>Magister Scientiae - MSc / Transition metals are a group of metals which are light in weight and have high hydrogen solubility. Their interaction with hydrogen is exorthermic and this phenomenon makes them “ideal” candidates for various applications of hydrogen storage systems. This explains why the phenomenon of hydrogen storage in Pd is used as a model for hydrogen storage systems because of the nature of absorption associated with it (like a sponge even at low temperatures). The hydrogenation process can be conducted at either room or high temperatures in a furnace under low pressure-low hydrogen gas concentration-short hydrogenation time (LP-LC-ST) and in intelligent gravimetric analyser under high pressurehigh hydrogen gas concentration-long hydrogenation time conditions. Most of the research on hydrogen storage sytems is based on gravimetric analysis of absorbed and desorbed hydrogen concentration. In this work, a comparison study of the hydrogen content in pure Pd, Pd-Pt coated systems, Pd-Pt alloys, commercially pure Ti and Ti-6Al-4V alloy determined by gravimetric methods and elastic recoil detection analysis (which is based on the detection of recoiled hydrogen after interaction with He+ ions) technique was investigated. The changes in the structural properties and the hydrogen content of the materials when exposed to a hydrogen gas environment for different durations at various system temperatures and pressures will be reported. These changes have an effect on the microstructure of CP-Ti and Ti-6Al-4V alloy and structural properties of all the hydrogenated materials. The results obtained from optical microscopy, scanning electron microscopy, x-ray diffraction, intelligent gravimetric analyser, digital balance, elastic recoil detection analysis and Vickers hardness test, show the following: it is found that hydrogenation of Pd at elevated temperatures (550 ˚C and 650 ˚C) does not yield hydrides under LP-LC-ST conditions. However, at room temperature the absorption of hydrogen occurred faster at the beginning of the process. Furthermore, the absorption of hydrogen increased with pressure where optimum absorption (0.67 wt. % hydrogen concentration) occurred under a system pressure of 2000 mbar. After pressure release, the remaining hydrogen content in the Pd sample was 0.6 wt. %. The Pd-Pt coated system provide hydrogen mobility at 550 and 650 ˚C where hydrides were formed under LP-LC-ST conditions. In addition to the decrease of hydrogen solubility in Pd-Pt alloys with an increase in Pt content, the probability of the alloys to achieve full saturation also decreases with an increase in Pt content under HP-HC-LT conditions. CP-Ti and Ti-6Al-4V alloy absorb substantial amount of hydrogen in the first hour of room temperature hydrogenation under LP-LC-ST conditions but hydrides were not formed. Therefore, under LP-LC-ST conditions at room temperature, Pd is able to store hydrogen in the form of hydrides whereas Ti and Ti-6Al-4V alloy could not. The 550 ˚C is the optimum temperature for hydrogenation of CP-Ti under LP-LC-ST conditions. The Ti- 6Al-4V alloy absorb optimum hydrogen at 650 ˚C under LP-LC-ST conditions. Consequently, the change of microhardness of CP-Ti and Ti-6Al-4V alloy was found to depend on hydrogenation temperature.

