The aim of this Master Thesis is to synthesize and study new macrocyclic ligands containing fluorine atoms for use as potencial contrast agents in 19 F magnetic resonance imaging. New ligands were designed as analogues of ligands used in contrast agents already utilized in clinical practice - macrocyclic ligand for complexation of trivalent lanthanide ions based on DOTA skeleton and series of macrocyclic ligands for complexation of divalent nickel ion based on cyclam skeleton. All designed ligands were synthesized a characterized. The complexes [LnIII (dotptfe )]− were tested in vitro and in vivo to study 19 F NMR relaxation times enhancements.

Les récepteurs métabotropes du glutamate (mGluR) sont des récepteurs couplés aux protéines G qui régulent la transmission synaptique. Ce sont des cibles de choix pour le traitement de maladies neurologiques et psychiatriques telles que la maladie de Parkinson et la schizophrénie.J'ai développé une stratégie d'étude de l'assemblage multimérique des protéines membranaires dans des cellules vivantes, à l'aide de techniques de marquage orthogonal et de FRET en temps-résolu. De façon inattendue, j'ai montré que certaines sous-unités de mGluR, en plus de former des récepteurs homodimériques, peuvent former des récepteurs hétérodimériques fonctionnels. D'autre part, j'ai appliqué ces techniques à l'étude du mécanisme d'activation des mGluR et de leur régulation allostérique. J'ai démontré qu'un mouvement relatif des domaines extracellulaires au sein du dimère était responsable de l'action du glutamate.Ce travail a permis de mieux comprendre le fonctionnement des mGluR, et permet la conception de nouveaux tests de criblage. / Metabotropic glutamate receptors (mGluRs) are G protein-coupled receptors that regulate synaptic transmission. They are relevant therapeutic targets for neurological and psychiatric disorders, such as Parkinson disease and schizophrenia.I developed a strategy to study the multimeric assembly of membrane proteins in living cells, through a combination of orthogonal labeling and time-resolved FRET. Unexpectedly, some subunits of mGluRs, in addition to forming homodimeric receptors, were found capable of forming functional heterodimeric receptors. Then, I applied these techniques to study the activation mechanism of mGluRs and their allosteric regulation. I demonstrated that a conformational change of the dimeric extracellular domain is responsible for the action of glutamate.In addition to increase our understandings of how mGluRs function, this work opens new avenues for the design of drug screening tests.

Molecular Nanomagnets have been proposed as plausible candidates in a variety of futuristic applications. Thorough understanding of the magnetic properties of these systems is therefore necessary to develop devices that include such units. The aim of this thesis is to synthesise and structurally and magnetically characterise a range of systems that could be used as elementary units in three proposed applications such as: data storage devices, magnetic refrigerants and qubits for quantum computing. A series of mixed 3d/4f metal complexes were synthesised through solvothermal reactions and characterised by X-ray single crystal diffraction and SQUID magnetometry. Through indirect methods it was possible to obtain high magnetic entropy change for some systems. It was also possible to obtain some insight into the magnetic interactions within the systems through modelling the magnetic data. The role of the 4f-4f and 3d-4f interactions in two sets of molecules is also described. The first study is in an asymmetric dysprosium dimer, where through a range of experimental techniques and advanced theoretical methods, such ab-initio calculations we are able to explain the role of the intramolecular interactions and their effect on the SMM properties of this system. Similarly, insight into the role of the 3d-4f interactions is achieved through the observation of the magnetic behaviour of a family of 27 tetranuclear systems, though SQUID data and ab-initio calculations. Finally, chemical functionalization of a well-proposed qubits, namely {Cr7Ni} and subsequent reaction with a redox active metal ion, CoII/III, two {Cr7Ni} systems are linked. The magnitude of the exchange interaction between the {Cr7Ni}-CoII-{Cr7Ni} was determined through Electron Paramagnetic Resonance. Furthermore, by chemical oxidation/reduction of the cobalt between paramagnetic and diamagneticstates, i.e. CoII and CoIII respectively, we demonstrate that the interaction can be switched ON/OFF. This characteristic makes of these systems candidates to function as a SWAP gate.

