Frothers are non ionic surfactants, commonly alcohols and polyglycols, used to provide two functions in flotation, namely to reduce bubble size and stabilize froth. Both functions imply an impact on the properties of the air-water interface (bubble surface) but there is no consensus on the mechanism, particularly with regard to bubble size reduction. Blending frothers is becoming common in flotation practice arguably enhancing performance by permitting independent control of the two frother functions. However, there have been no studies to determine this blend possible action. The blends used here focused on a small addition of polyglycol (F150 and DF250) to alcohols (1-pentanol and MIBC). The effect of blends on gas dispersion properties (bubble size and gas holdup) and froth properties (froth height and water overflow rate) was measured in three units, bubble column, 800 L and 5.5 L mechanical cells. Froth height and water overflow rate showed a strong blend effect, both increasing significantly compared to individual frothers. However, while the bubble size was decreased at blend concentration below the critical coalescence concentration (CCC) of the alcohol frother, bubble size was significantly larger above the alcohol CCC. Gas holdup data supported these effects on bubble size. This bubble size effect compromised testing the hypothesis of independent function control using blends. The thesis focussed on explaining the bubble size observations by designing coalescence and beak-up experiments. Coalescence time of bubbles generated from two horizontal capillaries did not show a blend effect. Break-up tests for one-bubble-at-a-time and from an air stream conducted for F150 – pentanol blends showed that the blend reduced bubble break-up compared to single frothers. The increase in bubble size above the base CCC therefore appears to due to decreased break-up. A mechanism based on the Marangoni effect is introduced to explain this phenomenon. / Les moussants sont des surfactants non ioniques. Ceux-ci sont communément des alcools et des polyglycols, utilisés en flottation pour assumer deux fonctions : réduire la taille de la bulle et stabiliser la mousse. Ces deux fonctions exercent une influence sur les propriétés de l'interface air-eau (surface des bulles). Cependant il n'existe pas de consensus sur le mécanisme, particulièrement en ce qui a trait à la réduction de la taille des bulles. Mélanger des moussants devient une pratique généralement acceptée en flottation, laquelle sans doute renforce la performance en permettant le contrôle indépendant des deux fonctions de moussants. Toutefois, aucune étude n'a été réalisée pour déterminer l'action possible du mélange des moussants. Le mélange utilisé dans ce travail est un ajout d'une petite quantité de polyglycol (F150 et DF250) aux alcools (1-pentanol et MIBC). L'effet des mélanges sur les propriétés de dispersion de gaz (taille des bulles, rétention de gaz) que sur celles de la mousse (hauteur de la mousse, vitesse du flux du trop plein d'eau) a été mesurée en trois unités, colonne à bulles, 800 L et 5.5 L cellules mécaniques. La hauteur de la mousse et la vitesse du flux du trop plein d'eau ont révélé un fort effet de mélange puisque les deux paramètres accusent dans le cas du mélange une augmentation significative en comparaison aux moussants individuels. Alors que la taille des bulles avait diminué lorsque la concentration du mélange avait atteint une valeur inferieure à la concentration critique de coalescence (CCC) de la mousse d'alcool, la taille de la bulle était devenue largement supérieure à celle d'alcool CCC, contre toute attente. Les données sur la rétention de gaz ont aussi corroboré ces effets affectant la taille des bulles. Cet effet de taille de la bulle compromise tester l'hypothèse de la contrôle indépendante de fonction utilisant des mélanges. Le pivot de cette thèse est l'explication des observations de la taille des bulles par le biais des tests de coalescence et de rupture. Le temps de coalescence de bulles générées à partir de deux capillaires horizontaux n'a pas démontré un effet du mélange. Les tests de rupture réalisés avec une seule bulle à la fois et à partir d'un courant d'air pour le mélange F150 – pentanol, démontre que le mélange a réduit la possibilité de rupture par rapport à un seul moussant. L'augmentation de la taille de la bulle au-delà de la base CCC pourrait être due à la diminution de l'effet de rupture. Afin d'expliquer ce phénomène, un mécanisme basé sur l'effet de Marangoni est introduit.

