Friction stir processing (FSP), a solid-state post-processing technique that can provide localized modification and control of microstructures in near-surface layers of metallic components is developed based on the principles of friction stir welding (FSW). In this study, FSP was carried out in two phases. In Phase I, FSP was applied to one inch thick, sand casting A206 aluminum and 0.25 inch A367 bending fatigue sample. Optical microscope, SEM and EDS measurements have been performed to reveal a homogenous microstructure in the stirred zone, in which the porosities were quantitatively reduced, the second phase particles have been broken up and distributed in the Al matrix and there were some new components generated. Thermocouples have been applied to acquire the temperature gradient and thermo history during FSP. A range of material properties, including hardness and tensile measurement in A206, and bending fatigue test in A367 were examined. FSP resulted in an increase in hardness of the nugget compared to the hardness profile of the T4 condition. A significant improvement of the ductility of A206 was found after FSP compared to the ductility of T4 A206, but the FSP process slightly reduced the strength of A206. In Phase II, FSP was used for composite fabrication in the top layer of A206 substrate. The discontinuously reinforced aluminum (DRA) which contained 15% weight percent SiC was introduced into one inch thick sand cast A206 substrate. The surface composite layer is well bonded to the aluminum alloy substrate. Defects were not visible validating that FSP is an effective way for composite fabrication in Al cast alloys.

Electron irradiation may be a useful method for treating inorganic cosmetic powders such as talc for undesirable microorganisms if it can be established that no loss of quality or efficacy results as a consequence of the technique. A commercial electron accelerator was used to prepare bulk samples by exposure to electron irradiation under different conditions. A morphological characterisation of the control talc was carried out using microscopy techniques and focused on the particle size, shape and surface features of individual particles. Asperity dimensions vs. Roughness values were compared. Assessment of the bulk properties of the powders as a function of absorbed dose showed no serious impact on the quality or efficacy of the powder for cosmetic applications. The particle size and the level of dust expelled during impact were maintained. Changes to the surface chemistry of the irradiated powders were apparent from an increase in pH and water retention with increasing dose. Improvements to the fragrance retention occurred for both irradiated powders tested. Importantly a satisfactory microorganism level was achieved with the lowest absorbed dose level tested i.e. 5 kGy. Analysis of the crystallography did not show the development of a new phase. Mechanical testing using a slip-peel tester found an increase in the stick-slip behaviour occurred for the powder exposed to the highest absorbed dose only. Atomic force microscopy (AFM) testing of the lift-off force showed an increase in adhesion with increasing dose. Surface roughness increased with dose, while no change in elastic properties was found using nanoindentation suggesting the differentiating factor is due to surface features. Microstructural analysis used the transmission electron microscope (TEM). Voids were observed, which decrease in size, yet increase in number with increasing voltage. Prior to void development a loss of crystalinity is seen using electron diffraction. Explorative data analysis using factor analysis and Independent Component Analysis was performed on the void development data to discover the controlling mechanisms. Two distinct normally distributed populations were identified, each driven by 2-3 critical mechanisms. The distinct behaviour differences of the mechanisms may be utilised for characterisation of more complex properties of crystal microstructures.

Thesis (Ph.D.)--University of Delaware, 2007. / Principal faculty advisor: Eric M. Furst, Dept. of Chemical Engineering. Includes bibliographical references.

Le travail de thµese porte sur l'application de la methode non-lineaire d' homogeneisation dite du second- ordre de Ponte Casta~neda (2002) pour estimer le comportement effectif des materiaux poreux viscoplastique. A titre de rappel, cette methode est basee sur la construction des principes variationnels appropries en utilisant un composite lineaire de comparaison pour produire des evaluations correspondantes µa des milieux poreux non-lineaires. Ainsi, l'objectif principal de ce travail est de proposer un modµele constitutif general qui tient compte de l' evolution de la microstructure, et par consequent, de l'anisotropie induite par l'application de deformations nies au materiau poreux. Le modele est construit pour reproduire exactement le comportement d'un assemblage de sphere composites dans la limite des chargements hydrostatiques, et concide donc avec la limite hydrostatique du critere de Gurson (1977) pour des materiaux poreux plastiques avec des microstructures isotropes. En consequence, ce nouveau modµele ameliore les estimations d'homogeneisation existantes, lesquelles sont satisfaisantes pour de faibles triaxialites mais excessivement raides pour des triaxialites et des non-linearites elevees. En outre, les estimations obtenues par le modµele dependent de la troisieme invariable du tenseur macroscopique des contraintes, lequel porte un effet non negligeable sur la reponse eective du materiau pour de moyennes et hautes triaxialites. De plus, les resultats cites ci-dessus ont ete generalises µa des microstructures anisotropes complexes (par exemple : des microstructures avec des formes et des orientations arbitraires des pores) et a des chargements tridimensionnels, conduisant a la reponse anisotrope globale du materiau poreux. Le modele est ensuite etendu pour tenir compte de l'evolution de la microstructure lorsque le materiau est soumis µa des deformations nies. Enn, la validation du modµele propose a ete realisee par le biais de calculs par elements ¯nis sur des mi-crostructures axisymetriques periodiques, et donnent des resultats pertinents pour l'ensemble des triaxialites et des non-linearites envisagees.

