Fatigue and creep of adhesively bonded joints
Adhesives are extensively used in modern structural engineering as they can deliver specific benefits over other joining methods. For example, adhesives allow more flexibility in component design and eliminate the weight and space penalties of other joining methods such as riveting and bolting. The removal of the requirement to drill the structure for joining improves the stress distribution in the joint and thereby increase fatigue resistance and service life. Adhesives are extensively used in the manufacture of high strength to weight ratio components such as the honeycomb structures used in aerospace applications. Even in some miniature components, adhesives have been employed to increase reliability and performance and reduce the number of assembly components. In order to ensure reliable in-service performance, it is vital that the long term behaviour of bonded joints should be understood. In terms of loading the two distinct, but related, features of a joint that determine its long-term performance are its fatigue and creep behaviour. Recent work [1.1-1.8] has shown that creep can be significant in adhesive materials, even at ambient temperatures, and the accumulation of creep strain under normal service loading conditions can lead to premature failure in adhesively bonded structures. Creep-fatigue interaction is an important problem that must be addressed in the design of many bonded structures. It is generally believed that when there is creep-fatigue interaction, such as under low frequency fatigue or high frequency fatigue combined with extended hold periods, a crack will grow faster than under either fatigue or creep loading alone. This is a practical consideration in many industrial areas such as the aerospace and automotive sectors. The aims of the current project were therefore to investigate the creep and fatigue performance of adhesively bonded joints and to evaluate any creep-fatigue interactions. An extensive range of experimental data has been generated in this work to provide the information required for the development and assessment of creep and fatigue failure criteria. A creep rig has been designed to collect creep data and tensile tests have been conducted to obtain the mechanical properties of the adhesive. Adhesively bonded double-cantilever beam (DCB) samples have been tested in fatigue at various frequencies (0.1-10 Hz) and temperatures (22 ±1 -120°C). The adhesive used in this work was a toughened epoxy (FM300-2M) and the substrates used were a carbon fibre reinforced polymer (CFRP) and mild steel. Results showed that the crack growth per cycle increases and the fatigue threshold decreases as the test frequency decreases. The locus of failure with the CFRP adherends was predominantly in the adhesive layer whereas the locus of failure with the steel adherends was in the interfacial region between the steel and the adhesive. The crack growth was faster, for a given strain energy release rate, and the fatigue thresholds lower for the samples with steel adherends. Increasing the temperature to 120°C drastically reduced the threshold fatigue value. Tests with variable frequency loading were also carried out at a range of temperatures. The effects of creep, fatigue and creep-fatigue were investigated through a series of experiments using steel-epoxy DCB samples. This entailed testing under quasi-static, fatigue and creep loading and an investigation of the effects of test frequency and temperature on the fatigue life and the crack growth rate. The experimental work was supported by linear and non-linear finite element analysis of the joints and an investigation of elastic, elastic-plastic and time dependent fracture mechanics parameters. Various correlations between the fracture parameters and crack growth for time and frequency dependent fracture have been explored. Four methods to predict crack growth under creep-fatigue conditions have been investigated. The first method assumed that crack growth can be described by an empirical crack growth law and the crack growth law constants can be determined experimentally as functions of frequency and temperature. The Second method assumed creep and fatigue crack growth are competing mechanisms and the crack growth rate is determined by whichever is dominant. The third method assumed that crack growth can be partitioned into cyclic dependent (fatigue) and time dependent (creep) components. The fourth method was used to treat the effect of creep/fatigue interaction. The predicted crack growth using these methods agreed well with the experimental results. Each of these methods was shown to be useful in predicting crack growth and the selection of the most appropriate method is dependent on the data available and the loading /environmental conditions.