The major topics discussed are the phosphorescence sensitization in the lanthanides via energy transfer and in the organics by heavy atom effects. The f-f transitions in lanthanides are parity forbidden and have weak molar extinction coefficients. Upon complexation with the ligand, ttrpy (4'-p-Tolyl-[2,2':6',2"]-terpyridine) the absorption takes place through the ligand and the excitation is transferred to the lanthanides, which in turn emit. This process is known as "sensitized luminescence." Bright red emission from europium and bright green emission from terbium complexes were observed. There is ongoing work on the making of OLEDs with neutral complexes of lanthanide hexafluoroacetyl acetonate/ttrpy, studied in this dissertation. Attempts to observe analogous energy transfer from the inorganic donor complexes of Au(I) thiocyanates were unsuccessful due to poor overlap of the emissions of these systems with the absorptions of Eu(III) and Tb(III). Photophysics of silver-aromatic complexes deals with the enhancement of phosphorescence in the aromatics. The heavy atom effect of the silver is responsible for this enhancement in phosphorescence. Aromatics such as naphthalene, perylene, anthracene and pyrene were involved in this study. Stern Volmer plots were studied by performing the quenching studies. The quenchers employed were both heavy metals such as silver and thallium and lighter metal like potassium. Dynamic quenching as the predominant phenomenon was noticed.

Interest in the synthesis of novel luminescent lanthanide complexes stems from the wide area of application in various fields of science. Thus, the theoretical design becomes an important tool in the search for efficient luminescent complex. Within this context, emerged the Sparkle model that were developed for predicting the ground state geometries of lanthanide complexes. In this work we present a new parameterization of the ion Eu (III) for the AM1 and RM1 semiempirical methods, with the explicit inclusion of orbitals s, p and d in the basis set. With this reasoning, we have limitations in the calculation of lanthanide complexes in the solvent with MOPAC and initiate an improvement in the description of links that show the highest degree of covalency. The parameterization process of this model had a robust statistical analysis to select the set of parameterization, since our whole universe has 144 structures of lanthanide complexes. Thus, we used two methods to identify the structures that should be part of the parameter set: (i) the method of hierarchical grouping known as AGNES and (ii) the DIANA method. The models that we developed (AM1/Eu and RM1/Eu) showed an improvement compared to the versions of the Sparkle model, considering the low average absolute errors for the distances Eu L (L = O, N, C, S, P, F, Cl, Br and Eu), maintaining the commitment to the low computational cost involved in performing the calculations of the geometries of the ground state. We also investigated the accuracy of quantum-chemical methods in the reproduction of the coordination polyhedron of lanthanide ion complexes. We compared semiempirical methods (the PCC approach and Sparkle model) and also those with the ab initio methodology RHF/ECP/STO-3G. After this study, we conclude that the only semiempirical Sparkle model (hundreds of times faster) present similar accuracy to what can be obtained by present-day ab initio/ECP full geometry optimization calculations on such lanthanide complexes. In addiction it further indicates that the PCC approach is inappropriate to predict the coordination polyhedron geometries of lanthanide complexes. / O interesse na síntese de novos complexos de lantanídeos luminescentes advém da vasta área de aplicação em diversos campos da ciência. Deste modo, o design teórico torna-se uma importante ferramenta na busca de eficientes complexos luminescentes. Dentro desse contexto, surgiram as versões do modelo Sparkle que foram desenvolvidas para a predição da geometria