In this study, conductive particles will spray- coated onto the surface of the MPL to function as micro pillar structures that has been proved to be able to significantly increase the performance of a PEMFC. Contrast with the previously used nano imprint technology , the new method is cheap,fast, and especially suitable for the fabrication of large-area electrodes. The results show that a 30% increase in performance can be acquired by using 20um graphite particles. An increase of 57%, max power of 737mW/cm2 , may be achieved with irregular graphite flakes. The understanding of the distribution and the development of the produced water inside the cathode are essential to associate the performance increase with the microstructures. The electrode is dried gradually until a sharp increase in its impedance appeared, which indicates that the surface of the PEM begins to lose water. Then, i-v performance is measured through a cyclic test, i.e. , form a small load current to a large one and, then, buck to the small again. The performance improved with each cycle, because more and more water is produced along the test. A large performance ¡§jump¡¨ appearded at the 4th cycle only for the electrodes with the microstructure that indicates that the major reaction sites have shifted to the location of microstructures. It is concluded that, along with other evidence, a general water ¡§surface¡¨ exists and migrates form PEM towards MPL.

The study focused on analysis of microstructure of sintered TiO2 bulk by using XRD, SEM-EBSD, and TEM. The experiment results showed that sintered samples were rutile phase, besides also found growth twins in sintered sample. The twin plane {101 ̅) was confirmed by SEM-EBSD, TEM-SADP and mathematics transformation matrix. Finally traditional contrast analysis was adopted for determining twin boundary were £\-boundaries and displacement vectors R = 1/n[011].

Abstract Physical chemistry and microstructure of polyimide / clay nanocomposites were investigated in this study. Diffusion of diamine monomers into the clay gallery prior to polymerization was carefully controlled in order to demonstrate the physical constraint during the formation of the nanocomposites. In addition, diamines with different chemical reactivity were employed to show the effect of chemical reaction on the resultant composites. The microstructure and morphology were identified with X-ray diffraction (XRD) and transmission electron microscopy (TEM), while thermal gravimetric analysis (TGA) was performed to demonstrate the thermal stability of the composites. XRD and TEM results indicate that the final properties and microstructure of the PI / clay nanocomposites were profoundly affected by both the diffusion and chemical reactivity of the monomers prior to the formation of polyamic acid and the subsequent thermal imidization had almost no effect. It should be emphasized that diffusion and chemical reactivity of the monomers are highly competitive. An exfoliated or intercalated nanocomposites can be formed only with a low-reactivity diamine and long enough diffusion time; this ensures polymerization was taken place within the clay gallery. On the other hand, a PI / clay blend is formed with insufficient diffusion time and high diamine reactivity. Under this circumstance, nearly all monomers were captured outside the clay gallery and a homo-polyimide was formed independent of clay. These results were consistent with observations from TGA. The maximum degradation temperature (Td) of the nanocomposite increased with increasing diffusion time; Td of the nanocomposites with low-reactivity 1,3-phenylenediamine was 30oC greater than the nanocomposites with high-reactivity 4,4¡¦-oxydianiline.

In this experiment the characteristics and microstructure of ZrN films between Si and Cu or Al were investigated. The ZrN films were deposited by RF magnetron reactive sputtering system. The ZrN films were annealed at various temperatures in order to observe the grain growth and the inter-diffusion of atoms between interfaces. X-ray analysis showed that in the ZrN/Al system the ZrN film has¡]220¡^preferred orientation after annealing at 600¢J for 1 hour. In the Al/ZrN/Si system of the same annealing condition, the ZrN film showed ¡]200¡^preferred orientation with the concurrent formation of the ZrSi2 and AlN phases. In the Cu/ZrN/Si system, the Cu film showed ¡]111¡^preferred orientation after annealing at 700¢J for 1 hour. TEM analysis showed that in the ZrN/Al system there were no Al-Zr compound was found after annealing at 600¢J for 1 hour. In the Al/ZrN/Si system, although ZrSi2¡BAlN and£\-Si3N4 were present after annealing at 600 ¢J for 1.5 hour, but the diffusion barrier is still effective. In the Cu/ZrN/Si system the CuZr2 was formed after annealing at 700¢J for 1 hour, but the diffusion barrier is also still effective. The results up to now suggest that ZrN layer can be a successful candidate as a diffusion barrier between Si and Cu or Al.

In order to understand the mechanical functions of the cardiac muscle it is important to first understand the microstructure of the tissue. Young et al. (1998) realized that quantitative three-dimensional information about the ventricular myocardium is necessary to analyze myocardial mechanics. They developed a technique using confocal fluorescence laser scanning microscopy to obtain three-dimensional images. While this method worked well in rebuilding the myocardial tissue image by image, it was quite extensive and costly. Costa et al. (1999) developed a method that was used to perform three-dimensional reconstruction as well. Their method, while less expensive and much less time consuming, required sheet assumptions and did not look directly at the cross-fiber plane. From Dr. Criscione's previous work on canines (Ashikaga et al., 2004), we found that the sheet structure can be accurately determined from cross-fiber sections without making any sheet assumptions. We have now expanded on those ideas and created a method to perform the quantitative histological investigation of the rat hearts in a way that is both timely and cost effective. We developed a processing method that preserves the orientation of the fiber and sheet angles. This method was carried out using plastic embedding since the dehydration process used in paraffin embedding has a tendency to grossly distort tissue. Once the heart was fixed in formalin, we then removed the septum and sliced it several times vertically. This allowed us to image the tissue at several depths and find an average fiber angle for each slice. Next, the specimen was hardened, and the sheet orientation was evaluated using polarized light. Once both fiber and sheet angles were obtained from several depths within the septum, we then constructed a three-dimension model of the wall. This method was both cost effective and less time consuming than previous ones and will be a method that can be used in the future to compare the myocardial tissue of diseased and healthy rat hearts so that we may better understand the mechanical functions of the heart as it remodels due to disease.

none