates and aliphatic hydrocarbon species formed at the beginning of the reaction and unsaturated/aromatic hydrocarbons and graphite-like species with increasing time on stream. The deactivation in several dehydrogenation– regeneration cycles was attributed to a decrease in the number of actives sites, which was possibly caused by clustering of the active phase into more three-dimensional structures. Acidic hydroxyl species of exposed alumina support may have contributed to the side reactions observed during dehydrogenation. Chromium catalysts prepared on unmodified alumina and on alumina modified with basic aluminium nitride-type species were compared in an attempt to increase the activity and selectivity in dehydrogenation. However, the presence of nitrogen in the catalyst was not beneficial for the dehydrogenation activity. A kinetic model was derived for the rate of dehydrogenation of isobutane on chromia/alumina. The dehydrogenation results were best described by a model with isobutane adsorption, possibly on a pair of chromium and oxygen ions, as the ratedetermining step. Satisfactory description of the reaction rate depended upon inclusion of the isobutene and hydrogen adsorption parameters in the mathematical model. The activation energy of the rate-determining step was estimated to be 137±5 kJ/mol. 

Transition metal nitrides, metal silicides, and metal-silicon-nitrides are considered the most promising diffusion barrier materials for next generation ultra large scale integration (ULSI) microelectronics. The semiconductor industry has long used Ti, Ta, and W based materials, and their material properties have been very well studied. Recently, tantalum-based materials have been attracting particular interest. The barrier properties of materials based on other transition metals have been little studied. In this work, tantalum nitride films were deposited, with four new reducing agents used to reduce tantalum and obtain the desired TaN phase. As well, the deposition of niobium and molybdenum nitride films was investigated. All films were deposited by the atomic layer deposition (ALD) method, which ensures excellent conformality and large area uniformity of the films.

The problem in depositing TaN films by ALD is that in volatile tantalum precursors the tantalum usually exists in oxidation state +V which is difficult to reduce to the +III state needed in cubic TaN. The new reducing agents examined in this study were trimethylaluminum (TMA), tert-butylamine (tBuNH2), allylamine (allylNH2), and tris(dimethylamino)silane (TDMAS). In addition to reducing tantalum, TMA also acted as a carbon and aluminum source, tBuNH2 and allylNH2 as nitrogen sources, and TDMAS as a silicon precursor.

ALD of niobium nitride and molybdenum nitride films was studied at lower temperatures than reported earlier. Both NbNx and MoNx films were deposited from the corresponding metal chloride precursors (NbCl5 and MoCl5, respectively) using ammonia as nitrogen source. No additional reducing agent was required.

The deposition parameters, compositions, crystallinity, and electrical properties were studied for all deposited films. Barrier characteristics were investigated for Ta(Al)N(C), NbNx, and MoNx films. The work function values were measured for Ta(Si)N films deposited at two different temperatures.
 

Noble metal thin films have several potential applications for example in integrated circuits. In this work, new noble metal processes have been developed for atomic layer deposition (ALD), which is a gas phase thin film deposition method based on alternate saturative surface reactions. The self-limiting film growth mechanism of ALD leads to films with excellent conformality and good large area uniformity. In addition, the film thickness can be accurately controlled by the number of the applied growth cycles. ALD processes for ruthenium, platinum, iridium, rhodium, and palladium were studied. All the processes are based on the reaction of the metal precursor with oxygen, the process temperatures being in the range of 200–450 °C. Metallic ruthenium films with low resistivity (< 20 μΩ⋅cm) and low impurity contents (< 0.2 at.% H, < 0.2 at.% C, and < 0.4 at.% O) were grown from a cyclopentadienyl precursor RuCp2. Ruthenium films grown from a β-diketonato precursor Ru(thd)3 had higher resistivities, higher impurity contents, and longer incubation time for onset of the film growth. High quality platinum films were grown from MeCpPtMe3. The films had strong (111) orientation even at the lowest growth temperatures. Iridium films with low resistivities (< 18 μΩ⋅cm), low impurity contents (< 1.0 at.% H, < 0.3 at.% C, and < 0.5 at.% O), and smooth surface morphology were grown from Ir(acac)3 and oxygen. Metallic rhodium films were grown from Rh(acac)3 and oxygen. ALD of palladium was also studied but self-limiting film growth was not obtained. Reaction mechanism studies were performed in order to gain better understanding of the chemistry in the studied noble metal ALD processes. It was found that adsorbed oxygen atoms react with the ligands of the noble metal precursor during the metal precursor pulse. Unreacted ligand species that remain on the surface after the metal precursor pulse react with oxygen during the following oxygen pulse. The main reaction by-products detected during the both reaction steps were water and carbon dioxide. 

In the present thesis it has been demonstrated that the nanoporous membranes of anodic aluminium oxide have a large number of applications for fabrication of nanostructures. The AAO membrane can be manufactured in large quantities and there are numerous possibilities for tailoring the di mensions of the membrane well into the mezo-porous size range (2-50 nm). The thickness of the membrane can be monitored by changing the anodiza tion time. The inter-pore distance as well as the pore diameters can be tai lored by changing the anodization voltage. The degree of order of the pores can be determined by changing the anodization parameters, such as the elec trolyte concentration and temperature. By exposing the fabricated AAO membranes to phosphoric acid the pores can be further widened (of course the pore diameter is limited by the inter-pore distance). After the anodization the membranes can be left on the aluminium substrate or the aluminium can be etched away in mercury chloride, depending on the application of the membrane. There are numerous ways to fabricate nanoparticles, by wet-chemical techniques, by physical means in the gas phase etcetera. Two of the great challenges in the fabrications of nanoparticles are: 1. to get a narrow size distribution of the nanoparticles. 2. To get immobilized nanoparticles on different surfaces. In this thesis both these challenges were fulfilled for pal ladium nanoparticles, Prussian blue nanoparticles and for copper nanoparti cles. All fabricated by different techniques and in all cases it was possible not only to get a narrow size distribution, but also to tailor the sizes in wide ranges. Pd and Prussian blue nanoparticles were fabricated through sequen tial electroless techniques, while copper nanoparticles were made through Atomic Layer Deposition (ALD). The fabrication of monodisperse nanotubes are also described in this thesis. AAO membranes have been used as templates for deposition of Prus sian blue nanotubes as well as niobium oxide nanotubes. The Prussian blue nanotubes were polycrystalline and fabricated using the same sequential techniques as was used with Prussian blue nanoparticles. The niobium oxide nanotubes were amorphous and fabricated using ALD. In both cases the outer diameter of the nanotubes were strictly restricted by the pore diameters of the AAO membrane used as template. The thicknesses of the tube walls could be tailored in both cases and by using a template which was well or dered with respect to the pores. The produced nanotubes could be arranged in a well ordered fashion as well. By depositing nanoparticles or thin films of materials along the pore walls, the surface properties of the AAO membranes could be modified. If it is desirable to instead achieve other bulk properties, but still benefit from the well ordered pore arrangement, the pattern must be transferred to another material. Pattern transfer has been achieved using MeV ion beam lithogra phy. AAO membranes were used as masks and heavy ions were irradiated through the masks onto an underlying substrate of another material. Prior to irradiation a careful alignment of the mask with respect to the ion beam were necessary. This was possible by a technique described in this thesis, which involves a gold marker layer and Rutherford backscattering spectroscopy (RBS). After alignment and irradiation it was possible to selectively etch in the regions of the underlying substrate which had been exposed to the highly energetic ion beam. The pattern from the AAO membranes were transferred to monocrystalline TiO2 and amorphous SiO2 substrates with good results. By using this technique it was possible to transfer the pattern of the AAO membranes to other materials on large areas (several mm2 , depending on the optics). With the techniques described in this thesis individually or combining them with each other or other known techniques, it is possible to fabricate new nanomaterials. These new materials can be of use in a wide variety of application, such as in sensors and in the field of photo-cleavage of water. By depositing different thin films along the pore walls of AAO membranes interesting properties can be expected, which makes use of the large micro scopic surface and the possibility to precisely tailor the properties of the thin films deposited. A first step is to deposit bi- and multilayered structures, and thereby produce multilayered nanotubes. These nanotubes can make use of properties from all the materials deposited and will surely have interesting properties and future applications.

Atomic layer deposition (ALD) is a powerful ultra-thin film deposition method that uses sequential self-terminating surface reaction steps for preparing a variety of materials. Typically, the process permits nano-scale control of materials in the vertical direction. To develop the method for three-dimensional control of materials, we have been investigating an area-selective ALD technique which will enable and ultimately nano-scale definition of the lateral structure. Many ALD processes are very sensitive to the conditions of the substrate surface. As a consequence, the surface functional groups can be manipulated prior to ALD to carry out an area selective ALD process. Our approach is to chemically modify the substrate surface in order to impart spatial selectivity to ALD. The materials we focus on are both dielectrics (e.g. Hf02 and Zr02 high-lc materials) and metals (e.g. Pt). We have investigated several different types of self-assembled monolayers (SAMs) as resists against ALD. Oxide-coated substrates (e.g. Si02) have been protected using organosilane-based SAMs by silylation reaction; hydrogen-terminated Si (Si-H) and hydrogen-terminated Ge (Ge-H) protected by reaction with Ialkenes or 1-alltynes via hydrosilylation and hydrogermylation, respectively. We have followed the SAM properties as a function of molecular structure and formation time using several experimental techniques and have correlated the properties of the SAMs with their efficacy as ALD resists for both classes of monolayers. With the successful ALD resists, area-selective ALD has been carried out using different patterning methods to define the lateral structure. Both micro-contact printing of the SAMs and selective functionalization of a Si02/Si structure by SAMs have been used to achieve area-selective ALD of Hf02 and Pt films. We have compared the selectivity between these methods, and have described the differences in the context of the SAM resist requirements. We have also shown that by choosing either silylation- or hydrosilylation-based chemical fimclionalization, a single patterned oxide substrate can be used for either positive or negative pattern transfer into the ALD film. The ability to achieve area selective ALD for both dielectrics and metals enables potential gate stack fabrication. Simple capacitor devices were fabricated using area selective ALD and their electrical performance was measured. The comparison of Hf02 electrical characteristics by different surface treatments prior to ALL) suggested further surface and interface modifications are important to achieve better electrical performance.

Atomic Layer Deposition (ALD) is a chemical, gas-phase thin film deposition method. It is known for its ability for accurate and precise thickness control, and uniform and conformal film growth. One area where ALD has not yet excelled is film deposition at low temperatures. Also deposition of metals, besides the noble metals, has proven to be quite challenging. To alleviate these limitations, more aggressive reactants are required. One such group of reactants are radicals, which may be formed by dissociating gases. Dissociation is most conveniently done with a plasma source. For example, dissociating molecular oxygen or hydrogen, oxygen or hydrogen radicals are generated. The use of radicals in ALD may surmount some of the above limitations: oxide film deposition at low temperatures may become feasible if oxygen radicals are used as they are highly reactive. Also, as hydrogen radicals are very effective reducing agents, they may be used to deposit metals. In this work, a plasma source was incorporated in an existing ALD reactor for radical generation, and the reactor was used to study five different Radical Enhanced ALD processes. The modifications to the existing reactor and the different possibilities during the modification process are discussed. The studied materials include two metals, copper and silver, and three oxides, aluminium oxide, titanium dioxide and tantalum oxide. The materials were characterized and their properties were compared to other variations of the same process, utilizing the same metal precursor, to understand what kind of effect the non metal precursor has on the film properties and growth characteristics. Both metals were deposited successfully, and silver for the first time by ALD. The films had low resistivity and grew conformally in the ALD mode, demonstrating that the REALD of metals is true ALD. The oxide films had exceptionally high growth rates, and aluminium oxide grew at room temperature with low cycle times and resulted in good quality films. Both aluminium oxide and titanium dioxide were deposited on natural fibres without damaging the fibre. Tantalum oxide was also deposited successfully, with good electrical properties, but at slightly higher temperature than the other two oxides, due to the evaporation temperature required by the metal precursor. Overall, the ability of REALD to deposit metallic and oxide films with high quality at low temperatures was demonstrated.

Photonic crystals (PCs) manipulate the flow of light via Bragg reflections. Potential applications for such crystals include low threshold lasers, low loss waveguides, highly efficient light bulb filaments, and many other exciting products. While predictions for PC applications grow very rapidly, the realization of these structures has been relatively slow. This thesis explores ways in which atomic layer deposition (ALD) may be used to create, or modify PCs. One-dimensional PCs can be constructed entirely from ALD films. Thin alternating layers of tungsten (W) and alumina (Al2O3) were deposited in a viscous flow reactor and studied using X-ray reflectivity, X-ray diffraction, quartz crystal microbalance, secondary ion mass spectrometry, and transmission electron microscopy. The optimization of thin film growth and nucleation presented in this thesis led to thin film stacks that displayed ultrahigh reflectivity in the hard X-ray regime, and very low thermal conductivity. Three-dimensional PCs were modified with ALD. The first in depth investigation of intensity and position of a Bragg reflection as a function of high index fill fraction will be shown. This study investigated Al2O3 ALD growth rates inside of PCs, and allows for predictions of how all other ALD systems are expected to behave in a similar system. The extent of red-shift for the Bragg peak of this system also revealed the degree of disorder present in the PC prior to deposition. Three-dimensional PCs were also coated with W metal. This system created a photonic band gap (PBG), which is a section photon energy that cannot propagate through the PC. The location of the PBG was adjusted by varying the lattice constant of the PC. This system was tuned to interact with infrared (IR) and ultraviolet (UV) light. This was the first demonstration of a 3-dimensional metal PBG in the UV region. Modified 3-dimensional PCs were studied with scanning electron microscopy to examine the structure of the crystals. UV-visible-IR spectroscopy was used to track optical response changes of the crystal as it was coated with ALD. The reflectance spectra were compared with transfer matrix method numerical simulations to understand the rate and uniformity of ALD on the crystal.

Since the introduction of the MOSFET transistor (metal-oxide-silicon field ef fect transistor) in 1960, the semiconductor technology underwent rapid devel opment. This advancement consisted mainly of the capability to make tran sistors with ever decreasing dimensions and resulted in integrated circuits like the current Intel pentium 4 processor with 178 million transistors. The down scaling continuously improved not only the complexity and speed, but also for example the energy consumption and cost price, to become attractive to end users. While the current characteristic dimension of a MOSFET transistors is about 100 nm, this was a factor 100 larger about 30 years ago. By now, a number of physical limits have been reached and further development of the transistor is not possible without drastic changes to the production process. The goal of this project was to contribute to the solution of an important barrier in the progress of semiconductor technology. Downscaling has resulted in a SiO2 gate dielectric layer of only a few atomic layers thick. Electrons are able to tunnel through this layer at relatively low operating voltage, resulting in an unacceptable high energy consumption of chips. A solution can be found in the dielectric constant of the material: a higher dielectric constant allows 

The reduction of the SOFC operating temperature causes several problems such as ohmic drop through the electrolyte and electrode overpotentials. To overcome these cell performance losses, the reduction of the electrolyte resistivity is necessary. This can be achieved by lowering the thickness of the electrolyte. Therefore, thin layers of CGO and overlayers of YSZ have been deposited separately or together in a bilayer system by Atomic Layer Deposition (ALD) and DC reactive magnetron sputtering on porous and dense substrates. The Chemical Bath Deposition (CBD) of CGO has also been studied as a cheaper deposition technique. Several techniques (SEM, X-Ray diffraction, EDX) have been used to characterize the physico-chemical properties of the thin layers. Their electrical properties have been studied by impedance spectroscopy and the difficulty of such measurements on ultrathin layers has been underlined. These physico-chemical and electrical characterizations allowed us to highlight and compare the leading characteristics of these elaboration techniques and to show the influence of the choice of the deposition technique on the electrical behaviour of the thin layers. Finally, a preliminary study on the ALD elaboration and the physico-chemical and  

Thin-film solar cells based on Cu(In, Ga)Se, contain a thin buffer layer of Cds in their standard configuration. In order to avoid cadmium in the device for environmental reasons, Cd-free alternatives are investigated. In this thesis, ZnO-based films, containing Mg or S, grown by atomic layer deposition (ALD), are shown to be viable alternatives to CdS.

The CdS is an n-type semiconductor, which together with the n-type ZnO top-contact layers form the pn-junction with the p-type Cu(In, Ga)Se,. From device modeling it is known that a buffer layer conduction band (CB) position of 0-0.4 eV above that of the Cu(In, Ga)Se, layer is consistent with high photovoltaic performance. For the Cu(In, Ga)Se/ZnO interface this position is measured by photoelectron spectroscopy and optical methods to -0.2 eV, resulting in increased interface recombination. By including sulfur into ZnO, a favorable CB position to Cu(In, Ga)Se, can be obtained for appropriate sulfur contents, and device efficiencies of up to 16.4% are demonstrated in this work. From theoretical calculations and photoelectron spectroscopy measurements, the shift in the valence and conduction bands of Zn(O,S) are shown to be non-linear with respect to the sulfur content, resulting in a large band gap bowing.

ALD is a suitable technique for buffer layer deposition since conformal coverage can be obtained even for very thin films and at low deposition temperatures. However, deposition of Zn(O,S) is shown to deviate from an ideal ALD process with much larger sulfur content in the films than expected from the precursor pulsing ratios and with a clear increase of sulfur towards the Cu(In, Ga)Se, layer.

For (Zn, Mg), single-phase ZnO-type films are obtained for Mg/(Zn+Mg) < 0.2. In this region, the band gap increases almost linearly with the Mg content resulting in an improved CB alignment at the heterojunction interface with Cu(In, Ga)Se, and high device efficiencies of up to 14.1%.

Keywords: solar cells, Cu(In Ga)Se2, atomic layer deposition, ZnO, Zn(O S), (Zn Mg), band alignment, photoelectron spectroscopy

This thesis describes the processing of thin films of lanthanide (Ln) oxides by atomic  layer deposition (ALD) technique. Deposition of all binary lanthanide oxides was  studied, excluding terbium oxide and the unstable promethium oxide. In addition,  gadolinium oxide-doped cerium dioxide films were grown by combining the  respective binary processes developed in this work. Films were characterized by a  wide range of analytical techniques for structural, compositional, electrical, and  surface properties. As background for the study, some promising application areas  for lanthanide and rare earth (RE) oxide thin films are briefly introduced, and the  ALD technique is explained. Reported ALD processes for RE oxides are then  reviewed.  Ln(thd)3 and ozone were successfully utilized for deposition of most members of the  Ln2O3 series. The deposited films were nearly stoichiometric Ln2O3 with only low  concentrations of carbon, hydrogen, and fluorine impurities. Films were also uniform  and smooth. Relative permittivity values were in the range of 8.4−11.1.  In addition to Er(thd)3, Er2O3 films were also grown with (CpMe)3Er and  Er(tBu2amd)3 as metal precursors. All processes resulted in pure and nearly  stoichiometric Er2O3 films. The growth rate of 1.5 Å/cycle obtained with the  (CpMe)3Er/H2O process was approximately four and six times the rates measured for  erbia films grown by the Er(tBu2amd)3/O3 and Er(thd)3/O3 processes, respectively.  Cerium dioxide films were successfully deposited with use of Ce(thd)4 or  Ce(thd)3(phen) and ozone as precursors. Gadolinium oxide-doped CeO2 (CGO) films  were then grown by combining the Ln(thd)x/O3 processes for the respective binary  oxides. ALD-grown CGO films were dense and conformal, but the Ce:Gd ratio in the  films could not be optimized to the level required in solid oxide fuel cells.  

The atomic layer deposition (ALD) method was applied for fabricating high permittivity  (high-k) dielectrics, viz. HfO2, ZrO2 and rare earth oxides, which can be used to replace  SiO2 as gate and capacitor dielectric. The dielectrics were processed by ALD using novel  cyclopentadienyl (Cp, -C5H5) precursors together with water or ozone as the oxygen  source. ALD, which has been identified as an important thin film growth technique for  microelectronics manufacturing, relies on sequential and saturating surface reactions of  alternately applied precursors, separated by inert gas purging. The surface-controlled  nature of ALD enables the growth of thin films of high conformality and uniformity with  an accurate thickness control.  The ALD technique is introduced and ALD processes for HfO2, ZrO2 and rare earth oxide  films, as well as the applications of the high-k dielectrics in microelectronics are reviewed.  The need for developing new ALD processes for the high-k materials is emphasized.  ALD processes for HfO2 and ZrO2 were developed using Cp-type precursors. The effect of  different oxygen sources, namely water or ozone, on the film growth characteristics and  properties of the ALD-processed films was examined in detail. The oxide films were  stoichiometric, with impurity levels below even 0.1 at-% for C or H. Electrical  measurements showed promising dielectric properties such as high permittivity values and  low leakage current densities. Other properties, such as structure, interfacial layer thickness  and morphology, were also characterized. Compared to films processed by water, the  ozone-processed films on H-terminated Si showed improved dielectric properties, as well  as higher density, lower roughness and better initial growth rate. In addition, in situ gasphase  measurements by quadrupole mass spectrometry (QMS) were performed in order to  study the ZrO2 growth mechanism.  A number of Cp-precursors were tested for the ALD of several rare earth oxide films. The  thermal stability of many of the precursors was limited, but nevertheless, ALD-type  processes were developed for Y2O3 and Er2O3 films. High reactivity of the Cp-precursors  towards water resulting in high growth rates (1.2-1.7 Å/cycle) and purity of the Y2O3 and  Er2O3 films were realized. Despite the detected partial decomposition of the (CpMe)3Gd  precursor, Gd2O3 films with high growth rate and purity as well as effective permittivity of  about 14 were deposited.  Finally, promising processes for ternary scandates, namely YScO3, GdScO3, and ErScO3,  were developed using either Cp- or β-diketonate-based processes. These as-deposited  ternary films were amorphous exhibiting high effective permittivity (14-15), low leakage  current density, and resistance towards crystallization upon annealing even up to 800°C.  

The deposition of the dielectric layer to replace SiO2 remains a major challenge for the electronics industry in the coming years. After the invention of the technique of atomic layer deposition ALD, in parallel with its continuous improvement, many experimental studies have been carried out in this direction. In addition, in the last decade, numerical simulations based on quantum calculation, ab initio or semi-empirical, have begun to provide the additional information necessary for the development of this new technology. Yet, in the current stage of know-how, the deep understanding at the atomic level of the reaction mechanisms that will generate, layer by layer, new gate oxide will be essential for the pursuit and completion of this challenge. In this manuscript we have, firstly, reviewed the main problems and challenges related to the miniaturization of integrated circuits. We discussed the alternative of replacing SiO2, as gate oxide in CMOS or DRAM applications, by oxides with higher permittivities, as well as the main selection criteria for future dielectrics. Al2O3, HfO2 and ZrO2 have demonstrated remarkable thermodynamic compatibility with the Si substrate and are considered today among the most interesting candidates to replace SiO2. The ALD technique, to which we have devoted a section detailing the principles, characteristics and advantages over conventional techniques, was often used to test the three oxides.
Al2O3 remains the reference oxide for ALD, as it lends itself well to the basic models of this technique. Yet Al2O3 remains an alternative in the medium term given its slightly higher permittivity compared to SiO2. The other two, more interesting from this point of view, are unfortunately more difficult to handle at the technological level for the conformity of the deposit, in particular in the initial phase. The main difficulty remains the formation of oxide nanocrystals from the initial phase of the deposition, which induces leakage currents through the oxide along the grain boundaries between the nanocrystals. This rather complex problem has been little studied and the proposed mechanisms have not been studied explicitly by ab initio calculation. However, the reactions of ALD precursors with water constitute a very good start for the understanding of the formation of these oxides. In this thesis we approached this complex subject in a less conventional way, by trying to answer several problems and specific growth conditions (temperature in the reaction chamber, concentration of vapors of gaseous precursors, temperature and preparation of the surface of the substrate, gas phase reactions and surface reactions).
In the methodological part, we detailed the principles of the ab initio and DFT methods. In this context, we have focused our discussion on the treatment of electronic correlation and its correct application according to the systems to be studied. We have also devoted a section to methods and numerical algorithms for the optimization of molecular structures, where we have inserted some tips accumulated with experience, in particular for the search for transition states in chemical reactions. Our own work began with an extensive methodological study, using as molecular systems the molecules used as precursors for the formation of the three candidate oxides: Al2O3, ZrCl2 and HfO2. We tested the accuracy of the B3LYP hybrid density functional, by comparison, both with more accurate post Hartree-Fock type methods (CCSD(T) or QCISD(T)) and with different function bases, and with existing experimental data. The calculations were carried out on the static properties, in the fundamental states, as well as on the dynamic properties, of the vibrational spectra of these states. This led us to validate the very reliable B3LYP/TZVP calculation model for the prediction of all the properties studied.
In particular, we are explicitly interested in potential surfaces in the space of strongly anharmonic internal motions of the methyl groups of the TMA, which constitutes a first for the systems in question. Also, we have proposed two methods to calculate the scale factors of the normal modes for different thermodynamic functions and for different temperature regimes. Our results, and in particular the observations that we have made concerning the necessity of the treatment of strongly anharmonic motions for TMA, will serve as very precise and complete data for the study of the kinetics or/and the precise prediction of the thermodynamic properties of the systems. Then we presented our study of the reactions of H2O with the three molecular systems in the gas phase. Three-step hydrolysis mechanisms for TMA, and four-step hydrolysis for ZrCl4 and HfCl4 have been proposed and the results discussed in their most detailed aspects. These mechanisms correspond to conditions of very low water concentration, i.e. residual water in the containment. In particular, we noticed and discussed the strongly anharmonic movements in the formed complexes. We concluded that the reactions of TMA with H2O were exothermic, whereas the corresponding reactions for the two tetrachlorides, ZrCl4 and HfCl4, were found to be endothermic. We also discussed the possible consequences of the exothermicity of TMA reactions with H2O on the surface reactivity. This opens perspectives to study the possible modulation of the growth rate of the film in its initial phase. Subsequently, we presented the study of similar mechanisms, this time between the precursors formed at the surface (SiO2/Si(001)-2x1) in the initial phase of deposition by chemisorption/recombination. First, we made some important remarks from a methodological point of view on the choice and construction of surface aggregate models.
Concerning the results part, we focused our discussion on the main differences and similarities with the results of hydrolysis in the gas phase. We have again concluded to the strong reactivity of H2O with hydroxymethyl aluminum complexes. With regard to Zirconium and Hafnium hydroxychloride complexes, apart from the fact that their surface chemistry is very similar, the endothermic nature of the reactions studied was also confirmed. We have discussed these results with respect to very specific deposition conditions, i.e. low concentrations of OH active sites, and low concentration of water vapour. In the last part of our thesis, we presented and discussed our preliminary results on more complex mechanisms of the reaction of water with surface ALD complexes. These mechanisms, as well as the resulting complexes, are more appropriate for the usual deposition conditions, where the concentrations of the OH sites are high - which induces considerable interactions between the neighboring surface complexes - or high vapor pressure pulses of water. We have modeled such conditions by explicitly treating the effects of cooperative interactions of water molecules in the vicinity of the complexes.
We have presented here the cases of Aluminum and Hafnium. The results revealed mechanisms not yet considered, to our knowledge, and demonstrated the need to question classical models of reactions in ALD, at least for HfO2 (and implicitly for ZrO2, given their very similar chemistry). Even if our study is in an initial phase, and there are certainly other aspects to be discovered, we believe that we have captured one of the essential aspects of reactions with water, and in particular of the hydrolysis cycle. Finally, we proposed for study a reaction mechanism responsible for Cl contamination in ZrO2 and HfO2 films. Our results constitute a very important step in the understanding of ALD growth of these films. However, the conjunction of non-equilibrium thermodynamic constraints, of the non-linear structure of the evolution equations of the macroscopic physicochemical processes and of the nature of the initial state (very flexible), makes the complexity of the evolutionary dynamics of the systems envisaged inconceivably treatable by modeling studies only at the atomic scale. Thus, the modeling of the deposition by ALD of the three oxides must be part of a multi-scale strategy, capable of structuring and coherently interconnecting the results of numerical simulations at different scales. In this context, firstly, the ab initio results obtained in this work will be used to parameterize a Kinetic Monte Carlo simulation software, a first version of which, incorporating mechanisms appropriate to ALD boundary conditions, is already available at LAAS-CNRS. Work is currently underway to improve this software.
 

High dielectric constant (high-k) insulators and metal gate electrodes are important for advanced MOS devices to limit gate leakage by increasing gate capacitance with ultimately thicker films and eliminate poly-depletion & dopant diffusion, respectively. Reactions between dielectric/substrate and gate electrode/dielectric during deposition or postdeposition processing lead to an increase in interfacial layer formation, and the mechanisms that control the changes need to be well understood. We investigate yttrium-based and hafnium-based high-k dielectrics and ruthenium-based gate electrodes formed by various processing methods such as physical vapor deposition (PVD), chemical vapor deposition (CVD), and atomic layer deposition (ALD) on Si(100). Characterization techniques include IR, XPS, TEM, EELS, AES, and IV and CV electrical analysis. During deposition and postdeposition treatments the interfaces have some extent of interfacial layer formation. The extent of the intermixing depends on substrate surface preparation, process conditions, and annealing conditions. The transition metal alluminate dielectrics show evidence on flatband voltage tuning via charge compensation. Also, the ruthenium gate electrodes show that process condition can have a direct effect the electronic and chemical properties of MOS structures such as in-situ versus ex-situ capacitor fabrication and the role of subsurface adsorbed oxygen in ruthenium.

Solid-state sensors are one of the most effective tools for detecting toxic and combustible gases, and semiconducting tin oxide is the most widely used material. However, present understanding of the mechanism of gas detection is still immature, and disadvantages such as lack of reproducibility and insufficient sensitivity are frequently observed. This research is aimed at understanding the sensing mechanism of metal oxide semiconductor based gas sensors fabricated by atomic layer deposition (ALD) techniques, and exploring the ways to optimize their sensing performance. The ALD of tin oxide thin films has been examined using in situ quartz crystal microbalance (QCM) and Fourier transform infrared (FTIR) techniques. The SnOx films were deposited using sequential exposures of SnCl4 and H2O2 at temperatures from 150-430ºC. The linear growth of the tin oxide ALD films was observed by both the mass gain during QCM measurements and the background infrared absorbance increase during FTIR investigations. A growth rate of ~0.7 Å/cycle at 325 ºC was achieved assuming a density of 6.9 g cm-3. Additional ex situ surface analysis such as X-ray photoelectron spectroscopy (XPS), Auger electron spectroscopy (AES), X-ray diffraction and Atomic force microscope (AFM) results revealed that the SnOx ALD films deposited on Silicon wafers were sub-stoichiometry tetragonal phase polycrystalline SnO2 with very smooth surface. The gas sensing process of O2 and CO on ultra thin tin oxide ALD films were studied by in situ FTIR and electrical measurements. Under oxygen, the background absorbance decreases which is consistent with resistivity increase, the sensing process is slow and may be companied with diffusion process; under CO, the background absorbance increases which is consistent with resistivity decrease, whereas the process is quick compared with the oxygen process, showing it is not a complete reverse process of oxygen response. It is also shown that oxygen is not necessary for CO sensing process. The temperature shows a complicated effect on the response of ultra thin SnO2 gas sensors to O2 and CO gases. The thickness effect sensitivity was studied using hotplate prototype gas sensor. The sensitive films with various thicknesses were deposited on the hotplate and the gas sensing experiemnts were conducted under various CO and O2 concertrations. The highest sensitivity was obtained when film thickness is around 25 angstrom, which is the Debye length. When the thickness of the sensitive film exceeds the Debye length, the resistance of the underlayer plays a important role in the sensitivity and it is hardly influenced by the sensing process which occurs on the surface; When the thickness of the sensitive film is less than the Debye length, the whole film will be influenced by the surface gas sensing processes. A mathematic model describes the effect of film thickness on sensitivity. The model shows consistance with the experiment results for sensitive films with thickness exceeds the Debye length. 

In the history of Chemistry, materials chemists have developed their ideas mainly by doing experiments in laboratories. The underlying motivation for this laboratory work has generally been ?sure curiosity or the ambition to fmd a solution to a specific problem. Minor changes in the composition or structure of a material can cause major changes in its properties. The development of powerful computers has now opened up the possibility to calculate properties of new materials using quantum mechanical methods. The Chemistry of different boron-related materials has been evaluated in this thesis by Density Functional Theory (DFT). Cubic boron nitride (c-BN) is a most interesting material for the microelectronics and tool industry. During thin film deposition of c-BN, several problems arise which most often result in unwanted BN isomorphs. Chemical processes at the (110) and (111) surface of c-BN have been investigated in order to shed light upon some of these complex processes. Typically adsorption energies and surface reconstruction were found to differ significantly between the two surfaces. Other materials investigated are layered transition-metal diborides (MeB,). Incorporation of transition-metal atoms into elemental boron in its most fundamental structure, it-boron, has also been investigated. The calculations on Mek focused on the stability of the planar compared to the puckered structure of Me1:12. Stability was investigated by calculating Density of States (DOS) and bond populations. Deviations in the cell parameters from their ideal values were also considered. A separate project concerned reactivity of the TiB2(00 I ) surface. Molecular and dissociated adsorption energies and adsorption geometries were calculated for 1-12, 11,0 and O.. It was concluded that the titanium surface was more reactive than the boron surface and that the adsorption energies were comparable to or stronger than other well known surface-active compounds like TiO2.

1. For the first time, a comprehensive experimental and quantum-chemical study of vanadium-oxygen structures chemically bonded to the Si02 surface by a different number of Si-0-V bonds was carried out, which made it possible to describe the spectral, energy, and adsorption characteristics depending on their stoichiometry and local structure.

2. It is shown that for vanadium-containing clusters built on the basis of the minimal model of the silanol group - USi-OH - the spectral, structural and energy characteristics are predicted at a semi-quantitative level.

3. Quantum-chemical analysis showed that the energy effect of the addition of vanadium-oxygen groups depends on their functionality, and in the absence of steric hindrances, the formation of polyfunctional groups is energetically more favorable.

4. The possibility of spectral identification of vanadium-oxygen structures of different functionality has been experimentally shown and quantum-chemically substantiated: in the region of 920-940 cm"1, one should expect the manifestation of Si-OV stretching vibrations of vanadium-oxygen structures that form two and three bonds with eighteen the surface of the substrate, and in the range of 950-970 cm "1 - monodentant groups with a single Si-0-V bond.

5. On the basis of diffuse reflectance electron spectroscopy, adsorption studies and quantum chemical modeling, it has been shown that the interaction of vanadium-oxygen groups with water vapor can lead to competing processes: adsorption interaction and hydrolytic destruction of Si-0-V bonds.

6. Quantum-chemical analysis has shown that the energy effect is successively reduced by the adsorption of water by a hydrogen bond, adsorption by the coordination mechanism, and hydrolytic destruction of Si-0-V bonds, and the latter process may be of an activation nature. At the same time, monodentate vanadium-oxygen groups are characterized by the greatest energy effect during adsorption, but at the same time they are the least resistant to degradation under the action of water.

7. For the first time, a quantum-chemical analysis of possible chemical transformations as a result of two MN cycles has been carried out. In the second MN cycle, it is energetically more favorable to form bidentate vanadium-oxygen groups chemically bonded to structures that form one Si-0-V bond with the silica surface.

Hydroformylation is an important commercial process for the conversion of alkenes, carbon monoxide and hydrogen into aldehydes to be further used in the production of various chemicals. The industrial processes operate in a homogeneous mode. Therefore, the development of a solid catalyst would solve problems related to catalyst separation. The purpose of this work was to study supported cobalt and rhodium catalysts in heterogeneous hydroformylation both in liquid and gas-phase conditions. The effect of different preparation methods, precursors, support modifications and pretreatments on the characteristics of the catalysts was investigated. Atomic layer deposition (ALD) is a promising technique for the preparation of dispersed Co(A)/SiO2 catalysts using a Co(acac)3 precursor. Higher activity in ethene hydroformylation was obtained with Co(A)/SiO2 catalysts compared to impregnated Co(N)/SiO2 catalyst prepared from nitrate precursor. The dispersion, and consequently the activity and oxo-selectivity of the Co(A)/SiO2 catalyst, was further improved by inert handling of the catalyst. Moreover, by varying the metal content of the Co(A)/SiO2 catalysts, a clear correlation between metal dispersion and oxo-selectivity was found. The basic AlN modification of the silica support did not enhance hydroformylation activity due to low dispersion of the Co(A)/n⋅AlN/SiO2 catalysts. For the carbon supported catalysts, the best hydroformylation activity was obtained with coconut-shell based Rh/C(C) catalyst. The presence of dispersed active sites and unreduced rhodium enhanced CO insertion, and also unintentional promotion by potassium was possible. Furthermore, without any pretreatment the catalyst exhibited even better propanal yields than with hydrogen pretreatment, apparently due to the better dispersed active sites. Pretreatment with carbon monoxide partially blocked the catalyst surface with carbonaceous residues, which improved CO insertion selectivity, but suppressed the overall activity. The fibrous polymer-supported Rh-phosphine catalyst, FibrecatTM, prepared using a Rh(acac)(CO)2 precursor, was the most promising rhodium catalyst in ethene hydroformylation: high propanal selectivity (95%) and high activity were obtained under the mild reaction conditions of 100 ºC and 0.5 MPa. The 31P NMR characterisations suggested the formation of both a Rh-monophosphine species, Rh(acac)(CO)(PS-PPh2), and a Rh-bisphosphine species, Rh(CO)2(PS-PPh2)2, on FibrecatTM, which were transformed in contact with CO/H2 to the active Rh-carbonyl hydrides. In the liquid-phase hydroformylation of 1-hexene, the activity of the Rh/C catalysts appeared to correlate with the support: the larger the pores, the better the mass transfer and the higher the activity. In addition, C21 products were only formed on a support with sufficiently large pores – an indication of the heterogeneous functionality of the catalysts. With the carbonyl based cobalt catalysts, problems were encountered with the catalyst preparation and handling procedure due to the air sensitivity of the carbonyl precursors. The Co/SiO2 catalysts were stable in gas-phase hydroformylation at 173 °C and 0.5 MPa, whereas Rh/C catalysts lost 10-30% of the metal deposited, mostly due to the formation of volatile carbonyls. However, at a lower temperature, i.e. 100 °C and 0.5 MPa, no volatile carbonyls were formed on FibrecatTM, as confirmed by quantitative 31P NMR characterisations. In liquid-phase conditions, 20–50% of the metal deposited was dissolved from the cobalt and rhodium catalysts. Therefore, the stability of the catalysts in hydroformylation was related to the ability of the catalytic metal to form volatile or soluble carbonyls and thus, to the reaction conditions used.