do estado fundamental de complexos de lantanídeos. Nesta dissertação, buscaremos parametrizar o íon Eu(III) para os métodos semiempíricos AM1 e RM1, com a inclusão explícita de orbitais do tipo s, p e d no conjunto de base. Com essa fundamentação, acabamos com limitações no cálculo de complexos lantanídicos em meio solvente com o MOPAC e iniciamos uma melhoria na descrição de ligações com grau de covalência mais elevados. O processo de parametrização deste modelo contou com uma análise estatística robusta para a seleção do conjunto de parametrização, uma vez que o nosso conjunto universo apresenta 144 estruturas de complexos de európio. Desta forma, utilizamos dois métodos para identificar as estruturas que deveriam fazer parte do conjunto de parametrização: (i) o método hierárquico de agrupamento conhecido como AGNES e (ii) o método de desagrupamento DIANA. Os modelos que desenvolvemos (AM1/Eu e RM1/Eu) mostraram uma evolução em relação às versões do modelo Sparkle, tendo em vista os baixos erros médios absolutos para as distâncias Eu L, (L = O, N, C, S, P, F, Cl, Br e Eu), mantendo o compromisso com o baixo custo computacional envolvido na execução dos cálculos das geometrias do estado fundamental. Também investigamos a exatidão dos métodos químico-quânticos na reprodução do poliedro de coordenação de complexos de íons lantanídeos. Nós comparamos métodos semiempiricos (a abordagem PCC e o modelo Sparkle) e ainda esses com a metodologia ab initio RHF/ECP/STO-3G. Após este estudo, concluímos que a abordagem PCC é ineficiente para reproduzir o poliedro de coordenação cristalográfico desses complexos. Os resultados obtidos confirmaram também que os modelos Sparkle apresentam exatidão semelhante, à que pode ser obtida pela otimização completa da geometria, utilizando as atuais metodologias ab initio/ECP.

Orientador: Fernando Aparecido Sigoli / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Química / Made available in DSpace on 2018-08-26T22:25:31Z (GMT). No. of bitstreams: 1 Petiote_Lanousse_M.pdf: 4326996 bytes, checksum: a438622b7ca8e3a70bbbec5abf7dc735 (MD5) Previous issue date: 2015 / Resumo: Esse trabalho reporta o estudo das propriedades fotoluminescentes de uma série de complexos dinucleares de lantanídeos trivalentes contendo os ligantes 2 tenoiltrifluoro-acetilacetonato (tta), óxido de [2 (difenilfosforil)etil]difenilfosfina (dppeo), e óxido de trifenilfosfina (tppo). Os complexos de formula geral [{Ln(tta)3(tppo)}2(µ-dppeo)] foram obtidos a partir do controle da estequiometria e da ordem de adição dos reagentes em reações envolvendo os ligantes fosfinóxidos e os complexos precursores [Ln(tta)3(H2O)2]. A caracterização foi realizada por análise elementar, espectroscopia vibracional na região do infravermelho, espectroscopia eletrônica na região do visível, ressonância magnética nuclear de 1H e de 31P e análise termogravimétrica. Para os complexos de európio(III), são apresentados cálculos semi-empíricos realizados baseando-se nas geometrias de coordenação do estado fundamental desses complexos usando o modelo SPARKLE/AM1. Os espectros de fotoluminescência exibem as bandas de emissão características dos íons lantanídeos utilizados confirmando a atuação do efeito antena nos mesmos. Os espectros de emissão dos complexos do íon Eu(III) apresentam elevados valores da razão assimétrica I(5D0?7F2)/ I(5D0?7F1) e de ?2 evidenciando a presença dos íons Eu(III) em ambientes químicos de baixa simetria e a atuação do acoplamento dinâmico. Os complexos de európio(III) mostram um aumento do tempo de vida de emissão e do rendimento quântico com a substituição das moléculas de água devido a diminuição das taxas de decaimentos não radiativos do nível emissor 5D0. A energia do nível tripleto do complexo [{Eu(tta)3(tppo)}2(?