Interfacial reaction mode and its influence on tensile strength in laser joining Al alloy to Ti alloy

Abstract

In order to obtain a robust dissimilar joint of Al/Ti alloys, a filler wire of Al–12 wt-%Si and 45° V shape groove on the base metal were used during CO₂ laser welding. Heat input had evident influence on the interfacial reaction mode. It was found that a dissolution mode for low heat input and a melting mode for high heat input exist at the joint interface, which was analysed from thermodynamic point of view. Tensile strength of the joints in the dissolution mode reached as high as 296 MPa, which was significantly higher than that in the melting mode.     

Unsteady interfacial phenomena during inward weld pool flow with an active surface oxide

Abstract

Particle image velocimetry (PIV) was applied to estimate the velocity field on a weld pool surface with an oxide layer. A positive surface tension gradient drives an inward flow pattern from the boundary to centre of the weld pool, resulting in particles collecting on the centre line of the pool at the surface. Unsteady flow motions were observed by experimental visualisation of the particle flow, and the computed velocity field shows strong unsteady interfacial movement during inward flow. These unsteady flow motions are related to clockwise and counterclockwise vortices on the weld pool surface, which in turn affect both mass flow and energy transport. Both the qualitative flow motion and quantitative flow velocity are described in this work, which contributes to explaining the characteristic unsteady fluid flow in the weld pool, the measurements also provide experimental data for validation of numerical fluid flow models of the weld pool.     

Effect of surface oxygen content and roughness on interfacial adhesion in carbon fiber–polycarbonate composites

The effect of surface chemistry and rugosity on the interfacial adhesion between Bisphenol-A Polycarbonate and a carbon fiber surface subjected to surface treatment to add surface oxygen groups was investigated. The surface oxygen content of PAN based intermediate modulus IM7 carbon fibers was varied by an oxidative surface treatment. The oxygen content of the carbon fiber surface increased from 4 to 22% by changing the degree of surface treatment from 0 to 400% of nominal commercial surface treatment levels. The oxidative surface treatment also causes an increase in surface roughness by creating pores and fissures in the surface by removing carbon from the regions between the graphite crystallites. To decouple the effects of surface roughness and the surface oxides on the interfacial adhesion, the oxidized fiber surface was passivated via hydrogenation at elevated temperature. Thermal hydrogenation removes the oxides on the surface without significantly altering the surface topography. The results of interfacial adhesion tests indicate that an increase in the oxygen content of the fiber does not increase the fiber-matrix interfacial adhesion significantly. Comparing adhesion results between oxidized and hydrogen passivated fibers shows that the effect of the surface roughness on the interfacial adhesion is also insignificant. Overall, dispersive interactions alone appear to be the primary factor in adhesion of carbon fibers to thermoplastic matrices in composites.     

Peak force as function of the embedded length in pull-out and microbond tests: effect of specimen geometry

We have derived the equations which explicitly express the peak force, F _max, and the apparent interfacial shear strength, τ _app, measured in the pull-out and microbond tests, as functions of the embedded length. Three types of test geometries were considered: (1) a fiber embedded in a cylindrical block of the matrix material; (2) microbond test with spherical matrix droplets; and (3) pull-out test in which the matrix droplet had the shape of a hemisphere. Our equations include the local interfacial shear strength (IFSS), τ _d, and the frictional interfacial stress, τ _f, as parameters; the effect of specimen geometry appeared in the form of dependency of the effective fiber volume fraction on the embedded length. The values of τ _d and τ _f were determined by fitting our theoretical curves to experimental F _max (l _e) plots by using the least squares method. Our analysis showed how the local IFSS and the frictional interfacial stress affected the measured F _max and τ _app values. In particular, it was revealed that intervals of embedded lengths could exist in which frictional interfacial stress had no effect on F _max and τ _app, even if the τ _f value was high. We also derived an equation relating the scatter in the interfacial strength parameters (τ _d and τ _f) to the scatter in τ _app, which is experimentally measurable, and proposed a procedure to determine the standard deviations of τ _d and τ _f from experimental pull-out and/or microbond test data.     