Ultrathin polymer films can be fabricated using the gas phase method known as molecular layer deposition (MLD). This process typically uses bifunctional monomers in a sequential, self limiting reaction sequence to grow conformal polymer films with molecular layer control. In this study, terephthaloyl chloride (TC) and phenylenediamine (PD) were used as the bifunctional monomers to deposit poly(p-phenylene terephthalamide) (PPTA) thin films. 3-aminopropyl trimethoxysilane (APMS) or ethanolamine (EA) was used to functionalize the surface to prepare an amine-terminated surface prior to the PPTA MLD. The surface chemistry and growth rate during PPTA MLD at 145°C were studied using in situ transmission Fourier transform infrared (FTIR) spectroscopy experiments on high surface area powders of SiO2 particles. PPTA MLD thin film growth at 145°C was also examined using in situ transmission FTIR experiments on flat KBr substrates with an amine-terminated Al2O3 ALD overlayer. The integrated absorbance of the N-H and amide I stretching vibrations was measured and used to determine the thin film thickness. X-ray reflectivity (XRR) experiments were also employed to measure the film thickness after PPTA MLD at 145°C and 180°C. The experiments revealed that TC and PD reactions displayed self-limiting surface chemistry. The surface species alternated with sequential TC and PD exposures and the PPTA MLD films grew continuously. However, the growth rates per MLD cycle were less than the ideal expectations and varied between 0.4-2.9 Å per TC/PD reaction cycle. The lower growth rates are explained by the growth of a limited number of polymer chains on the substrate. The variability in the growth rate is attributed to the difficulties with the bifunctional monomer precursors. Alternative surface chemistries for polymer MLD are proposed that would avoid the use of bifunctional monomers.

The ability to create novel inorganic-organic-metal ordered structures with molecular level precision opens the possibility of developing multifunctional textiles for a myriad of applications including active filtration, bio-separation of proteins, catalytic mantles, and electronic fabrics as well as novel barrier and anti-counterfeiting materials. Due to the high curvature and heterogeneous nature of textile fibers, existing surface modification technologies are not capable of providing complete coverage of a fiber/fabric surface. The use of self-assembly techniques and self-limiting reactant adsorption processes offer the possibility of achieving fully conformal, uniform functionalization of textile fibers of any continuous shape. Atomic layer deposition and electrostatic self-assembly have been used in the semiconductor industry to produce uniform self-organized molecular assemblies over large areas of uniform and clean surfaces such as silicon wafers. However, the use of these techniques in textiles has been largely unexplored. The goal of this research was to determine the feasibility of using layer-by-layer and atomic layer deposition as methods of textile modification. This research has also investigated the optimum processing conditions that allow the selective and controlled deposition of organic, inorganic, and metallic substances on textile substrates via selfassembled nanolayers and atomic layer deposition techniques.

This thesis reports on the development of silicon-based microphotonic waveguide components, which are targeted in future optical telecommunication networks. The aim of the work was to develop the fabrication of silicon microphotonics using standard clean room processes which enable high volume production. The waveguide processing was done using photolithography and etching. The default waveguide structure was the rib-type, with the waveguide thickness varying from 2 to 10 µm. Most of the work was done with silicon-oninsulator (SOI) wafers, in which the waveguide core was formed of silicon. However, the erbium-doped waveguides were realised using aluminium oxide grown with atomic layer deposition. In the multi-step processing, the basic SOI rib waveguide structure was provided with additional trenches and steps, which offers more flexibility to the realisation of photonic integrated circuits. The experimental results included the low propagation loss of 0.13 and 0.35 dB/cm for SOI waveguides with 9 and 4 µm thicknesses, respectively. The first demonstration of adiabatic couplers in SOI resulted in optical loss of 0.5 dB/coupler and a broad spectral range. An arrayed waveguide grating showed a total loss of 5.5 dB. The work with SOI waveguides resulted also in a significant reduction of bending loss when using multi-step processing. In addition, a SOI waveguide mirror exhibited optical loss below 1 dB/90° and a vertical taper component between 10 and 4 µm thick waveguides had a loss of 0.7 dB. A converter between a rib and a strip SOI waveguides showed a negligible loss of 0.07 dB. In the Er-doped Al2O3 waveguides a strong Erinduced absorption was measured. This indicates potential for amplification applications, once a more uniform Er doping profile is achieved. 

High dielectric constant (high-k) - TiO2 thin films were deposited by Atomic Layer Deposition (ALD) with Ti(Oi-C3H7)4 (TTIP) and O3 as metal and oxygen source. TiO2 is well known as material having several polymorphs : rutile and anatase. Dielectric constant of anatase is 45 along a-axis and 25 along c-axis, respectively. On the other hand, rutile structured TiO2 have large dielectric constants at room temperature (170 along c-axis and 86 along a-axis). Although TiO2 thin films grown by ALD have been extensively studied, most of the TiO2 films grown by ALD have anatase structure, on the other hand, there are few reports on the formation of rutile-structured TiO2 films. However, in this study, the growth of rutile TiO2 thin films on Ru electrodes, having a k-value of 83-100, by using ALD technique at a growth temperature of 250 ℃ is reported. The structural compatibility of the RuO2 that was in-situ formed by the strong oxidation power of O3 with rutile TiO2 may constitute the main reason for obtaining rutile structured TiO2 with the high εr value from the O3-TiO2/Ru samples. Also, TiO2 thin films were grown on O3 pre-treated Ru electrodes by atomic layer deposition method with TTIP and H2O as reactants. TiO2 film grown on O3 pre-treated Ru electrode was crystallized in to rutile structure whereas crystalline structure of TiO2 film on Ru electrode without O3 pre-treatment was anatase and the dielectric constant of that TiO2 film was approximately doubled in the value without O3 pre-treatment.
It verifies that the growth of rutile structured TiO2 films on Ru electrodes is caused by structural compatibility with RuO2 formed in the interface. The crystallization behavior of TiO2 films was also promoted due to the local epitaxy growth.
For the application of the material and ALD process to DRAM devices with three-dimensional structures, contact hole type 3D MIM capacitors composed of Ru/TiO2/Ru layers were fabricated with different hole diameters and distances between the holes. The grown films exhibited a very high bulk k value (~100) due to the formation of randomly oriented rutile phase material. The conformity in film thickness and dielectric properties over the entire structured surface area was confirmed by capacitance variation vs. hole surface area experiments.
The equivalent oxide thickness of TiO2 thin films could be reduced up to 8 Å and the leakage current density of the films was less than 10-7 A/cm2 at the applied voltage of 0.8 V. Large leakage current of TiO2 thin films may be induced by low schottky barrier height of TiO2. In order to enhance the schottky barrier height, variation of the properties of the TiO2 films according to O3 concentration and feeding time were investigated. However, n-type of nature of TiO2 films did not change by the variation of the O3 concentration and feeding time even though the atomic O/Ti ratio of the films analyzed by XPS was about 2.2. Bulk properties such as density, crystalline structure, and composition were not also varied from the variation of the O3 concentration and feeding time. However, the higher O3 concentration and the longer O3 feeding time, leakage current of the films increased by increase of the roughness, which was induced by oxidation of Ru due to strong oxidation potential of O3.
Doping of Al ions in the TiO2 films were performed in order to improve the electrical properties eventually. Al and Ti precursor feeding ratio could control Al concentration in the films and Al ions were uniformly distributed in the films. The growth rate of Al-doped TiO2 films decreased with increasing Al precursor feeding ratio. It indicated that chemisorption site density of Ti precursor onto the Al-O layer might be lower than that of Ti precursor onto the Ti-O layer.
Although the dielectric constant of the Al-doped TiO2 films decreased with increasing the Al precursor feeding ratio, the dielectric constant of films which is in the range from 47 to 81, is higher than that of un-doped anatase structured TiO2, It means that the crystalline structure of Al-doped TiO2 films is not anatase but rutile structure after even Al doping. The rather higher leakage current of TiO2 films, which were induced by low Schottky barrier height, remarkably decreased by increasing the Schottky barrier height with the Al doping. In addition, leakage current density of the films was further improved by smoothing the film surface using a longer O3 purge time. Consequently, a minimum toxeq. of 4.8 Å with leakage current density < 1 x 10-7A/cm2 at applied voltage of 0.8 V. were achieved.
However, it is important to control proper Al concentration in the films through the variation of Al/[Al+Ti] precursor feeding ratio, because excess Al ions in the film generate defects like the dipole complexes.


 

The present study was carried out in order to determine the conduction mecha-nisms in thin atomic layer deposited TiO2-containing films. For this purpose, MIM and MIS stacks with thin TiO2 and Al203-TiO2 films were prepared by ALD method. The electrical characteristics of resulting stacks were related to phase, chemical and morphological composition which in turn were influenced by the deposition parameters and electrode materials. The effect of post-deposition annealing on the electrical properties was also studied. The analysis of results was carried out on the basis of a number of samples prepared at the same conditions. As a rule, in the case of Mo/Ti02/A1 stacks, space charge limited currents were assigned as dominant conduction mechanism and the electrodes had only slight effect on the currents. The current densities were determined by the TiO2 layer and depended strongly on deposition parameters. At deposition tempera-tures higher than 200°C, the films grew polycrystalline and possessed extremely high leakage currents. At lower deposition temperatures the films grew domi-nantly amorphous and became more insulating. The decrease in temperature increased the amount of defects which further decreased the leakage currents. At the same time, the long term stability also decreased. Besides the growth temperature, the currents depended also on the precursor chemistry. The annealing procedure used in the present study did not improve the insulating properties of TiO2 films in metal-insulator-stack. In the case of some Mo/Ti02/A1 stacks the interface layer between aluminum electrode and TiO2 became insulating. The insulating layer controlled the leakage currents and complicated the analysis of conduction mechanism. Similar insulating layer appeared also in the case of Si substrate. Quality of the Si substrate, i.e. the presence of SiO2 interfacial layer between TiO2 and Si substrate had a significant effect on the conductivity of the TiO2-based capacitor stacks. Numerical calculations were used to account the effect of interface layers on the conduction mechanism. The calculations indicated that at low voltages, trap assisted tunneling through TiO2 layer was prevailing whereas at higher voltages the tunneling through thin interface layers dominated. The comparison of calculations and experimental results allowed one to evaluate the affmity of our TiO2 films which was between 3.8-4.0 eV. Use of electrodes with higher work-function decreased the current through metal-insulator-metal stacks. In the case of Au electrode, Fowler-Nordheim emission through Au/TiO2 barrier controlled the currents. In the case of Pt electrode, tunneling trough traps apparently reduced the effect of higher barrier at Pt/TiO2 interface and the conductivity of stacks was controlled by Poole-Frenkel effect. The latter result highlights the influence of electrode preparation teclmology on the dominant conduction mechanism. 
The use of Al203 together with TiO2 in mixture and nanolaminate con-figurations resulted in more insulating films. Both experimental results and numerical calculations demonstrated that the insulating properties of Al203-TiO2 mixture and laminate films will still remain inferior to pure Al203 in MIS structures. Nevertheless, the use of electrodes with higher work function electrodes would make TiO2 containing films more attractive due to higher permittivity value. 
 

Challenges are encountered during atomic layer deposition (ALD) on large quantities of nanoparticles. The particles must be agitated or vigorously mixed to perform the ALD surface reactions in reasonable times and to prevent the particles from being agglomerated by the ALD film. The high surface area of nanoparticles also demands efficient reactant usage because large quantities of reactant are required for the surface reactions to reach completion. To address these challenges, a novel rotary reactor was developed to achieve constant particle agitation during static ALD reactant exposures. In the design of this new reactor, a cylindrical drum with porous metal walls was positioned inside a vacuum chamber. The porous cylindrical drum was rotated by a mag- netically coupled rotary feedthrough. By rotating the cylindrical drum to obtain a centrifugal force of less than one gravitational force, the particles were agitated by a continuous “avalanche” of particles. The effectiveness of this rotary reactor was demonstrated by Al2O3 ALD on ZrO2 particles. A number of techniques including transmission electron mi- croscopy, Fourier transform infrared spectroscopy, scanning Auger spectroscopy and x-ray photoelectron spectroscopy confirmed that the Al2O3 ALD film confor- mally coats the ZrO2 particles. Combining static reactant exposures with a very high surface area sample in the rotary reactor also provides unique opportunities for studying the surface chemistry during ALD. Sequential, subsaturating doses can be used to examine the self-limiting behavior of the ALD reactions in the rotary reactor. This dosing method is the first demonstration of self-limiting ALD on bulk quantities of nanoparticles. By combining these sequential, subsaturating doses with quadrupole mass spectrometry, ALD reactions can be analyzed from the gas phase using full mass spectrum analysis. The reaction products are present in a high enough concen- tration to discern a gas phase mechanism for reactions which previously only had surface studies as mechanism determination and characterization. 

The research described in this thesis allowed comparison of several ALD processes for deposition of TiO2, ZrO2 and HfO2. The processes that were studied included those described earlier as well as new ones (TiOC3H7)4-H2O2, ZrC14-H2O-H2O2 and Hf14-02) reported in the original publications of this thesis for the first time. For the TiOC3H7)4-H2O2 and ZrC14-H2O-H2O2 processes and also for several earlier-known ALD processes the real-time characterization of the reaction mechanisms was performed for the first time. The results of the work and comparison of those with literature data demonstrated that chlorides were the most stable metal precursors from those studied so far. They could be used in wide ranges of deposition temperatures and showed high reactivity in surface reactions.

The dependence of the growth rate on the substrate temperature was demonstrated to be in agreement with the changes in the growth mechanism, when crystallization did not influence the surface roughness and adsorption of precursors. In the opposite cases, the effect of crystallization on the growth rate might even exceed the changes related to variations in the mechanisms of exchange reactions. Most significantly the crystal growth influenced ALD of Tio, films. Tio, of the anatase phase grew markedly faster than the amorphous TiO2 phase did. This led to significant surface roughening of the TiO2 films, which contained mixtures of amorphous and anatase phases. Faster growth of anatase and the increase of the surface area with surface roughening resulted in a dramatic growth-rate increase with the transition from the amorphous to anatase phase. This kind of crystallization-related increase of the growth rate was observed with the increase of the growth temperature as well as film thickness. Thus, a common assumption that the film thickness is proportional to the number of ALD cycles employed does not apply in the cases, when the degree of crystallinity increases and the texture becomes more developed with increasing film thickness. The development of crystallinity and texture with increasing film thickness is, however, a very usual phenomenon in ALD of polycrystalline thin films.

The growth rate of ZrO2 and HfO2 thin films did not depend on crystallization as strongly as the growth rate of TiO2 did. Nevertheless, evidence of this kind of effect was found in original studies of this thesis as well as in publications of other authors. The studies performed also demonstrated that the choice of metal precursors and deposition temperature significantly influenced ALD of HfO2 in the initial stage of the deposition on silicon substrates. A marked delay and the three-dimensional nature of the film growth was observed in the high-temperature chloride processes. Much more uniform growth without measurable delay was obtained for the iodide-based and low-temperature (300°C and lower) chloride-based ALD processes.

Crystallization of ZrO2 and HfO2 films in the ALD processes caused an expected increase of the optical density and dielectric constant. In the case of TiO2, by contrast, inhomogeneous crystallization of non-epitaxial films sometimes resulted even in the decrease of the mean density compared with that of amorphous films. The optical density of epitaxial films was, however, always higher than that of the amorphous phase and reached the values of respective single crystals.

The optical band gap of amorphous TiO2 was found to be wider than the band gaps of crystalline TiO2 phases. On the contrary, the band gaps of amorphous phases of ZrO2 and HfO2 films were comparable to or even narrower than the band gaps determined for crystalline phases of corresponding oxides. Optical studies also revealed that the absorption spectra of monoclinic ZrO2 and HfO2 phases markedly differed from the absorption spectra of amorphous, cubic and tetragonal phases. Although similar differences between the absorption spectra of amorphous and crystalline (monoclinic) HfO2 films have recently been observed in works of other authors, too, and explanations to these differences given, verification of the main reasons for this effect needs further studies.

Atomic layer deposition (ALD) is a method for thin film deposition which has been  extensively studied for binary oxide thin film growth. Studies on multicomponent oxide  growth by ALD remain relatively few owing to the increased number of factors that  come into play when more than one metal is employed. More metal precursors are  required, and the surface may change significantly during successive stages of the  growth. Multicomponent oxide thin films can be prepared in a well-controlled way as  long as the same principle that makes binary oxide ALD work so well is followed for  each constituent element: in short, the film growth has to be self-limiting.  ALD of various multicomponent oxides was studied. SrTiO3, BaTiO3, Ba(1-  x)SrxTiO3 (BST), SrTa2O6, Bi4Ti3O12, BiTaO4 and SrBi2Ta2O9 (SBT) thin films were  prepared, many of them for the first time by ALD. Chemistries of the binary oxides are  shown to influence the processing of their multicomponent counterparts. The  compatibility of precursor volatilities, thermal stabilities and reactivities is essential for  multicomponent oxide ALD, but it should be noted that the main reactive species, the  growing film itself, must also be compatible with self-limiting growth chemistry. In the  cases of BaO and Bi2O3 the growth of the binary oxide was very difficult, but the  presence of Ti or Ta in the growing film made self-limiting growth possible.  The application of the deposited films as dielectric and ferroelectric materials  was studied. Post-deposition annealing treatments in different atmospheres were used to  achieve the desired crystalline phase or, more generally, to improve electrical  properties. Electrode materials strongly influenced the leakage current densities in the  prepared metal–insulator–metal (MIM) capacitors. Film permittivities above 100 and  leakage current densities below 1􀂘10-7 A/cm2 were achieved with several of the  materials.  

This thesis describes the deposition of binary lead oxide and ternary lead titanate, lead zirconate, bismuth silicate, and bismuth titanate films by atomic layer deposition (ALD) and characterization of structural, compositional and surface properties of the films. The first part of the thesis reviews the principles of the ALD technique and the relevant literature on perovskite oxides and films and the deposition of lead and bismuth films by ALD, and the second part summarizes the experimental work reported in the five appended publications. On the basis of the binary lead oxide depositions, the Ph4Pb/O3 process was chosen for the ternary oxide studies. Careful optimization of the pulsing ratio of the binary oxides allowed processing of stoichiometric perovskite PbTiO3 and PbZrO3 thin films. Crystalline PbTiO3 on Si(100) was detected after annealing at 600 °C. In the case of lead zirconate, the perovskite phase (PbZrO3) was obtained on SrTiO3(100) after annealing at 600 °C. In both cases, a slight excess of lead enhanced the crystallinity. Roughness values were nevertheless higher than values obtained in binary processes. A new bimetallic precursor Bi(CH2SiMe3)3 was introduced for the deposition of bismuth silicate. With ozone as oxidizing agent, ALD-window for Bi-Si-O thin film growth was found at 250-350 °C. The Si to Bi atomic ratio in this region was about 2. Addition of a second bismuth precursor, BiPh3, increased the bismuth content. Combination of the BiPh3/O3 process and the Ti(O-i-Pr)4/H2O process allowed successful deposition of bismuth titanate. Good control of the film stoichiometry was achieved at the deposition temperature of 250 °C. Both as-deposited ternary bismuth oxides were amorphous. After annealing at 600 °C, the a-axis-oriented Bi2SiO5 phase was detected. Higher annealing temperatures were necessary for bismuth titanate. The most textured film of Bi4Ti3O12 was obtained in N2 atmosphere at annealing temperature of 1000 °C. Roughness values of the thin films were reasonable, being in the range of 0.3-1.3 nm. 

Thin solid films are standard components in a wide variety of technologies such as integrated circuits, wear resistant coatings, energy cells, displays, and micro-electromechanical devices. This work explores four principle technology needs in thin film materials research: inorganic gas diffusion barriers on organic polymer encapsulation, thin nucleation layers to seed film deposition, metal conduction layers on dielectric polymer particles incorporated into bulk electromagnetic interference shield structures, and low oxygen content transition metal coatings on cobalt nanoparticles incorporated into bulk hardmetal. The fabrication of these nanometer dimensioned thin film structures requires the atomic level thickness control and conformability of the Atomic Layer Deposition (ALD) method. An Al2O3 ALD inorganic diffusion barrier on organic polymers is fabricated and characterized to elucidate the general mechanism of deposition on porous polymer substrates. The model developed for inorganic thin film growth on organic polymers is based on extensive Quartz Crystal Microbalance (QCM) studies and involves precursor diffusion into the near surface region, nucleation cluster formation, cluster coalescence to form an impermeable barrier, and subsequent thin film growth above the surface. A W ALD conductive layer is fabricated and characterized on polymer films and polymer particles. X-Ray Reflectivity (XRR) and X-Ray Photoelectron Spectroscopy (XPS) studies reveal that W ALD is greatly facilitated by an Al2O3 ALD nucleation layer and resistivity measurements affirm the conductive nature of thin tungsten films on polymers. A W ALD coating is fabricated on hydrogen reduced cobalt nanoparticles and cobalt thin films and a model of the oxygen content of the particles is developed based on geometry, profilometry, oxygen analysis, and XPS studies. This work demonstrates that Al2O3 and W ALD thin films are ideal components in thin film technologies with polymers and nanoparticles. 

The continued miniaturisation of transistor components is confronting scientists with increasingly difficult technological nodes to solve. In particular, the integration of the copper diffusion barrier in interconnects requires a change of process associated with the use of organometallic precursors, which are more reactive at low temperatures (of the order of 250 °C), to solve compliance problems. Thus, the present work aimed to advance the understanding of the growth of tantalum nitride films by ALD using the organometallic precursor PDMAT, Ta[N(CH3)2]5, and ammonia. The ambition to understand the deposition processes is not new. This understanding requires knowledge of the gaseous species that are transported through the lines and also those that reach the substrate. Many studies on halogenated precursors used in ALD and CVD exist. The particularity of the ALD process over the classical CVD process is the non-interaction of the precursors with each other, which aims at "eliminating" the gas phase reactions between precursors and facilitates part of the study. However, the use of organometallic precursors such as PDMAT makes the understanding of the growth mechanisms significantly more complex due to the complex structure of the precursor and the small number of studies and therefore data reported in the literature. We have conducted parallel modelling/development/characterisation actions combining experimental thermodynamics, modelling and experimentation on a prototype ALD reactor. Different models - ab-initio, statistical, dimensional - have been tested and allow in some cases to estimate the missing thermodynamic data for organometallics. In particular, the molecule Ta[N(CH3)2]4 could be studied more seriously in order to deduce an enthalpy of formation independently. The importance of calorimetric measurements of the enthalpy of formation of solids or liquids should also be noted, provided that their chemical "purity" and/or molecular composition can be attested at the same time. Thermodynamics is fully useful in understanding the mechanisms and allows us to know the state of equilibrium. Indeed, the vaporisation of organometallic precursors in bubblers is at thermodynamic equilibrium. However, given the injection times of the reagents used in ALD reactors, intermediate states may exist due to slower gas decomposition or solid precipitation kinetics, not predicted by thermodynamics. In order to study the behaviour of the precursor PDMAT in the gas phase as a function of temperature, we used the SIMAP localized mass spectrometer.

As PDMAT is an extremely reactive molecule in contact with the atmosphere, a new sealed effusive reactor adapted to the mass spectrometer was developed. This reactor allows the study of saturation vapour pressures when mounted with a single effusion cell. It also allows the analysis of the thermal cracking of the vapours of the organometallic precursor studied when it is set up with "tandem" cells, consisting of an evaporation cell and a cracker. The particularity of the effusion/evaporation cell is that it can be loaded with precursor under a glove box (controlled atmosphere), closed and kept tight during the assembly of the reactor in the desired configuration and the evacuation of the reactor. The cell is then opened remotely and allows the spectrometric study. This cell can also be weighed for calibration by mass loss. The operation of the reactor was validated by performing saturation vapour pressure measurements of a well known organometallic, Y(tmhd)3 and spectrometric measurements with mercury in both configurations against physical flow models. The present spectrometric study of the vaporisation and thermal cracking of the precursor showed that: - PDMAT in solid form at room temperature vaporises as 3 gaseous species Ta[N(CH3)2]5, Ta[N(CH3)2]4 and O-Ta[N(CH3)2]4 - the Ta[N(CH3)2]4 molecule remains by far the most stable gaseous species up to 400°C during vaporisation and cracking. - the decomposition products, mainly the amine compound HN(CH3)2, are derived from the cracking of PDMAT to Ta[N(CH3)2]4 or TaN3C6H18. - The deposits observed in the cracker are mixtures of Ta, O, N and C but we cannot certify that they are defined compounds or solid solution. All this allowed us to perform some thermodynamic simulations of the ALD process. It then appeared that the amine HN(CH3)2 in question observed only exists because its decomposition step has slow kinetics (>1s) and that it is therefore a transition species. These simulations are still incomplete because species such as TaN3C6H18 are not taken into account in the calculations due to a lack of thermodynamic data. Prospects for improving the spectrometric reactor would be to integrate an additional gas introduction to study specific and targeted reactions. A cracker operating at thermodynamic equilibrium, e.g. by increasing the residence time, would be valuable for studying reactions at thermodynamic equilibrium. Crackers with intermediate residence times would allow the study of different kinetic regimes and in particular the consequences of the pulse times used in ALD deposition. This thesis also describes in detail the ALD reactor designed by AST, a rather complex prototype reactor using a virtual valve. The geometry of this reactor leads to the elaboration of thin films of heterogeneous thickness due to an inhomogeneous distribution of the reagent inputs on the substrate surface. This geometry needs to be improved in order to obtain perfectly conformal thin films (homogeneous in thickness) typical of ALD deposits. Despite the inhomogeneity of the thickness of the deposits, the latter have allowed us to understand the technical and chemical issues related to the implementation of ALD processes. Leads on process optimisation were obtained thanks to spectrometric studies, deposition with PDMAT alone and PDMAT and NH3 and flow simulations. All this explains the difficulties encountered for the deposition of pure TaN, the incorporation of impurities such as O and C, which are difficult to avoid. The negative effect of this contamination on the diffusion barrier performance of these layers remains to be assessed. More surface-based approaches - study of the reactions between the precursors and the substrate - deserve to be explored in greater detail by means of deposition carried out by ALD or CVD, followed in situ by microbalance and/or by qualitative quadrupole spectrometric monitoring of the reactive species. The results show the interest of using the spectrometric tool to understand the growth of thin films by ALD from organometallic precursors. Thermodynamic simulation, spectrometric studies and the elaboration and characterisation of deposits are perfectly complementary techniques for the understanding and control of these emerging and complex processes

I consider the most important achievements of my dissertation: 1) developing a method of low-temperature growth of ZnO layers with using zinc acetate as a zinc precursor (and also as monoprecursor) 2) demonstrating the correlation between the growth temperature and parameters electricity (although this issue is not discussed at length in of this dissertation, was relevant to the project we were carrying out Of the European Union, code-named VERSATILE) 3) development of a technology for obtaining uniform ZnMnO layers decomposes the manganese admixture 4) showing that the layers thus obtained with low contents manganese impurities do not show ferromagnetic signals. At this point, I would like to point out that the developed methodology of growth (low temperature and proper selection of zinc cycles and magnetic ion) gave at the moment very promising results for ZnCoO layers performed outside the field the current hearing. 

Hf-Si oxynitride alloys were deposited by a remote plasma-enhanced chemical vapor deposition (RPECVD) system using the combination of the process conditions for Si oxynitride and Hf silicate alloys. The N-KLL, O-KLL, and Hf-NVV intensities measured by on-line AES (Auger Electron Spectroscopy) spectra were used to calculate the composition of alloys. The composition of Hf-Si oxynitride alloys can be tuned by controlling the N2 / (N2 + N2O) ratio, Hf source flow rate, and the amount of He dilution. FTIR (Fourier Transform Infrared Spectroscopy) and XPS (X-ray Photoelectron Spectroscopy) measurements were performed off-line and the results were used to investigate changes in film internal structure with (i) composition, and (ii) post-deposition annealing temperature. As deposited alloys show single feature in FTIR spectra, and there is no evidence of metallic Hf-Si, Hf-N, Hf-Hf, or Si-Si bonding in XPS core-level spectra. The (HfO2)X(SiO2)1-X alloys and (Si3N4)x(SiO2)0.5•(1-x)(HfO2)0.5•(1-x) alloys with low Si3N4 concentration (x = 0.07 and 0.17) show changes in FTIR absorption spectra after anneal at 900 ~ 1100 60s ℃ , in Ar. However, (Si3N4)x(SiO2)0.5•(1-x)(HfO2)0.5•(1-x) alloys with high Si3N4 concentration (x = 0.33 and 0.49) show no change in FTIR spectra even after 1100 anneal. ℃ The results of XPS O1s spectra of the pseudo-binary (HfO2)X(SiO2)1-X alloys and the pseudo-ternary (Si3N4)x(SiO2)0.5•(1-x)(HfO2)0.5•(1-x) alloys correspond to the results of FTIR spectra. The chemical phase separation in Hf-Si oxynitride alloys was suppressed when the amount of Si3N4 phase is above 33%. Micro/nanotubes with precisely defined nanoscale walls have attracted considerable attention with a variety of different processes and materials. Since ALD provides excellent step coverage on aggressive topographic structures, ALD has expanded rapidly its application fields. In some cases, ALD can be conducted at the generally lower temperature (~100℃ or less). This makes ALD attractive for coating on temperature-sensitive materials. We explored the low temperature metal ALD process for the fabrication of nanostructures. In this work, Cobalt thin film deposition using atomic layer deposition process sequencing was studied between 30 and 130°C using Co2(CO)8 and H2 gases using on-line quadrupole mass spectrometry and Auger electron spectroscopy. Similar experiments using cobalt cyclopentadienyl dicarbonyl and H2 reactants were also performed between 140 and 350°C. For the dicobalt octacarbonyl precursor, mass spectroscopy and growth rate analysis showed precursor dissociation with non-self-limiting adsorption leading to continuous film growth at temperatures as low as 60°C, whereas the cyclopentadienyl dicarbonyl precursor showed evidence for CO cleavage and volatile Co(cyclopentadienyl) resulting in no film growth until ~300°C. The continuous film growth with the Co2(CO)8 is related to the zero-valent metal center, where no reduction step is required to produce a reactive surface for adsorption. Evidence for Fisher-Tropsch catalytic production of CH4 is observed by mass spectroscopy during the initial cycles of Co film growth.

Metal oxides are a class of materials which plays a major role in many present applications, ranging from optical coatings to microelectronics, photovoltaics and gas/moisture diffusion barrier technology. Thin metal oxide films can be obtained using different deposition techniques, such as physical vapor deposition (i.e., sputtering) and chemical vapor deposition. In the present project, an expanding thermal plasma metal organic chemical vapor deposition (ETP-MOCVD) technique was used for the deposition of zinc oxide (ZnO) and aluminum oxide (Al203) thin films. ZnO polycrystalline films have been intensively studied in the recent years as transparent conductive oxides for applications such as, among others, channel/gate layers in thin film transistors or front electrodes in solar cells, as well as applications which require both it and ptype films, i.e., light emitting diodes. Al203 amorphous dielectric films, on the other hand, have shown great potential in high-k applications and, more recently, in gas/moisture diffusion barrier applications. Understanding the thin film growth and controlling it in terms of structure, morphology and opto-electrical properties is a necessary step in order to extend the application range of the deposited layers. In this work the evolution of the AI-doped ZnO (AZO) film properties during growth was investigated by an extensive set of ex situ and in situ techniques. In particular, the dependence of the intrinsic properties (crystallinity, stoichiometry, doping level, etc.) and extrinsic properties (grain size, morphology) on the film thickness was studied and correlated with the electrical characteristics of the deposited layers. As a result, it was shown that the working pressure plays an important role in controlling the development of the electrical and morphological film properties during growth. At 1.5 mbar ("high pressure") the AZO films are characterized by a low nucleation density, a large sheet resistance gradient with film thickness and high root-mean-square values, i.e., >4% of the film thickness. By decreasing the pressure from 1.5 mbar to 0.38 mbar ("low pressure"), the initial layer becomes denser, the sheet resistance gradient is significantly reduced and the films become smoother, i.e., <1% of the film thickness. The sheet resistance gradient and the surface roughness development correlate with the grain size evolution, indicating the transition from pyramid-like at high pressure to pillar-like growth mode at low pressure. The in situ use of the spectroscopic ellipsometry (SE) technique, a novelty in the ZnO field, allowed the real time identification of the growth mode by monitoring the thickness development in the initial growth stage and the roughness evolution during film growth. A slower growth rate was observed for the pyramid-like films during the initial growth phase compared to the bulk, while the pillar-like films exhibited a linear increase in thickness at all stages of growth. A saturation behavior in the roughness evolution for films thicker than 150-200 nm was observed for the pyramid-like structure, while for pillar-like films the roughness scales linearly with the film thickness. The relation between these differences and the two growth modes was validated by comparison with ex situ measurements, such as time-of-flight secondary ion mass spectrometry TOF-SIMS (initial growth) and atomic force microscopy AFM (roughness). Moreover, the SE measurements proved to be useful in determining the in grain electronic properties of the AZO films, which is essential to define the role of grain boundaries in limiting the electron transport in ZnO films. The results obtained demonstrate excellent in grain mobility values, i.e., above 100 cm2/Vs (pyramid-like growth) and 50 cm2/Vs (pillar-like growth), independent of the film thickness. These values are much higher than the ones provided by the ex situ measurements (Hall), which indicates that the limiting factor for the electron transport in these films is the scattering at grain boundaries. Controlling the film growth mode is very important from an application point of view. The low resistivity and high roughness of the pyramid-like films make them suitable as front electrodes in a-Si:H and in pc-Si solar cells applications. However, the sheet resistance gradient with thickness and the low nucleation density makes them unsuitable vvlieti the AZO films are deposited on the solar cell as a substrate or in applications where thinner and smoother layers are required, such as thin film transistors. In these cases the pillar-like films or a combination of the two modes might be more appropriate. Moreover, the film growth studies, both ex Sal and in situ, presented in this thesis, indicate that a valid route for further improving the conductivity of the AZO films is to increase the grain size at the initial stage of film growth by, for example, increasing the substrate temperature or using a ZnO buffer layer as substrate. Another challenge in the ZnO field is to obtain ptype conductivity, which, in combination with the more easily obtainable n-type ZnO, would allow the fabrication of ZnO homojunctions. In this work initial studies on the plasma chemistry and its influence on the doping efficiency were performed in the case of nitrogen-doped ZnO. Because the expanding thermal plasma has a high dissociation degree for N2, allowing a large flux of N radicals and/or N-containing species towards the substrate, it could be an excellent source for p-type N-doping of ZnO films. Nitrogen incorporation in the ZnO films was successfully obtained using this technique. The nitrogen was found to incorporate preferentially as -CEN (nitrile bond), which is electrically inactive. As a consequence, no p-type conductivity was generated in the N-doped ZnO films. The detection of CN presence in the film using infrared spectroscopy is a novelty in the field and it is found to be corroborated by the formation of HCN in the plasma, suggesting an inherent limitation in any deposition process which combines the use of a metalorganic precursor with a highly reactive nitrogen environment. Using the knowledge acquired in this project, a valid route to overcome this limitation can be proposed, i.e., to combine the advantages of both ETP and sputtering techniques, by using a metal or metal oxide target, sputtered in an expanding thermal Ar/N2 plasma environment. The second part of this thesis work was dedicated to extending the applicability of the ETP-MOCVD technique to obtain dense Al203 films at relatively low substrate temperatures (< 400 0C) compared to other CVD processes. While, initially, the ETP-deposited Al203 film properties were found to be rather poor, i.e., low refractive index (<1.5 at 633 nm) and high hydrogen content (>30 at%), a key parameter to obtain film densification was identified. Through the addition of ion bombardment to the ETP-MOCVD process by means of an external rf bias applied to the substrate, films with high refractive index (1.6 at 633 nm) and low hydrogen content (540/0) can be obtained at temperatures even below 150 ciC These films are potentially suitable as water permeation barrier layers on polymers, as preliminary investigations have already indicated.

