O processo de injeção pelas ventaneiras dos altos-fornos (Pulverized Coal Injection - PCI) é uma das tecnologias mais promissoras para a incorporação de biomassas termicamente tratadas na siderurgia e um dos meios de alcançar uma redução consistente nas emissões de CO2 no setor. O objetivo deste trabalho foi avaliar a combustibilidade e reatividade ao CO2 de biomassas de madeira e caroço de azeitona tratadas em laboratório desde temperaturas de torrefação (250°C) até de carbonização (450°C) e comparar com carvões típicos utilizados em PCI, correlacionando com as características ocorridas devido aos tratamentos térmicos. Além da caracterização química, as transformações devido aos tratamentos térmicos das biomassas foram avaliadas via testes de combustão em termobalança, técnicas de microscopia ótica e eletrônica, espectroscopia de infravermelho por transformada de Fourier (FTIR) e técnicas de adsorção para análise da porosidade. Testes de combustibilidade foram conduzidos em um forno de queda livre (Drop Tube Furnace - DTF) em atmosferas convencional (O2/N2) e de oxi-combustão (O2/CO2) e os chars resultantes destes testes foram caracterizados quanto à estrutura e à reatividade ao CO2 em termobalança. Além disso, foram feitos testes de reatividade ao CO2 de misturas de eucalipto termicamente tratado e carvões em termobalança. A torrefação manteve o alto teor de voláteis das biomassas, enquanto que as biomassas carbonizadas apresentaram teores de carbono e poder calorífico semelhantes aos dos carvões de mais alto rank, com as vantagens típicas de biomassas de manterem um baixo teor de cinzas e enxofre. No entanto, o elevado teor de álcalis e fósforo nas cinzas pode ser um fator limitante na composição de misturas para PCI. O tratamento térmico das biomassas levou a gradual decomposição dos componentes da madeira com uma progressiva homogeneização da estrutura celular, associada a um aumento de aromaticidade e porosidade. De uma maneira geral, quanto menor foi a temperatura de tratamento térmico das biomassas, maior foi o burnout obtido no DTF. Comparada à atmosfera convencional (O2/N2), a atmosfera de oxicombustão (O2/CO2) levou a maiores burnouts para os chars de todas as biomassas e carvões. As biomassas carbonizadas apresentaram burnouts mais elevados que o carvão de mais baixo rank e o caroço de azeitona carbonizado apresentou baixa conversão, equivalente a um carvão de alto rank. Os chars das biomassas torrefeitas apresentaram estruturas cenosféricas isotrópicas de elevada porosidade nas paredes enquanto que os chars das carbonizadas preservaram a morfologia apresentada nas amostras originais. Os chars das biomassas foram altamente porosos, com áreas superficiais de meso e microporos em média 15 e 5 vezes maior que os chars dos carvões, respectivamente. Com relação aos testes de reatividade ao CO2 em termobalança, em geral, a reatividade dos chars das biomassas torrefeitas foi maior do que a reatividade dos chars das biomassas carbonizadas e estes foram pelo menos 10 vezes mais reativos ao CO2 do que o chars do carvão de mais baixo rank. Além das maiores áreas superficiais, principalmente o ordenamento da estrutura carbonosa e a morfologia foram fundamentais nas diferenças de reatividade ao CO2 entre os chars das biomassas e dos carvões. As misturas do carvão de mais baixo rank com a biomassa carbonizada apresentaram os melhores resultados em termos de aditividade na reatividade ao CO2. / Pulverized Coal Injection (PCI) in the blast furnace tuyeres is a promising technology for incorporation of thermally-treated biomasses and it is a way to reduce CO2 emissions in ironmaking processes. The aim of this work was to evaluate combustibility and CO2 reactivity of laboratory torrefied (250°C) and carbonized (450°) olive stone and woody biomasses, comparing with typical PCI coals. The transformations produced in biomasses due to torrefaction and carbonization were evaluated by chemical analyses, combustion tests in thermobalance, Fourier Transform Infrared Spectroscopy (FTIR) and optical and electron microscopy and adsorption