This thesis deals with the development and application of Monte Carlo simulations to study electron transport in bulk GaN in the wurtzite crystal structure and the properties of field effect transistors made from the material. There is a particular emphasis on transport in the high electric field regime and transistors operating at high voltages. The simulation model includes five sets of non-parabolic conduction band valleys which can be occupied by electrons during high field transport. The effects on electron transport of impurities and the relevant phonon scattering mechanisms have been considered. Results for electron transport at both low and high electric field are presented and compared with the properties of GaN in the zincblende structure, of other group-III nitride semiconductors, and of GaAs. The dependence of the transport properties on the material parameters is discussed and also with regard to the temperature, donor concentration and electric field magnitude and direction. The transport properties of electrons in wurtzite GaN n+-i(n)-n+ diodes are also explored, including the effect of the upper valleys and the temperature on hot electron transport. Simulations have also been carried out to model the steady-state and transient properties of GaN MESFETs that have recently been the subject of experimental study. It has been suggested that traps have a substantial effect on the performance of GaN field effect transistors and we have developed a model of a device with traps to investigate this suggestion. The model includes the simulation of the capture and release of electrons by traps whose charge has a direct effect on the current flowing through the transistor terminals. The influence of temperature and light on the occupancy of the traps and the /- V characteristics are considered. It is concluded that traps are likely to play a substantial role in the behaviour of GaN field effect transistors. Further simulations were performed to model electron transport in AlGaN/GaN hetero-junction FETs. So called HFET structures with a 78 nm Alo.2Gao.8N pseudomorphically strained layer have been simulated, with the inclusion of spontaneous and piezoelectric polarization effects in the strained layer. The polarization effects are shown to not only increase the current density, but also improve the electron transport by inducing a higher electron density close to the positive charge sheet that occurs in the channel.

