Aplicações móveis que não podem ser recarregadas durante operação, como sensores biomédicos e aplicações da Internet das Coisas, dependem da extração de energia do próprio meio onde se encontram. Tensões de alimentação típicas são normalmente maiores que as disponiveis por métodos de extração de energia do meio e requerem uma conversão de nivel DC que invariavelmente resulta em perdas proporcionais ao fator de conversão. Consequentemente, aplicações projetadas para tensões de alimentação mais próximas da tensão nominal da fonte melhora a eficiência energética. Entretanto, topologias de circuitos elétricos para tensões típicas de alimentação sao impróprias para tensões extremamente baixas. Neste trabalho foram propostas topologias de amplificadores de saída unipolar e diferencial para tensões de alimentaçãoo na casa de centenas de milivolts. As técnicas propostas se baseiam no uso de pares pseudodiferenciais com terminais de corpo polarizados diretamente para vários propósitos, incluindo rejeição de modo comum e polarização de modo comum de saída e corrente DC. Adicionalmente, um oscilador baseado na mesmas técnicas de polarização foi proposto e projetado para duas classes de aplicações: um oscilador de referência intrinsicamente estável e um oscilador controlado por tensão para conversão analógica-digital com melhor linearidade. / Nomadic applications which cannot be recharged while at operation, such as biomedical sensors and Internet of Things applications, rely on energy harvesting from the environment. Typical supply voltages are usually higher than those achieved by energy harvesting methods and requires DC-DC conversion levels, which invariably results in energy loss proportionally to the step of voltage conversion. Consequently, designing at supply voltages closer to the nominal voltage of the energy source improves power efficiency. However, extremely low supply voltages bring design challenges, as circuit topologies for typical voltages employ techniques not suitable for extremely low supply voltages. In this work, single ended and fully differential amplifier topologies for voltage supplies in the range of few hundreds mV were proposed. The proposed approaches use the pseudo differential pairs with the transistor bulk terminals with forward biasing voltages for several purposes, including common mode rejection, output common mode voltage and DC current biasing. Additionally, a ring oscillator based in the same biasing techniques was proposed and designed for two main classes of applications: an intrinsically stable reference oscillator and a voltage controlled oscillator for analog-digital conversion with linearity improvements.

Today, medical implants such as cardiac pacemakers, neurostimulators, hearing aids anddrug delivery systems are increasinglymore important and frequently used in the health caresystem. This type of devices have historically used inductive coupling as communicationmedium. Newdemands on accessibility and increased performance in technology drives newresearch toward using radio communications. The FCCMICS radio standard are specificallydevoted for implantable devices.Basically all published research on transmitters in this area are using frequency shift keying(FSK) modulation. The purpose of this thesis is to explore the viability of using phase shiftkeying (PSK) modulation in ultra low power transmitters and suggest suitable architectures.

Ultra-low-power circuits are becoming more desirable due to growing portable device markets and they are also becoming more interesting and applicable today in biomedical, pharmacy and sensor networking applications because of the nano-metric scaling and CMOS reliability improvements. In this thesis, three main achievements are presented in ultra-low-power adders. First, a new majority function algorithm for carry and the sum generation is presented. Then with this algorithm and implied new architecture, we achieved a circuit with 75mV supply voltage operation. Last but not least, a 64 bit current-mode majority-function adder based on the new architecture and algorithm is successfully tested at 75mV supply voltage. The circuit consumed 4.5nW or 3.8pJ in one of the worst conditions.

With the increasing stringency on sulfur content in petrodiesel, there is a growing tendency of broader usage of ultra low sulfur diesel (ULSD) with sulfur content of 15 ppm. Refineries around the world should develop cost-effective and sustainable strategies to meet these requirements. The primary objective of this work is to analyze alternatives for producing ULSD. In addition to the conventional approach of revamping existing hydrotreating facilities, the option of blending petrodiesel with biodiesel is investigated. Blending petrodiesel with biodiesel is a potentially attractive option because it is naturally low in sulfur, enhances the lubricity of petrodiesel, and is a sustainable energy resource. In order to investigate alternatives for producing ULSD, several research tasks were undertaken in this work. Firstly, base-case designs of petrodiesel and biodiesel production processes were developed using computer-aided tools ASPEN Plus. The simulations were adjusted until the technical criteria and specifications of petrodiesel and biodiesel production were met. Next, process integration techniques were employed to optimize the synthesized processes. Heat integration for petrodiesel and biodiesel was carried out using algebraic, graphical and optimization methods to maximize the integrated heat exchange and minimize the heating and cooling utilities. Additionally, mass integration was applied to conserve material resources. Cost estimation was carried out for both processes. The capital investments were obtained from ASPEN ICARUS Process Evaluator, while operating costs were calculated based on the updated chemical market prices. The total operating costs before and after process integration were calculated and compared. Next, blending optimization was performed for three blending options with the optimum blend for each option identified. Economic comparison (total annualized cost, breakeven analysis, return on investment, and payback period) of the three options indicated that the blending of ULSD with chemical additives was the most profitable. However, the subsequent life-cycle greenhouse gas (GHG) emission and safety comparisons demonstrated that the blending of ULSD with biodiesel was superior.

