In order to understand the intrinsic and extrinsic degradation mechanism, the degradation behaviors of the hydrogen absorption capacity and the hydriding kinetics of $LaNi_5$, $LaNi_{4.7}Al_{0.3}$ and $MmNi_{4.5}Al_{0.5}$ are investigated during the hydriding-dehydriding cycling in pure hydrogen and mixed gases ($H_2$-Co and $H_2$-$O_2$). And then, the degradation behaviors of the hydrogenation properties of $LaNi_5$ during the cycling are simulated. These theoretical results are compared with the experimental ones. Also, the kinetic studies on the hydriding reaction are carried out, because informations on the rate controlling step for the hydriding reaction are required for the computer simulation.
In Chapter II, the intrinsic degradation behavior of $LaNi_5$ is investigated by the changes of Pressure-Composition-Temperature (P-C-T) curves of $LaNi_5-H_2$ system and the hydrogen desorption technique after the pressure induced hydrogen absorption-desorption cycling. Thermal desorption experiment of $LaNi_5$ hydrides shows only one peak in the hydrogen evolution rate vs. temperature plot for the activated samples but two peaks for the degraded samples. Low temperature peak of the degraded samples and a peak of the activated samples are related to the hydrogen evolved from the sites which can absorb and desorb reversibly. The high temperature peak of the degraded samples seems to be associated with the hydrogen from the stable hydrides such as several series of La-Ni stoichiometric compound hydrides and $LaH_2$. Such stoichiometric compound hydrides and $LaH_2$ will be formed by the disordering of the lattice and the changes of local composition of the degraded $LaNi_5$ hydrides. Ni clusters formed at the surface by the cycling are confirmed by the magnetization measurement, which indicates the formation of a series of stable hydrides. Thus it is concluded that the formation of the various stoichiometric compound hydrides and $LaH_2$ were responsible for the intrinsic degradation of $LaNi_5$ through the cycling.
In chapter III, in order to inspect the effects of CO impurity on the hydrogenation properties of $LaNi_5$, $LaNi_{4.7}Al_{0.3}$ and $MmNi_{4.5}Al_{0.5}$alloys, the changes of the amount of the absorbed hydrogen on each cycle are measured during the pressure induced hydriding-dehydriding cycling in hydrogen containing CO as an impurity. For all alloys, the amount of the absorbed hydrogen of the cycling decreased continuously as the number of cycling is increased. Thus, $LaNi_5$ and $LaNi_{4.7}Al_{0.3}$ are degraded completely within 20 cycles, $MmNi_{4.5}Al_{0.5}$ within 80 cycles. The lost of the hydrogen storage capacities is caused by the deactivation of the active sites for the dissociative chemisorption of hydrogen molecules by the preferential adsorption of the CO impurity. It is confirmed by the thermal desorption experiments for the fully degraded samples and the analysis of the composition of the gases evolved from the degraded specimens during the thermal desorption by the gas chromatography. Partial substitution of Ni by Al improves the resistance of the alloys to the CO impurity, which is due to the changes of the surface electronic structure induced by the partial substitution of Al for Ni. But the resistance of $LaNi_{4.7}Al_{0.3}$ alloy to the CO impurity is much poorer than that of $MmNi_{4.5}Al_{0.5}$ alloy, which is caused by the successive accumulation of the retained hydrogen in La $Ni_{4.7}Al_{0.3}$ which is not desorbed completely during the dehydriding period of each cycle.
In Chapter IV, the extrinsic degradation behaviors of $LaNi_5, LaNi_{4.7}Al_{0.3}$ and $MmNi_{4.5}Al_{0.5}$ alloy are examined by the changes of the amount of the absorbed hydrogen of each cycles during the pressure induced hydriding-dehydriding cycling extended up to 2450 cycles in hydrogen containing $O_2$ as an impurity. For all alloys, the absorbed hydrogen contents are reduced drastically at first cycle and recovered by the subsequent cycling, which may be due to the formation of the metallic Ni clusters on the surface of specimen. However, as the cycling is extended over a few thousand cycles, the storage capacity is decreased gradually again. The gradual decrease is caused by the decomposition of the alloy into the hydroxides of the rare earth metals and the metallic Ni clusters. The decomposition is verified through the observation of the characteristic peaks of $La(OH)_3$ and Ni clusters in the x-ray diffraction pattern of the fully degraded $LaNi_5$ samples after 2272 cycles. The decomposition is confirmed detecting the metallic Ni clusters in the degraded samples by the magnetization measurements. $LaNi_4.7Al_0.3$ and $MmNi_4.5Al_0.5$ alloy can absorb hydrogen after 2450 cycles, though $LaNi_5$ is fully degraded. The substituted Al atoms will reduce the amount of the absorbed oxygen by the formation of $Al_2O_3$ and retard the decomposition by restricting the change between La and Ni atoms in the neighboring sites because its atom size is much larger than that of Ni atoms. Therefore, the resistance of Al substituted alloys to $O_2$ are improved.
In Chapter V, the hydriding kinetics of $LaNi_5$ and $LaNi_4.7Al_0.3$ are studied at a constant pressure. In order to avoid the effect of the heat evolved in the hydriding reaction and to obtain the intrinsic kinetic data, the samples are mixed with a large amount of Al powders and the Cu tube reactor with good thermal conductivity is used. The dependence of the hydrogen absorption rate on the reacted fraction, F, and the applied hydrogen pressure shows that the dissociative chemisorption of hydrogen molecules on the sample surface is the rate controlling step up to F=0.6-0.8 and F=0.4-0.5 at the low applied hydrogen pressure for the hydriding reaction of $LaNi_5$ and $LaNi_4.7Al_0.3$ respectively. At the high applied hydrogen pressure, the rate controlling step is the nucleation and growth of the hydride phase at the site of the hydrogen chemisorption. At a later stage of the hydriding reaction, irrespective of the applied hydrogen pressure, the hydrogen diffusion through the hydride phase is the rate controlling step. In both alloys, the rate controlling step have changed from the dissociative chemisorption of hydrogen at the surface or the nucleation and growth to the hydrogen diffusiion through the hydride phase. The transition of the rate controlling step take place at much earlier stage in the hydriding reaction of $LaNi_4.7Al_0.3$ than $LaNi_5$, which is caused by the slower diffusion rate of hydrogen atom through the hydride phase in $LaNi_4.7Al_0.3$ than in $LaNi_5$ particles.
In Chapter VI, the degradation behaviors of the hydrogenation properties of $LaNi_5$ under the cyclic operations in the mixed gases ($H_2$-$O_2$ and $H_2$-CO) are simulated. For the simulations, the rate equations for the adsorption of CO, $O_2$ and $H_2$ gases are derived and two models which are different in the contamination types are considered. One is that the gaseous impurity poisons some particles which are come in touch with it immediately after the mixed gases are introduced newly during the hydriding reaction. The other is that the surface of all particles is contaminated evenly and partially on each cycle. In the former case, the alloy is deactivated strongly by the gaseous impurity and the shapes of the hydrogen absorption kinetic curves are not changed with the cycling. On the other hand, in the latter case, the effect of the gaseous impurity on the hydrogenation properties of the alloy is relatively mild. However, the hydriding rate is retarded and the shapes of the kinetic curve are changed continuously with the increase of the number of the cycles. Comparison of the changes of the theoretical kinetic curves with the experimental results shows that the alloy is degraded by the former type with the cycling in $H_2$-CO and by the latter type in $H_2$-$O_2$.
