Abstract The defect structure of CuMX2 (MºIn, Ga and XºS, Se, Te) was investigated. The defect concentrations were derived as a function of nonmolecularity( X) and nonstoichiometry( Y), and the carrier concentrations were calculated quantitatively using a theoretical defect model, which could be used to select the proper region of stoichiometry for device designs with specified carrier concentrations. The compensation reaction between dopants and intrinsic defects in CuMX2, compound semiconductors is quantitatively illustrated. If for a small increment of temperature the Fermi level EF is shifted by £GEF and the concentration of free majority carriers is increased by £Gn, then the ratio£Gn /£GEFis a measure of the defect-level concentration within£GEF. The electrical and optical properties of CuMX2 have been investigated using various types of electrically active intrinsic defects caused by deviations from the ideal stoichiometry of the compound. The defect complexes are formed by reactions among the impurity and native defects and/or among the native defects during crystal growth or during processing, and play a significant role in determining the characteristics and performance of devices by affecting lifetimes, degradation and breakdown etc.

Neste trabalho foi realizado um estudo sistemático, das interações hiperfinas (campo hiperfino magnético e campo hiperfino elétrico), nos compostos do tipo RMn2Si2 e RMn2Ge2 (onde R = La, Nd, Pr), sendo que, para a amostra de La foi estudada a série LaMn2(Si(1-x)Gex)2 (onde x = 0, 0.2, 0.4, 0.6, 0.8, 1), por meio da técnica de correlação angular gama-gama perturbada (CAP), utilizando os núcleos de prova 111In(111Cd) e 140La(140Ce). Além disso, foram realizadas medições para a caracterização estrutural utilizando a técnica de difração de raios X, para a caracterização magnética, a nível macroscópico, através de medições de magnetização, caracterização nuclear utilizando a técnica de espectrometria da radiação gama e um estudo complementar por cálculos de primeiros princípios somente para as amostras de LaMn2Ge2 e LaMn2Si2. A maior contribuição para o comportamento magnético nesses compostos tem origem no ordenamento magnético da sub rede do Mn, que aparece em temperaturas relativamente altas, da ordem de 480 K, sendo que, esses compostos possuem um ordenamento antiferromagnético e próximo a temperatura ambiente passam a possuir um ordenamento ferromagnético. Além disso, para o caso especifico dos compostos PrMn2Ge2 e NdMn2Ge2, abaixo de 40K é observado, também, o ordenamento ferromagnético dos spins da sub rede da terra rara. O núcleo de prova 111In(111Cd) foi utilizado para medir, tanto as interações de quadrupolo elétrico, como de dipolo magnético na rede do Mn (interação matrizmatriz). Esses resultados evidenciaram a transição de fase magnética (antiferromagnética para ferromagnética), sendo que, o campo hiperfino magnético, para as duas fases magnéticas, segue o comportamento da função de Brillouin. Já o núcleo de prova 140La(140Ce) permitiu medir a interação de dipolo magnético, tanto originada pela rede do Mn, como pela rede da terra rara. Mas, nesse caso, pode ser verificado somente o campo hiperfino magnético da fase ferromagnética da rede do Mn. Para a análise dos resultados PAC levou-se em conta uma interação matriz-impureza, pois o íon Ce3+ possui um elétron 4f desemparelhado, que pode contribuir para o campo hiperfino total. O composto LaMn2Si2 apresentou o comportamento do campo hiperfino magnético com a temperatura que segue o comportamento da função de Brillouin. Já os compostos LaMn2Ge2, PrMn2Ge2 e NdMn2Ge2 apresentaram um comportamento anômalo ao da função de Brillouin. Esse comportamento pode ser associado a uma forte hibridização da banda 4f do Ce com a banda 3d do Mn, tal conclusão foi reforçada pelos resultados do DOS (densidade de estados), resultados dos cálculos de primeiros princípios, dos compostos LaMn2Si2 e LaMn2Ge2. / This thesis reports a systematic study of the hyperfine interactions (magnetic hyperfine field and electric field gradient) in the RMn2Si2 and RMn2Ge2 ( R = La, Nd, Pr) compounds as well as in LaMn2(Si(1-x)Gex)2 (where x = 0, 0.2, 0.4, 0.6, 0.8, 1) mixed compounds. The hyperfine interactions were measured through Perturbed Angular Correlation Gamma-Gamma Spectroscopy technique using 111In(111Cd) and 140La(140Ce) as probe nuclei. Structural, magnetic and chemical analysis characterization were carried out by X-ray diffraction, magnetization measurements and gamma radiation spectrometry, respectively. An additional study by first-principles calculations for LaMn2Ge2 e LaMn2Si2 samples were also performed. The magnetic behavior in these compounds is mainly associated with the ordering of the magnetic moments of Mn sublattice, which appears at high temperature around 480 K. All studied compounds undergo a transition from antiferromagnetic to ferromagnetic ordering around room temperature. Specifically for PrMn2Ge2 and NdMn2Ge2 compounds, an additional ferromagnetic ordering due to the polarization of Pr or Nd magnetic moments is also observed below 40 K. 111In(111Cd) probe nuclei were used to measure the temperature dependence of the electric quadrupole and the dipole magnetic interactions at the Mn sublattice (matrix-matrix interaction). The results show the magnetic transition (antiferromagnetic to ferromagnetic phase), where the magnetic hyperfine field associated to both magnetic phases follows the Brillouin function. For the 140La(140Ce) probe nuclei, only dipole magnetic interaction measurements in the ferromagnetic phase were possible, also caused by the Mn sublattice and, for R = Pr, Nd, by the rare earth sublattice. The analysis of results considers a matrix-impurity interaction because the Ce3+ has an unpaired 4f electron, which can contribute to the total hyperfine field. The compound LaMn2Si2 presents a behavior for the temperature dependence of the magnetic hyperfine field that could be fitted by a Brillouin function. On the other hand, LaMn2Ge2, PrMn2Ge2 and NdMn2Ge2 compounds showed an anomalous Brillouin function behavior. This behavior can be associated with a strong hybridization Ce 4f band with the Mn 3d band. Such conclusion was reinforced by the DOS (density of stats), first-principles calculations, results to LaMn2Si2 and LaMn2Ge2.

