Cette thèse contribue à l'étude des aimants à effet ressort, connu sous le nom de "Exchange Spring Magnets" (ESM). Nous avons divisé nos travaux en deux parties : une première dite expérimentale et une seconde dite théorique. Dans la première partie, nous avons étudié la synthèse de ESM. Le processus de fabrication débute par le mélange des nano-poudres magnétiques dures et douces. Cette étape a été réalisé par broyage mécanique ainsi que par ultrasons. Les mélanges sont ensuite frittés via la technique de frittage par plasma d'arc (Spark Plasma Sintering (SPS)) afin d'obtenir des aimants massifs. La structure ainsi que les propriétés magnétiques de ces aimants sont étudiées. Les structures obtenues présentent des phases intermédiaires ainsi qu'une réduction de la phase magnétique dure. Nous avons trouvé que ces aimants ont des cycles magnétiques à deux phases indiquant un couplage faible entre les phases précurseurs. De plus, les aimants obtenus à partir du mélange à ultrasons possèdent des meilleurs propriétés magnétiques que ceux obtenus par broyages mécanique ainsi qu'une plus grande capacité à obtenir un meilleur couplage. Dans la seconde partie, nous avons étudié l'utilité des ESM pour l'enregistrement magnétique perpendiculaire. Il a été montré qu'en y appliquant un faible champ micro-onde , l'ESM présente un basculement facilité ainsi qu'une meilleur stabilité thermique, ce qui est désirable pour l'enregistrement magnétique. Nous avons étudié les dynamiques non-linéaires de basculement magnétique dans un ESM perpendiculaire en utilisant l'équation de Landau-Lifshitz. A partir d'un nouveau code FORTRAN basé sur la méthode des différences finies, nous avons étudié les solutions numériques de cette équation. Le système bi-couche présente des solutions stables nommés modes-P globaux et une diminution du champ nécessaire au basculement pour une grande gamme de fréquence. Le profil de la magnétisation dans le système étendu à été étudié numériquement. L'étude analytique des dynamiques est commentée en faisant l'approximation que deux modes-P sont couplés par un champ d'interaction. L'"ansatz" utilisé pour décrire le champ d'interaction est validé par comparaison avec les résultats numériques. Cette approche s'est révélée être équivalente à l'étude de deux macro-spins couplés. / This thesis is divided into two parts, experimental part presenting the synthesis of exchange spring magnets and theoretical part showing the magnetization dynamics of exchange spring magnets. For the synthesis, ferrite and metal alloy based magnets using mechanical milling and ultrasonic mixing are studied. This part discusses the difficulty in controlling the microstructure during synthesis. Several samples with varying volume fractions of the soft phase are synthesized. It is found that mechanical milling initiates a reaction and this leads to disintegration of the ferrite phase into its intermediate phases. Samples prepared using ultrasonic mixing however show presence of the ferrite phase up to very large volume fraction of the soft alloy phase which provides better perspective for the synthesis. Dynamics of the exchange spring system is studied theoretically using micromagnetic theory. Microwave assisted magnetization reversal are studied in the bulk bilayer exchange coupled system. We investigate the nonlinear magnetization reversal dynamics in a perpendicular exchange spring media using the Landau-Lifshitz equation. In the limit of the infinite thickness of the system, the propagation field leads the reversal of the system. The reduction of the switching field and the magnetization profile in the extended system are studied numerically. The possibility to study the dynamics analytically is discussed and an approximation where two P-modes are coupled by an interaction field is presented. The ansatz used for the interaction field is validated by comparison with the numerical results. This approach is shown to be equivalent to two exchange coupled macrospins.