-dppeo)] determinada experimentalmente e calculada apresentam boa concordância. Os cálculos das taxas de transferência e retro-transferência de energia mostram que, no caso da transferência de energia dos ligantes para o íon Eu(III), as rotas mais eficientes são T?5D1 e T?5D0. As coordenadas de cromaticidade da Commission Internacionale d¿Éclairage (CIE) indicam que a luz vermelha emitida pelos complexos contendo íons Eu(III) apresenta alta pureza. Os complexos com emissões na região do infravermelho próximo apresentam tempos de vida de emissão 10 vezes menor do que os complexos de európio devido baixa eficiência do efeito antena e a desativação não radiativa dos níveis emissores dos íons Er(III) e Yb(III). Palavras-chave: Lantanídeos. Complexos dinucleares. Luminescência. Efeito Antena / Abstract: This work reports on photoluminescent properties of a series of lanthanide dinuclear complexes containing 2- thenoyltrifluoroacetylacetonate, [2 (diphenylphosphoryl)ethyl](diphenyl)phosphine oxide and triphenyl oxide as ligands. The complexes have the general formula [{Ln(tta)3(tppo)}2(µ-dppeo)] and were obtained by controlling the stereochemistry and reagents addition in reactions involving the phosphine oxide ligands and precursor complexes [Ln(tta)3(H2O)2] The characterization was carried out using elemental analysis, IR vibrational spectroscopy, UV-Vis electronic spectroscopy, 1H and 31P NMR and thermogravimetric analysis techniques. Semi-empirical calculations have been carried out using the fundamental geometries of europium(III) complexes optimized by the SPARKLE/AM1 model. The photoluminescence spectra under excitation in the ligand show the narrow bands characteristics of lanthanide ions used, confirming that the antenna effect is operating in all complexes. The emission spectra of Eu(III) complexes show high values of either asymmetric ratio I(5D0?7F2)/ I(5D0?7F1) and the intensity parameter ?2 indicating low-symmetry chemical environments occupied by Eu(III) ions and contribution of the dynamic coupling mechanism. The Eu(III) complexes show an increasing of the luminescence lifetime and the quantum yield as the water molecules have been substituted. This was attributed to a reduction of the non-radiative decay of the 5D0 level in complexes without water molecules in the coordination sphere of lanthanide trivalent ion. The experimental value of the triplet energy level for the complex [{Eu(tta)3(tppo)}2(?-dppeo)] is consistent with the calculated one. Calculations of the energy transfer rates in the europium complexes show that the most efficient routes are T?5D1 e T?5D0. The chromaticity coordinates indicate high purity for the red light emitted by the europium complexes. The complexes with emission in the near infrared region show emission life 10 times lower than the europium complex. This was attributed to different factors such as low relative efficiency of the antenna effect and the efficient non-radiative deactivation of emitting levels in Er (III) and Yb (III) mainly by C-H oscillators in the ligands. Keywords: Lanthanides, Dinuclear complexes, Luminescence, Antenna effect / Mestrado / Quimica Inorganica / Mestre em Química

Dreiwertige Actinide (An(III)) und Lanthanide (Ln(III)) stellen im Falle ihrer Inkorporation eine ernste Gefahr für die Gesundheit des Menschen dar. An(III) sind künstlich erzeugte, stark radioaktive Elemente, die insbesondere bei der nuklearen Energiegewinnung in Kernkraftwerken entstehen. Durch Störfälle oder nicht fachgerechte Lagerung radioaktiven Abfalls können sie in die Umwelt und die Nahrungskette des Menschen gelangen. Ln(III) sind hingegen nicht radioaktive Elemente, die natürlicherweise vorkommen und für vielfältige Anwendungen in Technik und Medizin abgebaut werden. Folglich kann der Mensch sowohl mit An(III) als auch Ln(III) in Kontakt kommen bzw. sie inkorporieren. Es ist daher von enormer Wichtigkeit, das Verhalten dieser Elemente im menschlichen Körper aufzuklären. Während makroskopische Vorgänge wie Verteilung, Anreicherung und Ausscheidung bereits sehr gut untersucht sind, ist das Wissen hinsichtlich der chemischen Bindungsform (Speziation) von An(III) und Ln(III) in Körperflüssigkeiten noch sehr lückenhaft. In