O trabalho anteriormente realizado pelo nosso grupo, onde foram simulados defeitos de nitrogênio em nanotubos de carbono, apresentou resultados interessantes relativos às energias e propriedades eletrônicas. A interpretacão dos resultados teóricos obtidos levou à proposta da Divacância rodeada por 4 Nitrogênios como estrutura mais estável para o nitrogênio tipo piridina, em constraste com aquela proposta pelos experimentais, uma Monovacância rodeada por 3 Nitrogênios. Os cálculos das propriedades eletrônicas da Divacância reproduziram as medidas experimentais na investigação de sensores de amônia. Dessa forma, como informação adicional na determinaçã da estrutura mais estável, o presente trabalho investigou as propriedades vibracionais daqueles sistemas que apresentaram menor energias de formação. Foram calculadas as freqüências vibracionais dos seguintes três defeitos: Nitrogênio Substitucional (1N), Monovacância rodeada por 3 Nitrogênios (3NV) e Divacância rodeada por 4 Nitrogênios (4ND) e comparadas com os resultados para os tubos puros. Utilizou-se a aproximação de supercélula, com 140 átomos para um tubo metálico (5,5) e 160 para um tubo semicondutor (8,0). Como o objetivo é identificar as características de cada sistema, focalizou-se na comparação dos valores das freqüências Raman ativas mais intensas. Os cálculos foram realizados com o código SIESTA, utilizando DFT com o formalismo dos pseudopotenciais e a aproximação GGA-PBE. As freqüências foram obtidas pelo Método Direto pelo mesmo programa. Os resultados mostraram diferenças quanto à quebra de degenerescências, que ocorre devido à quebra da simetria do sistema puro e quanto à mudança dos valores das freqüências dos modos. Como característica geral, os defeitos fazem com que as freqüências da banda mais baixa de energia do espectro Raman sofram shifts negativos, ou seja, afastam os picos para energias mais baixas. O modo de freqüência intermediária sofre um shift positivo e os modos da banda G voltam a apresentar valores negativos. Os splittings, bem como os valores numéricos dos shifts variam conforme o tipo de defeito e o tipo de sistema dopado (armchair ou zig-zag). Apesar de não apresentar diferenças consideravelmente grandes para os valores de shifts e splittings entre os defeitos, o comportamento qualitativo distinto para os modos RBMs é uma boa ferramenta para a diferenciação desses defeitos através de espectroscopia vibracional. / A previous work developed in our own group on which nitrogen defects on carbon nanotubes were simulated presented very interesting results regarding the energetics and the electronic properties. The interpretation of the theoretical outcomes led us to propose the Divacancy surrounded by 4 nitrogen atoms as the most stable structure for a pyridine-like nitrogen, in contrast to the one proposed by the experimentalists, namely the Monovacancy surrounded by 3 nitrogen atoms. Calculations of the electronic properties of the Divacancy have reproduced the experimental data. In this way as additional information for determining the actual most stable structure the present work investigated the vibrational properties of those systems that showed the lowest formation energies. We performed the calculations of the vibrational frequencies for the following three defects: Substitutional nitrogen atom (1N), Monovacancy surrounded by 3 nitrogen atoms (3NV) and Divacancy surrounded by 4 nitrogen atoms (4ND). Then the frequencies were compared to those ones from the pure tubes. We used the supercell approximation with 140 atoms for a (5,5) metallic tube and 160 for a (8,0) semiconducting tube. Since the present work aims to identify the main features of each system we focused on the comparison of the values of the strongest Raman active modes. All the calculations were carried out by the SIESTA code, using DFT with the pseudopotential formalism and GGA approximation. Then the frequencies were evaluated using the Direct Method. The results showed differences on the degeneracy splittings, which are caused by the symmetry-breaking due to the introduction of defects, and also differences on the shifts of the numerical values of the frequencies. As general feature, the defects caused the low band frequencies modes of Raman spectrum to have a negative shift, i.e. they push the peaks further to lower energies. The intermediate mode shifts positively and the G band modes show negative shifts again. The splittings as well as those shifts change depending on the type of the defect and the type of the doped system (armchair or zig-zag). Although not showing significant differences for shifts and splittings between the defects, the qualitatively distinct behavior for RBMs modes is a good tool to tell them apart using vibrational spectroscopy.

Dans le cadre d'une étude sur les dispositifs à nanocristaux (NC) de silicium, tels que les mémoires flash à nanocristaux et le transistor à un électron, ce travail de thèse a pour objectif de modéliser avec précision le transport d'électrons par effet tunnel entre deux nanocristaux ; cela impose de tenir compte de l'élargissement des niveaux d'énergie électronique dans les NC induit par le couplage entre les électrons et les phonons. Les modes de vibration des NC de silicium de tailles variables ont dans un premier temps été calculés en utilisant l'Adiabatic Bond Charge Model (ABCM). Les résultats obtenus présentent une très bonne concordance avec les spectres Raman expérimentaux. Les densités d'états (DOS) des nanocristaux de grande taille ont été comparées avec la DOS du silicium massif. Il apparaît que les DOS sont globalement très proches, excepté dans des gammes de fréquences spécifiques où des modes de surface, inexistants dans le cristal massif, ont été identifiés. L'interaction électron-phonon a alors été prise en compte par un calcul de fonctions spectrales. Les fréquences d'interaction électron-phonon obtenues sont très grandes par rapport aux fréquences de transfert tunnel : l'interaction avec les phonons suffit donc à garantir un transport séquentiel. Le transfert tunnel d'un nanocristal à l'autre a été modélisé grâce à ces fonctions spectrales. Le courant à travers un dispositif comprenant deux nanocristaux a été calculé. L'étude de l'influence des différents paramètres sur ce courant à travers le dispositif montre une évolution conforme aux résultats attendus.

