Nanocrystalline (NC) and Ultrafine-grained (UFG) metal films exhibit a wide range of enhanced mechanical properties compared to their coarse-grained counterparts. These properties, such as very high strength, primarily arise from the change in the underlying deformation mechanisms. Experimental and simulation studies have shown that because of the small grain size, conventional dislocation plasticity is curtailed in these materials and grain boundary mediated mechanisms become more important. Although the deformation behavior and the underlying mechanisms in these materials have been investigated in depth, relatively little attention has been focused on the inhomogeneous nature of their microstructure (particularly originating from the texture of the film) and its influence on their macroscopic response. Furthermore, the rate dependency of mechanical response in NC/UFG metal films with different textures has not been systematically investigated. The objectives of this dissertation are two-fold. The first objective is to carry out a systematic investigation of the mechanical behavior of NC/UFG thin films with different textures under different loading rates. This includes a novel approach to study the effect of texture-induced plastic anisotropy on mechanical behavior of the films. Efforts are made to correlate the behavior of UFG metal films and the underlying deformation mechanisms. The second objective is to understand the deformation mechanisms of UFG aluminum films using in-situ transmission electron microscopy (TEM) experiments with Automated Crystal Orientation Mapping. This technique enables us to investigate grain rotations in UFG Al films and to monitor the microstructural changes in these films during deformation, thereby revealing detailed information about the deformation mechanisms prevalent in UFG metal films.

In this work, unirradiated and irradiated model body centered cubic (BCC) and face centered cubic (FCC) materials are investigated using advanced electron microscopy techniques to quantitatively measure local stresses and strains around defects, with the overarching goal of obtaining a fundamental understanding of defect physics. Quantitative in-situ transmission electron microscopy (TEM) tensile tests are performed with Molybdenum-alloy nano-fibers, functioning as a model BCC structural material. Local true stress and strain around an active Frank-Read type dislocation source are obtained using quantitative load-displacement data and digital image correlation. A mixed Frank-Read dislocation source, b=a/2[-1-11](112) with a line direction 20° from a screw orientation and length 177 nm, is observed to begin operating at a measured local stress of 1.38 GPa. The measured local true stress values compare very well to estimated stresses using dislocation radius of curvature, and a line-tension model of a large bow-out configuration, with differences of only ~1%. The degree to which the local true stresses can be measured is highly promising. However, the ultimate failure mode of these fibers, sudden strain softening after dislocation starvation and exhaustion, cannot be captured at the typical camera frame rate of 30 frames per second. Thus, fibers are mechanically tested while under observation with the Gatan K2-IS direct electron detector camera, where the frame rate is an order of magnitude larger at 400 fps. Though the increase in frame rate adds to the overall understanding of the sudden failure, by definitively showing that the nano-fibers break rather than strain soften, the failure mechanism still operates too quickly to be observed. In the final investigation of this BCC model structural alloy, the mechanical behavior of heavily dislocated, but unirradiated, and He1+ and Ni2+ irradiated nano-fibers are compared. Remarkable similarities are found in the mechanical data, as the two defect conditions exhibit similar yield strengths, ultimate tensile strengths, and number and size of load-drops. This similarity implies that, even if materials contain dissimilar individual defects, the collective defect behavior can result in similar mechanical properties. Thus, the origin of mechanical properties can be ambiguous and caution should be taken when extrapolating to different size scales. Furthermore, such similarities highlight the importance of in-situ observation during deformation. These experiments provide a key test of theory, by providing a local test of behavior, which is much more stringent than testing behaviors averaged over many regions. Advanced electron microscopy imaging techniques and quantitative in-situ TEM tensile tests are performed with Au thin-film as a model FCC structural material. These investigations highlight the various hurdles experimental studies must overcome in order to probe defect behavior at a fundamental level. Two novelly-applied strain mapping techniques are performed to directly measure the matrix strain around helium bubbles in He1+ implanted Au thin-film. Dark-field inline holography (DFIH) is applied here for the first time to a metal, and nano-beam electron diffraction (NBED) transient strain mapping is shown to be experimentally feasible using the high frame rate Gatan K2 camera. The K2 camera reduces scan times from ~18 minutes to 82 seconds for a 128x256 pixel scan at 400 fps. Both methods measure a peak strain around 10 nm bubbles of 0.7%, correlating to an internal pressure of 580 MPa, or a vacancy to helium ion ratio of 1V:2.4He. Previous studies have relied on determining the appropriate equation of state to relate measured or approximated helium density to internal bubble pressure and thus strain. Direct measurement of the surrounding matrix strain through DFIH and NBED methods effectively bypasses this step, allowing for easier defect interaction modeling as the bubble can be effectively simplified to its matrix strain. Furthermore, this study demonstrates the feasibility of fully strain mapping, in four dimensions, any in-situ TEM experiment. The final set of experiments with this model FCC structural material shows the attempted correlation of defect interactions and deformation behavior at the nano-scale. Experimental comparison of mechanical behavior from quantitative in-situ TEM tensile tests of focused ion beam (FIB) shaped, He1+ implanted, and FIB-shaped He1+ implanted Au thin-film show a wide range of behavior that could not be directly linked to irradiation condition. This is due to the large role that overall microstructural features, such as grain boundary orientation and texture, play in mechanical behavior at this size scale. However, these tests are some of the first to in-situ TEM mechanically strain single grain-boundaries free of FIB-damage. It is expected that, with well-defined grain orientations and boundaries, real conclusions can be made.