der vorliegenden Arbeit wurde daher erstmals die chemische Bindungsform von Cm(III) und Eu(III) in natürlichem menschlichem Urin (in vitro) spektroskopisch aufgeklärt und die gebildeten Komplexe identifiziert. Hierzu wurden auch grundlegende Untersuchungen zur Komplexierung von Cm(III) und Eu(III) in synthetischem Modellurin sowie mit den urinrelevanten organischen Modellliganden Harnstoff, Alanin, Phenylalanin, Threonin und Citrat durchgeführt und die noch unbekannten Komplexbildungskonstanten bestimmt. Abschließend wurden alle experimentellen Ergebnisse mit Literaturdaten und Vorherberechnungen mittels thermodynamischer Modellierung verglichen. Auf Grund der hervorragenden Lumineszenzeigenschaften von Cm(III) und Eu(III) konnte insbesondere auch die Eignung der zeitaufgelösten laserinduzierten Fluoreszenzspektroskopie (TRLFS) als Methode zur Untersuchung dieser Metallionen in unbehandelten, komplexen biologischen Flüssigkeiten demonstriert werden. Die Ergebnisse dieser Arbeit liefern damit neue Erkenntnisse zu den biochemischen Reaktionen von An(III) und Ln(III) in Körperflüssigkeiten auf molekularer Ebene und tragen zu einem besseren Verständnis der bekannten, makroskopischen Effekte dieser Elemente bei. Darüber hinaus sind sie die Grundlage weiterführender in-vivo-Untersuchungen.

In dieser Arbeit ist das Wachstumsverhalten von unter 10nm großen, aufwärtskonvertierenden NaYF4-Nanokristallen der hexagonalen und kubischen Phase untersucht worden. Die Ostwald-Reifung solcher Partikel in Ölsäure/1-Octadecen führt zu einer Verbreiterung der Partikelgrößenverteilung, falls das Kolloid nur aus Partikeln einer Kristallphase besteht. Eine schmale Teilchengrößenverteilung tritt nur dann auf, wenn die Partikel der b-Phase in Anwesenheit eines Überschusses von Partikeln der a-Phase wachsen. Solche binären Gemische aus Partikeln der a- und b-Phase entstehen, sobald Kolloide aus Partikeln der a-Phase für eine genügend lange Zeit erhitzt werden, denn bei hohen Temperaturen nukleieren Keime der b-Phase nach einer gewissen Zeit. Weil durch die Anzahl der Keime die finale Partikelgröße des Produkts der b-Phase bestimmt wird, ist eine Kontrolle der Nukleation wichtig um die finale Partikelgröße steuern zu können. Es wird gezeigt, dass die Anzahl der Keime der b-Phase stark abhängig von der Zusammensetzung der Partikel der a-Phase ist, die als Ausgangsmaterial verwendet werden. Die a-Phase ist dafür bekannt Na1−xYF4−x-Mischkristalle mit Zusammensetzungen von x=0 bis x=4/9 zu bilden. Bei natriumarmen Partikeln der a-Phase entstehen eine geringe Anzahl an Keimen der b-Phase, wobei Partikel der a-Phase mit einem hohen Gehalt an Natrium eine große Anzahl an Keimen bilden. Durch die Ausnutzung dieser Abhängigkeit und der dementsprechenden Modifizierung der Synthese der Partikel der a-Phase, können in Ölsäure/1-Octadecen kleine, phasenreine b-NaYF4:Yb,Er-Partikel mit Größen von unter 6 nm, sowohl als auch deutlich größere Partikel erhalten werden. Weiterhin wird gezeigt, dass Mischungen von a- und b-Partikeln auch sehr gut für die Synthese von Kern/Schale-Partikeln geringer Größe geeignet sind. Durch die Reaktion von Natriumoleat, Seltenerdoleat und Ammoniumfluorid werden zunächst etwa 5nm große b-NaYF4:Yb,Er-Kernpartikel gebildet, wobei ein hohes Verhältnis von Natrium- zu Seltenerdionen die Nukleation einer großen Anzahl von Keimen der b-Phase begünstigt. Anschließend wird eine etwa 2nm dicke Schale aus b-NaYF4 aufgewachsen, wobei 3–4nm große a-NaYF4-Partikel als Vorläuferpartikel für das Schalenmaterial dienen. Im Gegensatz zu den Kernpartikeln werden diese Partikel der a-Phase jedoch mit einem geringen Verhältnis von Natrium- zu Seltenerdionen hergestellt, welches die unerwünschte Nukleation von b-NaYF4-Partikeln während des Schalenwachstums wirksam unterdrückt. Diese neue Methode zur Herstellung sehr kleiner aufwärtskonvertierender Kern/Schale-Nanokristalle kommt ohne zusätzliche Kodotierung der Kernpartikel aus und liefert Partikel im Gramm-Maßstab.

