The interfacial fracture properties of a representative polymer/metal interface commonly found in microelectronic applications are examined. The double cantilever beam (DCB) configuration was used to investigate the effects of environmental variables on interfacial adhesion and progressive delamination under monotonic and cyclic fatigue loading conditions. The steady-state interfacial fracture energy, Gss, taken from the plateau of the R-curve, of a representative silica-filled Phenol-Novolac epoxy on a Nielectroplated Cu substrate showed little sensitivity to the presence of moisture. On the other hand, both the initiation interfacial fracture energy, Gi, and the entire progressive debond curve under fatigue loading were remarkably sensitive to moisture and temperature, respectively. Debonding is modeled in terms of interface structure, chemistry using chemical reaction rate theory, and relaxation process at the debond tip. The activation energy for stage I debond growth is found to be 140 kJ/mol and 63 kJ/mol for stage II for the current polymer/metal interface.

For several microelectronics applications, Cu/dielectric adhesion is a key reliability issue. Either electroplating or local galvanic coupling under moist operating conditions may result in hydrogen induced interface weakening. The present study compared hydrogen effects on Cu/SiO2 adhesion for sputter deposited f'dms with and without Ti underlayers. Thin Cu and Cu/Ti films ranging from 80–3000 nm have been evaluated. Direct observations of the surface during electrolytic charging have shown no evidence of film/substrate debonding for Cu/Ti systems. In contrast, extensive delaminations have been observed for Cu films. Indentation testing immediately after charging indicated up to 100% decrease in the practical adhesion for Ti/Cu films. The observed effect resulted from a true interfacialfracture energy reduction from 4 to 2 J/m2, Plastic energy dissipation was assumed unaffected as no yield stress changes were detected after charging. Even with the deleterious effect of hydrogen, adhesion strength of Cu/Ti films remained higher than that of non-charged Cu films.

Indentation-induced delamination of thin films provides the basis for adhesion calculations. In the case of ductile Cu films plastic deformation usually prevents a film from debonding from the substrate. Deadhesion is facilitated by the use of a hard W superlayer, which promotes indenter-induced Cu film failure, increasing the delamination area by an order of magnitude. Radial as well as annular cracking acts like a secondary mechanism in the strain energy release, and can be resolved from excursions on the load-displacement curves. For the thicker Cu films no excursions were observed, though radial cracking took place. It is important to identify fracture events as they occur in order to understand the system behavior and accurately apply the analysis. An acoustic emission signal is used to detect the magnitude of fracture events in thin Cu films. For the films of different thickness from 40 nm to 3 microns the corresponding interfacial fracture energy ranged from 0.2 to over 100 J/m2. Limits of plastic energy dissipation are determined with the lower limit, the true work of adhesion, being associated with a dislocation emission criterion. Crack arrest marks were found upon the blister removal, and are proposed to represent the shape of the crack tip. Total acoustic emission energy was found to be inversely proportional to the strain energy release rate.

Silicon nitride thin films used for transistor spacer are usually under large tensile residual stresses. When a film is thick enough, it can crack by channel cracking mechanism. The toughness of these films, used in models for failure criteria, can be obtained by determining the critical thickness for channel cracking of blanket films with known residual stresses. However, depositing films of a range of different thickness to determine the critical thickness is tedious, and this process also introduces risks of producing particles in the deposition chamber when the film exceeds the critical thickness. An alternative technique is to deposit films of thickness below critical thickness and then to apply additional tensile stresses to the film by bending the substrate. A critical film stress can be determined for channel cracking from which the film toughness is obtained. This paper describes the experiments performed for a new SiN film and presents the results.

Fatigue (slow) crack growth in silane bonded epoxy/glass interfaces was studied under static and cyclic loading at 23°C, 95% RH using the double cleavage drilled compression test. Crack growth rates under cyclic loading were significantly greater than under static loading, in contrast to crack growth results in monolithic glass. In addition, cyclic crack growth rate data after aging up to 94°C in distilled water were not significantly different from unaged samples; however, after aging at 98°C in distilled water and 70°C in a KOH aqueous solution of pH 10, crack growth became cohesive and fractal in nature. These results are discussed in terms of possible mechanisms for fatigue crack growth at silane bonded epoxy/glass interfaces.

Microwedge Indentation (MWIT) and Precracked Line Scratch (PLST) tests have been developed for adhesion measurements of microscale fine lines. In the MWIT a symmetric wedge-shaped probe is driven downward into a thin film line debonding it from the substrate. For the PLST, a precrack is introduced at one end of a thin metal line on a substrate. The line is loaded from the precrack end by an asymmetric diamond wedge until the crack propagates. When the crack reaches its critical length at a certain critical load, the film buckles.

The mechanics of these tests are based on several assumptions of thin line behavior, which are not experimentally observable due to small line sizes and equipment limitations. Testing similar but larger lines allows for the direct observation of line delamination and buckling, which can be used for calculating in-situ crack lengths as well as instantaneous strain energy release rates.

Both MWIT and PLST models [1–4] were experimentally verified on the macroscopic scale using polycarbonate as a line material bonded to a steel substrate with cyanoacrylate. New features such as initial crack growth as well as instantaneous crack growth at buckling were observed. Modified Kic tests were performed for strain energy release rate comparison. Adhesion values ranging from 200 to 1000 J/m2 were measured for this system, and had very strong mode mixity dependence. Connectivity to the microscale is through previous experiments by de Boer, et al using fine line W/SiO2/Si systems [ 1–4].

Silicon high-power devices are commonly bonded to Mo electrodes using Al films. Bonding stress will inevitably be introduced into the Si substrate by such a process. In this work, the infrared (IR) photoelasticity (PE) method was employed to measure the stress distribution in the Si substrates induced by high temperature bonding process of Si/Al/Mo structures commonly used in the production of silicon thyristors. It is demonstrated that quantitative information on both the directions and magnitudes of the stress can be obtained. The dependence of the magnitude of the stress on the geometrical parameters of the structure has also been studied. The experimental results are shown to agree well with the calculated results derived from a theory of interlaminar stresses in composites.