L’industrialisation de la production des conducteurs utilisant les cuprates à haute température de transition pendant les deux dernières décennies offre aujourd’hui la possibilité de nouveaux développements dans le domaine du magnétisme supraconducteur. En particulier, nous nous intéressons ici aux applications des rubans supraconducteurs REBCO. Ceux-ci font partie des rares candidats disponibles à l’heure actuelle capables de fonctionner sous des inductions supérieures à 20 T aux températures de cryogénie classique (4,2 K), ce qui les rend quasiment incontournables dans le contexte des applications haut-champ. Parallèlement, leurs températures critiques élevées apportent également la possibilité de nouvelles conceptions d’aimants pouvant fonctionner à des températures allant jusqu’à 80 K environ. Ceci ouvre notamment la porte à de nouvelles options de cryogénie moins coûteuses, comme le refroidissement par azote liquide. Néanmoins, la mise en œuvre des rubans REBCO à l’intérieur de nouveaux aimants supraconducteurs se heurte dans certains cas au problème posé par les courants d’écrantages. Ce terme désigne les supercourants induits dans le matériau supraconducteur en réponse au champ extérieur. Ces courants parasites engendrent alors des perturbations sur le champ magnétique total produit par l’aimant : réduction de son amplitude, distorsions spatiales et instabilités temporelles. Ces multiples effets entrent généralement en collision avec les exigences sur la qualité de champ produite. Ainsi, comprendre le phénomène des courants d’écrantages, être capable de prédire sa dynamique et, idéalement, disposer de solutions afin d’y remédier, sont des tâches cruciales pour assurer la réussite des futurs projets d’aimants supraconducteurs REBCO. Cette thèse propose des éléments de réponse vis-à-vis de cette problématique. Le phénomène de courants d’écrantages y est tout d’abord approché de manière expérimentale au travers de la mesure de ses effets perturbateurs sur le champ magnétique de petits aimants tests REBCO. Les données expérimentales récoltées sont ensuite employées pour étayer le développement d’un modèle numérique de simulation magnétodynamique sous l’environnement open-source GetDP. Ce modèle est par ailleurs mis à l’essai avec de bons résultats dans le cas du dipôle 5,4 T EuCARD du CERN. Enfin dans une dernière partie, une investigation expérimentale est menée sur deux techniques capables de réduire, voire de supprimer intégralement, les impacts négatifs des courants d’écrantages sur le champ des aimants REBCO : le vortex shaking et l’overshoot en courant. Au travers de ces développements, cette thèse entend fournir aux concepteurs de futurs aimants supraconducteurs REBCO des moyens d’anticiper et de traiter les problèmes posés par les courants d’écrantages. / The industrialization of the production of high-transition temperature, cuprate-based conductors during the last two decades brings nowadays the possibility of new developments in the field of superconducting magnetism. In particular, we are interested here in the applications of superconducting REBCO tapes. Those belong to the few currently available candidates that are able to operate under magnetic induction above 20 T at classical cryogenic temperature (4.2 K), what makes them almost unavoidable in the scope of high-field applications. Besides, their high critical temperatures also bring the possibility for new magnet designs capable to operate up to temperatures of about 80 K. This notably opens the door to new, less expensive, cryogenic design options, like liquid nitrogen cooling.However, the practical implementation of REBCO tapes inside new superconducting magnets encounters a problem posed by screening currents. The latter term designates the supercurrents that are induced in the superconducting material in response to an external field. These parasitic currents then give rise to perturbations in the total magnetic field generated by the magnet: lowering of its magnitude, spatial distortions and time instabilities. Those various effects collide most of the time with the requirements on the produced field quality. Thus, understanding the phenomenon of screening currents, being able to predict its dynamic and, ideally, having of solutions to remedy this issue, are crucial tasks to ensure the success of future REBCO superconducting magnet projects.This Ph.D. work suggests some elements of answer with regard to this problematic. The phenomenon of screening currents is first approached experimentally, through the measurement of its disruptive effects of the magnetic field generated by small REBCO tests magnets. The experimental data gathered are then used to support the development of a numerical model for magnetodynamic simulations in the framework of the open-source environment GetDP. This model is furthermore tested with satisfactory results on the scenario of the 5.4 T EuCARD dipole developed by CERN. Finally, in a last section, an experimental investigation is conducted on two techniques capable at reducing, or even entirely suppressing, the negative impacts of the screening currents on the field of REBCO magnets: the vortex shaking and current overshoot.Across these developments, this Ph.D. work aims at providing the designers of future REBCO superconducting magnets with tools to anticipate and handle the issues caused by the screening currents.