Ce travail est axé vers la compréhension détaillée des propriétés de luminescence de composés de certains métaux lourds. La première partie de ce mémoire décrit la caractérisation spectroscopique d'un radical de type nitronyle nitroxyde, 2-(2-pyridinyl)-4,4,5,5-tétraméthyl-4,5-dihydro-1H-imidazolyl-1-oxyl-3-oxyde, abrégé (NIT2-Py), et de ses complexes avec les cations Tb(III), [Tb(hfac)3NIT2-Py], et Y(III), [Y(hfac)3NIT2-Py]. La variation de la température affecte les spectres de luminescence qui montrent de la structure vibronique résolue. Les maxima de ces transitions vibroniques se rapprochent au fur et à mesure que la température augmente. Ces variations des maxima en fonction de la température ne correspondent pas à des variations de fréquences vibrationnelles et sont de l'ordre de 200 cm-1 entre 80 K et 240 K. La variation de la température n'a pas d'influence significative sur la structure moléculaire, comme atteste la variation mineure des maxima des spectres Raman entre 80 K et 300 K. La comparaison des spectres expérimentaux à des spectres calculés montre que ces variations peuvent être reproduites par l'utilisation d'une combinaison de fréquences vibrationnelles. Le paramètre dont la variation est très significative est la résolution du spectre de luminescence, représentée par la largeur à mi-hauteur des transitions vibroniques qui forment le spectre de luminescence. La deuxième partie de ce mémoire décrit les propriétés de luminescence d'une série de complexes d’or(I). Elles sont comparées aux changements structuraux à pression et température variable. Les interactions aurophiles ont une grande influence sur la luminescence. La variation de la température et de la pression est une approche efficace pour varier la luminescence. Les effets observés dans les spectres d'émission de ces complexes dépendent des changements de structure induits par variation de la température et de la pression. Ces petites variations structurales mènent à des changements importants, à titre d'exemple à un déplacement du maximum de la bande de luminescence de 60 cm-1/ kbar vers les faibles énergies pour un des complexes de l'or(I) étudiés au cours de ce projet. / The overall goal of this work is to contribute to the understanding of luminescence properties of transition metal compounds. The first part of this thesis describes the spectroscopic characterization of a pyridine-substituted nitronyl nitroxyde radical 2-(2-pyridyl)-4,4,5,5-tetramethyl-4,5-dihydro-1H-imidazolyl-3-oxide-1-oxyl abbreviated (NIT2-Py) and its complexes with Tb(III) and Y(III), [Tb(hfac)3NIT2-Py] and [Y(hfac)3NIT2-Py], respectively. Their luminescence spectrum show resolved vibronic structure and variations with temperature The energy difference between vibronic luminescence maxima decreases with increasing temperature. This decrease is on the order of 200 cm-1 between 80 K and 240 K and is not due to decreasing Raman frequencies. The variations of temperature do not have a significant influence on the molecular structures, as illustrated by the very small variation of Raman maxima between 80 K and 300 K. Luminescence spectra were calculated using the time dependent theory of spectroscopy. Calculated luminescence spectra show that the variations with temperature are reproduced by using specific combinations of experimental frequencies and by adjusting the width of the vibronic transitions, determining the resolution of the luminescence spectrum. The second part of the thesis describes the luminescence spectroscopic properties of a series of trimetallic gold(I) complexes at variable temperature and pressure. Aurophilic interactions are very important for these compounds and influence the luminescence spectra. The spectroscopic features are compared to structural changes at variable temperature and pressure. Even small changes in structure lead to significant changes in luminescence, for example a shift of the maximum of the luminescence band by 60 cm-1/ kbar to lower energy for one of the gold(I) complexes studied.

