Chapter I presents an overview of mass spectrometry on double focusing instruments. Special attention is paid to ionisation methods which may be of use in mixture analysis. In chapter II, following a brief introduction to chemometrics, results of the application of factor analysis to the problem of determining the components of a mixture from mass spectra of simple mixtures are given. After applying the technique to simulated data, the results of applying the technique to simple mixtures and to the deconvolution of overlapping GC/MC spectra are given. It is concluded that provided the pure compounds have 'unique' peaks in their mass spectra, and that the statistical variations in the intensities of peaks due to different mixtures are not correlated, then the technique will yield good results. Chapter III deals with the application of CI techniques and factor analysis to the analysis of engine oils. Attempts to use the factor analysis technique to yield a meaningful 'type' analysis of engine oil fractions was not successful. The use of proton transfer reagents and charge transfer reagents for CI spectra of engine oil fractions was also unsuccessful. It is concluded that the complexity of engine oil mixtures and the chemical similarity of the constituent molecules makes them difficult to differentiate by CI mass spectrometry. In the final chapter, details are gave of an investigation into the use of CI mass spectrometry and FAB mass spectrometry for the analysis of engine oil additives. The additives studied were calcium 2,2'-bis (4-alkylphenyl) sulphides. The FAB mass spectrometry of these sulphurised phenates gave poor results with glycerol, triethanolamine, triethanolamine / sodium sulphate or squalane used as FAB matrix. Of the CI techniques used, electron capture ionisation of the hydrolysed sample gave the most promising results. The molecular ion peak intensities were higher than those encountered in the EI spectra. In addition, the fragmentation in the EC spectra was simpler than the fragmentation in the EI spectrum. Metastable ion studies of the EI and EC spectra of the sulphurised phenols, using the technique of metastable mapping, gave some useful results, but the poor mass resolving power and the long run times were a major drawback. It is concluded that the technique of metastable mapping is of no use where mass resolution is important, and the use of inlet systems, such as an AGHIS, which allow long sample lifetimes is recommended. Where possible, the use of tandem mass spectrometry or Fourier transform mass spectrometry for metastable ion studies for mixture analysis, is also recommended.

Blends of lubricant base stocks are commonly employed to produce lubricants with optimised performance. However, the influence of blending on lubricant behaviour within high pressure, mechanical contacts, particularly the film forming capacity and friction, have largely gone unstudied. This thesis examines both of these aspects for a range of base fluids and their blends. Film thickness generation and fluid friction are reviewed in the context of elastohydrodynamic (EHD) lubrication. Various models addressing the origin and mechanism of liquid viscosity are reviewed. The derivation of viscosity as a thermodynamic property is also reviewed and its relevance to the current study is discussed. Consideration is also given to two commonly accepted yet contrasting models of EHD traction, resulting in one being adopted for this research. The relationship of molecular structure to film generation is examined experimentally by studying a number of single component fluids with widely varying chemical structures. It is shown that pressure-viscosity coefficients derived from film thickness are strongly related to the respective chemical “family” of the lubricant. EHD friction (traction) measurements are also made and analysed in order to establish a reliable method for comparing the influence of fluid composition on traction. The method, based on a well regarded fluid model, allows accurate description of full fluid traction by means of the Eyring stress and pressure-viscosity coefficient in the central EHD contact. Binary blends of well-defined base fluids are studied. It is found that the pressure- viscosity coefficient, as derived by the aforementioned methods, varies nonlinearly with the composition and tends toward the lower value of the individual components, at times attaining values lower than either of the individual components.

A high-traction fluid is one of the vital components of a traction drive; a mechanism where input power is transformed into an output force when pressure and shear force are applied to the fluid. In this environment they need to withstand high pressures, temperatures and shear forces, and must be able to lubricate at the same time. So far there has been very little research relating engineering performance to molecular structure. Previous work with model hydrocarbon traction fluids has shown that molecular rigidity about the centre of the molecule appears to be important. This work has now been extended to cyclohexyl esters. A series of cyclohexyl esters was synthesised, and where possible, x-ray structures were obtained. Extensive 13C T1 and NOE relaxation data over a range of radiofrequencies and temperatures were obtained and used to calculate correlation times, for both overall and internal motion, using both the "model-free" and reduced Lorentzian models. This provided information on the rigidity of the molecules studied, In addition molecules have been modelled using molecular dynamics techniques to calculate order parameters and torsion angle distributions. ESR studies have been conducted to measure the viscosity of each fluid. A spin probe, with a similar structure to one of the molecules studied, was synthesised. This allowed complimentary correlation time measurement, and showed that the molecule rotated isotropically. Difficulties were encountered in fitting the NMR data to the motional models and the need for higher radiofrequency data is indicated, to check the validity of the models used. The NMR, ESR and molecular dynamics results did, however, provide a consistent indication of the differing rigidities and motions of these molecules.

No description available.

No description available.

No description available.

