CRACK GROWTH OF STRUCTURAL ADHESIVE JOINTS IN HUMID ENVIRONMENTS

The adhesive fracture energy, G _c , of aluminiumalloy and steel joints bonded with a rubber-toughened epoxy adhesive has been measured using monotonicallyloaded tests. Such tests have been conducted at different levels of relative humidity, and two surface pretreatments have been employed for the substrates prior to bonding: a simple grit-blast and degrease (GBD) pretreatment or a silane primer (GBS) pretreatment. When G _c was plotted against the crack velocity, three regions of fracture behaviour could be distinguished. At low rates of displacement the crack grew in a stable manner, visually along the interface, and relatively low crack velocities could be readily measured. This was termed “Region I”, and here the value of the adhesive fracture energy was relatively low and decreased steadily as the relative humidity was increased. On the other hand, at relatively high rates of displacement the crack grew in a stick-slip manner mainly cohesively in the adhesive layer at approximately 20 km/min. This was termed “Region III”, and here the value of G _c was relatively high and independent of the relative humidity. In this region the crack was considered to grow faster than the water molecules were able to reach the crack tip, which explains the independence of G _c upon the test environment. In between Region I and Region III a transition region was observed, which was designated “Region II”. The major effect of the GBS pretreatment, compared to which the GBD pretreatment, was to increase the value of G _c both in Regions I and III, although the presence of the silane primer had the greater effect in Region I.     

Mode-l Fracture Behaviour of Adhesive Joints. Part I. Relationship Between Fracture Energy and Bond Thickness

The fracture properties of adhesive joints of aluminium were investigated using a rubber-modified tough epoxy resin system (G_IC = 2.76 kJ/m²) as adhesive material. Compact tension (CT) adhesive joints were manufactured for a wide range of bond thickness t (from 0.05mm to 10mm) and fracture tests conducted under static load. Scanning electron microscopy (SEM) was used to examine the fracture surface morphology. A large deformation elastic- plastic finite element model was developed to evaluate the J-integral value for different bond thickness. The fracture energy, J_c , was found to be highly dependent on the bond thickness and was lower than that of the bulk adhesive. As the bond thickness was increased J_c also increased, though not monotonically, towards the fracture energy of the bulk adhesive. This result was caused by the complicated interactions between the stress and strain fields, plastic deformation of the adhesive around the crack tip, constraint from the adherends and the failure path. It was shown that values of J_c as a function of bond thickness correlated well with the variation of plastic zone height. Scanning electron micrographs from the fracture surfaces of the CT adhesive joints illustrated that the failure path was mainly cohesive through the centre-plane of the adhesive layer. Brittle fracture mechanisms were observed for thin bonds (0.04mm < t< 0.5 mm) but tough fracture mechanisms were identified for thick bonds (t > 1 mm).     

Fatigue crack growth in polymers. II. The interaction of mean stress,frequency, and temperature

Fatigue crack propagation was studied in large centernotched plate specimens of two polymers, poly(methyl methacrylate) and polycarbonate, under tensile cycling conditions. Tests were performed at frequencies of 0.1, 5.0, and 20 Hz, and at two temperatures, ?60 and +21°C. The interaction of mean stress intensity, frequency and temperature was investigated. It was found that in tests performed at constant frequency or temperature, the fatigue crack propagation rates were dependent on both the range of the applied stress intensity factor and its mean value, K_m. All propagation rates increased with increasing K_m. Also, the threshold stress intensities decreased with increasing K_m, suggesting very low levels of ΔK for non-propagating cracks, certainly below the 0.4 K_c, the minimum level investigated here. In PMMA, decreasing crack growth rates with increasing frequency were established for a wide range of K_m. However, in PC crack growth rates increased substantially with increasing frequency. Finally, “upper and lower transition points” were noted on crack growth curves of both materials.     

