Call number: LD2668 .T4 1958 M17

Call number: LD2668 .T4 1960 C37

A study of ammonia-water desorption in compact counter-flow geometries was conducted. Two novel vapor generation units, comprising integrated desorber, analyzer, and rectifier segments that use microchannel geometries, were conceptualized. The branched-tray concept features a desorber segment that uses predominantly pool-boiling mechanisms for desorption, while the vertical column desorber relies on falling-film evaporation and boiling mechanisms. Both concepts rely on falling-film heat and mass transfer mechanisms in the analyzer and rectifier sections. Segmented heat and mass transfer models, based on available correlations and modeling methodologies, were developed and used for the design of branched tray and vertical column test sections. An experimental facility was designed and constructed to evaluate desorption and rectification heat and mass transfer processes within these components, under realistic operating conditions. Data were analyzed to determine the boiling/evaporation (desorber) and condensation (rectifier) heat transfer coefficients, and to determine values of the desorber liquid and vapor mass transfer coefficients. Additionally, high-speed video and images were used to gain insights into the hydrodynamic phenomena and heat transfer mechanisms in these vapor generation units. Results of the heat and mass transfer analysis were compared with the predictions of correlations and modeling methods in the literature. The vapor generation unit (VGU) test sections were evaluated across a range of concentrated solution mass fractions (0.400 – 0.550), desorber coupling-fluid inlet temperatures (170 – 190ᵒC), and concentrated solution flow rates (0.70 – 1.3 g s-1). Flow rates in this range correspond to desorber liquid Reynolds numbers of approximately 175 to 410 for the branched tray design, and desorber film Reynolds numbers of approximately 90 to 215 for the vertical column. Pressures observed within the VGU test sections ranged from approximately 1620 to 2840 kPa during testing. The novel VGUs were shown to achieve ideal cooling capacities as high as 432 and 323 W for the branched tray and vertical column, respectively. This parameter indicates the cooling capacity that would be achieved by an idealized cooling system using the refrigerant stream produced by the experimental VGU. Ideal COPs of 0.561 and 0.496 were demonstrated for the branched tray and vertical column, respectively. Experimental heat transfer coefficients were found to range from approximately 1860 to 11690 W m-2 K-1 for the pool-boiling desorption of the branched tray VGU. A new correlation was proposed and shown to provide good agreement with the data, achieving average and average absolute deviation of -5.2 and 16.1%, respectively, across the range of conditions tested. Falling-film evaporation/boiling heat transfer coefficients, determined for the desorption process in the vertical column VGU, were found to range from approximately 1290 to 4310 W m-2 K-1. Rectifier condensation heat transfer coefficients ranging from approximately 160 to 250 W m-2 K-1 were observed. Mass transfer coefficients for the desorbers of both concepts were also quantified. These results were used to develop revised heat and mass transfer models of the VGU concepts. The revised models were demonstrated to predict component-level performance with reasonable accuracy, and may be used in the design of future compact VGUs with similar geometries and operating conditions.

Shoddy fibres or"Shoddies" are a mixture of post-consumer and post-industrial fibres diverted from textile waste streams and recycled into their raw fibre form. They have found widespread use as a raw material for manufacturing sound absorbers that include, but are not limited to: automotive, architectural and home appliance applications. The purpose of this project is to develop a simple acoustic model to describe the acoustic behaviour of sound absorbers composed primarily of Shoddy fibres. The model requires knowledge of the material's bulk density only. To date, these materials have not been the focus of much published research and acoustical designers must rely on models that were developed for other materials or are overly complex. For modelling purposes, an equivalent fluid approach is chosen to balance complexity and accuracy. In deriving the proposed model, several popular equivalent fluid models are selected and the required input parameters for each model identified. The models are: the model of Delaney and Bazley, two models by Miki, the model of Johnson in conjunction with the model of Champoux and Allard and the model of Johnson in conjunction with the model of Lafarge. Characterization testing is carried out on sets of Shoddy absorbers produced using three different manufacturing methods. The measured properties are open porosity, tortuosity, airflow resistivity, the viscous and thermal characteristic lengths and the static thermal permeability. Empirical relationships between model parameters and bulk density are then derived and used to populate the selected models. This yields several"simplified" models with bulk density as the only parameter. The most accurate model is then selected by comparing each model's prediction to the results of normal incidence sound absorption tests. The model of Johnson-Lafarge populated with the empirical relations is the most accurate model over the range of frequencies considered (approx. 300 Hz - 4000 Hz) Characterization testing yields specific values for intrinsic material parameters that allow for comparison to other porous materials. Individual parameter relations allow users to substitute measured or theoretical values as needed. A new empirical acoustical model is proposed to describe the behaviour of Shoddy-based fibre absorbers. The model requires knowledge of the bulk density only. This parameter is easily measured making application of the model elementary.

