The boundary element method (BEM) has become a powerful method for the numerical solution of boundary-value problems (BVPs), due to its ability (at least for problems with constant coefficients) of reducing a BVP for a linear partial differential equation (PDE) defined in a domain to an integral equation defined on the boundary, leading to a simplified discretisation process with boundary elements only. On the other hand, the coefficients in the mathematical model of a physical problem typically correspond to the material parameters of the problem. In many physical problems, the governing equation is likely to involve variable coefficients. The application of the BEM to these equations is hampered by the difficulty of finding a fundamental solution. The first part of this thesis will focus on the derivation of the boundary integral equation (BIE) for the Laplace equation, and numerical results are presented for some examples using constant elements. Then, the formulations of the boundary-domain integral or integro-differential equation (BDIE or BDIDE) for heat conduction problems with variable coefficients are presented using a parametrix (Levi function), which is usually available. The second part of this thesis deals with the extension of the BDIE and BDIDE formulations to the treatment of the two-dimensional Helmholtz equation with variable coefficients. Four possible cases are investigated, first of all when both material parameters and wave number are constant, in which case the zero-order Bessel function of the second kind is used as fundamental solution. Moreover, when the material parameters are variable (with constant or variable wave number), a parametrix is adopted to reduce the Helmholtz equation to a BDIE or a BDIDE. Finally, when material parameters are constant (with variable wave number), the standard fundamental solution for the Laplace equation is used in the formulation. In the third part, the radial integration method (RIM) is introduced and discussed in detail. Modifications are introduced to the RIM, particularly the fact that the radial integral is calculated by using a pure boundary-only integral which relaxes the “star-shaped” requirement of the RIM. Then, the RIM is used to convert the domain integrals appearing in both BDIE and BDIDE for heat conduction and Helmholtz equations to equivalent boundary integrals. For domain integrals consisting of known functions the transformation is straightforward, while for domain integrals that include unknown variables the transformation is accomplished with the use of augmented radial basis functions (RBFs). The most attractive feature of the method is that the transformations are very simple and have similar forms for both 2D and 3D problems. Finally, the application of the RIM is discussed for the diffusion equation, in which the parabolic PDE is initially reformulated as a BDIE or a BDIDE and the RIM is used to convert the resulting domain integrals to equivalent boundary integrals. Three cases have been investigated, for homogenous, non-homogeneous and variable coefficient diffusion problems.

Stabilised zirconias have a remarkable variety of technological and commercial applications, e.g., thermal barrier coatings, gas sensors, solid oxide fuel cells, ceramic knives and even fashion jewelry. This amazing versatility seems to originate from the creation of atomic defects (oxide ion vacancies) in the zirconia crystal. Indeed, these vacancies, and their interactions with other vacancies or cations, dramatically affect the structural, thermal, mechanical and electrical properties of zirconia. This thesis is concerned with the study of the role of the vacancy interactions on the conducting properties of these materials. This study was performed by using realistic, first-principles based molecular dynamics simulations. The first system studied in this thesis is Zr0:5 0:5xY0:5+0:25xNb0:25xO7. This has a fixed number of vacancies across the series but its conductivity changes by almost two orders of magnitude as a function of x. For this reason, Zr0:5 0:5xY0:5+0:25xNb0:25xO7 represents an ideal test-bed for the role of the cation species on the defect interactions and therefore on the ionic conductivity of these materials. Realistic inter-atomic potentials for Zr0:5 0:5xY0:5+0:25xNb0:25xO7 were developed on a purely first-principles basis. The observed trends of decreasing conductivity and increasing disorder with increasing Nb5+ content were successfully reproduced. These trends were traced to the influences of the cation charges and relative sizes and their effect on vacancy ordering by carrying out additional calculations in which, for instance, the charges of the cations were equalised. The effects of cation ordering were considered as well and their influence on the conductivity understood. The second part of this thesis deals with Sc2O3–doped (ScSZ) and Y2O3–doped (YSZ) zirconias. These systems are of great academic and technological interest as they find use in solid oxide fuel cells. Inter-atomic potentials were parametrised and used to predict the structural and conducting properties of these materials, which were found to agree very well with the experimental evidence. The simulations were then used to study the role of the vacancy interactions on the conducting properties of these materials. Two factors were found to influence the ionic conductivity in these materials: cation-vacancy and vacancy-vacancy interactions. The former is responsible for the difference in conductivity observed between YSZ and ScSZ. Vacancies, in fact, prefer to bind to the smaller Zr4+ ions in YSZ whereas there is not a strong preference in the case of ScSZ, since the cations have similar sizes in this case. This effect is observed at temperatures as high as T = 1500 K. Finally, it was found that vacancies tend to order so that they can minimise their mutual interaction and that this ordering tendency is what ultimately is responsible for the observed anomalous decrease of the ionic conductivity with increasing dopant concentration. The consequences of such a behaviour are discussed.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