Neste trabalho foi realizado um estudo sistemático, das interações hiperfinas (campo hiperfino magnético e campo hiperfino elétrico), nos compostos do tipo RMn2Si2 e RMn2Ge2 (onde R = La, Nd, Pr), sendo que, para a amostra de La foi estudada a série LaMn2(Si(1-x)Gex)2 (onde x = 0, 0.2, 0.4, 0.6, 0.8, 1), por meio da técnica de correlação angular gama-gama perturbada (CAP), utilizando os núcleos de prova 111In(111Cd) e 140La(140Ce). Além disso, foram realizadas medições para a caracterização estrutural utilizando a técnica de difração de raios X, para a caracterização magnética, a nível macroscópico, através de medições de magnetização, caracterização nuclear utilizando a técnica de espectrometria da radiação gama e um estudo complementar por cálculos de primeiros princípios somente para as amostras de LaMn2Ge2 e LaMn2Si2. A maior contribuição para o comportamento magnético nesses compostos tem origem no ordenamento magnético da sub rede do Mn, que aparece em temperaturas relativamente altas, da ordem de 480 K, sendo que, esses compostos possuem um ordenamento antiferromagnético e próximo a temperatura ambiente passam a possuir um ordenamento ferromagnético. Além disso, para o caso especifico dos compostos PrMn2Ge2 e NdMn2Ge2, abaixo de 40K é observado, também, o ordenamento ferromagnético dos spins da sub rede da terra rara. O núcleo de prova 111In(111Cd) foi utilizado para medir, tanto as interações de quadrupolo elétrico, como de dipolo magnético na rede do Mn (interação matrizmatriz). Esses resultados evidenciaram a transição de fase magnética (antiferromagnética para ferromagnética), sendo que, o campo hiperfino magnético, para as duas fases magnéticas, segue o comportamento da função de Brillouin. Já o núcleo de prova 140La(140Ce) permitiu medir a interação de dipolo magnético, tanto originada pela rede do Mn, como pela rede da terra rara. Mas, nesse caso, pode ser verificado somente o campo hiperfino magnético da fase ferromagnética da rede do Mn. Para a análise dos resultados PAC levou-se em conta uma interação matriz-impureza, pois o íon Ce3+ possui um elétron 4f desemparelhado, que pode contribuir para o campo hiperfino total. O composto LaMn2Si2 apresentou o comportamento do campo hiperfino magnético com a temperatura que segue o comportamento da função de Brillouin. Já os compostos LaMn2Ge2, PrMn2Ge2 e NdMn2Ge2 apresentaram um comportamento anômalo ao da função de Brillouin. Esse comportamento pode ser associado a uma forte hibridização da banda 4f do Ce com a banda 3d do Mn, tal conclusão foi reforçada pelos resultados do DOS (densidade de estados), resultados dos cálculos de primeiros princípios, dos compostos LaMn2Si2 e LaMn2Ge2. / This thesis reports a systematic study of the hyperfine interactions (magnetic hyperfine field and electric field gradient) in the RMn2Si2 and RMn2Ge2 ( R = La, Nd, Pr) compounds as well as in LaMn2(Si(1-x)Gex)2 (where x = 0, 0.2, 0.4, 0.6, 0.8, 1) mixed compounds. The hyperfine interactions were measured through Perturbed Angular Correlation Gamma-Gamma Spectroscopy technique using 111In(111Cd) and 140La(140Ce) as probe nuclei. Structural, magnetic and chemical analysis characterization were carried out by X-ray diffraction, magnetization measurements and gamma radiation spectrometry, respectively. An additional study by first-principles calculations for LaMn2Ge2 e LaMn2Si2 samples were also performed. The magnetic behavior in these compounds is mainly associated with the ordering of the magnetic moments of Mn sublattice, which appears at high temperature around 480 K. All studied compounds undergo a transition from antiferromagnetic to ferromagnetic ordering around room temperature. Specifically for PrMn2Ge2 and NdMn2Ge2 compounds, an additional ferromagnetic ordering due to the polarization of Pr or Nd magnetic moments is also observed below 40 K. 111In(111Cd) probe nuclei were used to measure the temperature dependence of the electric quadrupole and the dipole magnetic interactions at the Mn sublattice (matrix-matrix interaction). The results show the magnetic transition (antiferromagnetic to ferromagnetic phase), where the magnetic hyperfine field associated to both magnetic phases follows the Brillouin function. For the 140La(140Ce) probe nuclei, only dipole magnetic interaction measurements in the ferromagnetic phase were possible, also caused by the Mn sublattice and, for R = Pr, Nd, by the rare earth sublattice. The analysis of results considers a matrix-impurity interaction because the Ce3+ has an unpaired 4f electron, which can contribute to the total hyperfine field. The compound LaMn2Si2 presents a behavior for the temperature dependence of the magnetic hyperfine field that could be fitted by a Brillouin function. On the other hand, LaMn2Ge2, PrMn2Ge2 and NdMn2Ge2 compounds showed an anomalous Brillouin function behavior. This behavior can be associated with a strong hybridization Ce 4f band with the Mn 3d band. Such conclusion was reinforced by the DOS (density of stats), first-principles calculations, results to LaMn2Si2 and LaMn2Ge2.

