A very simple microscopic technique has been developed for the observation of uni-molecular steps. This technique utilises the diffraction effect at the edge of such steps, and steps of 4.6 A and 7.9 A have been observed on beryl crystals. This effect is discussed and a theoretical account given of methods of microscopic examination likely to result in the greatest sensitivity. It is concluded that steps of 1 A , or less, should be visible when reflexion phase contrast equipment is used. A review is given of past work on crystal growth in which it is shown that only the dislocation theory of growth is capable of satisfactorily explaining much of the experimental data. The microscopic diffraction technique and multiple beam interferometric techniques are applied to the study of the topography of faces of beryl and quartz crystals. As the microscopic technique enables one to observe dislocations directly, experimental proof is given that these crystals have grown by a dislocation mechanism, and that the dislocation theory of growth must therefore be generally applicable. This theory is shown to be correct in its many detailed predictions and quantitative information is obtained about the conditions in which some of these crystals have grown. The activity of dislocation groups, the cross-linking of layers, the "bunching" of layers, and "rod-formation" on beryl, are discussed. The treatment by the dislocation theory of dissolution as the simple reverse of the growth process is shown to be correct. Information is obtained on the density of dislocations and they are shown to occur frequently on transition surfaces and limited slip-zones. Direct visual proof is given that the dislocation theory of slip is correct. The topography of a cleavage face of topaz is briefly discussed.

Growth patterns upon the (001) faces of crystals of stearic acid, CH3 (CH2)16 COOH, and behenic acid CH3 (CH2)20 COOH, are described. The crystals were grown from solution in chloroform at temporatures~268°K, and studied by optical, interferometric and electron microscopic techniques. The crystals were found to be imperfect and to have grown by a spiral mechanism, in accordance with the theory proposed by Burton, Cabrera and Frank. Two independent methods were used to determine the height of the spiral steps. Inter-ferometric measurements showed that the mean step-height was an integral multiple of c sin beta. There were four exceptions, in which the step-height was 1/2 N c sin beta where N was an odd integer. Electron microscopic measurements on crystals of behenic acid revealed mainly bimolecular growth steps (c sinbeta). Six examples of trimolecular steps 9delta/2 c sin beta) were also found. These results showed that Frank's hypothesis that the normal component of the Burgers vector of a dislocation is equal to an interplanar spacing of the lattice was not always valid. An approximate calculation showed that many of these crystals had grown at supersaturations as low as 2%. Evidence of unit slip and internal slip was given .It was shown that internal slip did not perfectly replicate the underlying surface structure; internal slip occurred through thicknesses as much as 1000 - 1500 A°. Patterns due to internal slip also suggested that the dislocation lines were not necessarily parallel to the c-axis or perpendicular to (001). The shape of the growth spirals was discussed. It was found to depend upon the crystal structure, the temperature at growth, and the inclination of the molecules to the basal plane. It was shown that layer disorders occur in crystals of the fatty acids. These disorders were repeated by spiral growth, leading to polytypism or polysynthetic twinning, and characteristic interlaced growth patterns. The component layers were related by rotations through an angle phi = 180°, 106°, or 74°. When phi = 106° or 74° the layers were in twinned orientation. Occasionally random values of phi occurred. This, together with observations of helicoidal crystals suggested that the dislocations were due to non-uniform distributions of stresses leading to buckle and shear of the thin crystal plates, as proposed by Frank.

Part I of this thesis deals with a brief historical review on diamond together with the studies made on crystal growth with the existing information on diamond. Part II deals with the experimental techniques employed for the present investigation including an interference technique whereby, it has been shown for the first time, that very sharp multiple-beam interference fringes can be produced with a high power 3 mm. lens of numerical aperture 0.95. The optical conditions necessary to secure high definition in the fringes are discussed together with the arrangement used which can resolve a lateral extension about 0.7 and a depth of 50 A0. Volume elements as small as 5 x 10 cc. can be resolved. Part III deals with the optical studies of growth features on the octahedron faces of some diamonds. Some features of particular interest are described namely :-1 - A number of anomalies in trigons. These are discussed in detail.2 - An occasional mode of growth sheets is shown to operate on the octahedral faces of diamond which leads to the formation of six-sided growth features containing alternate angles of approximately 90[degrees] and 150[degrees]. It is shown that the edges of these are effectively parallel to the directions 431.These may possibly arise through intersections of (221) with (111)3 - A remarkable case of multiple linear processes has been observed on five faces of a well formed good octahedron. This has been well interpreted as due to crystallographic slip in diamond. Part IV deals with the microstructure on the dodecahedral faces of some diamonds. Some new surface structures are described. A mosaic structure exists. At least three main different structures for the (110) faces can possibly arise during the growth of diamond namely :-1 - Striated surfaces either smooth or coarse.2 - A network structure.3 - A parallelogram structure related to 2.These structures are described and discussed in detail.

