No description available.

No description available.

No description available.

This thesis is primarily concerned with the characterisation of the native defects present in CdTe crystals grown from the vapour phase using two different methods. The principal characterisation techniques used were TEM, SEM and etching/optical microscopy. Crystal defects in bulk CdTe are of great importance since this material is in demand as a substrate for the epitaxial growth of Cd(_x)Hg(_1-x)Te for use in large area infra-red imaging devices. Large single crystals grown using the 'Durham' technique invariably contain first order twin boundaries lying on {111} planes and these are of the 'ortho'-type. Twin boundaries on other planes were characterised using the coincidence site lattice models of twin boundaries in sphalerite which were developed during this work. First order lateral twin boundaries lying on {511}-{111}, {112}-{112}, {001}-{221} and {110}-{114} and second order twin boundaries lying on {111}-{115} {114}-{114} and {22l}-{22l} are discussed. Twinning in CdTe is thought to be due to post-growth stress rather than to a growth phenomenon. The dislocations which are present in crystals grown by the 'Durham’ method are predominantly concentrated into well-formed arrays which are sometimes associated with other crystal defects. These arrays were shown by EBIC studies to be electrically active and the average sub-grain dimension was found to be ~ 150 µm. It is thought that the arrays form as a result of dislocation polygonisation during the latter stages of growth. Precipitates, which were identified as being comprised of Te, are found in the bulk of the crystals although they are more often seen on twin boundaries. The relationship between boundary type and the density of precipitates is discussed. Sub-grain boundaries and precipitates were also examined in a small number of CdTe crystals which had been grown by a modified 'Piper-Polich' technique. The differences between the defects in this material and those in CdTe crystals grown by the 'Durham' method are fully explained in terms of the differences in the crystal growth conditions employed in the two techniques. Networks of dislocations in Cd(_0.95) Zn(_0.0 5) Te crystals grown by the 'Durham' technique were also investigated. A mechanochemical polishing machine which was developed during the course of this work and which is capable of producing high quality chemically polished surfaces is described. Also the use of chemical etchants to determine the crystallographic polarity of {111} oriented surfaces is clarified. Important features of this work include; the development of coincidence site lattice models of first order lateral and second order twin boundaries in the sphalerite structure and the thorough characterisation of these boundaries in as-grown CdTe crystals, and the comparison of the sub-grain boundaries and precipitates present in material grown from the vapour phase by two different techniques.