O composto magnético geometricamente frustrado Gd2Ti2O7 da família dos pirocloros apresenta um comportamento bastante interessante, sendo que a natureza da fase magnética em baixas temperaturas se encontra ainda sob intenso debate. Este material entra em um estado antiferromagnético parcialmente ordenado à temperatura T1N ~ 1 K, apresentando outra transição de fase em T2N ~ 0.7 K. Neste trabalho é investigada a física de baixas temperaturas de amostras de Gd2Ti2O7 com defeitos estruturais tais como vacâncias de oxigênio e diluição de ítrio. Amostras policristalinas com composição Gd2Ti2O7 e Gd2-xYxTi2O7 foram sintetizadas em diferentes condições por uma rota alternativa conhecida como método sol-gel. O refinamento de um modelo para os dados de difração de raios X mostra que vacâncias de oxigênio são os principais defeitos estruturais neste material. As vacâncias de oxigênio resultam numa ligeira diminuição de T1N e numa redução da magnetização de saturação. A diluição da rede com ítrio leva a uma clara diminuição de T1N e da temperatura de Curie-Weiss. Medidas de calor específico evidenciaram as duas transições T1N e T2N no composto com menor grau de vacâncias de oxigênio. A análise da contribuição magnética ao calor específico Cm em baixas temperaturas, 0.39 K < T < 0.68 K, revelou um comportamento proporcional a T^2 previamente discutido na literatura. Entretanto, verificamos que uma dependência T^3, usualmente encontrada em antiferromagnetos convencionais, descreve igualmente bem nossos dados experimentais resultando em uma velocidade de magnons consistente com a apresentada por outros pirocloros. / The geometrically frustrated compound Gd2Ti2O7 of the pyrochlore family displays such an interesting behaviour that the nature of the ordered magnetic phase at low temperatures is still under intense discussion. This material enters in a partially ordered magnetic state at a temperature T1N ~ 1 K, and there is another phase transition at T2N ~ 0.7 K. In this thesis we study the low temperature physics of Gd2Ti2O7 with structural defects such as oxygen vacancies and yttrium dilution. Polycrystalline samples of Gd2Ti2O7 and Gd2-xYxTi2O7 were synthesized in different conditions by an alternative route known as the sol-gel method. The refinement of a model for the X-ray diffraction data reveal that the oxygen vacancies are the leading defects in this material. The oxygen vacancies result in a slight decrease of T1N and in a reduction of the saturation magnetization. The yttrium dilution of the lattice leads to a clear reduction of T1N and of the Curie-Weiss temperature. Specific heat measurements display both transitions T1N and T2N in the compound with lower degree of oxygen vacancies. The analysis of the magnetic contribution to the specific heat Cm at low temperatures, 0.39 K < T < 0.68 K, reveals a behaviour proportional to T^2 previously discussed in the literature. However, we verify that a dependence T^3, usually found in standard antiferromagnets, describes similarly well our experimental data resulting in a velocity of magnons consistent with the ones exhibited for another pyrochlores.

Realizamos medidas Mossbauer a temperatura ambiente das ligas de SM IND.2 (CO IND.1-XFE IND.X) ind.17, COM X = 0.1 A 0.6, E DAS LIGAS sm (CO IND.0.90-VFE IND.VCU IND.0.08ZR IND.0.02) ind.8.35, Com V variando de 0.23 A 0.27. Estas amostras foram caracterizadas por difração de raio X. elas apresentaram a estrutura romboédrica do tipo th ind.2zn ind.17 (R3barraM) com os átomos de Fe ocupando quatro sítios cristalográficos distintos todos os espectros obtidos apresentaram desdobramento magnético e um nível de complexidade alto, resultado da sobreposição de quatro sítios distintos de átomos de fe. Os espectros foram ajustados e através desse ajuste determinados os parâmetros hiperfinos associados a cada sítio e suas respectivas populações. Os resultados obtidos das populações dos sítios nas series de compostos Sm ind.2 (CO IND.1-XFE IND.X) ind.17 revelaram a forte preferência dos átomos de FE de ocupar os sítios halteres (2c). No modelo proposto por ray, à função do ZR e o de deslocar pares de FE dos sítios halteres, substituindo-os por pares ZR-vacância. Através dos resultados obtidos pudemos também comprovar esta hipótese através da análise das populações dos sítios das duas series de amostras. / Mössbauer measurements were carried out at room temperature for Sm2(Co1-xFex)17 alloys with x=0,1 to 0,6 and Sm2(Co0.9-vFevCu0.08Zr0.02)8.35, with v=0.23 to 0.27. All samples were characterized by X-ray diffraction. They showed rhombohedral structure of th2Zn17 type (R3m), where Fe atoms occupy four crystallographic sites. All spectra showed magnetic splitting and a high complexity resulting from the superposition of four Fe sites. Hyperfine parameters and site populations were obtained by least-squares fitting of the spectra. For Sm2(Co1-xFex)17 alloys, Fe atoms show a strong preference for 2c (dumbbell) sites. In the Zr-containing alloys, on the other hand, this tendency is significantly reduced, confirming A. E. Rays hypothesis on the role of Zr in the improvement of the magnetic properties of these alloys through metallurgical treatments.