Heat insulation is the process of blocking the transfer of thermal energy between objects at different temperatures. Heat transfer occurs due to conduction, convection, or radiation, as well as any combination of these three mechanisms. Fibrous insulations can completely suppress the convective mode of heat transfer for most applications, and also help to reduce the conductive and radiative modes to some extent. In this study, an attempt has been made to computationally predict the effects of microstructural parameters (e.g., fiber diameter, fiber orientation and porosity) on the insulation performance of fibrous materials. The flexible simulation method developed in this work can potentially be used to custom-design optimal multi-component fibrous insulation media for different applications. With regards to modeling conductive heat transfer, a computationally-feasible simulation method is developed that allows one to predict the effects of each microstructural parameter on the transfer of heat across a fibrous insulation. This was achieved by combining analytical calculations for conduction through interstitial fluid (e.g., air) with numerical simulations for conduction through fibrous structures. With regards to modeling radiative heat transfer, both Monte Carlo Ray Tracing and Electromagnetic Wave Theory were implemented for our simulations. The modeling methods developed in this work are flexible to allow simulating the performance of media made up of different combinations of fibers with different materials or dimensions at different operating temperatures. For example, our simulations demonstrate that fiber diameter plays an important role in blocking radiation heat transfer. In particular, it was shown that there exists an optimum fiber diameter for which maximum insulation against radiative transfer is achieved. The optimum fiber diameter is different for fibers made of different materials and also depends on the mean temperature of the media. The contributions of conduction and radiation heat transfer predicted using the above techniques are combined to define a total thermal resistance value for media with different microstructures. Such a capability can be of great interest for design and optimization of the overall performance of fibrous media for different applications.

Au cours de ce travail de thèse, nous avons étudié la formation de nanogaps par électromigration dans des nanofils d'or. Cette technique consiste à provoquer la rupture d'un nanofil en lui appliquant de fortes densités de courant et peut être utilisée pour la caractérisation électrique de nano-objets. Néanmoins, les mécanismes de formation des nanogaps ne sont, à ce jour, pas encore totalement compris. L'étude en temps réel du processus d'électromigration par microscopie à force atomique a permis d'apporter un éclairage nouveau de la dynamique du processus. En effet, il apparaît que la structure globale du dispositif est définie dans les premiers temps de l'électromigration et nous avons montré que cette structure est directement liée à la microstructure du film métallique. Pour la première fois, des nanogaps ont été élaborés par électromigration dans des films monocristallins. Malgré l'absence de joints de grain, il est possible de former des nanogaps dans un matériau épitaxié. L'utilisation de ces matériaux permet d'obtenir des nanogaps avec une morphologie plus reproductible. Les propriétés de transports des nanogaps obtenus à partir de films polycristallins ont été caractérisées. Les caractéristiques obtenues présentent toutes des signatures particulières, attribuées à la présence d'agrégats d'or provenant de la procédure d'électromigration et/ou de polymères issus du procédé de nanofabrication. Ces résultats montrent la difficulté à réaliser des mesures à l'échelle de la molécule unique / We have studied the formation of nanogaps by électromigration of gold nanowires. Electromigration relies on large current densities to break a thin and narrow metallic wire and can be used for the electrical characterization of nanometer scale objects. Nevertheless, a complete description of the electromigration process is lacking. Real time atomic force microscopy during the electromigration process gave a new view the dynamic of the process. Indeed, this study reveals that the major structural changes appear at the early stage of the process and that the final global structure of the device is directly linked to the pre-existing microstructure. For the first time, we make nanogaps by électromigration of monocrystalline nanowires. We show that despite the lacking of grain boundaries, it is possible to form nanogaps in epitaxial materials. The morphology of those nanogaps is more reproducible. The electrical transport properties of the polycristalline nanogaps have been measured. The entire obtained characteristics show particular signature that can be attributed to the presence of gold cluster formed during the électromigration process and / or to polymer from the nanowire fabrication. These results show the difficulty to measure at the molecular scale