Release of metallic fission products from fully intact tri-structural isotropic (TRISO) fuel particles raises serious concern on the safety of high temperature gas-cooled reactors (HTGRs). In TRISO particles, SiC and/or ZrC coating is considered as the major barrier for the migration of the fission products. This thesis focuses on the migration mechanism study of Ag in SiC and Pd in ZrC.The mechanism of the migration of Ag in SiC is a long-lasting mystery. None of the currently existing models could satisfactorily explain the reported experimental facts. In this work, a new mechanism, termed as the “reaction-recrystallization” model, is proposed to explain the Ag migration behavior through SiC. Designed SiC/Ag diffusion couple experiments were carried out, and the results indicate that Ag migrates in SiC by the following three steps. First, Ag reacts with SiC to form an Ag-Si alloy (reaction). Second, carbon precipitates as a second phase and subsequently reacts with the Ag-Si alloy to form new β-SiC (recrystallization). Third, the Ag-Si alloy penetrates through the SiC layer by wetting its grain boundaries (migration). The validity of the proposed model was supported by thermodynamic calculations. (Chapter 3) The finding that SiC could be recrystallized in the presence of Ag inspires the idea of Ag-assisted crack healing in SiC. Cracks were intentionally generated by indenting the bulk SiC by a Vickers indenter. After vacuum annealing with Ag powder, the indent impressions were healed by newly-formed β-SiC grains with a recovery ratio of~ 60%. Median cracks were fully healed by both newly formed SiC and Ag-Si nodules. TEM observation reveals that the newly formed β-SiC layer is presented between the Ag-Si nodule and pristine SiC crack surface and smooths the tortuous crack surface. The above result is in potential to solve the problem of brittleness of SiC as a structural material. (Chapter 4)ZrC is considered as a candidate to replace SiC in TRISO fuel particles. The migration behavior of Pd in ZrC was investigated by designed Pd/ZrC diffusion couple experiments. It is found that ZrC reacts with Pd at temperatures higher than 600 °C to form Pd3Zr and amorphous carbon. The reaction kinetics parameters, i.e., the activation energy and the reaction order, along with the inter-diffusion coefficients of Zr and Pd, were calculated based on established models. These results provide preliminary explanation to the Pd migration in ZrC (Chapter 5).

The solid-state interaction of palladium (Pd) with single crystal silicon carbide (6H- SiC) before and after annealing has been investigated using Rutherford backscattering spectrometry (RBS) in conjunction with RUMP simulation package, time-of-flight elastic recoil detector analysis (ToF-ERDA), glancing incident X-ray diffraction (GIXRD) and scanning electron microscopy (SEM). A thin layer of Pd (500 A) was deposited onto a clean 6H-SiC substrate at room temper- ature. The prepared difusion couples were then annealed in vacuum at different annealing temperatures for a maximum period of 1 h. The annealing temperature ranged from 2000C to 8000C. The composition of the as-deposited and the annealed samples was measured by using a He+ beam with an energy of 1.6 MeV. The ToF-ERDA measurements were per- formed on the as-deposited sample by using a high energy copper beam (about 30 MeV) for elemental depth distribution. The GIXRD measurements performed on the samples were able to identify the phases that form before and after annealing. The SEM micrograph obtained during this study gave some insight on the surface morphology of the samples before and after annealing. Our results obtained during this study showed that Pd reacts with SiC after annealing at 4000C resulting in the formation of metal-rich silicides and some unreacted Pd. Annealing at higher temperatures (5000C and 6000C) produced metal-rich silicides, which continued to grow until all the Pd has been consumed. Annealing at even higher temperatures (7000C and 8000C), the metal-rich silicides disappear and the silicon rich silicides start appearing. These appear by simply consuming the metal-rich silicides, resulting in the formation of two or more phases. The behaviour of the interaction between Pd with 6H-SiC is different than the Pd-Si system. The reaction temperature of the Pd/SiC are much more higher than those of the Pd/Si system. That is, Pd reacts with Si at temperatures as low as 2500C, while it starts to react with SiC at an annealing temperature of 4000C. In addition to this silicides such as Pd9Si2, Pd4Si form at the initial reaction temperature followed by the formation of the Pd2Si phase at the temperatures above 6000C for the Pd/SiC system. Meanwhile in the Pd/Si system the Pd2Si phase remains stable even after annealing at 8000C. No carbon compounds were observed in the temperature range used in this study and the formation of silicides were found to be accompanied by the formation of free carbon which remained immobile in the system. / Dissertation (MSc)--University of Pretoria, 2012. / Physics / unrestricted

The ability to sense gas such as methane can provide an early warning system to protect human lives. High demand for the ability to sense the world around us has provided an extensive area of research for sensor technology. In particular, current sensor technology, specifically for methane, has provided sensors that require a heated environment to function. The majority of current methane sensors function at temperatures between 150°C and 450°C [1-3]. This thesis will explore an approach to produce a room temperature methane sensor.