The use of in situ spectroscopic ellipsometry (SE) during a variety of atomic layer deposition (ALD) processes was exemplified in this work, clearly demonstrating the merits of the combination of in situ SE and ALD. In situ SE yields accurate information on the film thickness and the growth rate per cycle and it was shown that this information can be used to determine ALD saturation curves, to study initial film growth (nucleation effects), and even to probe half-cycles with a submonolayer sensitivity with respect to the surface chemical species. These aspects are all very relevant for ALD film growth, both in terms of fundamental understanding and in terms of process optimization and control. Additionally, it was shown that in situ SE yields also important information on the material properties during the ALD process when the film is still being deposited. Besides the optical properties such as the refractive index and the optical band gap, also insight into the electrical properties of conductive (metal nitride) films can be obtained non-intrusively. The latter aspect can be used to study the influence of film thickness and ALD growth condition (e.g., the setting of precursor/reactant dosing, deposition temperature, etc) on the electrical properties of the films. Furthermore, the crystalline phase of the thin film materials can be established from the in situ SE measurements and phase transitions can even be probed during the ALD growth process. As is generally valid for ellipsometry measurements, the validity and the accuracy of the information extracted from the measurements depends critically on the interpretation of the data in terms of optical modeling. Corroboration of the results by other (ex situ) techniques is therefore key, especially when addressing absolute values of physical quantities. The situation is mitigated when mainly addressing specific trends in the data such that in situ SE is particularly useful during ALD film growth for process monitoring and control. The value of SE in monitoring the film growth has recently been recognized by ALD tool manufacturers and has led to the development of commercial ALD reactors with integrated in situ spectroscopic ellipsometry capability. It is, therefore, anticipated that in situ SE has a bright application prospect in the field of ALD. Finally we want to note that in situ SE cannot only be used to study, optimize and design ALD processes. The combination of in situ SE and ALD is also very powerful for fundamental studies of physical and chem[1]ical effects related to ultrathin films. The precise growth control and the relatively high level of understanding of the surface chemistry associated with ALD can, e.g., be exploited in studies of the optical properties of surface layers of chemical species or in studies of (quantum) size effects in ultrathin semiconductor and metal films.

The atomic layer deposition (ALD) technique has recently gained considerable interest as a suitable method for the fabrication of nanoscale thin films. The virtue of this technique is that the deposition is controlled at the atomic level by self-limiting surface reactions through the alternate exposure of the substrate surface to different gaseous precursors. ALD provides ultimate control of film thickness and has the potential to achieve uniform film properties over the entire substrate surface, even in high aspect ratio structures. Over the past years it has been proven that with ALD high quality, atomically smooth, and conformal thin films of a wide variety of materials can obtained. Currently, several ALD processes are on the verge of being incorporated into the production of devices, such as CMOS transistors and DRAM memory. A recent development to broaden the applicability of ALD is the use of a plasma as an alternative reactant source. Because the activation of the reactive species already takes place in the gas phase, this so-called plasma-assisted ALD, can provide certain benefits. In this thesis work, the plasma-assisted ALD of the metal oxides (Al2O3, Ta2O5 and HfO2) and metal nitrides (TiN) were investigated. For this purpose, a versatile plasma-assisted ALD reactor based on a remote plasma configuration was constructed. To study the plasma-assisted ALD processes, in situ diagnostics were employed. New in this respect was the use of spectroscopic ellipsometry (SE) to measure film thickness in situ and optical emission spectroscopy (OES) to study the electronically excited reaction products in the plasma. Furthermore, also a quartz crystal microbalance and quadrupole mass spectrometer were employed to monitor the mass uptake per half-cycle and the reaction products created, respectively. The composition, microstructure, and electrical properties of the films were determined by ex situ techniques. During the project, the collaboration with Oxford Instruments contributed to the design of one of the first commercially available R&D tools for plasma-assisted ALD, the FlexAL. Currently, a beta version of the FlexAL tool is installed at our university and the first results on the deposition of TiN and HfO2 on 200 mm wafers were reported in this thesis work. The merits of plasma-assisted ALD have been studied and made apparent for the materials investigated. The improvement of material properties by the plasma-based process was demonstrated for the case of TiN. Good material properties in terms of impurity content and electrical resistivity were obtained for TiN, also at a deposition temperature as low as 100 °C. The feasibility of depositing good quality Al2O3, Ta2O5, and HfO2 films by plasma-assisted ALD at low substrate temperatures was also demonstrated, even down to room temperature for the case of Al2O3. The reaction mechanisms of plasma-assisted ALD have been studied for the deposition of Al2O3 and Ta2O5 from metal-organic precursors in which an O2 plasma is used as oxidant source. Two different types of metal-organic precursors, a metal-alkyl (Al(CH3)3) and metal-alkylamide (Ta[N(CH3)2]5), were used. During the O2 plasma exposure, the presence of CO, CO2 and H2O was detected in both processes. Furthermore, the depletion of the O2 source gas indicated the consumption of O radicals. These observations demonstrated that combustion-like reactions in which the surface groups are converted by O radicals into combustion products occur at the surface. Secondly, in the Al2O3 process the detection of CH4 during the O2 plasma exposure indicated that the production of H2O has an effect on the surface chemistry during the Al2O3 deposition process. The produced H2O apparently forms an alternative thermal ALD-like reaction pathway in concurrence with the combustion-like reactions. The possible existence of more reaction pathways is suggested by the presence of C2Hx and CN species during the O2 plasma exposure in the Al2O3 and Ta2O5 deposition, respectively. Additionally, the dissociation and excitation of molecules are other reactions that can take place in the plasma. This was demonstrated by the light emission coming from the plasma during Al2O3 and Ta2O5 processing, which changes in the presence of reaction products released from the surface.

This work concentrates on two topics: (i) GaAs surface passivation methods using different materials and (ii) formation of InAs islands on GaAs and transformation of InAs islands into quantum rings (QRs). All the samples are fabricated by metalorganic vapor phase epitaxy and characterized by optical spectroscopy and atomic force microscopy. InGaAs/GaAs near-surface quantum well (NSQW) structures were used in the GaAs surface passivation studies because of their sensitivity to surface states. Ultra-thin InP, Gag GaN layers were grown in-situ on top of the NSQW structure as the passivation layer. As-P and As-N exchange on the GaAs surface were also applied for passivation. In all the passivation methods, the photoluminescence (PL) intensities and carrier lifetimes of the NSQWs were significantly increased. The enhancement factor of the PL intensity was up to two orders of magnitude. The study of time durability of the passivation after keeping the samples for months in air ambient showed that those passivation methods protect the samples against oxidation while the unpassivated samples degrade severely. The passivation effects of these materials were also studied using NSQWs fabricated on (110)-oriented GaAs substrates. The suitability of atomic layer deposited (ALD) titanium nitride layer on GaAs surface as an ex-situ passivation layer was also investigated. Although the enhancement factor of the PL intensity is smaller than that obtained by in-situ methods, smooth surface morphology and notable extension of carrier lifetime were observed in the ALD passivated samples. It is known that island fomiation is severely suppressed on the GaAs (110) surface. This limitation can be overcome by using a thin strain reducing layer, e.g., an InGaAs layer. Relatively uniform InAs islands with an average areal density of 109 cm -2 were obtained on GaAs (110) substrate at 400 °C using a thin InGaAs strain reducing layer. Transformation of InAs islands into rings was realized by partially capping the InAs islands and annealing under tertiarybutylarsine flow. Effects of growth conditions on ring evolution were studied by varying the thickness of the partial capping layer, annealing time and annealing temperature. It was concluded that the temperature dependence of the diffusion anisotropy of the indium atoms plays an important role in the ring evolution. The annealing process of the partially capped islands affects significantly the ring shape and the optical properties of the QR structure. 
 

This thesis deals with x-ray reflectivity (XRR) analysis. XRR is a very accurate technique for the metrology of thin films but the analysis of measurements has been difficult thus limiting every day material research. In this thesis, novel genetic algorithms (GAs) for XRR curve fitting and statistical error analysis methods are developed. The XRR analysis utilizes very accurate Parratt's formalism combined with Nevot–Croce interface roughness. The analysis concentrates on the atomic layer deposited materials by using models mimicking their properties. The properties of GAs are studied using aluminium oxide/zinc oxide nanolaminate models. Models of aluminium oxide layers on silicon substrate are used in the case of the error analysis.
The demonstrated novel GAs are utilizing the rotation of coordinates during the crossover phase to reduce interparameter dependencies. The new basis is formed from the eigenvectors of Hessian and statistical covariance matrices. The crossover is performed in the rotated coordinates and the new combinations are transformed back to the original coordinates. It is shown that the coordinate rotation improves the convergence properties of GAs in complex XRR curve fitting problems and a statistical approach is more powerful than the Hessian matrix method. Furthermore, a GA using independent component analysis gives additional robustness to the curve fitting by utilizing a nonorthogonal linear transformation technique.
The interdependency of XRR parameters is studied using fitness landscapes. The fitness landscape analysis utilizes subspace projection of the original parameter space where the projection is done using an experimental model. The work reveals that the error in the determined mass density can compensate the error in surface roughness thus diminishing the accuracy of both of these parameters. This result is also verified later with other methods.
The effect of Poisson noise on the accuracy of XRR analysis is studied statistically. Thickness determination accuracy of an aluminium oxide layer is ±0.09 nm with 99% confidence in the studied case which represents the lower limit for the error. Here the analysis assumed a perfect fit to the measurement. The upper error is achieved by taking into account a nonideal fit by separating the effect of noise from the fitness value. In a case of the studied measurement, the determined thickness error is ±0.12 nm with 99% confidence.

The field of atomic layer deposition (ALD) has grown substantially over the past several decades. The development of ALD as a method to deposit materials has been predominantly led by the microelectronics industry, which continues to move to submicron dimensions. This momentum towards submicron dimensions has pushed conventional thin film deposition techniques to their limits. Atomic layer deposition is a thin-film deposition technique based on sequential, self limiting surface reactions. The reactions are performed in an ABAB… binary reaction sequence to deposit a controlled and conformal film. A growing number of materials can be deposited by ALD including oxides, nitrides, sulfides, and metals. ALD has the ability to control numerous film properties such as thickness, morphology, crystallinity, conformality, and electrical properties. This thesis studied the chemistry of reactions leading to the ALD of MgO, MnO, TaN, and SiO2. Additionally, the novel chemistry for the rapid ALD of SiO2 was investigated. In situ Fourier transform infrared (FTIR) spectroscopy and quartz crystal microbalance (QCM) were conducted to monitor surface species during each half reaction and verify saturation behavior. Once the surface chemistry was understood, the films were then deposited on Si(100) substrates using the optimal deposition conditions. The films grown on Si(100) substrates were used for numerous ex situ thin film analysis techniques. X-ray reflectivity experiments were conducted to yield both film thicknesses and film density. X-ray diffraction experiments were used to determine film crystallinity. Four-point probe measurements were conducted to determine film resistivities. XPS sputter depth profiling, Rutherford backscattering measurements and in some cases SIMS measurements were also conducted to obtain the chemical composition of the films. Transmission electron microscopy and scanning tunneling microscopy were utilized to visualize the conformality of the films.

All of the work carried out in one hundred studies focused on the resolution of the problems related to the ohmic drop within the electrolyte and the reduction of the reaction kinetics at the cathode generated by the lowering of the operating temperature of the batteries. SOFC. Our study focused on the search for new materials and new architectures for SOFC cells operating at intermediate temperatures. Indeed, this field has grown with the development of high-performance thin-film deposition techniques which are increasingly used to solve problems of performance problems related to electrolyte/electrode interfaces, by inserting a thin layer of the electrode or iketrolyte material. test in this context that fits the first part of this work. Cathode/electrolyte half-cells with thin interfacial layers of YSZ, LSM and La1121iO4 have been produced on dense YSZ substrates by various techniques (ALD, PVD and dip-coating). The thick cathode layers, LSM or La2NiO4, were produced by painting or screen printing. This study allowed us to compare these techniques in different configurations. The electrical performance of the half-cells was evaluated by impedance spectroscopy using an unsymmetrical two-electrode configuration. He has shown the benefit brought by a thick porous layer deposited on a thin interfacial layer of 80 nm and the interest of the choice of the synthesis technique used on the quake of the interface. The presence of a thin layer of YSZ deposited by ALD improves the polarization resistance of the interface in comparison with a layer deposited by dip-coating or PVD. However, the presence of a thin layer of the same cathode material provides the best electrochemical performance. This study confirmed that the cathode material, La2NiO4, emits more performance than the classically used LSM perovskite. Indeed, the reduction of oxygen in k case (an essentially electronic conductor (LSM) can take place at the TPB points at the electrolyte/cathode interface; (the inverse of a mixed conductive material such as LaNith oil Its electrons and Its oxygen molecules can react on any k volume of the cathode. This work has shown the complexity of the role of thin interfacial layers of electrolyte or electrode which appears essential for Its SOFCs at intermediate temperature. In a second paper, we were interested in the elaboration by ALD on a porous cathode substrate (LSF) of a material in thin layers of zirconium oxide doped with indium oxide presenting a composition gradient making it possible to gradually pass from an ionic conductivity has an electronic conductivity by increasing the indium content.This material has conduction properties minus the level of dopant incorporated.For this, three thin layers of compositions ranging from 31 4 to 77.3 mol% of InOi.s were successively deposited on an LSF substrate. Each of the compositions was also synthesized separately on different substrates in order to study their structure and/or their electrical behavior. ALD made it possible to deposit thin and even ultra-thin (<100 nm) uniform, adherent, covering layers of large quake microstructural iris and directly crystallized at low temperature (300°C). The study of the electrical properties of these thin layers elaborated by ALD was carried out by impeclance spectroscopy using a point electrode of platinum in transverse configuration. Indeed, the use of a transverse geometry is very important because it allows the study of the electrode/electrolyte interface. Moreover, this measurement configuration corresponds to the actual operation of the SOFC cell. The electrical characterization of these samples showed a different tnIs electrical behavior from that of bulk materials. The two deposits least concentrated in indium oxide (31.4 and 54.7mol%) exhibit an ionic character with an activation energy which increases with the indium content. On the other hand, the resistance (normalized with respect to the thickness of the sample) measured at high temperature for the sample presenting a composition gradient is lower than that of the ionically conductive thin layers which compose it. Indeed, the composition gradient would favor the passage of charge carriers through the intermediate layers constituting it. This shows the interest of a mixed conductor such as Zr02-111203, in particular when it is produced by ALD in the form of thin layers with a composition gradient in order to improve the performance of IT-SOFCs by reducing the ohmic drops and the overvoltages on the cathodic side. The last part of this work has been the study of a new composite electrolyte material, GDC•carbonates. The goal is to characterize this composite material and to highlight its electrochemical properties by impedance spectroscopy. Thermogravimetric (ATG) and differential thermal (ATD) analyzes coupled with mass spectrometry have enabled us to show that the endothermic peaks observed at high temperature are perfectly superimposed on those of the mixture of carbonates; they correspond to the fission points of the eutectic, according to the phase diagrams. X-ray temperature diffraction (DXHT) analysis shows that only peaks corresponding to gadolinium-doped ceria (GDC) are clearly visible. The peaks relating to the carbonates are of low intensity and clearly muddy visible. Scanning electron microscopy analysis showed the presence of two well-separated phases with different grain sizes. A gray phase with an agglomerated distribution is attributed to the mixture of carbonates and a white and finer phase corresponds to the ceria doped with gadolinium (GDC). The study of the electrical properties by impedance spectroscopy in symmetrical and unsymmetrical configurations under different conditions has shown a discontinuity in the ionic conductivity digraphs around the melting temperature of the mixture of carbonates with a rapid increase in the conductivity values. Low values of activation energy are obtained at high temperature. The conductivity in this zone is ensured mainly by the ions of the carbonate phase which are more mobile than the O 2 ions of the GDC phase. The study of cycling and aging of the composite GDC-carbonates (Li/K) shows a high chemical stability of the mixture of carbonates. This study is the first in-depth test ever performed on this composite. The aging test carried out at 600°C in air for 1528 hours shows particularly encouraging results; they tend to show the stability of the composite and a high conductivity value (0.66 S.cni1). This work opens several perspectives. The first part of this work has clearly shown the breadth of study that combines advanced fundamental knowledge with aspects of technological know-how. The final goal is to manufacture a complete cell with anodic support, including an electrolyte of the order of pm &boron by ALD and which will be covered by a thin interfacial layer of cathode by dip-coating or PVD followed by a thick layer of cathode by paint. The study of indie zircon in the form of a thin layer with a composition gradient may lead to a more in-depth study of the associated electrochemical properties in order to confirm the beneficial relationship of such a system for SOFCs. With regard to the GDC-Carbonates composite material, all the results obtained tend to show the interest of this material as an electrolyte for a SOFC fuel cell. Engineering and Technology of the Royal Institute of Technology (KTH, Sweden) is still in progress and further aging tests under different atmospheres are possible.

Rooth, M. 2008. Metal Oxide Thin Films and Nanostructures Made by ALD. Acta Universitatis Upsaliensis. Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology. 56 pp. Uppsala. 
Thin films of cobalt oxide, iron oxide and niobium oxide, and nanostructured thin films of iron oxide, titanium oxide and multilayered iron oxide/thanium oxide have been deposited by Atomic Layer Deposition (ALD). The metal oxides were grown using the precursor combinations Col,/02, Fe(Cp)2/02, Nb110, and Ti14/11,0. The samples were analysed primarily with respect to phase content, morphology and growth characteristics. Thin films deposited on Si (100) were found to be amorphous or polycrystalline. depending on deposition temperature and the oxide deposited; cobalt oxide was also deposited on MgO (100), where it was found to grow epitaxially with orientation (001)1100]Co20,11(001)(100)Mg0. As expected, the polycrystalline films were rougher than the amorphous or the epitaxial films. The deposition processes showed properties characteristic of self-limiting ALD growth; all processes were found to have a deposition temperature independent growth region. The deposited films contained zero or only small amounts of precursor residues. The nanostructured films were grown using anodic aluminium oxide (AAO) or carbon nanosheets as templates. Nanotubes could be manufactured by depositing a thin film which covers the pore walls of the AAO template uniformly; free-standing nanotubes retaining the structure of the template could be fabricated by removing the template. Multilayered nanotubes could be obtained by depositing multiple layers of titanium dioxide and iron oxide in the pores of the AAO template. Carbon nanosheets were used to make titanium dioxide nanosheets with a conducting graphite backbone. The nucleation of the deposited titanium dioxide could be controlled by acid treatment of the carbon nanosheets. 
 

This thesis introduces important improvements in fabrication of microfluidic devices on silicon and glass. With the main aim in surface and volume manipulation of aqueous solutions for subsequent biochemical analysis, the backbone of the work has been the development of plasma etching processes for silicon and glass. As the silicon microfabrication technologies are combined with deep anisotropic etching of glass, the processability of microfluidic applications with surface and volume manipulation of fluid is diversified. Several mask materials have been studied with respect to deep plasma etching of glass. As the demand for depth of microfluidic devices extends past 150 µm, the number of usable masking schemes becomes limited. To reach an etch depth beyond 350 µm with aspect ratio of over 3:1 including the mask, silicon shadow mask was used. The results of process development on Al2O3, AlN and TiO2 masks show that a very high etching selectivity on glass can be achieved with these mask materials. The described masking technologies enable e.g. high density of through-a-wafer holes or nearly vertical structuring of glass with great depth. Also, a silicon shadow mask was used for local tuning of hydrophobicity of C4F8 polymer on silicon and glass surfaces by pattering the polymer with O2 plasma through the shadow mask. For both purposes, one silicon shadow mask wafer can be re-used to enable lower processing costs. Thermal manipulation of fluid allows polymerase chain reaction on silicon and glass microchips, but also triggering of capillary action. However, the results of a novel method indicate possible lack of biocompatibility of oxidized silicon surfaces, which may limit the usable microchip surface materials. Microfluidic components with hydrophilic patterning for controlled capillary action can be combined with microphotonics through excitation of fluorescence with evanescent field, which has been characterized with a grating-coupled laser beam. 

The effect of support material on the dehydrogenation performance of chromia catalysts was studied with zirconium oxide (zirconia), aluminum oxide (alumina), and zirconia/alumina as the support materials. The dehydrogenation performance of the supports and chromia catalysts was studied by in situ infrared and Raman spectroscopies and activity measurements. The active surface sites of the supports and catalysts were characterized by in situ infrared and Raman spectroscopies using probe molecules, and the surface of zirconia was studied in more depth by modeling with density functional theory (DFT). The characterization experiments and DFT calculations revealed the amphoteric character of the hydroxyl groups of zirconia and the presence of coordinatively unsaturated (c.u.s.) acid–base pairs. The hydroxyls of alumina exhibited similar basicity to those of zirconia, but their acidity was lower. Lewis acid and Lewis base sites were observed for alumina, but they did not form c.u.s. acid–base pairs as on zirconia. The deposition of zirconia on alumina decreased the Lewis acidity of the c.u.s. sites and the acidity of the hydroxyls, while the total basicity of the material appeared to increase. The addition of chromium also appeared to increase the basicity. Zirconia was the most active and selective support material in the dehydrogenation of isobutane. The high activity was suggested to originate from the acid–base pairs that were present only on the zirconia surface. The performance of zirconia/alumina resembled that of alumina more than that of zirconia. The benefit of zirconia deposition on alumina was a lower coke deposition rate than on alumina due to the lower Lewis acidity of the zirconia/alumina. However, the cracking activity of alumina was not influenced by zirconia deposition and this was attributed to the presence of similar hydroxyls. The chromia/zirconia catalyst was the most active dehydrogenation catalyst. The deposition of zirconia on alumina decreased the activity of the chromia catalysts. This was attributed to an incomplete monolayer of zirconia on alumina, which enabled chromium to interact with both zirconia and alumina. In contrast to the supports, the rate of coke deposition was not influenced by the acid–base properties of the catalysts but followed the dehydrogenation activity, and the formation of cracking products was due to thermal cracking. It was concluded that deposition of zirconia on alumina is not beneficial for chromia dehydrogenation catalysts. High surface area zirconias should be investigated instead. Keywords dehydrogenation, zirconia, alumina, chromia, infrared, Raman, DFT 

Transfer from aluminum to copper metallization and decreasing feature size of integrated circuit devices generated a need for new diffusion barrier process. Copper metallization comprised entirely new process flow with new materials such as low-k insulators and etch stoppers, which made the diffusion barrier integration demanding. Atomic Layer Deposition technique was seen as one of the most promising techniques to deposit copper diffusion barrier for future devices. Atomic Layer Deposition technique was utilized to deposit titanium nitride, tungsten nitride, and tungsten nitride carbide diffusion barriers. Titanium nitride was deposited with a conventional process, and also with new in situ reduction process where titanium metal was used as a reducing agent. Tungsten nitride was deposited with a well-known process from tungsten hexafluoride and ammonia, but tungsten nitride carbide as a new material required a new process chemistry. In addition to material properties, the process integration for the copper metallization was studied making compatibility experiments on different surface materials. Based on these studies, titanium nitride and tungsten nitride processes were found to be incompatible with copper metal. However, tungsten nitride carbide film was compatible with copper and exhibited the most promising properties to be integrated for the copper metallization scheme. The process scale-up on 300 mm wafer comprised extensive film uniformity studies, which improved understanding of non-uniformity sources of the ALD growth and the process-specific requirements for the ALD reactor design. Based on these studies, it was discovered that the TiN process from titanium tetrachloride and ammonia required the reactor design of perpendicular flow for successful scale-up. The copper metallization scheme also includes process steps of the copper oxide reduction prior to the barrier deposition and the copper seed deposition prior to the copper metal deposition. Easy and simple copper oxide reduction process was developed, where the substrate was exposed gaseous reducing agent under vacuum and at elevated temperature. Because the reduction was observed efficient enough to reduce thick copper oxide film, the process was considered also as an alternative method to make the copper seed film via copper oxide reduction 

Thin films of various metal fluorides are suited for optical coatings from infrared (IR) to ultraviolet (UV) range due to their excellent light transmission. In this work, novel metal fluoride processes have been developed for atomic layer deposition (ALD), which is a gas phase thin film deposition method based on alternate saturative surface reactions. Surface controlled self-limiting film growth results in conformal and uniform films. Other strengths of ALD are precise film thickness control, repeatability and dense and pinhole free films. All these make the ALD technique an ideal choice also for depositing metal fluoride thin films. Metal fluoride ALD processes have been largely missing, which is mostly due to a lack of a good fluorine precursor. In this thesis, TiF4 precursor was used for the first time as the fluorine source in ALD for depositing CaF2, MgF2, LaF3 and YF3 thin films. TaF5 was studied as an alternative novel fluorine precursor only for MgF2 thin films. Metal-thd (thd = 2,2,6,6-tetramethyl-3,5-heptanedionato) compounds were applied as the metal precursors. The films were grown at 175–450 °C and they were characterized by various methods. The metal fluoride films grown at higher temperatures had generally lower impurity contents with higher UV light transmittances, but increased roughness caused more scattering losses. The highest transmittances and low refractive indices below 1.4 ( = 580 nm) were obtained with MgF2 samples. MgF2 grown from TaF5 precursor showed even better UV light transmittance than MgF2 grown from TiF4. Thus, TaF5 can be considered as a high quality fluorine precursor for depositing metal fluoride thin films. Finally, MgF2 films were applied in fabrication of high reflecting mirrors together with Ta2O5 films for visible region and with LaF3 films for UV region. Another part of the thesis consists of applying already existing ALD processes for novel optical devices. In addition to the high reflecting mirrors, a thin ALD Al2O3 film on top of a silver coating was proven to protect the silver mirror coating from tarnishing. Iridium grid filter prototype for rejecting IR light and Ircoated micro channel plates for focusing x-rays were successfully fabricated. Finally, Ir-coated Fresnel zone plates were shown to provide the best spatial resolution up to date in scanning x-ray microscopy 

Most electrical, magnetic or optical devices are today based on several, usually extremely thin layers of different materials. In this thesis chemical synthesis processes have been developed for growth of less stable and metastable layers, and even multilayers, of nickel containing compounds. A chemical vapor deposition (CVD) method for deposition of metastable Ni3N has been developed. The deposition process employs ammonia as nitrogen precursor. An atomic layer deposition (ALD) process for deposition of both polycrystalline and epitaxial NiO and using low oxygen activity, has also been developed. Both deposition processes utilizes bis(2,2,6,6-tetramethy1-3,5-heptanedionato)nickel(II) (Ni(thd)2) as the metal precursor. The Ni3N deposition proceeds via surface reactions. The growth rate is very sensitive to the partial pressure of ammonia, why adsorbed —NH, species are believed to be of importance for the film growth. Similar reactions can be expected between the metal precursor and H2O. For ALD of NiO a large excess of water was needed For the multilayered structures of Ni3N/NiO, growth processes, working at low activities of oxygen and hydrogen, are needed to avoid oxidation or reduction of the underlying layer. Chemical vapor growth methods such as CVD and ALD are often suffering from using high activities of hydrogen or oxygen to deposit metals and oxides. An alternative deposition pathway for metal deposition, without any hydrogen in the vapor, has been demonstrated. The metal has been formed by decomposition of the metastable nitride Ni3N in a post-annealing process. Ni3N decomposes via different mechanisms, depending on environment in the annealing process. The different mechanisms result in different degrees of ordering in the resulting Ni films. From the knowledge gained about the chemical growth of NiO and Ni3N as well as the decomposition of Ni3N, well-defined multilayer structures have been produced in different combinations of NiO, Ni3N and Ni. 
 

We are interested in the development of processes for depositing thin films of TiN-C and W-N-C by PEALD (Plasma Enhanced Atomic Layer Deposition) for the production of Metal/Insulator/Metal (MIM) capacitor electrodes MIM capacitors are used in many applications such as DRAM memories, analog/digital converters, RF filtering or decoupling. In order to meet the required performance, the Ta2O5 material was chosen as the dielectric. Good voltage linearity and high dielectric permittivity make it an excellent material for this application. In the case of electrodes, the choice was made for the Ti-N-C and W-N-C ternary systems which have low resistivity, high work function and good thermodynamic stability with Ta2O5. The PEALD method was chosen as the deposition method because it allows conformal deposition in the trenches and the deposition temperatures are compatible with the processes for developing integrated circuits (less than 400°C). We started the analysis of Ti-N-C and W-N-C deposits by performing thermodynamic simulations of the ALD deposit from the precursors TDMAT and BTBMW, respective precursors of Ti-N-C and W-N-C deposition. Regarding the Ti-N-C system we obtain the following results: the film obtained is composed of a solid solution Ti(C,N) + Cgraphite. the amount of carbon in the solid solution increases as the temperature increases and the pressure decreases. For the W-N-C system, we obtain a WN-WC mixture but as soon as the temperature increases and/or the pressure decreases, WN being unstable, the nitrogen disappears from the composition of the solid. We then showed that the PEALD deposits of our films meet the ALD deposit criteria. the growth rate is independent of the temperature in the “ALD window” the saturation of the surface of the substrate is obtained when the duration of injection of the precursor increases Studies of the influence of the deposition parameters on the properties of Ti films -N-C and W-N-C have shown that the temperature and the plasma play a very important role on the properties of the deposited films.

Indeed, the increase in temperature makes it possible to reduce the resistivity of the deposited films. The plasma remains the most important parameter. First of all, by its nature, in fact a nitrogen plasma will lead to depositing films of high resistivity, whereas a hydrogen plasma tends to reduce this resistivity by eliminating the carbon impurities and by promoting the bonds carbon-metal which lead to less resistive films than films containing many C-H bonds. Nitrogen plasma promotes C-H bonds. Then, the increase in the power or the duration of the plasma promotes the decrease in the resistivity by modifying the type of majority bonds. Indeed, at high power the formation of metal-carbon bonds (Ti-C or W-C) is favored. This change in linkage with the increase in power will also lead to a decrease in work output. Indeed, nitrogen being more electronegative than carbon, the film deposited at low power will have a higher work function. The microstructure of the deposited films also depends on the deposition power. Ti-N-C films are composed of a solid solution between TiC and TiN: Ti(N,C), as shown by thermodynamic simulations. When the deposition power is increased, the percentage of TiC in Ti(N,C) increases. As for W-N-C films, these are composed at low power of the solid solution W(N,C) whose composition limits are W2N and WC1-x. At high power, the films obtained are composed of WC1-x. We then proposed PEALD growth mechanisms for Ti-N-C deposits as well as for W-N-C deposits. In the case of Ti-N-C deposition from TDMAT, the mechanism first involves a transamination step. Then, the surface compound reacts with hydrogen radicals from the plasma. At low power, the dimethylamino –N(CH3)2 species will gradually accumulate in the Ti-N-C film after each cycle and lead to a low density of the deposited film. This facilitates the oxidation of the TiN film under exposure to air and increases its resistivity.

At high power, on the other hand, a rearrangement of carbon and nitrogen atoms, or a transposition reaction will occur. This reaction will tend to promote the formation of Ti—C bonds, which leads to a decrease in the resistivity of the films. Regarding W-N-C deposition from BTBMW, two growth mechanisms are likely to occur. In the first case, the mechanism is similar to TDMAT, it first includes a transamination step where the amine radicals are released from the BBTBMW by reaction with hydrogen, forming Me2C=CH2. Then the surface compound will react with the hydrogen radicals from the plasma. This makes it possible to create the W-N and/or W-C links. In the second hypothesis, the molecule decomposes before adsorption, radicals will be generated and the compound WC being thermodynamically more stable than WN, it will have a greater tendency to form. Finally, we studied the electrical characteristics of MIM capacitors integrating Ti-NC and W-N-C as an electrode. From a morphological point of view, the PEALD deposits of Ti-N-C are consistent in the trenches and homogeneous in thickness and composition. The low PEALD deposition temperature compared to MOCVD leads to low leakage currents because the dielectric is less degraded. The results obtained with W-N-C films, integrated as electrodes in MIM capacitance structures, are as follows: the capacitance values are higher than those obtained with TiN alone. This can be attributed to a thinner parasitic interface layer in the case of W-N-C. The voltage linearity is degraded. This could be due to the hydrogen plasma effect which generates defects in the dielectric, and degrades the linearity by trapping charges for example. In negative polarization the leakage currents are degraded. This is probably due to the plasma effect, as it degrades the electrode/dielectric interface. On the other hand, the leakage currents have been improved in positive polarization thanks to the higher work function of W-N-C compared to TiN.

The W-N-C resistivity is stronger than the resistivity of TiN, which can generate a resistance in series with the capacitance, and degrade the performance of MIMs It would be interesting later, to study in more detail the interface between the W-N-C electrodes (or Ti-N-C) and the dielectric in order to understand the improvement of the capacitance value and the degradation of the linearity. The TEM images showed an absence of interface layer between the dielectric and the W-N-C electrodes, however these results must be coupled with XPS and XRR analyses. We have shown that the electrical characteristics such as resistivity and work function are a function of the properties of the deposited material, and that these depend on the deposition parameters. It would then be interesting to use another precursor to compare the physico-chemical and electrical properties of the films obtained. It appears after this study that electrodes made from Ti-N-C or W-NC ternary compounds improve the electrical performance of MIM capacitors compared to TiN usually used in microelectronics. On the other hand, the PEALD deposition process is not suitable for the deposition of electrodes, especially the upper electrode. In fact, the plasma degrades the properties of the dielectric and therefore degrades the electrical performance of the capacitors. It would be interesting to use a process with a remote (or indirect) plasma that does not directly affect the substrate.
 

The aim of this work was to understand on the effect of thermal treatments, precursor and support on the interaction between the support and cobalt species, and further how the interaction affects the reducibility and dispersion of the catalyst. Silica and alumina supported cobalt catalysts were prepared, characterised and tested for catalytic activity. The catalysts were prepared by gas phase deposition techniques from cobalt acetylacetonate and cobalt carbonyl and by incipient wetness impregnation from cobalt nitrate. One of the goals was to investigate whether atomic layer deposition (ALD) using cobalt acetylacetonate precursors can produce well dispersed reducible cobalt catalysts. The cobalt acetylacetonates, Co(acac)2 and Co(acac)3, were found suitable for ALD. Silica supported catalysts were prepared by chemisorption of Co(acac)3, while Co(acac)2 was used on alumina. The main mode of interaction on silica was the ligand exchange reaction with OH groups. On alumina both the ligand exchange reaction and dissociative adsorption occurred. Steric hindrance limited the amount of precursor on the support. The acac ligands were removed through calcination at 450 °C. The cobalt loading was increased by repeating the precursor addition and air calcination steps up to five times; samples with about 2 to 8 Co atoms per nm2 were achieved on both silica and alumina. Calcination of the cobalt acetylacetonate modified samples led to the formation of silicate- or aluminate-type species, which decreased the reducibility of the catalysts. The reducibility was enhanced when the calcination step after the last precursor reaction step was omitted (‘uncalcined’ catalysts). High reduction temperatures were still needed: the maximum metal surface area was obtained after reduction at 500–600 °C. The cobalt dispersion on the uncalcined ALD catalysts was, in general, higher than on the calcined catalysts. Furthermore, the dispersion was higher on the alumina supported catalysts than on corresponding silica supported ones. The interaction between cobalt and silica on the nitrate based catalysts was found to be weak, which led to high reducibility but modest dispersion. The main cobalt species on the catalysts was Co3O4. The use of reduction temperatures above 400 °C induced sintering or migration of silica, which decreased the cobalt surface area significantly. Dicobalt octacarbonyl, Co2(CO)8, was adsorbed on silica by vapour-phase adsorption in a fluidised bed reactor under CO. The carbonyl interacted through hydrogen bonding and rearranged to Co4(CO)12 on the support. The amount of precursor that adsorbed on the support was limited by steric hindrance. Decarbonylation was achieved by heat treatment, and higher cobalt loadings were obtained by repeating the deposition and decarbonylation steps. Chemisorption of hydrogen on cobalt was found to be activated and highly reversible. The effect was stronger on alumina than on silica supported samples. The ALD and nitrate based catalysts were tested for gas phase hydrogenation of toluene, and the activity was found to correlate with the available surface area of metallic cobalt.

Research relevance . The problem of regulating the functional properties of polymeric materials (thermal, electrophysical, adhesive, barrier, etc.) is one of the most important in obtaining products for various purposes. One of the promising directions for solving this problem is the modification of the surface of solid polymeric materials, which does not require changes in the existing technology for the production and processing of organic polymers. These methods include the method of molecular layering (ML) [1], based on the implementation of chemical reactions between reagents supplied from outside and functional groups on the surface of a solid substrate under conditions of maximum distance from equilibrium.

