No description available.

No description available.

Monte Carlo techniques have been applied to the field of electrophotography in order to simulate a variety of xerographic dark discharge mechanisms. The work is divided into two sections, each of which concentrates on a particular discharge mechanism. The first is concerned with the transport of surface carriers that drift through the bulk of a sample under the influence of a local internal field (Surface Injection model). Simulations were performed for both trap-free carrier transport and trap-limited band transport in which the trapping centres were distributed over a range of energies. Results for these simulations were found to be in agreement with both the experimental and the theoretical findings of other workers. The surface injection model was also used to account for the cross-over effect observed in xerographic dark decay experiments performed upon polyethylene photoconductors. It was found that the field dependent partial injection of carriers from the sample surface accounted for such a phenomenon. The second part of this work concentrates on the xerographic dark decay associated with the build up of thermally generated charge in the bulk (Bulk Generation model). A bipolar bulk generation (BBG) model was suggested in which both carrier types generated within the bulk are mobile. The results from the BBG simulations produce a dark decay that occurs within a continuous time regime. Such a feature is found to occur within experimental dark decays within <I>a</I>-Si:H and <I>a</I>-Si<SUB>1-x</SUB>C<SUB>x</SUB>:H photoreceptors. It was also shown that the BBG model possessed the same thickness and initial surface charge dependencies as found experimentally. Therefore, it was suggested that dark discharge within <I>a</I>-Si:H and <I>a</I>-Si<SUB>1-x</SUB>C<SUB>x</SUB>:H occurs due to the BBG mechanism.

The detection and monitoring of toxic and explosive gases is often performed using semiconductor gas sensors. The substrate forms an important part of these sensors and current designs were investigated and tested. Various new designs were developed and thick and thin-film technologies employed to fabricate substrates and complete sensors. Substrateless sensors were also analysed and alterations performed to fashion new devices. A number of ceramic materials were tested for their suitability as semiconductor gas sensor substrates. The adhesion of thin-films and thermal conductivity were found to be the most crucial properties, in addition to those typical of ceramics, such as high temperature stability. Alumina is routinely used in semiconductor gas sensors and many other substrates and its performance was compared with less commonly used materials such as beryllia and aluminium nitride. These materials have a much greater thermal conductivity than alumina, and this was shown to improve sensitivity. A semi-empirical formula was derived to enable the prediction of sensitivity loss of a semiconductor gas sensor fabricated on a substrate with high temperature gradients, compared with one where gradients are minimal. The heaters used to raise the temperature of the substrate are typically made from platinum films. The longevity of thin platinum films depends on the film thickness and substrate surface, but for a given film thickness on a given substrate, additional adhesion layers of various metals were also shown to change the films stability and hence lifetime. Various substrate geometries were investigated to optimise temperature distribution and sensitivity. Predominantly subtle effects were observed, but a significant increase of sensitivity was found with an increased surface area. Electrical circuitry used to control and monitor sensors was summarised and a new substrate developed which could be used in conjunction with switching circuitry, the main advantage being that the fabrication of the substrate was more economical than standard substrate configurations.

In this thesis the numerical simulation of advanced semiconductor devices is considered. In order to simulate devices such as the semiconductor laser or the electron wave diffraction transistor advanced physical models must be included. These models are first derived and then applied to particular devices of interest. Initally the fundamentals of heterostructure device modelling are considered with descriptions of the control region approximation and the Scharfetter-Gummel algorithm for the calculation of the current densities. This model is then developed to cope with degenerate statistics using an additional parameter in the Maxwell-Boltzmann exponential. To simulate optical devices such as semiconductor lasers the optical field must be known and hence the solution to the wave equation is considered. Two methods are used, the effective and weighted index methods, and it is found that the weighted index method has important advantages for wave guides of reduced size. In either method a one dimensional algebraic eigenvalue equation must be solved and a highly efficient method for the solution to this equation is presented. The thermal properties of buried heterostructure lasers are investigated using a coupled approach to the electrical, thermal and optical equation sets. The lasing mode profiles, carrier distributions, threshold currents and temperature characteristics are analysed and good agreement is found with experimental results, including the temperature dependence of the threshold current and the prediction of a break-point temperature. To model quantum effects Schröinger's equation is solved using the transfer matrix technique and this is coupled with Poisson's equation and the continuity equations. Scattering is introduced via Lorentzian broadening and a new method for incorporating a finite capture time is derived. This model is applied to 'coherent electron emitters' and a new device is proposed which provides highly coherent emission in the direction of propagation.

No description available.

The preparation and characterisation of robust, transparent nanocrystalline thin films of anatase TiO<SUB>2</SUB> and their use in photochemical systems for the photodegradation of organics (4-chlorophenol, methylene blue and stearic acid), and the evolution of hydrogen from aqueous solutions of sacrificial electron donors (SEDs) is described. Quantum yields for the photooxidation of 4-chlorophenol in TiO<SUB>2</SUB> are 0.5-1% and these depend upon the total number of photons absorbed and are independent of both the absorbed photon flux and the distribution of charge carriers in the film. In contrast the rate of photoreduction of methylene blue appears to be surface limited as quantum yields for methylene blue reduction fall from 1.7% to 0.5% as film thickness increases. Quantum yields for photooxidation of solid films of stearic acid are 1-4%. In studies of hydrogen evolution from aqueous solution in the presence of SEDs, the quantum yields are low and surface limited unless a Pt or Au cathode is present. The geometry and size (over a ten-fold range) of the Pt cathode is not influential in determining hydrogen evolution quantum yields, nor is the spatial separation of the electrodes. At constant total ion concentration (0.1M), quantum yields are 0.112 (0.007) for pH 2-6 and 0.029 (.003) for pH 8-12. In acidic solution hydrogen evolution occurs at both the TiO<SUB>2</SUB> surface (30%) and the Pt surface (70%) whereas the alkaline solution hydrogen is evolved exclusively from the TiO<SUB>2</SUB> surface. Hydrogen evolution yields are highly dependent upon the nature and concentration of the SED. The maximum quantum yield obtained was 0.27 for 1 M HCl in 50/50 v/v methanol/water. The attempted preparation of visible light absorbing WO<SUB>3</SUB> and dye-sensitised TiO<SUB>2</SUB> films in these photosystems is also described.