Five exterior hot benches have been installed in Gävle, in Kyrkogatan street by the company Gävle Energi with the aim of achieving the wellnes of people that sit on them. This system uses the residual heat from the district heating, representing consequently a non-polluting system. However, the temperature desired on the surface, 35°C is not always achieved before different exterior conditions. For this reason, Gävle Energi is interested in carrying out a study about enhancements that could be made in the system in order to take them into account for future projects of this kind of technology.   The aim of this project is analysing if it would be possible to achieve the requirements established by Gävle Energi, changing with this objective all the necessary system parameters of the current system such as diameter of the pipes, materials, number of turns... These requirements consist of working with a supply temperature of 40, 45 and 50°C when the exterior conditions are 0, -5 and -10°C respectively, accomplishing always 35°C on the surface. Moreover, in case that it was not possible, providing the company with the characteristics of the system that would make the system as efficient as possible, specifying for different exterior temperatures the mass flow, pressure drop, velocity and needed power.   The study has been developed by different simulations with the software COMSOL, whose use requires a high knowledge on heat transfer. After several simulations, it has been checked that it is not possible to accomplish the requirements established by the company. However, a new more efficient design has been designed because the supply temperatures of the system to accomplish an average temperature of around 35°C on the surface have been minimised. For that, several changes have been carried out. The number of pipes turns have been increased from 12 to 17, their total diameter from 20mm to 30mm and the distance between the centres of the pipes from 5.5cm to 4cm. The 2mm of outer plastic thickness of the pipes has been replaced by copper and the height of the pipes has been moved 2cm upwards.   With all these changes, the final length of the pipes inner the stones has a value of 40.6m and the supply temperatures reach 46, 47 and 49°C for the 0,-5 and -10°C exterior conditions respectively. Apart from the supply temperatures for the study cases, the ones necessary to accomplish always the temperature desired on the surface for other exterior temperatures have been provided together with the amount of power necessary, velocity flow, volumetric flow and pressure drop for all the different cases. These values would allow the company to work always at the optimum point as well as to design the heat pump for the system.

L’objectif de ce travail consiste à accroître la connaissance sur le thermoformage du verre. Ce procédé, largement mis en œuvre par le CERFAV, est basé sur la déformation du verre au sein d’un four électrique radiant sous l’effet de la température. Deux axes de travail ont été identifiés : le développement d’un outil de simulation numérique, et la conception d’un banc expérimental original. D’un point de vue mécanique, le modèle viscoélastique rhéologique choisi permet de décrire le comportement du verre successivement élastique linéaire, viscoélastique puis visqueux, de l’ambiance jusqu’à 800°C. D’un point de vue thermique, le couplage de la conduction avec le rayonnement, validé pour des cas tests issus de la littérature, a été appliqué dans des conditions similaires au thermoformage sans déformation. Pour ce faire, le code radiatif RAD2D développé au LEMTA a été généralisé au verre et implémenté dans le logiciel MSC MARC©. La phase expérimentale a été marquée par la conception et le développement d’un banc d’essais original autour d’un four de thermoformage. Les mesures sans contact de déplacement vertical et de température de la face supérieure du verre ont ainsi été comparées qualitativement aux résultats numériques. Par ailleurs, le four a fait l’objet d’un protocole de qualification (température et flux) qui permettra d’affiner la modélisation. En parallèle, une campagne de détermination des propriétés thermomécaniques du verre a été initiée en vue de paramétrer plus fidèlement le modèle numérique. En conclusion cette première étude a permis la mise en place d’outils performants de calcul et de mesure, les premiers résultats probants devront être complétés par une seconde étude. / This study aims at increasing knowledge on glass forming. This process, studied and developed by the CERFAV, is based on the deformation of glass by increasing temperature within an electric radiant furnace. Two investigation tools were developed in parallel: a computational program dedicated to the numerical simulation, and an original experimental set up. The rheological viscoelastic model implemented into the MSCMARC© software for mechanics was chosen to describe the behavior of glass at temperature ranging from ambient to 800°C in turn elastic, viscoelastic and viscous. The coupled code (conduction with radiation), validated for test cases found in the literature, were applied to benchmarks similar to glass forming with no deformation. The radiative home-made RAD2D code based on the Finite Volume Method was adapted to glass and implemented into MSCMARC©. The experimental part of the study consisted in designing and developing an original set up based on a thermoforming furnace. The no-contact-measurements of vertical displacement and temperature on the upper face of the glass were qualitatively compared to our innovative numerical results. In addition, the furnace was characterized in terms of temperature and fluxes and a determination campaign of the thermomechanical properties of the glass was initiated in order to provide data for the refinement of the model. To conclude, efficient tools for calculation and measurement were developed in the present study. The first results are convincing and will be supplemented by further investigations.