No intuito de simplificar a gastrostomia endoscópica percutânea (GEP) suprimindo- se a utilização de um endoscópio ou métodos de imagem, a gastrostomia magnética percutânea (GMP) foi concebida como nova opção técnica. Objetivo: investigar factibilidade de nova técnica para GMP por estudo experimental controlado, comparada à técnica GEP. Método: quatorze porcos foram submetidos a GEP (grupo ENDO, 7 animais) ou a GMP (grupo MAG), e submetidos a eutanásia 7 dias após o procedimento. Foram avaliados sucesso técnico, peso, tempo de procedimento e número de ocorrências/complicações (complicação menor – sem necessidade de nova intervenção). Para análise estatística foram aplicados o teste não paramétrico de Mann-Whitney e o teste exato de Fisher. Valores de p<0.05 indicaram significância estatística. Resultados: Todos os procedimentos realizados foram tecnicamente bem-sucedidos. Doze animais apresentaram evolução pós-operatória favorável. Dois animais (um de cada grupo) foram a óbito precocemente, não atribuível ao acesso gástrico. Houve no total 5 ocorrências/ complicações menores (4 no grupo MAG), sendo 3 dessas relacionadas ao tubo de gastrostomia. Houve maior perda de peso no grupo MAG comparado ao grupo ENDO, com significância estatística (peso médio em gramas 115±131(desvio-padrão) e -83±128, respectivamente, p=0,04). O tempo de procedimento foi maior para o grupo MAG comparado ao grupo ENDO, com significância estatística (tempo médio em segundos 471±140 e 882±239, respectivamente, p=0,001). Conclusão: a gastrostomia percutânea magnética é tecnicamente factível, porém necessita de melhorias no dispositivo devido a complicações relacionadas ao tubo de gastrostomia. Comparado ao método tradicional endoscópico, a gastrostomia magnética apresentou maior tempo de procedimento e maior perda de peso. / In order to simplify a Percutaneus Endoscopic Gastrostomy (PEG) procedure and obviate the need of imaging methods, a Percutaneous Magneticallyguided Gastrostomy (PMG) has been conceived. OBJECTIVE to investigate preclinical application of a novel magnetic device for PMG through an experimental controlled trial compared to PEG. METHODS fourteen domestic pigs were assigned for ENDO group (undergoing PEG, 7 animals), and MAG group (undergoing PMG). All animals were euthanized 7 days after procedure. Aspects related to technical success, procedure duration and clinical outcome (weight changes, event/complication rate) were evaluated. These were statisctically evaluated using Mann-Whitney (non-parametric) and Fisher’s exact test. Values of p<0.05 were considered statistically significant. RESULTS technical success was achieved for all animals undergoing the procedures. Of these, 12 had a favorable clinical outcome. Two animals died in less than 24 hours (one from each group), not directly related to gastric access. There were 4 minor complications (4 of them for group MAG), and 3 of these were tube-related. Weight loss was statistically significantly higher for group MAG (mean weight in grams 115±131(standard-deviation) e - 83±128, respectively, p=0.04). The procedural time was statistically significant higher for group MAG compared to ENDO (mean time in seconds 471±140 e 882±239, respectively, p=0.001). CONCLUSION a magnetically-guided percutaneous gastrostomy without imaging is technically feasible, however a device improvement is needed because of tube-related complications. Compared to percutaneous endoscopic gastrostomy, the magnetically-guided percutaneous gastrostomy technique is more time-consuming and had more weight loss.