In this research, the optoelectronic and structural properties of thin film silicon carbide (SiC) prepared by plasma enhanced chemical vapour deposition and excimer (ArF) laser crystallisation are presented. These materials have been utilised as p-i-n electroluminescent devices, including development of various novel device structures. A wide-ranging series of experiments aimed at optimising the deposition conditions of amorphous and microcrystalline SiC films are described. Dark conductivity, photoconductivity, photoluminescence, optical absorption by Swanepoel's method and constant photocurrent method (CPM), scanning and transmission electron microscopy, infrared spectroscopy, and elastic recoil detection analysis were employed to characterise the films. Absorption spectra and the density of states profile of amorphous silicon carbide as found by CPM are reported. As the carbon content increases, the valence band tail becomes broader. At the same time, the deep defect density of states increases and also becomes broader. The CPM data also verified that the band gap widening is due to the conduction band shifting with increasing carbon content. It is shown that H<SUB>2</SUB> dilution leads to an improvement of electronic properties via a decrease in the density of localised states. A novel method has been developed to prepare highly conductive and wide band gap doped microcrystalline silicon carbide (μc-SiC) by excimer (ArF) laser crystallisation. After crystallisation, this material has Tauc gap of around 2.0 eV and exhibits a dark conductivity as high as 20 (Ωcm)<SUP>-1</SUP>, more than ten orders of magnitude higher than before the laser irradiation. This is shown to be mainly correlated to structural change. The dopant concentration plays a dominant role in the electrical transport properties of μc-SiC, regardless of type of dopant and carbon concentration up to 30 at.%. Laser crystallised μc-SiC can be utilised not only as the carrier injection layer in a-SiC:H based electroluminescent devices, but also as a luminescent layer. EL devices fabricated with μc-SiC as a hole injector possess the highest electroluminescent intensity, the most stable emission and the longest operating life-time among all the investigated device structures. The electroluminescence from these devices is possibly related to the formation of some form of porous SiC by laser crystallisation.

The band discontinuities or band offsets in semiconductor heterojunctions are crucial parameters that affect the transport and optical properties of the heterojunction devices. The ability to control and modify the band offsets to desired values would enable the performance of solid devices to be significantly improved. The effect of a one monolayer intralayer of an ionic material (ZnSe) has been investigated for band offset engineering at Ge-Si junctions. Soft x-ray photoelectron spectroscopy (SXPS) was performed to probe the Ge(111)/Si and Ge(111)/ZnSe/Si interface formation. The evolution of the valence band edges and the movements of core levels were monitored as the interfaces were formed under ultra-high vacuum conditions. It was found that the ZnSe intralayer dramatically modified the valence band offset of the Ge(111)/Si heterojunction. The valence band offset was increased by ˜0.57 eV due to the presence of the ZnSe intralayer, as compared to a negligible valence band offset for the intralayer-free junction. The effect of band bending on the interpretation of the valence band offset was found to be negligible for the interfaces studied. The dipole effect was bound to be destroyed at higher coverages (>10 Å of Si) due to the out-diffusion of Se. X-ray photoemission was used to examine the effect of placing a ZnSe intralayer at the Si(111)/Si homojunction. An effective change of ˜0.2 eV in band profile was measured. The initial growth of ZnSe layer on Si(111) substrate was investigated with STM/STS technique. STM images indicated the growth of ZnSe islands on the substrate and STS measurements illustrated the effect of the ZnSe clusters on the electrical properties of the substrate. The ZnSe intralayer-induced modifications of the valence band offsets for these interfaces are interpreted in terms of the charge transfer at the interfaces. It is believed that the orientation and structural geometry of the ZnSe molecule at the interface are of vital importance in determining the magnitude and direction of the dipole moment and hence the band offset. Strain and stability of chemical bonds are also addressed.

A detailed numerical model of the electronic properties of hydrogenated amorphous silicon has been developed and shown to be a useful tool for the analysis of the performance and optimization of the design of solar cells. The method of simulation involves solving Poissons's equation, and the electron and hole continuity equations, in conjunction with the transport equations for the electrons and holes. From the solutions of these equations we obtained the electrostatic potential, the electron and hole concentrations and the current densities. A set of realistic material parameters has been used. We have modelled the density of states to consist of two exponential band tails and the dangling bonds. Recombination in both the band tails and the dangling bonds has been taken into consideration in the model. We investigated the effect of the cell performance on varying dangling bond densities (10<SUP>16</SUP>cm<SUP>-3</SUP>-10<SUP>17</SUP>cm<SUP>-3</SUP>) for various cell thicknesses of p-i-n hydrogenated amorphous silicon solar cells, for incident blue and red light. Our results agree well with experiments for solar cells in the undegraded state. However for the degraded state the fill factors appear to be higher than the experimental values. This might be because we have only assumed a single level dangling bond density in our model. It is suggested that future work might undertake the incorporation of the spatial dependence of the dangling bond density in the model.