techniques. Combustion experiments were carried out in a Drop Tube Furnace (DTF) under conventional (O2/N2) and oxy-fuel (O2/CO2) atmospheres and the chars collected were characterized by its structure and CO2 reactivity in thermobalance. Reactivity tests were also conducted in thermobalance with blends of thermally-treated eucalyptus and coals. Torrefied samples maintained high contents of volatile matter, typical of raw biomasses, while carbonized biomasses showed carbon contents and high heating values similar to that of high rank coals, retaining low ash and sulfur contents. However, its high alkali and phosphorus contents could be a limiting factor to the use in blends for PCI. The thermal treatments of biomasses lead to a gradual decomposition of wood components and to a progressive homogenization of cell structure, associated to an increase in aromaticity and porosity. In general, the lower the thermal treatment temperature, the higher was the burnout in the DTF. Compared to conventional atmosphere, oxy-fuel combustion led to the highest burnouts for all biomass chars. The carbonized biomasses showed higher burnouts than the high-volatile coal and olive stone showed burnouts similar to a low-volatile coal. The chars from the torrefied biomasses showed isotropic cenospheric structures with high porosity within the walls and the chars from the carbonized biomasses preserved the morphology seen in original carbonized samples. The biomass chars presented highly porosity, with micro and mesoporosity in average, 5 and 15 times greater than the coal chars, respectively. In relation to the CO2 reactivity tests, in general, the torrefied biomass chars were more reactive than the carbonized biomass chars. However, due to its higher surface areas, structure arrangement and morphology, the carbonized biomass chars were at least 10 times more reactive than the high-volatile coal chars. The blends of high-volatile coal and carbonized eucalyptus showed good additivity in the CO2 reactivity tests in thermobalance.

O processo de injeção pelas ventaneiras dos altos-fornos (Pulverized Coal Injection - PCI) é uma das tecnologias mais promissoras para a incorporação de biomassas termicamente tratadas na siderurgia e um dos meios de alcançar uma redução consistente nas emissões de CO2 no setor. O objetivo deste trabalho foi avaliar a combustibilidade e reatividade ao CO2 de biomassas de madeira e caroço de azeitona tratadas em laboratório desde temperaturas de torrefação (250°C) até de carbonização (450°C) e comparar com carvões típicos utilizados em PCI, correlacionando com as características ocorridas devido aos tratamentos térmicos. Além da caracterização química, as transformações devido aos tratamentos térmicos das biomassas foram avaliadas via testes de combustão em termobalança, técnicas de microscopia ótica e eletrônica, espectroscopia de infravermelho por transformada de Fourier (FTIR) e técnicas de adsorção para análise da porosidade. Testes de combustibilidade foram conduzidos em um forno de queda livre (Drop Tube Furnace - DTF) em atmosferas convencional (O2/N2) e de oxi-combustão (O2/CO2) e os chars resultantes destes testes foram caracterizados quanto à estrutura e à reatividade ao CO2 em termobalança. Além disso, foram feitos testes de reatividade ao CO2 de misturas de eucalipto termicamente tratado e carvões em termobalança. A torrefação manteve o alto teor de voláteis das biomassas, enquanto que as biomassas carbonizadas apresentaram teores de carbono e poder calorífico semelhantes aos dos carvões de mais alto rank, com as vantagens típicas de biomassas de manterem um baixo teor de cinzas e enxofre. No entanto, o elevado teor de álcalis e fósforo nas cinzas pode ser um fator limitante na composição de misturas para PCI. O tratamento térmico das biomassas levou a gradual decomposição dos componentes da madeira com uma progressiva homogeneização da estrutura celular, associada a um aumento de aromaticidade e porosidade. De uma maneira geral, quanto menor foi a temperatura de tratamento térmico das biomassas, maior