The Israelachvili surface force apparatus has been used extensively over the past decade to make detailed measurements of surface forces and adhesion between very smooth solids in various liquid and vapor environments. Most of those measurements have been made with mica surfaces, but we have recently developed a method of preparing smooth silica surfaces for use in place of the mica. The silica surfaces adhere in dry and humid atmospheres, but do not adhere when immersed in water.

The use of a second material not only broadens the scope of the Israelachvili technique, but also enables studies of forces and adhesion between dissimilar materials. In this work, we present the results of measurements of adhesion in air and forces in aqueous solution between two silica surfaces; we also report preliminary results of the adhesion between a mica surface and a silica surface.

The contact stresses resulting from indentation of a compliant coating on a rigid substrate cause debond cracks to initiate and propagate along the interface. The indentation debonding behavior of four polymer coatings on glass substrates is investigated. A model is developed for calculating interfacial fracture energy based on data of indenter load vs. debond crack size. Based on the experimental observations and the data for the four coating systems, the limitations of the model are discussed.

An experimental approach has been developed using a single-fiber pullout test to measure intrinsic interface properties for ceramic composites. The properties are determined from a pull-out, force-displacement curve, which is directly related to reinforcement toughening via fiber/matrix debonding and frictional pull-out. They were evaluated for a model composite system of continuous SiC fibers with various surface treatments in a borosilicate glass matrix. For the processing conditions used, the interface fracture toughness and the interface frictional shear resistance were found to be 1.0 ± 0.5 J/m2 and 3.3 ± 0.6 MPa, respectively, for as-received fibers. Experiments conducted with long embedded fiber lengths allowed the shear resistance to be deconvolved into an interface friction coefficient of 0.05 ± 0.01 and an initial fiber-clamping pressure of 65 ± 6 MPa. Nitric acidwashed fibers gave an increased interface toughness of 3.6 ± 0.1 J/m2 and friction coefficient of 0.08 ± 0.02, but nearly the same initial clamping pressure, 72 ± 12 MPa. Calculations of the clamping pressure from the fiber/matrix thermal expansion mismatch and from stress birefringence measurements in the glass were in general agreement with this value.

In situ observations of fracture mechanisms and crack bridging effects in model copper-glass laminated composites are described and compared with measured resistance curves.

The effects of reinforcement debonding and work hardening on ductile reinforcement toughening of γ-TiAl have been examined. Debonding has been varied by either the development of a brittle reaction product layer or by depositing a thin oxide coating between the reinforcement and matrix. The role of work hardening has been explored by comparing Nb reinforcements that exhibits high work hardening with solution hardened Ti-Nb alloy that exhibits negligible work hardening. It is demonstrated that a high work of rupture is encouraged by extensive debonding when the reinforcement exhibits high work hardening. Conversely, debonding is not beneficial when the reinforcement exhibits low work hardening.

Au/Al2 O3 interfaces are created by bonding highly textured Au films (25 pm thick) to single crystal A12 O3. Mechanical tests are done under the optical microscope to examine the effect of plastic deformation on the energy of fracture of the interface. Fracture occurs at the interface and is accompanied by plastic deformation in the Au. The relatively large value for the fracture energy measured (50-70J/m2) is attributed to the plastic deformation in the Au. It is also observed that fracture occurs subcritically due to stress corrosion at fracture energies from 10-20J/m2. Topographic features on the fracture surfaces are characterized and discussed briefly.

The mechanical properties of fiber composites are strongly influenced by the debonding of fibers. When an embedded fiber is loaded from one end, debonding can occur at both the loaded end and the embedded end. Existing theories neglect the possibility of debonding from the embedded end and are thus limited in applications to cases with low fiber volume fraction, low fiber modulus, high interfacial strength/interfacial friction ratio or short fiber length. A new twoway fiber debonding theory, which can extend the validity of one-way debonding theories to all general cases, has recently been developed. In this paper, the physical reason for the occurrence of two-way debonding is discussed. The limit of validity for one-way debonding theories is considered. One-way and two-way debonding theories are then compared with respect to the prediction of composite behaviour. The determination of interfacial parameters from the fiber pull-out test will also be described.