The results of photoluminescence (PL) spectra of TlGaS2 single crystal were reported in the 500-1400 nm wavelength and in the 15-115 K temperature range. Three broad PL bands with an asymmetric Gaussian lineshapes were observed to be centered at 568 nm (A-band), 718 nm (B-band) and 1102 nm (C-band). The shift of the emission band peak energy as well as the change of the half-width of the emission band with temperature and excitation laser intensity were also studied. We analyzed the observed results using the configurational coordinate (CC) model. The powder diffraction patterns of TlInS2 and TlGaS2 crystals were obtained and the diffraction data were indexed using CRYSFIRE computer program packet. TlInS2 has hexagonal system with parameters a = 3.83 and c = 14.88 Ao. TlGaS2 has monoclinic system with parameters a = 9.62, b = 4.01 and c = 7.52 Ao with &amp / #946 / = 96.30o. Our diffraction studies at low temperatures did not reveal any phase transition for TlInS2 as reported in the literature. The specific heat capacities of both TlInS2 and TlGaS2 crystals calculated from Differential Scanning Calorimetry (DSC) measurements at low temperatures are reported in the thesis.

CdTe an II-VI semiconductor has been a leading thin film photovoltaic material due to its near ideal bandgap and high absorption coefficient [1]. The typical thin film CdTe solar cells have been of the superstrate configuration with CdS (Eg-2.42eV) as the n-type heterojunction partner. Due to the relatively narrow bandgap of CdS, a wider bandgap n-type window layer has recently emerged as a promising substitute: alloys of MgyZn1-yO have been successfully used as the emitter or window layer. The benefits in the usage of MgyZn1-yO (MZO) are its tunable bandgap and wide optical spectrum on optoelectronic devices. Due to an increasing bandgap of the window layer, the carrier collection can be improved in the short wavelength range (<500 nm). In addition alloys of CdSexTe1-x (CST) have also been used in the absorber layer (i.e., CST/CdTe) for the fabrication of CdTe devices to improve the carrier collection and lifetime [2]. The lower bandgap of the CST alloy can lead to higher short-circuit current (JSC), but it can also result in lower open circuit voltage (VOC). Another critical aspect of the CdTe solar cell is the use of copper as a p-type dopant, which is typically incorporated in the cell during the fabrication of the back contact. The most challenging issue related to further advancing the CdTe solar cell efficiency is the relatively low level of p-type doping, which limits the VOC. Efforts to dope CdTe with group V dopants are yet to produce the desired results. ZnO has been used as an effective high resistivity transparent. When CdTe is deposited directly on sputtered ZnO, VOC of typically 500-600 mV is produced. Band alignment measurements indicate that a negative conduction band offset with CdS exists; alloying with MgO to produce MgyZn1-yO with a composition of y = 0.15 can produce a flat conduction band alignment with CdS. This material has an additional benefit for improving the energy bandgap of the MZO for better UV light transmission in the short wavelengths. By changing the magnesium content from y = 0 to 0.30 allowed researchers to make the tunable conduction band offset from a “cliff” to a “spike,” with both increased open-circuit voltage and fill factor as increasing magnesium compositions [3] — the bandgap gains as expected with increased magnesium composition. The large compositions (y > 0.30) of MgyZn1-yO cause the enormous spike result in S-kink in the IV measurement so that the FF decreases. Besides, due to the instability of MZO material, the fabrication process has to proceed carefully. The properties of CST films and cells were investigated as a function of Se composition (x), substrate temperature (TSUB), and ambient used during the CSS deposition. The higher ratio of Se in CST alloy causes the smaller grain structures and lower bandgap, which profoundly detrimental to the device performance (VOC). However, the CST can be deposited in various substrate temperatures and different inert ambient gas to improve the grain structure by utilizing the especial Close Space Sublimation (CSS) deposition system. Therefore, despite the fact that the CST (25% Se) has the optical bandgap (1.37eV), the improvement of grain structure can slightly increase the doping concentration and decrease the grain boundary (GBs) due to increased alloys grain size 3X larger, which is contributed to improving the VOC [4]. The study of higher ratio Se of CST alloy is significant to achieve the high efficiency polycrystalline CST/CdTe photovoltaic devices. The effect of Cu doping back contact in CdSexTe1-x (CST)/CdTe solar cells with varying amounts of Se (x) has been investigated. The Cu-based back contact was annealed at different thermal temperatures in order to vary the amount of Cu in-diffusion. Net p-type doping was found to increase as the back-contact annealing temperature increased. All cells exhibited a decrease in VOC with increased annealing temperature (i.e., higher Cu concertation), presumably due to a degradation of the lifetime with increased amounts of Cu [5]. However, cells with the highest Se composition appeared to exhibit a higher degree of tolerance to the amount of Cu – i.e., they exhibited a smaller loss in VOC with the increased amount of Cu. Extrinsic p-type doping of CdSeTe can be fabricated using two different experimental processes. Firstly, by using group I elements such as, Cu to substitute Cd, which is promising during the back contact process. Secondly, using group V (P, As, Sb) elements to substitute Te, and this is suitable for Cd-rich of intrinsic CdTe. Intrinsic CST alloy has lower hole density concentration as higher Se composition with limitation of the VOC. Thus, in order to increase the p-type net doping up to 1016 cm-3 the extrinsic P or As doping have been widely investigated recently. The research studies show the CST/CdTe devices lead to improve VOC up to 850 mV with higher hole density in higher Se compositions of As doped CST alloys. Nevertheless, the group V doped CdTe still cause the formation of compensating defects limits the upper boundary of dupability on the CdTe thin film solar cells. Even if a high hole density concentration is achieved for intrinsically-doped p-type CST/CdTe, it is believed the poor carrier lifetime in the CdTe side would still limit the VOC.

Η ανάπτυξη νέων τεχνολογιών ανανεώσιμων πηγών ενέργειας τα τελευταία χρόνια, έχει γίνει αναγκαία με τη συνεχόμενη περιβαλλοντική ρύπανση και την αλόγιστη σπατάλη των φυσικών πόρων. Οι τεχνολογίες υδρογόνου και ειδικά η παραγωγή υδρογόνου μέσω της διαδικασίας της ηλεκτρόλυσης έχει αρχίσει να κερδίζει σημαντικό έδαφος τα τελευταία χρόνια με τη συνεχιζόμενη ανακάλυψη νέων υλικών, αποδοτικών και με χαμηλό κόστος. Στην παρούσα εργασία θα αναλύσουμε την τεχνολογία παραγωγής υδρογόνου με ηλεκτρόλυση καθώς και την παρασκευή υλικών με τη μέθοδο της ηλεκτροχημικής εναπόθεσης διμεταλλικών και τριμεταλλικών λεπτών υμενίων μεταβατικών μετάλλων με βάση το Νικέλιο (Ni). Τα κράματα Νικελίου με προσθήκη άλλων μετάλλων όπως Fe, Co, Mo κ.α. έχουν τεράστιο ερευνητικό ενδιαφέρον λόγω των πολύ καλών ηλεκτροκαταλυτικών ιδιοτήτων που παρουσιάζουν όσον αφορά την παραγωγή υδρογόνου, καθώς παρουσιάζουν καλύτερες ιδιότητες και από την Πλατίνα(Pt), που αποτελεί το αποδοτικότερο αλλά και το ακριβότερο μέταλλο που έχει χρησιμοποιηθεί ως καταλύτης για παραγωγή υδρογόνου. Η τεχνολογία της ηλεκτροχημικής εναπόθεσης λεπτών υμενίων προσφέρει υλικά πορώδη, αποδοτικά, φθηνά και ανθεκτικά σε συνθήκες έντονο αλκαλικού και όξινου περιβάλλοντος. Τα υλικά που παρασκευάσαμε ήταν NiFe, NiFeZn, NiCoZn, NiCoFeZn και NiMoZn. Κύριος στόχος της εργασίας αυτής ήταν η μελέτη των συνθηκών της ηλεκτροχημικής εναπόθεσης και ο τρόπος με τον οποίο επηρεάζουν τη μορφολογία και τις ιδιότητες των λεπτών υμενίων. Αρχικά μελετήσαμε τα υμένια σε πραγματικές συνθήκες ηλεκτρόλυσης, μετρώντας την παραγωγή υδρογόνου για συγκεκριμένες πάντα συνθήκες. Στη συνέχεια έγιναν μετρήσεις υπερδυναμικού παραγωγής υδρογόνου και από τα δεδομένα αυτά και με τη βοήθεια των διαγραμμάτων Tafel υπολογίσαμε ηλεκτροκαταλυτικούς συντελεστές όπως η πυκνότητα