An experimental study of the quenching effects on the galvanomagnetic and thermomagnetic properties of bismuth single crystals is presented. Measurements made include components of the magneto resistivity and magnetothermoelectric power tensors at low and intermediate magnetic fields for samples in both the annealed and quenched conditions, at temperatures between 77 K and room temperature. Marked effects of quenching on each set of tensor components have been observed. After quenching the anisotropy ratio p(_11)/ p(_33) of the zero field resistivity inverts from 0.9446 to 1.030 and the thermomagnetic data show a general reduction in magnetic field dependence. The quenching effects anneal out below room temperature. To allow a detailed quantitative investigation, a formalism has been developed by which the thermomagnetic data and the predictions of both the phenomenological theory and o f the band transport theory can be compared on the same basis as the galvanomagnetic effects. The expressions provide for the first time an analytical explanation for the Umkehr effect and lead to its prediction. The occurrence of Umkehr effect and sign reversal in the thermomagnetic power of bismuth directly follows from the nature o f the Fermi surface. Measurements a t low and intermediate magnetic fields have been analysed in terms of a two-band multi - valley Fermi surface model, using a least-means-square procedure. Reasonable agreement obtains between the model parameters computed from different sources. Magneto resistivity data at low and intermediate fields evidence consistent quenching-induced changes on the model parameters. Quenching results in an increase in carrier densities from 4.4 10(^23)m(^-3) in annealed state to 4.6 10(^23)m(^-3) and 5.3 10(^23)m(^-3) for electrons and holes respectively. Carrier mobilities in the xy-plane are reduced but those along the trigonal (z) axis are increased slightly. The energy separation between the band edges increases markedly. Quenched-in defects have predominantly accepter-like character.

An evaluation is presented of some of the photoconductive and luminescent properties of CdS crystals grown by sublimation in sealed tubes under excess pressures of cadmium and sulphur. It is demonstrated that the use of impure starting material, for crystal growth, can lead to anomalous results. Eight discrete sets of electron traps were found in the present samples using the thermally stimulated current technique. The trapping parameters and the variation of trap density with growth conditions are given in the text. It is suggested that one of these traps, which had a thermal activation energy of 0.15 eV, is associated with cadmium interstitials. The densities of two traps, with activation energies of 0.53 eV and 0.60 eV, changed in magnitude when differing conditions of illumination were used prior to measurement. Such changes are shown to be commensurate with traps of a -photochemical nature. Three and four probe measurements are also described, which were made to determine how the crystal surface and the indium contacts modify the thermally stimulated current. Considerable differences were found in the spectral response and infrared quenching of photocurrent measured on cadmium and sulphur rich samples. A disparity was found between the thermal and optical separations of the sensitising Class 11 centres from the valence band, which indicates that these centres are singly negatively charged when occupied by electrons. The structured luminescence, in the region 1.6 µm to 2.0 µm, has been attributed to radiative recombination of electrons from the sensitising Class 11 centres with holes in the upper branches of the valence band.

Hall effect and photo Hall measurements have been made on a number of crystals. These crystals were grown (in this laboratory) using a modified method of the technique of growth by vacuum sublimation. The object of the photo Hall measurements was to determine some of the parameters of the imperfection centres with energy levels in the forbidden gap in photosensitive cadmium sulphide crystals. Two crystals (crystals 78 and 79) with dark conductivities less than 10(^-7) mho cm(^-1)were chosen for the investigation. Their conductivities could be increased to 10(^-2) mho cm(^-1) by optical excitation with an intensity of 3200 ft-c. The photo Hall effect was investigated at different temperatures. Measurements were made to determine the variation in the value of Hall mobility as a function of the location of the electron Fermi level which was achieved by changing the intensity of photoexcitation. From the plots of 1/µ versus E(_fn) (at different temperatures) four electron trapping levels with energy depths of 0.12, 0.l6, 0-22 and 0.33 eV and six electron trapping levels with energy depths of O.O98, 0.13, 0'19, 0.23, 0.33 and 0.42 eV below the conduction band were obtained for crystals 78 and 79 respectively. The photoHall data were also used to determine the charge state and the scattering cross-sections of the imperfection centres. The experimental values of the scattering cross-sections were of the -11 2order of 10cm. From a study of the temperature dependence of concentration of photoexcited carriers, the height of the energy levels of the sensitizing centres above the valence band was found to be 1.04 eV. The Hall coefficient was also measured for a number of semiconducting samples between 14 K and 300 K. The donor ionization energies and the donor and acceptor concentrations were determined from the variation of carrier concentration with temperature above 30 K. Below 30 K, the variation of carrier concentration with temperature was found to be dominated by an impurity band conduction mechanism (non-metallic type). Polar optical mode scattering was the dominant intrinsic scattering mechanism at the higher temperatures. The experimental mobility data could be fitted to theoretically computed values of µ assuming that polar optical mode, piezoelectric and ionized impurity scattering processes were operative. The effective mass was used as an adjustable parameter. The best fit was obtained with m*(_e) = 0.19 m.

