Table of contents

Volume 34

Number 21, 7 November 2001

Previous issue Next issue

RAPID COMMUNICATION

L109

, , and

The high photon energy 157 nm F2 laser has potential applications as a source for modifying the refractive index of glasses for telecommunication device fabrication, e.g. fibre Bragg gratings. At this wavelength even `pure' silica glass exhibits significant absorption and the cladding on fibres may produce non-negligible beam attenuation. Here we report new results using a fibre Bragg grating for in situ monitoring of the temperature rise produced in a silica fibre by 157 nm F2 laser exposure. These results, coupled with thermal modelling, allow the fibre cladding absorption coefficient (~7750 m-1) in the vacuum ultraviolet to be determined.

PAPERS

APPLIED INTERFACES AND SURFACES

3097

Heavily doped zinc oxide films are used as transparent and conductive electrodes, especially in thin film solar cells. Despite decades of research on zinc oxide it is not yet clear what the lower limit of the resistivity of such films is. Therefore, the electrical parameters of zinc oxide films deposited by magnetron sputtering, metal organic chemical vapour deposition and pulsed laser ablation are reviewed and related to the deposition parameters. It is found that the lowest resistivities are in the range of 1.4 to 2×10-4 Ω cm, independently of the deposition method. The highest reported Hall mobilities are about 60 cm2 V-1 s-1. The thin film electrical data are compared with the corresponding values of single crystalline zinc oxide and with that of boron and phosphorous doped crystalline silicon. From this comparison it can be seen that the dependence of the Hall mobilities on the carrier concentration n are quite similar for silicon and zinc oxide. In the region n>5×1020 cm-3, which is most important for the application of zinc oxide as a transparent and conductive electrode, phosphorous doped silicon has a mobility only slightly higher than zinc oxide. The experimental data on the electron and hole mobilities in silicon as a function of the impurity concentration have been described by a fit function (Masetti et al 1983), which can also be applied with different fitting parameters to the available zinc oxide mobility data. A comparison of the experimental data with the well known ionized impurity scattering theories of Conwell-Weisskopf (1946) and Brooks-Herring-Dingle (1955) shows that these theories are not able to describe the data very well, even if the non-parabolic band structure is taken into account. As in the case of silicon, an additional reduction of the mobility also occurs for zinc oxide for concentrations n>5×1020 cm-3, which can be ascribed qualitatively to the clustering of charge carriers connected with increased scattering due to the Z-2 dependence of the scattering cross section on the charge Z of the scattering centre. The presented review of the charge carrier transport in zinc oxide indicates that a physical limit due to ionized impurity scattering is reached for homogeneously doped layers. Due to the universal nature of this limitation it is suggested that it also applies to the other important materials indium-tin (ITO) and tin oxide. Experiments are proposed to overcome this limit.

3109

, , , , and

The effects of white light illumination during the deposition of CdS thin films in a quasi-closed volume on the structural, photoelectrical and optical properties are investigated. The films were highly c-axis oriented with an increasing intensity of (002) reflection as the illumination increases. The room temperature resistivity values of the CdS films decreased in the range of 107-104 Ω cm. The photosensitivity in the fundamental absorption region and the transparency in the transmission region considerably increased as the illumination increased. Under 100 mW cm-2 insolation, the efficiencies of the CdS/CdTe solar cells based on CdS window materials which were deposited: (1) in the dark; and (2) under an illumination of 150 mW cm-2; were found to be 1.8% and 7.3%, respectively.

3113

The layer thickness and tungsten metal volume fraction of W-AlN cermet solar selective absorbing coatings on a W, Cu or Al infrared reflector with a surface aluminium oxynitride (AlON) or Al2O3 ceramic anti-reflector layer were optimized using physical modelling calculations. Due to limited published data for the refractive index of AlN, and likely oxygen contamination during reactive sputtering of AlN ceramic materials, AlON was used as the ceramic component and the published value of its refractive index was employed. The dielectric function and then the complex refractive index of W-AlON cermet materials were calculated using the Ping Sheng approximation. The downhill simplex method in multi-dimensions was used in the numerical calculation to achieve maximum photo-thermal conversion efficiency at 350 °C under a concentration factor of 30 for a solar collector tube. Optimization calculation results show that the initial graded (ten-step layers) cermet films all converge to something close to a three-layer film structure, which consists of a low metal volume fraction cermet layer on a high metal volume fraction cermet layer on a metallic infrared reflector with a surface ceramic anti-reflection layer. The optimized three-layer solar coatings have a high solar absorptance of 0.95 for AlON and 0.96 for the Al2O3 anti-reflection layer, and a low hemispherical emittance of 0.073 at 350 °C. For the optimized three-layer films the solar radiation is efficiently absorbed internally and by phase interference. Thermal loss is very low for optimized three-layer films due to high reflectance values in the thermal infrared wavelength range and a very sharp edge between low solar reflectance and high thermal infrared reflectance. The high metal volume fraction cermet layer has a metal-like optical behaviour in the thermal infrared wavelength range and makes the largest contribution to the increase of emittance compared with that of the metal infrared reflector.