Cette thèse est consacrée à l'étude de la dynamique des porteurs de charges sous forte concentration. La méthode Monte Carlo " Full-Band " a été utilisée pour la modélisation du transport et la relaxtion des porteurs de charge dans les semi-conducteurs III-V (GaAs, InAs, GaSb, In0.53Ga0.47As et GaAs0.50Sb0.50). Les structures électroniques ont été calculées par la Méthode des Pseudo-potentiels Non-Locaux Empiriques, ce qui a notamment permis de traiter le cas de l'alliage ternaire GaAs0.50Sb0.50 dans une approche de type Cristal Virtuel, matériau qui souffre d'un manque de caractérisations expérimentales. Dans ces semi-conducteurs polaires fortement dopés, le couplage entre phonons optiques polaires et plasmons a été pris en compte via le calcul de la fonction diélectrique totale incluant les termes associés à l'amortissement dans le système phonon-plasmon auto-cohérents. Ce phénomène de couplage phonon-plasmon, est apparu primordial pour l'analyse de la mobilité des électrons dans GaAs, In0.53Ga0.47As et GaAs0.50Sb0.50 en fonction de la concentration en accepteurs. Dans des semi-conducteurs fortement photo-excités, la relaxation des électrons et des trous a été étudiée en tenant compte du chauffage de la population de phonon (qui ralentit la relaxation des porteurs) avec un modèle Monte Carlo dédié à la dynamique des phonons (Thèse de H. Hamzeh). L'étude a montré que le ralentissement de la relaxation dépend fortement des concentrations de porteurs photo-excités à cause du couplage phonon-plasmon dans ces matériaux. Les processus de génération et recombinaison de porteurs tels que l'absorption optique, la recombinaison radiative, l'ionisation par choc et les recombinaisons Auger, ont été implémentés. Les taux de génération et recombinaison associés sont calculés directement sur les distributions de porteurs modélisées, sans supposer des distributions à l'équilibre. Ces processus sont cruciaux pour l'optimisation de Cellules Solaires à Porteurs Chauds. Le photo-courant de ce type de cellule théorique à haut rendement de 3ème génération avec un absorbeur en In0.53Ga0.47As a été étudié.

Afin de satisfaire aux exigences de plus en plus contraignantes imposées par la Roadmap ITRS, l'industrie microélectronique doit aujourd'hui envisager un certain nombre de révolutions dans ses procédés de fabrication des composants. En effet, la seule miniaturisation des dimensions du transistor à effet de champ Métal-Oxyde-Semiconducteur (MOSFET) ne suffit plus à améliorer les performances des dispositifs électroniques et de nouvelles approches doivent être imaginées. Parmi les solutions envisagées, l'une des plus prometteuses consiste à remplacer l'isolant de grille «historique» en oxyde de silicium (SiO2) et la grille en polysilicium par un couple constitué d'une grille métallique et d'un matériau isolant possédant une plus forte permittivité diélectrique. Ce travail présente ainsi les effets du couple grille TiN/dioxyde d'hafnium HfO2 sur les performances électriques d'un MOSFET en étudiant un paramètre caractéristique du transport électrique dans le canal de conduction, à savoir la mobilité des porteurs libres en régime d'inversion. Pour ce faire, une étude théorique des différentes interactions limitant la mobilité des porteurs dans ces nouvelles architectures a d'une part été réalisée. D'autre part, des techniques expérimentales innovantes d'extraction de la mobilité ont été développées (magnétorésistance, split C-V pulsé) pour caractériser finement nos dispositifs. La conjonction de ces deux approches a ainsi permis de déterminer avec précision les interactions prédominantes dans la réduction de mobilité des porteurs liées à l'utilisation d'une grille métallique TiN et d'un oxyde de grille de forte permittivité HfO2.

