Nucleation and bluntness of nanocracks were studied through in situ tensile tests for thin crystals of 310 stainless steel by transmission electron microscopy (TEM). A dislocation free zone (DFZ) could form after the dislocation emission had just ceased. The DFZ is an elastic zone so that the local stress near the crack tip in the DFZ is possibly up to the cohesive strength, because of which a nanocrack could initiate in the DFZ or at the crack tip. The nanocrack in the DFZ or at the crack tip would blunt into a void or a notch through the increment and movement of dislocations in the plastic zone even when keeping constant displacement. If constant displacement was kept for a long time, nanovoids could initiate in the DFZ through diffusion and enrichment of supersaturation vacancies. The connection of the nanovoids would result in the initiation of nanocracks.
The thermoelastic transformation of the CuZnAl shape memory alloy has been studied in situ and in real time by synchrotron X-ray topography. The white nature of the beam leads access to crystallographic relations between crystals in β and 9R phases. From our original method deduced from the Kajiwara theory the experimental and simulated orientations have been compared and the self-accommodating character of the transformation has been shown.
Polymer composites with 3D woven graphite fiber reinforcement (3D interlock weaves) have been tested in compression-compression fatigue under load control. As under monotonic loading, the principal mechanism of failure is kink band formation in the primary load bearing tows. Observations of kink bands and microcracking in sectioned specimens suggest that fatigue progresses by the accumulation of damage to the resin within individual tows. It is conjectured that resin damage leads to failure by lowering the critical stress for kink band formation on a single cycle. If resin damage is assumed to accumulate at a rate proportional to some power of the local axial shear stress in a misaligned tow, then a simple formula follows for the cycles to kink band formation. Under load control, only a few kink bands are required for specimen failure. Then the formula is also the basis for estimates of fatigue life. Fatigue life data and measured misalignment angles, which determine the local axial shear stress, support the fatigue model.
The influence of fluorine addition on the grain/phase boundary structures and their viscous behavior at high temperature were systematically investigated in Si3N4/SiC composites. As a reference, a simple system densified by hot isostatic pressing (HIP) and containing only SiO2at the boundaries was selected for this basic investigation. In addition, increasing amounts of F dopant were incorporated into the composite bodies by adding Teflon during the mixing procedure of the raw powders and then pre-firing the mixture under high vacuum at 1200°C. Analytical transmission electron microscopy showed that fluorine remained localized at the grain boundary films and triple points, constituting an amount up to a few percent by weight of the intergranular glassy-SiO2phase. Detailed structural characterizations of both grain and phase boundaries were performed by using high-resolution electron microscopy (HREM) and atomic force microscopy (AFM). The high-temperature mechanical behavior of the undoped and F-doped SiO2phases was characterized by both measurements of torsional creep rate and variation of internal friction at temperatures up to 1600°C. F-doped materials showed creep rates several orders of magnitude higher compared to the undoped sample and damping temperature curves markedly shifted to lower temperature values. According to the above set of microstructural and mechanical data, the inherent viscosity of the SiO2intergranular phase could be quantitatively evaluated and the viscous-sliding mechanism under stress modeled.
The coalesced γ″ precipitate in the aged nickel-base superalloy has been investigated using a high resolution electron microscopy. The encounter of γ″ precipitates during aging formed the coalesced precipitate that consists of several domains which keep the original orientation before encountering. The thickness of boundary in the coalesced γ″ precipitate was in the range of zero to a few atom layers.
The logarithmic decrement, δ, was measured in Nb-V-O alloys with the V content of 0.5, 1.0, 1.9, 7, 12, 20 and 50 at.% using a low frequency (f= 3-7 Hz) inverse torsion pendulum in the temperature range from 20 to 800°C. The complex oxygen atom Snoek peaks were analysed with the help of a computer and individual Debye form constituent peaks-jwere extracted. The parameters of these peaks were used to derive information of two sorts: (a) distribution of oxygen atoms over octahedral interstices which differ in the number jof V atoms in the nearest neighbour lattice sites (static parameter) and (b) the potential barriers for oxygen atoms diffusing from j-positions (dynamic parameter). The first was used to evaluate the binding energies of oxygen with vanadium atoms in jV-O complexes, the second made understandable the structure of potential barriers for diffusion of the oxygen atoms from positions j. The differences between measurements of static and dynamic characteristics are discussed.
During isothermal pearlitic transformation of high carbon austenite, the formation of proeutectoid phases precedes that of pearlite. Widmanstatten cementite is the major proeutectoid phase that forms at all temperatures. Within this constituent the precipitation of ε-Cu always occurs, together with the formation of stabilized coarse austenite entrapments. Within pearlitic cementite interphase precipitation of ε-Cu occurs at all transformation temperatures, while within pearlitic ferrite a banded interphase precipitation of ε-Cu occurs only at temperatures close to the eutectoid point. As the temperature of the isothermal pearlitic transformation is lowered the formation of non-partitioned pearlitic ferrite is possible and ε-Cu precipitation only forms within this ferrite as a result of ageing occurring during prolonged isothermal holding. It was concluded that interphase precipitation of ε-Cu always occurred on the migrating austenite/cementite interphase boundaries, but ε-Cu interphase precipitation occurred on the migrating austenite/ferrite boundaries only at temperatures close to the eutectoid point.
An experimental study has been made of a laminate consisting of monolithic thin alumina plates alternating with unidirectional carbon/epoxy (C/E) prepreg tapes. The main advantages of this system over the traditional means of reinforcing ceramics, are the avoidance of large flaws due to processing, which occur in fiber reinforced brittle matrix composites, and the nearly isotropic behavior under biaxial loading. In addition, the multiple fracture mechanism occurring in the system gives rise to pseudo ductile behavior and enhanced strain energy dissipation. The mechanical behavior of the laminate is explored. The effects of the number of layers, volume fraction and transverse properties are also investigated. The loss of stiffness with increase of the applied strain is estimated using a simple shear lag theory, which includes the plastic behavior of the interface.
The hardening rate of the matrix mean stress in planar random fibre aluminium composites with a range of fibre volume fractions is measured in cyclic Bauschinger experiments at room temperature and 77 K. A particular problem with predicting the mean stress hardening rate in these composites is that the plastic strain is not isotropic in the transverse directions. A method for quantifying the experimentally determined plastic flow in a form useful for this analysis is proposed. The difference between the predicted and measured mean stress hardening rates is smallest for low fibre volume fractions at 77 K. A model for relaxation of the plastic mean stress in the matrix is developed. It is based on the assumption that the rate controlling mechanism in these composites is cross slip of screw dislocations. The measured activation energy is independent of fibre volume fraction but the activation volume decreases with increasing fibre content. The magnitudes of activation energy and activation volume support the assumptions of the model.