This thesis describes work involving the measurement of the uniaxial extensional viscosity of mobile elastic liquids using three methods namely: the falling mass technique of Matta and Tytus (1990), the rotating drum technique of Jones <i>et al </i>(1986) and the tube arrest (T-A) experiment of Chesterman (1952). The work assesses the application of each of these methods to liquids of varying shear viscosity and elasticity. The falling mass experiment is found to be insensitive to the stretching forces generated in low viscosity liquids, but adequate for highly elastic liquids. The development of elastic instabilities is observed to occur above critical values of strain and extension rate. Using video imaging the rotating drum technique has been revised and improved theory developed. The technique is found suited to highly elastic, high viscosity liquids but cannot repeatably produce Newtonian filaments: it is shown to produce perfect uniaxial extension at moderate extensional rates. Fibrillation is observed involving an instability restricted to a single contact point. A study of the origins of the stretching force reveals that the influence of the flow instability on the stress generated in the filament is markedly reduced. The modified T-A experiment is used to produce high velocity liquid filaments, formed from collapse of a bubble beneath a liquid's surface: it facilitates the study of low viscosity, low elasticity liquids, at high extensional rates, accompanied by large strains. Studies of aqueous solutions of high molecular weight polyacrylamide show that at high rates, a marked deceleration of the filament occurs due to a significant increase in resistance to extensional flow. The relative contributions of the inertial and extensional flow components are assessed by volumetric analysis procedures. For polymer enhanced multi-grade motor oils, it was shown that the filaments contain a significant extensional component, with an upper limit of extensional rate far in excess of that achievable in commercial rheometers.

This work is concerned with a theoretical and experimental investigation of pulsed neutron beams used in elemental <i>in vivo</i> analysis. The main object of the theoretical work is to develop a Monte Carlo simulation programme to study the history of the incident deuteron beam on the deuterated titanium target in the Van de Graaff neutron generator used as a neutron source for <i>in vivo</i> analysis. Experimental measurements are reported for a 5 x 2.5 cm organic liquid scintillator NE-213 as a neutron detector after the spectrometer was systematically calibrated, using a new technique, to determine the parameters used in the unfolding programme. The angular distributions and the background contribution of the neutrons produced by the Van de Graaff generator were measured at three different angles (0<SUP>o</SUP>, 45<SUP>o</SUP> and 90<SUP>o</SUP>) under the same conditions as those used in the theoretical study. A small neutron spectrometer was designed and constructed, using a 1 x 1 cm organic liquid scintillator cell and a new photomultiplier tube which approximately matches the small cell cross-section. This detector was tested and its gain measured experimentally at two different voltages. The results showed good discrimination between short (gamma-ray) and long (neutron) pulses, but the gain was not large enough for accurate calibration. The theoretical results showed good agreement with the experimental measurements and with previous work, but showed a small number of additional neutrons in the higher energy ranges. This occurred since the scattering angle of the incident deuteron in the target was allowed for in the computations. The dose equivalent of the neutron beam was computed from the experimental energy spectrum and was in good agreement with the result obtained under the same conditions using the neutron dosimeter.

Conventional, forced, non-resonant oscillatory rheometers can possess limitations in the measurement of some gels, where they display a pronounced mechanical weakness, are highly shear sensitive and exhibit marked time dependence. Instead, shear wave propagation techniques are proposed as one alternative. One instrument that employs shear wave propagation techniques in the determination of the dynamic shear moduli, G' and G", is the Pulse Resonance Rheometer. The Pulse Resonance Rheometer uses two modes of operation, measurements of the velocity of pulse propagation and measurements of the resonant bandwidth. This thesis uses computer simulations to determine the errors inherent the in use of a pulse velocity to calculate the dynamic moduli of a gel at the gel point, and also a pre gel point viscoelastic liquid. It compares these results to the measurements from simulations of groups and a corresponding group velocity as an inherently more accurate method of gel point determination. The simulations at gel point are carried out using constitutive equations that result from the Winter and Chambon criteria. Simulations are carried out for the pre gel point viscoelastic liquid using the Gross-Marvin ladder model to mimic the development of the gel network. A fractal dimension can be found to be associated with this build up of the network, with different fractal dimensions corresponding to different types of Winter-Chambon gel that are characterised by the viscoelastic exponent a.

The Monte Carlo neutron-photon transport code MCNP has been used to design a system for thermal neutron activation analysis (TNAA) of aluminium (Al) concentration in the bones of the hand. The clinical motivation for the work is the need for a non-invasive technique for the measurement of Al body burden in patients with chronic renal failure. It is necessary to moderate the high (≈ 75%) proportion of fast neutrons present in the un-moderated source spectrum. This serves to improve the gamma-ray yield from the thermal neutron induced <SUP>27</SUP>Al(n,γ)<SUP>28</SUP>Al reaction, while reducing interference from a competing fast neutron induced reaction on phosphorus in bone, <SUP>31</SUP>P(n,α)<SUP>28</SUP>Al. Initially the MCNP code was benchmarked against an existing fast neutron activation analysis (FNAA) system at Singleton Hospital, Swansea. The MCNP calculations were found to be accurate within inherent statistical limits (≤ 3%). Modifications of the FNAA system were then investigated using MCNP. This study indicated that due to high levels of interference from a fast neutron reaction on phosphorus the system was unlikely to be able to measure Al with acceptable sensitivity. Finally a completely new design of a TNAA system was developed using MCNP. The design was based on a cylindrical D<SUB>2</SUB>O moderator with the <SUP>252</SUP>Cf source positioned along the central axis of the cylinder and is designed to be sufficiently compact to be used in a clinical environment. Surrounding the moderator by a graphite reflector showed further improvement in the thermal-to-fast flux ratio. The dose equivalent delivered to a hand phantom in 20 minutes using a 1 mg source was 0.12 mSv. Results of the simulations indicate that the system could be made sufficiently sensitive to measure bone Al concentrations down to levels close to the normal range.

Thin film optical sensors in which a halide sensitive fluorophore is immobilised in a thin (~ 50 micron) polymer film have been developed and characterised. The sensor films use rhodamine, 6-methoxyquinoline or harmane dyes which have been functionalised and bound to a hydrophilic copolymer. The selection, synthesis, functionalisation, and polymer-binding of the dyes is described, as is the synthesis and physical characterisation of the polymer supports. Two hydrophilic copolymers have been synthesised and characterised and twenty novel halide sensitive compositions have been synthesised, or formulated, and studied. Steady-state luminescence spectroscopy and single photon counting time resolved emission spectroscopy have been used to characterise the luminescence spectra and emission decay kinetics of the lumophores, both in free solution and bound and unbound in polymer films. For all dye/polymer formulations studied the emission decay kinetics are complex. Fitting to discrete exponential analysis gives at least two, and often three decay components, while distribution analysis often gives either a very broad distribution or even groups of distributions. As a consequence the steady state response gives non linear Stern-Volmer analysis when examined over a wide dynamic range. Even so, a reasonable linear Stern-Volmer response is observed for halide over physiologically and technologically important concentration ranges. The sensor films described in this work are reversibly capable of determining aqueous iodide and halide with ~ 4% accuracy at concentrations of around 0.001 mol dm<SUP>-3</SUP>. The 90% response time to molar halide is ~ 30 s. A combination of sensor films allows the simultaneous and rapid determination of both I<SUP>-</SUP> and Br<SUP>-</SUP> in a mixed halide solution. These sensors are potentially useful for the online determination of aqueous halide in a variety of technological applications.

No description available.

No description available.

No description available.

No description available.

No description available.