The "slot-drill" completion method, which utilizes a mechanically cut high-conductivity "slot" in the target formation created using a tensioned abrasive cable, has been proposed as an alternative stimulation technique for shale-gas and other low/ultra-low permeability formations. This thesis provides a comprehensive numerical simulation study on the "slot drill" completion technique. Using a Voronoi gridding scheme, I created representative grid systems for the slot-drill completion, as well as for the case of a vertical well with a single fracture, the case of a horizontal well with multiple hydraulic fractures, and various combinations of these completions. I also created a rectangular slot configuration, which is a simplified approximation of the actual "slot-drill" geometry, and investigated the ability of this rectangular approximation to model flow from the more complicated (actual) slot-drill configuration(s). To obtain the maximum possible diagnostic and analytical value, I simulated up to 3,000 years of production, allowing the assessment of production up to the point of depletion (or boundary-dominated flow). These scenarios provided insights into all the various flow regimes, as well as provided a quantitative evaluation of all completion schemes considered in the study. The results of my study illustrated that the "slot-drill" completion technique was not, in general, competitive in terms of reservoir performance and recovery compared to the more traditional completion techniques presently in use. Based on my modeling, it appears that the larger surface area to flow that multistage hydraulic fracturing provides is much more significant than the higher conductivity achieved using the slot-drill technique. This work provides quantitative results and diagnostic interpretations of productivity and flow behavior for low and ultra-low permeability formations completed using the slot-drill method. The results of this study can be used to (a) help evaluate the possible application of the "slot-drill" technique from the perspective of performance and recovery, and (b) to establish aggregated economic factors for comparing the slot-drill technique to more conventional completion and stimulation techniques applied to low and ultra-low permeability reservoirs.

This study investigated the fuel properties, engine performances, and emissions of two biodiesels and diesel. The fuels examined were D100 (ultra-low sulfur diesel), B20 (20% palm biodiesel +80% ultra-low sulfur diesel) and B100 (palm biodiesel). The fuel properties analysis results showed that the benefits of biodiesel were high cetane value, extremely low sulfur and aromatic contents, and good lubricity. While the defects of biodiesel were high pour point. The particulates emitted from the burning of D100, B100, B20 were mainly fine particulates, also known as young aerosols. Particles smaller than 2.5 µm easily enter the trachea and bronchus via the upper respiratory tract, finally deposit on the alveolus, which could cause severe injury to human health. The emission of soluble organic fraction (SOF) from diesel engine using D100, B100 and B20 were 23.2%, 19.9% and 20.2%, respectively. The SOF of D100 is slightly higher than B100 and B20. It suggested that adding biodiesel into diesel can decrease SOF and thus reduce the potential danger to human health. The original total PAHs concentration of tail gas emitted from engines using D100, B100 and B20 were 241, 50.6 and 98.8 µg/m3, respectively. Adding 20% biodiesel into D100 could reduce 59.0% of PAHs emission. Moreover, the original total BaPeq concentration of tail gas emitted from diesel engines using D100, B100 and B20 were 0.714, 0.509 and 0.570 µg/m3, respectively. Adding 20% biodiesel into D100 could also reduce 20.2% of total BaPeq emission. Hence, adding biodiesel into diesel can effectively reduce the emission of PAHs and the potential danger to human health. The emission factors of carbonyl compounds from diesel engines using D100, B100 and B20 were 395, 1,170 and 326 mg/BHP-hr, respectively. carbonyl compounds of B100 were obviously higher than D100 and B20. The results indicated that using pure palm biodiesel in diesel engine can increased the emission of carbonyl compounds. However, adding 20% biodiesel into D100 can effectively reduce 17.5% of carbonyl compounds emission. Keyword: ultra-low sulfur diesel, palm biodiesel, fuel properties¡BThe emission of soluble organic fraction (SOF)¡BPAHs¡Bcarbonyl compounds¡C