Two titanium alloys were welded using linear friction welding (LFW) techniques. The two alloys consisted of one α + β alloy, Ti-6Al-4V, and one near-β alloy, Ti-5553. The welding conditions were varied in order to assess the effect of each parameter on the mechanical properties, microstructure, and crystallographic texture of the materials. Axial pressures from 50 MPa to 150 MPa, oscillation frequencies from 30 Hz to 110 Hz, and oscillation amplitudes from 1.5 mm to 3 mm were employed.The linear friction welded (LFWed) samples of Ti-6Al-4V and Ti-5553 were examined using electron backscatter diffraction techniques (EBSD) to relate the texture and phase changes to the thermomechanical conditions. Characterization of the welds included analysis of the microstructural features of the weld region and thermomechanically affected zone (TMAZ) in relation to the parent material. Mechanical properties were evaluated using tensile tests and microhardness measurements.The maximum strains and strain rates experienced by the material during LFW for each set of welding parameters were estimated based on the process parameters and the measured time of oscillation. A heat input equation was developed in order to estimate the temperature at different points in the joint and temperature measurements were made during welding to corroborate the calculated temperatures. The strains, strain rates and temperatures measured and calculated for the welding conditions employed were found to be sufficient to initiate dynamic recrystallization in both alloys. This finding is in agreement with the microstructures and textures observed in the weld centres.The near-β Ti-5553 alloy was examined not only in the as-welded state but also in two post-weld-heat-treated (PWHT) conditions. The TMAZ and weld centre of this alloy were weakened by welding due to the reduction of the α phase volume fraction during rapid cooling from super-transus temperatures in and near the weld. With the restoration via PWHT plus aging of the α phase fraction, the UTS's for the welded samples were restored to the literature values for the heat-treated condition of this alloy.In addition to the two titanium alloys comprising the main focus of the study, a number of other materials were examined in the context of linear friction welding. These were: IMI-834, a near-α titanium alloy (nominally Ti-5.8Al-4Sn-3.5Zr-0.7Nb-0.5Mo-0.3Si); CMSX-4, a single crystal Ni-based superalloy (Ni‑9.5Co-6.4Cr-6.4Ta-6.4W-5.6Al-2.9Re-1Ti-0.6 Mo-0.1Hf wt %); stainless steel 316L; mild steel A42; as well as dissimilar pairs aluminum alloy 6063 to commercial purity Cu and stainless steel 316L to Zr alloy Zr702. All of these materials were successfully linear friction welded after some refinement of the welding parameters. / Deux alliages de titane ont été soudés par friction linéaire (linear friction welding - LFW). Le premier alliage, Ti-6Al-4V, est biphasé α + β, alors que le second, Ti-5553, est principalement constitué d'une phase bêta. Différentes conditions de soudage (pression axiale variant de 50 à 150 MPa, fréquence d'oscillation entre 30 et 11 Hz et amplitude d'oscillation entre 1.5 et 3 mm) ont été utilisées afin d'étudier leur effet sur les propriétés mécaniques, la microstructure ainsi que la texture des deux alliages. Les cordons de soudure ainsi que les régions adjacentes ont été analysés par microscopie électronique, plus particulièrement au moyen de la technique de diffraction des électrons rétrodiffusés, afin de mieux comprendre l'effet qu'ont les paramètres de soudage sur la texture et la transformation de phase cristallographique. Des essais de microdureté et traction ont aussi été effectués afin de déterminer les propriétés mécaniques des soudures. Pour chaque combinaison de paramètres de soudage, les déformations et vitesses de déformation maximales obtenues durant la soudure par friction linéaire ont été estimées à partir des conditions de soudage et des temps d'oscillation. De plus, une équation paramétrique à été développée afin d'estimer la température aux différents endroits du cordon de soudure. La validité de cette équation à été confirmée à partir de mesures de la température effectuées lors de la soudure. Les déformations, les vitesses de déformation ainsi que les températures générées durant la soudure sont suffisamment élevées pour que la microstructure recristallise dynamiquement. L'analyse des microstructures et de la texture a en effet révélé des traces évidentes de recristallisation dynamique dans les deux alliages. Les zones centrales et TMAZ de la soudure sur l'alliage Ti-5553 sont affaiblies par la diminution du volume de la phase α qui se produit lors du refroidissement rapide suivant le soudage. L'application de traitements thermiques après le soudage permet la restauration complète de la microstructure. Finalement, des essais de soudure par friction linéaire ont été faits sur plusieurs autres matériaux, dont IMI-834 (Ti-5.8Al-4Sn-3.5Zr-0.7Nb-0.5Mo-0.3Si), mono cristal CMSX-4, (Ni‑9.5Co-6.4Cr-6.4Ta-6.4W-5.6Al-2.9Re-1Ti-0.6 Mo-0.1Hf wt %), l'acier austénitique 316L, l'acier doux A42, etc.