Surface structure of the spiral is studied employing multiple beam interferometric techniques. The spiral has a slope perpendicular to a radius vector equal to h/2piA per radian, where h is the step height of the spiral. The surface of the spiral is highly curved; the curvature decreasing with distances away from the dislocation. The angle of inclination of the growth surface with the underlying (0001) face of SiC crystal is determined at various sections and is found to be greater than the theoretical radial slope Tan-1 h/a where d is the distance between tv/o consecutive loops of the spiral. The hollow dislocations predicted by Prank are associated with surface distortions of the sinking in type. Sometimes the large stress accompanying a giant dislocation produces a percussion figure on the crystal surface. Employing interferometric, X-ray diffraction and electron-microscopic techniques, it is shown that multi-molecular steps do exist in SiC and that they are sharp, single and vertical. Analysing the structure of multi-molecular steps, it is shown that the average distance between two kinks along the step edge should be about 1000A in SiC. It is established that at least in SiC, the step height is determined by the Burgers vector of the dislocation. In addition to the movement of unit dislocation causing unit slip, evidence for the movement of giant dislocations has been obtained for the first time. The experimental evidence for the operation of a modified cone source as an alternative to the Frank-Read source which is also common in SiC is presented. In this type of source, the slip plane of the sweeping dislocation rotates as the dislocation spirals. This surface study has also revealed the nature of dislocations present in SiC. In addition to the pure screw dislocations with Burgers vector perpendicular to the crystal surface, inclined dislocations which can he resolved into edge and screw components have been observed. In the light of these inclined dislocations, a modification to Frank's theory of polytypism is suggested. This explains the production of almost all the rhombohedral polytypes in SiC without invoking periodic break in the stacking sequence. Dislocations having screw component usually equal to 15 A have been observed on the step lines of the growth spiral. Deep channels terminating on these dislocations have been observed along the step lines. These dislocations cause among other things, fault lines, dissociation and grouping of steps and disorder in rhombohedral class of crystals. This latter has been confirmed by X-rays. Optical and interferometric study of the dendrites of germanium and tin grown from undercooled melt and by the action of Zn on SnCl2 solution respectively has been carried out. In the case of tin, tv/o types of crystals, viz: (1) flat dendrites with perpendicular branching and (2) needles without branching have been found to occur. It has been possible both in the case of Ge and Sn to correlate the surface features with their space lattice. Since most of the observations have been made employing the multiple beam interferometric techniques, the contouring efficiency of thin films used in these techniques is studied. It is shown for the first time that silver films accurately contour the underlying topography. Dielectric multilayers can be used for qualitative work but for quantitative measurements they do not seem to satisfy contouring requirements. The thesis commences with a brief account of the theories of crystal growth. The previous observations are reviewed in brief in the light of the present investigation.

Part I of the thesis commences with a general picture of the various concepts of hardness and the relationship of indentation hardness to other physical properties. Types of measurement of hardness are reviewed with special reference to static indentation tests. The optical techniques and the apparatus made and used are described, including the interferometric techniques employed. The following aspects of hardness of silicon carbide are studied: (1) The variation of hardness with load, the relation between 'recovered' and 'unrecovered' hardness and the surface distortions around the indentations. (2) The variation of hardness with direction: it is found that the [1120] and [0110] are respectively the directions of maximum and minimum hardness. An attempt is made to explain the results in terms of the shear stress on the slip plane. (3) Ring cracks formed by steady pressure made by a diamond ball: the distortions round the cracks and the profiles of sections are studied by multiple-beam interferometry and phase-contrast microscopy. Hertz's classical equation is applied to compare the results. The micro-slip lines observed on a rare trigonal pyramid face of a silicon carbide crystal due to diamond ball impacts are described. A description is given of the analysis of silicon carbide crystals by X-rays. Part II of the thesis gives a description of various types of micro-features observed on transparent crystals of silicon carbide. The features are studied by applying the techniques of multiple-beam interferometry and phase-contrast microscopy. The special point about these features is that they are depressions. An attempt is made to explain their existence on the basis of a theory postulated by Tolansky and the dislocation theory of evaporation of crystals by Cabrera and Levine. A description is given of spiral-like features observed at the edge of some silicon carbide crystals. An appendix to the thesis gives a brief account of interferometric studies made on sawn surfaces of diamond.