ρεύματος ανταλλαγής και η κλίση Tafel. Τέλος μετρήσεις σε ηλεκτρονικό μικροσκόπιο σάρωσης (SEM) και στοιχειακές αναλύσεις (EDX) μας έδωσαν μια πιο ολοκληρωμένη εικόνα για την επίδραση των συνθηκών της ηλεκτροχημικής εναπόθεσης στην μορφολογία της επιφάνειας των λεπτών υμενίων και των ιδιοτήτων τους. Μελετήσαμε τις ηλεκτροκαταλυτικές ιδιότητες ενός νέου υλικού (NiCoFeZn), καταλήγοντας στο συμπέρασμα ότι είναι ένα πολλά υποσχόμενο υλικό για την παραγωγή υδρογόνου και για το λόγο αυτό πρέπει να μελετηθεί εκτενέστερα, καθώς σήμερα χρησιμοποιείται κυρίως ως μαγνητικό υλικό. / The continuous environmental pollution and the depletion of natural resources have brought forward the need for the development of new renewable energy technologies. Hydrogen technologies and especially electrolytic hydrogen production, has gained considerable ground in the recent years due to the discovery of new, efficient and low cost alloy compounds. In this work we analyze the electrolytic hydrogen production and the fabrication of electrodeposited binary and ternary Nickel-based thin film alloys. The addition of the Fe, Co, Mo transition metals in Nickel-based alloys is of great interest in terms of research, because of their great electrocatalytic properties and high hydrogen production. These alloys even surpass Platinum (Pt), which is the most efficient yet also one of the most expensive metal used as hydrogen production catalyst. The electrodeposition of thin film alloys, offers porous, inexpensive and resistive in basic and acidic environment, compounds. The alloys we fabricated were NiFe, NiFeZn, NiCoZn, NiCoFeZn and NiMoZn. Our main goal was to study the electrodeposition process conditions and the manner in which they affect the morphology and the properties of the thin film alloys. At first we studied the electrocatalytic properties of the thin film alloys during the electrolytic production process using the same conditions every time. The next step was to measure the hydrogen production overvoltage and the estimation of electrocatalytic constants such as the exchange current density and the Tafel slope using the Tafel diagrams. Finally SEM micrographs and EDX stoichiometry analysis gave more accurate results on the morphology and the electrocatalytic properties of the thin film alloys. We also studied the electrocatalytic properties of a new compound (NiCoFeZn), which is in fact a very promising compound for the electrolytic hydrogen production. Today the research around this material is mainly focused on its magnetic properties, so a thorough study of its catalytic properties would be of great importance.

L’hydrogène est un des moyens envisageables pour réduire les émissions des gaz à effet de serre. Celui-ci est un carburant très abondant, et sa combustion est très énergétique que le pétrole (3 fois supérieure au pétrole). L’un des obstacles de son utilisation est son stockage. Le stockage à l’état solide présente de gros avantages en termes de capacité volumique (i.e. 100 à 200 g/L) et de sécurité. L’hydrure de magnésium MgH2 est le candidat qui présente les meilleurs résultats en termes de capacité massique (7,6 %wt.). Cependant, il est défavorisé par des cinétiques d’hydruration lentes et une température d’utilisation élevée (i.e. hydrure très stable).Nous nous sommes intéressés aux intermétalliques riches en magnésium TR-M-Mg (TR = Nd, Gd et M = Cu, Ni). Nous avons mis en évidence 3 nouvelles compositions : NdNiMg5, GdCuMg4 et GdCuMg12. Un traitement post-fusion à 700°C pendant une semaine suivi d’un refroidissement lent à 6°C / h jusqu’à 300°C permet d’obtenir ces phases. Seule la première composition a pu faire l’objet d’une étude complète. Elle présente un ordre antiférromagnétique à 12 K et sa capacité réversible de stockage est de 2,8 wt.