This thesis investigates conduction mechanisms of covalently and non covalently functionalised single wall carbon nanotube (SWCNT) networks. Unlike previous strategies where diamines were used, a novel route to covalently bridge SWCNTs by organic molecular linkers is proposed. The bridging relies on using modified Sonogashira and Ullmann couplings, which have the advantage of using spectroscopic evidence to ascertain the success of the bridging. Platinum-enriched SWCNTs were produced by coordinating Pt to pyridine ligands grafted on SWCNTs. Networks of covalently bridged SWCNTs, Pt-enriched SWCNTs and their SWCNT precursors were fabricated by vacuum filtration. In addition to these networks, networks of non covalent ly functionalised SWCNTs were built up using layer-bylayer (LbL) deposition. This second approach required the wrapping up of SWCNTs by ionic surfactants to exploit their electrostatic interactions. Electrical properties, such as current- voltage and the current dependence on temperature and electrode separation are discussed for both filtered SWCNT and SWCNT LbL networks. Combined analyses of these characteristics were carried out to identify dominant conduction mechanisms. In this study, a modified quantum tunnelling model was proposed to best describe the in-plane electrical behaviour of the filtered SWCNT networks. As for SWCNT LbL networks, the in-plane conduction was shown to be governed by the Poole-Frenkel mechanisms while direct tunnelling dominates the out-of-plane conduction. Furthermore, the charge storage capacity of cut-SWCNT LbL networks integrated into metal- insulator-semiconductor devices are discussed in view of organic memory device applications.

This thesis describes an investigation of the longitudinal magnetostriction of a single crystal of nickel over the temperature range - 180ºC. to 360ºC. The most important measurement to be made at any temperature was of the saturation magnetostriction, but there is considerable interest in measuring the strain at different stages in the magnetization process, and this has been done at several temperatures. Measurements have been confined to two ellipsoidal specimens cut from the same crystal of nickel with their major axes along cube edge and cube diagonal crystallographic directions. Part 1 of the thesis describes the background to the work. The development of the theory of ferromagnetism is outlined, and the subject of magnetostriction introduced. Previous magnetostriction measurements, particularly on single crystals of nickel, are discussed. Part 2. ' surveys the apparatus used by previous workers for measuring magnetostriction over a range of temperatures, and discusses the design most suitable for the present investigation. Sections 2.3 to 2.7 describe in detail the apparatus used for the high temperature measurements, and section 2.10 the modifications which were necessary for use below room temperature. The preparation of the specimens from the single crystal is described in section 2.8. In Part 3 the results are presented, and their significance is discussed in relation to theories of the magnetization process.

For some time past there has been a need for a more positive identification of the nature of the defect centres which give rise to energy levels in the forbidden gap in cadmium sulphide. Electron spin resonance (e.s.r.) techniques have proved very useful in this type of investigation in other materials. Since little similar work on CdS has previously been reported, the purpose of the research described in this thesis, has been to examine the usefulness of the technique in studying CdS. Initially it was necessary to construct a sensitive x-band microwave spectrometer, which operates at temperatures down to 1.5ºK and has provision for continual. illumination of the samples, Electron spin resonance absorption signals have been detected in undoped single crystals of GdS which can be attributed to four different defect centres. The occurrence of the resonance signals can be correlated with the resistivities and edge luminescence spectra of the samples. This indicates that the centres responsible for the resonance absorption are important in determining the electrical and optical properties of CdS and are those which this work was initiated to study. Tentative models for the various defects have been proposed. The most important feature of the work reported in this thesis is the isolation of an e.s.r. signal which is thought to be associated with the class 2 centres which provide the photo- conductive sensitisation and possibly act as I.R. emission centres in CdS. The model proposed for such centres is of compensated acceptor complexes with levels approximately 0.7eV above the valence band. A centre consists of four cadmium vacancies in nearest neighbour association. The models for the other three centres have not been discussed as fully as that mentioned above because of the lack of experimental data. However it is evident that e.s.r. techniques are a very valuable tool for investigating the nature of defect centres in CdS and that continuation of the work should prove invaluable in providing an unabiguous identification of the atomic structure of the defect complexes.