foi o burnout obtido no DTF. Comparada à atmosfera convencional (O2/N2), a atmosfera de oxicombustão (O2/CO2) levou a maiores burnouts para os chars de todas as biomassas e carvões. As biomassas carbonizadas apresentaram burnouts mais elevados que o carvão de mais baixo rank e o caroço de azeitona carbonizado apresentou baixa conversão, equivalente a um carvão de alto rank. Os chars das biomassas torrefeitas apresentaram estruturas cenosféricas isotrópicas de elevada porosidade nas paredes enquanto que os chars das carbonizadas preservaram a morfologia apresentada nas amostras originais. Os chars das biomassas foram altamente porosos, com áreas superficiais de meso e microporos em média 15 e 5 vezes maior que os chars dos carvões, respectivamente. Com relação aos testes de reatividade ao CO2 em termobalança, em geral, a reatividade dos chars das biomassas torrefeitas foi maior do que a reatividade dos chars das biomassas carbonizadas e estes foram pelo menos 10 vezes mais reativos ao CO2 do que o chars do carvão de mais baixo rank. Além das maiores áreas superficiais, principalmente o ordenamento da estrutura carbonosa e a morfologia foram fundamentais nas diferenças de reatividade ao CO2 entre os chars das biomassas e dos carvões. As misturas do carvão de mais baixo rank com a biomassa carbonizada apresentaram os melhores resultados em termos de aditividade na reatividade ao CO2. / Pulverized Coal Injection (PCI) in the blast furnace tuyeres is a promising technology for incorporation of thermally-treated biomasses and it is a way to reduce CO2 emissions in ironmaking processes. The aim of this work was to evaluate combustibility and CO2 reactivity of laboratory torrefied (250°C) and carbonized (450°) olive stone and woody biomasses, comparing with typical PCI coals. The transformations produced in biomasses due to torrefaction and carbonization were evaluated by chemical analyses, combustion tests in thermobalance, Fourier Transform Infrared Spectroscopy (FTIR) and optical and electron microscopy and adsorption techniques. Combustion experiments were carried out in a Drop Tube Furnace (DTF) under conventional (O2/N2) and oxy-fuel (O2/CO2) atmospheres and the chars collected were characterized by its structure and CO2 reactivity in thermobalance. Reactivity tests were also conducted in thermobalance with blends of thermally-treated eucalyptus and coals. Torrefied samples maintained high contents of volatile matter, typical of raw biomasses, while carbonized biomasses showed carbon contents and high heating values similar to that of high rank coals, retaining low ash and sulfur contents. However, its high alkali and phosphorus contents could be a limiting factor to the use in blends for PCI. The thermal treatments of biomasses lead to a gradual decomposition of wood components and to a progressive homogenization of cell structure, associated to an increase in aromaticity and porosity. In general, the lower the thermal treatment temperature, the higher was the burnout in the DTF. Compared to conventional atmosphere, oxy-fuel combustion led to the highest burnouts for all biomass chars. The carbonized biomasses showed higher burnouts than the high-volatile coal and olive stone showed burnouts similar to a low-volatile coal. The chars from the torrefied biomasses showed isotropic cenospheric structures with high porosity within the walls and the chars from the carbonized biomasses preserved the morphology seen in original carbonized samples. The biomass chars presented highly porosity, with micro and mesoporosity in average, 5 and 15 times greater than the coal chars, respectively. In relation to the CO2 reactivity tests, in general, the torrefied biomass chars were more reactive than the carbonized biomass chars. However, due to its higher surface areas, structure arrangement and morphology, the carbonized biomass chars were at least 10 times more reactive than the high-volatile coal chars. The blends of high-volatile coal and carbonized eucalyptus showed good additivity in the CO2 reactivity tests in thermobalance.