Determination of critical micelle concentration of surfactant by ultraviolet absorption spectra

A method was investigated for determining the critical micelle concentration (CMC) by the shift of absorption maxima when an organic compound (I) with ultraviolet absorption was added to an aqueous solution of a surfactant. When I was added to the surfactant solution at higher concentrations (above the CMC), λ_max of I approached the value inn-octane, since I was solubilized in the hydrocarbon atmosphere of the inner part of the surfactant micelle. At lower concentrations (below the CMC), however, I was present in the water phase and λ_max approached the value in water. The curve of λ_max vs. surfactant concentration declined from the high concentration values as the CMC was approached and at the CMC, the curve broke upward sharply. Then, it rose for some time and approached the value in water. N,N′-diethylaniline was used because it exhibited larger shifts of λ_max. The standard amount used was 0.002 ml/3–10 ml of aqueous solution of the surfactant. The CMC values obtained agreed with those obtained by the electric conductivity method, dye adsorption method and light scattering method, for surfactants such as tetradecyldimethylbenzylammonium chloride, sodium dodecyl sulfate and polyoxyethylene cetyl ether.     

Model prediction of the ESCR of semicrystalline polyethylene: Effects of melt cooling rate

The effects of melt cooling rate on the morphology and environmental stress cracking resistance (ESCR) of a commercial‐grade high‐density polyethylene (HDPE) were investigated by DSC, WAXS, Raman spectroscopy, DMTA, microhardness, and standard Bell test. The results showed exclusion of short chain branches from the crystallites leading to their perfection by decrease in the melt cooling rate. Accordingly, samples' ESCR increased because of the aforementioned crystal thickening. In addition, quantitative evaluation of crystal strength was performed for the first time by microhardness technique. Crystal strength or the required energy for plastic deformation of unit area of the crystals, Δh·?_c, complemented crystal thickening hypothesis. The product of Δh and average crystal thickness, ?_c, was also proportional with storage modulus of the samples at ESCR test temperature. The evaluation of the results using the recently proposed model based on the analogy of crack growth through amorphous phase of semicrystalline polymers in harsh environments with crack growth at adhesive polymer–substrate interfaces showed reasonably good correlation. Finally, the comparison of literature and current research data based on the new model delineated new directions for phenomenon generalization. © 2009 Wiley Periodicals, Inc. J Appl Polym Sci, 2009     

Failure criterion for the fracture of structural adhesive joints

The long-term strength of stressed, structural adhesive joints, consisting of aluminium alloy substrates bonded with an epoxide adhesive, has been investigated. The applied adhesive fracture energy, G_Ic, is shown to be linearly dependent upon the logarithmic time-to-failure; the failure time decreases as the value of G_Ic is increased. The fracture of these joints over eight decades of time is uniquely described by the hypothesis that there is a critical plastic-zone size developed at the crack tip at failure.     

Impact and dynamic fracture resistance of crystalline thermoplastics: Prediction from bulk properties

The behavior of tough, crystalline thermoplastics in notched impact tests leads to the definition of crack initiation resistance and propagation resistance as two distinct properties, G_c and G_D. It is shown here that a single criterion—adiabatic thermal failure of a crack-tip cohesive zone—can be applied to predict both. Dynamic fracture resistance G_D emerges as a geometry independent, though crack speed and temperature dependent, material property, whose minimum value G_D,min depends only on temperature and bulk physical properties. G_D,_min can be measured using a simple pressurized-tube test. Crack initiation resistance G_c, however, is inherently influenced by geometry and impact speed, although its lower bound is also G_D,min. Craze extension and failure of a notched impact specimen, and hence G_c, can be predicted for a specific temperature, given bulk thermal property data and a dynamic stress/strain curve measured by impact bending of an unnotched beam. For materials that comply with the model, sharp-notched Charpy type impact tests will not arrive at a unique G_c value, while Izod type tests, for which a revised compliance calibration is presented, may fail to establish any G_c value at all.     

The Adhesive Fracture Energy of Bonded Thermoplastic Fibre-Composites

The present paper first discusses the problems that occur when thermoplastic-based fibre-composite materials are bonded using structural engineering adhesives, such as epoxy and acrylic adhesives. A double-cantilever beam joint has been employed and it is shown that the value of the adhesive fracture energy, G_c , is very low when a simple abrasion/solvent wipe pretreatment is used for the thermoplastic fibre-composites. This arises from crack growth occurring along the adhesive/composite interface, which is relatively weak when such a pretreatment is employed. Secondly, it is demonstrated how very effective a corona surface pretreatment may be for these materials. Indeed, when such a pretreatment is used, interfacial crack growth is no longer observed but the crack now propagates either cohesively in the adhesive or through the composite substrate; both failure modes lead to relatively high values of G_c , with the former resulting in the highest values of G_c being recorded. Finally, from measuring the fracture properties of the composites and combining these data with a detailed analysis of the stresses in the DCB joint, calculated using a finite element analysis, the reasons for these different loci of failure may be readily understood and predicted.     

