Effect of surface treatment for steel adherends on static and fatigue crack growth of epoxy adhesives

The fracture toughness of unmodified, glass-bead-reinforced and CNBR-modified epoxy adhesives under mode I loading is not improved by acid surface treatment of steel adherends since cohesive failure always occurs for all adhesives with or without acid surface treatment. On the other hand, the fatigue crack growth resistance greatly increases due to acid surface treatment of steel adherends. Especially, the threshold dramatically increases. The crack grows cohesively at all stages of crack velocity for DCB specimens treated with acids while it grows at the interface between the adherend and the adhesive layer for the specimens whose polished surface of adherends is only decreased with solvent. An optical microscope observation revealed that adherend surfaces treated with acids were rougher than ones without acid treatment, although XPS examination for the surfaces did not show significant difference in their chemical elements among the specimens with and without acid treatment.     

Stage-I fatigue crack studies in order to validate the dislocation-free zone model of fracture for bulk materials

Stage-I fatigue cracks are commonly described by the model of Bilby, Cottrell and Swinden (BCS model). However, since several experimental investigations have shown a dislocation-free zone (DFZ) in front of crack-tips, it is necessary to validate the new DFZ model and to examine the deviations to the BCS model. Therefore, the dislocation density distribution is derived from height profiles of slip lines in front of stage-I fatigue cracks in CMSX4® single crystals measured by contact-mode atomic force microscopy. This is possible, because the cracks are initiated at notches milled by focused ion beam technique directly on slip planes with a high Schmid factor. Consequently, the directions of the Burgers vectors are well known; it is possible to calculate the dislocation density distributions from the height profiles. The measured distributions are compared to the calculated distribution function of the DFZ model proposed by Chang et al. The additionally measured microscopic friction stress of the dislocations is then used to calculate the influence of grain boundaries on the dislocation density distribution in front of stage-I cracks. The calculation is done by the extended DFZ model of Shiue et al. and compared with the measured distribution function in polycrystalline specimens. Finally, the crack-tip sliding displacement as a measure for the crack propagation rate is compared for the DFZ model and the BCS model with the experimentally revealed values. The important result: the often used BCS model does not reflect the experimental measurements. On the contrary, the DFZ model reflects the measurements at stage-I cracks qualitatively and quantitatively.     

From monoscale to multiscale modeling of fatigue crack growth: Stress and energy density factor

The formalism of the earlier fatigue crack growth models is retained to account for multiscaling of the fatigue process that involves the creation of macrocracks from the accumulation of micro damage.The effects of at least two scales,say micro to macro,must be accounted for.The same data can thus be reinterpreted by the invariancy of the transitional stress intensity factors such that the microcracking and macrocracking data would lie on a straight line.The threshold associated with the sigmoid curve disappears.Scale segmentation is shown to be a necessity for addressing multiscale energy dissipative processes such as fatigue and creep.Path independency and energy release rate are monoscale criteria that can lead to unphysical results,violating the first principles.Application of monoscale failure or fracture criteria to nanomaterials is taking toll at the expense of manufacturing super strength and light materials and structural components.This brief view is offered in the spirit of much needed additional research for the reinforcement of materials by creating nanoscale interfaces with sustainable time in service.The step by step consideraton at the different scales may offer a better understanding of the test data and their limitations with reference to space and time.     

Scatter of fatigue data owing to material microscopic effects

A common phenomenon of fatigue test data reported in the open literature such as S-N curves exhibits the scatter of points for a group of same specimens under the same loading condition.The reason is well known that the microstructure is different from specimen to specimen even in the same group.Specifically,a fatigue failure process is a multi-scale problem so that a fatigue failure model should have the ability to take the microscopic effect into account.A physically-based trans-scale crack model is established and the analytical solution is obtained by coupling the micro-and macro-scale.Obtained is the trans-scale stress intensity factor as well as the trans-scale strain energy density(SED)factor.By taking this trans-scale SEDF as a key controlling parameter for the fatigue crack propagation from micro-to macro-scale,a trans-scale fatigue crack growth model is proposed in this work which can reflect the microscopic effect and scale transition in a fatigue process.The fatigue test data of aluminum alloy LY12 plate specimens is chosen to check the model.Two S-N experimental curves for cyclic stress ratio R=0.02 and R=0.6 are selected.The scattering test data points and two S-N curves for both R=0.02 and R=0.6 are exactly re-produced by application of the proposed model.It is demonstrated that the proposed model is able to reflect the multiscaling effect in a fatigue process.The result also shows that the microscopic effect has a pronounced influence on the fatigue life of specimens.     

A random covering interpretation for the phase transition of the random energy model

The random energy model is related to a random covering of the real line. The phase transition is interpreted as the passage from a regime where a family of random intervals covers the line (high temperature) to a noncovering regime (low temperature).     

A qualitative model for the growth mechanism of silver clusters in polymer solution

A simple and high-reproducible method for the synthesis of polymer-protected silver cluster of controlled size is described. UV-visible spectroscopy has been used for investigating the influence of the aging of the protective poly(vinylpyrrolidone) layer on the cluster growth rate at different reaction temperatures and poly(vinylpyrrolidone)/ethylene glycol weight ratios. The obtained results show that the aging time of the polymeric stabilizer solution plays a fundamental role in the reproducibility of the cluster growth process. A model for the metal cluster formation-grow process is also proposed. Received 18 July 2001 and Received in final form 3 October 2001     

A simple model for the growth kinetics of Fe2B iron boride on pure iron substrate

The present work evaluates the growth kinetics of Fe₂B iron boride forming on iron substrate by means of a diffusion model in the temperature range 1223-1323 K. The model takes into account the effect of the boride incubation time during the formation of Fe₂B phase. The parabolic growth constant at the (Fe₂B/Fe) interface and the mass gain generated by this treatment were estimated. Likewise a simple relationship was proposed to describe the variation of the parabolic growth constant as a function of both the temperature and the boron content in the Fe₂B phase. Furthermore, the simulation results show a good agreement with our experimental results.