This thesis is concerned with the study of the magnetophonon effect in a wide range of semiconductors. The five materials investigated are n-InP, n-CdSe, n-Ge, p-Ge and p-GaAs. The effect which was first predicted in a theoretical paper by Gurevich and Firaov (1951) arises from the resonant coupling of optical phonons with conduction electrons in an applied magnetic field. Extrema occur in the magnetoresistance when the phonon energy approaches an integral number of Landau level spacings. The resonance condition is Nω_c = ^NeB⁄_m* = ω_e ; N = 1,2,3 ... (1) where ω_e is the phonon frequency, ω_c the cyclotron frequency, m* the carrier effective mass and B the applied magnetic field. There are three main applications of the magnetophonon experiments to the study of the properties of conduction electrons. Firstly, equation (1) may be used to determine a precise value of the band edge effective mass provided that small corrections for band non-parabolicity and polaron effects are taken into account, secondly, the data are a test of recent theories concerning the problem of electrical conduction in an applied magnetic field. Thirdly, the magnetophonon structure which appears in the non-ohmic magnetoresistance at low temperatures can reveal new mechanisms for the energy relaxation of warm electrons. The first application of magnetophonon resonance is of particular relevance to the study of n-InP and n-CdSe since previously the conduction band effective masses had not been known precisely. In n-InP (Eaves et al., 1971) up to eleven extrema have been observed in the transverse orientation over the temperature range from 77 K to 300 K. The low frequency cyclotron effective mass deduced from the field positions of the peaks is 0.082 m₀ ± 0.001 m₀ at 77 K and 0.078 m₀ ± 0.001 m₀ at 300 K. The dependence of the peak amplitudes in InP on magnetic field and ionised impurity concentration are in agreement with the theoretical treatment of the transverse magnetophonon effect by Barker (1972), The magnetophonon structure in n-CdSe (Eaves et al., 1972) is considerably weaker in amplitude and can only be observed in a limited temperature range around 80 K. After corrections are applied for band non-parabolicity and the polaron effect, the low frequency cyclotron masses are deduced to be 0.122 m₀ for B ⊥ c-axis and 0.127 m₀ for B || c-axis. The relatively polar nature of both InP and CdSe provides a test of two recent theoretical studies of the magnetophonon effect. First, a comparison of the value of m* derived directly from equation (1) with that obtained from recent cyclotron resonance experiments shows that the optical polaron correction factor by which m* must be multiplied to obtain the low frequency cyclotron mass is (1 + 0.73 α/3)^-1, in excellent agreement with the predictions of Meare et al., (1968) and the theoretical estimate by Palmer (1970). Second, the displacement of the magnetic field positions of the minima in the longitudinal magnetoresistance ρ_zz away from the resonance fields given by equation (1) is in qualitative agreement with the damping theory of Barker (1972a). The magnetophonon structure which has been observed in n- and p-Ge (Eaves et al., 1970) is the most extensive yet reported. In both materials the anisotropy of the peaks has been studied for orientations of B in the (110) crystal plane and accurate values are deduced for the cyclotron effective masses of both the electrons and holes over the temperature range from 55 K to 260 K. In n-Ge, the structure can be interpreted solely in terms of intravalley scattering by optical phonons at the centre of the Brillouin zone. No peaks can be identified as arising from intervalley transitions and such processes are estimated to be at least an order of magnitude weaker than intravalley scattering processes. This result is in opposition to that of Sokolov and Tsidil'kovskii (1967) who attributed some of the minima which they observed in the longitudinal magnetoresistance of n-Ge to intervalley optical phonon scattering. The striking series of peaks in p-Ge are the first definitive observation of the magnetophonon effect due to optical phonon transitions in the valence band of a semiconductor. The complex splitting of the harmonics at fields above 50 kG is rather difficult to interpret precisely but appears to reflect the complex dependence of the heavy hole Landau level energies on k_B, the hole wavevector in the direction of the applied field. A series of magnetophonon peaks has also been observed in the transverse magnet ore sistance of p-GaAs at a temperature of about 100 K. The magnetic field positions of the five light hole peaks provide an estimate of the band-edge light hole effective mass