The report is an overview article, as a result of our investigation at the field of laser welding applied to electromobility cells manufactured in an aluminium housing. This project was proposed by the University of Skövde in collaboration with ASSAR Centre. The key results presented are based on the study of the following parameters: laser type and power, shielding gases, welding modes, patterns and layout. The conclusions of the project define the final selection of each parameter in order to achieve minimum defects and optimal electrical performance by minimizing the contact resistance.

För att nå utsatta klimatmål i Sverige och på EU-nivå är begränsningen av fossil fordonstrafik en hörnsten. Lösningen kan vara eldrift, men dilemmat vid långa transporter är påfyllnad av drivmedel. Utmaningen blir att tillgodose den ökade efterfrågan på energi – både gällande distribution och produktion – på ett miljömässigt försvarbart sätt. En litteraturstudie samt en intervju utförs för att undersöka om vägar kan göras smartare för att främja en grönare, eldriven fordonstrafik och ge möjlighet för energiutvinning i vägområdet samt elektrisk energidistribution till fordon under färd. Teknologin till energidistribution och energiutvinningssystem finns, men den kan alltid utvecklas för billigare tillverkning och effektivare produkter. Studerade energidistribueringssystem är induktion och konduktion. Energiutvinningssystemen som har behandlats är solenergi, piezoelektricitet och vindkraft. Pilotprojekt inom de bägge områdena existerar och fler projekt är under utveckling. / Sweden and the EU has a lot of climate goals, to achieve these goals the use of fossil fuels must be reduced. One solution can be electric powered vehicles, when travelling long distances there is a problem to replenish fuel. The challenge is to meet the increased demand for energy - both in terms of production and distribution - in an environmentally sound manner. A literature study as well as an interview is carried out to investigate whether roads can be made smarter to promote a greener, electric vehicle traffic and the possibility of energy harvesting systems and energy distribution in the road area. The technology for energy harvesting systems and energy distribution is available, but it can always be further developed for cheaper manufacturing and more efficient products. Energy distribution systems studied are induction and conduction. The energy harvesting systems in this report are solar energy, piezoelectricity and wind power. Pilot projects in both areas exists and more projects are under development.

O Tempo de Condução Total Atrial (TACT) reflete a condução dos estímulos elétricos no território atrial. Tal parâmetro pode ser mensurado por meio do estudo eletrofisiológico ou por métodos não invasivos, como a ecocardiografia tecidual. Em medicina sabe-se que a mensuração do tempo decorrido entre o início da onda P do eletrocardiograma ao pico da onda A do gráfico do Doppler tecidual (intervalo PATDI) é um preditor independente da ocorrência de fibrilação atrial secundária a diversas causas. Em medicina veterinária não existem estudos publicados sobre o assunto até o momento. Objetivou-se estabelecer valores normais de TACT para cães e investigar quais variáveis podem influenciar os resultados obtidos. Além disso, propôs-se a investigação do TACT como um preditor da ocorrência de fibrilação atrial em cães com valvopatia mitral mixomatosa. Realizou-se estudo retrospectivo utilizando um banco de imagens ecocardiográficas de cães que participaram de projetos de pesquisa no Serviço de Cardiologia do VCM/HOVETUSP. O TACT foi mensurado empregando o Doppler tecidual guiado por cores, sendo o intervalo PA-TDI medido com o cursor do Doppler tecidual posicionado na parede lateral do átrio esquerdo, logo acima do ânulo da valva mitral. Para análise estatística as variáveis foram submetidas ao teste de Shapiro-Wilk e estimou-se o coeficiente de correlação de Pearson ou de Spearman com o TACT. Para verificar as diferenças do TACT quanto às variáveis utilizaram-se o teste t de Student, a ANOVA para medidas não repetidas, o procedimento de Bon Ferroni quando necessário e a ANOVA 2-fatores. Todos os testes estatísticos foram considerados significativos quando p<0,05. No que concerne à investigação das variáveis associadas à fibrilação atrial, foi realizada uma regressão múltipla de Cox considerando o tempo para a