The Verwey transition in magnetite is a remarkable and challenging phenomenon which is still not understood despite many years of intense research. When the temperature is lowered below the transition temperature (Tv = 121 K) there is a sudden drop in the electrical conductivity and the structure- changes from cubic above to monoclinic below. The importance of this phase transition goes further than just this simple material. Magnetite stands as a model system in many areas of the physical sciences, but is probably most known for its use in the theory of 'metal-insulator' transitions, that are known to occur in many of the transition metal oxides. For a long time it was thought that the mechanism responsible for this transition was ionic charge ordering of the iron atoms. This model has since been proven to be incorrect and new models with partial charge ordering have largely taken its place. This work is concerned with using the neutron as a probe to study both the vibrational and magnetic excitations within magnetite and offers insight into the lattice dynamics around the phase transition. The transverse acoustic phonons have been intensively studied and interesting broadening effects observed. Furthermore the entire spin wave spectrum has been observed, with the top mode (known as (1)2) and a direction other than [1 0 0] reported for the ftrst time. From the work presented here it is suggested that the structural phase transition in magnetite is better understood using a band type model where both phases of magnetite (above and below) are in fact semiconducting. The broadening effects of the phonons are described by the presence of an anharmonic potential above the transition temperature and the sudden drop in the electrical conductivity is explained by the widening of the band gap below the transition.

No description available.

Neutron scattering and susceptibility measurements as a function of temperature are reported on alloys between the light rare earths Pr and Nd and the heavy rare earths Tb, Dy and Ho over the entire composition range. It is clear from the results that the alloys form three distinct crystallographic and magnetic phases dependent upon composition. The heavy rare earth rich alloys remain in the h.c.p. phase like the parent heavy rare earth metal and the magnetic properties are essentially those of a magnetic dilution system. The ordering temperatures are shown to follow a 2/3 power law with reduced deGennes function in common with many other rare earth alloy systems. Alloys containing approximately equal proportions of light and heavy rare earth metals adopt the complicated Sm. structure and, as a result, display a complex magnetic structure. The light rare earth rich alloys display the d.h.c.p. crystallographic structure like Pr and Nd. The magnetic properties of the ferrimagnetic d.h.c.p, alloys are accounted for in terms of a two sublattice model of magnetism.

No description available.

Electron spin resonance techniques were applied to study crystals of ZnSe both doped and undoped. Crystals grown by sublimation at 1300ºC were found to have a mixed cubic-hexagonal structure. Annealing these crystals at 1050ºC increased the hexagonal component. This is the opposite of what is usually observed in crystals produced by the flow process. ZnSe crystals produced at 850ºC by the iodine transport method were found to have a cubic structure. No cubic → hexagonal phase transition could be achieved in these crystals. From the experimental data it is concluded that the hexagonal phase in crystals grown at 1300ºC occurred as a result of the presence of an impurity, probably oxygen. Electron spin resonance of Mn was studied in detail in cubic, in twinned and in cubic-hexagonal ZnSe crystals. Comparing the Mn concentration and distribution in crystals produced at 1300ºC and at 850ºC, it was found that the solubility of Mn was larger in crystals grown at 1300ºC. Attempts were made to dope ZnSe with Mn by diffusion, but results indicated that Mn does not diffuse into ZnSe. However, spin resonance results showed that annealing ZnSe: Mn crystals in molten Zn at 850ºC removes some of the Mn impurity. Electron spin resonance investigations were also carried out in ZnSe crystals doped with I, CI, Al, and In. In these crystals spin resonance signals characteristic of mobile donor electron were detected. These signals were all found to be affected by annealing in molten Zn, and furthermore they were all photosensitive. Addition of 0.01% Cr to ZnSe was found to prevent charge transport. Spin resonance of a sintered sample containing Cr indicated a trapping level associated with Cr at 0.51 eV below the conduction band.

No description available.

No description available.

No description available.

No description available.

In recent years there has been much interest in the general problem of point imperfections in solids, particularly in the context of the technologically important semiconducting materials. Quite apart from the possible practical implications there are also challenging academic reasons for pursuing studies of this subject. A number of features characteristic of low symmetry defect centres within an otherwise perfect crystal lattice await more gull understanding by theoreticians. One of the most fundamental problems, and one which has e to be dealt with in a totally satisfactory manner, is that of determining the response of the valence electrons to the presence of localised defects. (We shall treat this many-body problem by resource to the perfect crystal screening function but this is obviously an appropriate means.) In addition it would be valuable to better appreciate the precise numerical significance of the Jahn-Teller effect and other distortion-dependent phenomena such as the Stoke’s shift and various aspects of the electron-phonon interaction. Unfortunately, much of the theory required for a completely details understanding of these topics, and others, has yet to be developed in a satisfactory form. Although some characteristics of defect systems have been accounted for by relatively simple approaches the necessity for large-scale computer studies is increasingly apparent if good numerical agreement is to be obtained. However, even with the aid of powerful computing facilities there are still many difficulties to overcome. A vast amount of experimental information concerning semiconductor properties has been accumulated but much of this has still to be explained on a firm theoretical basis. This situation will almost certainly prevail for some time to come. Meanwhile, we must continually employ theoretical methods and models which may ultimately lead to more complete understanding. The task is not an easy one, however, and even some of the more accepted results are continually in need of revision. Consequently it is not surprising that some of the less well understood properties of semiconductors, particularly those in relation to chemical impurities and intrinsic defects, are still subject to considerable discussion and indeed confusion.