of 0.091 m₀. In the ohmic electric field regime, the structure in the magnetoresistance disappears as the temperature is lowered below about 50 K due to the decreasing importance of optical phonon scattering in limiting the electron mobility. However, fay applying electric fields of sufficient strength to heat the electron temperature above that of the lattice, extrema can be made to reappear in the magnetoresistance. The phonons responsible for the structure are those which dominate the energy relaxation of the electrons and the reappearance of the peaks arises from the oscillatory variation with magnetic field of the energy relaxation time of the electrons. In InP, at a temperature of about 10 K f the most prominent series of peaks is caused by electron capture at a shallow ionised donor site accompanied by the emission of a single L.O. phonon. At electric fields below about 5 V/cm a second series of five extrema, accurately periodic in ¹/B, appears in the magnetoresistance. This structure is attributed to energy relaxation of electrons by the simultaneous emission of a pair of band-edge transverse acoustic phonons. The field positions of the peaks satisfy the condition ^Nheb⁄_m* = 2hω (T.A.X) (2) where hω (T.A.X) is the T.A. phonon energy at the X-point of the Brillouin zone. In addition to the structure arising from phonon emission processes, additional peaks appear in the magnetoresistance of all the n-InP samples studied. The field positions of these extrema appear to correspond to an energy relaxation mechanism in which the electrons inelastically scatter from neutral donor sites which are thereby excited from the ground state to the lowest energy Landau state. The energies of the donor states which are required for the interpretation of the warm electron magnetophonon data are deduced from a series of photoconductivity experiments on n-InP (Stradling et al., 1972). This work involved the use of two cryostats designed for infra-red measurements at temperatures down to 1.2 K. The spectra reveal the presence of two shallow donor species whose ground states are separated in energy by an amount equivalent to 0.7 cm^-1. The mean ionisation energy of the two impurities is 61.0 cm^-1. The photoconductive response also shows an additional peak at 33.8 cm^-1 which appears to have an intensity directly related to the width of the shallow donor lines.

Piezotronics is a new field, as first explored by Professor Zhong Lin Wang (Georgia Institute of Technology, Atlanta, USA), which describes the exploitation of the piezoelectric polarization and internal electric field inside semiconductor nanostructures by applying strain, to develop electronic devices with new functionality. Such concepts find applications in both III-V and II-VI semiconductor compounds, in optics, optoelectronics, catalysis, and piezoelectricity, sensors, piezoelectric transducers, transparent conductor and nanogenerators. In this work I explore the strain dependence of the piezoelectric effect in wurtzite ZnO crystals. The Linear and quadratic piezoelectric coefficients of III-V (GaP, InP, GaAs and InAs) wurtzite semiconductors are also calculated using ab-initio density functional theory. The polarization in terms of the internal anion–cation displacement, the ionic and dipole charges is written and the ab initio Density Functional Theory is used to evaluate the dependence of all quantities on the strain tensor. The piezoelectric effect of III–V semiconductors are nonlinear in the strain tensor. The quadratic piezoelectric coefficients and a revised value of the spontaneous polarization are reported. Furthermore, the ZnO nanowires is found to be non-linear piezoelectric effect and leads to predictions in some cases opposite to those obtained using the widely used linear model. The predicted magnitude of such coefficients are much larger than previously reported and of the same order of magnitude as those of III-N semiconductors. We also model the bending distortion created on a III-V wurtzite nanowire by an atomic force microscope tip induced deflection to calculate the piezoelectric properties of both homogenous and core shell structures. A number of combinations of III-V materials for the core and the shell of the nanowires, are shown a favour much increased voltage generation. The largest core voltages in core/shell combinations of InAs/GaP, InP/GaP, GaP/ InAs and GaP/InP are observed which can be theoretically 3 orders of magnitude larger than the typical values of ±3V in homogenous nanowires. Also considering properties such as bandgap discontinuity and mobility, III-V wurtzite core shell nanowires are candidates for high performance components in piezotronics and nanogeneration.