O composto magnético geometricamente frustrado Gd2Ti2O7 da família dos pirocloros apresenta um comportamento bastante interessante, sendo que a natureza da fase magnética em baixas temperaturas se encontra ainda sob intenso debate. Este material entra em um estado antiferromagnético parcialmente ordenado à temperatura T1N ~ 1 K, apresentando outra transição de fase em T2N ~ 0.7 K. Neste trabalho é investigada a física de baixas temperaturas de amostras de Gd2Ti2O7 com defeitos estruturais tais como vacâncias de oxigênio e diluição de ítrio. Amostras policristalinas com composição Gd2Ti2O7 e Gd2-xYxTi2O7 foram sintetizadas em diferentes condições por uma rota alternativa conhecida como método sol-gel. O refinamento de um modelo para os dados de difração de raios X mostra que vacâncias de oxigênio são os principais defeitos estruturais neste material. As vacâncias de oxigênio resultam numa ligeira diminuição de T1N e numa redução da magnetização de saturação. A diluição da rede com ítrio leva a uma clara diminuição de T1N e da temperatura de Curie-Weiss. Medidas de calor específico evidenciaram as duas transições T1N e T2N no composto com menor grau de vacâncias de oxigênio. A análise da contribuição magnética ao calor específico Cm em baixas temperaturas, 0.39 K < T < 0.68 K, revelou um comportamento proporcional a T^2 previamente discutido na literatura. Entretanto, verificamos que uma dependência T^3, usualmente encontrada em antiferromagnetos convencionais, descreve igualmente bem nossos dados experimentais resultando em uma velocidade de magnons consistente com a apresentada por outros pirocloros. / The geometrically frustrated compound Gd2Ti2O7 of the pyrochlore family displays such an interesting behaviour that the nature of the ordered magnetic phase at low temperatures is still under intense discussion. This material enters in a partially ordered magnetic state at a temperature T1N ~ 1 K, and there is another phase transition at T2N ~ 0.7 K. In this thesis we study the low temperature physics of Gd2Ti2O7 with structural defects such as oxygen vacancies and yttrium dilution. Polycrystalline samples of Gd2Ti2O7 and Gd2-xYxTi2O7 were synthesized in different conditions by an alternative route known as the sol-gel method. The refinement of a model for the X-ray diffraction data reveal that the oxygen vacancies are the leading defects in this material. The oxygen vacancies result in a slight decrease of T1N and in a reduction of the saturation magnetization. The yttrium dilution of the lattice leads to a clear reduction of T1N and of the Curie-Weiss temperature. Specific heat measurements display both transitions T1N and T2N in the compound with lower degree of oxygen vacancies. The analysis of the magnetic contribution to the specific heat Cm at low temperatures, 0.39 K < T < 0.68 K, reveals a behaviour proportional to T^2 previously discussed in the literature. However, we verify that a dependence T^3, usually found in standard antiferromagnets, describes similarly well our experimental data resulting in a velocity of magnons consistent with the ones exhibited for another pyrochlores.

Imãs são corpos de materiais magnetizáveis utilizados para gerar um forte campo magnético em sua vizinhança. Essa característica faz com que sejam empregados em diversas aplicações na vida moderna, tais como motores para a indústria eletroeletrônica e automobilística, como elementos de fixação e em acoplamentos magnéticos na indústria mecânica. O mercado de ímãs permanentes movimenta atualmente cerca de US$ 5 bilhões por ano e está em plena expansão, sendo estimado um crescimento de 12% por ano até o final desta década de 90. Tal crescimento é atribuído aos \"novos materiais\" desenvolvidos a partir da década de 80 (ímãs de terras-raras) e aos novos mercados, que foram gerados pelos próprios \"novos materiais\" (Hart, 1996). Tratando-se de materiais tão ligados às facilidades da vida moderna, a pesquisa na área de imãs está intimamente ligada ao seu desenvolvimento tecnológico. Procura-se desenvolver imãs com as melhores propriedades magnéticas para as suas aplicações, investigando ao mesmo tempo os fenômenos que regem os mecanismos físicos de magnetização. / Permanent magnets composed of magnetic powders bonded with a polymer represent the fastest growing sector of the magnetic materials market since they are ideal for the fabrication of small motors. This work presents a magnetic and structural characterization of TIve eommereial bonded magnets, Reversible and irreversible components of the total magnetization as well as magnetic interactions in the five commereial magnets are also studied, the magnets are composed by ferrite and MQP-Q nanoerystelline powders and mixtures of these two powders with 80%, 60% and 40% ferrite. Magnelie interactions were analyzed by Henkel plots, gM plots and switching field distributions. In bonded magnets, since the magnetic particles are separated from each other by a binder, it is expected that interactions are mainly dipolar in Nature. There is a progressive change in the data as the fraction of MQP-Q powder is increased. The sample with 100% ferrite shows strong magnetizing interactions at low fields. Date for hybrid magnets present increasing demagnetizing interactions as the fermion of MQP-Q increases and for the 100% MQP-Q sample, the data indicate demagnetizing effects. Reversible and irreversible magnetization components were obtained by applying two methods commanly used in magnetic materials characterization, the DCD -IRM method and the reversible susceptibility method. For the 100% ferrite magnet, in which the reversible component is small, the methods lead lo similar results. The results for both methods diverge as the reversible component! Increases, which in this case occurs with the increase of the MQP-Q powder fraction. The divergence is attributed to the idealized conditions of non-inleracting particles assumed by the DCD-IRM method. Magnetic interactions and lotai magnetization components were also studied in a melt-spun nanocrystalline Nd9Fe85B5 sample. This composition is similar to that of the MQP-Q powder and the magnetic behavior of both the bonded magnetic and the nanocrystalline precursor could be compared. Micromagnetic simulations allowed the evaluation of exchange, anisotropy and magnetostatic interactions on the magnetization reversal of nanocryslalline romposlle systems. The Monte Carlo method was applied lo a linear array of 300 magnetic moments distributed in three grains, two magnetically hard with a soft grain between them.