3121

, , , and

Lead scandium titanate (PST) thin films were deposited by RF dual magnetron sputtering and then annealed either by vacuum furnace or combined rapid thermal annealing (RTA) and furnace anneal. The film structure was investigated by x-ray diffraction, scanning electron microscopy and transmission electron microscopy (TEM) and energy dispersive x-ray spectroscopy techniques. Lead loss was more severe using furnace annealing than the combined RTA and furnace anneal. The annealed films were characterized by the presence of voids and exhibited relaxor ferroelectric characteristics. PST perovskite crystal grains were found to co-exist with pyrochlore matrix in the furnace-only annealed films, whilst in RTA annealed films no apparent pyrochlore morphology was observed in the TEM image. Lead was found to diffuse through the bottom electrode Pt layer during the annealing. Films treated by combining RTA and furnace annealing have shown pyroelectric coefficients under field of up to 500 µC m-2 K-1, a dielectric loss of below 0.007 and a merit figure for thermal detection of 2.5×10-5 Pa-1/2.

3130

, , and

The recombination zone in triple-layer electroluminescent (EL) devices with the structure indium-tin-oxide (ITO)/N,N'-diphenyl-N,N'bis(3-methylphenyl)-1,1'-biphenyl-4,4'diamine (TPD)/(tris-hydroxyquinolinato (Alq3) + dopant)/Alq3/Mg:Ag has been determined based on their emission quantum efficiency (φEL), and examined as a function of applied voltage and dopant concentration in the emitter layer (EML). Devices containing an EML of 5,6,11,12-tetraphenylnaphthacene (rubrene) doped into Alq3 generate yellow emission with a peak external φEL of 2.5% photons/carrier at an applied field of ~0.6 MV cm-1. Devices containing an EML of 6,13-diphenylpentacene (DPP) doped into Alq3 exhibit a saturated red emission and a maximum φEL of 1.5% photons/carrier at ~0.7 MV cm-1 at low dopant concentration levels. The field (F) dependence of φEL shows a maximum at Fmax dependent on the dopant concentration. This observation may be explained in terms of the field evolution of the recombination zone in the low-field range, and the field-assisted dissociation of the excited states in the high-field region. Other mechanisms of the high-field decrease in φEL such as singlet-singlet and singlet-charge carrier annihilation are discussed. They seem to be inapplicable under the operation conditions of these EL devices.

EXPERIMENTAL, COMPUTATIONAL AND THEORETICAL PARTICLE BEAM AND PLASMA SCIENCE

3139

, , and

In this paper, we present the results of the calculations of high-frequency electroconductivity, dielectric permeability and refraction and reflection coefficients of a fully ionized gaseous plasma in an external electric field. The electron density, Ne, and temperature, Te, varied in the domains 1017Ne⩽1019 cm-3 and 10 000⩽Te⩽50 000 K, while the electric field was in the microwave and far-infrared frequency regions. The calculations are based on the modified random phase approximation theory. The final results are also shown in a parametrized form suitable for laboratory applications.

3145

, , , and

During deep penetration laser welding, the plasma over the keyhole absorbs beam energy and reduces the power efficiency. The thermal movement of laser produced plasma is analysed theoretically and experimentally. A method to increase the electric current in the plasma is put forward. The principle and feasibility of controlling the plasma by electric and magnetic fields are discussed. An experimental procedure involving elevating the nozzle during laser welding is used to evaluate the effect of increasing the power efficiency by driving away the charged particles. The power efficiency increases with increasing magnetic field intensity. There is an optimal electric field intensity at which the power efficiency reaches its highest value. It is indicated that by applying proper electric and magnetic fields the charged plasma particles can be driven away and the power efficiency is increased.

3150

and

Improved positive column simulation techniques are needed because of the non-local nature of typical low-pressure discharges used for lighting. In a local model, the power balance between Joule heating and collisional losses must hold for each volume element of the discharge separately while a non-local model requires only a global power balance. The departure from locality increases as either gas density ng or radius R is decreased. Despite this, most current fluorescent lamp software is based on the local concept. We present a non-local kinetic particle-in-cell Monte Carlo collisions (PIC-MCC) code to simulate low-pressure, small-radius, positive column discharges. This code is also compared to a non-local fluid code, a non-local kinetic Monte Carlo code and to experimental data. The PIC-MCC code made the least approximations and assumptions and was accurate and stable over a wider parameter regime than the other codes. Also, 1d3v PIC-MCC simulation speeds are quite competitive even on moderate workstations. Finally, we analyse the PIC-MCC simulation results in detail, especially the power balance and the radial electron kinetic energy flux Hr(r). We found that for low ngR< 1×1015 cm-2, the electron kinetic energy flux is directed radially outward while for higher ngR, it is directed radially inward except right near the wall.