It has been established that the introduction of element-containing structures into the surface layer of phenol-formaldehyde, polyamide, epoxy and other polymeric materials by the MN method makes it possible to influence their thermal-oxidative properties and flammability, change the electrical characteristics of polyethylene and polytetrafluoroethylene [2]. It is shown that as a result of the interaction of the surface reactive centers of polymers (-OH, =NH, -CH2-, etc.) with volatile reagents (PCb and VOCb), chemical grafting of modifying structures occurs. However, given the complexity of the "polymer-modifier" system, to understand the nature of the observed changes in functional properties, a deeper study of the structure of the surface layer of the initial modified materials, their morphology, and the distribution of chemically active centers in them is required.

Polyethylene, polyvinyl chloride and products made from them are widely used both in engineering and in everyday life. Depending on the intended purpose, it is necessary to enhance their various performance characteristics. At the same time, in a number of cases, certain physical effects (UV, laser radiation, corona discharge) are used to increase the efficiency of chemical modification of the surface of solids.

In this regard, it is relevant to study the structural-chemical transformations on the surface of polyethylene and polyvinyl chloride films under the influence of halide vapors of elements of various nature, their influence on the surface-energy, thermal-oxidative, and other properties of modified polymers, as well as the study of the effect of corona pretreatment of the surface of solid-phase matrices on the effectiveness of chemical modification and functional properties of the obtained products.

Research on the topic of the dissertation was carried out within the framework of the state budget topics of St. Petersburg State Technical University (TU) (order-orders No. 1.7.04, I.1.03), grants from the Russian Foundation for Basic Research (grants No. 05-03-08172, 07-03-00330) and the government of St. Petersburg (grant series PSP Xo080191).

Purpose of the work: to study the relationship of structural-chemical transformations and functional properties of the products of the interaction of films of high-pressure polyethylene (LDPE) and polyvinyl chloride (PVC) with halogenide vapors of nioc^ora, titanium, vanadium and silicon, incl. combined with exposure to a corona discharge.

The main objectives of the study:

- development of modes of synthesis of element-containing structures on
the surface of HDPE and PVC films by gas-phase treatment with chlorides
PC1 ₃ , VOCl ₃ , T1CI4 and Si(CH3)2Cb, including
pre-treatment in a corona discharge;

study of the relationship between chemical and structural transformations of solid-phase polymers upon interaction with vapors of the indicated halides;

study of the influence of the chemical composition, structure and morphology of the surface of modified polymeric materials on their thermal-oxidative, surface-energy and electrophysical properties.

Scientific novelty:

it has been shown that in the presence of oxygen, the grafting of element-containing structures to the surface of LDPE and PVC is carried out due to the formation of the E-O-C bond (E = P, V, Ti, Si). At the same time, according to AFM data, as a result of treatment with phosphorus chloride and vanadium oxochloride vapors, hydrated structures appear on the polymer surface in the case of phosphorus-containing films and cavities in the case of vanadium-containing films, with lateral dimensions of 200–300 nm, the adhesive characteristics of which differ significantly from the rest of the surface. The synthesis of titanium- and silicon-containing structures contributes to the formation of a uniform surface layer;

It was established that the structures formed on the surface of LDPE are the main Brønsted centers with pK _A -10.5. Treatment of PVC films with halide vapor leads to the appearance of acid Brønsted centers with pK _A of 1.3 and 2.5. The presence on the polymer surface of groupings containing -E=0 bonds contributes to an increase in the concentration of Lewis basic centers with pK _A -4.4 and -0.29.

- the relationship of the chemical composition and topography
of the surface of polymeric materials with its energy characteristics was
revealed. It has been established that in the case of phosphorus- and vanadium-containing LDPE
and PVC, the components of surface
energy are redistributed with a significant increase in the polar component and total
energy by 1.4-1.8 times. When the surface is treated with pairs of titanium
and silicon halides, a decrease in surface energy by 30-40% is observed due to a
decrease in its dispersion component;

- it is shown that the introduction of surface element-containing structures
has an effect on the initial stage of the thermal-oxidative destruction
of LDPE, as evidenced by an increase in the temperatures of the beginning of decomposition and
10 ⁰ /o-th weight loss by 15-75C;

- the effect of pre-treatment of the initial matrix by a corona discharge on the efficiency of the chemical modification of LDPE and PVC with phosphorus and vanadium halides and the functional properties of the resulting products was studied. It has been established that the impact of a positive corona discharge contributes to a further decrease in the surface energy by 10-20% compared to chemically modified samples, and a negative one - its increase by 10%. In the case of a preliminary physical impact on the PVC surface, complete wetting of the films with test liquids is observed, regardless of the sign of the charge.

The practical significance of the work. The results obtained can be used in the development of the technology for modifying the surface of polymeric materials according to the (MN) method to control their thermal-oxidative stability and wetting, which is important for applying various coatings, paints, and controlling the barrier properties of materials.

Approbation of work.The main results of the work are presented at scientific conferences and seminars: International Scientific Conference "Thin Films and Nanostructures" (Moscow, 2005); II, III and IV St. Petersburg Conference of Young Scientists "Modern Problems of Polymer Science" (St. Petersburg, 2006, 2007 and 2008); III All-Russian Conference (with international participation) "Surface Chemistry and Nanotechnology", (St. Petersburg-Khilovo, 2006); International scientific and technical conference "Fundamental problems of radio-electronic instrumentation" (Moscow, 2006); All-Ukrainian Conference of Young Scientists with International Participation "Nanomaterials in Chemistry, Biology and Medicine", (Kyiv, 2007); XVIII Mendeleev Congress on General and Applied Chemistry, (Moscow, 2007); III International Scientific and Technical Conference "Polymer Composite Materials and Coatings", (Yaroslavl, 2008); International scientific and technical school-conference "Young scientists - science, technology and professional education" (Young scientists - 2008), (Moscow, 2008).

Publications. The materials of the dissertation work are reflected in 12 scientific publications, including 5 articles, one of which was published in a journal included in the list of the Higher Attestation Commission, and abstracts of 7 reports.

The volume and structure of the dissertation. The dissertation is presented in 149 steps of matpinppig.nlgg. tekt.tya splgpchgit CHA ppgapyu ”23 of Table II consists of an introduction, a literature review, including two chapters, four chapters describing the experimental part, conclusions, a list of references, including 174 titles of works by domestic and foreign authors

As current silicon-based microelectronic devices and circuits are approach[1]ing their fundamental limits, the research field of nanoelectronics is emerging worldwide. With this background, the present thesis focuses on semiconduc[1]tor nanoelectronic devices based on ballistic and quantum effects. The main material studied was a modulation doped In0.75Ga0.25As/InP semiconductor two-dimensional electron gas grown by metal-organic vapor phase epitaxy. The thesis covers mainly three types of devices and their twofold integra[1]tion: in-plane gate transistors, three-terminal ballistic junctions and quan[1]tum dots. Various advanced nanofabrication tools were used to realize the devices, such as electron beam lithography, focused ion beam lithography and atomic layer deposition. The theories behind the analysis of the experimen[1]tal data include principles of field effect transistors, the Landauer-Büttiker formalism, the constant interaction model, etc. The principles of in-plane gate transistors can be explained by a classical theory. The source, drain, one-dimensional channel and two side gates were in the same plane; a setup that can be obtained by single step lithography. The gating efficiency of the two independent gates was voltage-dependent, which resulted in a simplified circuitry for implementing a logic function. At room temperature, an SR latch with a signal gain of ∼4 was realized by the integration of two in-plane gate transistors. Three-terminal ballistic junctions are nonlinear devices based on ballistic electron transport. When two terminals are applied with voltages, the third terminal will output a voltage close to the more negative voltage in the two inputs, as opposed to a simple average of the two. From numerical calcula[1]tions, this ballistic effect persists up to room temperature. Three-terminal ballistic junctions are so robust that nonlinearity is observable in asymmet[1]ric devices and relatively large devices. They can be fabricated on several materials by assorted techniques. The junctions find their applications in analogue frequency mixers, phase detectors and digital SR latches and the circuits are simpler than conventional designs. The intrinsic speed of the devices is in the GHz or THz regime by virtue of the ballistic transport. It is believed that as-built junctions have a potential as building blocks in future nanoelectronics. Quantum dots are zero-dimensional boxes for electrons with a decent resemblance to natural atoms. Due to their nanoscale size, numerous inter[1]esting quantum effects can be observed. Gate-defined quantum dots were fabricated in InGaAs/InP by incorporating a high-k HfO2 (20-30 nm thick, grown by atomic layer deposition) as the gate dielectric. The gate leakage was suppressed and the gating efficiency improved. At 300 mK, charge stabil[1]ity diagrams of single and double quantum dots were measured and studied in detail. Zeeman splitting in a parallel magnetic field and charge sensing by nearby quantum point contacts were also investigated. The single and dou[1]ble quantum dots are expected to be useful in fields including single electron logic, stochastic resonance, spintronics, quantum computing, etc.

This thesis is concerned with the chemical vapour deposition (CVD) of tungsten and zirconium nitride and carbonitride thin films. [W(µ-Nt Bu)(Nt Bu)Cl2(H2Nt Bu)]2, [W(Nt Bu)2Cl2(TMEDA)] (TMEDA = N,N,N’,N’-tetramethylethylenediamine) and [W(Nt Bu)2Cl2(py)2] (py = pyridine) have been investigated as potential precursors. Additionally, two novel precursors, [W(Nt Bu)2Cl{N(SiMe3)2}] and [W(Nt Bu)2(η5 -Cp’)(η1 -Cp’)] (Cp’ = methylcyclopentadienyl), have been synthesised via metathesis reactions of [W(Nt Bu)2Cl2(py)2] with the appropriate lithium or sodium amide salt. The attempted synthesis of [W(Nt Bu)2Cl{N(H)NMe2}] by lithium salt metathesis was unsuccessful due to polymerisation. Cyclopentadienyl-based precursors to zirconium carbonitride have been investigated, including [ZrCp2(NMe2)2], [ZrCp’2(NMe2)2] and four novel compounds: [ZrCp2(η2 -MeNCH2CH2NMe)], [ZrCp’2(NEt2)2], [ZrCp’{(i PrN)2CNMe2}2Cl] and [ZrCp’2{(i PrN)2CNMe2}Cl] (Cp = 5 -cyclopentadienyl, Cp’ = 5 - methylcyclopentadienyl). The compounds were synthesised via sodium or lithium metathesis reactions. All compounds synthesised were characterised by NMR, mass spectrometry, elemental analysis and IR, and their decomposition was investigated by thermogravimetric analysis (TGA). The molecular structures of [ZrCp’{(i PrN)2CNMe2}2Cl] and [ZrCp’2{(i PrN)2CNMe2}Cl] were determined by single crystal X-ray diffraction. Thin films of tungsten carbonitride, zirconium carbide and zirconium carbonitride were deposited by low pressure CVD (LPCVD) at 550-600 °C and 0.1 Torr using the aforementioned precursors. In addition, aerosol-assisted CVD (AACVD) of [W(µ-Nt Bu)(Nt Bu)Cl2(H2Nt Bu)]2 in toluene was investigated as a comparison. All films deposited via CVD were characterised using wavelength dispersive X-ray spectroscopy (WDX), glancing-angle XRD, UV/Vis spectroscopy (for reflectance and transmission measurements) and their morphology and thickness were investigated using SEM imaging. Plasma-enhanced atomic layer deposition (PEALD) using [ZrCp2(NMe2)2] and [ZrCp2(η2 -MeNCH2CH2NMe)] was studied with the view to depositing stoichiometric ZrN. It showed these precursors to be insufficiently volatile for the ALD system as the growth rates were low. The ALD progress was analysed using in situ ellipsometry and mass spectrometry.

PENG, QING. Nanoscale Engineering Materials with Supercritical Fluid and Atomic Layer Deposition. (Under the direction of Gregory N. Parsons.) With the development of material science and technology, modification of substrates, which have random geometry and high aspect ratio three dimensional (3D) complex structures, with desired functional, reactive and stable coatings becomes important and challenging. The ability to fabricate mono- or multi-layers of functional materials with precisely controlled dimensions, finely tuned composition and molecular structures, attracts significant interests in materials science and is the key to construct such devices and structures at nano- and micro- scale with desired properties. In this study, supercritical carbon dioxide (scCO2) has been studied as an alternative route for modifying substrates due to the unique gas-like (low viscosity, high diffusivity and zero surface tension) and liquid-like properties (high density). 1) The reaction kinetics of metal oxides thin film deposition from pyrolysis of metal organics in scCO2 was studied in detail. This method was demonstrated as a powerful technique to coat oxides, including Al2O3, Ga2O3 and others, into 3D high aspect ratio complex structure of carbon nanotubes (CNTs) forest. 2) The low temperature scCO2 based hydrogenolysis process was developed as a useful way to functionalize aligned CNTs forest with dense Nickel nanoparticles. On the second part of this work, atomic layer deposition (ALD) /molecular layer deposition (MLD), as a vapor phase, stepwise and self-limiting vacuum based deposition process, was demonstrated as a powerful way to form highly conformal and uniform film onto substrates, even into highly complex 3D complex structures. In this study, 4) Metal oxide ALD is applied onto 3D electrospun polymer microfiber mats template to illustrate an effective and robust strategy to fabricate long and uniform metal oxide microtubes with precisely controllable wall thickness. Designer tubes of various sizes and different materials were demonstrated by using this method. 5) By further extending this technique, complex coaxial Al2O3/ZnO/Al2O3 multilayed microtubular structure is fabricated, which provides an unique platform to study the solid state reaction and diffusion process (Kirkendall Effect) between Al2O3 shells and the confined middle ZnO layers by annealing the samples at 700 ˚C. 6) The extension of ALD-MLD process of polyamides, zinc hybrid, aminosilane self assembly monolayers were studied by various techniques to illustrate the surface reaction mechanism.

NA, JEONG-SEOK. Nanoscale Assembly for Molecular Electronics and In Situ Characterization during Atomic Layer Deposition. (Under the direction of Dr. Gregory N. Parsons.) The work in this dissertation consists of a two-part study concerning molecular-based electronics and atomic layer deposition (ALD). As conventional “top-down” silicon-based technology approaches its expected physical and technical limits, researchers have paid considerable attention to “bottom-up” approaches including molecular-based electronics that self assembles molecular components and ALD techniques that deposit thin films with atomic layer control. Reliable fabrication of molecular-based devices and a lack of understanding of the conduction mechanisms through individual molecules still remain critical issues in molecular-based electronics. Nanoparticle/molecule(s)/nanoparticle assemblies of “dimers” and “trimers”, consisting of two and three nanoparticles bridged by oligomeric ethynylene phenylene molecules (OPEs), respectively, are successfully synthesized by coworkers and applied to contact nanogap electrodes (< 70 nm) fabricated by an angled metal evaporation technique. We demonstrate successful trapping of nanoparticle dimers across nanogap electrodes by dielectrophoresis at 2 VAC, 1 MHz, and 60 s. The structures can be maintained electrically connected for long periods of time, enabling time- and temperature-dependent current−voltage (I−V) characterization. Conduction mechanisms through independent molecules are investigated by temperature dependent I−V measurements. An Arrhenius plot of log (I) versus 1/T exhibits a change of slope at ~1.5 V, indicating the transition from direct tunneling to Fowler−Nordheim tunneling. Monitoring of the conductance is also performed in real-time during trapping as well as during other modification and exposure sequences after trapping over short-term and long-term time scales. The real-time monitoring of conductance through dimer structures during trapping offers immediate detection of a specific fault which is ascribed to a loss of active molecules and fusing of the nanoparticles in the junction occurring mostly at a high applied voltage (≥3 VAC). After successful trapping, the sample exposure to air reveals a small rapid decrease in current, followed by a slower exponential increase, and eventual current saturation. This work also reports on the dependence of electron transport on molecular length (2 to 4.7 nm) and structure (linear-type in dimers and Y-type in trimers). The extracted electronic decay constant of ~0.12 Å-1 and effective contact resistance of ~4 MΩ indicate a strong electronic coupling between the chain ends, facilitating electron transport over long distances. A three terminal molecular transistor is also demonstrated with trimers trapped across nanogap electrodes. The source-drain current is modulated within a factor of 2 with a gate bias voltage of −2 to +2 V. A subthreshold slope of ~110 mV/decade is obtained. Finally, we report on both fundamental understanding and application of atomic layer deposition. First, in situ analysis tools such as quartz crystal microbalance and electrical conductance measurements are combined to reveal direct links between surface reactions, charge transfer, and dopant incorporation during ZnO and ZnO:Al ALD. Second, the ability of ALD to form uniform and conformal coating onto complex nanostructures is explored to improve the ambient stability of single molecules/nanoparticle assemblies using Al2O3 ALD as an encapsulation layer. In addition, the ability to shield the surface polarity of ZnO nanostructures using Al2O3 + ZnO ALD, leading to hierarchical morphology evolution from one-dimensional ZnO nanorods to three-dimensional ZnO nanosheets with branched nanorods during hydrothermal growth is investigated.

The present PhD thesis is devoted to the design and fabrication of multi-functional Fe2O3-based nanomaterials by means of vapor phase techniques, such as chemical vapor deposition, both thermal (CVD) and plasma enhanced (PECVD), atomic layer deposition (ALD) and sputtering, either as such or combined into original preparation strategies. The performed research activities have covered the entire material production chain, encompassing the preparation of the molecular precursor, the material development and chemico-physical characterization, up to the ultimate functional validation for energy and environmental applications. In particular, the attention has been initially devoted to the synthesis and charac terization of a novel Fe(II) precursor [Fe(hfa)2TMEDA (hfa = 1,1,1,5,5,5-hexafluoro 2,4-pentanedionate; TMEDA = N,N,N’,N’- tetramethylethylenediamine)], possessing im proved properties for use in CVD processes with respect to the iron compounds proposed so far. The utilization of this compound in thermal CVD experiments yielded not only the most stable and widely used α-Fe2O3 phase, but also the rare and scarcely inves tigated β- and  -Fe2O3 polymorphs, that could be selectively obtained as pure phases with controlled nano-organization. In addition, Fe(hfa)2TMEDA was used in PECVD experiments as molecular source for both Fe and F thanks to the unique reactivity of non-equilibrium cold plasmas, resulting in the obtainment of F-doped α- and β-Fe2O3 nanosystems. Following the efforts devoted to the preparation of single-phase nanoma terials with improved functional performances, the fabrication of metal/oxide (M/Fe2O3, with M = Pt, Ag, Au) and oxide/oxide (CuO/Fe2O3, Fe3−xTixO4/Fe2O3) nanocomposites has finally been accomplished through the combination of CVD with sputtering or ALD. The study of the interplay between processing conditions, system features and func tional activities was proved to be a successful tool of the whole PhD research activity. To this regard, a thorough characterization of the material composition, morphology and spatial organization, micro- and nano-structure and optical properties, was carried out by the use of forefront and complementary analytical techniques. In addition, the func tional performances of selected nanosystems were investigated in view of their possible use in a variety of technological end-uses [magnetism, Li-ion batteries, gas sensing of flammable/toxic analytes, and photo-activated applications (photo-induced hydrophilic ity, photocatalytic pollutant decomposition, photocatalytic and photoelectrochemical H2 production)]. The results obtained in this PhD work demonstrate that the preparation of iron(III) oxide systems, either as such or in combination with others guest phases, with selected phase composition (α- or β- or  -Fe2O3) and nano-organization, represents a valuable an swer to meet open challenges in various high-tech applications. In particular, the adopted approaches involving vapour phase-related routes offer the possibility of future up-scaling and commercialization of the studied materials, one of the key issues for their technological exploitation in advanced devices.

Gibbs, Zachary Michael Conway (MS, Chemical Engineering) Growth and Characterization of Polymer Thin Films Grown Using Molecular Layer Deposition with Heterobifunctional Precursors Thesis Directed by Professor Steven M. George In this work, growth of thin polymer films using molecular layer deposition with heterobifunctional precursors is investigated. Several growth phenomena are observed including: loss or gain of reactive sites as a result of precursor reactivity or vapor pressure; precursor diffusion and reaction within the porous polymer film; and crosslinking. Reactions were investigated using quartz crystal microbalance, Fourier transform infrared spectroscopy, and various ex situ techniques. Reactions involving 4-azidophenylisothiocyanate and 4-aminobenzonitrile were shown to stop growth after only a few cycles which is attributed to a loss in reactive sites which was modeled by an exponentially decaying growth rate. Growth of 4- carboxyphenylisothiocyanate with TMA and water was investigated as well. Active site multiplication as a result of the trifunctionality of the TMA molecule was proposed to explain the significantly higher growth rate for TMA/CI films. TMA/H2O/CI films showed the ability to crosslink through aluminum hydroxyl condensation reactions. Upon increasing the reaction temperature, reactant diffusion was observed in the form of mass removal upon TMA exposure. This same phenomena is thought to be occurring in films grown using Diels-Alder reactions in the third section of this thesis. These films showed a strong growth rate dependence upon reactant purge time and growth temperature. FTIR seems to weakly support Diels-Alder reaction, but it appears that the primary film growth mechanism is through CVD-like diffusion and condensation reactions.

With device scaling down, metal deposition techniques with high conformality are required for contact fabrication. Since atomic layer deposition (ALD) has many advantages such as high conformality and atomic thickness controllability, it is suitable for nanoscale contact fabrications. In this study, Co ALD has been developed by using several Co precursors, CoCp(CO)2, CoCp2, and Co(iPr-AMD)2, and reactants, gas and plasma, and then studies on fundamentals and applications were performed. Co films were not deposited from thermal ALD (TH-ALD) using NH3 and H2 gas as a reactant while plasma enhanced ALD (PE-ALD) produced pure Co films by using NH3 plasma. For PE-ALD, CoCp2 showed better film quality and more suitable growth characteristics for ALD growth than CoCp(CO)2. The resistivity was very low, down to 10 μΩcm for PE-ALD Co from CoCp2. Quadrupole mass spectroscopy (QMS) and synchrotron radiation X-ray reflectivity (SR-XRR) were employed to investigate growth mechanism and initial growth of PE-ALD Co. The results showed that an interlayer was formed between PE-ALD Co and Si substrate, and the interlayer was a key role of Co deposition. In addition, the Co was grown by island growth during the initial growth stage but not layer by layer growth. The results on the interlayer from ellipsometry and transmission electron microscopy (TEM) were consistent with QMS and SR-XRR results, which indicated that the interlayer was amorphous SiNx, and it was formed due to direct exposure of NH3 plasma to Si substrate. In addition, the epitaxial CoSi2 was obtained from annealing of PE-ALD Co. The epitaxial growth was correlated with the presence of a-SiNx interlayer, in other words, the epitaxial CoSi2 was formed through interlayer mediated epitaxy mechanism. For Co(iPr-AMD)2 precursor, pure Co films were obtained by TH-ALD as well as PE-ALD. TH-ALD produced excellent conformal Co films in nanosize contact holes up to 95 % of step coverage. Co films deposited by TH-ALD with NH3 as a reactant showed superior film properties such as high density and low resistivity to that with H2. PE-ALD Co and TH-ALD Co were applied to area selective ALD (ASALD) by using octadecyltrichlorosilane (OTS) self-assembled monolayers (SAMs) for novel contact fabrications. No deposition selectivity was observed in PE-ALD Co because OTS was rapidly degraded by NH3 plasma exposure. In contrast to PE-ALD, AS-ALD was successfully achieved by using TH-ALD Co. By using patterned OTS substrate, Co line patterns with 3 μm-width were obtained without etching process. 

The research activity described in this thesis was mainly carried out in the framework of the European project REALISE “Rare earth atomic layer deposition for innovation in electronics” [1]. The principal project aims were to: • deposit high permittivity rare earth oxide layers with sub-nanometer control • integrate these dielectrics into innovative memory and communication devices The idea was to develop and optimize the deposition process of novel ultra-thin dielectric films and to investigate and characterize their physical and chemical properties before testing them for applications in innovative microelectronic devices. The growth process which was the subject of the research is atomic layer deposition (ALD). Nowadays, this technique is one of the leading technologies employed for deposition of nanometer-scale films at an industrial level. Indeed ALD allows deposition of conformal ultra-thin layers with an extremely precise thickness control. Moreover, ALD growth processes are scalable up to 8” or 12” large area substrates making this technique very promising for the necessities of high-throughput industries. Most of the materials investigated in this study belong to the rare earth oxides (REOs) family. Rare earth-based binary and ternary oxides are high dielectric constant (high-k) materials which might be successfully employed in several microelectronic fields. Indeed, the scaling down of the device dimensions requires the employment of ever-thinner insulating dielectric layers for logic, memory and communication applications. Appropriate physical and chemical stability are required together with stringent electrical requisites in terms of permittivity value and leakage current. Thus, a variety of highk dielectrics was identified, deposited using ALD and characterized in order to asses the potential implementation in advanced nano-scaled devices. The REOs were deposited on different semiconductor substrates in order to address different oxide/semiconductor interfaces. In addition, an in situ study of the ALD processes and of the film and interface optical properties was performed using spectroscopic ellipsometry during the deposition of the various stacks. Therefore, the research activity was balanced between the development of new ALD processes and the assessment and the discussion of more fundamental scientific issues connected with the structural, chemical and electrical properties of the thin films grown by ALD. 

Transparent electronics was interesting field for the application to the transparent display, solar cells, smart window, defroster, and etc. Until now conducting oxide was investigated the most enthusiastically among transparent electronics with high performance and functionality. However, recently, the application of ZnO thin films as an active channel layer of transparent thin film transistor (TFT) has become of great interests due to it’s transparency, low process temperature and high performance, such as high mobility, since ZnO is the interesting material since it correspond to the properties. Thus many groups report on the transparent TFT with ZnO active layer. First we deposited ZnO thin films by atomic layer deposition (ALD) from diethyl Zn (DEZ) as a metal precursor and water as a reactant at growth temperatures between 100 ~ 250 °C. The lowest carrier concentration was about 1017 cm-3 at the Ts = 100 °C. However, we expect that the carrier concentration was too high to apply the films to TFT, since negative VTH was observed when we applicate the ZnO to TFT active layer. Thus we introduced nitrogen doping in ZnO to reduce carrier concentration. For this purpose, we used NH4OH solution as a reactant. As a result, the carrier concentrations were decreased down to 6.13×1013 /cm3 by the nitrogen incorporation in ZnO. Additionally, we introduced oxygen plasma as a reactant to for reducing carrier concentration. We deposited ZnO thin films by plasmaenhanced atomic layer deposition (PE-ALD) technique, and investigated structural, electrical, and optical properties of the films. ALD mode growth was achieved at between 200 and 250 °C. The resistivity was too high to measure under 200 °C of the growth temperature. However, the resistivity decreased with increasing growth temperature. According to the photoluminescence spectra, at low growth temperature band to band and oxygen interstitial emissions, act as acceptor, were observed, however the oxygen interstitials decreased and oxygen vacancies increase with increasing growth temperature. This result implies that oxygen defect is strongly related to the electrical properties of PE-ALD ZnO films. Thin film transistors were fabricated using ALD ZnO:N thin films with different N contents as active channel layers. Due to the electrical properties changes in ZnO:N films, the device properties were significantly changed by amount of nitrogen incorporation. Especially, threshold voltages were changed from 20.0 to 3.1 V by adjusting nitrogen doping. And also, DC bias stability was increased by the increment of nitrogen concentration. And also, we fabricated ALD ZnO:N TFT on the flexible PEN substrate to study the bending effect on the device properties. The threshold voltage was changed to 14.8 V and 16.2 V from 15.5 V during upward and downward bending with 0.83 cm of radius of curvature respectively. And also, the saturation mobility was changed a little from the 3.3 cm2/Vs to 3.1 cm2/Vs and 3.6 cm2/Vs under upward and downward bending with 0.83 cm of radius of curvature respectively. The downward bending raises the threshold voltage and reduces saturation mobility. On the contrary, the upward bending reduces the threshold voltage and raises saturation mobility. From the C-V measurement of the MIM and MISM structure, the device properties shift is attributed not by the gate insulator but by the properties change in the ZnO:N depending on gate insulator. The properties change in the ZnO:N is originated from the piezoelectric effect of ZnO:N. And also, we fabricated plasma enhanced atomic layer deposition (PEALD) ZnO TFT. However the TFT did not modulate within the gate voltage sweep range, due to high resistance of our PEALD ZnO films. After we expose UV light to the PE-ALD ZnO active layer, the device show the TFT modulation by gate voltage, and we found the VTH moved to the negative gate voltage direction with increasing UV exposure time. In detail, VTH moves from 17.8 V to -6.5 V with increasing UV exposure time from 3 min to 120 min. The UV exposure is effective way of activating resistive ZnO layer and control of VTH under low temperature. Additionally, we fabricated back gate type ALD ZnO:N TFT with enhancement mode to study UV exposure effect. We exposed UV light to the TFT and measured the change of VTH in the device to developing easy way for changing the enhancement to depletion mode. We obtained depletion mode TFT after UV exposure, however the properties are not stabilized and the VTH return to the initial value. Thus we passivated ZnO:N with ALD Al2O3 after UV exposure. We prove that the passivation was effective way for stabilize the UV exposure TFT. 

The downscaling of the DRAM device is necessary to achieve higher speed with less power consumption. It is getting difficult to meet the new requirements with the existing SiO2 or Si3N4 due to their low dielectric constants and tunneling leakage currents through the thin layers. For this reason, high-k materials enabling high-k and low leakage currents with physically thicker film have received considerable attention. Among the candidate for DRAM capacitor dielectrics, SrTiO3 is a promising candidate for giga-bit scale dynamic random access memory (DRAM) capacitors because of its high dielectric constant, high breakdown strength and good thermal stability.

SrTiO3 films were deposited on 20 nm-Ru/25 nm-TiN/p-type Si (100) substrates by plasma-enhanced atomic layer deposition at a deposition temperature of 225oC and a deposition pressure of 3 Torr using 0.2M Sr(DPM)2 dissolved in butyl acetate and TTIP as precursors and O2 plasma as an oxidant.

SrO and TiO2 films were grown separately to investigate the ALD characteristics. The thickness per cycle of SrO and TiO2 are saturated to 0.054 nm/cycle 0.036 nm/cycle at 225℃, respectively. The composition of STO films was controlled by changing the number of each precursor cycles, and stoichiometric SrTiO3 films were obtained when one super-cycle consisted of six TiO2 cycles and seven SrO cycles.

The deposited-SrTiO3 films were crystallized after annealing at 600oC for 10min under N2 ambient and the dependence of the dielectric constant on SrTiO3 film thickness was investigated for less 50 nm-thick SrTiO3 films after the annealing process at 600oC for 10min under N2 ambient. The dielectric constants of the films having thickness higher than 20 nm were not as sensitive to the film thickness with a relatively constant value of about 65. However, the dielectric constants of the films with thickness under 15 nm were dramatically decreased with decreasing the film thickness. This change was related to the film crystallinity. Moreover, it was confirmed that non-stoichiometric region near the interface of SrTiO3 film and Ru bottom electrode existed and was intermixed with SrTiO3 and Ti-O phases. The dielectric constant of only SrTiO3 film being excluded the interface layer was about 85. As the crystallization of deposited-SrTiO3 film by annealing at 600℃ for 10min in ambient N2, the leakage current density abruptly increased from the level of 10-7~10-8 at ±1V to the level of 10-1~10-2 at ±1V owing to formation of the grain boundary as the leakage current path, irrespective of the film thickness.

To improve dielectric properties, crystallization the seed-layer was introduced. First, seed layer was prepared by depositing 2.7-nm SrO and post-annealing in ambient N2 at 600°C for 10min before SrTiO3 deposition. By inserting of the SRO seed-layer between the SrTiO3 thin film and the Ru bottom electrode, the crystallinities of the annealed-SrTiO3 films were enhanced, especially film thickness with below 15 nm. In aspect of dielectric properties, SRO seed-layer helped to increase the dielectric constant of the SrTiO films, especially films with thicknesses below 15 nm (3the dielectric constant of 10 nm-SrTiO3 films was increased from 15.7 to 50.3) and the thickness dependency of the dielectric constant was reduced. Moreover, it was confirmed that the low-k interfacial layer between SrTiO film and bottom electrode was reduced by inserting of the seed-layer. For optimization of SRO seed layer formation, the dependence of dielectric properties of SrTiO films on the thickness of the inserted SrO layer was investigated. 33As the SrO layer was reduced below 1.35 nm, the dielectric constants of the SrTiO3 films drastically decreased and the dielectric constants of SrTiO3 thin films decreased continuously as the thickness of the inserted SrO layer was increased beyond 1.35 nm. From the above results, it appears that the optimized thickness of the inserted SrO layer for forming a seed layer is 1.35 nm, at which had the highest dielectric constant.

If the RuO2 layer was used as the substrate instead of Ru, it was thought that the sufficient oxygen supply might be possible for transformation of deposited-SrO to the SrRuO3 layer (SrO + RuO2 → SrRuO3) and the O2 ambient annealing could be possible as well as N2 ambient annealing for crystallization of SrTiO3 thin films. Therefore, SrRuO3 was introduced as a crystallization seed layer and formed through deposition of a SrO layer on a RuO2 substrate followed by O2-annealing instead of on a Ru substrate followed by N2-annealing, as the second method for seed layer formation. The SrRuO3 layer was successfully formed after annealing of a 2.7-nm SrO/RuO2 sample at 600°C. As the results of introducing SrRuO seed layers, 3the dielectric constant of 10 nm-thick SrTiO3 thin films increased to 83 compared with films deposited on Ru directly and seed formed on Ru substrate, respectively.

Finally, the effect of alumina (Al2O3) insertion on electrical properties of SrTiO3 was investigated. To investigate the variation of electrical properties of Al2O3-added SrTiO3 films, the added Al2O3 were inserted in SrTiO3 thin films by two different ways. The first method for reducing leakage current by addition AlOto SrTiO films is the doping of AlO in SrTiO thin film and the second method is the insertion of nano-scale-thick AlO layer in SrTiO thin film as the leakage current blocking layer. In the case of SrTiO film deposited on SrRuO seed layer formed on RuO substrate, the best optimized condition was obtained by 23 3233233332insertion of Al2O3 layer with 10 cycles (corresponding thickness of about 1.2 nm) and in this case, the leakage current density at 1V and dielectric constant were 9x10-7 A/cm3 and 53, respectively.

Copper-based multi-level metallization systems in today’s ultralarge-scale integrated electronic cir cuits require the fabrication of diffusion barriers and conductive seed layers for the electrochem ical metal deposition. Such films of only several nanometers in thickness have to be deposited void-free and conformal in patterned dielectrics. The envisaged further reduction of the geometric dimensions of the interconnect system calls for coating techniques that circumvent the drawbacks of the well-established physical vapor deposition. The atomic layer deposition method (ALD) allows depositing films on the nanometer scale conformally both on three-dimensional objects as well as on large-area substrates. The present work therefore is concerned with the develop ment of an ALD process to grow copper oxide films based on the metal-organic precursor bis(tri n-butylphosphane)copper(I)acetylacetonate [(n Bu3P)2Cu(acac)]. This liquid, non-fluorinated β diketonate is brought to react with a mixture of water vapor and oxygen at temperatures from 100 to 160°C. Typical ALD-like growth behavior arises between 100 and 130°C, depending on the respective substrate used. On tantalum nitride and silicon dioxide substrates, smooth films and self saturating film growth, typical for ALD, are obtained. On ruthenium substrates, positive deposition results are obtained as well. However, a considerable intermixing of the ALD copper oxide with the underlying films takes place. Tantalum substrates lead to a fast self-decomposition of the copper precursor. As a consequence, isolated nuclei or larger particles are always obtained together with continuous films. The copper oxide films grown by ALD can be reduced to copper by vapor-phase processes. If formic acid is used as the reducing agent, these processes can already be carried out at similar temperatures as the ALD, so that agglomeration of the films is largely avoided. Also for an integration with subsequent electrochemical copper deposition, the combination of ALD copper and ruthenium proves advantageous, especially with respect to the quality of the electroplated films and their filling behavior in interconnect structures. Furthermore, the ALD process developed also bears potential for an integration with carbon nanotubes.