ABSTRACT Our Atlantic Ultra-Low-Temperature Tuna Fishing fleets were under attack from Japan during the 2004 ICCAT (International Commission for the Conservation of Atlantic Tunas) Annual Meeting. Japan openly called for tough penalty for our fleet such as significantly cutting our fishing quotas in the Atlantic and revoking our national membership in the ICCAT for over-fishing, violating International Conservation Regulations and washing fish right on the ocean. This resulted in catastrophic damage in our Ultra-Low-Temperature fishing operation in the pacific and undue pressure from our international competitors. Under the strict monitoring of international conservation groups, our government was forced reform its regulations of the fishing industry and focus on the long-neglected area of offshore fishing. To express its sincerity in active management of the industry, the government offered a series of matching measures. The first of these measures was the reduction of the number of ships to achieve cutback in fishing capacity. The three-year plan spanning from 2005 to 2007 involved a joint venture by the government and the offshore industry with joint contribution of 12.5 million US dollars to disassemble 183 large Tuna Long line Fishing Vessels in the three oceans. Fishing vessels in the Atlantic were reduced from 100 to 76. The next measures were to improve the monitoring of the fishing industry and to end any and all illegal, unreported and unregulated fishing. How will our offshore fishing industry recover after a series of attack? How will our industry thrive in the austere environment of conservatism? Do we still possess any competitive advantage after the vessel-reduction measure? This thesis will cover the analysis of the competitive advantage of the Offshore Ultra-Low-Temperature Fishing Industry after the vessel-reduction, the analysis of the supply-and-demand of the Ultra-Low-Temperature market and SWOT analysis. It will also discuss in depth the post-reform fishing industry management strategy for reference.

Ultra-low-power circuits are becoming more desirable due to growing portable device markets and they are also becoming more interesting and applicable today in biomedical, pharmacy and sensor networking applications because of the nano-metric scaling and CMOS reliability improvements. In this thesis, three main achievements are presented in ultra-low-power adders. First, a new majority function algorithm for carry and the sum generation is presented. Then with this algorithm and implied new architecture, we achieved a circuit with 75mV supply voltage operation. Last but not least, a 64 bit current-mode majority-function adder based on the new architecture and algorithm is successfully tested at 75mV supply voltage. The circuit consumed 4.5nW or 3.8pJ in one of the worst conditions.

Aplicações móveis que não podem ser recarregadas durante operação, como sensores biomédicos e aplicações da Internet das Coisas, dependem da extração de energia do próprio meio onde se encontram. Tensões de alimentação típicas são normalmente maiores que as disponiveis por métodos de extração de energia do meio e requerem uma conversão de nivel DC que invariavelmente resulta em perdas proporcionais ao fator de conversão. Consequentemente, aplicações projetadas para tensões de alimentação mais próximas da tensão nominal da fonte melhora a eficiência energética. Entretanto, topologias de circuitos elétricos para tensões típicas de alimentação sao impróprias para tensões extremamente baixas. Neste trabalho foram propostas topologias de amplificadores de saída unipolar e diferencial para tensões de alimentaçãoo na casa de centenas de milivolts. As técnicas propostas se baseiam no uso de pares pseudodiferenciais com terminais de corpo polarizados diretamente para vários propósitos, incluindo rejeição de modo comum e polarização de modo comum de saída e corrente DC. Adicionalmente, um oscilador baseado na mesmas técnicas de polarização foi proposto e projetado para duas classes de aplicações: um oscilador de referência intrinsicamente estável e um oscilador controlado por tensão para conversão analógica-digital com melhor linearidade. / Nomadic applications which cannot be recharged while at operation, such as biomedical sensors and Internet of Things applications, rely on energy harvesting from the environment. Typical supply voltages are usually higher than those achieved by energy harvesting methods and requires DC-DC conversion levels, which invariably results in energy loss proportionally to the step of voltage conversion. Consequently, designing at supply voltages closer to the nominal voltage of the energy source improves power efficiency. However, extremely low supply voltages bring design challenges, as circuit topologies for typical voltages employ techniques not suitable for extremely low supply voltages. In this work, single ended and fully differential amplifier topologies for voltage supplies in the range of few hundreds mV were proposed. The proposed approaches use the pseudo differential pairs with the transistor bulk terminals with forward biasing voltages for several purposes, including common mode rejection, output common mode voltage and DC current biasing. Additionally, a ring oscillator based in the same biasing techniques was proposed and designed for two main classes of applications: an intrinsically stable reference oscillator and a voltage controlled oscillator for analog-digital conversion with linearity improvements.