Ce mémoire présente l'étude d'un système de trigénération ou CCHP (combined cooling, heating and power) au gaz naturel fournissant les besoins thermiques à un immeuble à logement typique de 13 étages. Un groupe électrogène permettant de générer de l'électricité tout en fournissant de l'eau chaude assure le chauffage du bâtiment et de l'eau chaude sanitaire tandis qu'une machine à absorption ammoniac-eau permet de transformer cette chaleur en puissance frigorifique afin de climatiser le bâtiment en période chaude. Le modèle de cette machine à absorption a spécialement été conçu pour fonctionner avec une température de désorption basse et permet d'évaluer les performances de telles machines dans des situations hors design. Ce modèle a été vérifié à l'aide d'autres modèles avérés et propose des coefficients de performance de l'ordre de 0,7 à 0,8. L'évaluation des performances horaires de ce système est présentée pour une année standard sous trois climats canadiens. Ces performances sont ensuite comparées aux différentes méthodes employées pour assurer ces besoins ainsi que pour générer l'électricité supplémentaire. Les résultats démontrent d'excellentes performances même lorsque comparées aux plus récentes technologies. Les résultats passent d'une légère augmentation de la consommation de gaz naturel dans un climat chaud comme Toronto à une importante réduction dans un climat froid comme Edmonton. Ce système permet donc de combler les besoins thermiques d'un bâtiment tout en générant de l'électricité dans des périodes de pointes annuelles et possiblement journalières. Plusieurs études dont notamment l'optimisation du système et le couplage à d'autres types de bâtiment permettrait cependant de mieux évaluer les possibilités complètes de la trigénération.

Dissertation made openly available per email from author, 6/8/2016.

published_or_final_version / Mechanical Engineering / Doctoral / Doctor of Philosophy

For understanding the mechanisms of low level laser/light therapy (LLLT), accurate knowledge of light interaction with tissue is necessary. In order to have a successful therapy, laser energy needs to be delivered effectively to the target location which depending on the application can be within various layers of skin or deeper. The energy deposition is controlled by input parameters such as wavelength, beam profile and laser power, which should be selected appropriately. This thesis reports a numerical study that investigates the laser penetration through the human skin and also provides a scale for selection of wavelength, beam profile and laser power for therapeutic applications. First, human skin is modeled as a three-layer participating medium, namely epidermis, dermis, and subcutaneous, where its geometrical and optical properties were obtained from the literature. Both refraction and reflection are taken into account at the boundaries according to Snell’s law and Fresnel relations. Then, a three dimensional multi-layer reduced-variance Monte Carlo tool was implemented to simulate the laser penetration and absorption through the skin. Local profiles of light penetration and volumetric absorption densities were simulated for uniform as well as Gaussian profile beams with different spreads at 155 mW average power over the spectral range from 1000 nm to 1900 nm. The results showed that lasers within this wavelength range could be used to effectively and safely deliver energy to specific skin layers as well as to achieve large penetration depths for treating deep tissues, without causing any skin damage. In addition, by changing the beam profile from uniform to Gaussian, the local volumetric dosage could be increased as much as three times for otherwise similar lasers. In the second part of this thesis, a three-dimensional single-layer reduced-variance inverse Monte Carlo method was developed to find the optical properties of the skin using the experimental values of transmittance and reflectance. The results showed that both transmittance and reflectance scale well with transport optical thickness. Moreover, it was also shown that penetration depth is highly sensitive to the laser wavelength and varied within the range from 1.7 mm to 4.5 mm. / text