Lead zirconate titanate (PZT) is a oxide, which posseses a perovskite type structure. The material is ferroelectric making it technically useful for a diverse range of applications from thermal imaging to non-volatile integrated memory devices. Thin films of PZT are commonly deposited onto platinised-silicon (Pt(111)/Ti/SíOX/Si) bottom electrodes by chemical solution deposition (CSD). Depending on the conditions used a transient intermetallic phase Pt3Pb can form in-situ with a (111) preferred orientation during processing, which reduces the lattice mismatch between the desired perovskite phase and bottom electrode. Thereby, making it easier for the perovskite phase to both nucleate and take on a preferred (111) orientation. In the work presented experiments were conducted on modifying the surface of platinised-silicon to achieve a better lattice-match to Pb(Zr0_3Ti0_7)O3 (PZT30/70) through coating, alloying or reacting the platinum with silver (Ag), gold (Au) or lead (Pb). Single layer and multi-layer PZT thin fihns were deposited by a CSD method onto these modified bottom electrodes and plain platinised-silicon. Two precursor sols were used one that had 10% excess Pb added (PZT30/70) and one that had no excess Pb added (StoichPZT30/70). In general, two drying regimes prior to crystallisation were employed for the precursor fihns. i. Dried at 200°C for 3min ii. Dried at 300°C for 5min Crystallisation of the dried films was conducted i general at 480°C. The primary result of the work, showed that the presence of a lattice-matched surface is highly beneficial for nucleation of perovskite PZT and the subsequent transformation from amorphous precursor into perovskite PZT at low crystallisation temperatures < 500°C. All the modified bottom electrodes performed well when used i conjunction with fihns dried at 300°C for 5min, which otherwise would not transform on plain platinised-silicon. It was found that the Au and Ag modified platinised-silicon behaved i a similar manner to plain platinised-silicon with respect to in-situ Pt3Pb, except that the Ag modified platinised-silicon appeared to impede its formation and promote its oxidation. I general the perovskite phase was detected by x-ray diffraction earlier on Ag modified platinised-silicon than on either Au modified or plain platinisedsilicon. The PZT30/70 thin films dried at 200°C for 3min on Au and Ag modified Abstract bottom electrodes appeared to have a higher crystallinity than similar than similar films deposited onto plain platinised-silicon. This was not observed for the StoichPZT30/70 sol as the crystallinity of these films, dried at 200°C for 3mjn were similar i value with all being greater than comparable PZT30/70 sol thin films. It was also proposed that the transformation rate from pyrochlore to perovskite could be affected by precursor sol Pb content and its distribution within the film during crystallisation. This was illustrated when bottom electrodes containing Pt3Pb before the film deposition (ex-situ) were used. PZT30/70 films dried at 200°C for 3min, provide conditions for forming in-situ Pt3Pb and this produced poor quality perovskite PZT thin films on ex-situ Pt3Pb. Implying that i the concentration of Pb is too high at the film/electrode, not only is the preferred orientation of the perovskite thin film compromised but also the degree of transformation from an amorphous to the perovskite phase will be compromised. However, StoichPZT30/70 films dried at 300°C for 5min do not form in-situ Pt3Pb on crystallisation but When crystallised on ex-situ Pt3Pb they appear to transform rapidly into the perovskite phase with a preferred (111) orientation.

The thesis describes the development of laser-assisted bonding methods for assembly of microfluidic devices and MEMS packaging. A laser microwelding technique for assembly of transparent polymer substrates for fabrication of microfluidic devices was studied. The transparent PMMA substrates were bonded together using a high power diode laser system with a broad top-hat beam profile and an intermediate titanium thin film consisting of 0.7 mm diameter spots. A tensile strength of 6 MPa was achieved for this novel method which is comparable to that of the previous work in laser welding of polymers. It has been demonstrated that the method is capable of leak free encapsulation of a microfluidic channel. Furthermore, a novel laser-based method using an LCP film for packaging of MEMS, sensors and other microelectronic devices has been investigated. The results show that it is possible to use a laser based method with an LCP polymer for high quality substrate bonding applications. Glass-glass based cavities allow optical transmission and have potential applications for optical sensors and other photonic devices. For glass-glass bonding, it was shown that thin film titanium material can be used as an effective optical absorber in the laser based LCP bonding technique. Laser bonding of glass and silicon using an LCP film has also been achieved but in this case the silicon substrate acted as the absorber to capture the laser power. Laser bonding of a silicon cap to a molded LCP package has also been demonstrated successfully. The results of temperature monitoring using embedded sensors show that the temperature at the base of the LCP package (~130C) is substantially lower than the bonding temperature (> 280C). The results of shear and leak test show good reliability and hermeticity of the laser bonded microcavities. Both two-dimensional and three-dimensional models of heat transfer are developed and studied using the COMSOL Multiphysics software tool to understand the localised laser heating effects. The results are in good agreement with those of the practical work.