Development of a PCI card for PCI bus based host systems, which allows for analysis of ATM data streams on ATM transmission links with a data rate of 622Mbit/s. Implementation and previous deliberations and therefrom and from given specifications arisen partial tasks Fields of application of the ATM-Analyzer board Depiction of particular development steps Functional description of the ATM-Analyzer Explanations on how to put the PCB and the employed test software into operation Assessment of test results / Entwicklung einer Einsteckkarte für PCI-Bus-basierte Hostsysteme zur Analyse von Datenströmen auf ATM-Übertragungsstrecken mit einer Geschwindigkeit von 622 Mbit/s. Realisierung und damit verbundene Vorüberlegungen und daraus und aus gegebenen Spezifikationen erwachsene Teilaufgaben Anwendungsgebiete der Analyzer-Karte Darstellung der einzelnen Entwicklungsschritte Funktionsbeschreibung des ATM-Analyzers Erläuterungen zur Inbetriebnahme der Leiterkarte und der verwendeten Testsoftware Bewertung der Testergebnisse

The thesis From the sickle and hammer to the oak. The role of the opinion currents in the Italian Communist Party during the transformation process deals with the transformation of the Italian Communist Party (Partito Comunista Italiano, PCI) into the post-communist Democratic Party of the Left (Partito Democratico della Sinistra, PDS) and with the role which the intraparty opinion currents played in this process. It will be evaluated their influence on the very own process of the party transformation and their key role in influencing the membership base. The thesis will also focus on analysing of political behavior of the actors on the middle level of decision-making in the party (secretaries of local sections, bosses of trade unions, local and regional politicians) in the light of precipitous events of the examined period. It will be also mentioned the external factors which influenced actions of single groups and how these groups responded to the new challenges in front of which were put almost ad hoc. Subject of this work is defined by the open outbreak of the intraparty crisis during the succession of Achille Occhetto to the post of the secretary general in 1988 and by the parliamentary elections in 1994 in which the party entered fully prepared after the complete conversion and nevertheless...

Den snabbt växande utvecklingen av AI-baserade applikationer och den stora mängden data dessa applikationer behandlar ställer ökade krav på prestanda och optimering av datorsystemen. För att tillfredsställa de växande datorbehoven används hårdvaruacceleratorer som förbättrar databehandlingshastigheten genom att avlasta den befintliga utrustningen genom att hjälpa till med uppgifter och komplexa beräkningar. De befintliga lösningarna som används i dagsläget är kostsamma och MT-FoU på Umeå Universitetssjukhus efterfrågar därför en alternativ lösning i form av att kombinera mindre integrerande acceleratorer på ett större PCIe-kort. I detta examensarbete designas och implementeras en AI-accelerator bestående av fyra Google Coral Dual Edge TPU M.2 på ett 16x PCIe-kort. Arbetet genomfördes på MT-FoU och målet med examensarbetet var att undersöka om den tilltänkta konstruktionen kan förbättra prestandan hos AI-baserade system och fungera som ett billigare alternativ i verksamheten. Schemaritning och PCB-design utfördes i KiCad och information om gränssnitt och komponenter hämtades främst från tillverkares hemsidor och datablad. Kretsen består i huvudsak av fyra stycken M.2 E key kontaktdon, en 16port/16lane packetswitch och en 16x PCIe-anslutning. Switchen delar upp banorna från PCIe porten så att Edge TPU’erna kan anslutas parallellt i M.2 kontakterna. Edge TPU’erna använder pipelineparallellism för att fördela arbetsuppgifter på varje TPU så att större, mer komplexa program kan exekveras. Vid monteringen av kretskortet uppstod problem med fastlödningen av vissa komponenter. För att