Surfactant bound stable colloids of Au, Ag, and Pd were prepared by the solvated Metal Atom Dispersion (SMAD) method, a method involving co-condensation of metal and solvent vapors on the walls of a reactor at 77 k. The as=prepared dodecanethiol-capped Au and Ag colloids consisting of polydisperse nanoparticles were transformed into colloids consisting of highly monodisperse nanoparticles by the digestive ripening process. In the case of Pd colloids, digestive ripening led to the formation of thiolate complexes. The [Pd(SC12H25)2]6 complex formed from the dodecanethiol-capped Pd nanoparticles was found to be a versatile precursor for the synthesis of a variety of Pd nanophases such as Pd(0), PdS, and Pd@PdO by soventless thermolysis. Co-digestive ripening of as-prepared dodecanethiol-capped Au or Ag colloids with Pd colloid resulted in Au@Pd and Ag@Pd core-shell nanoparticles, respectively; attempts to transform the core-shell structures into alloy phases even at high temperatures were unsuccessful. Phosphine-capped Au nanoparticles were also prepared by the SMAD method and refluxing of this colloid resulted in an Ostwald ripening process rather than the expected digestive ripening due to the labile nature of bound PPh3. The labile nature of the bound phosphine was studied using 31P NMR spectroscopy and utilized in the adsorption of CO. Palladium nanoparticles obtained from the SMAD Pd-butanone colloids and Pd@PdO nanoparticles prepared by the solventless thermolysis of Pd-dodecanethiolate complex were found to be good catalysts for the generation of H2 from AB via either hydrolysis and methanolysis. The active hydrogen atoms produced during the hydrolysis and methanolysis diffuse into the Pd lattice. It was also noticed that hydrogen atoms that were buried deep inside the Pd lattice cannot be removed completely by heating the sample even at 600°C. Wet chemical reduction method was employed for the synthesis of PVP capped, nearly monodisperse, spherical Ir nanoparticles which undergo a polymer driven self-assembly at 80°C to afford rectangular structures and interlinked particles.

No description available.

After introducing fundamentals of catalysis with noble metal surfaces especially Pt metal for CO oxidation and subsequent developments on nano-crystalline Pt metals supported on oxide supports, an idea of Pt ion in reducible oxide supports acting as adsorption sites is proposed in chapter 1. Idea of red-ox cycling of an ion in an oxide matrix is presented taking Cu ion in YBa2Cu3O7 as an example. Noble metal ions in reducible oxides such as CeO2 or TiO2 acting as adsorption sites and hence a red-ox catalyst was arrived at from chemical considerations. Among several reducible oxide supports, TiO2 was chosen from crystal structure and electronic structure considerations. A good redox catalyst for auto exhaust and related applications should have high oxygen storage capacity (OSC). Any new material that can work as a redox catalyst should be tested for its OSC. Therefore we designed and fabricated a temperature programmed reduction by hydrogen (H2¬TPR) system to measure OSC. This is presented in chapter 2. We have synthesized a number of oxides by solution combustion method. Structures were determined by powder XRD and Rietveld refinement methods. Fe2O3, Fe2-xPdxO3-δ, Cu1-xMnAl1+xO4, LaCoO3, LaCo1-xPdxO3-δ, CeO2, Ce1¬xPdxO2-δ, TiO2, Ti1-xPdxO2-δ and many other oxide systems were synthesized and their structures were determined. OSC of these systems were determined employing the H2/TPR system. TPR studies were carried out for several redox cycles in each case. Except Pd ion substituted CeO2 and TiO2 other oxide systems decomposed during redox cycling. Pd ion substituted TiO2 gave highest OSC and also it was stable paving way to choose this system for further study. In chapter 3, we have described lattice oxygen of TiO2 activation by the substitution of Pd ion in its lattice. Ti1-xPdxO2-x (x = 0.01 to 0.03) have been synthesized by solution combustion method crystallizing in anatase TiO2 structure. Pd is in +2 