Planar InAs avalanche photodiodes (APDs) are reported as low noise, high gain photon detectors operating across the electromagnetic spectrum from 1.5 to 3.5 µm. This work includes a study of post-growth selective area doping techniques in InAs required for forming planar junctions, through to developing the first planar and lateral InAs APDs to realise high gain. Be ion implantation and annealing are developed for selective area P-type doping InAs. An implantation and annealing procedure was optimised to maximise Be activation and recovery, whilst minimising Be diffusion. A planar fabrication procedure was developed and InAs planar photodiodes were characterised with high uniformity and low surface leakage achieving a detectivity of 6.08 × 108 cmHz^-1/2W^-1 at room temperature. InAs APDs were fabricated using a planar fabrication process. The activation energy of the bulk and surface leakage components of the dark current were analysed to determine the dominant leakage mechanisms. Utilising an optimised structure to minimise tunnelling current, a record high gain in excess of 300 was achieved at -26 V at 200 K. The maximum gain was limited by breakdown, and the breakdown mechanism was found to be due to an unusual thermal runaway effect within an electric field hotspot at the planar junction edge. To mitigate the formation of electric field hotspots, planar APDs with guard rings were designed and characterised. Planar APDs with optimised guard ring placements were characterised with lower dark current near breakdown, and an increased breakdown voltage at 200 K compared to unguarded APDs. The gain limitations of InAs APDs utilising a conventional structure are discussed. To overcome such limitations a novel lateral APD structure was proposed. A range of lateral APD structures were simulated to evaluate the evolution of a lateral electric field that may lead to enhanced lateral gain. N-type selective area doping using Si implantation and annealing were developed, and a range of lateral APD structures were fabricated utilising Be and Si implantation and annealing. Lateral APDs were characterised and an optimised structure was identified. Finally a discussion of the recommended work to be carried out on InAs APDs is presented.

Two terminal electrical measurements of freestanding multiwalled carbon nanotubes employing tunnelling contacts have been made. Previous electrical characterisation on both single and multiwalled carbon nanotubes have revealed a zero bias anomaly, in which the current is characterised by a power law (I &prop; Va+1). In the case of single wall nanotubes this behaviour has been explained within Luttinger liquid theory, the strong repulsive electron-electron interactions found in one dimension causing Fermi liquid theory to break down. The origin of the power law in the characteristics of multiwalled nanotubes is less clear and is the subject of considerable debate. In part, the debate is fuelled by the very similar predictions of the various theories, particularly those of Luttinger liquid and single junction Coulomb blockade theory. The measurements presented in this thesis employ a unique combination of a freestanding geometry with high resistance tunnelling contacts, which are necessary to probe electron-electron interactions. The results are quantitatively very different to all previous reports and are explained using the environmental quantum fluctuations found in single junction Coulomb blockade. The size of the quantum fluctuations is strongly influenced by the high frequency impedance of the nanotube. The freestanding geometry causes reflections within the nanotube, resulting in an impedance-frequency characteristic significantly different to that expected for conventional on-substrate geometries. It is proposed that this is the cause of the large quantitative difference between the results gained from the freestanding geometry and the on-substrate geometry. The exponent, a which characterises the power law (I &prop; Va+1) is typically around 0.3 in the conventional on-substrate geometries (both theoretically and experimentally). The original results presented in this thesis show a ~ 3.5 and the exponent shows considerable variation. The non-conventional freestanding geometry, splits for the first time, the predicted exponent from Luttinger liquid and environmental quantum fluctuations theories as they are applied to multiwalled carbon nanotubes. The later interpretation is favoured not only because of the quantitative agreement with the exponent, a but unlike Luttinger liquid theory it is also in agreement with other experimental observations, such as offset ohmic behaviour. The agreement is such that these results could be the best demonstration of environmental quantum fluctuations in a single junction Coulomb blockade. The power law in MWNTs appears to have several manifestations. Uniting them all in a single explanation could well be impossible. So, while this work may not end the debate on the origin of the power law in MWNTs it certainly will make a significant contribution.