In this work we present spectroscopic studies of the exciton dynamics in colloidal spherical cadmium sulfide CdS nanocrystals (NC) in the vicinity of a single indium gallium nitride InGaN quantum well (QW) in dependence of temperature. QWs of the alloy material InGaN exhibit a dependence of the exciton dimensionality on the thermal energy available. It was demonstrated that this dependence in uences the rate for uorescence resonant energy transfer from the QW to a layer of CdS-NC deposited on top of its capping layer. Investigations of different capping layer thicknesses demonstrated the dependence of the exciton dimensionality on the disorder potential of the QW. Furthermore, spectroscopic measurements of elongated asymmetric cadmium selenide/- cadmium sulfide CdSe/CdS nanorods (NR) under the application of external magnetic and electric fields are discussed. Asymmetric CdSe/CdS-NR represent a special case of elongated NR as the analytical treatment of spherical NC can be combined with numerical methods of calculating the electron and hole energies and wave functions. The results for the excitonic fine structure splitting in these nanomaterials is used to explain the dependence of the exciton dynamics under an external magnetic field. For the first time, a separate measurement of the Zeeman splitting and the magnetic field induced spin admixture in colloidal NR was performed. Electric field mediated carrier separation in asymmetric CdSe/CdS-NR is measured in time resolved luminescence quenching experiments. Retrieval of stored excitations is demonstrated employing a synchronised ultrafast voltage pulse detection scheme.

The concept of electron crystallography is by no means new, although the majority of crystallographers will never have contemplated trying the technique, to solve a structural problem. Electron crystallography is fraught with problems from sample preparation, and data collection to structure solution, all have unique and challenging problems when being used in the context of electron crystallography. New techniques and methodologies for structure solution using electron crystallographic data are presented with the aim of making structure solution a more routine part of the procedure for electron crystallography. The problems of phase extension into the missing cone region, structure solution and refinement of the structures are all dealt with here.

A clear and predictive understanding of the propensity for crystallisation of one polymorph over another is lacking, and in this regard glycine is a model system due to difficulties in crystallisation of the thermodynamically stable gamma polymorph. The preferential crystallisation of gamma-glycine in the presence of micellar CTAB (Cetyltrimethylammonium bromide) as opposed to the alpha form commonly crystallised from pure solution was observed. A rationale for this result was sought through the observation of the nucleation and growth kinetics of the alpha and gamma polymorphs of glycine (and DL-alanine) using in situ microscopy, the measurement of induction times and following the solution mediated phase transformation of alpha-glycine. These observations help explain the dominant crystal form produced in a number of solutions. The nucleation and growth rates of alpha-glycine were shown to be orders of magnitude greater than those of gamma-glycine in pure solution. Also, the addition of a cationic surfactant (such as CTAB) or modification of the solution pH were shown to dramatically accelerate the nucleation and growth of polar gamma-glycine and DL-alanine, a rarely reported phenomenon. In addition, the growing (00-1) faces of gamma-glycine and DL-alanine, at which growth was accelerated, were shown to be macroscopically rough, indicating a growth mechanism dominated by nucleation rather than the growth of layers. The most likely cause of the inhibited kinetics of gamma-glycine and DL-alanine is water bound electrostatically at the negatively charged (00-1) faces, while the growth acceleration inferred by the additives is related to their ability to release water from these surfaces. Other mechanisms which may play a role include the adsorption of adventitious impurities, strong electrostatic repulsion between like-charged carboxylate groups at the (00-1) surface resulting in structural disorder, and the effect of surface energy on the rate of surface nucleation. This research provides an important example of nature’s complexity in selecting crystal form in polymorphic systems, gives further insight into the causes of the asymmetric growth of polar crystal structures, and introduces the possibility that the crystallisation kinetics of ‘difficult’ slow growing compounds may sometimes be modified through the use of additives.