undvika att dessa problem uppstår och möjliggöra avlägsnandet av dessa felkällor bör montering istället beställas av fabrik där lödrobot finns tillgängligt. På grund av att tiden för kursen tog slut hann en sådan beställning inte göras och evaluering av den framtagna designen var därför inte möjlig att genomföra. Den design som togs fram var dock betydligt billigare än de existerande lösningarna och med pipelineparallellism förväntas designen kunna utföra komplexa beräkningar och därmed förbättra prestandan i befintliga system. / The rapidly growing development of AI-based applications and the large amount of data these applications process place increased demands on the performance and optimization of conventional computer systems. To satisfy these growing computing requirements, hardware accelerators are used to improve the data processing speed by offloading the existing equipment by executing models and complex calculations. The existing solutions currently used are costly and MT-R&D at Umeå University Hospital is therefore requesting an alternative solution by combining smaller integrating accelerators on a larger PCIe card. In this thesis, an AI accelerator using four Google Coral Dual Edge TPU M.2 on a 16x PCIe card is designed and implemented. The work was carried out at MT-R&D and the goal of the thesis was to investigate whether the intended design can improve the performance of AI-based systems and serve as a cheaper alternative in the institution. Schematic and PCB were designed in KiCad and information on interfaces and components was obtained from manufacturers' websites and data sheets. The circuit’s main components are four M.2 E key connectors, a 16port/16lane packet switch and a 16x PCIe connection. The switch divides the lanes from the PCIe port so that the Edge TPUs can be connected in parallel in the M.2 connectors. The Edge TPUs use pipeline parallelism to distribute models across each TPU so that larger, more complex programs can be executed. When assembling the circuit board, problems arose with the soldering of certain components. In order to avoid these sources of error, assembly should instead be ordered from a factory where a soldering robot is available. Due to the fact that the time for the course ran out, such an order could not be placed and evaluation of the design was therefore not possible to carry out. However, the design that was produced was significantly cheaper than the existing solutions and by using pipeline parallelism, the design is expected to be able to perform complex calculations and thus improve the performance of existing systems.

Purpose of Review: Fractional flow reserve (FFR) and instantaneous wave-free ratio (iFR) have emerged as the invasive diagnostic tools of choice for hemodynamic assessment of the severity of CAD (coronary artery disease). We sought to comprehensively review the evidence on the utility of hemodynamic assessment of the coronary stenoses after percutaneous coronary intervention (PCI) using FFR/iFR, mechanisms of positive post-PCI iFR/FFR, and the clinical impact of significant residual ischemia. Recent Findings: The evidence on the utility of the post-PCI hemodynamic assessment has accumulated over the last few years. The post hoc analysis from the FAME 1 and FAME 2 data shows that higher post-PCI FFR is associated with better symptomatic improvement and lower event rate with larger increase in delta FFR (∆ FFR: post-PCI FFR – pre-PCI FFR). Unlike pre-PCI FFR, a consensus has not been established on the optimal value of post-PCI FFR, though multiple studies point toward better clinical outcomes with higher post-PCI FFR and larger ∆ FFR. Summary: Visual assessment of adequate stent apposition by coronary angiography is insufficient in evaluating for residual ischemia. The hemodynamic evaluation of residual ischemia by post-PCI FFR/iFR yields clinically relevant data and allows for appropriate post PCI optimization.