A novel mathematical procedure to evaluate the effects of surface topography on the interfacial state of stress. Part I: Verification of the method for flat surfaces

A mathematical procedure is developed to utilize the complementary energy method, by minimization, in order to obtain an approximate analytical solution to the 3D stress distributions in bonded interfaces of dissimilar materials. The stress solutions obtained predict the stress jumps at the interfaces, which cannot be captured by current FEA methods. As a novel method, the penalty function is used to enforce the displacement boundary conditions at the interfaces. Furthermore, the mathematical procedure developed enables the integration of different interfacial topographies into the solution procedure. In order to incorporate the effects of surface topography, the interface is expressed as a general surface in Cartesian coordinates, i.e. F (x, y, z) = 0. In this paper, the flat interface problem, i.e. y = 0 surface is considered for verification of the method by comparison with the FEA method. A comparison of the results reveals our new mathematical procedure to be a promising and efficient method for optimizing interface topographies.     

Influence of the interfacial resin layer on delamination initiation in broken ply composite laminates

The present study has investigated the influence of a resin layer on the delamination initiation at the interface of broken and continuous plies in the case of GR/E (graphite/epoxy) laminates with broken central plies. A full three-dimensional (3D) finite element (FE) analysis was performed with each layer of the laminate modelled as homogeneous and orthotropic. The interface between the broken and the continuous plies was modelled with a thin resin-rich layer. Eight-noded isoparametric layered elements were used to model the laminate specimen. Also, 3D contact elements were used to prevent inter-penetration of the delaminated faces at the interface. Based on the results of the 3D FE analysis, strain energy release rates were calculated at the delamination front using Irwin's 'crack closure integral'. Using the concepts of linear elastic fracture mechanics (LEFM), the strain energy release rate was used as a parameter for assessing delamination initiation. The effects of various factors such as resin layer stiffness, resin layer thickness, and fibre orientation at the interface on the three components of the strain energy release rates, namely G_I, G_II and G_III, were studied for laminates with various crack sizes of the broken ply, and the influence of the resin layer in the delamination initiation was established. It was observed that delamination initiation is a mixed-mode phenomenon even in the case of uniaxial loading and the dominance of the mode of delamination is governed by the resin layer stiffness, thickness, and lamina orientation at the interface. The present work also concludes that an increase in the resin layer modulus leads to an increase in the probability of mode I delamination while the probability of mode II delamination decreases. A 0/90 interface exhibits a higher chance of delamination in modes I and II, while mode III delamination is maximum for 0/30 and 0/60 fibre orientation interfaces. It was also observed that the larger the crack width, the greater the probability of delamination initiation at the interface.     

A new understanding on the mechanism of fountain solution in the prevention of ink transfer to the non-image area in conventional offset lithography

In conventional offset lithographic printing, it has been well established that the existence of a continuous layer of fountain solution (FS) on the surface of the non-image area is an essential condition to ensure correct operation of lithography. However, the mechanistic function of FS in preventing the ink from being transferred onto the non-image area has not been fully understood. Several major mechanistic interpretations can be found in the literature, which are based either on comparing of static works of adhesion and cohesion of ink and FS, or on the splitting of the 'weaker' FS layer. Although the latter becomes more accepted, direct experimental evidence is difficult to find in the literature. On the other hand, confusing information found in the literature showed that the ink-transfer (or non-transfer) observations reported in many case studies correlate well with simple comparisons of works of adhesion, cohesion and spreading data of ink/FS, ink/plate and FS/plate obtained under the static condition. These results, therefore, imply that, in explaining the function of FS in preventing ink transfer to the non-image area, the ink/FS interfacial adhesion failure would be the dominant mechanism. The work presented in this study covered two specific areas in order to address and better understand the responses of ink and FS layers and their interface to forces encountered during ink transfer. Firstly, an analysis of lithographic plates contaminated with a cationic polymer revealed that the violation of the ink non-transfer condition of the plate non-image area due to contamination could be predicted by traditional criteria of plate wetting and works of adhesion and cohesion. However, these traditional criteria cannot reliably predict the non-transfer condition of the ink on the clean non-image area that was covered by FS. Secondly, in some novel experiments conducted in this study using ice or Teflon as a substrate, the works of adhesion and cohesion were not able to predict ink transfer in most cases. Direct experimental evidence from this work revealed that splitting of the FS layer was involved in the prevention of ink transfer to the non-image areas, and that the thickness of the FS layer was critical in allowing the splitting to occur.     