%. Un échantillon presque pur pour la phase GdCuMg12 a pu être obtenu (a = 9,9721(8) Å et c = 7,775(6) Å et G.E. P4/m). Dans le cas de GdCuMg4, les mêmes conditions expérimentales nous ont permis d’obtenir un échantillon presque pur. Sa structure n’a pas encore pu être déterminée. / Hydrogen is one of the means to reduce emissions of greenhouse gas emissions. This is a very abundant fuel and its combustion is highly energetic (3 times more than petrol). An obstacle to its use is its storage. Storage in the solid state has significant advantages in terms of volume capacity (100 to 200 g/L) and safety. Magnesium hydride MgH2 is the candidate who shows the best results in terms of specific capacity (7,6 %wt) . However, it is disadvantaged by slow hydrogenation kinetics and high temperature use (very stable hydride). We are interested in intermetallic magnesium rich RE-TM-Mg (RE = Nd, Gd and TM = Cu, Ni). We highlighted three new compositions: NdNiMg5, GdCuMg4 and GdCuMg12. A post- fusion treatment at 700°C for one week followed by slow cooling at 6°C / h up to 300°C allows to obtain these phases. Only the first composition has been the subject of a comprehensive study. It has an antiferromagnetic ordering at 12 K and reversible storage capacity of 2,8 %wt. An almost pure sample for GdCuMg12 phase could be obtained (a = 9,9721(8) Å, c = 7,775(6) Å and space group: P4 / m). In the case of GdCuMg4, the same experimental conditions allowed us to obtain a nearly pure sample. Its structure has not yet been determined

L’utilisation des combustibles fossiles est responsable de l’augmentation de la concentration en gaz à effet de serre dans l’atmosphère. Parmi les solutions de remplacement envisagées, l’hydrogène constitue un vecteur d’énergie très intéressant. Toutefois, cette solution ne sera envisageable que lorsque les problématiques liées à la production de l’hydrogène et à son stockage seront résolues.Le premier objectif de cette thèse porte sur la synthèse et la caractérisation de composés ternaires à base de magnésium dans le système ternaire TR-M-Mg (avec TR = Terres Rares et M = métaux de transition) qui pourraient être de bons candidats pour le stockage de l’hydrogène. Ces composés pourraient de plus avoir d’autres applications, notamment comme matériaux de structure, du fait de leur très faible densité. La composition NdNiMg15 a fait l’objet d’une étude complète. Cette phase cristallise selon une symétrie quadratique avec a= 10,0602(1) et c= 7,7612(2) Å et un groupe d’espace P4/nmm. Un ordre antiferromagnétique à 9 K est observé et la capacité massique de stockage réversible est de 4 %mass. Cette nouvelle phase a montré un effet durcissant sur le magnésium.Le deuxième objectif de cette thèse concerne la production d’hydrogène par hydrolyse i) des composés ternaires TR-M-Mg qui pourrait être considérée comme une possibilité économique et énergétique pour valoriser les déchets de ces composés et ii) des mélanges ternaires TR-M-Mg élaborés par broyage mécanique. Le broyage a permis la création des défauts favorisant ainsi la corrosion des métaux. De plus, la production d'hydrogène par hydrolyse des composites Mg-NdNiMg15 (70, 80 et 90 %mass. Mg) a été réalisée et comparée à celle du composé NdNiMg15 (64 %mass. en Mg). Le mécanisme de corrosion principal déduit des essais électrochimiques sur les composites est la corrosion galvanique. / The use of fossil fuels (non-renewable energy) is responsible for the increase of the concentration of greenhouse gases in the atmosphere. Among the considered alternatives, hydrogen is seen as the most attractive energy vector. Production and storage of hydrogen is one of the key challenges in developing the hydrogen economy.The first objective of this thesis deal with the synthesis and characterization of magnesium-based ternary compounds in the RE-TM-Mg ternary system (with RE = Rare Earth and TM = transition metals) which could be good candidates for hydrogen storage. These compounds could also have other applications than the hydrogen storage in the future such as light structured material. The NdNiMg15 compound has been the subject of a completed study. This phase crystallizes with a tetragonal symmetry (a= 10.0602(1) and c= 7.7612(2) Å and a space group P4/nmm). It showed an antiferromagnetic ordering at 9 K and a reversible hydrogen storage capacity of 4 %mass. This phase exhibited a hardening effect respect to magnesium compound.The second objective of this thesis concerns the hydrogen production by hydrolysis of i) RE-TM-Mg ternary compounds, which could be considered as an economic and energetic possibility to valorize the waste of these compounds and ii) RE-TM-Mg ternary mixtures prepared by ball milling. The grinding creates defects thus promoting the corrosion of the metals. In addition, the hydrogen production by hydrolysis of the Mg-NdNiMg15 composites (70, 80 and 90 %mass. Mg) was carried out and compared with that of the NdNiMg15 compound (64 %mass. Mg). The main corrosion mechanism determined from the electrochemical measurements of the composites is the galvanic corrosion.