No description available.

The spin-lattice relaxation times,T(_1), of Cr(^3+)ions in a range of samples of artificial ruby have been measured at 4.2ºK, in the direct ( one phonon ) relaxation process regime, at frequencies of 9.3, 16, 22, 35.5, and 71 GHz, corresponding to magnetic fields, H, up to 25 kG. These results have been compared with the predictions of the Van Vleck theory of paramagnetic relaxation, that T(_1) should be proportional to H(^-2) for a non-Kramer's ion and H(^-4) for an isolated Kramer's doublet. The observed dependencies, between -0.3 and -1, at low magnetic fields are not in agreement with theory, but at high fields (between fields corresponding to 55.5 and 71 GHz) the average dependency for ΔM=1 transitions at polar angle 90 is T(_1) -H(^-2.55) indicating that the energy levels are tending to behave as isolated Kramer's doublets, due to their increasing separation in energy and the decreasing degree of mixing of pure spin states. At low fields the high degree of mixing of states and the masking effect of the zero field splitting on the field dependent energy separation of the energy levels appear to cause the low degree of dependence of T(_1) on H. In contrast to other published collations of T(_1) measurements performed en different samples at various frequencies, all of these results have been obtained under the same experimental conditions from the same samples, frequency of observation (magnetic field) being the only variable. In addition to the variation of T(_1) with H, a linear variation of T(_1) with mean lattice strain, ε , derived from observations of the line width broadening in the samples used, has been observed. The design and construction of the spectrometers used for these measurements is described in detail

Polarised spectroscopic techniques were used to investigate the underlying physics of steady-state and ultrafast fluorescence depolarisation in conjugated polymers. Depolarisation is due to fluorescence anisotropy: the angular difference between the absorption and emission transition dipole moments of a molecule. Polarised spectroscopy results from a polymer with a flexible backbone, poly (9,9-di(ethylhexyl)fluorene) were compared with those from two rigid backbone polymers: methyl-substituted ladder-type poly (para-phenylene) and the newly synthesised naphthylene ladder-type polymer (2,6-NLP). This revealed that there is an intrinsic anisotropy directly associated with the molecular backbone. This work is the first reported on 2,6-NLP. Fluorescence anisotropy was shown to be dependent upon the conjugation length; the transition dipole moments show larger angles for short lengths, tending to a minimum as the length increases. For rigid-rod polymers, this behaviour is replicated at each vibronic position. In the flexible polymer, planarisation of the backbone elongates the excited state over more conjugated bonds, changing the angle between the transition dipole moments, whereas in rigid-rod polymers, such elongation can only be electronic. Linear dichroism results obtained for all the polymers has shown the angle between the absorption transition dipole moment and the molecular backbone is large and that the emission transition dipole moment is aligned with the backbone. “Off-chain” to “on-chain” transition dipole moments arise from transitions from localised to delocalised states suggesting that the excited state in conjugated polymers is delocalised. Time-dependent measurements show that the main fluorescence depolarisation mechanism occurs in under 5 ps for both flexible and rigid polymers. The ultrafast timescale and the similarity of the two systems requires the process to be electronic in origin and not linked to a physical deformation. This work proposes that ultrafast fluorescence depolarisation is a result of the delocalisation of the electronic state as the conjugation length extends over more of the polymer.

Polycrystalline and single crystal specimens of silicon iron were strained up to the yield point by the application of external stresses. A tending technique was found to be the most convenient method for doing this. Domain patterns were observed by means of the Bitter technique. Tensional stresses were applied along the [110] and 001] directions of (110) surfaced specimens and the result ing structures described in detail. This necessitated an analysis of the effect of stress on Bloch Wall energies for various wall orientations. Domain patterns were also examined on (100) surfaced specimens under the effect of compressive forces in the [110] and [100] directions. In the first case the maze patterns so produced were compared with theoretical considerations while in the second example the nucleation and growth of transverse domains were examined. A new model of fir tree structures, proposed by Spacek, was evaluated using a different expression for the magnetostatic energy and this was then applied to measurements on the effect of stress on fir tree closure domains. The theory did not compare very favourably with experiment. The work of Chikazumi and Suzulci, in which large stresses were applied to a (l00) surfaced specimen by scratching the surface, was extended to (110) surfaces.