A wide variety of novel complementary-metal-oxide-semiconductor (CMOS) devices that are strong contenders for future high-speed and low-noise RF circuits have been evaluated by means of static electrical measurements and low-frequency noise characterizations in this thesis. These novel field-effect transistors (FETs) include (i) compressively strained SiGe channel pMOSFETs, (ii) tensile strained Si nMOSFETs, (iii) MOSFETs with high-k gate dielectrics, (iv) metal gate and (v) silicon-on-insulator (SOI) devices. The low-frequency noise was comprehensively characterized for different types of operating conditions where the gate and bulk terminal voltages were varied. Detailed studies were made of the relationship between the 1/f noise and the device architecture, strain, device geometry, location of the conduction path, surface cleaning, gate oxide charges and traps, water vapour annealing, carrier mobility and other technological factors. The locations of the dominant noise sources as well as their physical mechanisms were investigated. Model parameters and physical properties were extracted and compared. Several important new insights and refinements of the existing 1/f noise theories and models were also suggested and analyzed. The continuing trend of miniaturizing device sizes and building devices with more advanced architectures and complex materials can lead to escalating 1/f noise levels, which degrades the signal-to-noise (SNR) ratio in electronic circuits. For example, the 1/f noise of some critical transistors in a radio receiver may ultimately limit the information capacity of the communication system. Therefore, analyzing electronic devices in order to control and find ways to diminish the 1/f noise is a very important and challenging research subject. We present compelling evidence that the 1/f noise is affected by the distance of the conduction channel from the gate oxide/semiconductor substrate interface, or alternatively the vertical electric field pushing the carriers towards the gate oxide. The location of the conduction channel can be varied by the voltage on the bulk and gate terminals as well by device engineering. Devices with a buried channel architecture such as buried SiGe channel pMOSFETs and accumulation mode MOSFETs on SOI show significantly reduced 1/f noise. The same observation is made when the substrate/source junction is forward biased which decreases the vertical electric field in the channel and increases the inversion layer separation from the gate oxide interface. A 1/f noise model based on mobility fluctuations originating from the scattering of electrons with phonons or surface roughness was proposed. Materials with a high dielectric constant (high-k) is necessary to replace the conventional SiO2 as gate dielectrics in the future in order to maintain a low leakage current at the same time as the capacitance of the gate dielectrics is scaled up. In this work, we have made some of the very first examinations of 1/f noise in MOSFETs with high-k structures composed by layers of HfO2, HfAlOx and Al2O3. The 1/f noise level was found to be elevated (up to 3 orders of magnitude) in the MOSFETs with high-k gate dielectrics compared to the reference devices with SiO2. The reason behind the higher 1/f noise is a high density of traps in the high-k stacks and increased mobility fluctuation noise, the latter possibly due to noise generation in the electron-phonon scattering that originates from remote phonon modes in the high-k. The combination of a TiN metal gate, HfAlOx and a compressively strained surface SiGe channel was found to be superior in terms of both high mobility and low 1/f noise. / QC 20100928

No description available.