In den letzten Jahren hat sich maschinelles Lernen als hilfreiches Werkzeug zur Vorhersage von simulierten Materialeigenschaften erwiesen. Somit können aufwendige Berechnungen mittels Dichtefunktionaltheorie umgangen werden und bereits bekannte Materialien besser verstanden oder sogar neuartige entdeckt werden. Eine zentrale Rolle spielt dabei der Deskriptor, ein möglichst interpretierbarer Satz von Materialkenngrößen. Diese Arbeit präsentiert einen Ansatz zur Auffindung von Deskriptoren für periodische Multikomponentensysteme, deren Eigenschaften durch die genaue atomare Anordnung mitbeinflusst wird. Primäre Features von Einzel-, Paar- und Tetraederclustern werden über die Superzelle gemittelt und weiter algebraisch kombiniert. Aus den so erzeugten Kandidaten wird mittels Dimensionalitätsreduktion ein geeigneter Deskriptor identifiziert. Zudem stellt diese Arbeit Strategien vor bei der Modellfindung Kreuzvalidierung einzusetzen, sodass stabilere und idealerweise besser generalisierbare Deskriptoren gefunden werden. Es werden außerdem mehrere Fehlermaße untersucht, die die Qualität der Deskriptoren bezüglich Genauigkeit, Komplexität der Formeln und Berücksichtung der atomaren Anordnung charakterisieren. Die allgemeine Methodik wurde in einer teilweise parallelisierten Python-Software implementiert. Als konkrete Problemstellungen werden Modelle für die Gitterkonstante und die Mischenergie von ternären Gruppe-IV Zinkblende-Legierungen "gelernt", mit einer Genauigkeit von 0.02 Å bzw. 0.02 eV. Datenbeschaffung, -analyse, und -bereinigung werden im Hinblick auf die Zielgrößen als auch auf die primären Features erläutert, sodass umfassende Analysen und die Parametrisierung der Methodik an diesem Testdatensatz durchgeführt werden können. Als weitere Anwendung werden Gitterkonstante und Bandlücken von binären Oktett-Verbindungen vorhergesagt. Die präsentierten Deskriptoren werden mit den Fehlermaßen evaluiert und ihre physikalische Relevanz wird abschließend disktutiert. / In the last years, machine learning methods have proven as a useful tool for the prediction of simulated material properties. They may replace effortful calculations based on density functional theory, provide a better understanding of known materials or even help to discover new materials. Here, an essential role is played by the descriptor, a desirably interpretable set of material parameters. This PhD thesis presents an approach to find descriptors for periodic multi-component systems where also the exact atomic configuration influences the physical characteristics. We process primary features of one-atom, two-atom and tetrahedron clusters by an averaging scheme and combine them further by simple algebraic operations. Compressed sensing is used to identify an appropriate descriptor out from all candidate features. Furthermore, we develop elaborate cross-validation based model selection strategies that may lead to more robust and ideally better generalizing descriptors. Additionally, we study several error measures which estimate the quality of the descriptors with respect to accuracy, complexity of their formulas and the capturing of configuration effects. These generally formulated methods were implemented in a partially parallelized Python program. Actual learning tasks were studied on the problem of finding models for the lattice constant and the energy of mixing of group-IV ternary compounds in zincblende structure where an accuracy of 0.02 Å and 0.02 eV is reached, respectively. We explain the practical preparation steps of data acquisition, analysis and cleaning for the target properties and the primary features, and continue with extensive analyses and the parametrization of the developed methodology on this test case. As an additional application we predict lattice constants and band gaps of octet binary compounds. The presented descriptors are assessed quantitatively by the error measures and, finally, their physical meaning is discussed.

Thermoelectric (TE) materials directly convert heat energy into electrical energy. The conversion efficiency of the TE devices depends on the performance of the materials. The conversion efficiency of available thermoelectric materials and devices is low. Therefore, the development of new materials for improving thermoelectric device performance is a highly essential. As the performance of the TE materials depends on TE figure of merit [zT=S2P T ] which consist of three material properties such as Seebeck coefficient (S), electrical resistivity ( ) and thermal conductivity ( ). Thermoelectric figure of merit can be improved by either increase of power factor or decreasing of thermal conductivity or by both. In the present thesis, Cu based chalcogenide compounds are chosen for the study of thermoelectric properties because of their complex crystal structure, which leads to lower values of thermal conductivity. Also, the power factor of these materials can be tuned by the partial substitution doping. In the present thesis, Cu based chalcogenide compounds quaternary chalcogenide compound (Cu2ZnSnSe4), ternary compounds (Cu2SnSe3 and Cu2GeSe3) and tetrahedrite materials (Cu12Sb4S13) have been prepared by solid state synthesis. The prepared compounds are characterized by XRD for the phase identification, Raman Spectroscopy used as complementary technique for XRD, SEM for surface morphology and EPMA for the phase purity and elemental composition analysis respectively. For the evaluation of zT, thermoelectric properties of all the samples have been studied by measuring Seebeck coefficient, resistivity and thermal diffusivity. In the chapter 1, a brief introduction about thermoelectricity and its effects is discussed. Thermoelectric materials parameters such as electrical resistivity, Seebeck coefficient and thermal conductivity for