The development of more fundamental knowledge of the tribology of the cam and follower mechanism calls for a more comprehensive theoretical analysis and experimental investigation than has been previously reported. A mixed lubrication analysis has been applied to the problem to give an estimation of the nominal minimum film thickness and friction force associated with the contact in such mechanisms. The analysis showed that the roughness height and the distribution of the roughness between the two contacting surfaces had important effects on the lubrication performance of the contact. A full numerical transient EHL analysis was carried out allowing the normal velocity to vary along the conjunction. This revealed that local squeeze film velocity provided an increased damping effect which contributed to the persistence of the minimum film thicknesses in the two zero entraining velocity regions. An approximate technique for determining the minimum film thickness of a transient EHD line contact associated with rough surfaces was developed and applied to the mixed lubrication analysis of a four-power polynomial cam and non-rotating flat faced follower arrangement. The results demonstrated that under certain circumstances mixed lubrication predominated in the conjunction of the cam and follower with the surfaces being separated by an EHL film on the cam flanks. Existing experimental apparatus was improved to test the effects of altering the bulk temperature and camshaft rotational speed by measuring the friction torque and electrical resistivity across the contact. By adopting advanced techniques for data sampling and processing the instantaneous friction torque was successfully obtained with the camshaft rotational speed exceeding (2000 rpm). The wear characteristics were also examined. The bulk temperature showed a mild effect on the wear characteristics of the cam and follower as it was increased from (75° C) to (105° C), whilst, a substantial influence was found as the temperature was further increased to (120°C). Increasing the bulk temperature caused an increase in both the friction torque and power loss o f the valve train, but this increase was not considerable. Based upon the theoretical analyses and experimental observations, a theoretical model for evaluating the tribological performance of the valve train was developed. A multi-aspect comparison between theoretical and experimental results was made. The excellent agreement between theoretical and experimental results showed that the model provided a reliable prediction o f the tribological characteristics of the cam/flat faced follower. Three critical portions of the cycle could be identified — one over the cam nose and two in the vicinity of the zero entraining velocity regions. The minimum separation between the cam and follower occurred near the falling flank of the cam.

Lubricating oils subjected to the operating conditions in an automobile engine degrade by many routes. The most important of these routes is oxidation, which at room temperature and atmospheric pressure is almost immeasurably slow. The performance of a new oil formulation is ultimately assessed by the use of a standard engine test. Such engine tests are very expensive and time consuming, which makes the use of an initial screening test a necessity. Screening tests are used to eliminate any oils that would fail the full engine test, thereby saving considerable amounts of time and money. In the present research the concern has been to compare two screening tests for the analysis of antioxidants based on differential scanning calorimetry (DSQ and cyclic voltammetry (CV). The two techniques have been applied to the analysis the of the antioxidants, dioctyldiphenyl amine (DODPA) and Topanol '0' in three base oils. DSC is a commonly used technique in which the time to oxidation is measured when the sample is maintained at a constant temperature in a high pressure atmosphere of oxygen. CV has not been used widely to examine lubricating oils and the present work has included the development of reliable experimental techniques for the two types of antioxidants. The results obtained for both freshly prepared oil formulations and those partially oxidised clearly reveal the different nature of the tests. DSC is a non specific technique with respect to the antioxidant and leads to an indication of the overall oxidative stability of the oil blend, whereas the voltammogram obtained in CV is specific to certain groups of antioxidant. The value of the current flowing through the circuit reaches a maximum value at a potential specific to and dependant on the concentration of the additive being analysed. A kinetic analysis of an antioxidant by CV leads to information about the diffusion coefficient of the molecule and the rate of electron transfer at the electrode surface. It was found that the electrochemically oxidised species of the amine antioxidant underwent a chemical reaction to produce an electro-inactive species over the potential range studied. The kinetic data for this reaction was obtained but their relevance to the oxidation of an oil in an engine environment is doubtful, whereas the results obtained by DSC are probably more applicable. The kinetic data obtained from isothermal DSC curves were employed in a combined kinetic scheme to model the shapes of DSC curves.

No description available.

This dissertation is concerned with the physical processes which determine friction and lubrication in metal rolling in the mixed lubrication regime, with particular attention paid to the conditions encountered when rolling aluminium foil. Two areas of relevance to the analysis of the rolling process are initially investigated. Firstly, the rheological properties of a typical aluminium foil rolling oil at high pressures and shear rates have been measured using a disc machine. The behaviour of the oil was found to be well described by the Eyring viscous model, at the shear rates and pressure likely to be found in metal rolling. Secondly, the deformation of asperities when the bulk material is deforming has been examined. The theory developed here was found to agree reasonably with experiments. The results of these investigations are used in the analysis of lubrication in metal rolling, considering the hydrodynamic buildup of oil pressure in the entry region and the crushing of the asperities both in the entry region and at the beginning of the work zone. The contact between roll and strip is divided into two regions, that under the asperities and that in the intervening valleys. Calculations for conditions appropriate to strip and foil rolling give the proportion of the two types of contact and the film thicknesses in each region. Measurements of film thicknesses with an experimental mill in a regime where roughness is unimportant were not found to agree well with an existing simple theory of lubrication. This was ascribed to uneven lubrication in the experiments. After taking this into account, the experiments in a regime where roughness was important were found to agree reasonably with the theory developed here. The effect of roughness on traction is measured in a disc machine with elastic contacts. Its behaviour is found to be determined by the bulk properties of the lubricant at the pressures and strain rates under the asperities. Theory and experiments presented in this dissertation lead to a greater understanding of the physical processes determining friction in metal rolling in the mixed lubrication regime. Film thicknesses and friction coefficients in metal rolling may now be estimated with more confidence.