ocorrência da fibrilação atrial. De 264 estudos ecocardiográficos disponíveis, 144 foram selecionados. O intervalo PA-TDI médio dos cães livres de doenças cardiovasculares foi de 47 ± 9,09 ms. Observou-se que dentre os fatores estudados, o valor encontrado foi diferente entre animais de porte mini ou pequeno comparados aos animais de porte grande ou gigante (p=0,01) e que sexo, castração e obesidade não influenciaram nos valores de p, respectivamente, 0,50, 0,24 e 0,98. A duração do TACT apresentou correlação com a duração do intervalo PR, frequência cardíaca, peso e concentração de eosinófilos/mm3 de sangue nos animais sem doenças cardiovasculares. Quanto aos animais doentes, o intervalo PA-TDI foi significativamente mais longo nos indivíduos com valvopatia mitral mixomatosa estágios B2 e C (respectivamente, 55,4 ± 7,7 e 55,7 ± 9,2 ms). Além disso, observou-se aumento de 13% de chance de desenvolver fibrilação atrial a cada unidade aumentada do intervalo PA-TDI (p<0,01). Com base nos resultados encontrados, aventa-se a hipótese de que tanto o aumento atrial isolado quanto o advento da ICC possam já acarretar atrasos de condução interatrial e que o TACT mensurado pelo intervalo PA-TDI pode prever a ocorrência de fibrilação atrial secundária à valvopatia mitral mixomatosa em cães, embora a baixa frequência dessa arritmia na população estudada possa ter mascarado resultados mais expressivos neste sentido. / Total Atrial Conduction Time (TACT) reflects the conduction of the electrical stimuli in the atrial territory. It can be measured by electrophysiological study or by noninvasive methods, such as tissue Doppler echocardiography. In human medicine it is known that the time measured between the beginning of the electrocardiogram Pwave to the peak of A-wave tissue Doppler graph (PA-TDI interval) represents and independent predictor of the occurrence of atrial fibrillation secondary to several causes. In veterinary medicine there are no published studies on the subject so far. The objective was to establish normal TACT values for dogs and investigate which variables may influence the results obtained. In addition, TACT was proposed as a predictor of the occurrence of atrial fibrillation in dogs with myxomatous mitral valvopathy. A retrospective study was carried out using a bank of echocardiographic images of dogs that were enroled in research projects at the Cardiology Service of HOVET/FMVZ-USP. TACT was measured using color-coded tissue Doppler, and PATDI interval was measured with tissue Doppler sample positioned on the lateral wall of the left atrium, just above the mitral valve annulus. For statistical analysis, variables were submitted to Shapiro-Wilk test and Pearson or Spearman correlation coefficient with TACT was estimated. Student\'s t-test, ANOVA for non-repeated measurements, Bon Ferroni procedure when necessary, and the 2-factor ANOVA were used to verify the TACT differences. All statistical tests were considered significant when p <0.05. Regarding the investigation of the variables associated with atrial fibrillation, a multiple Cox regression was performed considering the time for the occurrence of atrial fibrillation. From 264 available echocardiographic studies, 144 were selected. The mean PA-TDI interval of dogs free from cardiovascular diseases was 47 ± 9.09 ms. It was observed that among the factors studied, the value found was different between small dogs compared to large or giant animals (p = 0.01) and that gender, castration and obesity had no influence (p values were, respectively, 0.50, 0.24 and 0.98). TACT duration correlated with the duration of the PR interval, heart rate, weight, and blood eosinophil concentration/mm3 in animals without cardiovascular disease. PA-TDI interval was significantly longer in patients with myxomatous mitral valvopathy stage B2 and C (55.4 ± 7.7 and 55.7 ± 9.2 ms, respectively). In addition, there was a 13% increase in the chance of developing atrial fibrillation at each increased unit of PA-TDI interval (p <0.01). On the basis of the results found, it is hypothesized that both isolated atrial strech and the advent of CHF may already lead to atrial conduction delays and that the TACT measured by the PATDI interval may predict the occurrence of atrial fibrillation secondary to myxomatous mitral valve disease in dogs. The low frequency of atrial fibrillation in the studied population may have masked more expressive results.