The world demand for energy is estimated to increase by up to 70% from 2016 to 2040. To meet this demand in a sustainable way, the power density of electric motors and generators can be increased by using superconducting materials. In particular, trapped field superconducting magnets, where the field is generated by a circulating persistent current in the sample, can create magnetic fields an order of magnitude higher than possible using conventional ferromagnets, thus increasing the power density of motors and generators. This is of great interest where added weight comes at a premium, such as for naval propulsion motors, wind generators and motors/generators for future electric aircraft. This work investigates the suitability of stacked tape layers of second generation high temperature superconductors (HTS), such as YBa2Cu3O7-x (YBCO) for trapped field applications. The present limits for trapped field magnitude have been determined, which provide a basis for the optimization of pulsed field magnetization techniques for in-situ magnetization in motors and generators. Trapped fields were increased by optimising the magnetic pulse sequence, using thermally conductive material to reduce temperature rise during pulse and changing the duration of the magnetic field pulse. Finite element method computer modelling was used to model and predict the behaviour of the trapped field magnets made of HTS tape with good agreement to experiment for both field cooling and pulsed field magnetisation. The models rely on critical current data for the HTS tape and its dependence on magnetic field and temperature. For this reason, a critical current testing facility was developed and constructed as a part of this work capable of measuring critical current up to 900 A, magnetic field of 1.5 T and down to temperatures of ~10 K in forced and dynamically controlled helium vapour flow. Lastly, first steps into scaling up by pulse magnetising an array of HTS tape stacks were made, allowing for larger overall trapped flux values. Such an array exhibits geometry, similar to what is going to be used in a functional motor prototype being developed in our research group (Applied Superconductivity and Cryoscience Group, ASCG). The work done culminated in the highest trapped field achieved to date using both field cooling (13.4 T between two stacks) and pulsed field magnetization (2.1 T above a single stack), for this type of trapped field magnet.

O presente trabalho teve como objetivo: projetar, construir e testar os imas principais do acelerador race-track microtron booster do ifusp, que tem como funcao defletir e focalizar o feixe de eletrons, permitindo a sua recirculacao. Esses imas desempenham papel fundamental quanto a qualidade do feixe obtido que, por sua vez, constitui um fator primordial para experiencias em fisica nuclear. O trabalho desenvolve a teoria do funcionamento do acelerador e os calculos analiticos que fundamentam a geometria dos imas. A ausencia da uniformidade do campo magnetico e a presenca de campos de borda inadequados comprometem o sincronismo. Utilizou-se bobinas corretoras confeccionadas na forma de circuitos impressos, que limitaram a falta de uniformidade a valores relativos de campo magnetico da ordem de \'10 POT.-5\'. Os imas possuem pecas polares auxiliares (clamps) que criam campos de borda reversos ao campo principal. Os clamps foram projetados por meio de simulacoes realizadas com os programas poisson e ptrace, de forma a se obter uma distribuicao para o campo magnetico que propicie aceleracao para uma distancia conveniente entre os imas / The present work had the purpose of designing, building and testing the bending magnets of the IFUSP race-track microtron booster that deflect and focus the electron beam and allow its recirculation. These magnets play a fundamental role in terms of the beam quality, which is important in Nuclear Physics research. This work develops the theory conceming the functioning of the accelerator and the analytical calculations that establish the geometry ofthe magnets. The lack of magnetic field uniformity and the presence of inadequate fringing fields damage the synchronism. The magnetic fields were corrected using coils made of etched printed circuit boards. This method provides uniformity of a few parts in 10 \'10 POT.-5\' . The magnets have auxiliary poles (clamps) that produce reverse fringing fields. The clamps were designed using Poisson and Ptrace programs in order to get a magnetic field distribution that allows acceleration for a convenient distance between the magnets. The beam trajectories determined by Ptrace program were confirmed with agreement of 0,1% using the floating wire technique.

Thesis (Ph.D.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 1981. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND ENGINEERING. / Vita. / Includes bibliographical references. / by Joseph Vito Minervini. / Ph.D.