Degree Name

Doctor of Philosophy


Department of Materials Engineering


Rechargeable nickel-metal hydride (Ni-MH) batteries employing hydrogen storage alloys as their negative electrode materials have been some of the most promising power sources for many important applications. For most commercial Ni-MH batteries, their negative electrodes are of rare-earth system alloys, which have low theoretical discharge capacities (e.g., LaNi5 alloy has a theoretical discharge capacity of only about 370 mAh/g). The capacities of the negative alloy electrodes restrict further improvement of the capacities of nickel-metal hydride batteries.

Magnesium-based hydrogen storage alloys are promising energy conversion and storage materials because they possess very high theoretical hydrogen storage capacity, eg., the theoretical discharge capacity of Mg2Ni alloy is approximately mAh/g. Also magnesium is abundant in nature, light in weight and relatively low incost. FeTi-type hydrogen storage alloys are also prime candidates for hydrogen storage systems and have high theoretical discharge capacities of about 500 mAh/g. FeTi-type alloys are also the cheapest of the promising hydrogen conversion and storage alloys. However, for a long period, it has been thought that Mg-based and FeTi type hydrogen storage alloys were unsuitable for Ni-MH negative electrodes because of their slow hydriding/dehydriding kinetics at ambient temperature and their susceptibility to corrosion in alkaline solution.

Amorphous and nanocrystalline metals and alloys have the features of new alloy compositions and new atomic configurations, which are different from those of crystalline alloys. These features enable various kinds of characteristics to be achieved such as excellent mechanical properties, useful physical properties and unique chemical properties, which have not been obtained for conventional crystalline alloys. Thus, there appear to be some possibilities to achieve high discharge capacities for amorphous and nanocrystalline Mg-based and FeTi-type alloys.

The purposes of this study are to exploit the electrode properties of novel amorphous and nanocrystalline Mg-based and FeTi-type hydrogen storage alloys as high performance and low cost negative materials for the Ni-MH batteries. For these purposes, amorphous and nanocrystalline Mg-based and FeTi-type hydrogen storage alloys were prepared by mechanical alloying and milling methods using a planetary ball mill. Their structures were determined by X-ray diffraction, differential thermal analysis and transmission electron microscopy. Their electrode behaviour was investigated in a 6 M KOH solution at room temperature.

A nanocrystalline FeTi alloy with an average grain size of about 8 nm can be obtained by mechanically milling a crystalline FeTi alloy for 40 hours. Alternatively,an amorphous FeTi alloy can be obtained by mechanically alloying mixtures of Feand Ti powders for 40 hours. Crystalline FeTi alloy has an extremely low discharge capacity of only 3 mAh/g at a discharge current density of 50 mA/g, which is negligibly small compared with its theoretical discharge capacity. Therefore, crystalline FeTi alloy is not suitable for use as a rechargeable anode material in alkaline solution at room temperature. The nanocrystalline FeTi alloy has an improved discharge capacity of 54 mAh/g at the first discharge, which is much higher than that of crystalline FeTi alloy, but is only around 10% of the calculated theoretical electrochemical capacity of FeTi alloy. Due to its good discharge capacity during the first twenty cycles, it may be possible to improve the electrochemical discharge capacity of FeTi-type alloy with good cycle life by achieving a nano crystalline structure. The amorphous FeTi alloy achieves its highest discharge capacity of 220 mAh/g at the first charge-discharge cycle at a discharge current density of 50 mA/g. This discharge capacity is almost comparable with the capacities of commercial alloys used in Ni-MH batteries. Although the discharge capacity of the amorphous FeTi alloy decreases rapidly with prolonged charge-discharge cycles,it still has a discharge capacity of 78 mAh/g at the twentieth cycle. Thus, it is effective method in achieving high discharge capacity for the FeTi-type alloy to form an amorphous or nanocrystalline structure.

In order to determine the effect of the preparation method for Mg-based alloys ontheir discharge capacity, the crystalline Mg-based alloys used were obtained in ways: sintering elemental powders and induction melting. At a discharge current density of 50 mA/g, crystalline Mg2Ni alloys prepared by sintering Mg and Ni powders have extremely low discharge capacities of only 15 or 18 mAh/g. Induction melted crystalline Mg2Ni alloy also has an extremely low discharge capacity of 18mAh/g. The discharge capacity of sintered crystalline Mg2Nio.6Coo.4 alloy is still very poor (about 20 mAh/g), demonstrating that cobalt addition does not alter the discharge capacity of crystalline Mg-Ni alloys. All the discharge capacities are negligibly small in comparison with their theoretical discharge capacity. Thus, crystalline Mg2Ni-type alloy is not suitable for use as a rechargeable anode material in an alkaline solution at room temperature.

A nanocrystalline Mg2Ni alloy with an average grain size of about 13 nm can be fabricated by mechanical milling induction melted (TM) crystalline Mg2Ni alloy. initial discharge capacity of the nanocrystalline Mg2Ni alloy is about 111 mAh/g,which is much higher than that of the TM crystalline Mg2Ni alloy (only 18 mAh/g).Like the nanocrystalline FeTi alloy, the discharge capacity of nanocrystalline Mg2Ni alloy is only 10% of the calculated theoretical electrochemical capacity of Mg2Ni alloy.

All the mechanically alloyed sintered Mg-Ni and Mg-Ni-Co alloy electrodes studied have completely or mainly amorphous structures, showing very high initial discharge capacities (around 300 mAh/g) compared with their crystalline counterparts. The addition of a small amount of Co to amorphous Mg-Ni alloy can increase its initial discharge capacity and cycle life. Increasing the Co addition in the amorphous Mg-Ni alloy results in a lower initial discharge capacity but a slightly better cycle All the results indicate that the amorphous structure is a key factor in order to high initial discharge capacities for Mg-based alloy electrodes. In the mechanical alloying process, the ball milling parameters have significant effects on the dischargecapacities of the Mg-based alloy electrodes. Increasing the ball milling time and ratio of ball to sample weight are effective methods to further improve the discharge capacity for Mg-based alloy electrodes.

Assuming that the composition of amorphous alloys can be obtained in a certain range, non-stoichiometric miiform amorphous MgNixVy alloys (x = 1, 1.28; y = 0,0.1, 0.2, 0.4) have been firstly synthesised by mechanically alloying induction melting Mg2Ni alloy, Ni and V powders based on a stoichiometric amorphous MgNi alloy component. The results indicate that non-stoichiometric amorphous Mg-based alloys can be obtained by either increasing the Ni content or adding a range of vanadium or both, through the mechanical alloying method. The non-stoichiometric Mg-based alloy electrodes studied have shown improved initial discharge capacities(more than 400 mAh/g) compared with the stoichiometric amorphous MgNi alloy (about 330 mAh/g). These results describe a novel method of achieving better M g -based alloy electrodes with high discharge capacities. The method enables a larger composition range to be achieved with a range of different elemental additions.

Finally, it is demonstrated that the initial discharge capacity of amorphous Mg-alloy electrodes is higher than that of rare-earth system and zirconium-based Laves phase alloys in commercial batteries. Thus, Mg-based hydrogen storage alloys are very promising negative materials for the Ni-MH battery. Also the results have shown that FeTi-type alloys having improved discharge capacity and good cycle life can be obtained by achieving an amorphous or nanocrystalline structure.



Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.