A series of heavy alkaline earth complexes containing TpR2 -based ligands has been synthesized, structurally characterized, and their properties investigated. Salt metathesis routes involving MI2 (M = Ca, Sr, Ba) with either KTpEt2 or KTpnPr2 afforded MTpEt2 2 or MTpnPr2 2 in good to moderate yields. All of these complexes are volatile and exceptionally thermally stable, and have acceptable properties for use as group 2 atomic layer deposition (ALD) precursors. In addition, a series of group 2 complexes containing BpR2 -based ligand systems were synthesized. Treatment of MI2 (M = Ca, Sr, Ba) with two equivalents of TlBptBu2, KBp, and KBpiPr2 in tetrahydrofuran at ambient temperature afforded MBptBu2 2(THF)n, MBp2(THF)n, and MBpiPr2 2(THF)2, respectively. In the MBptBu2 complexes, the BptBu2 ligands undergo deformation in order to avoid intraligand and interligand tert-butyl group steric repulsions. Some of the BptBu2 complexes have enough volatility and thermal stability to be used as ALD or CVD precursors; however, the Bp and BpiPr2 complexes decompose prior to volatilization. The thermolysis of three equivalents of CaBp2(THF)2 at 190-200 °C/0.05 Torr afforded one equivalent of CaTp2 and two equivalents of TpCaBH4 in good yields. Treatment of TpCaBH4 with methanol and ethanol, followed by the appropriate work up procedure, led to the isolation of [(TpCa(MeOH)2)2(μ-B(OMe)4)][B(OMe)4] and [(TpCaB(OEt)3O)3B]∙EtOH, respectively. The ALD film growth of MB2O4 (M = Ca, Sr, Ba) has been demonstrated using CaTp2, SrTp2, or BaTpEt2 2 and water. For the ALD growth of BaB2O4, an ALD window was observed from 250-375 °C with a growth rate of 0.23 Å/cycle. In the ALD growth of CaB2O4 and SrB2O4, an ALD window was observed from 300-375 °C, respectively, with a consistent growth rate of approximately 0.36 and 0.47 Å/cycle, respectively. In the temperature versus growth rate plots for CaTp2, SrTp2, and BaTpEt2 , precursor saturation occurred at ≥ 3.0, ≥ 3.0, and ≥ 1.0 s, respectively. Elastic recoil detection analysis (ERDA) of representative samples prepared within the ALD windows indicated that the films were consistent with the composition of MB2O4, which demonstrates that TpR2 -based ligands are able to act as a single source precursor for metal and boron, and confer a precise 2:1 boron/metal ratio.

This thesis comprises four intercomplementary parts that introduce new approaches to brittle reaction layers and mechanical compatibility of metalloceramic joints created when fusing dental ceramics to titanium. The first part investigates the effects of TiO2 layer structure and thickness on the joint strength of the titanium-metalloceramic system. Three groups of standard metalloceramic samples with different TiO2 layer thickness and crystal structure were tested. The TiO2 layers were produced using atomic layer deposition (ALD). Scanning acoustic microscopy (SAM), three-point bending (TPB), cross-section microscopy, scanning electron microscopy (SEM), and energy dispersive X-ray spectroscopy (EDS) were employed. Samples with all TiO2 thicknesses displayed good ceramics adhesion to Ti, and uniform TPB results. The fracture mode was independent of oxide layer thickness and structure. Cracking occurred deeper inside titanium, in the oxygen-rich Ti[O]x solid solution surface layer. During dental ceramics firing TiO2 layers dissociate and joints become brittle with increased dissolution of oxygen into metallic Ti and consequent reduction in the metal plasticity. To accomplish an ideal metalloceramic joint this needs to be resolved. The second part introduces photoinduced superhydrophilicity of TiO2.  Test samples with ALD deposited anatase TiO2 films were produced. Band gap energy (EBG) for the TiO2 layers was estimated from transmittance measurements. Samples were irradiated with UV light (> EBG) to induce superhydrophilicity of the surfaces through a cascade leading to increased amount of surface hydroxyl groups. Samples were divided into two groups D and E to study the required irradiation time. Hydrophilicity of the TiO2 surfaces was assessed by sessile drop contact angle measurements. The reference contact angle prior to UV radiation was ~55˚. Superhydrophilicity (contact angle ~0˚) was achieved within 2 minutes of UV radiation. After initial partial recovery during the first 10 minutes, the contact angle remained below 20˚ for 1h. Total recovery was not observed within 24h storage. Photoinduced ultrahydrophilicity can be used to enhance wettability of titanium surfaces, an important factor in dental ceramics veneering processes. The third part addresses interlayers designed to restrain oxygen dissolution into Ti during dental ceramics fusing. The main requirements for an ideal interlayer material are proposed. Based on these criteria and systematic exclusion of possible interlayer materials silver (Ag) interlayers were chosen. Six groups of standard metalloceramic samples were studied. Groups F, J, K and L were Al2O3-blasted, Groups G and H were left polished. Thin silver interlayers were produced on Groups G, H, J and K by using the DIARC®  plasma coating method and thicker interlayers on Group J by electrochemical baths. Analysis methods were as in the first part. Good ceramics adhesion to titanium was observed in all test groups save for G and H, which both exhibited several areas of poor contact. SEM/EDS analyses revealed attachment of alumina particles on the Al2O3-blasted titanium. Ag covered this contamination in Group L. TPB results were significantly better in Group L samples compared to Group F. Generally, cracking occurred inside titanium in oxygen-rich Ti[O]x solid solution (F, G, H, J, K), locally also between Ti and dental ceramics (K, L). In Group L multiple cracks occurred inside dental ceramics, none inside Ti structure. Ag interlayers of 5 μm on Al2O3-blasted samples can be efficiently used to retard formation of the brittle oxygen-rich Ti[O]x layer, thus enhancing metalloceramic joint integrity. Based on the literature, isolation of alumina blasting particle contamination was also considered beneficial. The most brittle component in metalloceramic joints with 5 μm Ag interlayers was bulk dental ceramics instead of Ti[O]x. The fourth part investigates the importance of mechanical interlocking and presents a new approach to overcome mechanical problems of brittle reaction layers. Mechanically polished, Al2O3-blasted, and photolithographically etched standard metalloceramic samples, Groups M, N and P, showed no significant TPB test differences. Cracking occurred through Ti[O]x, but in photolithographically etched samples also locally through dental ceramics. Hence, the significance of mechanical interlocking achieved by conventional surface treatments can be questioned as long as the formation of the brittle layers (mainly oxygen-rich Ti[O]x) cannot be sufficiently controlled. Photolithographically etched pits can be used to cause cracking of dental ceramics instead of the more brittle reaction layers. The current depth and steepness of the pits, however, were insufficient for extensive stress redistribution. In summary – in contrast to former impressions of thick titanium oxide layers – this thesis clearly demonstrates diffusion of oxygen from sintering atmosphere and SiO2 to Ti structures during dental ceramics firing and the following formation of brittle Ti[O]x solid solution as the most important factors predisposing joints between Ti and SiO2-based dental ceramics to low strength. This among other predisposing factors such as residual stresses created by the coefficient of thermal expansion mismatch between dental ceramics and Ti frameworks can be avoided with Ag interlayers. 

Pent, E. 2010. Nano-structured 3D Electrodes for Li-ion Micro-batteries. Acts Universitatis Upsaliensis. Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 719. 119 pp. Uppsala. ISBN 978-91-554-7732-5. A new challenging application for Li-ion battery has arisen from the rapid development of micro-electronics. Powering Micro-ElectroMechanical Systems (MEMS) such as autonomous smart-dust nodes using conventional Li-ion batteries is not possible. It is not only new batteries based on new materials but there is also a need of modifying the actual battery design. In this context, the conception of 3D nano-architectured Li-ion batteries is explored. There are several micro-battery concepts that are studied; however in this thesis, the focus is concentrated on one particular architecture that can be described as the successive deposition of battery components (active material, electrolyte, active material) on free-standing arrays of nano-sized columns of a current collector. After a brief introduction about Li-ion batteries and 3D micro-batteries, the electrodeposition of Al through an alumina template using an ionic liquid electrolyte to form free-standing columns of Al current collector is described. The crucial deposition parameters influencing the nucleation and growth of the Al nano-rods are discussed. The deposition of active electrode material on the nano-structured current collector columns is described for 2 distinct active materials deposited using different techniques. Deposition of TiCII using Atomic Layer Deposition (ALD) as active material on top of the nano-structured Al is also presented. The obtained deposits present high uniformity and high covering of the specific surface of the current collector. When cycled versus lithium and compared to planar electrodes, an increase of the capacity was proven to be directly proportional to the specific area gained from shifting from a 2D to a 3D construction. Cu2Sb 3D electrodes were prepared by the electrodeposition of Sb onto a nano-structured Cu current collector followed by an annealing step forcing the alloying between the current collector and Sb. The volume expansion observed during Sb alloying with Li is buffered by the Cu matrix and thus the electrode stability is greatly enhanced (from only 20 cycles to more than 120 cycles). Finally, the deposition of a hybrid polymer electrolyte onto the developed 3D electrodes is presented. Even though the deposition is not conformal and that issues of capacity fading need to be addressed, preliminary results attest that it is possible to cycle the obtained 3D electrode-electrolyte versus lithium without the appearance of short-circuits.

SURI, RAHUL. Investigation of MOS Interfaces with Atomic-Layer-Deposited High-k Gate Dielectrics on III-V Semiconductors. (Under the direction of Dr. Veena Misra). The purpose of this research work was to investigate the surface passivation methods and metal gate/high-k dielectric gate stacks for metal-oxide-semiconductor devices (MOS) on III-V compound semiconductor materials – (i) GaAs for future high-speed low-power logic devices and (ii) AlGaN/GaN heterostructure for future high-speed high-power devices. GaAs is a candidate material for high-mobility channel in a NMOS transistor to extend the CMOS scaling up to and beyond the 16-nm technology node. AlGaN/GaN heterostructure is useful in a MOS-high electron mobility transistor (MOS-HEMT) device for providing a high current-carrying two dimensional electron gas (2DEG) channel. The interaction of GaAs surface with atomic layer deposition of high-k dielectrics was investigated to gain fundamental insights into the chemical properties of GaAs surface oxides and high-k/GaAs interface. Electrical characterization of devices was performed to understand the impact of high-k/GaAs interface on MOS device characteristics in order to form a suitable metal/high-k/GaAs gatestack for future high-speed logic and power devices. Reduction of native oxides on GaAs was found to occur during atomic layer deposition (ALD) of high-k dielectrics- HfO2 and Al2O3/HfO2 nanolaminates on GaAs. Reaction between ALD metal precursor and native oxides on GaAs was identified to be the cause for consumption of native oxides. It was established that the ALD growth temperature has a strong impact on this phenomenon. During post-dielectric annealing the residual arsenic oxides at the interface decomposed leading to an increase in the interfacial gallium oxides. Presence of gallium oxide, Ga2O3 was identified as a cause for observed frequency dispersion in MOS capacitance-voltage curves indicative of a high interface state density. The chemical properties of the AlGaN/GaN heterostructure surface prepared by wet chemical treatment using HCl/HF and NH4OH solutions were investigated and compared. Both HCl and NH4OH solutions were effective in etching the native oxide layer and reducing the surface carbon content; HCl treatment being slightly more effective. Atomic layer deposition of Al2O3 on AlGaN/GaN surface revealed a reduction of surface gallium oxides due to the reaction between metal precursor and Ga2O3. This oxide reduction provides an in situ ALD surface cleaning action and provides a passivation effect useful for suppressing surface states. The interface and electrical properties of Al2O3 and SiO2 grown by ALD on HCl-treated AlGaN/GaN surface were investigated. An upward band bending in the semiconductor was observed; Al2O3 resulting in a greater band bending at the interface than SiO2. SiO2 based device yielded a more positive threshold voltage than Al2O3 suggesting the potential use of a thin SiO2 interface passivation layer to achieve enhancement mode operation. Energy band alignment of ALD dielectrics- SiO2, HfO2, HfAlO and Al2O3 on GaN was determined using x-ray photoelectron spectroscopy. Fundamental chemical properties of the AlGaN/GaN heterostructure surfaces, interaction of the AlGaN/GaN surfaces with atomic layer deposition of dielectrics and electrical properties of AlGaN/GaN based MOS devices were studied and are the key to improving the device performance of MOS-HEMT transistors for high-power applications.

Consistent downsizing of microelectronics due to Moore's law scaling of integrated circuits has pushed traditional design techniques to their micro-scale limitations. When approaching the nano-scale, the electrical properties of the widely used poly-silicon based metal-oxide-semiconductor field-effect transistor (MOSFET) change drastically and render the devices un-useable. For this reason, the development of new materials that maintain favourable electrical properties on a nano-scale, such as titanium, aluminum, and zirconium based metals, metal-nitrides and metal-oxides, have been of growing interest in research. Future circuit design will require an efficient means for controlled and uniform coating of ultra-thin films of next-generation microelectronic materials. Two interesting candidates for depositing thin films are atomic layer deposition (ALD) and chemical vapour deposition (CVD). The ability to control the uniformity of thin films using ALD and CVD depends on locating effective precursors for deposition. An effective precursor should be thermally stable, volatile, chemically reactive and self-limiting. Compounds with guanidinate and amidinate ligands make promising precursors due to their facile tunability, their volatility and their self-limiting properties. In this work amidinate and guanidinate precursors for ALD and CVD are developed and characterized. This work includes the synthesis of a novel iminopyrrolidine ligand and characterization to determine its potential as an ALD precursor ligand. The ligand showed thermal stability and a tunable melting point trend demonstrating potential as an effective ligand for ALD and CVD precursors. In addition, once the iminopyrrolidinate ligand was reacted with aluminum and titanium species, tunable melting points were observed for the resulting metal containing precursor species, offering potential flexibility in ALD process design. The synthesis and thermal chemistry of other novel heteroleptic titanium and zirconium species are also presented. The most promising precursors include the heteroleptic titanium +3 guanidinate and amidinate species for the deposition of TÌN/TÌO2 films. One of the guanidinate species was chosen for an exposure experiment on high surface area silica to study and determine the precursor chemistry with respect to nucleation between titanium and the substrate. The novel heteroleptic zirconium species presented in this work demonstrated two different bonding arrangements within the guanidinate family.

The present thesis describes atomic layer deposition (ALD) of ternary rare earth (RE) oxides and characterization of compositional, structural and electrical properties of the films. The REScO3, LaLuO3 and ErxGa2-xO3 thin films investigated are potential high-κ materials for future metal-oxide-semiconductor field-effect transistors, i.e. MOSFETs. The dissertation consists of five peer reviewed publications. As a background for the work, issues related to the miniaturization of MOSFETs and the feasibility of rare earth oxides as new high-κ dielectrics are discussed. Also some challenges of manufacturing Ga-based MOSFETs with high quality gate oxide having satisfactory interface properties and the role of rare earth oxides in GaAs passivation are presented. In addition the basic principle of the ALD method is briefly introduced and recent literature concerning deposition of rare earth oxides is reviewed. A series of REScO3 thin films was deposited by ALD using rare earth β-diketonate precursors RE(thd)3 together with ozone. The films were characterized for growth rate, elemental composition, crystallization upon annealing and electrical properties. Amorphous films of high quality with low impurity contents and promising electrical characteristics were produced. Several gradually evolving properties of the films were examined and the effect of the RE3+ cation size was discussed. YScO3 films were also deposited using novel cyclopentadienyl metal precursors and water. Deposition of LaLuO3 films having similar properties but even higher dielectric constant (κ ≈ 30) than ternary scandates was examined at two different temperatures. The relationship between the crystallization behaviour and the dielectric constant of REScO3 and LaLuO3 thin films was discussed. Finally deposition of a possible gate oxide for GaAs MOSFETs, viz. ErxGa2-xO3 by two different precursor approaches was investigated. In addition to β-diketonate metal precursors novel cyclopentadienyl and amidinate metal precursors together with water as oxygen source were utilized. For both YScO3 and ErxGa2-xO3 films the choice of precursor system affected e.g. the electrical properties and the crystallization behavior.

The continued pursuit of smaller device dimensions by the semiconductor industry has led to an increased interest in functional organic films. Organics have great potential as advanced materials, owing to the versatility of organic moieties and vast knowledge base of organic reactivity. In order to implement organic films into semiconductor devices, the inorganic/organic hybrid interfaces must be investigated, so that the reactivity at these pivotal features is well-known. In this work organic films are studied in two environments: the Ge(100)-2xl surface and the SiC>2 surface. The reconstructed Ge(100)-2xl surface offers a well-defined substrate, ideal for fundamental reactivity studies. Organic reactants are deposited under ultrahigh vacuum conditions, allowing reactions between gas-phase organic molecules and the surface to be isolated and analyzed by in situ spectroscopic techniques. By use of infrared (IR) spectroscopy, x-ray photoelectron spectroscopy (XPS) and density functional theory (DFT) modeling, we investigate the reactivity of phenyl and tert-butyl isocyanate and isothiocyanante at the Ge(100)-2xl surface. The isocyanate and isothiocyanate moieties are both highly reactive groups consisting of a cumulated double bond containing two heteroatoms, allowing for many potential products with the Ge surface. We find that dative bonding through the heteroatoms plays an important role in the surface reactions, functioning as either reaction intermediates or final products depending on the adsorbate. Various cycloaddition products are also observed at the surface, with prominent reactivity trends resulting from the differences in oxygen and sulfur reactivity. In order to study the practical implementation of organic films, molecular layer deposition (MLD) reactions are studied on the hydroxlyated Si02 surface. MLD is a layer-by-layer technique, where films are deposited one molecular unit at a time, allowing for film tailorability and composition control down to single molecule specificity. Coupling reactions between isocyanate and isothiocyanate moieties with amines to form polyurea and polythiourea films, respectively, are studied. The MLD films are analyzed by ex situ ellipsometry, IR spectroscopy, XPS, and DFT. A constant growth rate and monomer dose saturation is observed for both the urea and thiourea coupling chemistry, and chemical composition of the films agrees well with theoretical models. The ability to precisely control film composition is demonstrated through the deposition of polyurea blends, with a homogenous composition throughout the film, and polyurea laminates, with layers of distinct composition within the film. Organic films have shown promise as copper diffusion barrier layers, and the effectiveness of the MLD films as copper barrier layers is investigated through thermal stability, adhesion, and copper penetration measurements. The films demonstrate potential in this application, but further modification of the films is necessary to meet all the requirements necessary for barrier layer implementation. The thesis concludes with some perspectives on the future of organic films on semiconductors.

In the last decades several new nanostructures of carbon have been discovered, including carbon nanotubes (CNTs), and the recently discovered 2-dimensional graphene. These new materials exhibit extraordinary and unique properties—making them extremely interesting both for fundamental science and for future applications. It is, however, of crucial importance to develop new and improved fabrication and processing methods for these carbon nanomaterials. In this thesis the concept of applying solution chemistry and solution-based techniques to fabricate and to deposit graphene and other carbon nanomaterials is explored. An area-selective deposition method was developed for CNT and carbon-coated iron nanoparticles. By utilizing organic functionalization the properties of the nanomaterials were tuned, with the purpose to make them soluble in a liquid solvent and also enable them to selectively adsorb to non-polar surfaces. The first step of the functionalization process was an acid treatment, to introduce defects in the materials. This method was also used to create defects in so-called carbon nanosheets (CNS). The effect of the defect formation on the electric properties of the graphene-like CNS was studied; it was found that the resistance of the CNS could be reduced to 1/50 by acid treating of the sample. Also, the effect of the created defects on gas adsorption to the surface of the CNS has been investigated. This was done using atomic layer deposition (ALD) of TiO2 on the CNS, and a clear change in nucleation be-havior was seen due to the acid-treatment. Furthermore, a solution-based new method for fabrication of graphene was developed; this method combines intercalation of bromine into graphite with ultrasonic treatment to exfoliate flakes into a solvent. From the solvent the flakes can be deposited onto an arbitrary substrate. Several important parameters in the method were investigated in order to optimize the process. One important parameter proved to be the choice of solvent in all steps of the procedure; it was shown to influence sonication yield, flake size, and deposition results. Toluene was identified as a suitable solvent. A mild heat-treatment of the starting material was also identified as a way to increase the exfoliation yield. Using this method, fabrication of few-layer graphene sheets was achieved and areas down to 3 layers in thickness were identified—this is in the very forefront of current solution-based graphene fabrication techniques. 

In science and technology there is a steadily increased demand of new materials and new materials production processes since they create new application areas as well as improved production technology and economy. This thesis includes development and studies of a chemical vapour deposition (CVD) process for growth of thin films of the metastable material copper nitride, Cu3N, which is a semiconductor and decomposes at around 300 °C. The combination of these properties opens for a variety of applications ranging from solar cells to sensor and information technology. The CVD process developed is based on a metal-organic compound copper hexafluoroacetylacetonate, Cu(hfac)2, ammonia and water and was working at about 300 °C and 5 Ton. It was found that a small amount of water in the vapour increased the growth rate considerably and that the phase content, film texture, chemical composition and morphology were strongly dependent on the deposition conditions. In-situ oxygen doping during the CVD of Cu3N to an amount of 9 atomic % could also be accomplished by increasing the water concentration in the vapour. Oxygen doping increases the band gap of the material as well as the electrical resistivity and changes the stability. The crystal structure of Cu3N is very open and contains several sites which can be used for doping. Different spectroscopic techniques like X-ray photoelectron spectroscopy, Raman spectroscopy and near edge X-ray absorption fine structure spectroscopy were used to identify the oxygen doping site(s) in Cu3N. Besides the properties, the oxygen doping also affected the morphology and texture of the films. By combining thin layers of different materials several properties can be optimized at the same time. It has been demonstrated in this thesis that multilayers, composed of alternating Cu3N and Cu2O layers, i.e. a metastable and a stable material, could be grown by CVD technique. However, the stacking sequence affected the texture, morphology and chemical composition. The interfaces between the different layers were sharp and no signs of decomposition of the initially deposited metastable Cu3N layer could be detected. 
Keywords: Chemical vapour deposition, copper hexafluoroacetylacetonate, copper (I) nitride, copper (I) oxide, multilayers, oxygen doping
 

CdS is conventionally used as a buffer layer in Cu(In,Ga)Se2, CIGS, solar cells. The aim of this thesis is to substitute CdS with cadmium-free, more transparent and environmentally benign alternative buffer layers and to analyze how the material properties of alternative layers affect the solar cell performance. The alternative buffer layers have been deposited using Atomic Layer Deposition, ALD. A theoretical explanation for the success of CdS is that its conduction band, E., forms a small positive offset with that of CIGS. In one of the studies in this thesis the theory is tested experimentally by changing both the E0 position of the CIGS and of Zn(O,S) buffer layers through changing their gallium and sulfur contents respectively. Surprisingly, the top performing solar cells for all gallium contents have Zn(O,S) buffer layers with the same sulfur content and properties in spite of predicted unfavorable E. offsets. An explanation is proposed based on observed nonhomogenous composition in the buffer layer. This thesis also shows that the solar cell performance is strongly related to the resistivity of alternative buffer layers made of (Zn,Mg)O. A tentative explanation is that a high resistivity reduces the influence of shunt paths at the buffer layer/absorber interface. For devices in operation however, it seems beneficial to induce persistent photoconductivity, by light soaking, which can reduce the effective E. barrier at the interface and thereby improve the fill factor of the solar cells. Zn-Sn-0 is introduced as a new buffer layer in this thesis. The initial studies show that solar cells with Zn-Sn-0 buffer layers have comparable performance to the CdS reference devices. While an intrinsic ZnO layer is required for a high reproducibility and performance of solar cells with CdS buffer layers it is shown in this thesis that it can be thinned if Zn(O,S) or omitted if (Zn,Mg)O buffer layers are used instead. As a result, a top conversion efficiency of 18.1 % was achieved with an (Zn,Mg)O buffer layer, a record for a cadmium and sulfur free CIGS solar cell. Keywords: Cu(In, Ga)Se2, Solar cells, Thin film, Buffer layer, Window layer, ZnO, Zn(O, S), (Zn, Mg)O, Zn-Sn-0 

Recent advances in fiber processing technology have allowed the formation of fibers with sub-micron and nanoscale dimensions. Such fibers have received much recent attention due to their potential in wide ranging applications including but not limited to tissue scaffolds, affinity membranes, and other advanced filtration applications. This is due to their ability to provide high surface area, high porosity and good mechanical properties. However, in order to expand the range of applications and improve chemical and physical functionality, the ability to uniformly and predictably modify the surface of fibers in highly dense and tortuous nonwoven structures, in an environmentally friendly manor, is required. A variety of different techniques are currently used to modify the surface of planar polymers as well as nonwoven fiber structures. In this study, a vapor phase film deposition technique called atomic layer deposition (ALD) has been used to modify the surface of planar films of fiber forming polymers as well as nonwoven fiber structures. In-situ gravimetric and chemical analyses have been employed to investigate film precursor and polymer chemical and physical interaction, as well as the effect of process variables such as temperature on the nucleation and growth of ALD films on polymer materials. 

Well-controlled atomic layer deposition of high-quality thin films of ZrO2 and HfO2 can be realized using novel cyclopentadienyl-based precursors in normal laboratory conditions. Cyclopentadienyls of zirconium and hafnium, viz. (CpMe)2ZrMe. (CpMe) Zr(OME)Me, CpMe) Zr(NM22), (CpEt) Zr(NMe2), (CpMe) HfCOMeMe, and CpHf(NMe2)3 (Cp = CsHs, Me = CH3, and Et = C2H5) together with ozone as the oxygen source are thus appropriate precursors for ALD process, providing dense dielectric and insulating films on semiconductor or metallic substrates. Appreciably high conformality (step-coverage) on three-dimensional substrates can be achieved.
Both HfO, and Zro, films deposited at 300–350°C consisted predominantly of the monoclinic phase. Upon decreasing the film thickness down to 5–15 nm, the significance of metastable cubic or tetragonal phases increased, especially in Zro, films, which was promising in terms of higher dielectric permittivity and therefore increasing capacitance. The precursors containing two Cp-ligands and oxygen seemed to be more stable and yielded films with slightly higher phase homogeneity compared to precursors containing two Cp-ligands without methoxy-groups (OME). These films were also more insulating, although not necessarily superior in terms of electronic defect density. Leakage currents in HfO2 were lower than those in ZrO2, although the ZrO2 films could possess even lower interface trap densities than HfO.
Conduction mechanisms were quite alike in films grown from the different precursors, but the ultrathin films grown from (CpMe) Zr(OME) Me appeared more insulating possessing higher breakdown fields than those e.g. in the films grown from CpMe) ZrMe2. The dominant conduction mechanism was bulklimited field-assisted excitation of charge carriers, although at low voltages thermal excitation over interfacial barriers could also be taken into account.
Since the ability of metastable cubic/tetragonal phases of HfO2 and/or ZrO2 to withstand post-deposition annealing procedures without transformation to lower-permittivity monoclinic phase was a likely issue when fabricating highdensity capacitors, the films were further doped or mixed with rare earth metal oxides. In the case of HfO2, quite a recently developed monocyclopentadienylbased precursor CpHf(NMe2)3 was chosen and used in the experiments devoted to the studies on Yoz-doped HfO). In the case of Zro, a well-behaving compound (CpMe) Zr(OMe Me was chosen for the preparation of Zro, films doped or nanolaminated with Gd2O3 and Er2O3.
Doped HfO2:Y films were amorphous in as-deposited state but crystallized in the form of metastable polymorphs after heat-treatments above 500°C, possessing higher capacitance and lower equivalent oxide thickness compared to those of non-doped HfO2. The leakage currents remained significant, probably due to somewhat inhomogeneous crystallization. ZrO2:Gd and ZrO2:Er films were crystalline already in as-deposited state, and the cubic polymorph of ZrO, was retained upon annealing at 650°C in the doped films, whereas lower
permittivity monoclinic ZrO2 became apparent in the nondoped films already in the as-deposited state. The dielectric permittivity value of 31 was achieved in the ZrO2:Er2O3 films with an Er:Zr cation ratio of 0.09 and 30 in the ZrO2:Gd2O3 films with a Gd:Zr cation ratio of 0.027, whereas in non-doped films permittivity values above 20-25 could not be measured. Concerning the bottom electrode materials, the best results in terms of permittivity and leakage currents were achieved with Ru, allowing equivalent oxide thickness below 1 nm and a current density of 3x10 A/cmat 1 V. In general, at electric fields below 2-3 MV/cm, normal and trap-compensated Poole-Frenkel conduction mechanisms were competing, whereas at higher fields, Fowler-Nordheim and/or trap-assisted tunneling were to be considered. In Er2O3-ZrO2 and Gd2O3ZIO, nanolaminates the cubic ZrO, and rare earth oxide phases dominated, but the capacitance increased after annealing in the films with relatively low rare earth metal content and decreased in the case of higher rare earth content, being indicative of the sensitivity of dielectric behavior on the contribution of ZrO2 phases.
The processes and resulting films examined within the present study may well become considered as those relevant to the fabrication of high-performance capacitor dielectric materials. Further studies might become concentrated on the optimization of the dopant content and deposition (cycle) time parameters, in order to improve the structural stability, capacitance density, pre-breakdown leakage currents and conformal growth over 3D substrates. In addition, the search for even better cyclopentadienyl-based precursors for both host and dopant materials may continue.

The light emitted by flat panel displays (FPD) can be generated in many different ways,  such as for example alternating current thin film electroluminescence (ACTFEL), liquid  crystal display (LCD), light emitting diode (LED), or plasma display panel (PDP)  technologies. In this work, the focus was on ACTFEL devices and the goal was to  develop new thin film processes for light emitting materials in ACTFEL devices. The  films were deposited with the atomic layer deposition (ALD) method, which has been  utilized in the manufacturing of ACTFEL displays since the mid-1980s. The ALD  method is based on surface-controlled self-terminated reactions and a maximum of one  layer of the desired material can be prepared during one deposition cycle. Therefore, the  film thickness can be controlled simply by adjusting the number of deposition cycles. In  addition, both large areas and deep trench structures can be covered uniformly.  During this work, new ALD processes were developed for the following thin film  materials: BaS, CuxS, MnS, PbS, SrS, SrSe, SrTe, SrS1-xSex, ZnS, and ZnS1-xSex. In  addition, several ACTFEL devices were prepared where the light emitting material was  BaS, SrS, SrS1-xSex, ZnS, or ZnS1-xSex thin film that was doped with Ce, Cu, Eu, Mn, or  Pb. The sulfoselenide films were made by substituting the elemental selenium for sulfur  on the substrate surface during film deposition. In this way, it was possible to replace a  maximum of 90% of the sulfur with selenium, and the XRD analyses indicated that the  films were solid solutions. The polycrystalline BaS, SrS, and ZnS thin films were  deposited at 180-400, 120-460, and 280-500 °C, respectively, and the processes had a  wide temperature range where the growth rate of the films was independent of the  deposition temperature. The electroluminescence studies showed that the doped  sulfoselenide films resulted in low emission intensity. However, the emission intensities  and emission colors of the doped SrS, BaS, and ZnS films were comparable with those  found in earlier studies. It was also shown that the electro-optical properties of the  different ZnS:Mn devices were different as a consequence of different ZnS:Mn  processes. Finally, it was concluded that because the higher deposition temperature  seemed to result in a higher emission intensity, the thermal stability of the reactants has  a significant role when the light emitting materials of ACTFEL devices are deposited  with the ALD method.  

Photocatalytic TiO2 thin films can be highly useful in many environments and applications. They can be used as self-cleaning coatings on top of glass, tiles and steel to reduce the amount of fouling on these surfaces. Photocatalytic TiO2 surfaces have antimicrobial properties making them potentially useful in hospitals, bathrooms and many other places where microbes may cause problems. TiO2 photocatalysts can also be used to clean contaminated water and air. Photocatalytic oxidation and reduction reactions proceed on TiO2 surfaces under irradiation of UV light meaning that sunlight and even normal indoor lighting can be utilized. In order to improve the photocatalytic properties of TiO2 materials even further, various modification methods have been explored. Doping with elements such as nitrogen, sulfur and fluorine, and preparation of different kinds of composites are typical approaches that have been employed. Photocatalytic TiO2 nanotubes and other nanostructures are gaining interest as well. Atomic Layer Deposition (ALD) is a chemical gas phase thin film deposition method with strong roots in Finland. This unique modification of the common Chemical Vapor Deposition (CVD) method is based on alternate supply of precursor vapors to the substrate which forces the film growth reactions to proceed only on the surface in a highly controlled manner. ALD gives easy and accurate film thickness control, excellent large area uniformity and unparalleled conformality on complex shaped substrates. These characteristics have recently led to several breakthroughs in microelectronics, nanotechnology and many other areas. In this work, the utilization of ALD to prepare photocatalytic TiO2 thin films was studied in detail. Undoped as well as nitrogen, sulfur and fluorine doped TiO2 thin films were prepared and thoroughly characterized. ALD prepared undoped TiO2 films were shown to exhibit good photocatalytic activities. Of the studied dopants, sulfur and fluorine were identified as much better choices than nitrogen. Nanostructured TiO2 photocatalysts were prepared through template directed deposition on various complex shaped substrates by exploiting the good qualities of ALD. A clear enhancement in the photocatalytic activity was achieved with these nanostructures. Several new ALD processes were also developed in this work. TiO2 processes based on two new titanium precursors, Ti(OMe)4 and TiF4, were shown to exhibit saturative ALDtype of growth when water was used as the other precursor. In addition, TiS2 thin films were prepared for the first time by ALD using TiCl4 and H2S as precursors. Ti1-xNbxOy and Ti1-xTaxOy transparent conducting oxide films were prepared successfully by ALD and post-deposition annealing. Highly unusual, explosive crystallization behaviour occurred in these mixed oxides which resulted in anatase crystals with lateral dimensions over 1000 times the film thickness. 

The ALD method was used to obtain semiconductor and conductive ZnO layers as well as ZnO: Al layers for organic electronics. Technological parameters were optimized in order to obtain layers ZnO semiconductors with a low concentration of free electrons (n ≤ 1017 cm-3) and deposited at low growth temperatures (Tgr ≤ 100 ° C). This made it possible to use these layers in the organic photovoltaic structure as pn junction element with p-type organic material - nickel phthalocyanine (NiPc). The obtained ITO / NiPc / ZnO / Al structure was characterized by better properties photovoltaics compared to the identical structure but without the ZnO layer, ie ITO / NiPc / Al. Moreover, the low growth temperature allowed the layer to be deposited ZnO on the NiPc layer, thus protecting it against contact with air. This significantly improved the time stability of such a structure compared to the structure without ZnO layers. Optimization of technological parameters allowing to obtain undoped ZnO layers having the lowest possible appropriate resistances while maintaining high optical transmission in the range visible. One of the main parameters was the growth temperature. Obtained values of ρ = 1.7 × 10-3 Ωcm and T ≈ 90% (at Tgr = 200 ° C) are among the best parameters reported in the literature for transparent, conductive and undoped ZnO layers obtained by various methods. The resulting conductive undoped ZnO layers were used as transparent electrode in OLED diode with Alq3 active layer (ZnO / CuI / Alq3 / PEGDE / Al). The obtained IV and LV characteristics, including high luminance ~ 3 × 103 cd / m2 and the luminous efficiency of 3 cd / A are comparable and even better compared to similar structures with an Alq3 layer, but with a ZnO: Al layer or ITO used as electrodes. To further improve conductivity, the ZnO layers were doped with aluminum. ZnO: Al layers having the lowest resistances were obtained at a temperature of 200 ° C equal to 8.2 × 10-4 Ωcm (for thicknesses of 200 nm) and 7.1 × 10-4 Ωcm (for thicknesses ≈1 µm) and high optical transmission (T ≈ 90%) in the visible range. The obtained layers are characterized by a homogeneous distribution of aluminum for Al content ≥ 2% at. Periodicity was observed at lower contents Al decomposition disappears after heating in nitrogen at 300 ° C. Based on the research techniques used (SEM, SIMS, XRD), no the presence of foreign Al phases was observed in the obtained ZnO: Al layers, even at high Al contents of 7.8% at. ρ values are among the lowest reported so far in the literature for ZnO: Al layers obtained by the ALD method. However, compared to the lowest reported resistance of ZnO: Al layers obtained by sputtering magnetron and the PLD method ( ρ = 2 × 10-4 Ωcm) they are higher, but the same row. ZnO: Al layers were obtained on a flexible polymer - PET substrate, for use as transparent electrodes in organic and flexible instruments electronic. The growth temperature was 110 ° C. Parameters obtained electrical components are comparable to those reported in the literature for ZnO: Al / PET layers obtained by magnetron sputtering. In the literature known to the author So far, there are no studies devoted to the characterization of the obtained ZnO: Al layers by the ALD method on a flexible substrate. The influence of native defects and unintentional admixtures was examined on the obtained electrical parameters of ZnO and ZnO: Al layers obtained by the ALD method. It was found that in the non-doped aluminum layers a fundamental role native defects play. Among them, the predominant influence on high conductivity most likely has interstitial zinc, not oxygen gaps. It means bigger stability of ZnO layers caused, among others, by greater resistance to possible diffusion oxygen from the air. It turned out that the presence of hydrogen does not play a significant role in increasing the conductivity ZnO layers obtained by the ALD method. In the case of undoped layers aluminum, a decrease in the concentration of n is observed with an increase in concentration hydrogen. Comparing the results of the cross-sectional tests of the ZnO and ZnO layers: Al with the profile of the hydrogen content in these layers it can be concluded that hydrogen it accumulates in greater amounts at the grain boundaries which may contribute to lower electron mobility in these layers. An additional argument this is due to the fact that both in the ZnO and ZnO layers: Al the increase in hydrogen concentration was always accompanied by a decrease in the mobility of µ. 