[pt] O aumento do consumo de eletroeletrônicos e o desenvolvimento da tecnologia, proporcionou o surgimento de uma gama de produtos com diferentes funcionalidades, cada vez mais complexos e menores. As placas de circuito impresso (PCIs) são consideradas a parte principal dos dispositivos eletrônicos, sendo o cobre o seu componente elementar. O desenho e espessura das trilhas do circuito são determinantes para caracterizar a passagem de corrente elétrica nos equipamentos eletroeletrônicos e seu funcionamento está diretamente ligado a qualidade da confecção das trilhas do circuito. Este trabalho tem como objetivo estudar o processo de lixiviação do cobre, mediante a reação do ácido clorídrico, cloreto de cobre II e fluxo de ar. Além de investigar as condições experimentais ótimas do processo, que tem como principal característica a possibilidade de regeneração e reutilização da solução. Modelos para avaliar o efeito da concentração de ácido e fluxo de ar na corrosão das placas de circuito impresso foram desenvolvidos através da aplicação do planejamento experimental (pelo método clássico e por algoritmo genéticos em modelos polinomiais) e redes neurais artificiais. Visando encontrar as melhores condições experimentais para o sistema proposto, além de investigar a melhor técnica de predição do mesmo. Os resultados obtidos pelas previsões foram comparados com os resultados experimentais reais. As modelagens foram comparadas pela análise dos coeficientes de correlação (R2) e índices de erro (SSE, MSE e RMSE). Constatando-se que o modelo polinomial foi o mais adequado para prever a resposta. Através da investigação da superfície de resposta e curvas de contorno, foram identificadas as condições otimizadas para o processo. Das quais as concentrações ótimas de ácido clorídrico, cloreto de cobre II e fluxo de ar foram 1 mol.L-1, 0.3 mol.L-1 e 0.5 L/ min, respectivamente. / [en] The increased consumption of consumer electronics and the development of technology has led to the emergence of a range of products with different features, increasingly complex and smaller. Printed circuit boards (PCIs) are considered the main part of electronic devices, with copper being their elementary component. The design and thickness of the circuit tracks are crucial to characterize the passage of electric current in electronic equipment and its operation is directly linked to the quality of the circuit tracks. This work aims to study the copper leaching process through the reaction of hydrochloric acid, copper chloride II and airflow. In addition to investigating the optimal experimental conditions of the process, which has as its main feature the possibility of regeneration and reuse of the solution. Models to evaluate the effect of acid concentration and airflow on PCB corrosion were developed by applying experimental design (by the classical method and by the genetic algorithm in polynomial models) and artificial neural networks. Aiming to find the best experimental conditions for the proposed system, besides investigating the best prediction technique. The results obtained by the predictions were compared with the actual experimental results. The modeling was compared by analysis of correlation coefficients (R2) and error indices (SSE, MSE, and RMSE). Noting that the polynomial model was the most appropriate to predict the response. Through investigation of the response surface and contour curves, the optimized conditions for the process were identified. Of which the optimal concentrations of hydrochloric acid, copper chloride II and airflow were 1 mol.L-1, 0.3 mol.L-1 and 0.5 L / min, respectively.

International Telemetering Conference Proceedings / October 26-29, 1998 / Town & Country Resort Hotel and Convention Center, San Diego, California / The necessity to acquire and analyze data dates back to the beginning of science itself. Long ago, a scientist may have run experiments and noted the results on a piece of paper. These notes became the data. The method was crude, but effective. As experiments got more complex, the need for better methodologies arose. Scientists began using computers to gather, analyze, and store the data. This method worked well for most types of data acquisition. As the amount of data being collected increased, larger computers, faster processors, and faster storage devices were used in order to keep up with the demand. This method was more refined, but still did not meet the needs of the scientific community. Requirements began to change in the data acquisition arena. More people wanted access to the data in real time. Companies producing large data acquisition systems began to move toward a network-based solution. This architecture featured a specialized computer called the server, which contained all of the data acquisition hardware. The server handled requests from multiple clients and handled the data flow to the network, data displays, and the archive medium. While this solution worked well to satisfy most requirements, it fell short in meeting others. The ability to have multiple computers working together across a local or wide area network (LAN or WAN) was not addressed. In addition, this architecture inherently had a single point of failure. If the server machine went down, all data from all sources was lost. Today, we see that the requirements for data acquisition systems include features only dreamed of five years ago. These new systems are linked around the world by wide area networks. They may include code to command satellites or handle 250 Mbps download rates. They must produce data for dozens of users at once, be customizable by the end user, and they must run on personal computers (PCs)! Systems like these cannot work using the traditional client/server model of the past. The data acquisition industry demands systems with far more features than were traditionally available. These systems must provide more reliability and interoperability, and be available at a fraction of the cost. To this end, we must use commercial-off-the-shelf (COTS) computers that operate faster than the mainframe computers of only a decade ago. These computers must run software that is smart, reliable, scalable, and easy to use. All of these requirements can be met by a network of PCs running the Windows NT operating system.