Alors que la miniaturisation des transistors suivant la loi de Moore semble ralentir dû à des limites physique, technologique et économique, il devient essentiel de trouver des alternatives afin de répondre à la demande croissante en électronique : informatique et télécommunication, objets intelligents et interconnectés, domaine médical et biologique… En cela, l’utilisation de la troisième dimension, par opposition à la fabrication planaire de composants électrique, semble être une option prometteuse. L’intégration 3D permet en effet d’incorporer plus de composants sur une même surface en les empilant à un coût technologique et économique plus faibles que celui de la miniaturisation. En particulier, l’intégration séquentielle ou CoolCubeTM au CEA-Leti permet de profiter pleinement de la troisième dimension en fabriquant successivement les uns sur les autres chaque étage d’une puce, permettant un alignement optimal des transistors unitaires à chaque niveau. Néanmoins, plusieurs verrous technologiques particuliers à l’intégration 3D Séquentielle doivent alors être levés.Dans ce manuscrit, nous nous intéresserons à la réduction du budget thermique pour la fabrication des transistors supérieurs, nécessaire afin de ne pas endommager les étages inférieurs lors de la réalisation des composants sus-jacents. Nous commencerons par définir le budget thermique maximal afin de ne pas dégrader les couches inférieures avant d’identifier les briques technologiques impactées lors de la fabrication d’un transistor. Nous verrons alors dans ce manuscrit qu’il sera non seulement nécessaire d’étudier de nouveaux matériaux, mais aussi de nouveaux procédés voire de nouvelles techniques de recuit. Plus particulièrement, nous évaluerons tout d’abord l’utilisation des diélectriques low-k comme espaceurs de grille permettant notamment d’améliorer les performances dynamiques des composants. Ensuite, nous présenterons différentes stratégies de préparation de surface et de croissance épitaxiale à basse température pour la réalisation des sources et drains surélevés. Enfin, nous étudierons l’impact d’un budget thermique faible ainsi que de nouvelles techniques de recuits micro-onde et laser sur les propriétés de l’empilement de grille. Nous verrons en particulier que la difficulté principale d’une intégration à bas budget thermique est l’obtention d’une bonne fiabilité des transistors. Toute cette étude nous permettra alors de proposer des solutions à l’intégration d’un transistor à un bas budget thermique compatible avec l’intégration 3D Séquentielle. / As the scaling of transistors following Moore’s law seems to slow down due to physical, technological and economical barriers, it becomes mandatory to find alternatives to cope with the increasing demand in electronics: computing and telecommunication, smart and interconnected objects, medical and biological fields… To that end, the use of the third dimension, in opposition to the planar processing of electronical devices, appears to be a promising option. Indeed, 3D integration allows incorporating more devices per area by stacking them at a lower technological and economical cost than scaling. More specifically, 3D sequential or CoolCubeTM at CEA-Leti allows benefiting fully from the third dimension by processing successively one on top of each other each level of a die, allowing an optimal alignment of single transistors at each layer. However, several technological barriers specific to 3D Sequential Integration need then to be alleviated.In this work, we will study the reduction of thermal budget for the transistors fabrication, which is required to not damage bottom levels during the processing of top devices. First, we will define the maximal thermal budget in order not to degrade bottom layers prior to identifying the technological modules impacted during the fabrication of a transistor. We will then see in this work that not only new materials need to be studied, but also new processes and new annealing techniques. Specifically, we will first evaluate the use of low-k dielectrics as gate offset spacers, allowing the improvement of devices dynamic performance. Then we will present different strategies of surface preparation and epitaxial growth at low temperature for the formation of raised sources and drains. Finally, we will study the impact of a low thermal budget process flow along with novel microwaves and laser annealing techniques on the gate stack properties. In particular, we will see that the biggest challenge in a low thermal budget integration is to get a good reliability of transistors. This study leads to a proposed low thermal budget process flow for transistor fabrication compatible with 3D Sequential Integration.