The influence of flexible interlayers/interphases on the performance of unidirectional fiber reinforced composites is studied. Micromechanical analysis based on the embedded composite cylinders model is used to study the stiffness as well as the internal stress distributions within the matrix phase of composites. Based on the results of the analysis, a criterion is proposed for the selection of optimal interlayer thickness for fiber composites. The proposed criterion gives results which seem to correlate well with the experimental data published in the literature.

The effects of a frictional interface on the load diffusion from a broken fiber to the surrounding matrix material and the extent of debonding near the fiber break in a single-fiber reinforced composite of infinite extent are studied by using the finite element method. The normal and shear stresses on the interface, the extent of the slip zone and the axial load of the fiber are evaluated for different frictional coefficients and material parameters for both the fiber and the matrix. A shear-lag analysis is also carried out to obtain a closed form approximate solution of the fiber load diffusion problem. The extent of the slip zone and the stresses predicted by the shear-lag model are compared with the finite element method results.

The effect of fiber surface treatments on the relationship between the tensile strength of a filament and the shear strength of its interphase is one of the central issues facing composite materials technologists today. We demonstrate here that analysis of fragmentation phenomena in monofilament composites can simultaneously yield information about these two parameters. Characterization of shear stress transfer zones in non-critical fragments has led us to the determination of interphase strength.

A phenomenological treatment that highlights the role of the matrix in the fragmentation process is presented here. This analysis considers issues such as the strain energy exchange between a failing fiber and the matrix, as well as interphase relaxation due to the viscoelastic nature of the matrix. Our observations of the fragmentation phenomena in AU4/polycarbonate monofilament composites indicate that the fiber/matrix interaction in this system is governed by micromechanical locking.

The microbond method has been applied with increased frequency to characterize interfacial adhesion in fiber reinforced composites. Nevertheless, a number of serious questions remain regarding the interpretation of experimental data. This paper addresses material and experimental variables in the microbond test procedure including the cure behavior of thermoset test specimens, matrix heterogeneity, locus of failure in test specimens and load application techniques. The theoretical basis for the method is examined by viewing experimental results in terms of existing theoretical interpretations of interfacial failure. Conclusions are presented regarding the limitations and the potential of the microbond method for determining fiber/matrix adhesive bond strength.

Microstructural investigations of HIP'ed silicon nitride reinforced with silicon carbide whiskers were performed using a 400 kV high resolution analytical transmission electron microscope. Two different whisker sources as well as uncoated and carbon coated whiskers were evaluated for their effectiveness in improving mechanical properties.

This paper focuses on microstructural and microchemical analysis of interface and grain boundary regions within the composites. Combined HREM and AEM investigations reveal that variations in fracture toughness are mainly influenced by the microstructural and chemical development of the interface. Fracture toughness can be related to the surface chemistry of the reinforcements.

The reactions between coated SiC continuous fibres and TiAl, Ti3A1, or Ti3A1 + β matrices, have been investigated using optical metallography, scanning electron microscopy (SEM), and transmission electron microscopy (TEM), with chemical information obtained by energy dispersive x-ray analysis (EDX) and electron energy loss spectroscopy (EELS).

Similar sequences of phases were found in all the composites; carbides based on TiC and TixAIC (x=2,3) and titanium silicides based on Ti5Si3 forming in narrow bands outside of the complex carbides. The occurrence of individual phases can be related to published phase diagram information. However, the sequences in which the reaction products occur within the reaction zones cannot be fully rationalised from the present results.