Mechanical behavior in thin films continues to be a growing field of interest in the materials research community. This behavior can critically influence the design, performance and reliability of thin-film structures used in every area of thin-film technology. Examples of affected areas include semiconductor and magnetic recording technology, as well as protective and hard-coating technology. As a result, it is important to study fundamental issues involved in film-substrate adhesion, the development of intrinsic stresses and the mechanisms of plastic deformation, strain relaxation, and fracture in thin films. This book addresses issues towards improving existing, as well as developing new, mechanical property characterization techniques such as more sensitive ultrasonic methods for elastic behavior determination and low-load indentation methods to investigate yield, creep, and fracture behavior. Experimental, theoretical and modelling work is presented. Topics include: novel testing methods; low-load indentation; metallization and reliability; structural and mechanical stability; surface and tribological properties; adhesion; deformation mechanisms; stresses in thin films - generation mechanisms and measurement techniques; multilayered and superlattice thin films and structure/property/processing relationships.

Measurement of localized strain states on the nanoscale is of a field of interest in materials science because the information is vital to futhering the understanding of the processes of microstructural development. There are a variety of existing techniques that are capable of nano-scale strain measurement to one degree or another, however, in order to use this information in real time to inform the visualization of microstructural evolution processes such as boundary migration, an ideal technique would be one that could be carried out in a transmission electron microscope (TEM). Although techniques such as geometric phase analysis have been developed to do just that, they have limitations for some applications. For that reason, there is a demand for a new technique for the measurement of localized strain that can be carried out automaticalIy in TEM and used on a variety of sample types ranging from single crystal thin films to bulk polycrystalline materials. In this study, a technique is proposed to measure local misorientation gradients in TEM as an approximation of strain in the form of gwmetrically necessary dislocation density that is analogous to similar approaches used to treat electron backscatter diffraction data. The utility of the proposed technique is tested through its implementation in conjunction with an in situ TEM heating study on grain boundary migration and twin formation in copper. However, the mechanisms that result in the GBE microstructure are still unclear. In this study oxygen-free electronic copper was characterized in TEM both ex situ and during in situ annealing experiments in order to local strain effects on twin boundary generation and migration. The proposed strain mapping technique was used to measure local orientation gradients before and after processing. The technique was validated by comparison with visible defects as well as a comparison with geometric phase analysis (GPA). The spatial resolution of the strain maps produced was an order of magnitude larger than for GPA, showing strain values averaged across the step area, due to the spot size of the instrument used.