different class of materials are mentioned. The selection of thermoelectric materials and the motivation for choosing the Cu based chalcogenide compounds for thermoelectric applications are discussed. In chapter 2, the details of the experiments carried out for Cu based chalcogenide compounds are presented. In chapter 3, the effect on thermoelectric properties by the cation substitution on quaternary chalcogenide compound Cu2+xZnSn1 xSe4 (0, 0.025, 0.05, 0.075, 0.1, 0.125, and 0.15) is studied. The electrical resistivity of all the samples decreases with an increase in Cu content except for Cu21ZnSn09Se4, most likely due to a higher content of the ZnSe. All the samples showed positive Seebeck coefficients indicating that holes are the majority charge carriers. The thermal conductivity of doped samples was higher as compared to Cu2ZnSnSe4 and this may be due to the larger electronic contribution and the presence of the ZnSe phase in the doped samples. The maximum zT = 0.23 at 673 K is obtained for Cu205ZnSn095Se4. In chapter 4, the effect of multi{substitution of Cu21ZnSn1 xInxSe4 (0, 0.05, 0.075, and 0.1) on transport properties were studied. The Rietveld powder X-ray diffraction data accompanied by electron probe microanalysis (EPMA) and Raman spectra of all the samples con firmed the formation of a tetragonal kesterite structure. The electrical resistivity of all the samples exhibits metallic-like behavior. The positive values of the Seebeck coefficient and the Hall coefficient reveal that holes are the majority charge carriers. The co-doping of copper and indium leads to a significant increase of the electrical resistivity and the Seebeck coefficient as a function of temperature above 650 K. The thermal conductivity of all the samples decreases with increasing temperature. Lattice thermal conductivity is not significantly modified as the doping content may infer negligible mass fluctuation scattering for copper zinc and indium tin substitution. Even though, the power factors (S2 ) of indium-doped samples Cu21ZnSn1 xInxSe4 (x=0.05, 0.075) are almost the same, the maximum zT=0.45 at 773 K was obtained for Cu21Zn09Sn0925In0075Se4 due to its smaller value of thermal conductivity. In chapter 5, thermoelectric properties of Zn doped ternary compounds Cu2ZnxSn1 xSe3 (x = 0, 0.025, 0.05, 0.075) were studied. The undoped com\pound showed a monoclinic crystal structure as a major phase, while the doped compounds showed a cubic crystal structure confirmed by powder XRD (X-Ray Diffraction). The electrical resistivity decreased up to the samples with Zn content x=0.05 in Cu2ZnxSn1 xSe3, and slightly increased in the sample Cu2Zn0075Sn0925Se3 . This behavior is consistent with the changes in the carrier concentration confirmed by room temperature Hall coefficient data. Temperature dependent electrical resistivity of all samples showed heavily doped semiconductor behavior. All the samples exhibit positive Seebeck coefficient (S) and Hall coefficient indicating that the majority of the carriers are holes. A linear increase in Seebeck coefficient with increase in temperature indicates the degenerate semiconductor behavior. The total thermal conductivity of the doped samples increased with a higher amount of doping, due to the increase in the carrier contribution. The total and lattice thermal conductivity of all samples decreased with increasing of temperature, which points toward the dominance of phonon scattering at high temperatures. The maximum zT = 0.34 at 723 K is obtained for the sample Cu2SnSe3 due to a low thermal conductivity compared to the doped samples. In chapter 6, thermoelectric properties of Cu2Ge1 xInxSe3 (x = 0, 0.05, 0.1, 0.15) compounds is studied. The powder X-ray diffraction pattern of the undoped sample revealed an orthorhombic phase. The increase in doping content led to the appearance of additional peaks related to cubic and tetragonal phases along with the orthorhombic phase. This may be due to the substitutional disorder created by indium doping. The electrical resistivity ( ) systematically decreased with an increase in doping content, but increased with the temperature indicating a heavily doped semiconductor behavior. A positive Seebeck coefficient (S) of all samples in the entire temperature range reveal holes as predominant charge carriers. Positive Hall coefficient data for the compounds Cu2Ge1 xInxSe3 (x= 0, 0.1) at room temperature (RT) con rm the sign of Seebeck coefficient. The trend of as a function of doping content for the samples Cu2Ge1 xInxSe3 with x = 0 and 0.1 agrees with the measured charge carrier density calculated from Hall data. The total thermal conductivity increased with rising doping content, attributed to an increase in carrier thermal conductivity. The thermal conductivity decreases with increasing temperature, which indicates the dominance of Umklapp phonon scattering at elevated temperatures. The maximum thermoelectric figure of merit (zT) = 0.23 at 723 K was obtained for Cu2In01Ge09Se3. In chapter 7, thermoelectric properties of Cu12 xMn1 xSb4S13 (x = 0, 0.5, 1.0, 1.5, 2.0) samples were studied. The Rietveld powder XRD pattern and Electron Probe Micro Analysis revealed that all the Mn substituted samples showed a single tetrahedrite phase. The electrical resistivity increased with increasing Mn due to substitution of Mn2+ on the Cu1+ site. The positive Seebeck coefficient for all samples indicates that the dominant carriers are holes. Even though the thermal conductivity decreased as a function of increasing Mn, the thermoelectric figure of merit (zT) decreased, because the decrease of the power factor is stronger than the decrease of the thermal conductivity. The maximum zT = 0.76 at 623 K is obtained for Cu12Sb4S13. In chapter 8, the summary and conclusion of the present work is presented.