The dislocation structure of Pt-(001)NiO interfaces was studied using electron microscopy and electron diffraction techniques. Specimens were produced by hot pressing polycrystalline Pt films on to thin NiO single crystals, and bulk Pt single crystals on to bulk NiO single crystals. The polycrystalline Pt specimens were used to determine the favored orientation relationships between the NiO and Pt, while the bulk NiO-Pt specimens were used to study the detailed structure of the interface. Three categories of orientation relationships were identified: exact epitaxy with (001)Pt ║ (001)NiO, [110]Pt ║ [110]NiO; small rotations away from exact epitaxy about the common [001] direction; high index planes of Pt parallel to (001) of NiO. Theoretical calculations of the expected dislocation structures of interfaces with the first two orientation relationships were made using a Bollmann-type analysis. The experimental observations and theoretical predictions were shown to be in good agreement. The energies of the interfaces having the first two orientation relationships were shown to be similar which is believed to be the reason why they both occur.

Solid state bonding of Si3N4 and Ni is considered to be a simple ceramic/metal diffusion bonding as the case of Al2O3/Nb. There are two kinds of bonding interfaces observed by transmission electron microscope(TEM). In addition to the directly bonded interface, an amorphous-like phase was found in some regions of the Si3N4/Ni interface. It is responsible for the degradation of bonding strength. In view of the experimental results, the existence of the amorphous-like phase in bonding interface results from the formation of SiO2 and accumulation of Al2O3 and Y2O3, which were used as sinterfng binders of Si3N4, via intergranular diffusion during the bonding process.

As a preliminary step to detailed study of the interfaces in SiC particulatereinforced Al alloys, a number of methods of sample preparation have been explored; the resultant artefacts on the thinned foils observed in transmission electron microscopes are discussed and improvements to the preparation procedures are suggested. Initial observations of the microstructure of two alloys differing significantly in their matrix silicon content are presented.

The interface between an epoxy adhesive and sheet molding compound (SMC) was studied by TEM. The interface width is less than 20 nm where it is sharp, and on the order of 40 nm where it is more diffuse. The interfacial area in the TEM sample is located first by the difference between adhesive and SMC microstructures. Where there are no microstructural features we can investigate further the location of the interface by using the Mg Kx peak from magnesia thickener in the polyester matrix of the SMC. Calcium carbonate crystals in the SMC extend to within less than 25 nm of the interface. Measurements of interface extent and roughness indicate that there is intimate contact and good mechanical bonding between the adhesive and SMC. The surface and interfacial energetics of this epoxy-SMC system were evaluated using sessile drop advancing contact angles. Good mechanical bonding observed by TEM is consistent with a small advancing contact angle of 20°the epoxy adhesive on the SMC at 121 °C which shows a good thermodynamic driving force for the intimate contact during the bonding process.

Transcrystallization of semicrystalline polymers, such as PEEK, PEKK and PPS, in high performance composites has been investigated. It is found that PPDT aramid fiber and pitch-based carbon fiber induce a transcrystalline interphase in all three polymers, whereas in PAN-based carbon fiber and glass fiber systems, transcrystallization occurs only under specific circumstances. Epitaxy is used to explain the surface-induced transcrystalline interphase in the first case. In the latter case, transcrystallization is probably not due to epitaxy, but may be attributed to the thermal conductivity mismatch. Plasma treatment on the fiber surface showed a negligible effect on inducing transcrystallization, implying that surface-free energy was not important. A microdebonding test was adopted to evaluate the interfacial strength between the fiber and matrix. Our preliminary results did not reveal any effect on the fiber/matrix interfacial strength of transcrystallinity.

The absorption of water in glass fiber reinforced epoxy resins is a complex process. With dynamic-mechanical measurements it is possible to distinguish locations of water in the resin-glass interphase, in micro-cracks of the resin, and diluted in the bulk polymer. The dominant mechanism of incorporation depends on the resin/hardener system, the surface treatment of the glass fabrics, and the sorption temperature.

Dynamic-mechanical measurements were used to study the properties of the interphase in glass fibre reinforced epoxy resins. The immobilisation of the polymer chains in the interphase is determined by the curing system, the annealing process, and the surface treatment of the glass fibres.