oxidation state and Ti is in +4 oxidation state in the catalyst as seen by XPS. Pd is more ionic in TiO2 lattice compared to Pd in PdO. Oxygen storage capacity defined by ‘amount of oxygen that is used reversibly to oxidize CO’ is as high as 5100 μmol/g of Ti0.97Pd0.03O1.97. Oxygen is extracted by CO to CO2 in absence of feed oxygen even at room temperature. Rate of CO oxidation is 2.75 μmol.g-1.s-1 at 60 0C over Ti0.97Pd0.03O1.97 and C2H2 gets oxidized to CO2 and H2O at room temperature. Catalyst is not poisoned on long time operation of the reactor. Such high catalytic activity is due to activated lattice oxygen created by the substitution of Pd ion as seen from first-principles density functional theory (DFT) calculations with 96 atom supercells of Ti32O64, Ti31Pd1O63, Ti30Pd2O62 and Ti29Pd3O61. The compounds crystallize in anatase TiO2 structure with Pd2+ ion in nearly square planar geometry and TiO6 octahedra are distorted by the creation of weakly bound oxygens. Structural analysis of Ti31Pd1O63 which is close to 3% Pd ion substituted TiO2 shows that bond valence of oxygens associated with both Ti and Pd ions in the lattice is 1.87. A low bond valence of oxygen is characteristic of weak oxygen in the lattice compared to oxygens with bond valence 2 and above in the same lattice. Thus, the exact positions of activated oxygens have been identified in the lattice from DFT calculations. Pt has two stable valencies: +2 and +4. Ti ion in TiO2 is in +4 state. Is it possible to substitute Pt exclusively in +2 or +4 state in TiO2? Implications are that Pt in +2 will have oxide ion vacancies and Pt in +4 states will not have oxide ion vacancies. Indeed we could synthesize Pt ion substituted TiO2 with Pt in +2 and +4 states by solution combustion method. In chapter 4, we have shown the positive role of an oxide ion vacancy in the catalytic reaction. Ti0.97Pt2+0.03O1.97 and Ti0.97Pt4+0.03O2 have been synthesized by solution combustion method using alanine and glycine as the fuels respectively. Both are crystallizing in anatase TiO2 structure with 15 nm average crystallite size. X-ray photoelectron spectroscopy (XPS) confirmed Pt ions are only +2 state in Ti0.97Pt0.03O1.97 (alanine) and only in +4 state in Ti0.97Pt0.03O2 (glycine). CO oxidation rate with Ti0.97Pt2+0.03O1.97 is over 10 times higher compared to Ti0.97Pt4+0.03O2. The large shift in 100 % hydrocarbon oxidation to lower temperature was observed by Pt2+ ion substituted TiO2 from that by Pt4+ ion substituted TiO2. After reoxidation of the reduced compound by H2 as well as CO, Pt ions are stabilized in mixed valences, +2 and +4 states. The role of oxide ion vacancy in enhancing catalytic activity has been demonstrated by carrying out the CO oxidation and H2 + O2 recombination reaction in presence and in absence of O2. There is no deactivation of the catalyst by long time CO to CO2 catalytic reaction. We analyzed the activated lattice oxygens upon substitution of Pt2+ ion and Pt4+ ion in TiO2, using first-principles density functional theory (DFT) calculations with supercells Ti31Pt1O63, Ti30Pt2O62, Ti29Pt3O61 for Pt2+ ion substitution in TiO2 and Ti31Pt1O64, Ti30Pt2O62, Ti29Pt3O61 for Pt4+ ion substitution in TiO2. We find that the local structure of Pt2+ ion has a distorted square planar geometry and that of Pt4+ ion has an octahedral geometry similar to Ti4+ ion in pure TiO2. The change in coordination of Pt2+ ion gives rise to weakly bonded oxygens and these oxygens are responsible in high rates of catalytic reaction. Thus, the high catalytic activity results from synergistic roles of oxide ion vacancy and weakly bonded lattice oxygen. In chapter 5, we have shown high rates of H2 + O2 recombination reaction by Ti0.97Pd0.03O1.97 catalyst coated on honeycomb monolith made up of cordierite material. This catalyst was coated on γ¬Al2O3 coated monolith by solution combustion method using dip-dry-burn process. This is a modified conventional method to coat catalysts on honeycombs. Formation of Ti0.97Pd0.03O1.97 catalyst on monolith was confirmed by XRD. Form