<p>This thesis contains two parts. The first part concerns the research work on free molecules using synchrotron-radiation-related techniques. Auger electron spectra of two free open-shell molecules, O₂ and NO, were studied experimentally and theoretically. Photoionization experimental technique with tunable synchrotron radiation source was used to induce core-level electron ionization and obtain the <i>KVV</i> normal Auger electron spectra. A quantitative assignment of O₂ normal Auger spectrum was obtained by applying <i>ab initio</i> CI calculations and LVI Auger line shape simulations including the bond length dependence of Auger transition rates. The photon energy dependence of normal Auger electron spectra was focused on with photon energies in the vicinities of core-ionization threshold energies. Consequently, the MAPCI (Molecular Auger Post Collision Interaction) theory was developed. Taking the near-threshold O₂ normal Auger spectrum as an example, the two extreme cases of MAPCI effect, “atomic-like PCI” and “molecular PCI”, were discovered and discussed. The effect of shape resonance on near-threshold molecular normal Auger spectrum was discussed taking NO near threshold normal Auger spectra as example.</p><p>The second part deals with research work on the chemisorption of small epoxy organic molecules, ethylene oxide, methyl oxirane, on Si (100) surfaces. Synchrotron radiation related techniques, UPS, XPS and NEXAFS, were applied. Based on the valence photoemission spectra, C 1<i>s</i> and Si 2<i>p</i> XPS spectra, the epoxy ring opening reactions of these molecules in chemisorption process were proved. Further tentative search for the surface-adsorbate CDAD effect was performed, and no evident circular dichroism was confirmed.</p>

Ce mémoire de thèse concerne l’étude des composés ternaires Terre rare–magnésium–nickel utilisable pour le stockage de l’hydrogène. Ces composés ont été obtenus par fusion ou par mécanosynthèse. Les intermétalliques YNi4-xAlxMg, dérivant des phases de Laves de structure cubique ont été étudiés. Ils réagissent de manière réversible avec l’hydrogène à P et T ambiantes. Le comportement structural lors d’une hydruration a été étudié par DRX in situ. Le composé conserve sa symétrie cubique mais avec diminution de la cristallinité. Cette étude est complétée par l’étude de composés : (i) riche en terre rare (e.g. Gd4NiMg) qui absorbe l’hydrogène à température ambiante de manière irréversible. La structure de l’intermétallique et de l’hydrure sont déterminées. La décomposition de l’hydrure à température supérieure à 90°C est expliquée. (ii) riche en magnésium. Nous avons pu identifié un nouveau composé de formulation proche de Mg77Gd9Ni14.5 de structure CFC. / This work deals with the study of ternary compounds Rare Earth – magnesium - nickel used ofr hydrogen storage. All the compounds are prepared by fusion and by mechanical alloying method.. The compounds YNi4-xAlxMg, derived from the cubic Laves phases have been studied in the first part. It reacts reversibly towards hydrogen at atmospheric pressure and room temperature. The structural behaviour during the hydrogen sorption has been studied by in situ XRD. The compound remains cubic with a decrease of the crystallinity. This study is completed by the study of compounds : (i) rich in rare earth (e.g. Gd4NiMg) that absorb hydrogen at room temperature but irreversibly. Structures of both the intermetallic and the hydride have been determined. The decomposition of the hydride at temperature higher than 90°C is also explained. (ii) rich in magnesium. A new compounds with a formulation closed to Mg77Gd9Ni14.5 has been identified and it crystallized with a cubic faces centred structure.