Many marine molluscs produce complex shells of calcium carbonate. These shells are formed under strict biological control to provide a range of functions to ensure the survival of the living organism. These inspiring biomineral structures can also provide an archive of environmental change via proxies such as δ18O and δ13C within the shell carbonate. However, the intimate relationship between the biological and environmental controls influencing biomineral production can often obscure the environmental signal, making it difficult to interpret environmental information from shell proxies. Understanding the design of biomineral structures will further our knowledge of biomineralisation as a whole, while understanding the controls that influence shell production will ensure that shell proxies applied to palaeoenvironmental studies are accurate. Oysters are sessile bivalve molluscs that have evolved since the Triassic and expanded to occupy a range of habitats with almost global distribution, providing an example of a highly successful biomineral system. This study investigates the ultrastructure, crystallography and stable isotope composition of the Pacific oyster, Crassostrea gigas from estuarine and marine environments. The method by which oysters adhere to hard substrates is also investigated. Both valves of the oyster shell are composed predominantly of low Mg calcite in three forms; an outer prismatic region, an inner foliated structure and chalk lenses which appear sporadically throughout the valves. Aragonite is restricted to the myostracum and parts of the hinge structure. Crystallographic analysis shows that, despite variations in structural morphology, the superimposed layers of the oyster shell maintain a single crystallographic orientation with the crystallographic c-axis orientated perpendicular to the outer shell surface. Varying crystal morphology, while maintaining crystallographic unity, may be a deliberate design to provide the oyster shell with both strength and flexibility. In general, there is no difference in the overall ultrastructure or microstructural arrangement of estuarine and marine oysters. Estuarine oysters contain significantly more chalk than marine equivalents suggesting that the appearance of chalk lenses is, to some degree, triggered by an external environmental stimulus. Stable carbon and oxygen isotope analysis of folia and chalk provides further insights into the appearance of chalk in the oyster shell structure. There is no significant difference in the isotope composition of folia or chalky layers, although patterns of δ13C and δ18O of folia and chalk reveal key differences between the two structures. Folia display a narrower range of δ18O values compared with chalk and exhibit significant interspecimen variation with respect to δ13C. Interspecimen variation, with respect to either δ18O or δ13C is absent in chalk samples. These patterns suggest that secretion of folia requires a more specific environmental stimulus and a greater input of metabolic carbon than chalk. Chalk is apparently secreted across a greater range of environmental conditions, with less metabolic regulation. Deviation from optimal environmental conditions, for example during periods of reduced salinity and/or cold or warm temperatures, may reduce metabolism causing the oyster to deliberately default from folia to chalk secretion. In general, folia and chalk in both estuarine and marine oysters is secreted in oxygen isotope equilibrium with the ambient environment. Another aspect of the oyster biomineral system is their ability to adhere tightly, and usually permanently, to a range of hard substrates by cementation of the left valve. Investigation of the contact zone between oyster shells and biological (other oyster shells) and inorganic (rock) substrates reveals the influence of both biogenic and non-biogenic processes in oyster cementation. Original adhesion is brought about by secretion of an organic adhesive which acts as a nucleating surface onto which crystals precipitate which have random orientation and are composed of high Mg calcite. The lack of orientation and elevated Mg content suggests that these crystals are nucleating outwith the biological control experienced by the shell biomineralisation process and are formed from inorganic precipitation from seawater. It is proposed that oysters do not control, or secrete, the crystalline cement but instead they secrete an organic film onto which crystals precipitate from seawater. Oysters thus provide excellent examples of both biologically induced and biologically controlled mineralisation.

In the past sixty years, X-ray, neutron and electron dffraction have emerged as the structural techniques of choice in the solid state. However, despite many advances in theory and instrumentation, these diffraction methods are still reliant on a number of assumptions. Chief amongst these is that the atoms in the crystal vibrate in a harmonic fashion. This thesis is concerned with understanding the effects of anharmonic motion on crystal structure determination and developing new ways of moving beyond the harmonic approximation used in crystallography. A method has been developed, using molecular dynamics simulations, to correct experimental structures to equilibrium structures. This has been applied to the crystal structures of phase-I deutero-ammonia, deutero-nitromethane and benzophenone. Path-integral molecular dynamics simulations have been used to obtain meaningful comparison with experimental data collected at low temperatures. The simulations also offer information on the probability density functions that describe thermal motion in solids. Using data from simulations of nitromethane and other compounds it has been demonstrated that the molecular dynamics-derived data can be used to assess and develop new functions for modelling thermal motion in crystal structure refinements. Finally, similar molecular dynamics techniques have been applied to determine the equilibrium structures of some polyhedral oligomeric silsesquioxanes in the gas phase. Some members of this class of compounds feature such strong anharmonic motion that refinement of the structures using gas electron diffraction is impossible without taking into account the effects of the anharmonicity.