DLTS, ODLTS and DLOS have been used to characterise the main deep level trapping centres in some II-VI semiconductors; these were single crystal CdS, (ZnCd)S, CdSe, CdTe and ZnS, and polycrystalline CdS films. Undoped, single crystal CdS contained four electron traps as detected by DLTS, at 0.29eV, 0.41eV, 0.61eV and 0.74eV below the conduction band (CB). The first two were observed in all samples and were due to native defects. The two states of highest energy were found only in material that had been annealed in S or Cd vapours. The 0.61ev level could be photoinduced by illumination at photon energies greater than about 1eV. It decayed in the dark with an activation energy of 0.25eV. The 0.61eV and 0.74eV centres were associated with electrically active extended defects (subgrain boundaries Such samples had dislocation densities of about 10(^10) cm(^-2). Copper was found to be a residual impurity in CdS. It produced two deep hole traps resulting from a crystal field splitting of the Cu d(^9) state. They were detected by ODLTS and DLOS and were found at 0.35eV and 1.lev above the valence band (VB).Introduction of the isoelectronic impurity tellurium into CdS induced a hole repulsive centre at 0.21eV above the VB. This is thought to be an inportant radiative recombination centre. The main electron trap in CdS at 0.41eV was found to shift to higher energy with incorporation of Zn. Replacement of 20% of the Cd with Zn shifted the energy to 0.63eV. The level appeared fixed to the VB and had a similar functional dependence on composition as the band gap. The activation energies of the copper centres observed in CdS remained unchanged with incorporation of Zn up to the composition (^Zn)0.45 (^cd)0.55(^s) showed that the crystal field splitting was constant and that these levels were also pinned to the VB. During the fabrication process of the (ZnCd)S/Cu(_2)S solar cell, a deep level was induced at about 1.2eV below the CB. This is thought to be a recombination centre and one of the contributory factors to the reduction observed in the current collection efficiency of these devices. Polycrystalline CdS films were prepared by silk screen printing (SP) and evaporation. The SP films were annealed at various times and temperatures to improve the crystallinity of the layers. At 640C for 1hr, deep states at 0.16eV and 0.48eV were detected. The levels disappeared when annealed at 670C-700C and a new level was observed at 0.13eV. CdS/Cu(_2)S heterojunctions were prepared on the material sintered at 670C; this induced a further trapping level at 1.1eV and one that was poorly resolved. Copper diffused into the CdS during the fabrication of the device so the states associated with copper were detected at 0.35eV and 1.1eV, The evaporated CdS layers showed that the defect signature was sensitive to the type of substrate. Using Ag instead of the usual SnO(_x), deep states were induced at 0.48eV and 0.98eV below the CB. These Ag-associated impurity centres prevent the indiffusion of Cu during the optimising heat treatment of the CdS/Cu(_2)S heterojunction. This maintains the stoichicmetry of the Cu(_2)S layer, thereby, preventing degradation of the devices. CdSe and copper doped CdSe were found to contain several important defect centres: a native sensitising centre (0.64eV from the VB), a class I recombination centre (0.9eV from the CB), a copper impurity centre (0.2eV from the CB) and two native defects (0.16eVand 0.45eV from the CB). n-type CdTe grown by the Piper-Polich technique contained6 electron traps at 0.15eV, 0.21eV, 0.40eV, 0.47eV, 0.53eV and 0.63eV. Their presence was shown to be dependent upon the method of growth of the crystal by comparing with material grown by other techniques. One or more of these states were thought to be due to extended defects or Te precipitates. Low resistivity ZnS contained two deep electron traps at 0.25eV and O.50eV as detected by DLTS. In addition DLOS showed the presence of four further states at 1.25eV, 1.37eV, 1.89eV and 2.19eV below the CB. The first two are thought to be the strong luminescence centres observed by other workers.

No description available.

No description available.

Currently, alignment films for use in liquid crystal displays are produced via a mechanical rubbing process. The dust produced by mechanical rubbing along with problems due to friction and uneven roller pressure lead to defects in the display. Therefore a novel method for aligning polymers films by irradiation with polarised light has been attempted. Anisotropy introduced into the films by selective irradiation affects liquid crystal alignment. The polymers used in this study are poly (vinyl cinnamate), poly (9- anthraceneoate ethyl methacrylate) and poly (p-azidobenzoate ethyl methacrylate). Poly (vinyl cinnamate) is a classical photoresist polymer which undergoes a [2+2] photocycloaddition in the presence of UV light. Poly (9-anthraceneoate ethyl methacrylate) and poly (p-azidobenzoate ethyl methacrylate) are both novel polymers which have the potential to undergo photo-crosslinking reactions. Poly (9-anthraceneoate ethyl methacrylate) contains an anthracene-terminated side chain which dimerises under the influence of UV light introducing anisotropy into the system. Poly (p-azidobenzoate ethyl methacrylate) contains an azido group which when irradiated with polarised light loses nitrogen to yield nitrenes which can combine to form azobenzene species. UV spectroscopy, infrared dichroism studies, birefringence measurements and fabrication of a simple liquid crystal cell show that poly (vinyl cinnamate) and poly (9-anthraceneoate ethyl methacrylate) undergo selective photoreaction. Poly (p- azidobenzoate ethyl methacrylate) is shown to undergo reaction but not to give the desired products.

No description available.

Photonic crystal slabs (PCS) are explored in the context of optofluidic refractive index (RI) sensing for portable, label-free, biosensing applications. The accuracy of RI sensors is limited by noise signals that cause a change in RI that cannot be differentiated from the signal of interest. For this reason self-referencing schemes that provide rejection of common mode signals, and an inherent temperature stabilization approach, are explored. A novel referencing method that allows for frequency shifts to be read out in the transmission power spectrum is proposed and characterized. In terms of improving sensing metrics the relevant characteristics of various PCS architectures are explored numerically. In addition, a novel suspended \emph{air-substrate} device that offers greatly improved sensitivity is proposed and characterized. An experimental measurement near the theoretical detection limit for a PCS is demonstrated. In understanding measurement errors the crossed-polarization effect and its practical limitations are explored numerically.