During the last 20 years, ceramic fiber reinforced metal matrix composites, referred to as MMCs, have provided a relatively new way of strengthening metals. High specific modulus and a good fatigue resistance in dynamic loading conditions or for high temperature applications make these composites very attractive for replacing classic alloys. The first requirement for the fabrication of MMCs, especially by processes involving liquid metals, is a certain degree of wetting of fibers by the liquid metal which will permit a good bonding between the two phases. / The conventional experimental approach to wettability consists of measuring the contact angle of a drop of the liquid metal resting on flat substrate of the ceramic reinforcement materials. / This work deals with the fabrication of eutectic $ rm ZrO sb2/Al sb2O sb3 (ZA), ZrO sb2/Al sb2O sb3/TiO sb2$ (ZAT), and $ rm ZrO sb2/Al sb2O sb3/SiO sb2$ (ZAS) ceramic substrates and the study of their wetting behavior by different classes of Al alloys. Wetting experiments were performed under high vacuum or under ultra high purity Ar atmosphere. Four major variables were tested to study the wetting behavior of the different ceramic/metal systems. Variables include holding time, melt temperature, alloy and ceramic compositions. / Ceramic materials were sintered under vacuum at temperatures ranging from 1500$ sp circ$C to 1790$ sp circ$C for 2.5 hours, and achieved over 96% of the theoretical density. An experimental set-up was designed to measure in-situ contact angles using the sessile drop method. For any ceramic substrate, a temperature over 950$ sp circ$C was necessary to observe an equilibrium wetting angle less than 90$ sp circ$ with pure Al; by alloying the aluminum, wetting could be observed at lower temperatures ($ theta$ = 76-86$ sp circ$ at 900$ sp circ$C for Al-10wt%Si, $ theta sim72 sp circ$ at 850$ sp circ$C for Al-2.4wt%Mg). Finally, ZAS specimens reacted with molten Al alloys over 900$ sp circ$C to produce Zr-Al based intermetallics at the metal/ceramic interface.

Recently Al was replaced by Cu as an interconnecting material. The primary objective of the present research was to investigate the mechanism of texture and microstructure evolution and to study its influence on electromigration in damascene Cu interconnects. For this purpose electromigration experiments were performed on the Cu interconnects in vacuum in the SEM. The in-situ electron back-scatter diffraction (EBSD) investigation of Cu interconnect lines before and after the electromigration failure helped to identify the orientations which were associated with electromigration defect nucleation. A mechanism was proposed which shows the correlation between the texture and electromigration damage formation. / The second part of the study involved investigation of texture and microstructure in damascene Cu interconnects. The experiments were designed so as to allow an analysis of both pre and post-CMP (chemical mechanical polishing) annealing on texture and microstructure evolution not only as a function of line width but also line spacing. In order to investigate the mechanism of texture evolution in the damascene Cu interconnects in-situ EBSD study of Cu films was carried out during their thermal treatment in SEM. A similar separate experiment carried out on freestanding Cu film helped to ascertain the role of substrate in inducing the texture transformation at high temperature. Some models were proposed which establish the combined effect of dislocation activity and substrate on texture evolution. Both EBSD and X-ray diffraction (XRD) methods were used to measure the texture. Finally the residual stresses in the damascene lines were measured using XRD to evaluate their impact on the mechanical reliability of the chip interconnect system. / The EBSD and XRD studies helped to identify the mechanisms which govern the texture and microstructure evolution in Cu interconnect lines during annealing performed before and after CMP. The role of surface/interface and strain energy on formation of energy minimizing textures was established. The mechanism of texture evolution was explained explicitly as function of principal and shear stress, dislocation density, trench aspect ratio and top passivation layer. / Thus, from the electromigration studies it was possible to screen out the undesirable grain orientations. The XRD and EBSD investigations revealed the mechanism of texture and microstructure evolution in damascene lines and were used to explain the impact of various factors like stresses, dislocations, line width, line spacing, trench aspect ratio, passivation layer, substrate on texture evolution. The knowledge gained from these two studies opens a new door giving an opportunity to design the texture and microstructure in the damascene interconnect lines so as to eliminate undesirable grain orientations which are vulnerable to electromigration and mechanical failure. Since the mechanism of texture evolution is now known, one could vary the process parameters like say current density, barrier and passivation layer material and thickness, Cu seed and electroplated layer thickness, additive content in electroplating bath, annealing conditions etc. to obtain an optimum texture and microstructure which provides best reliability against electromigration and mechanical failures.