Growing environmental concerns related to the use of synthetic non-biodegradable polymers in the packaging industry have led to the need for new, especially bio based, materials. Currently, petroleum-based synthetic polymers are widely used due to their relatively low cost and high performance. Biodegradable plastics and fibre-based materials have been proposed as a solution to the waste problems related to these synthetic polymers. Fibre-based packaging materials have many advantages over their non-biodegradable competitors, such as stiffness vs. weight ratio and recyclability. However, poor barrier properties and sensitivity to moisture are the main challenges restricting their use. Application of a thin coating layer is one way to overcome these problems and to improve the barrier properties of such materials. Atomic layer deposition (ALD) is a well suited technique for depositing thin inorganic coatings onto temperature-sensitive materials such as polymer-coated boards and papers and polymer films. In the present work, thin and highly uniform Al2O3 coatings were deposited at relatively low temperatures of 80, 100 and 130 °C onto various bio-based polymeric materials employing the ALD technique. The study demonstrates that a 25-nm-thick ALD-grown Al2O3 coating significantly enhances the oxygen and water vapour barrier performance of these materials. Promising barrier properties were obtained with polylactide-coated board, hemicellulose-coated board as well as various biopolymer (polylactide, pectin and nanofibrillated cellulose) films after coating with a 25-nm-thick Al2O3 layer. Thin Al2O3 coatings can improve the properties of biopolymers, enabling the use of these renewable polymers in the production of high-performance materials for demanding food and pharmaceutical packaging applications. The future roll-to roll ALD technology for coating polymers with inorganic thin films will increase the industrial potential of these materials and could lead to further opportunities for their commercialization. 

A synthetic route for the general synthesis of Ce(IV) alkoxides is described utilising a reaction with ceric ammonium nitrate and fe/Abutoxide to create the in situ intermediate [Ce(OBut)4] by means of salt metathesis. A further alkoxy exchange reaction with the desired alkoxide resulted in the formation of alkoxide complexes [Ce(L)4] (L = mmp, dmae, dmap, dmop, dmomp). The compounds [Ce(mmp)4], [Ce(dmap)4] and [Ce(dmop)4] were characterised crystallographically. [Ce(dmae)4] was shown to decompose over time and gave a crystal structure of the cluster [Cei4(dmae)22(N03)20i4(0H)4]. DmomH was found to be an unsuitable ligand for the formation of a homoleptic complex. A range of heteroleptic complexes were also synthesised by reaction of 2 equivalents of an alkoxide (dmop, dmom, dmomp) or P-diketonate ligand (thd, dbm) with the complexes [Ce(mmp)4] and [Ce(dmap)4]. The thd ligand was found to be too strong a donor to make a heteroleptic complex resulting exclusively in the formation of [Ce(thd)4]. These reactions yielded the complexes [Ce(mmp)2(L)2] (L=dbm, dmom, dmomp) and [Ce(dmap)2(dbm)2]. The reaction between [Ce(mmp)4] and 2 equivalents ofdmop gave the dimer [Ce2(mmp)2(dmop)40] which was characterised crystallographically. [Ce(mmp)4] and [Ce(dmap)4] were tested for use as precursors for the liquid injection MOCVD and ALD of CeCh thin films on a Si(100) substrate. The results were compared with the known precursors [Ce(thd)4] and [Ce(OC(CH3)2Pr')4], [Ce(thd)4] being the most commonly used precursor for Ce02 thin film growth. Both new precursors were found to be more volatile and have a temperature window of 200-600 °C for MOCVD. [Ce(dmap)4] showed growth at temperatures as low as 100 °C in ALD though was susceptible to decomposition via CVD processes at 250 °C. [Ce(mmp)4] was self-limiting over the temperature range of 150-350 °C. A range of heteroleptic Ti(IV) alkoxide complexes were also synthesised by reaction of either [T^OBu^] or [Ti(OPrl)4] with 2 equivalents of the desired alkoxide in an alkoxy exchange reaction to give complexes with the formula [Ti(OR)2(L)2] (R = Bu1, Pr1, L - dmae, mmp, dmop, dmom). [Ti(OPr1)2(dmop)2] and [Ti(OBut)2(dmop)2] were characterised crystallographically. These compounds were used as precursors for the MOCVD of anatase Ti02 nanostructures on Si(100) and F-doped Sn02 substrates. [Ti(OPr1)2(dmae)2] provided the optimum conditions of growth on both substrates at 450 °C for 180 minutes giving full coverage and uniform nanostructures. The precursor also interestingly grew nanorods at 600 °C on Si(100)

The demand for novel functional materials is a never-ending challenge, as the development of many future applications depends on the new innovations made in the field of materials science. Ternary manganese oxides are a versatile group of materials with interesting magnetic properties and several potential applications, e.g. in microelectronics and solid oxide fuel cells. This thesis reports the preparation of several ternary manganese oxide materials through a selection of synthesis methods. Studies on the structural details, magnetic proper ties and oxygen stoichiometry of the bulk and thin-film samples are moreover included in this work. The thesis consists of four publications, discussed with relevant literature data. The synthesis of an entire series of the hexagonal RMnO3 system for R = Y, Ho-Lu, and the subsequent conversion of the hexagonal phases to orthorhombic perovskites through high-pressure treatment is realized in the present work. The synthesis methods of the hexagonal bulk samples reported here include the hydrothermal and the sol-gel methods. A systematic study of the structural evolution is presented for both polymorphs, and studies on the magnetic properties have been performed as well. High-quality thin-film samples of the hexagonal and orthorhombic RMnO3 families have been fabricated employing the atomic layer deposition (ALD) technique and post-deposition heat treatment. The formation of metastable orthorhombic RMnO3 perovskites of the small rare earths has been successfully realized even for the smallest R constituent, Lu, by depositions on coherent perovskite substrates with low lattice mismatch with the targeted structure. The challenging task of studying the magnetic properties of these thin-film samples is also approached in the present work. The Néel temperatures determined for the RMnO3 films featuring antiferromagnetic ordering are in good accordance with the corresponding results on powder samples, and the presence of cation vacancies is shown for the ALD-grown LaMnO3+δ sample. The reproducible fabrication of spinel-structured (Mn,Co)3O4 thin films by ALD has been achieved and reported. Precise control of the oxygen content of as-deposited MnCo2O4+δ films has been realized for the first time for an ALD-grown thin-film sample through post deposition heat treatments, as evidenced from the monotonous increases of both the unit-cell volume and the Curie temperature with increasing annealing temperature/decreasing oxygen partial pressure. The performance of ALD-grown MnCo2O4 protective coatings is also reported on ferritic stainless steel in solid oxide fuel cells with promising results.

Over the last four decades, the aggressive scaling of complementary metal oxide semiconductor field effect transistor (CMOSFET) devices has led to the limitations of conventionally used SiO2-based gate dielectric since its very thin thickness resulted in significant problems such as large leakage current. From this point of view, it is necessary to replace the SiO2 with high-k dielectrics enabling physically thicker gate oxides. Additionally, for the requirements of thickness controllability down to few nm range, large area uniformity and excellent conformality, high-k thin film should be grown by atomic layer deposition (ALD) method. Among several high-k dielectric materials satisfying the considerations for their introduction, lanthanum oxide (La2O3) and cerium oxide (CeO2) are very interesting and promising candidates. They themselves have received a great deal of attention as the high-k dielectric materials in dynamic random access memory (DRAM) capacitors as well as CMOS devices due to their superior properties such as the high dielectric constant, high dielectric breakdown strength and large band gap and conduction band offset with Si as well as thermodynamical stability in contact with Si. In addition, their particular applications into Hf-based dielectric can be the promising solutions for the challenging issues such as work function engineering and further equivalent oxide thickness (EOT) scaling required in the advanced gate stack technologies. Nevertheless, current researches on ALD of La2O3 and CeO2 are still in their infancy. Therefore, in this thesis, I will describe the overall studies on ALD of the La2O3 and CeO2 from their ALD processes and properties to potential applications into Hf-based dielectrics. In the first and second sections (chapter III and IV), I investigated thermal and plasma enhanced atomic layer deposition (T-ALD and PE-ALD, respectively) of La2O3 thin films by using tris(isopropyl-cyclopentadienyl)lanthanum [La(iPrCp)3] precursor with H2O and O2 plasma. The growth characteristics, film properties and electrical properties were discussed by several analysis techniques, mainly focusing on the comparative studies. It revealed that PE ALD La2O3 exhibited the higher growth rate, better film quality and superior electrical properties that those of T-ALD. Then, I investigated the flat band voltage (VFB) modulation by insertion of the PE-ALD La2O3 capping layer into HfO2 gate dielectrics for effective work function engineering. The location of La2O3 insertion layer in the HfO2 were precisely modulated at bottom, middle or top to clarify which location plays the dominant role for effective work function modulation by the interface dipole layer. Thereby, I proposed that the main mechanism of VFB modulation by La2O3 capping layer is dipole moment formation at the interfacial layer between high-k dielectric and Si substrate. In the following sections (chapter V and VI), I investigated PE-ALD of CeO2 thin films by using newly synthesized tris(isopropyl-cyclopentadienyl)cerium [Ce(iPrCp)3] precursor with O2 plasma for the first time. After the evaporation characteristics of the Ce(iPrCp)3 precursor by simultaneous thermogravimetric-differential thermal analysis (TG-DTA) measurements, the growth characteristics, film properties and electrical properties were evaluated by several analysis techniques. It suggested that PE-ALD CeO2 has great benefits as the high-k gate dielectric. Then, systematic investigations were conducted to comprehensively demonstrate the energy band diagram including the electron carrier transports for Al/PE-ALD CeO2/p-Si structure by combining ellipsometry and photoemission spectroscopy techniques with Fowler-Nordheim (F-N) tunneling and Pool-Frenkel (P-F) conduction. Thereby, I proposed that the increased interfacial layer and reduced trapped oxide densities result in the increased effective barrier height and decreased trap energy levels with increasing O2 annealing temperature. In the final section (chapter VII), T-ALD of CeO2 thin film was investigated by using tetrakis(1-methoxy-2-methyl-2-propanolate)cerium [Ce(mmp)4] with H2O. After establishing the T-ALD CeO2 process, the effects of Ce doping into HfO2 gate dielectric were systematically investigated for significant enhancement of dielectric constant. For the realization of CeO2 doping into HfO2, ALD supercycles process was carried out. The growth characteristics, film properties and electrical properties were evaluated by several analysis techniques with a variety of Ce/(Ce+Hf) compositions. Thereby, I proposed that the maximum dielectric constant value was found to be ~39 for the Ce0.11Hf0.89O2 film with dominant tetragonal phases. Therefore, my research in this thesis will not only provide many valuable information and technical methods on ALD La2O3 and CeO2 gate dielectrics and their applications but will also extend essential and fundamental degree of freedom for their practical implementation toward the advanced gate stack technologies in the future microelectronic industry 

Current industrial atomic layer deposition (ALD) processes are almost wholly confined to glass or silicon substrates. For many industrial applications, deposition on polymer substrates will be necessary. Current deposition processes are also typically carried out at temperatures which are too high for polymers. If deposition temperatures in ALD can be reduced to the level applicable for polymers, it will open new interesting areas and applications for polymeric materials. The properties of polymers can be improved for example by coatings with functional and protective properties. Although the ALD has shown its capability to operate at low temperatures suitable for polymer substrates, there are other issues related to process efficiency and characteristics of different polymers where new knowledge will assist in developing industrially conceivable ALD processes. Lower deposition temperature in ALD generally means longer process times to facilitate the self limiting film growth mode characteristic to ALD. To improve process efficiency more reactive precursors are introduced into the process. For example in ALD oxide processes these can be more reactive oxidizers, such as ozone and oxygen radicals, to substitute the more conventionally used water. Although replacing water in the low temperature ALD with ozone or plasma generated oxygen radicals will enable the process times to be shortened, they may have unwanted effects both on the film growth and structure, and in some cases can form detrimental process conditions for the polymer substrate. Plasma assistance is a very promising approach to improve the process efficiency. The actual design and placement of the plasma source will have an effect on film growth characteristics and film structure that may retard the process efficiency development. Due to the fact that the lifetime of the radicals is limited, it requires the placement of the plasma source near to the film growth region. Conversely this subjects the substrate to exposure by other plasma species and electromagnetic radiation which sets requirements for plasma conditions optimization. In this thesis ALD has been used to modify, activate and functionalize the polymer surfaces for further improvement of polymer performance subject to application. The issues in ALD on polymers, both in thermal and plasma-assisted ALD will be further discussed.

The aim of this work was to investigate the potential of model materials designed by atomic layer deposition toward applications in catalysis research. Molybdenum based catalysts promoted with cobalt were selected as target materials, considering their important roles in various industrial processes. Particular attention was paid to understand the growth dynamics of the ALD processes involved and further to characterize the obtained materials carefully. It was of main concern to verify the feasibility to coat porous materials by ALD with our equipment. Another ambition was to confirm the advantages of the atomic layer technique to create model materials for industrial research projects in catalysis. Thin film growth of molybdenum trioxide has been demonstrated by the atomic layer deposition technique using molybdenum-hexacarbonyl, water and ozone as precursors. A narrow ALD-window is observed at relatively low temperatures, leading to amorphous films as deposited. The effect of different oxygen precursors on the growth mechanism of the molybdenum oxide has been assessed by QCM investigations. The chemical composition and Mo-oxidation state in MoO3 thin films grown by ALD have been investigated by two XPS approaches. The sputtering based studies affects strongly the analysis results by inducing reduction of the Mo-O film prior to XPS data collection. The ARXPS proves that molybdenum is in oxidation state VI throughout the bulk of the film. Molybdenum in a lower oxidation state is observed at the substrate interface, representing the initial stage of film formation. A convenient process to achieve thin film model materials of MoO3 polymorphs has been proposed, describing the crystallization behavior of the thin films from the as-deposited amorphous state via metastable β-MoO3 to the orthorhombic α-MoO3 phase. A significant mass transport, in particular during recrystallization into α-MoO3 is demonstrated. By means of combined AFM/Raman studies we have been able to relate morphology and vibration mode of α-MoO3 features. The ability to coat porous materials with our ALD equipment has been confirmed by means of coating anodiscs with the current molybdenum process and a cobalt process. ALD thin film growth of cobalt oxide has been obtained using cobaltocene and ozone as precursors. SEM and EDS investigations of coated porous anodiscs show the specific coverage profile to be dependent on the fluid dynamics in the reactor. Cobalt molybdate has been grown by atomic layer deposition, varying the cobalt oxide precursor between Co(thd)2 and CoCp2. The growth dynamics of the films, their composition and crystallization have been examined as function of the subcycles ratio, proving CoMoO4 to be the preferential composition and the excess of molybdenum to crystallize into α-MoO3. The films catalytic activity in the ammonia decomposition process is assessed at the laboratory scale. The growth dynamics have been investigated using quartz crystal microbalance (QCM) where it is evident that the different precursor chemistries affect each other’s growth. When water is combined in the reactions, a surface controlled mechanism takes place which guides the deposited stoichiometry towards the CoMoO4 phase over a range of different cobalt rich pulsed compositions. This is a rare example of how surface chemistry can control stoichiometry of depositions in ALD. The catalytic properties of cobalt-molybdenum oxide thin films deposited by ALD on industrial alumina carriers have been studied as function of the thickness of the films. Cobalt molybdenum multilayered thin films activity has proven to increase with the thickness of the films up to a certain extend. Multilayered films show a better activity in the HDS conversion as compared with cobalt molybdate single phase films. Finally, TEM imaging characterization of copper particles on top of a zinc oxide film has been achieved by first depositing an underlayer of Al2O3, thereafter coated with zinc oxide and copper oxide thin films by ALD at low temperatures. In situ TEM imaging of the multilayered film at 250 °C under H2 shows crystallization of the ZnO grains and reduction of the copper oxide film leading to Cu particles formation.

In this work, a new fabrication process is investigated for ultralow power, microelectronic hotplates. These hotplates are based on a small surface (0.0012 -0.12 mm2 ) that is heated by a heater to temperatures in the range of 300-400 °C. These hotplates can be used for instance as gas or flow sensors. Applied as gas sensor, an (exothermic) combustion reaction will take place at the (catalytic) surface in the presence of a flammable gas during which heat is generated. This reaction heat is detected by a change in the (temperature dependent) heater resistance. This means that in a hotplate the heater simultaneously acts as a heat source and thermal sensor. The research, discussed in this work, is part of the ‘Hot Silicon’ project in which ultralow power (10-6 W) hotplates are studied; the power consumption is a factor 1000 less than state-of-the-art ‘low power’ heaters. The ultralow power enables integration of hotplates as for instance gas sensors in portable (battery powered) systems for industrial or domestic applications, for instance for the detection of hazardous (harmful and/or flammable) gases. A combination of various sensors (in a ‘sensor array’ for instance) can be used for the detection of multiple gases. Furthermore, low power gas sensors are generally safer. In this work, the hotplate is based on a small high ohmic conducting cylinder (the ‘link’). The link is embedded in an insulating (glass) layer and positioned between two crossing electrodes. Fabrication of the link is based on etching a hole in the glass layer and coating it with an ultrathin (7-15 nm) titanium nitride (TiN). Due to the excellent step coverage of the ALD process, a perfectly hollow and conducting cylinder is created. In the first part of this work, the material properties of ALD TiN thin films are studied. In the second part, the realized hotplates are discussed, as well as the integration of ALD TiN in the fabrication process. In chapter 2, the resistivity (ρ) and the temperature dependence of the resistance (i.e. the temperature coefficient of resistance, TCR) of ultrathin ALD TiN films are measured using special test structures. The values of ρ and TiN are important parameters for the sensor design, as they can be used for an accurate temperature measurement of the hot surface. A relation is established between the TiN layer thickness and the resistivity and between the TCR and the resistivity: thin TiN layers have a higher ρ and TiN with a high resistivity have a lower TCR. Furthermore, it is shown that the TCR, measured between 25 and 175 °C, remains constant up to 600 °C. In chapter 3, the oxidation behaviour of thin TiN layers is investigated. During the fabrication process and operation hotplate, the ALD TiN layer should not oxidize. Despite the fact that TiN is considered as a very oxidation-proof material, little oxidation is necessary to modify the properties of a 7-15 nm thin film significantly in dry (O2) and wet (H2O) atmosphere at temperatures between 300 and 500 °C and during exposure to oxygen containing plasmas. The kinetics of the oxidation process have been studied. The composition of the ALD TiN film and the generated TiO2 is comparable to sputtered stoichiometric TiN layers. Using a suitable protection layer, the oxidation of ultrathin TiN layers can be prevented effectively. In chapter 4, the design, the novel fabrication process and the electrical characterization of link-based microelectronic hotplates are shown. Hotplates were fabricated with different link sizes: microlinks (∅ 2-6 μm) and nanolinks (∅ 100 nm) and for some hotplates, the silicon underneath the device is removed. By doing so, a suspended membrane is released for thermal insulation. Microlink-based hotplates have a low ohmic link. They reach a temperature of 250 °C with a power dissipation of 2.7 mW and cannot be heated without a suspended membrane. Nanolink-based hotplates can reach a temperature of 280 °C with a power consumption of 5.5 μW. Without a suspended membrane, only factor 2 more power is required for the same temperature. This makes a nanolink-based device without a suspended membrane and interesting candidate for a mechanically robust hotplate. In chapter 5, two techniques are investigated for measuring the temperature of a device using an alternative method than the temperature dependent resistance of the heater. The employed infrared (IR) thermometry method can be applied successfully to microlink-based hotplates. The employed polymer melting method can be applied successfully to measuring large (> 100×100 μm2 ) areas. However, it turned out that both methods, as employed in this work, are not sensitive enough to detect the small amount of heat that is generated by the small (< 1×1 μm2 ) surface of the nanolink based hotplates. Finally, in chapter 6, the unexpectedly high leakage current through the glass around the nanolink is discussed (together with a low breakdown field of the glass) that is observed for some hotplates. The high leakage current is related to a fundamental property of the ALD TiN process, required for manufacturing of the link. The excellent step coverage has the disadvantage that the sensitivity for process related errors in the device increases dramatically. Structural defects in the insulating glass layer are filled in with conducting TiN, leading to a low ohmic conducting path parallel to the link. Experiments using sputtered TiN layers with a worse step coverage show that these structural defects remain unnoticed otherwise. These experiments show the importance of the process integration for the introduction of new process steps and/or materials.

Surface passivation, i.e. the reduction of electronic recombination processes at semiconductor surfaces, is essential for realizing high Si solar cell efficiencies. In turn, the increase in the energy conversion efficiency is a major driver for reducing the costs of photovoltaic electricity. However, at present, effective surface passivation schemes have been implemented in only a fraction of industrial Si solar cells. Therefore, the development of suitable surface passivation schemes and related technology is currently a key topic in photovoltaic research. This is underscored by the immense attention that aluminum oxide (Al203) has captured in recent years after being introduced as an effective surface passivation material in 2006. Al203 thin films appeared to have some advantages over contending technologies for the passivation of the rear side of p-type Si solar cells and for the passivation of the p+ emitter in n-type Si cells. Along with the use of Al203, atomic layer deposition (ALD)—with its benefits of submonolayer growth control and excellent uniformity and step-coverage—was also introduced as a novel deposition method in the field of Si photovoltaics. 
This thesis addresses topics ranging from the fundamental mechanisms that govern the properties of nanolaycr surface passivation schemes to the industrial feasibility of the technology. These aspects are closely interwoven, as improved fundamental understanding forms the basis for developing, optimizing, implementing and commercializing the relevant technologies. The focus throughout the thesis is on Al203 and on ALD, both of which enabled new opportunities for developing and studying nanolayer surface passivation schemes. A share of the research was carried out in collaboration with strategic partners, including the solar cell manufacturer Q-Cells. 
The properties of the Al203 films were evaluated on multiple levels: Firstly, the surface passivation quality of the films was studied in relation to various technologically relevant parameters such as ALD conditions, annealing recipes, material properties and film thicknesses. It was found, for instance, that Al203 films with thicknesses down to 5.10 nm synthesized by plasma-assisted and thermal ALD induced ultralow surface recombination velocities, S,/y< 5 cm/s, on low resistivity n- and p-type Si in a relatively wide processing parameter window. Secondly, the chemical passivation (i.e., the reduction of the interface defect density) and field-effect passivation by negative fixed charges (i.e., the shielding of electrons from the surface), responsible for the passivation quality of Al203, are addressed in detail. While its negative fixed charge density is a distinguishing property relative to other relevant surface passivation materials, it is established that the effective chemical passivation, as demonstrated by very low defect densities of < 1011 me eV4 at mid gap, also plays a key role in the Al203 passivation properties. Finally, the fundamental mechanisms controlling the chemical- and field-effect passivation were addressed experimentally using innovative approaches. For example, the diffusion of hydrogen present in the Al203 films (typically —3 at.%) during annealing was studied with thermal effusion experiments and was correlated with the hydrogenation of interface defects. The
latter, in combination with the presence of an interfacial Si02 layer, are key to the low D„ values achieved by Al203. Regarding the field-effect passivation, the thickness of the interfacial Si02 was identified as a key parameter controlling the negative fixed charge density associated with Al203. This experimental study relied on using an ALD Si02 process for interface engineering with Angstrom resolution. It was combined with diagnostics such as electric-field induced second-harmonic generation for the contactless probing of the changes in charge distribution for thicker Si02 interlayers. 
This thesis also addresses surface passivation stacks, such as Al203/a-SiN„:H, SiOx/a-SiN„:H and Si02/Al203 stacks. The newly introduced SiO2/Al203 stacks are compatible with very low Se values, regardless of the Si02 synthesis method. In fact, the use of Al203 capping layers enabled an unprecedented high interface quality for low-temperature synthesized Si02. This appears to be mainly related to a very effective hydrogenation of the remote Si/Si02 interface during annealing. For both Si02/Al203 and Si02/a-SiN„:H stacks, field-effect passivation was reduced significantly compared to the corresponding single layers, which can avoid—sometimes undesirable—inversion conditions. Therefore, by using surface passivation stacks, not only the optical and chemical properties, but also the underlying passivation mechanisms can be controlled and optimized for solar cell and other electronic applications.
Regarding solar cell processing, various aspects pertaining to the feasibility of Al203 and related technologies are addressed. For instance, the thermal stability of Al203-based passivation schemes proved to be adequate. Moreover, films deposited using batch- or spatial ALE) methods—specifically designed by a number of companies to meet the throughput requirements for PV manufacturing—were shown to exhibit similar properties as obtained by single-wafer laboratory reactors. In addition, a pulsed-precursor PECVD process is reported as an alternative method for the fast deposition of Al203 and other materials. 
The development and understanding of Al203-based surface passivation schemes in conjunction with the new ways of investigating, controlling and manipulating their properties, as outlined in this thesis, are important for the ongoing developments in the field of photovoltaics aiming at higher efficiencies and lower costs per kilowatt-hour. Based on the recent announcements about enhanced efficiencies for industrial-type rear-passivated solar cells and the installation of high-throughput deposition systems for Al203 in solar cell pilot lines, it is expected that Al203-based surface passivation schemes will provide a leap in performance of a large share of commercially available solar modules in the coming years. In a broader context, the relevance of this thesis may extend to the field of (nano-)electronics in which the continuous reduction of device dimensions demands even more stringent requirements of thin film technology. Extrapolating the rapid developments in recent years, it is expected that ALD will play an increasingly important role for Si-based, but probably also other, photovoltaic applications in the near future.
 

Micromechanical devices have been fabricated out of silicon for decades, but only recently even smaller structures - nanodevices have become experimentally possible. Traditionally silicon devices are fabricated using separate lithography and various etching methods. This thesis work concentrates on developing fabrication techniques for silicon micro and nanostructures. The goal was to achieve nanometer-scale feature size and simultaneously significantly speed up the most time consuming phases. For testing purposes also functional devices were designed and fabricated. Main discoveries are related to the use of ion beam writing in a nonstandard manner. Instead of direct milling, methods were developed to directly use the beam to replace time consuming lithography step by the substrate treatment by ions. As a result, several silicon-based fabrication techniques were developed that require only a few processing steps and therefore can be realized in less than one day. The main achievement is in overcoming some of the limitations of serial writing methods such as those required in electron beam lithography or focused ion beam processing. High aspect ratio (laterally small, but tall) structures were successfully obtained using both technologies for the pattern transfer. Fabrication techniques, described in this thesis, open up an opportunity for the developers to almost instantly test their ideas using functional components by altering the way nanosystems are developed. The presented methods cannot easily b 

Self-assembled, 3D nanoporous templates present an opportunity to develop devices which are lithography-free, massively scalable and hence, highly manufacturable. Self-limited deposition processes on the other hand, allow functional thin films to be deposited inside such templates with precision and unprecedented conformality. Taken together, the combination of both processes provides a powerful ‘toolbox’ to enable many modern nano devices.
In this work, I will present data in three parts. First, I will demonstrate the capabilities of Atomic Layer Deposition (ALD), a self-limited thin film deposition technique in preparing nanoalloyed Al-doped ZnO (AZO) thin films. These films are visibly transparent and electrically conducting. Structure-property relationships are established that highlight the power of ALD to tailor film compositions at the nanoscale.

Next, I will use ALD ZnO films in conjunction with aged, ALD V2O5 films to form pn junctions which show rectification with an Ion/Ioff as high as 598. While, the ZnO is a well known n-type semiconductor, the discovery of p-type conductivity in aged V2O5 is surprising and is found to be due to the protonic (H+) conductivity of intercalated H2O in V2O5. Thus, we demonstrate a mixed electronic-ionic pn junction for the first time.

Finally, I combine the material set of the pn junction with self-assembled, anodic aluminum oxide (AAO) 3D nanoporous templates to create 3D nanotubular pn junctions. The pn junctions are built inside pores which are only 90nm wide and up to 2μm deep and show rectification with Ion/Ioff of 16.7.

Process development and integrations strategies will be discussed that allow for large scale manufacturing of such devices a real possibility.
 

This research describes the development of ALD processes for the deposition of nitride materials including tantalum, hafnium and gadolinium nitrides. Ta and Hf nitrides are of significant interests for sub-lOOnm silicon based electronic devices, while Gd nitride may be exploitable in future spintronic devices. ALD has been established a key manufacturing tool in microelectronics, the development of ALD processes for these nitrides are essential for future manufacturing of electronic devices and can benefit future manufacturing of spintronic devices. In the current research, these nitrides were deposited using ALD and the films were characterised using MEIS, AES, XRD, TEM, SEM, AFM, and a four point probe. Ta nitride films were grown at temperatures ranging from 200°C to 375°C using ALD with Pentakis(dimethylamino)tantalum, Ta(NMe2)5 as the metal source and either ammonia or monomethyl-hydrazine (MMH) as a nitrogen co-reactant. Self-limiting behaviour was observed for both ammonia and MMH processes, with growth rates of 0.6 and 0.4 A/cycle respectively at 300°C. Films deposited using ammonia were found to have a mono-nitride stoichiometry with a cubic microstructure and resistivities as low as 70 mQ.cm. In contrast, films deposited using MMH were found to be nitrogen rich TasNs with an amorphous microstructure and high resistivities (>4 Q.cm). A QCM was used to measure mass gain and loss during the cyclic ALD processes and the data was used in combination with MEIS to elucidate the Ta(NMe2)5 absorption mechanisms. For Hf nitride, films were firstly deposited using thermal ALD with tetrakis(dimethylamino)hafnium, Hf(NMe2)4 and ammonia between 100 and 400°C. Selflimiting behaviour was observed, however, the films exhibit a low density and were prone to oxidation during post-deposition exposure to air. A comparison between thermal and PE ALD was then made at 300°C with tetrakis(ethylmethylamino)hafnium, Hf(NEtMe)4 as the metal source and either molecular or plasma-cracked ammonia as a nitrogen source. PEALD allows shorter purge time, which significantly reduces the cycle length; PEALD also results in higher film density. The densities of the films deposited by PEALD and thermal ALD were found to be 11.6 and 9.7 g/cm3 respectively. Mass spectroscopy indicates that the process characteristics in PEALD are attributed to the nature of the co-reactants, namely, radicals of hydrogen and nitrogen. Their high reactivity and short life time are responsible for the resulted high density and the short required purge time. All films deposited were found to be insulators and with an amorphous microstructure. The films deposited by PEALD remain amorphous and stable with no interactions between Hf and Si after vacuum annealing up to 800°C. Gd nitride films were successfully deposited using a cyclic PEALD based process. The deposition was carried out with tris(methylcyclopentadienyl)gadolinium, Gd(MeCp)3, and remote nitrogen plasma exposure, separated by argon pulses. Films were deposited at temperatures between 150 and 300°C and capped with Ta nitride to prevent post deposition oxidation. Gd nitride with a 1:1 Gd:N ratio, low oxygen incorporation (5%), good thickness uniformity (95%), an amoiphous micro structure and smooth surface (Ra.=~0.7nm) have been deposited. Deposition with tris(silylamide)gadolinium, Gd{N(SiMe3)2}3, and either ammonia or MMH was also investigated. Although the process using ammonia was unsuccessful due to the insufficient reactivity of ammonia, the results show that a reaction between Gd{N(SiMe3)2}3 and MMh does take place. Gd{N(SiMe3)2}3 was found to be a self-limiting precursor, however, the as deposited films were found to be GdSixOy. The silicon incorporation was attributed to partial breakdown of silylamine groups, where the oxygen incorporation was attributed to the possible tetrahydrofuran (THF) contamination in the precursor. 

The world currently consumes over 16 TW of energy which is derived primarily  from carbon-based sources including natural gas, oil, and coal, and energy use is  expected to double by the year 2050. As concerns about energy security and carbon  emissions have increased over the past decade, the search for alternative and renewable  energy sources has garnered much attention. Photovoltaic (PV) technology is a leading  candidate to be a major contributor to future electricity production since sunlight is a vast  resource of energy and can be directly converted into usable electricity. As research into  photovoltaics has rapidly progressed, interfacial effects on the nanoscale have  increasingly come into focus; thus, the requirements for deposition techniques of PV  materials have become more stringent. Atomic layer deposition (ALD) has emerged as a  promising tool for studying and improving PV technology because of its unique  capabilities to coat nanoporous substrates, to controllably deposit films at sub-Ångstrom  thicknesses, and to manipulate compositions of very thin films. Understanding ALD  processes and the quality of deposited films is an important step in developing systems  with applications to PV manufacturing.  The II-VI semiconductor system is particularly interesting for its use in  transparent conducting oxides and in buffer layers for thin film PV. Of particular  relevance, the bandgap, crystal structure, growth rate, index of refraction, conductivity,  and resistivity of these materials can be tuned over large ranges by controllably  depositing tertiary alloys. ALD is one of the premier techniques for achieving this  control since it is a surface reaction rate-limited process in which a sub-monolayer of  material is deposited per ALD cycle. Thus, ALD allows for control of material  deposition at the Ångstrom level.  The equipment utilized for ALD material deposition is an important  consideration for any process and application. We have developed two ALD reactors:  one has been optimized for the deposition of II-VI alloy materials, and the other has been  designed to efficiently vaporize low vapor pressure precursors for relevant ALD  processes. With the first reactor, we have demonstrated a method for in situ generation  of small quantities of H2S for sulfide films, and we have expanded the knowledge of the  vi  II-VI system by ALD. The processes of ZnS, CdS, CdxZn1-xS, and ZnOyS1-y were  developed for testing as buffer layers in thin film photovoltaics, and we analyzed the  surface reactions that affect deposition of tertiary ALD films. Finally, we developed and  characterized the ALD process for CdO and CdxZn1-xO, which is the first step in  developing low resistivity transparent conducting oxides by ALD. The metalorganic  precursors utilized for each of the depositions affected the ALD growth properties, and  we performed experiments to show that the size of the ligand was an important  consideration for these processes. The growth and material properties of these films were  studied by spectroscopic ellipsometry, ultraviolet-visible spectroscopy, transmission  electron microscopy, atomic force microscopy, X-ray diffraction, scanning electron  microscopy.  The II-VI semiconductor project was concluded with a study of interfacial  engineering of CuIn1-xGax(S1-ySey)2 (CIGS) thin film photovoltaics in which the pn  heterojunction was formed via ALD of CdxZn1-xOyS1-y. Using these ALD materials, the  effect of thickness, surface treatment with solutions, alloy composition, and grading of  materials was analyzed. The devices were characterized by current-voltage (I-V) and  external quantum efficiency (EQE) measurements, which indicated that device  performance is strongly related to the treatment and to the composition of the film. This  thesis concludes with thoughts and perspectives of the future of ALD in PV  manufacturing.  

Microfabrication and thin film technologies were applied in the fabrication of miniaturized chemical sensors. Two types of devices were developed: an electrochemiluminescence device utilizing tunnel-emitted hot electrons, and a microhotplate semiconductor gas sensor with an atomic layer deposited (ALD) tin dioxide sensing film. The hot electron-induced electrochemiluminescence (HECL) device is an integrated microelectrode device that combines an insulator-covered working electrode and a platinum counter electrode on a single chip. Two types of fluidic systems were integrated on the same type of electrode chip: either an enclosed sample chamber made of polydimethylsiloxane (PDMS) elastomer, or hydrophobic sample confinement on the chip surface. Different metals were tested as the working electrode, and different types of insulator films made by various methods were tested as the tunneling dielectric, to determine the optimal working electrode structures for HECL. These were then used in the integrated microelectrode devices, which were fabricated on silicon and glass substrates. A variety of electrode geometries were tested with the different fluidic systems, and sub-nanomolar sensitivity and wide dynamic range were demonstrated with the best devices. Ongoing work with polymeric substrates is briefly presented. In the review, latest results are presented on the restoration of the hydrophilic properties of enclosed PDMS microfluidic channels. While a PDMS surface quickly reverts to its naturally hydrophobic state, thus preventing capillary filling, this plasma treatment enables capillary filling even after extended periods of storage. The gas sensor is a microhotplate (MHP) device, utilizing a tin dioxide sensing layer deposited by ALD for the first time in a MHP sensor. Unconventional solutions were developed for the fabrication sequence to accommodate the demands of the deposition method. Also, metallizations and intermetal dielectrics not commonly used in MHP devices were tested to enable rapid processing of prototype devices with available methods and equipment. Fast response to various analyte gases, as well as good recovery and short-term stability were observed, demonstrating the potential of ALD tin dioxide films in gas sensor applications.