<p>The objectives of this work were (i) to develop a new methodfor the preparation of thin cellulose model films, (ii) to usethese model films for swelling measurements and (iii) to relatethe swelling of fibers and films to the dry strength ofpaper.</p><p>In the new film preparation method, NMMO(N-methylmorpholine-N-oxide) was used to dissolve cellulose andDMSO (dimethyl sulfoxide) was added to control the viscosity ofthe cellulose solution. A dilute solution of the cellulose wasspin-coated onto a silicon oxide wafer and the cellulose filmthus prepared was then precipitated in deionised water. Asaturated layer of glyoxalated-polyacrylamide was used toanchor the film onto the silicon oxide wafer. This proceduregave films with thicknesses in the range of 20-270 nm. Thefilms were cleaned in deionised water and were found by ESCAanalysis and contact angle measurements (θ<20°)to be free from solvents. Solid state NMR measurements onfibers spun from NMMO also indicated that the model filmconsisted of about 50% crystalline material and that thecrystalline structure was of the cellulose II type.Determination of the molecular weight distribution of thecellulose surface material showed that the NMMO treatmentcaused only a minor breakdown of the cellulose chains and thatlow molecular mass oligomers of glucose were not created.</p><p>It was further shown that atomic force microscopy (AFM)measurements could be used to determine the thicknessof thecellulose films, in both the dry and wet states. The thicknesswas determined as the height difference between the top surfaceand the underlying silica wafer measured at a position where anincision had been made in the cellulose film. The cellulosesolutions were also directly spin-coated onto the crystal usedin the Quartz crystal microbalance (QCM-D), pre-treated withthe same type of anchoring polymer. With this application,these model surfaces were shown to be suitable for swellingmeasurements with the QCM-D. The extent of swelling and theswelling kinetics in the presence of electrolytes, such asNaCl, CaCl2 and Na2SO4, and at different pH were measured inthis way. The films were found to be very stable during thesemeasurements and the results were comparable to the swellingresults obtained for the corresponding pulps. The swelling ofboth fibers and films followed the general behavior ofpolyelectrolyte gels in the presence of electrolytes and was inaccordance with the Donnan equilibrium theory. The films havebeen shown to differ from fibers with regard to the absence ofa covalent interior network. This influences the evaluation ofthe deswelling effects measured on the model films. Theswelling effect seen with different electrolytes has also beenconsidered in relation to the tensile strength of paperprepared from a kraftliner-pulp. In this study, it was foundthat there was no direct relationship between the swelling ofthe fibers, measured as WRV, and the strength of the paper inthe presence of different electrolytes at pH 5.</p><p><b>KEYWORDS:</b>absorption, carboxymethyl cellulose,cellulose, cellulose fibers, dissolving pulps, donnanequilibrium, electrolytes, film, ion exchange, ionization,kinetics, liner boards, microscopy, spinning, surfaces,swelling, tensile strength, water, water retention value.</p>

Chitosan has been proposed as a "natural" coagulating agent to solve wastewater problems. The main hindrance in this commercial chitosan application has been its low cost effectiveness. The hypothesis in our research is that chitosan complexes with natural polyanions is more effective than chitosan alone, particularly in recovering low concentration proteins from food processing wastewater. Chitosan (Chi) was reacted with alginate (Alg), pectin (Pec) and carrageenan (Car) ex-situ to obtain chitosan-polyanion complexes (Chi-Pol). Analysis by Fourier Transform Infrared (FTER) spectroscopy confirmed electrostatic interactions as the mechanism for complex formation. Scanning electron microscopy revealed a tight, non-porous structure except for the porous Chi-Car complex. Tests with a bovine serum albumin solution revealed low adsorption rates with slightly higher values for Chi-Car suggesting the need for an improved complexation method. Chi-Pol complexes prepared in-situ at different monomeric weight ratios (MR) were evaluated using pH 6 adjusted Cheddar cheese whey and surimi wastewater (SWW). Complexes used at 30 mg complex/L whey showed higher turbidity reductions than at 10 mg/L. MR had no significant effect on turbidity reduction except for Chi-Alg at 30 mg/L; the value (72 %) at MR = 0.2 was higher than for MR = 0.8. UV-Vis spectroscopy confirmed in-situ complex formation with a preference for the adsorption of specific whey protein fractions. Complexes formed at 0.2 and 0.8 MR were evaluated at two concentrations for the treatment of SWW. Tests at 50 mg/L showed a turbidity reduction of up to 97 % at 24 h with a 81-90 % recovery of SWW proteins. At 150 mg/L, similar efficacy was achieved in only 1 h with turbidity reductions ranging 94-99 % and 78-94 % protein recovery. FTIR analyses confirmed the adsorption of proteins as indicated by similarities in the three amide bands for Chi-Alg recovered solids and untreated SWW. Differential scanning calorimetry (DSC) was employed to study interactions of SWW proteins and Chi-Alg complexes. Untreated and complex bound SWW proteins revealed single exothermic peaks at 23.3 and 38.0°C, respectively. This suggested Chi-Alg and SWW protein interactions increased the thermal stability of SWW proteins. However, further thermal analysis studies are needed to confirm this finding. / Graduation date: 2001