Modified shear lag model for fibers and fillers with irregular cross-sectional shapes

For fibers with irregular cross sections such as ultrahigh modulus polyethylene (UHMPE) fibers and ribbon-like carbon fibers, the original shear lag model would not provide accurate calculations for interfacial shear stress because it assumes a circular fiber cross section. In this study, a modified shear lag model is proposed to calculate the interfacial shear stress that reflects the change of fiber cross-sectional shape. Microbond test on a UHMPE fiber/epoxy system was used for verification of the model. The difference between the interfacial shear strength (IFSS) calculated using the modified model and that using the original model assuming an equivalent fiber diameter was found to be as high as 15% and it linearly increased as the irregularity of the cross-sectional shape increased. When the irregularity constant exceeds 1.12, the error in IFSS involved in using the original shear lag model and an equivalent fiber diameter is greater than 10%.     

The role of interfacial interactions and loss function of model adhesives on their adhesion to glass

The measurement of adhesion and the evaluation of influencing factors are of great scientific and technological importance. There are two distinct viewpoints on adhesion: (i) surface chemistry, and (ii) fracture mechanics. For elucidation of the relative importance of mechanical properties in the bonding of adhesives, the strength of adhesion between model adhesives and glass plates was measured by the wedge cleavage (WC) test method. Copolymers of methyl methacrylate (MMA) with n-butyl acrylate (nBA) and methyl methacrylate with styrene (S) were prepared as model adhesives. The results show that in MMA-nBA copolymers, by increasing the amount of nBA, both the loss function and the adhesion energy of the adhesives increase. However, by increasing the amount of nBA above a certain level, the adhesion strength begins to decrease. In this situation, the cohesive strength of the adhesive dominates the failure mechanism. On the other hand, a decrease in adhesion was expected upon increasing the amount of styrene in the poly(styrene-co-methyl methacrylate) adhesive, because methyl methacrylate, an interactive monomer with glass, is replaced by a non-interacting styrene monomer, while the loss function of the adhesive is almost constant. But our practical adhesion measurement technique was not sensitive enough to detect this adhesion loss.     

Surface oxidation of carbon fibre on tribological properties of PEEK composites

Abstract

In this work, ozone modification method and air oxidation were used for the surface treatment of polyacrylonitrile (PAN) based carbon fibre. The surface characteristics of carbon fibres were characterised by X-ray photoelectron spectroscopy. The interfacial properties of carbon fibre reinforced PEEK (CF/PEEK) composites were investigated by means of the single fibre pull-out tests. As a result, it was found that IFSS values of the composites with ozone treated carbon fibre are increased by 60% compared with that without treatment. X-ray photoelectron spectroscopy results show that ozone treatment increases the amount of carboxyl groups on carbon fibre surface, thus the interfacial adhesion between carbon fibre and PEEK matrix is effectively promoted. The effect of surface treatment of carbon fibres on the tribological properties of CF/PEEK composites was comparatively investigated. Experimental results revealed that surface treatment can effectively improve the interfacial adhesion between carbon fibre and PEEK matrix. Thus the wear resistance was significantly improved.