Atomic Layer Deposition (ALD) is a vapor-phase deposition technique in which ultra thin films are synthesized by repeating two subsequently executed half-cycles. Due to its characteristic self-limiting surface reactions, ALD offers sub-nanometer precision of film growth, uniform deposition over large substrate areas and conformal deposition in struc tures of high aspect ratio. Plasma-assisted ALD is a variant to the conventional thermal ALD technique where the surface is exposed to a plasma during the second half-cycle. The use of a plasma allows for more freedom in processing conditions and for a wider range of material properties compared with the conventional thermally-driven ALD me thod. Although it has been known from plasma-based techniques that photons and ions can also play an important role during processing, their contribution has never been sys tematically addressed for the specific case of plasma-assisted ALD. In this dissertation work, the plasma-surface interaction in plasma-assisted ALD is investigated. In the first part of this dissertation, the basics and status of plasma-assisted ALD have been discussed. A classification of plasma-capable ALD configurations has been made and examples for plasma-based processes of metal-oxide, metal-nitride and pure metal thin films prepared by plasma-assisted ALD have been used to illustrate its merits compa red to thermal ALD. Several challenges of plasma-assisted ALD, such as plasma-induced damage and surface-recombination of radicals, have also been identified. Several repor ted emerging applications of plasma-assisted ALD have been reviewed. Since the depo sition of thin films at low temperatures, and in particular at room temperature, is one of the most promising application areas of plasma-assisted ALD, special attention has been dedicated to this topic. Taking the deposition processes of Al2O3, SiO2, and TiO2 as examples, criteria for viable deposition at room-temperature using plasma-assisted ALD and ozone-based ALD have been identified. The second part deals with the identification and evaluation of ions and photons du ring plasma-assisted ALD. The presence and importance of ions have been discussed for four reactor configurations commonly used during plasma-assisted ALD. It has been shown that the energy and flux of ions towards the substrate surface is mainly determi ned by the reactor configuration, the gas pressure, the plasma power, and the electrical potential of the substrate stage. Under processing conditions typically employed during plasma-assisted ALD, ion energies are up to several tens of eV and ion-induced damage is therefore not a major issue during most processes. The energy flux of the ions toward the substrate surface can however be sufficient to promote some beneficial physical ef fects such as enhanced ligand-desorption, adatom migration and displacement of lattice atoms. With respect to the optical emission of plasmas, it has been shown that energetic vacuum ultraviolet (VUV) photons can be present in plasmas, which are able to induce electrical defects at interfaces. By varying the gas pressure and the plasma power, howe ver, the influence of ions and photons can be suppressed. The third part of this dissertation discusses the exploitation of the presence of ions during plasma-assisted ALD using substrate biasing in order to affect the properties of metal-oxide films. The ion energy was enhanced by applying a substrate bias signal at radio frequency to the substrate stage in a reactor equipped with an inductively-coupled plasma source. Alternatively, substrate-tuned biasing was used where no additional po wer source is required and where the substrate potential can be varied by enhancing capacitive coupling using an external electrical circuit. It has been demonstrated that both biasing techniques are viable and that ion energies have been increased up to a few hundreds of eV. The influence of these high-energy ions on ALD-synthesized thin films was illustrated for Al2O3, Co3O4 and TiO2 and it has been demonstrated that thin film properties can be tailored in terms of thin-film stress, composition, density and crystalli nity when ion energies are carefully tuned. To conclude, in this dissertation work important aspects of the plasma-assisted ALD technique have been elucidated leading to a better understanding of the fundamental and technological opportunities and limitations of the technique. The work will therefore contribute to the advancement and the acceptance of plasma-assisted ALD in science and technology.

The advancing complexity of advanced microelectronic devices is placing rigorous demands on currently used PVD and CVD deposition techniques. The ALD deposition method is proposed to meet the film thickness and conformality constraints needed by the semiconductor industry in future manufacturing processes. Unfortunately, there is a limited number of chemical precursors available that have high thermal stability, reactivity, and vapor pressure suitable for ALD film growth to occur. These properties collectively contribute to the lack of suitable transition metal precursors available for use in ALD. In this thesis, the discovery of a series of novel transition metal diazadienate precursors that promising properties deemed suitable for ALD is reported. The volatility and thermal stability of the new transition metal diazadienyl compounds were studied by preparative sublimation and capillary tube melting point/decomposition experiments. Thermogravimetric analyses (TGA) demonstrate precursor residues of less than 4% at 500 °C. In addition, sublimation data, melting points, and decomposition temperatures for all complexes are presented. The manganese diazadienyl complex has the highest decomposition temperature of the series of complexes produced (325 °C). During preparative sublimations, the product recoveries of all transition metal diazadienyl complexes were greater than 92.0% with nonvolatile residues of less than 7.0%. This is an excellent indication that these complexes may be suitable candidates as metal precursors for ALD. Nickel nitride (NixN) films have been studied as an intermediate material for the formation of both nickel metal and nickel silicide using chemical vapor deposition. Herein, the ALD growth of nickel nitride thin films from bis(1,4-di-tert-butyl-1,3-diazabutadiene) nickel(II) (Ni(tBu2DAD)2) and 1,1-dimethylhydrazine, is described. An ALD window for the deposition of nickel nitride films on 500 nm thermal SiO2 substrates was observed between 225 and 240 °C with a constant growth rate of 0.70 Å/cycle. X-Ray photoelectron spectroscopy (XPS) showed all expected ionizations with carbon concentrations below the detection limit after argon ion sputtering. Due to preferential nitrogen sputtering in XPS, Rutherford backscattering spectrometry (RBS) and nuclear reaction analysis (NRA) were performed and subsequently revealed Ni:N ratios between 2–4 for films deposited within the ALD window. AFM measurements revealed a RMS roughness value of 10.8 nm on an as deposited film at 225°C. All as-deposited films were amorphous as determined by X-ray diffraction. Copper is the primary interconnect material in microelectronics devices, due to its high conductivity and low affinity towards electromigration. With transistor gate lengths scheduled to reach 14 nm by 2014, there are severe demands upon the current film growth techniques used in device fabrication. The ALD film growth method is ideally suited for future microelectronics manufacturing, since it inherently provides highly conformal thin films, even in high aspect ratio nanoscale features, and allows sub-nanometer control over film thicknesses. In Chapter 4, the atomic layer deposition of high purity, low resistivity copper metal thin films using a three precursor sequence entailing Cu(dmap)2, formic acid, and hydrazine, is presented. In this process, Cu(dmap)2 is unreactive towards hydrazine but is transformed to copper(II) formate, which is then readily reduced to copper metal by subsequent hydrazine exposure. The present work therefore addresses a central problem with the ALD growth of metal thin films: low reactivity of metal precursors toward common reducing agents. A constant growth rate of 0.47-0.50 Å/cycle upon prime grade Si(100) was observed at substrate temperatures between 100 and 170 °C. Compositional analyses (XPS and TOF-ERDA) revealed copper films with low levels of carbon, oxygen, nitrogen, and hydrogen. Powder X-ray diffraction spectra of all films showed polycrystalline copper. The resistivities of films grown between 100 and 140 °C ranged between 9.6 and 16.4 μΩ•cm, demonstrating the growth of high purity, low resistivity copper films. An AFM measurement revealed a RMS roughness value of 3.5 nm on an as-deposited 50 nm Cu film at 120 °C.

This thesis describes three techniques for the characterization of living cells using micro-electro-mechanical systems (MEMS) based devices. The study of cellular function and structure is essential for bioprocess control, disease diagnosis, pa tient treatment and drug discovery. Microsystem technology enables characteriza tion of very small samples, minimal use of expensive reagents, testing of multiple samples in parallel, and point-of-care testing, all of which increase throughput and reduce the analysis cost. The three characterization techniques presented in this thesis could be integrated into a microfluidic cellular total analysis system to obtain complementary information of cellular function. The first part of the thesis presents the characterization of bovine adrenal cor tex capillary endothelial cells by impedance spectroscopy in a microsystem which was realized using microfabrication techniques. The microsystem consists of a small-volume cell culture area defined on PDMS walls on a glass substrate with gold electrodes coated with a self-assembled monolayer to enable cell attach ment. As the main result, it was possible to monitor the capillary formation of BACC endothelial cells in a microsystem using impedance spectroscopy. The second part describes calorimetric characterization of Saccharomyces cerevisiae yeast cells using a MEMS-based nanocalorimetric microsensor. The cells are introduced to the sensor membrane in small droplets (~1 µl), and the sensor thermopile voltage output is compared to the output of the reference water droplet to extract the effect of sample evaporation. The third part describes the design, process integration and fabrication of an electrically tunable Fabry-Perot interferometer (FPI) monolithically integrated on a photodiode for visible spectrum measurements. The options for the process inte gration of separate FPI optical filters are presented. The application of miniature spectrometers based on MEMS FPI technology in biological cell characterization is discussed.

The dye-sensitized solar cell (DSC) is a photovoltaic technology with the potential to efficiently and economically harvest and convert energy from the sun to electrical power. DSCs are built using abundant and low cost materials such as titanium dioxide (TiO2) and organic dye molecules. The dye molecule acts as a light absorber funneling electrons from its photo-excited state to the TiO2. A redox mediator which typical consists of iodide/tri-iodide undergoes redox reactions at the counter electrode and the oxidized dye molecule creating a circuit between the two. Solid-state versions of the DSC are also being investigated. In these devices the liquid electrolyte is exchanged with solid hole transporting material in order to both simplify the solar cell production as well as increasing the open-circuit potential and stability of the solar cell. One main draw-back, which limits the increase in conversion efficiency of solid-state DSC is the faster electron recombination dynamics between electrons in the TiO2 and holes in the solid hole transporter. Currently the highest performing liquid electrolyte DSC reaches a conversion efficiency of over 12 %, while the solid-state DSC is tailing with 7 %.
Materials development is crucial for further development of the DSC technology, hopefully leading to better stability and higher efficiency. Many types of dye molecules, redox mediators as well as hole transporting materials and working electrode materials have all been tested and modified in the past in order to improve DSC performance. Significant further improvement of DSC technology requires a better understanding of the operating principle behind the DSC and the interaction between the different components. This requires advanced characterization methods for materials and solar cells. In this thesis, new materials for DSC have been developed, tested and characterized using advanced methods. 
Atomic layer deposition was employed to develop a new working electrodes based on the core-shell SnO2-TiO2 material. These working electrodes were successfully used in both liquid and solid-state DSC to decrease the electron recombination dynamics and increase conversion efficiencies. The molecular structure of sensitizing dyes also plays a major role in electron recombination. Thus, investigating different molecular structures of sensitizing dyes is of importance when trying to improve DSC performance. Seven new molecular dye structures based on three different chromophore units were investigated in both liquid electrolyte and solid-state DSC. For example, adding a second anchoring group on the D35 molecular structure improved the light harvesting capabilities of the dye but did not result in DSC devices with higher conversion efficiency. Increasing the bulkiness of the molecular dye structure facing away from the TiO2 surface yielded on the other hand higher both slower electron recombination and higher conversion efficiencies.
The effects of oxygen on solid-state DSC using spiro-OMeTAD were also studied. The chemical oxidation of the solid-state hole transporting material was found to depend on both time and storing conditions of the complete DSC devices. Solar cells with higher conversion efficiency were found for solid-state DSC stored under ambient air conditions before measured.
Finally, a novel and efficient organic tandem solar cell was demonstrated built using a solid-state DSC and a bulk heterojunction solar. The 6 % efficient tandem cell almost perfectly added the photo-potentials of the subcells together while keeping the photo-current intact.
 

Electronic components, based on current semiconductor technologies and operating in radiation rich environments, suffer degradation of their performance as a result of radiation exposure. Silicon carbide (SiC) provides an alternate solution as a radiation hard material, because of its wide bandgap and higher atomic displacement energies, for devices intended for radiation environment applications. However, the radiation tolerance and reliability of SiC-based devices needs to be understood by testing devices under controlled radiation environments. These kinds of studies have been previously performed on diodes and MESFETs, but multilayer devices such as bipolar junction transistors (BJT) have not yet been studied. In this thesis, SiC material, BJTs fabricated from SiC, and various dielectrics for SiC passivation are studied by exposure to high energy ion beams with selected energies and fluences. The studies reveal that the implantation induced crystal damage in SiC material can be partly recovered at relatively low temperatures, for damage levels much lower than needed for amorphization. The implantation experiments performed on BJTs in the bulk of devices show that the degradation in device performance produced by low dose ion implantations can be recovered at 420 o C, however, higher doses produce more resistant damage. Ion induced damage at the interface of passivation layer and SiC in BJT has also been examined in this thesis. It is found that damaging of the interface by ionizing radiation reduces the current gain as well. However, for this type of damage, annealing at low temperatures further reduces the gain. Silicon dioxide (SiO2) is today the dielectric material most often used for gate dielectric or passivation layers, also for SiC. However, in this thesis several alternate passivation materials are investigated, such as, AlN, Al2O3 and Ta2O5. These materials are deposited by atomic layer deposition (ALD) both as single layers and in stacks, combining several different layers. Al2O3 is further investigated with respect to thermal stability and radiation hardness. It is observed that high temperature treatment of Al2O3 can substantially improve the performance of the dielectric film. A radiation hardness study furthermore reveals that Al2O3 is more resistant to ionizing radiation than currently used SiO2 and it is a suitable candidate for devices in radiation rich applications.

The increasing demand on available energy resources has led to a desire for  more energy efficient devices. The wide use of displays in consumer electronics, such as  televisions, cell phones, cameras and computers makes them an ideal target for improvement.  Organic light-emitting diodes (OLEDs) are a good candidate to replace traditional Si based  devices. However, the low work function metals typically used as electrodes in OLEDs are very  reactive with water and oxygen. Ultralow permeability gas diffusion barriers with water vapor  transmission rates (WVTRs) as low as <10-6 g/(m2*day) are required on the polymers used to  fabricate organic electronic and thin film photovoltaic devices.  Atomic Layer Deposition (ALD) uses self-limiting surface reactions to deposit thin  conformal films. ALD is capable of depositing thin, conformal, high quality barriers. WVTR  values as low as  5 x 10-5 g/(m2*day) have been measured for Al2O3 ALD films at 38C/85% RH  using the Ca test with optical transmission probing. The Ca test is a technique with very high  sensitivity to measure ultralow WVTRs. This test relies on measuring the oxidation of a Ca  metal film by monitoring the change in its optical or electrical properties. However, glass lid  control experiments have indicated that the WVTRs measured by the Ca test are limited by H2O  permeability through the epoxy seals. Varying results have been reported in the literature  using the electrical conductance of Ca to measure permeation. In this work, two approaches were applied to overcome the epoxy edge seal limitations.  The first approach was to deposit Al2O3 ALD barriers directly on Ca metal. While the Al2O3 ALD  barriers were successfully deposited, the measurement of an accurate WVTR was limited by  barrier pinholes. The presence of pinholes in the Al2O3 ALD barrier on Ca results in the localized  oxidation of the Ca sensor. Heterogeneous degradation of the Ca causes inaccuracies in the  conductance of the film. As oxidation regions merge, large percolation paths are severed  without complete Ca oxidation. To solve this problem, a new apparatus was developed that  measures the electrical conductance of Ca films. This new apparatus does not rely on epoxy  seals and separates the Ca metal from the barrier. Unfortunately, the electrical conductance of  the Ca film versus Ca oxidation was found to be extremely nonlinear. This nonlinearity severely  complicates the usual analysis to obtain WVTR values from the Ca test.  The new apparatus was useful for the examination of PEN polymer substrates using the  total lifetime of the Ca sensor. Polymer effects on the measurement of gas permeability for  polymer/barrier systems have largely been ignored. Experiments were performed to determine  the effect of the PEN polymer substrates on the WVTR measurement. The H2O permeation  activation energy in the PEN polymer, the effect of water saturation of the PEN polymer and  the dependence of the lifetime of the Ca sensor on the H2O flux on the PEN substrate were of  particular interest. The experiments obtained H2O permeation activation energies in the PEN  polymer of 12.4 kJ/mol. The Ca sensor lifetime was found to be linear with H2O flux. No  difference in Ca sensor lifetime was observed between dry or H2O-saturated PEN polymer  substrates.  

Infiltration method, developed at the University of Pennsylvania, is a unique analytical platform for investigating the effect of material properties and electrode microstructure on the performance of solid oxide fuel cell (SOFC) electrodes. During cell fabrication by infiltration, the ion-conducting electrolyte phase is sintered first, followed by the addition of the catalytically active perovskite phase into the pores of the electrolyte. The use of separate sintering steps for the electrolyte and the active phase gives one a high degree of control over the microstructure of both phases, unattainable with traditional fabrication methods. In this thesis, the infiltration approach has been used to conduct a systematic investigation into the factors that govern the performance and stability of solid oxide fuel cell cathodes. As a result, a number of microstructural and material properties, crucial for obtaining high electrode activity, were identified. In particular, the effect of varying the ionic conductivity of the porous electrolyte, the specific surface area of the perovskite as well as the specific surface area of the porous electrolyte, and the effect of solid-state reactions between the two phases were studied and were found to significantly affect performance. The experimental findings agreed well with the predictions of a mathematical model that was developed to describe the electrochemical characteristics of SOFC composite cathodes. Both theoretical and experimental evidence suggests the performance of SOFC cathodes prepared by infiltration is limited by slow oxygen adsorption on the perovskite surface. The chemical composition of the perovskite surface therefore plays an important role in determining the overall performance of the electrode. The last chapter of this thesis introduces a novel method that may allow one to characterize the active sites on the perovskite surface under SOFC cathode operating conditions (600-700°C, ambient air atmosphere, polarization), unattainable with traditional surface characterization techniques.

The main goal of this dissertation was to show that the principle of atomic layer deposition (ALD) can be applied to “endless” fibers. A reactor of atomic layer deposition has been designed, especially for coating depositions onto meter long bundles of fibers. Aluminum oxide (alumina), titanium oxide (titania), double layers of alumina and titania, as well as aluminium phosphate have been deposited onto bundles of carbon fibers using the home-built reactor. Scanning electron microscopic (SEM) and transmission electron microscopic (TEM) images indicate that the coatings were uniform and conformal onto fiber surface. There was a good adhesion of the coatings to the fibers. Alumina has been deposited using two separate aluminum sources (aluminum trichloride and trimethylaluminum), and water as a source of oxygen. In case of alumina deposition using aluminum trichloride and water, initial deposition temperature was 500 °C. In these conditions, a part of the fiber bundle has been damaged. Thus, the deposition temperature was decreased to 300 °C and the fibers were unaffected. In addition, during this process hydrochloric acid is formed as a byproduct which is a corrosive substance and affects the reactor and there was a chloride impurity in the coatings. Thus, aluminum trichloride precursor was replaced by trimethylalumium. Alumina deposition onto carbon fibers using trimethylaluminum and water was carried out at a temperature of 77 °C. SEM images revealed that the fibers were unaffected and the coatings were uniform and conformal. Oxidation resistance of the carbon fibers was improved slightly after alumina deposition. Oxidation onset temperature of the uncoated fibers was about 630 °C. The resistance was linearly increased with the coating thickness (up to 660 °C) and getting saturated over a thickness of 120 nm. Titania coatings have been deposited using titanium tetrachloride and water. The physical appearances of the titania coatings were similar to the alumina coatings. The oxidation onset temperature of the titania coated carbon fibers was similar to the uncoated fibers but the rate of oxidation was decreased than the uncoated fibers. Two double layer coatings were deposited, alumina followed by titania (alumina/titania), and titania followed by alumina (titania/alumina). If the fibers were coated with the double layer of alumina/titania, they had almost same oxidation onset as alumina coated fibers but the rate of oxidation was decreased significantly compared to alumina coated fibers. This feature is independent of the thickness of the titania layers, at least in the regime investigated (50 nm alumina followed by 13 nm and 40 nm titania). On the other hand, the oxidation onset temperature of fibers coated with titania/alumina (20 nm titania /30 nm alumina) was approximately 750 °C. The fibers were burned completely when temperature was further increased to 900 °C and held another 60 minutes at 900 °C. This is significantly better than any other coating used in this dissertation. ALD of titania and alumina in principle was known beforehand, this dissertation here applies this knowledge for the first time to “endless” fibers. Furthermore, this dissertation shows for the first time that one can deposit aluminum phosphate via ALD (planar surface as well as fibers). Aluminum phosphate might be special interest in the fiber coating because it is a rather soft material and thus might be used to obtain a “weak” coupling between fiber and matrix in composites. Aluminum phosphate was deposited using trimethylaluminum and triethylphosphate as precursors. Energy dispersive X-ray spectroscopy and solid state nuclear magnetic resonance spectra confirmed that the coating comprises aluminum phosphate (orthophosphate as well as other stoichiometries). Scanning electron microscopic images revealed that coatings are uniform and conformal. In cases of alumina and titania, it was observed that the coatings were delaminated from the ends of cut fibers and thus formed of clear steps. On the other hand, for aluminum phosphate coating it was observed that the border between coating and underlying fiber often being smeared out and thus formed an irregular line. It seems in case aluminum phosphate cohesion is weaker than adhesion, thus it might be act a “weak” interface between fiber and matrix. Alumina, titania, and double layer microtubes have been obtained after selective removal of the underlying carbon fibers. The carbon fibers were selectively removed via thermal oxidation in air at temperatures exceeding 550 °C. SEM and TEM images indicate that the inner side of the tube wall has the same morphology like the fibers. In addition, it was observed that the individual microtubes were separated from their neighbors and they had almost uniform wall thicknesses. The longest tubes had a length of 30 cm. 

Atomic layer deposition (ALD) is a vapor phase thin film deposition technique based on sequential self limiting surface reactions. In a typical ALD process, two reactants are introduced alternatively to a substrate, resulting in formation of a single atomic layer each exposure cycle. Repeating the surface-limiting reaction cycle allows ultra-uniform nanoscale films to be formed with precise thickness control over complex 3D surfaces. Similar self limiting surface reactions have been used for the deposition of polymer or organic-inorganic hybrid films. This extended ALD process is described as molecular layer deposition (MLD). In our studies, we are exploring new chemistry and applications of ALD/MLD. Arrays of ALD/MLD processes have been conducted on various polymer substrates. By coating the substrates with different materials, we were able to give new functionalities to the polymers. For example, hydrophilic, conductive, and photocatalytic coatings have been deposited. Moreover, recent studies of ALD on polymers showed that metal organic precursors often diffuse sub-surface into the polymer. This sub-surface diffusion and reaction could result in the change of the chemical composition and the physical properties of the bulk polymer. Surface functionalization and chemistry modification of polymers by ALD/MLD enabled a large number of new applications of polymers. In this work, several demonstration applications will be presented, and the mechanism of the ALD/MLD process on polymers will be discussed. 

Brass and bronze substrates were coated by atomic layer deposition (ALD) with alumina and titania in small and large scale batches. These films were evaluated for use as protective and cosmetic coatings. Optimization of deposition parameters for uniform coatings both on individual coins and across a batch for Al2O3 and TiO2 films was performed. The effect of film thickness on colour was examined. High-quality, uniform coatings were achieved with multi-pulse programs. The interference colours resulting from thin films of Al2O3 deposited by ALD on silicon were analyzed using a robotic gonioreflectometer. A series of thin films were deposited and their reflectivity values obtained for the visible spectrum. A comparison of these values with the predictions of computer simulations has revealed deviations from predicted reflectivities. The effect of these discrepancies on perceived colour appearance was investigated. Simulation predicts larger iridescence than what was observed. Alumina films were deposited by ALD on flat and nanostructured silicon substrates, and incorporated into PEDOT-Al2O3-silicon architectures that were then evaluated as photovoltaic devices. The reverse saturation currents observed on flat devices made with Al2O3 films were similar devices made with an SiO2 layer. The structured samples with Al2O3 showed a considerable increase in efficiency (of up to five times) over the equivalent flat samples. A new indium(III) guanidinate, (In[(NiPr)2CNMe2]3, was synthesized. Thermogravimetric analysis showed elemental indium was produced from the compound as a residual mass. Thermolysis in a sealed NMR tube showed carbodiimide and protonated dimethyl amine by 1H NMR. Chemical vapour deposition (CVD) experiments above 275 °C with air as the reactant gas produced cubic indium oxide films with good transparency. Dimeric silver(I) and gold(I) tert-butyl-imino-2,2-dimethylpyrrolidinates were synthesized and evaluated for thermal stability by thermal gravimetric analysis, differential scanning calorimetry and variable-temperature solution NMR. The compounds were used to deposit metallic films on silicon and glass substrates by CVD, with and without a flow of heated nitrogen gas. The compounds decomposed to produce metallic films at 140 °C for the silver compound and 300 °C for the gold compound. Additional CVD experiments with heated nitrogen gas flow improved film uniformity without sacrificing film purity.

c-BN has been in focus for several years due to its interesting properties. The possibility for large area CVD is a requirement for the realization of these different properties in various applications. Unfortunately, there are at present severe problems in the CVD growth of c-BN. The purpose with this research project has been to theoretically investigate, using DFT calculations, the possibility for a layer-by-layer CVD growth of c-BN. It could be established that, PEALD, using a BF3-H2-NH3-F2 pulse cycle and a diamond substrate, is a promising method for deposition of c-BN films. The gaseous species will decompose in the plasma and form BFx, H, NHx, and F species (x = 0, 1, 2, 3). The H and F radicals will uphold the cubic structure by completely hydrogenate, or fluorinate, the growing surface. However, surface radical sites will appear during the growth process as a result of atomic H, or F, abstraction reactions. The addition of NHx growth species (x = 0, 1, 2) to B radical sites, and BFx growth species (x = 0, 1, 2) to N radical sites, will then result in a continuous growth of c-BN.

The materials of today have intriguing properties. The mastering of phenomena at the nanometer range often forms the basis for the understanding of novel materials and their functional properties. In this thesis three materials - zinc oxide (ZnO), titanium dioxide (TiO2) and tungsten trioxide (WO3) - in the form of thin films less than 100 nanometers in thickness are being studied from the surface functionalization point of view. The film application method used was atomic layer deposition (ALD). First, the low-temperature deposition features of ZnO were studied. The deposition could be performed at as low as room temperature. Below 70 ºC the hexagonal wurzite structure of ZnO oriented along the c axis, and above this temperature orientation along the a axis was also observed. The crystallinity improved in post-deposition annealings at 400–600 ºC in argon and oxygen atmospheres while keeping the original preferential orientation unchanged. ZnO was also used to study the tailoring of a natural template, the wing surface of the cicada (Pomponia Intermedia) insect. The nanoscale pillar structure of the cicada wing could be area-selectively coated with a 100 nm thick ZnO film in the interstitial space between the pillars. By applying a thin aluminum oxide seed layer prior to ZnO deposition the wing nanostructure could be uniformly coated. Furthermore, the water wettability of the cicada wing was studied. The originally superhydrophobic wing surface was coated with ZnO while keeping the original superhydrophobicity almost intact. When exposed to ultraviolet (UV) light the surface was successfully turned hydrophilic and back to hydrophobic under storage in dark. The tunable wettability phenomenon was used to demonstrate directing fluid flows on planar, ZnO-coated quartz surfaces utilizing hydrophilic patterns irradiated on a hydrophobic surface by UV laser. The reversible patterning may find use in microfluidic and lab-on-a-chip devices. TiO2 is a common ALD film material capable of being deposited at low temperatures. Here its possibilities were demonstrated by depositing a TiO2 layer on nanofibrillated cellulose (NFC) template, and after removal of the template using the formed TiO2 nanotube network in a dropcast form as a humidity sensor. The usable relative humidity range was 40–80 % where the dropcast sensor layer gave repeatable resistive and capacitive response. Finally a new ALD process of WO3 films from W(CO)6 and O3 was introduced. The ALD temperature window was observed at 195–205 ºC with a deposition rate of 0.23 nm/cycle. This process provides a straightforward option for the ALD of WO3 as opposed to the methods based on in situ generated oxyfluoride intermediates published earlier.

Elastic recoil detection analysis (ERDA) with heavy ion beams has evolved into a universal ion beam analysis (IBA) method for simultaneous analysis of almost all elements, with an essentially constant detection sensitivity. The method is based on the detection and identification of recoiling atoms that have been elastically scattered from a sample by an incident heavy ion beam. The principal characteristics of heavy-ion ERDA are outlined and illustrated using examples of data obtained with time of flight (TOF) and ∆E −E detector systems. The potential and limitations of the quantitative analysis were explored. For this purpose, a number of thin layer samples were measured using different projectiles and energies. Desorption of the surface materials during ERDA measurements was determined as a function of the probing ion fluence. As the differential cross-sections for scattering were enhanced for heavy projectiles, the beam dose to which the sample was exposed to during measurements was reduced by using heavy ion beams. However the higher cross-sections caused an increase of the desorption. An essential part of this study was dedicated to study those topics that limit the accuracy of the analysis in heavy ion TOF-ERDA, namely: uncertain stopping forces, quantification accuracy, ir radiation induced damage, depth resolution, and the role of multiple and plural scattering. Possible approaches to improve the sample characterisation efficiency and accuracy were stud ied by using a gas ionisation detector. This study concentrates on the noise reduction, detection characterisation, and analysis procedures. The focus was upon the effect of the large solid angle and position sensitivity on the irradiation induced damage, depth resolution, mass resolution, and elemental sensitivity. The reliability of the concentration distributions obtained with heavy ion ERDA was strongly af fected by the surface structure, surface roughness and multiple scattering. These effects were stud ied by comparing Monte Carlo simulations with the experimental results. The analysis procedure was developed to enable the characterisation of novel materials such as atomic layer deposited thin films and nanoparticles. Data handling and storage was improved to facilitate and speed up the analysis procedures.

The electronics industry has been developing improvements in its products at a rapid pace for five decades, an achievement that stems from its ability to continuously decrease the smallest feature sizes in microelectronic devices. To keep step with the miniaturization of next-generation devices, the constituent polymeric films of microelectronics need to meet requirements such as providing conformal, uniform, pinhole-free and ultrathin coatings. Molecular layer deposition (MLD), as an analogue to atomic layer deposition, is a layer-by-layer technique that utilizes sequential, self-limiting reactions of organic precursors to deposit films with one molecular unit at a time, which in turn allows for fine tuning of the position and concentration of various functionalities in the deposited film. Hence MLD can be a powerful method for deposition of polymer films used in semiconductor device fabrication. In this thesis, novel MLD processes are developed for fabricating ultrathin films and improving the film properties with applications in semiconductor manufacturing. The first part of this thesis explores the application of MLD films as chemically amplified photoresist materials. Acid-labile groups are embedded in the backbone of the precursor and incorporated into the photoresist film with a uniform distribution. Two methods of incorporating photo acid generator (PAG) are employed. The first method is to directly soak the PAG into the resist film after deposition and the second approach is to form in-situ polymer-bound PAG. By this novel synthetic approach, several polyurea films were deposited by MLD and tested for patterning, including an aromatic polyurea film with a soaked-in PAG, an aromatic polyurea film with an in-situ polymer-bound PAG, and an aliphatic polyurea film with soaked-in PAG. All these films were successfully deposited and characterized for both materials properties and resist response. Ellipsometry measurements show that the film thicknesses have a linear dependence on the number of MLD cycles. The presence of the urea linkage is confirmed by infrared (IR) spectroscopy, and x-ray photoelectron spectroscopy (XPS) show that the films are deposited with stoichiometric composition. Both of the aromatic films show cross-linking behavior under e-beam exposure, probably due to reaction at the aromatic rings. Moreover, the in-situ polymer-bound aromatic PAG has a lower activity than the soaked-in aromatic PAG, likely due to a lower photoacid yield. Finally, among the three MLD films studied, the aliphatic film performs best as a photoresist material and good sensitivity and resolution are achieved. To be applied in semiconductor device fabrication, polymeric thin films need to be thermally stable. Two approaches are investigated to improve the thermal stability of the MLD films. First, a series of cross-linked polyurea thin films are deposited by using multifunctional precursors. The cross-linked films show constant growth rate, urea chemical bonding, and stoichiometric compositions. More importantly, they exhibit higher film density and thermal stability compared to the non-cross-linked polyurea film. Second, a MLD process for depositing inorganic organic hybrid carbosiloxane films is developed. Characteristic MLD growth behavior such as a constant growth rate and saturation behaviors are observed with this process as well. Significant improvement of film stability is achieved with the carbosiloxane films. This thesis concludes with thoughts and perspectives on the future of MLD in semiconductor device fabrication.

For decades, silicon as a high quality solid material has been playing a major role in developing state-of-the-art micromechtutical sensors. In this work, the performance of silicon etching processes is evaluated, and utilized in creating devices for high precision sensing of mechanical forces. Emerging methods such as atomic layer deposition (ALD) and focused ion beam (FIB) processing are combined with established technologies in order to achieve extended functionalities. 
The thesis is focused on the fabrication of components from single crystal silicon. In the developed processes, anisotropic tetramethylanunonium hydroxide (TMAH) wet etching has a central role. The material behavior related to substrate specifications, doping level, and the effects inherent to thin films, is studied both during the fabrication stage and in the final structures. Furthermore, the intrinsic residual stress generation in mechanical devices is investigated. 
The results were applied in successful production of mechanical sensors with superior accuracy in acoustic wave detection. The specific features of silicon-on-insulator (SOD technology were utilized in the fabrication of released, membrane-type cantilevers that serve as highly sensitive microphones in photoacoustic spectroscopy. Moreover, AID thin film coatings were shown to be advantageous in manipulating the functional properties of resonant silicon sensors. A step towards further device miniaturization was taken by using FIB implantation as a wet etching mask to realize 3D nanoscale structures. 
Some of the developed cantilever sensors have already been implemented in commercial photoacoustic gas analyzers. In a broader perspective, the achievements related to stress control in mechanical silicon structures are of great importance in reliable production of various micro- and nanomechanical systems. All the methods employed in this work are compatible with integrated circuit (IC) processing, which enables combining the developed processes with electronic chips and the adaptation to industrial manufacturing.
 

The demand of smaller, higher capacity and higher performance devices in microelectronics has driven the necessity of uniform, conformal, and pinhole-free thin film production. Furthermore, the design toward more complex structures and higher aspect ratios requires the processes to be highly controllable, down to the nanoscale. Atomic layer deposition (ALD) is a powerful technique that produces thin films with these desired properties, through a series of alternating self-limited surface reactions. The self-saturated nature of the technique allows for precise thickness control at the atomic scale. Despite increasing interest in ALD, there is still a lack of understanding of the mechanisms behind the process at a molecular level. The nucleation and growth fundamentals are crucial for better control and development of the process and, hence, need to be systematically studied. Due to the vulnerability of the reactions to ambient conditions, ex situ analysis techniques alone may not provide complete information on the surface chemistries needed to elucidate the mechanisms governing the processes. In situ analysis techniques, which allow surface investigation without disruption from contaminants and other species, are required. Therefore, in this work we have designed and constructed various in situ systems for this purpose. The in situ systems are ALD reactors integrated with different analysis tools, able to operate as fully functional deposition system so as to replicate the actual conditions of typical ALD reactors. Through in situ X-ray photoelectron spectroscopy (XPS), we studied ALD of TiO2 at 100 °C using titanium tetrachloride (TiCl4) and water (H2O) on two different surfaces. The initial growth rate on hydroxyl-enriched silicon dioxide (SiO2) is found to be higher than on hydrogen-terminated silicon. The XPS results provide evidence of Si-O-Ti bonds on the SiO2 surface and Si-Ti bonds on the hydrogen-terminated Si surface, without a trace of interfacial oxide. However, a silicon oxide layer forms at the interface between Si and TiO2 after vacuum annealing, concurrent with the reduction of TiO2. The results hence suggest TiO2 as an oxygen source for silicon oxidation under these conditions. In addition, we studied ruthenium thermal ALD using a new precursor, bis(2,4- dimethylpentadienyl) ruthenium, and oxygen. The process is achievable at a low operating temperature of 185 °C. Variation in the exposure time and pressure of oxygen has significant effects on the nucleation, growth rate and composition of the deposited ruthenium films. We propose that the subsurface oxygen formation, which involves slow diffusion of oxygen, is a rate-limiting step in the RuO2 formation process. The crystal growth and structures of Ru and RuO2 deposited on amorphous SiO2 by the same ALD process were measured by ex situ and in situ synchrotron X ray diffraction. Interestingly, in situ XRD studies reveal that RuO2 films initially nucleate as metallic Ru crystallites. The hindered formation of subsurface oxygen in small nanocrystals is hypothesized as the cause that prohibits the growth of the initial oxide. Although metallic ruthenium films are textured with a (002) preference in the growth direction, RuO2 films nucleating on the metallic Ru nanoparticles have no preferential orientation. We also studied surface chemistries of Ru reactions during half ALD cycles via in situ synchrotron photoemission spectroscopy (PES). After long oxygen exposures, Ru oxide and carbon-oxygen species, which localize near the top surface, were detected. The peak intensities of these species noticeably decreased after reaction with the Ru precursor, indicating the reactions of Ru precursor with both O-Ru and O-C species. In brief, we fabricated and utilized in situ ALD/analysis systems, together with ex situ analysis tools, for studies of TiO2 ALD and Ru/RuO2 ALD. The studies not only demonstrate the power of the in situ systems for mechanistic studies, but also provide information on possible bond formation, surface reactions, and nucleation and growth mechanisms in the ALD processes.