3164

, , and

The technique of spatially resolved cross-correlation spectroscopy (CCS) is used to carry out diagnostic measurements of the barrier discharge (BD) in air at atmospheric pressure. Quantitative estimates for electric field strength E(x,t) and for relative electron density ne(x,t)/nemax are derived from the experimentally determined spatio-temporal distributions of the luminosity for the spectral bands of the 0-0 transitions of the second positive system of N2 (λ = 337.1 nm) and the first negative system of N2+ (λ = 391.5 nm). These results are used to test the validity of some physical models of electrical breakdown in a BD. The influence of the spatio-temporal structure of the discharge on the chemical kinetics of ozone synthesis is studied by means of a semi-empirical method based on the results of spatially resolved CCS measurements.

3177

, , and

We investigated the cathode region of free-burning argon arcs at atmospheric pressure and at currents from 0.3 A up to 25 A. Apart from the well known diffuse mode of operation we also found a more constricted form which we call the blue-core mode. Different diagnostic methods are simultaneously applied to these modes of operation, that is: emission spectroscopy in conjunction with a partial LTE model for the density and temperature measurements; Thomson scattering for the electron temperature and electron density determination; and Rayleigh scattering for a determination of the gas temperature. The large differences in appearance, densities, temperatures or current formation between the discharge modes are well defined by the measurements.

3185

, , and

Time-resolved emission spectroscopy has been applied to the study of optical emission induced by pulsed positive primary streamers in coaxial geometry generated in high-purity nitrogen at atmospheric pressure. During the investigation of the post-discharge emission, the formation of N2(B 3Πg, C 3Πu, C'' 5Πu, C' 3Πu) states through N2(A 3Σu+) energy pooling reactions has been discovered. The evolution of the N2(A 3Σu+) metastable species has been inferred through the analysis of the time dependence of the band intensities of the Herman infrared N2(C'' 5Πu→ A' 5Σg+) system. It has been found that, in the case of pulsed positive streamers generated in high-purity nitrogen, the maximum density of the N2(A 3Σu+) metastables occurs during the early post-discharge period (0.5-1 µs). The production of NO(X 2Π) and OH(X 2Π) radicals originating from O2 and H2O impurities has been inferred from the emissions of the NO(A 2Σ+) and OH(A 2Σ+) states. The maximum densities of the NO(X 2Π) and OH(X 2Π) radicals occur in the 9 and 5 µs post-discharge times, respectively.

3191

, and

The electron velocity distribution function (EVDF) of electron swarms in N2 has a constricted part at an electron energy of around 2-3 eV. This feature has been explained by considering the barrier effect of vibrational excitation collisions that prevents the gain of energy by slow electrons. In order to investigate the barrier effect in detail further, we simulated the electron behaviour around the energy range of the barrier. We found that momentum transfer collisions also contribute to the barrier effect in collaboration with vibrational excitation. In addition, an evaluation of the electron flow in velocity space is thought to indicate a factor by which the presence of the barrier effect in N2 is revealed in the EVDF; the depression of the electron interchange across the barrier is not only due to the barrier effect itself but also due to the small electron transfer associated with the energy loss by electronic excitation.

OPTICAL, ELECTRONIC, ELECTRICAL AND SUPERCONDUCTING PROPERTIES AND APPLICATIONS

3197

and

We report measurements of low-frequency (10-2 Hz<f<20 Hz) conductance fluctuations (flicker noise) over a temperature range 150 K<T<450 K in single crystalline Si across the insulator-metal transition by doping with phosphorus and boron (~1019 cm-3). In this range of doping the noise (as quantified by the Hooge's parameter) is much larger than that seen in lightly doped Si. The f and T dependence shows a 1/f component whose strength increases exponentially with T (with activation energy ≈0.1-0.2 eV) although the resistivity is almost flat in this temperature range. For a sample on the metallic side of the transition the behaviour of the conductance fluctuation is very similar to that of disordered metals with the spectral power, SV(f)∝1/fα with α≈1-1.2. This is proposed to arise from the movement of defect complex involving P and interstitials/vacancies following a broad activated kinetics. Interestingly in a heavily doped sample, lying on the insulating side (but close to the insulator-metal transition) the spectral power contains discrete Lorentzians in addition to the 1/f term. The corner frequencies have an activated behaviour with activation energy ≈0.45-1 eV. This behaviour, most likely, originates from recombination-generation type mechanisms.

3203

, , , and

The dynamics of energy transfer processes in Tm+3-Tb+3 and Tm3+-Eu+3 co-doped LiYF4 crystal hosts were studied from time-resolved Tm3+ fluorescence analysis to estimate the optimal co-doping concentrations which maximize 1.5 µm laser emission from the 3H4 state of Tm3+. The analysis was carried out by finding a numerical solution to the general master equations that govern non-radiative energy transfer processes in crystalline materials and by using the Monte Carlo technique. Our analysis improves the description of experimental fluorescence decay curves. The predicted optimal co-doping concentrations and laser threshold for these luminescent systems are lower than those reported using traditional models for non-radiative energy transfer processes.