Ce travail est axé vers la compréhension détaillée des propriétés de luminescence de composés de certains métaux lourds. La première partie de ce mémoire décrit la caractérisation spectroscopique d'un radical de type nitronyle nitroxyde, 2-(2-pyridinyl)-4,4,5,5-tétraméthyl-4,5-dihydro-1H-imidazolyl-1-oxyl-3-oxyde, abrégé (NIT2-Py), et de ses complexes avec les cations Tb(III), [Tb(hfac)3NIT2-Py], et Y(III), [Y(hfac)3NIT2-Py]. La variation de la température affecte les spectres de luminescence qui montrent de la structure vibronique résolue. Les maxima de ces transitions vibroniques se rapprochent au fur et à mesure que la température augmente. Ces variations des maxima en fonction de la température ne correspondent pas à des variations de fréquences vibrationnelles et sont de l'ordre de 200 cm-1 entre 80 K et 240 K. La variation de la température n'a pas d'influence significative sur la structure moléculaire, comme atteste la variation mineure des maxima des spectres Raman entre 80 K et 300 K. La comparaison des spectres expérimentaux à des spectres calculés montre que ces variations peuvent être reproduites par l'utilisation d'une combinaison de fréquences vibrationnelles. Le paramètre dont la variation est très significative est la résolution du spectre de luminescence, représentée par la largeur à mi-hauteur des transitions vibroniques qui forment le spectre de luminescence. La deuxième partie de ce mémoire décrit les propriétés de luminescence d'une série de complexes d’or(I). Elles sont comparées aux changements structuraux à pression et température variable. Les interactions aurophiles ont une grande influence sur la luminescence. La variation de la température et de la pression est une approche efficace pour varier la luminescence. Les effets observés dans les spectres d'émission de ces complexes dépendent des changements de structure induits par variation de la température et de la pression. Ces petites variations structurales mènent à des changements importants, à titre d'exemple à un déplacement du maximum de la bande de luminescence de 60 cm-1/ kbar vers les faibles énergies pour un des complexes de l'or(I) étudiés au cours de ce projet. / The overall goal of this work is to contribute to the understanding of luminescence properties of transition metal compounds. The first part of this thesis describes the spectroscopic characterization of a pyridine-substituted nitronyl nitroxyde radical 2-(2-pyridyl)-4,4,5,5-tetramethyl-4,5-dihydro-1H-imidazolyl-3-oxide-1-oxyl abbreviated (NIT2-Py) and its complexes with Tb(III) and Y(III), [Tb(hfac)3NIT2-Py] and [Y(hfac)3NIT2-Py], respectively. Their luminescence spectrum show resolved vibronic structure and variations with temperature The energy difference between vibronic luminescence maxima decreases with increasing temperature. This decrease is on the order of 200 cm-1 between 80 K and 240 K and is not due to decreasing Raman frequencies. The variations of temperature do not have a significant influence on the molecular structures, as illustrated by the very small variation of Raman maxima between 80 K and 300 K. Luminescence spectra were calculated using the time dependent theory of spectroscopy. Calculated luminescence spectra show that the variations with temperature are reproduced by using specific combinations of experimental frequencies and by adjusting the width of the vibronic transitions, determining the resolution of the luminescence spectrum. The second part of the thesis describes the luminescence spectroscopic properties of a series of trimetallic gold(I) complexes at variable temperature and pressure. Aurophilic interactions are very important for these compounds and influence the luminescence spectra. The spectroscopic features are compared to structural changes at variable temperature and pressure. Even small changes in structure lead to significant changes in luminescence, for example a shift of the maximum of the luminescence band by 60 cm-1/ kbar to lower energy for one of the gold(I) complexes studied.