This thesis explores several novel material systems and innovative de vice concepts enabled by nanowire technology. State-of-the-art fab rication techniques such as electron beam lithography and atomic layer deposition are utilized to achieve high control and quality in the device fabrication. The devices in this thesis are based on two main types of design geometries, lateral and vertical, each of which have strengths and weaknesses. The  rst part of the thesis describes the goals of future metal oxide semiconductor  eld-e ect transistors (MOSFETs) and discusses the ultimate scalability surrounding exper imental results for 15-nm-diameter InAs nanowires and how they com pare to other state-of-the-art transistors. The extracted on-resistance (Ron = 250 Ω•µm) and drive currents (Ion = 1250 µA/µm) are compa rable to state-of-the-art high-electron-mobility transistors (HEMTs) from MIT and quantum-well  eld-e ect transistors from Intel. The outstanding performance is mainly attributed to the reduced access resistance achieved through an n +-i-n + doping pro le. The extracted mobilities also agree well with state-of-the-art and theoretical predictions for extremely scaled devices. The second part of the thesis discusses how nanowires may be employed to enable III-V complementary metal oxide semiconductor (CMOS) digital logic. Nanowires enable the formation of both n-type semiconductors and p-type semiconductors, which are a requirement for CMOS, in a single nanowire and allow for integration on a Si platform. III-V MOSFETs are frequently employed for analog ap plications, but there is a disconnect regarding p-type devices, which are also required for digital logic. The individual segments of the nanowire are evaluated as well as the entire nanowire in an inverter con guration. This thesis then presents a strategy for matching the drive currents n- and p-type MOSFETs. The  nal part of the thesis deals with a family of devices that op erate according to principles fundamentally di erent from those of a traditional MOSFET, namely tunnel FETs (TFETs). There is a de mand for steep-slope devices such as TFETs to enable supply-voltage scaling to reduce the power dissipation. Although devices have demon strated are much improved as compared to their axial counterparts when nor malized to the largest cross-sectional area of the nanowire, assuming a vertical device geometry, illustrating the advantage of a core shell architecture. The dimensions of the InAs shells are below 15-nm and display clear quantization e ects revealed in low-temperature electri cal characterization

The increasing demand for energy as standards of living and population sizes rise across the globe motivates the development of scalable resources to meet the forecasted doubling of energy consumption. This challenge is further compounded by the need to reduce the CO2 emissions associated with our current level of energy consumption in order to stave off costly changes in the global climate. Solar energy promises renewable, carbon-free energy and a resource that is orders of magnitude larger than alternative sources. Capturing this solar energy with photovoltaic devices has become an increasingly economical means of energy production, but further development is needed to make solar energy conversion inexpensive and ubiquitous. One particular class of photovoltaics—the dye-sensitized solar cell (DSSC)—is especially appealing because it can be constructed with abundant, inexpensive materials and be engineered in a modular fashion for a wide array of product applications. Challenges remain, however, in order to make DSSCs more efficient, more economical, and more practical. The DSSC architecture is quite different than traditional—i.e. silicon and thin film inorganic—photovoltaics in that light harvesting and charge collection are decoupled. This is achieved by depositing a thin nanometer scale layer of light-absorbing dye molecules atop a high surface area nanostructured TiO2 anode. The consequence of this architecture, however, is an abundance of interfacial area at which deleterious charge recombination processes can occur. Further improvements in DSSC performance therefore require a thorough understanding of and high control over the dual-interface that exists between the dye layer and the electron-conducting anode and the dye layer and the hole-transporting material (HTM). In the first part of this work we describe the application of thin, sub-nanometer insulating metal oxide layers grown by atomic layer deposition (ALD) to the TiO2 anode for the purpose of slowing down the undesirable back-recombination of electrons injected into TiO2 by the light-absorbing dye molecules. We use the well-characterized insulator Al2O3 as a recombination barrier material and perform a comprehensive study of different parameters that impact how such barriers change device performance. In a solid-state DSSC we demonstrate the importance of the dye chemical structure and the anode fabrication process in dictating whether or not improvements achieved through the recombination suppression outweigh device current losses that result from the insertion of an insulating layer. We apply these lessons to a new barrier layer material, In2O3, that unlike Al2O3 has not been previously well-characterized but is less-insulating and can be grown at an extremely low growth rate, providing excellent control over the competing consequences of barrier layers. With In2O3 we are able to demonstrate some of the highest ever reported open-circuit voltages for this class of DSSC and we observe an unexpected change in behavior as the In2O3 barriers reach a certain thickness. In our third barrier layer study, we achieve a near-doubling in efficiency when Al2O3 is applied to a quantum dot-sensitized solar cell, a close analogue of the DSSC, wherein recombination more severely limits device performance. In the second part of this work, we combine experimental and computational techniques to study the dye/anode interface and the orientation of dye molecules adsorbed on a TiO2 surface. Near-edge X-ray absorption fine structure (NEXAFS) spectroscopy is used to measure the angles of individual bonds and in turn deduce the full geometry of adsorbed dye molecules. This result is compared to computational simulations using density functional theory-molecular dynamics (DFT-MD) of the same dye/TiO2 system. Our results show remarkable correspondence between the experimental and computational approaches and signify important consequences for understanding the dye/anode interface in DSSCs as well as adsorbate/substrate studies more generally.

In 2007 there was an important change in the architecture of nanotransistors - the build ing blocks of modern logic and memory devices. This change was from utilising ther mally grown silicon dioxide as a dielectric to so-called high-κ hafnium oxide dielectrics grown by atomic layer deposition. The first production logic devices of this era used a hafnium oxide dielectric layer deposited by thermal atomic layer deposition; using HfCl4 and H2O as the precursors. Present day fabrication makes use of hafnium oxide-based atomic-layer-deposited dielectric films. The latest nanotransistor devices utilise a third generation hafnium oxide-based dielectric material. This thesis examines hafnium oxide-based thin film dielectric materials prepared by thermal atomic layer deposition on silicon substrates. Specifically the enhancement of the dielectric response of hafnium oxide by the addition of other elements is examined. Two ternary materials systems were deposited by thermal atomic layer deposition and analysed: titanium-hafnium oxide and cerium-hafnium oxide. Hafnium oxide films were deposited to be used as measurement benchmarks. Cerium oxide films were also de posited and analysed in their own right as potential dielectric layers. The hafnium oxide and both ternary deposition experiments used (MeCp)2Hf(OMe)(Me) as the hafnium precursor. The titanium-hafnium oxide growth used Ti(iOPr)4 as a ti tanium source and the cerium oxide and cerium-hafnium oxide work utilised Ce(mmp)4 as a cerium source. Post-deposition specimen sets consisted of an as-deposited sample, a sample spike-annealed in N2 at 850 ◦C and a sample annealed for 30 minutes at 500 ◦C. These annealing regimes were performed to mimic typical gate-first and gate-last transistor processing steps. The compositions and thicknesses of the films were mea sured using medium energy ion scattering. The structure of the films was analysed by X-ray diffraction and Raman spectroscopy. Capacitance-voltage and current density field measurements were taken from fabricated MOS capacitor specimens to assess the dielectric response of the films. X-ray diffraction and Raman measurements showed that un-doped HfO2 had monoclinic crystallinity as-deposited and after the two annealing regimes. The dielectric constant and leakage current density, 17 and 1.7×10−7 A/cm2 at -1 MV/cm respectively, are consistent with values reported in the literature for HfO2 films. The addition of titanium suppressed the crystallinity of the material resulting in amor phous films in compositions with Ti0.3Hf0.7O2 titanium and above. The optimum elec trical results were recorded for the titanium-hafnium oxide material in the composition Ti0.5Hf0.5O2 which had a dielectric constant of 35 as-deposited and a leakage current density of 1.0×10−7 A/cm2 at -1 MV/cm. This composition of film demonstrated sim ilar values after the 500 ◦C/30 min anneal but both dielectric constant and leakage current density suffered after the 850 ◦C/spike anneal; 22 and 1.8 ×10−6 A/cm2 at -1 MV/cm respectively. Films with compositions of Ti0.1Hf0.9O2 demonstrated much lower dielectric constant and higher leakage current density, especially after heat treatment. The addition of cerium in a Ce0.11Hf0.89O2 composition was found to suppress crys tallinity as-deposited and then provoke a lattice-substitutional phase change to the metastable tetragonal/cubic phase after both types of heat treatment. This cerium activated phase change resulted in a molar volume modulation compared to un-doped HfO2. An increased dielectric constant compared to un-doped HfO2 of 31 was recorded for the 500 ◦C/30 min anneal with the 850 ◦C/spike anneal resulting in a lower value of 21. Leakage current density was 1.3 ×10−7 A/cm2 and 3.2 ×10−7 A/cm2 at -1 MV/cm respectively for the same anneals. Deposition with Ce(mmp)4 and water was found to result in cubic crystalline films across a growth temperature range 150–350 ◦C. The frequency dependency of the dielectric properties was found to be influenced by the crystallite size which was governed by the deposition temperature. The highest dielectric constant, 42, was measured for the 150 ◦C growth temperature with C-V measurements performed at 1 MHz. The two doped HfO2-based materials systems studied have demonstrated potential as dielectric materials for use in future nanoelectronic devices.

In this thesis we have presented theoretical investigation of nucleation during the incubation period of ALD of dielectric thin films on technologically important substrates: III-V semiconductors and silicon nitride. Atomic-scale simulations in the framework of DFT were conducted on the bulk structures, surface models and gas phase molecules. Ab initio thermodynamics provided a bridge between zero-temperature, zero-pressure DFT and real experiments. We have analysed and described the interaction of the precursors with the considered substrates: starting from initial adsorption of the gaseous molecule, through its transformation in contact with native oxides, until formation of surface intermediates that can be lost as byproducts. The structure and chemical state of surface intermediates were described in detailed. Some kinetic aspects of the incubation mechanisms were also revealed. Based on the trends arising from the initial computations, we have developed a comprehensive model that allows us to compare the operation of different classes of precursor chemicals and assess their properties for self-cleaning ALD. On the example of TMA we identified two separate factors governing the clean-up effect: formation of the metal oxide as the driving force and affinity of the precursor ligand to the III-V oxide substrate as the ancillary force. ‘Clean-up’ of an oxide film is shown to strongly depend on the electropositivity of the precursor metal, and thus always results in formation of dielectric film from native oxide. However, self-cleaning ALD does not necessary result in substrate-enhanced growth, as the clean-up effect can be spread over a few cycles. The choice of ligand determines bonding at the interface and the overall type of clean-up. The predominant pathway for a metalloid oxide such as arsenic oxide is reduction, producing volatile molecules or gettering oxygen from less reducible oxides. An alternative pathway is non-redox ligand exchange, which allows non-reducible oxides to be cleaned-up. TMA was shown to be a superior precursor for self-cleaning ALD, considering the good properties of Al2O3 film as an interfacial layer in the transistor gate stack. Thanks to the lability of the methyl ligand that has affinity to both Lewis basic and acidic sites, a number of mechanisms are exothermic and competitive. We proposed TMA clean-up byproducts, interface bonding and the most thermodynamically abundant surface intermediates. TDMAH was shown to have slightly different properties to TMA. However we proved that this precursor is also a clean-up reagent, with a number of exothermic reactions participating. The dimethylamide ligand was shown to be resistant to oxidation when intact, however very prone to decomposition and subsequent oxidation. Some of the decomposition intermediates can migrate from Lewis acidic to Lewis basic site as in the TMA case. What is characteristic in the operation of alkylamide precursors is that their self-cleaning behaviour is increased with temperature and results in accumulation of As suboxides at the interface. Our kinetics study has revealed the reason for source of this behaviour. In general, we predict that the investigated methyl precursors are the best reagents for deposition of dielectrics and performing clean-up. Unfortunately, most of them are very unstable compounds. Having identified Mg precursors as the most effective at clean-up, one possible avenue for future research would be to evaluate the performance of some specific Mg precursor molecules on III-V surfaces. For instance, Mg(C5H5)2 is a commonly-used precursor in ALD [147]. It would be interesting to look in particular at whether this precursor too can carry out ‘clean-up’ and improve the electrical properties of the interface. Our study showed also that the chloride family cannot act according to the same mechanisms as organometallic and metalorganic precursors, but according to different mechanisms resulting in self-cleaning ALD that could also be investigated further. Successful ALD relies on growth reactions being faster than side-reactions. For these reasons, calculating the kinetics of self-cleaning ALD reactions could contribute to the research in this subject. However, DFT is not suitable for cases where detailed treatment of electron correlation is necessary. This includes some types of transition state (TS) that we study within this thesis to evaluate the kinetics of the ‘clean-up’ effect. Therefore, it would be profitable in the future to benchmark the DFT barriers reported here against more accurate calculations. The eigenvalue-following Trust Radius Image Minimization algorithm provides more accurate geometries and energetics of transition states. These geometries, at the DFT level, can be used as an input for the computationally-heavy configuration interaction (CI) calculations that account for electron correlation more completely than DFT and yield high-quality energy. It is worth mentioning that the heterodeposition (deposition on the initial substrate) mechanisms investigated here could serve as an input to the multi-scale simulation, e.g. kinetic Monte Carlo [148], of the actual film growth corresponding to an experimental timescale. The major aim of this thesis was to assist and to complement experiment and as a result to obtain better understanding of the mechanisms occurring during initial ALD cycles of dielectric films. This goal was definitely achieved. Some of the suggested here mechanisms were subsequently confirmed, in the case of TMA clean-up by Tallarida and co-workers, while other mechanisms provided explanations for prior experimental findings, in the case of alkylamides by Gougousi and co-workers. What is more, the clean-up by products reported here may be detectable with, e.g. FTIR or mass spectrometry experiments of sufficient surface sensitivity. Our calculations assisted analysis of the growth details on silicon nitride substrates. Theoretical calculations accounted for the different starting substrates and provided possible explanations for the experimentally-revealed anomalous growth step. 

As the demand in high power and high frequency electronics keeps increasing, standard semiconductors, such as silicon, show their limits. Approaches based either on new architectures or on wide band gap materials should allow to overcome these limitations. Among these, diamond offers superlative properties such as a wide band gap, a high breakdown electric field, an outstanding thermal conductivity and high carriers mobility. Thus, it is foreseen as the ultimate semiconductor that can not be surpassed by others. Besides, recent progresses in substrate and epilayer growth allow to consider seriously devices based on diamond semiconductor. Nevertheless, it also suffers from limitations, especially the difficulty to n-dope and the high ionization energy of the boron p-type dopant that results in a low carrier concentration at room temperature. Innovative solutions relying on 2D gas or/and field effect ionization have been proposed to overcome this problem. This PhD work was focused on two of these solutions, which might address different sectors of power electronics. On one hand, high frequency applications are aimed at using boron doped delta-field effect transistors involving a thin highly doped layer between two intrinsic layers, resulting in a conduction combining a high mobility (due to a confinement-induced delocalisation of carriers away from the ionized impurities) with a large carrier concentration (due to metallic behavior). In this work, the growth of nanometric sized delta-layer using an in-situ etching back technique was demonstrated using SIMS and TEM analysis and confirmed by electrical measurements. Then, the temperature dependence of electrical properties of several nanometric scaled delta boron doped layer were investigated experimentally (by Hall effect and four probe measurements) and theoretically over a large temperature range (6 K < T < 500 K). The samples can be classified in two categories with distinct behaviours: • Two samples showed an insulating non metallic conduction dominated by a hopping mechanism with an anomalous exponent of 0.7 which has already been observed in thin metallic films but is not yet well understood. • A metallic conduction was found in a wide variety of delta-layer thickness ranging from less than 2 nm to 40 nm (with sheet carrier from 1014 cm−2 to 5⇥1015 cm−2). Surprisingly, the same mobility value of 3±1 cm2/Vs was measured independently of the delta-layer thickness. Three dimensional and two dimensional mobility models were considered to explain this phenomenon. Scattering mechanisms by ionized impurity (3D) was found to be the best candidate for such low mobility. The model used in this work gives a relative agreement with this value even if a discrepancy remains. Future works are needed to fully understand this particular value. An exact calculation will certainly need a better knowledge of the valence bands (dispersion and/or boundary with localized states) at energy levels close to the Fermi level of such metallic B-doped diamond. Finally, the thinnest metallic layers (< 2 nm with pS ⇠1014 cm−2) reported in this work did not exhibit the expected improvement of the mobility (μ=3 cm2/V.s). The mobility was too low to consider that high frequency diamond delta field effect transistor are able to compete with existing AlGaN/GaN HEMT showing a mobility of 2200 cm2/V.s. On the other hand, high voltage applications are aimed at with metal oxide semiconductor field effect transistor (MOSFET) where the conducting or insulating behavior of the channel is based on the electrostatic control of the band curvature at the oxide/semiconducting diamond interface. During this PhD project, metaloxide- semiconductor (MOS) structures with aluminium oxide (Al2O3) as insulator and p−type (100) mono-crystalline diamond as semiconductor were fabricated and investigated by capacitance versus voltage C(V ) and current versus voltage I(V ) measurements. The aluminum oxide dielectric was deposited using low temperature atomic layer deposition on an oxygenated diamond surface. The C(V ) measurements demonstrate that accumulation, depletion and deep depletion regimes can be controlled by the bias voltage. In the accumulation regime, a frequency dispersion of capacitance value was noticed and has been attributed to the high series resistance of diamond but also to defects involved in a non ideal MOS structures. A band diagram was proposed to explain the surprisingly high leakage current flowing in accumulation regime: the barrier (at the SC/oxide) for holes seems to be very low (⇠ 0.3 eV). For positive voltage (larger than the threshold voltage) a MOS structure is expected to be in the inversion regime only if minority carriers could be provided to the channel. If not, the depletion of the semiconductor will continue until reaching the deep depletion regime. In the case of diamond which has a very large band gap, minority carriers could be thermally generated only by using very low measurement frequency (f n 1 Hz). Thus, the observation of deep depletion regimes in some samples is in accordance to the measurement frequency used and is indicative of a sufficiently clean oxide/diamond interface. It therefore opens the route for the fabrication of a diamond MOSFET in which the electrons will be provided to the channel by n-type drain and source. Nevertheless, a particular attention has to be paid to the oxide investigation in order to improve its quality (insulating properties, barrier for holes) as it will be a key point for the future diamond MOSFET. 

Consumer-driven demand for affordable, smaller and more sophisticated devices has driven advances in nanoscale engineering. Many of these products, whether they are consumer electronics, fuel cells or photovoltaics, are comprised of significant quantities of scarce, expensive and sometimes toxic materials. Manufacturing affordable and more environmentally benign products on decreasing length scales requires more economical use of scarce resources and the exploration of alternative and abundant materials. Platinum, as one example, is an outstanding catalyst used in automotive, fine chemical and fuel cell applications; however, it is extremely scarce and costly. Placing platinum only where needed and in small quantities to achieve desired performance may help economize its usage. Because nanoscale feature size and spacing influence catalytic behavior, being able to control these parameters may enable the careful engineering of more powerful platinum catalysts. Another scarce and expensive material whose usage follows rising consumer demand for electronics and photovoltaics is indium. Indium tin oxide is the industry standard transparent conducting oxide (TCO) for high-end electronics. The engineering of high-quality yet poorly-understood indium-free alternative metal oxides using earth-abundant and inexpensive materials may further the development of more powerful and advanced devices. In this dissertation, atomic layer deposition (ALD), a promising ultra-thin film growth technique, is used to selectively deposit small amounts of platinum and to alloy zinc tin oxide in order to address these concerns of scarcity and rising costs. ALD consists of self-limiting, gas-surface half-reactions separated by inert gas purges which allows for highly controlled growth of nanoscale materials. Because ALD is based on surface reactions, selectively blocking nucleation sites by passivating growth surfaces allows for the deposition of material only where desired. The spatial deposition of platinum using ALD is demonstrated in two studies. In the first, a watersoluble polymer, polymethacrylamide (PMAM) acts as a resist to ALD and allowed the deposition of very small features of platinum. In the second study, surface passivation by means of controlling defect site density in self-assembled monolayers (SAMs) also serves as a growth template to control the aerial density and size of platinum nanoparticles deposited via ALD. The highly tunable nature of the chemistry of each precursor half-reaction in ALD enables the careful alloying or doping of materials in any desired ratio. A range of compositions for the indium-free TCO alternative, zinc tin oxide (ZTO), is explored by varying the dosing cycle ratios of tin oxide (SnOx) and zinc oxide (ZnO) in each ALD super cycle. Because the effects of ZTO composition on conduction behavior and optical properties are not well understood, having nanoscale control over how the constituent metal oxides grow on one another is an effective approach to learning more about these properties. As a result, ALD is explored as a flexible system to allow for compositional sampling and variation of film lamination in ZTO. The characteristics of ZTO ALD and the structural and optical properties of these films were shown to vary significantly as a function of degree of lamination and composition. These findings suggest that having nanoscale control over these film characteristics via ALD can lead to further understanding of ZTO. The ALD of ZTO provides an excellent system for the nanoengineering of this material as an indium-free TCO or buffer layer in solar cells. 

The minimum feature size, including gate oxide thickness, of complementary metal oxide semiconductor field effect transistors (CMOSFETs) has decreased exponentially until now. However, with this method, scaling slows down at the 90 nm node as SiO2 runs out of atoms and further scaling is limited due to the increase of gate-leakage current. To continue the scaling of transistors, various high-k oxide materials have been studied intensively for the past few decades. The semiconductor industry has already converged on Hf-based oxides for the first generation CMOS products featuring high-k gate dielectrics and metal gate electrodes. However, even though various metal materials were already used on Hf-based oxides, there are still several crucial problems that need to be considered; an effective work function (EWF) modulation for adjusting the threshold voltage (Vt) of transistors and further scaling of equivalent oxide thickness (EOT), which was approximately 1.0 nm for the first generation high-k/metal gate device. Continued gate length (Lg) scaling for the 32 nm and beyond with a planer structure requires sub-nm EOT to suppress short-channel effects. Fully depleted device structures, such as FinFET or extremely thin SOI (ETSOI), improve short-channel control and thus relax the requirements for EOT scaling. However, the insertion point of such device architectures is expected to be the 22 nm and beyond and sub-nm EOT may be still required at those advanced technology nodes. To meet continued EOT scaling until the turning point in device architectures, mainly three possible EOT scaling approaches are studied in this work: (1) a high-k material with k-value greater than that of HfO2 (so-called “higher-k”), (2) the physical thickness reduction of interfacial layer (IL) (so-called “scavenging”), (3) suppression of low-k dielectric layer between metals and dielectrics (so-called “deadlayer effect”). Firstly, the relation between the permittivity and microstructures of atomic layer deposited Hf1-xSixOy (HfSiO) thin films with different Si concentrations as a function of post-deposition annealing (PDA) temperature was investigated. The PDA at high temperature results in the separation of crystallized HfO2 phase from the much higher Si- containing amorphous-like matrix. Tetragonal phase HfO2 formation with higher permittivity than the monoclinic HfO2 phase is induced with an appropriate Si concentration in the film (~10–20%). In the crystallized HfSiO film, the Si concentration in the phase-separated HfO2 (mainly consisting of HfO2) could be controlled by PDA temperature, which determines the degree of phase separation. The increased PDA temperature reduces the Si concentration in the phase-separated HfO2 which induced monoclinic phase formation. Therefore, the PDA temperature for maximized permittivity of the crystallized HfSiO films (maximized tetragonal phase portion in the film) depends on the Si concentration of the HfSiO film in the as-deposited state. However, considering the maturity of Hf-based high-κ gate dielectrics, scaling SiO2-based IL in conjunction with Hf-based oxides may be more practical in meeting the requirements for the 22 nm technology node and beyond. Secondly, lanthanum (La)-incorporated TiN metal gates, such as TiN/La/TiN (TLT) and TiLaN (TLN), on HfO2/Si substrates were investigated focusing on the flat band voltage (VFB) modulation for nMOS and IL scaling to almost zero. The maximum VFB modulation value of the TLT/HfO2/Si stack was −423 mV compared to the VFB of the TiN single metal case, which is superior to that of TLN (−247 mV). This is because the TiN barrier layer in the TLT metal stack prevents interfacial oxidation. Both TLT and TLN gate metals effectively shrink the IL thickness to values below 0.5 nm. In the case where the TLT metal gate was annealed at 600 o C for 30s, the IL thickness was almost zero and the equivalent oxide thickness (EOT) was decreased to 0.8 nm even though the maximum temperature was limited to 600o C. However, the La-incorporated TiN metal gates can not adopt for pMOS due to their low work function, and another method must be pursued to scale the EOT of p-type MOSFET. Thirdly, the influences of RuO2 metal gate on the dielectric performance of high-k HfO2 film on Si substrate were examined. Dielectric materials with a higher-k value also suffer from a dead-layer effect that the effective dielectric constant decreases with decreasing thickness, which becomes even more serious as the bulk k value increases. The EOT of HfO2 film can be scaled down by ~ 0.5 nm in the EOT range from 0.8 to 2.5 nm compared with the standard Pt gate case by using the electrically conducting RuO2 without sacrificing any other performance of the MOS capacitor. RuO2 is one of the rare materials, which contain polarizable ions, high electrical conductivity, and high work function (WF) which is necessary for the p-type MOSFET. This was attributed to the suppression of the dielectric dead-layer effect at the HfO2/RuO2 interface due to the possible ionic polarization of RuO2 within the screening length of the electrode. In addition, the estimated work function of RuO2 on HfO2 is ~ 5.0 eV suggesting the appropriateness of RuO2 for p-MOSFET. Finally, RuO2 metal gates were fabricated by a reactive sputtering method under the different O2 gas ratio. For the given sputtering power of 60 W, ~ 13 % O2 ratio was the critical level below or over which RuO2 film have hyperstoichiometric and stoichiometric compositions, which resulted in the effective work function difference by ~ 0.2 eV. The stoichiometric RuO2 film imposed almost no damaging effect to the underlying SiO2 and HfO2 gate dielectrics. RuO2 gate decreased the equivalent oxide thickness by ~ 0.5 nm and leakage current by ~ two orders of magnitude compared with the Pt-gated samples.

During the last several decades, titanium nitride (TiN) has gained much interest because of its low resistivity, chemical inertness and compatibility with complementary metaloxide- semiconductor (CMOS) technology. Thin films of TiN are commonly used as diffusion barrier and gate material for CMOS devices. Mostly sputtering and chemical vapor deposition (CVD) techniques were employed to deposit the TiN films. However, to achieve thin conformal films with accurate thickness control and excellent step coverage, atomic layer deposition (ALD) has become an ideal choice for making such high quality films in the thickness range of a few tens of nanometers. In this work, the growth mechanism and electrical properties of ALD TiN films are investigated. The films are grown on SiO2 substrate using the TiCl4/NH3 chemistry in the pressure range of 2.6−3.2 ×10-2 mbar and temperature range of 350−425 oC. The growth of the films is monitored by in situ spectroscopic ellipsometry (SE). The films are characterized by other complementary analysis techniques, including atomic force microscopy (AFM), high-resolution scanning/transmission electron microscopy (HRSEM/ TEM), X-ray fluorescence (XRF) and X-ray photoluminescence spectroscopy (XPS). We fabricated test structures to characterize electrical properties of the thin films. The main results of this research are presented in this thesis. In Chapter 2, the application of SE in studying the optical functions and measuring film thickness of TiN thin films is presented. We apply the Drude−Lorentz model to parameterize the dielectric functions of TiN. The film thickness obtained by SE is compared with the results obtained from HR-TEM/SEM and XRF measurements. The results show good agreement in a wide thickness range (up to 40 nm), indicating the reliable applicability of the SE technique and the Drude−Lorentz model. SE is employed to in situ study the growth (Chapter 3), the resistivity (Chapter 4) and the real-time oxidation of ALD TiN films (Chapter 5). In Chapter 3, the growth mechanism of ALD TiN films on SiO2 substrate is investigated. We have observed that the growth obeys Stranski−Krastanov model. The growth starts with a 2D mode (continuous layers) followed by a 2D-3D transition (onset of islanding). This transition is temperature independent and occurs at a film thickness of about 0.7 nm. This equals 3 monolayers of TiN. The growth of the 3D islands (on the continuous layers) eventually leads to their coalescence which occurs at 2.5 nm and 3.5 nm for the growth at 350 oC and 425 oC, respectively. Before the coalescence, new nuclei are constantly formed during the growth. Hereafter, the film grows with a constant (ALD) growth rate of 0.02 nm/cycle at both temperatures. In Chapter 4, the resistivity, temperature coefficient of resistance (TCR) and field effect in thin TiN films down to sub-nanometer scale are presented. The resistivity of TiN is determined by both spectroscopic ellipsometry (SE) and electrical test structures. The latter includes circular transfer length method (CTLM) and linear test structures. The results show that the resistivity increases significantly with decreasing film thickness. We compared the resistivity values obtained by the optical and electrical techniques. For films thicker than 4 nm, the values show good agreement. However, for films thinner than 4 nm, much higher values are found from the electrical measurements. This is attributed to the scattering effects at interfaces and grain boundaries that cannot be fully taken into account by the optical measurements. The TCR of TiN is determined from the electrical measurements in temperature range of 25−150 oC for different film thicknesses (0.65– 8.3 nm). With decreasing film thickness, the TCR values change sign from positive to negative. This change occurs at a thickness of about 2.5 nm. This effect is attributed to the metal-semimetal transition in these films. In the last part of this chapter, the electric field effect in ultra-thin TiN films in the metallic and the semimetallic states is presented. Upon electrical field, a change of the conduction current up to 22% is found for the films in the semimetallic state. In Chapter 5, the influence of native oxidation on electrical behavior of thin ALD TiN films is reported. For films thinner than 5 nm, this oxidation changes the I-V behavior from linear to non-linear. It shows an aging effect that causes an irreversible degradation of TiN. The native oxidation does not influence the linearity of the I-V characteristics for films thicker than 5 nm. Nevertheless, it reduces the electrical thickness of TiN. The kinetics and mechanism of thermal oxidation in the temperature range of 325−425 oC of films with a thickness of 5 and 15 nm are intensively studied. The oxidation in oxygen ambient was carried out directly after the TiN deposition in the same reactor (without vacuum break) and in situ spectroscopic ellipsometry monitoring was used. The oxidation of the films can be described as a four-stage process: (1) surface oxidation, (2) diffusion of oxygen along grain boundaries and lateral oxidation and sharpening the grains, (3) rapid oxidation of the sharp grains, and (4) diffusion controlled oxidation. Finally, in Chapter 6, the generation of atomic hydrogen (H) is presented. Atomic hydrogen is made by the dissociation of molecular hydrogen (H2) upon collision with a tungsten (W) filament kept at a high temperature (T ≈ 1600−1900 oC). We have demonstrated the ability to create atomic hydrogen and to introduce it in short pulses by experiments on etching of tellurium (Te) films in the reactor chamber. The generated atomic hydrogen (H) is further utilized to explore its impact on surface reactions in the TiCl4/NH3 precursor system. Atomic hydrogen is introduced in pulses additionally to TiCl4 and NH3 with different pulse sequences. The results show that for the TiCl4/NH3/H sequence, there is no influence on the process compared to the ALD without H-pulses. The growth rate remains at 0.02 nm/cycle and the oxygen content - at 3−5 at%. For the TiCl4/H/NH3 pulse sequence, the growth rate decreases to 0.01 nm/cycle and the oxygen content increases to 30−35 at%. This indicates the reduction of Ti-Cl surface groups by H and the creation of very reactive Ti surface. This surface can be easily oxidized by residual gases from the reactor and process gases. TiCl4/H pulses only result in no growth after the formation of approximately one monolayer. 

Atomic Layer Deposition of Platinum: from surface reactions to nanopatterning Platinum is a material that finds many applications in the fields of nanoelectronics and catalysis due to its catalytic activity, chemical stability, and high work function. The thin film deposition technique of atomic layer deposition (ALD) is gaining increasing interest for the deposition of Pt ultrathin films and nanoparticles, since it is able to deposit on demanding surfaces such as high-aspect-ratio structures and porous materials. In this dissertation, ALD of Pt was studied, aimed at the development of a novel bottom-up nanopatterning approach. Conventional patterning by lithography involves resist-films and lift-off steps that may yield compatibility issues with the envisioned nanoscale building blocks of future nanodevices, e.g. nanowires, carbon nanotubes, and graphene. The main goal was to develop a nanopatterning approach that enables direct and local fabrication of high-quality nanostructures without the need for additional lithography steps. Since ALD film growth depends critically on the properties of the surface, it is possible to chemically tailor the surface properties to achieve area-selective deposition. For the development of the nanopatterning technique, detailed understanding of the surface reactions of the ALD processes of noble metals turned out to be crucial. The reaction mechanism of Pt ALD was studied by evaluating which surface reactions take place at the catalytically active Pt surface during ALD, based on analogous surface reactions reported in surface science literature. This study led to new insights into the surface reactions that take place during the growth, the saturation of the half-reactions, and the temperature dependence of the process. Inspired by the conclusions drawn from the reaction mechanism study, an approach for plasma-assisted ALD at low substrate temperatures was developed. It was demonstrated that this new process enables the deposition of Pt at temperatures down to room temperature. Consequently, the Pt can be deposited on various temperature sensitive substrates such as polymers, textile and paper, which significantly broadens the possibilities for applications of Pt ALD. Furthermore, the nucleation behavior of Pt ALD was studied using spectroscopic ellipsometry and transmission electron microscopy. It was established that the pressure employed during the O2 half-reaction of the ALD process governs the nucleation behav200 ior, which can be exploited for controlling the nucleation of the Pt. This control enables nanoparticle deposition, thin film deposition with minimal nucleation delay, and areaselective ALD for nanopatterning. The developed nanopatterning approach is based on a combination of ALD with electron beam induced deposition (EBID). EBID is a direct-write patterning technique with nanometer scale resolution but its main drawback is that it gives material of poor quality. The newly developed approach comprises the deposition of a thin seed layer by EBID, followed by area-selective ALD. It was established that this so-called direct-write ALD technique yields high-quality Pt material (∼100% pure, 12 μΩcm), and an enhanced throughput comparable to that of electron beam lithography (EBL), while it allows for patterning of nanoscale line deposits of only ∼10 nm in width. To validate whether direct-write ALD is suitable for contacting applications, it was demonstrated that contacts can be patterned on multi- and single-walled carbon nanotubes. Additionally, it was evaluated whether direct-write ALD is a suitable technique for the fabrication of carbon nanotube field effect transistors (CNTFET). CNTFETs were synthesized by patterning of Pt contacts using direct-write ALD on single-walled carbon nanotubes. It was demonstrated by electrical characterization that these devices behave as a p-type transistors. In conclusion, in this work a novel bottom-up nanopatterning approach has been developed that is completely resist-free, and is especially suitable for the patterning of contacts on sensitive nanomaterials. In addition, the reaction mechanisms studies led to atomic level understanding of the surface reactions of Pt ALD, and thereby will contribute to the use of Pt ALD in a wide variety of applications. 

Atomic layer deposition (ALD) is a chemical gas phase deposition method to grow thin films which are highly uniform and conformal over large and complex substrate areas. Film growth in ALD is precise, remarkably repeatable, and combined with unparalleled control of the film thickness. These inherent properties make ALD an attractive method to deposit thin films for advanced technological applications such as microelectronics and nanotechnology. One material group in ALD which has matured in ten years and proven to be of wide technological importance is noble metals. The purpose of this study was to investigate noble metal oxide film growth by ALD. The ALD of noble metal oxides has been very limited compared to the noble metal growth. Another aim was to examine noble metal film deposition at temperatures lower than required in the earlier ALD noble metal processes. In addition, the selection of noble metals that can be grown by ALD was expanded with osmium. The results of the study showed that oxides of iridium, rhodium, platinum, and palladium can be deposited from the common noble metal precursors using ozone as the reactant at temperatures below 200 􀁱C. The development of ozone-based ALD noble metal oxide processes led further on to the low temperature deposition of noble metals by adding a reductive molecular hydrogen step after every oxidative ozone step. The noble metal deposition via noble metal oxide growth was achieved at lower temperatures than required with the common oxygen-based ALD noble metal processes. Film growth rates, resistivities, purities, and surface roughnesses resulting from the studied noble metal oxide and noble metal processes were reasonable. The processes showed some shortcomings but offer an alternative thermal ALD pathway to deposit noble metals and noble metal oxides compared to the oxygen-based ALD processes. Keywords: atomic layer deposition, ALD, noble metal oxide, noble metal, thin film, ozone 

Atomic layer deposition (ALD) is a thin film deposition method based on self-limiting reactions so that it enables to deposit high quality and ultra-thin films on complex structures with uniform thickness controlling thickness and chemical composition of films. ALD is available at low temperature around 100°C depending on precursors and co-reactants, which makes it possible to use plastic or biological substrates. The recent miniaturization in electronic devices and development in analytical tools accelerates the great interest in ALD and expands its applications from traditional semiconductor industries. Therefore, we made use of ALD to improve the performance of dye-sensitized solar cells (DSSCs) in our studies because DSSCs contain several interfaces based on mesoporous structure which need to engineer them for efficient electron transport system.

We synthesized ALD TiO2 on quartz fibers (QF) for the first time and investigated ALD TiO2 phase transition on QF upon annealing temperatures. Conformal and thin ALD TiO2 film is very stabilized even at 1050 °C and highly photocatalytic due to stable anatase TiO2 phase. For DSSCs, we employed coated QF as a light scattering layer for DSSCs and it showed larger diffused reflectance than conventional light scattering layer improving power conversion efficiency of DSSCs