Brazing is a low-cost process used extensively in manufacturing and repair of component assemblies in the aerospace industry. For stainless steel joints requiring an airtight seal, the brazing metal Pd-Ni-Cr-Si-B (PalNiCro) is commonly applied. However the underlying mechanisms of the brazing process are not well understood, which has precluded optimization of the brazing parameters and reduced the cost-effectiveness of the production process. To address these issues, the role of processing conditions on the wetting characteristics of austenitic AISI 347 and martensitic AISI 410 stainless steel was studied in this work using PalNiCro as the brazing alloy. Sessile drop tests were conducted to assess the effect of time, temperature and surface roughness on the wettability conditions between the stainless steel substrate and brazement. The effect of time and clearance on the microstructure and hardness of the brazement was then examined. / A brazing temperature of 980ºC and a surface roughness from 600 grit SiC grinding paper were determined as the optimal wetting conditions. Under such conditions, the value of the equilibrium contact angle for the AISI 347 and AISI 410 stainless steels was determined to be approximately 10º after 100 seconds of melting. Examination of the brazements produced using optimized wetting conditions revealed the formation of borocarbide (Mx(C,B) y) and boride (MxBy) precipitates primarily confined to the grain boundaries and some locations within the interior of the grains, which resulted from the diffusion of boron from the brazing metal. The microstructure of the joint seam consisted of the three distinct regions: (1) a soft ductile nickel solid solution with dissolved palladium and chromium (hardness ~ 220 HV) and (2) palladium silicide (over 600 HV) surrounding (3) very hard Ni-rich islands (over 1000 HV) with the presence of nickel borides. An increase in brazing time showed a slight increase in the maximum brazing clearance (MBC) with also an increase in the diffusion layer for both AISI 347 and AISI 410. The volume fraction of the secondary phases within the joint seam exhibited marked increases up to a clearance of 150 mum with a decrease thereafter and an estimated value of 300 mum as the maximum attainable clearance.

In this work, the crystallization of alpha-calcium sulfate hemihydrate in sulfuric acid solution and the correlation between its properties and preparation conditions by reaction of sulfuric acid with lime (CaO) have been thoroughly investigated. The research involved the study of conversion-dissolution of calcium sulfate dihydrate in H2SO4 solution, the measurement of solubilities, thermo-dynamic calculations and the preparation of alpha-CaSO4 ·0.5 H2O via different methods of reactive mixing of H2SO4 and CaO. It was found that the calcium sulfate solids can saturate the sulfuric acid solutions in only 5 minutes. The solubility of calcium sulfate hemihydrate in 0--3.0M H 2SO4 solution at 100°C was experimentally determined and thermodynamic calculations with the aid of FactSage and OLI have led to establishment of the phase diagram for the CaSO4-H2SO 4-H2O system. An operating window has been determined in terms of H2SO4 concentration, temperature and time within which alpha-hemihydrate can be produced by reaction of lime with H 2SO4. This window is defined as 0.6--1.1M H2SO 4 (steady-state concentrations), 98--105°C and 1 hour retention time. Dihydrate was found to form as intermediate phase quickly converting to hemihydrate. The kinetics of conversion depends on the acidity level. / For the standard preparation procedure of adding lime into hot sulfuric acid, alpha-hemihydrate grows in the c-axis direction much more rapidly than in other directions ending in the form of fine needle crystals. Also, independent of the shape of the seed particles, the resultant crystals of hemihydrate are needle-shaped, which suggests a "dissolution-recrystallization" mechanism. Upon prolonged equilibration in their acid-preparation solution hemihydrate needle-shape crystals become fibrous and eventually convert to anhydrite. It is believed that uptake of SO42- instead of Ca2+ is the rate-determining step in the hemihydrate crystallization process. The hot SO42--rich environment rendered most of the additives (particularly organic) tried ineffective. Trivalent cations such as Fe3+ and Al3+, are the only ones found to modify the crystal morphology from needle-shape to small "grain" type morphology. / Slow addition of H2SO4 solution to slaked lime - reverse procedure was found to favor the production of alpha-hemihydrate with column-shaped as opposed to needle-shaped crystal morphology within otherwise the same operating window, 0.6--1.1M H2SO4. Preliminary assessment of the properties of the alpha-hemihydrate materials synthesized in this work showed them to compare satisfactorily with other materials produced by conversion of dihydrate to hemihydrate.

Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 2006. / Source: Dissertation Abstracts International, Volume: 68-02, Section: B, page: 1240. Adviser: John Abelson. Includes bibliographical references. Available on microfilm from Pro Quest Information and Learning.

Thesis (Ph.D.)--Lehigh University, 2007. / Adviser: Jeffrey M. Rickman.

Thesis (Ph.D.)--Lehigh University, 2007. / Adviser: A. R. Marder.

Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 2007. / Source: Dissertation Abstracts International, Volume: 68-06, Section: B, page: 4065. Adviser: Nancy R. Sottos. Includes bibliographical references (leaves 98-103) Available on microfilm from Pro Quest Information and Learning.