the XPS spectra of Pd(3d) core level in Ti1-xPdxO2-δ, Pd ion is the formed to be +2 state. Ti0.97Pd0.03O1.97 showed high rates of H2 + O2 recombination compared to 2 at % Pd(metal)/γ-Al2O3, Ce0.98Pd0.02O2-δ, Ce0.98Pt0.02O2-δ, Ce0.73Zr0.25Pd0.02O2-δ and Ti0.98Pd0.02O1.98. Activation energy of H2 + O2 recombination reaction over Ti0.97Pd0.03O1.97 is 7.8 kcal/mole. Rates of reaction over Ti0.97Pd0.03O1.97 are in the range of 10 – 20 μmol/g/s at 60 0C and 4174 h-1 space velocity. Rate is orders of magnitude higher compared to noble metal catalysts. From the industrial point of view, solvent-free hydrogenation of aromatic nitro compounds to amines at nearly 1 bar pressure is an important process. In chapter 6, we showed that Ti0.97Pd0.03O1.97 is a good –nitro to –amine conversion catalyst under solvent-free condition at 1.2 – 1.3 bar H2 pressure. Nitrobenzene, p-nitrotoluene and 2-chloro-4-nitrotoluene are taken for the catalytic reduction reaction. The amine products were analyzed by gas chromatography and mass spectrometry (GCMS). Further, confirmation of compounds was done by FTIR, 1H NMR and 13C NMR. In presence of alcohol as solvent, 100% conversion of aromatic nitro compounds to amines took place at higher temperature and it required more times. In n-butanol solvent, 100% conversion of nitrobenzene and p-nitrotoluene occurred within 10 h and 12 h at 105 °C respectively. We have compared solvent-free reduction of p-nitrotoluene over different catalysts at 90 °C. Catalytic activity for reduction of p¬nitrotoluene over Ti0.97Pd0.03O1.97 is much higher than that reaction over 3 atom % Pd on TiO2 and Pd metal. Turnover frequencies (TOF) for nitrobenzene and 2-chloro-4-nitrotoluene conversion are 217 and 20 over Ti0.97Pd0.03O1.97 respectively. With increase of temperature, TOF of aromatic nitro compound reduction is also increased. We have compared the solvent-free reduction of aromatic nitro compound over Ti0.97Pd0.03O1.97 with others in the literature. Upto 3 cycles of reduction reaction, there was no degradation of Ti0.97Pd0.03O1.97 catalyst and stability of catalyst structure was analyzed by XRD, XPS and TEM images. Catalyst is stable under reaction condition and the structure is retained with Pd in +2 state. Finally, we have proposed the mechanism of -nitro group reduction reaction based on the structure of Ti0.97Pd0.03O1.97. Instead of handling nano-crystalline materials we proceeded with coating our catalysts on cordierite honeycombs. In chapter 7, we have shown high catalytic activity towards Heck reaction over Ce0.98Pd0.02O2-δ and Ti0.97Pd0.03O1.97 coated on cordierite monolith. XRD patterns of Ce0.98Pd0.02O2¬δ coated on cordierite monolith were indexed to fluorite structure. Heck reaction of aryl halide with olefins over Ce0.98Pd0.02O2-δ and Ti0.97Pd0.03O1.97 coated on cordierite monolith were carried out at 120 °C. The products were first analyzed by GCMS and for the confirmation of compounds, we have recorded 1H NMR and 13C NMR. Heck reaction was carried out with different solvents and different bases for choosing the good base and a solvent. Hence, we have chosen K2CO3 as base and N,N¬dimethylformamide (DMF) as solvent. We have compared the rates of Heck reactions over these two catalysts and Ti0.97Pd0.03O1.97 catalyst showed much higher catalytic activity than Ce0.98Pd0.02O2-δ. With increase of temperature from 65 °C to 120 °C, the catalytic activity of Ti0.97Pd0.03O1.97 on Heck reaction is also increased. The catalyst was reused for next Heck reaction without significant loss of activity. A mechanism for Heck reaction of aryl halide with alkyl acrylate has been proposed based on the structure of Ti0.97Pd0.03O1.97. In chapter 8, we have provided a critical review of the work presented in the thesis. Critical issues such as noble metal ion doping in TiO2 vs noble metal ion substitution, difficulty of proving the substitution of low % noble metal ion in TiO2, need for better experimental methods to study noble metal ion in oxide matrix have been discussed. Finally, conclusions of the thesis are presented.