Because of their outstanding characteristics, micro/nano-mechanical (MM) structures have found a plethora of applications in wireless communications and sensing. Many of these MM structures utilize mechanical vibrations (or phonons) at megahertz or gigahertz frequencies for their operation. On the other hand, the periodic atomic structure of crystals is the fundamental phenomenon behind the new era of electronics technology. Such atomic arrangements lead to a periodic electric potential that modifies the propagation of electrons in the crystals. In some crystals, e.g. silicon (Si), this modification leads to an electronic band gap (EBG), which is a range of energies electrons can not propagate with. Discovering EBGs has made a revolution in the electronics and through that, other fields of technology and the society. Inspired by these trends of science and technology, I have designed and developed integrated MM periodic structures that support large phononic band gaps (PnBGs), which are ranges of frequencies that phonons (and elastic waves) are not allowed to propagate. Although PnBGs may be found in natural crystals due to their periodic atomic structures, such PnBGs occur at extra high frequencies (i.e., terahertz range) and cannot be easily engineered with the current state of technology. Contrarily, the structures I have developed in this research are made on planar substrates using lithography and plasma etching, and can be deliberately engineered for the required applications. Although the results and concepts developed in this research can be applied to other substrates, I have chosen silicon (Si) as the substrate of choice for implementing the PnBG structure due to its unique properties. I have also designed and implemented the fundamental building blocks of MM systems (e.g., resonators and waveguides) based on the developed PnBG structures and have shown that low loss and efficient MM devices can be made using the PnBG structures. As an example of the possible applications of these PnBG structures, I have shown that an important source of loss, the support loss, can be suppressed in MM resonators using PnBG structures. I have also made improvements in the characteristics of the developed MM PnBG resonators by developing and employing PnBG waveguides. I have further shown theoretically, that photonic band gaps (PtBGs) can also be simultaneously obtained in the developed PnBGs structures. This can lead to improved photon-phonon interactions due to the effective confinement of optical and mechanical vibrations in such structures. For the design, fabrication, and characterization of the structures, I have developed and utilized complex and efficient simulation tools, including a finite difference time domain (FDTD), a plane wave expansion (PWE), and a finite elements (FE) tool, each of which I have developed either completely from scratch, or by modification of an existing tool to suit my applications. I have also developed and used advanced micro-fabrication recipes, and characterization methods for realizing and characterizing these PnBG structures and devices. It is agued that by using the same ideas these structures can be fabricated at nanometer scales to operate at ultra high frequency ranges. I believe my contributions has opened a broad venue for new MM structures based on PnBG structures with superior characteristics compared to the conventional devices.

The object of this research was to investigate dispersion in surface phononic crystals (PnCs) for application to a newly developed passive surface acoustic wave (SAW) ozone sensor. Frequency band gaps and slow sound already have been reported for PnC lattice structures. Such engineered structures are often advertised to reduce loss, increase sensitivity, and reduce device size. However, these advances have not yet been realized in the context of surface acoustic wave sensors. In early work, we computed SAW dispersion in patterned surface structures and we confirmed that our finite element computations of SAW dispersion in thin films and in one dimensional surface PnC structures agree with experimental results obtained by laser probe techniques. We analyzed the computations to guide device design in terms of sensitivity and joint spectral operating point. Next we conducted simulations and experiments to determine sensitivity and limit of detection for more conventional dispersive SAW devices and PnC sensors. Finally, we conducted extensive ozone detection trials on passive reflection mode SAW devices, using distinct components of the time dispersed response to compensate for the effect of temperature. The experimental work revealed that the devices may be used for dosimetry applications over periods of several days.

Comprehensive molecular design was used to identify new heat transfer fluids for direct immersion phase change cooling of electronic systems. Four group contribution methods for thermophysical properties relevant to heat transfer were critically evaluated and new group contributions were regressed for organosilicon compounds. 52 new heat transfer fluids were identified via computer-aided molecular design and figure of merit analysis. Among these 52 fluids, 9 fluids were selected for experimental evaluation and their thermophysical properties were experimentally measured to validate the group contribution estimates. Two of the 9 fluids (C6H11F3 and C5H6F6O) were synthesized in this work. Pool boiling experiments showed that the new fluids identified in this work have superior heat transfer properties than existing coolant HFE 7200. The radiative forcing and global warming potential of new fluids, calculated via a new group contribution method developed in this work and FT-IR analysis, were found to be significantly lower than those of current coolants. The approach of increasing the thermal conductivity of heat transfer fluids by dispersing nanoparticles was also investigated. A model for the thermal conductivity of nanoparticle dispersions (nanofluids) was developed that incorporates the effect of size on the intrinsic thermal conductivity of nanoparticles. The model was successfully applied to a variety of nanoparticle-fluid systems. Rheological properties of nanofluids were also investigated and it was concluded that the addition of nanoparticles to heat transfer fluids may not be beneficial for electronics cooling due to significantly larger increase in viscosity relative to increase in thermal conductivity.