In this thesis mid-infrared time-resolved pump-probe measurements are presented to evaluate spin dynamics in narrow gap semiconductors. The polarization of the pump beam is modulated and the transmittance change in the probe beam is measured. The optical polarization is measured versus time delay and its decay yields the spin lifetime. Results show that the spin lifetime decreases with increasing temperature. Between InAs and Insb films, although InAs shows longer spin lifetime, it has an accumulation layer in the surface with high current and very short spin lifetime, while InSb has a surface depletion layer. The spin dephasing dominates for the range of temperature 77 - 293 K for thick films, however thin films show stronger spin flip relaxation mechanism. For AlInSb/InSb quantum wells, the spin lifetime depends on the well width, mobility, and carrier concentration, and the dephasing mechanism dominates for high mobility, but for low mobility at low temperature the spin flip scattering is dominant. We have also investigated the Lande g-factor for the first time at room temperature, using the same technique by adding an external magnetic field causing spin precession. Our experimental results show that with increasing temperature and energy, the absolute g-factor, decreases. In comparison, simple predictions of k.p theory for temperature dependency of g-factor, which changes primarily because of the change in the band gap, suggest the opposite. We find leaving only the dilational part in the temperature dependence of the energy gap, improves the consistency between experimental and theoretical g-factors. Finally, a far-infrared time-resolved pump-probe study of population dynamics of shallow donor impurity states (P, As) in silicon shows that at low temperatures, the dominant linewidth broadening is due to direct phonon decay rather than dephasing collisions or inhomogeneous broadening. The achieved long lifetimes are convenient for applications such as quantum information processing.

This study investigates novel three-dimensional passive microwave components that make use of multilayer technology defined as a production process employing several alternately stacked layers of metal and dielectric structures. Components developed include conventional and equal delay Ruthroff transformers; Guanella baluns; stepped impedance lowpass filters; bandpass filters formed from quarter wavelength stubs and transmission line sections; and balanced transmission lines. The theory of operation of each component is described along with the required design procedures for fabrication in the form of a multilayer MMIC using the Caswell H40 process, or a multilayer MIC using the Hibridas photoimageable thick-film process. New practice includes the development of novel structures including a balanced transmission line with conductors arranged in a broadside configuration; a pseudo coplanar waveguide transmission line with a high characteristic impedance; and a new design of compensated microstrip T-junction with clearly defined reference planes between the three branch lines. An investigation into how certain components operate has resulted in two pieces of new theory which are: a design procedure for bandpass filters formed from quarter wavelength stubs and transmission line sections; and a new theory of operation of equal delay Ruthroff transformers and Guanella baluns which takes into account the parasitic common mode currents. Also included in the thesis are: a brief history of MICs and MMICs including an overview of the different fabrication processes used by industry; a comprehensive description of the Caswell H40 multilayer MMIC process, and the Hibridas photoimageable thick-film process; and a section devoted to test and measurement procedures for circuits fabricated in a MIC or MMIC form using microwave probes, which includes a critical analysis of different calibration and de-embedding methods, and details of suitable test equipment and procedures for testing multiport circuits and circuits with balanced ports.

