The rates of reaction of the constituents of sodium-potassium alloy with water have been determined in the temperature range 20 - 600C. They fall into two categories; the first is applicable to the instant the alloy meets the water, and the second applies to reaction of the metal through a bubble of hydrogen. The rates are widely different for these two stages, yet the activation energies are similar, being 38.3 and 33.0 KJ/mole respectively for sodium, and 24.5 and 27.3 KJ/mole respectively for potassium. The rate of reaction of sodium alone at 30 C, has also been determined. The behaviour of liquid metals when injected into water has been studied by high speed photography. Such jets disintegrate, after a short distance of travel, into small globules, each contained within a hydrogen gas bubble. These globules then travel upwards through water and consequently react much more slowly. The reaction rate may be reduced by the addition of small concentrations of mineral acids to the water, due to the, formation of salts at the metal-water interface which are less soluble than sodium hydroxide. Strong solutions of acid however, increase the rate of reaction. The addition of hydroxide ions as NH4OH has little effect on the rates. The metals undergo secondary reaction in that the hydrogen which is initially formed subsequently reacts with the metal to produce hydrides. These are eventually hydrolysed. The most probable reaction intermediate in the solution phase of the reaction is the solvated electron, e - (sq)' which has been detected photographically due to its absorption of light in the visible region of the spectrum. Overall reaction mechanisms for both reaction in solution and reactions at the metal surface have been proposed.

Aberration corrected transmission electron microscopy is a powerful tool for the structural characterisation of materials at the atomic scale, but the passage of high velocity electrons through the material can often induce structural changes via the transfer of large amounts of energy from the beam. The work in this thesis theoretically considers the nature of this transfer of energy and its impact upon the material being studied. The computational modelling of molecular species encapsulated inside carbon nanotubes and their response to electron irradiation is compared to results from TEM experiments, and used to explain the experimental observations. The high rate of destruction of C-H bonds under the beam is quantified, and its implications for TEM studies of organic materials considered. An effective solution for mitigating this rate of destruction is found, applied to a model system, and then confirmed experimentally. Using the considerations of stability under the beam, two experimentally witnessed reactions are investigated in detail, and careful comparison to intermediate structures observed in TEM allows full reaction mechanisms to be proposed. The dynamic motion of atomic defects in irradiated graphene is considered with the aid of a large library of experimental TEM images. A novel defect structure is observed, and is seen to undergo structural rearrangements on a quicker time-scale than accessible to TEM imaging. This species enables the very quick migration of defect structures across the graphene lattice, and is attributed to a trivacancy structure. The rates of beam induced reactions are considered in the framework of chemical kinetics, and a method for extracting kinetic parameters of a reaction from the statistics of a large number of TEM observations of it occurring is developed. This is used to obtain the first cross-sections for the formation and healing of the irradiation induced Stone-Wales bond rotation, and the first experimental activation energy for its healing. The latter agrees well with a theoretically predicted mechanism of catalysis, while the former demonstrates that the widely assumed process of direct knock-on damage cannot be responsible for the beam induced reaction. An alternative mechanism is proposed, resulting from the electronic excitation of the defect.