abstract / Chemistry / Doctoral / Doctor of Philosophy

The main aim of this thesis is the development of $^{17}$O solid-state nuclear magnetic resonance (NMR) spectroscopic techniques to study the local structure and ion dynamics of functional oxide materials for applications in energy conversion, in particular as electrodes and electrolytes in solid oxide fuel cells (SOFCs). Broadly, the work comprises two related areas: (1) application of a combined experimental and computational methodology to enable the first $^{17}$O solid-state NMR studies of paramagnetic oxides, in particular a class of perovskite-derived structures used as mixed ionic-electronic conductors (MIECs) for SOFC cathodes, and (2) further uses of multinuclear variable-temperature NMR spectroscopy, with emphasis on $^{17}$O NMR results, to elucidate mechanistic details of oxide-ion motion and sublattice exchange in a novel family of promising SOFC electrolyte materials based on $\delta$-Bi$_{2}$O$_{3}$. In the first section, $^{17}$O magic-angle spinning (MAS) NMR spectra of the paramagnetic MIEC, La$_{2}$NiO$_{4+\delta}$, are presented and rationalized with the aid of periodic DFT calculations. Advanced NMR pulse programming and quadrupolar filtering techniques are coupled to extract high-resolution spectra. In particular, these data reveal local structural distortions in La$_{2}$NiO$_{4+\delta}$ that arise from incorporation of interstitial oxide defects. Moreover, variable-temperature spectra indicate the onset of oxide-ion motion involving the interstitials at 130 °C, which is linked to an orthorhombic$-$tetragonal phase transition. By analyzing the ion dynamics on the spectral timescale, specific motional mechanisms are elucidated that prove relevant to understanding the functionality and conductivity of this phase. Next, a similar methodology is applied to the Sr-doped analogues, La$_{2-x}$Sr$_{x}$NiO$_{4+\delta}$, in an exploration of the defect chemistry and electronic structure of these phases (0 $\leq {x} \leq$ 1). By following the doping-induced evolution of spectral features assigned to interstitial and equatorial oxygen environments, changes in the ionic and electronic conductivity, respectively, are rationalized. This approach has been extended to the acquisition and assignment of $^{17}$O NMR spectra of isostructural Sm$_{2-x}$Sr$_{x}$NiO$_{4+\delta}$ and Pr$_{2-x}$Sr$_{x}$NiO$_{4+\delta}$ phases, promising SOFC cathode materials that exhibit paramagnetism on the A site (A = Sm, Pr). The final section details the characterization of oxide-ion motion in the fluorite-type phases Bi$_{1-x}$V$_{x}$O$_{1.5+x}$ and Bi$_{1-x}$P$_{x}$O$_{1.5+x}$ ($x$ = 0.087 and 0.148) developed as SOFC electrolytes. Variable-temperature NMR experiments between room temperature and 923 K reveal two distinct mechanisms. For the V-doped phases, an oxide-ion conduction mechanism is observed that involves oxygen exchange between the Bi-O sublattice and rapidly rotating VO$_{4}$ tetrahedral units. The more poorly conducting P-doped phase exhibits only vacancy conduction with no evidence of sublattice exchange, a result ascribed to the differing propensities of the dopants to undergo variable oxygen coordination. These initial insights suggest chemical design rules to improve the next generation of oxide-ion conducting materials.

Le composite LiF-Cu fonctionnant selon la réaction de conversion : Cu + 2 LiF →CuF2 + 2 Li+ + 2 e- a été synthétisé afin d’être étudié comme matériau d’électrode positivepour batterie au lithium. Pour faire réagir ces phases à l’état solide, les îlots de LiF nedoivent pas excéder quelques nanomètres et la percolation électrique dans l’épaisseur doitêtre établie. La technique retenue pour obtenir cette nanostructuration est la co-pulvérisationcathodique RF alternée. Afin de contrôler la synthèse du nanocomposite, la vitesse de dépôtet le flux d’adatomes pour chaque matériau ont été mesurés par profilométrie et ICP-OES.Leur composition et leur morphologie ont été caractérisées par microscopie électronique(SEM, TEM, STEM, HRTEM, AFM), diffraction (XRD, SAED) et spectroscopie (EELS, XPS).Les résultats ont montré que la morphologie obtenue était bien faite d’îlots nanométriques deCu et de LiF. La conductivité électrique du nanocomposite est de 5 ordres de grandeursupérieure à celle du LiF seul. Des premiers tests en batterie ont montré une fortepolarisation typique des matériaux de conversion et une faible cyclabilité. / The LiF-Cu nanocomposite reacting through the conversion reaction: Cu + 2 LiF→ CuF2 + 2 Li+ + 2 e- has been synthesized in order to be studied as positive electrodematerial for lithium battery. In order to perform this reaction in the solid state, the size of theLiF islands mustn't exceeds a few nanometers an the electric percolation through thethickness must be achieved. The technique used to obtain this nanostructuration is thealternated RF co-sputtering. In order to control the nanocomposite synthesis, the rate ofdeposition and the adatoms flux for each material have been measured using profilometryand ICP-OES. Their composition and morphology have been characterised by electronicalmicroscopy (SEM, TEM, STEM, HRTEM, AFM), diffraction (XRD, SAED) and spectroscopy(EELS, XPS). Results have shown that the morphology was indeed made of nanometricislands of Cu and LiF. The nanocomposite electric conductivity is 5 order of magnitudehigher than the LiF one. The first tests have shown a high polarization typical of conversionmaterials and a poor cyclability.