In recent years it has become increasingly apparent that there is a large demand for resilient electronics that can operate within environments that standard silicon electronics cease to function such as high power and high voltage applications, high temperatures, corrosive atmospheres and environments exposed to radiation. This has become even more essential due to increased demands for sustainable energy production and the reduction in carbon emissions worldwide, which has put a large burden on a wide range of industrial sectors who now have a significant demand for electronics to meet these needs including; military, space, aerospace, automotive, energy and nuclear. In extreme environments, where ambient temperatures may well exceed the physical limit of silicon-based technologies, SiC based technology offers a lower cost and a smaller footprint solution for operation in such environments due to its advantageous electrical properties such as a high breakdown electric field, high thermal conductivity and large saturation velocity. High quality material on large area wafers (150 mm) is now commercially available, allowing the fabrication of reliable high temperature, high frequency and high current power electronic devices, improving the already optimised silicon based structures. An important advantage of SiC is that it is the only wide band gap compound semiconductor that can be thermally oxidised to grow insulating, high quality SiO2 layers, which makes it an ideal candidate to replace silicon technologies for metal-oxide-semiconductor applications, which is the main focus of this research. Although the technology has made a number of major steps forward over recent years and the commercial manufacturing process has advanced significantly, there still remains a number of issues that need to be overcome in order to fully realise the potential of the material for electronic applications. This thesis describes the characterisation of 4H-SiC CMOS structures that were designed for high temperature applications and fabricated with varying gate dielectric treatments and process steps. The influence of process techniques on the characteristics of metal-oxide-semiconductor (MOS) devices has been investigated by means of electrical characterisation and the results have been compared to theoretical models. The C-V and I-V characteristics of both MOS capacitor and MOSFET structures with varying gate dielectrics on both n-type and p-type 4H-SiC have been analysed to explore the benefits of the varying process techniques that have been employed in the design of the devices. The results show that the field effect mobility characteristic of 4H-SiC MOSFETs are dominated at low perpendicular electric fields by Coulomb scattering and at high electric fields by low surface roughness mobility, which is due to the rough SiC-SiO2 interface. The findings also show that a thermally grown SiO2 layer at the semiconductor-dielectric interface is a beneficial process step that enhances the interfacial characteristics and increases the channel mobility of the MOSFETs. In addition to this it is also found that this technique provides the most beneficial characteristics on both n-type and p-type 4H-SiC, which suggests that it would be the most suitable treatment for a monolithic CMOS process. The impact of threshold voltage adjust ion implantation on both the MIS capacitor and MOSFET structures is also presented and shows that the increasing doses of nitrogen that are implanted to adjust the threshold voltage act to improve the device performance by acting to modify the charge at the interface or within the gate oxide and therefore increase the field effect mobility of the studied devices.

Photolithography, the standard pattern transfer technique, has many sustainable issues due to the application of a mask to the substrate. A ‘maskless’ pattern transfer method, called the Enface technique, has recently been proposed for metal plating and etching. This method introduces the idea of bringing a patterned tool and a substrate together in close proximity and a current or voltage is passed between them enabling metal to be selectively deposited or removed from the substrate. The process requires sufficient electrolyte agitation within a narrow inter-electrode gap and has previously been shown to hold in a vertical flow channel reactor. However, the process has to be adapted for tank-type systems for industrial implementation. Mass transfer during electrodeposition can be enhanced by ultrasonic waves. It has therefore been investigated whether this would be an appropriate agitation method for Enface. In order to scale-up the process, 3 types of Enface reactors were investigated; a vertical flow cell, a 500 ml lab-scale tank-type cell and an 18 L ultrasound plating tank. The limiting current technique was used to study the mass transfer in these systems. Electrodeposition of copper pattern features in 0.1 M CuSO4 was achieved in each of these geometries. The scalability was quantified by measuring the uniformity of deposit roughness and deposit thickness of the features across the substrate using profilometry. The lab-scale tank-type cell with a 20 kHz ultrasound probe was used to investigate the effect of ultrasound agitation within narrow inter-electrode gaps. Mass transfer correlations showed that turbulent flow becomes fully developed when using ultrasound in this narrow geometry. Limiting current experiments showed that relatively low ultrasound powers of 9 – 18 W/cm2 should be used and current distribution modelling showed that the ultrasound source should be placed no less than 30 mm from the substrate. Copper pattern features were deposited onto 10 mm diameter substrates and using long current pulses with bursts of ultrasound during the off-time was the most suitable plating mode. Specially designed electrode holders in the large-scale 18 L ultrasound tank was used to deposit copper patterns onto larger substrates. Features of μm-scale were deposited onto A7 size substrates, but there was an unacceptable variation in deposit thickness of ±80% due to the non-uniformity of the electrode gap across the plate. However, mm-scale features were successfully deposited onto A7 size substrates with an acceptable deposit thickness uniformity and deposit roughness uniformity of ±18% and ±40% respectively across the plate. Enface is therefore currently scalable for mm-scale features on substrates of this size.