Electrical and thermal properties of composite of liquid crystalline polymer filled with carbon black

The electrical resistivity and thermal conductivity of a liquid crystalline polymer (LCP) filled with a commercial carbon black (CB) of various volume fractions (?) is investigated. The percolation threshold (?_c) is found at about 3%, and the resistivity (ρ) as a function of (? ? ?_c) satisfies the exponential function. Although the pure LCP is highly anisotropic in thermal and mechanical properties after processing, the composite samples exhibit no preferential direction for electrical conduction. Samples of ? below ?_c exhibit a negative temperature coefficient of resistivity while those above ?_c show almost no temperature dependence from room temperature to 200°C. In addition, the samples at lower ? have higher thermal conductivity in the LCP flow direction than those measured in the transverse and thickness directions, and they approach the same value at higher ?. This result indicates that preferential molecular alignment of the matrix LCP is responsible for the behavior of the thermal conductivity of the composites. © 2001 John Wiley & Sons, Inc. J Appl Polym Sci 82: 1549–1555, 2001     

Determination of fracture toughness of poly (ethylene terephthalate) film by essential work of fracture and J integral measurements

Abstract

The plane stress fracture toughness of a semicrystalline poly (ethylene terephthalate) (PET) film of thickness 0·125 mm has been measured as a function of specimen size, specimen geometry, loading rate, and temperature using the essential work of fracture (EWF) approach. It was found that the specific essential work of fracture w _e was independent of specimen width, specimen gauge length, and loading rate, but was dependent upon specimen geometry and test temperature. Below the glass transition temperature (93°C), w _e for double edge notched tension (DENT) type specimens was temperature insensitive, but increased with temperature for single edge notched tension (SENT) type specimens. The w _e value for SENT specimens was consistently higher than for DENT specimens. Estimation of w _e via crack opening displacement was reasonable using the relationship w _e = σ_n e _0,y; estimations made via similar type equations were either too high or too low and were generally unsatisfactory. It was found that values of J integral obtained by power law regression and linear extrapolation of the J–R curves to zero crack growth were lower than w _e. The power law regression of the J–R curves with ?a taken as half the crack opening displacement value at maximum load gave J _c values which agreed reasonably well with w _e.     

Moisture effects on the behavior of graphite/polyimide composites

The effect of moisture on the material behavior of Magnamite® IM7 Graphite/Avimid® K3B thermoplastic polyimide composite laminates has been investigated. Laminates consisting of a 62 vol% fiber were laid up with several different stacking sequences: [90]₁₀ or unidirectional, [0₂/90]_S or thin cross-ply, [0₂/90₂/0₂]_S or thick cross-ply, [45/0/-45/90]_S or quasi-isotropic, and [0/90] laminates. In this study, the glass-transition temperature, T_g, and the intralaminar fracture toughness, G_IC, were measured for dry and moisture-saturated unidirectional samples. When laminates were saturated with moisture, the value of T_g was found to decrease from its baseline (dry) value, but recovered upon redrying the samples. This observation is consistent with the effects of moisture on the T_g of other polymer composites. Permanent toughness losses have not been observed in samples conditioned in room-temperature 75% and 100% relative humidity environments. However, during long-term conditioning of cross-ply and quasi-isotropic samples in liquid water, transverse cracks initiated in the absence of an applied mechanical load. Moisture uptake curves for conditioning in room-temperature liquid water, 80°C (176°F) liquid water, and at room temperature and 75% relative humidity were used to calculate Henry's Law constants and diffusion coefficients. Non-Fickian behavior, consisting of a postsaturation increase in moisture uptake, was observed in crossply and quasi-isotropic laminates and might be due to the observed transverse cracking.     

The Fracture Toughness of Adhesive-Bonded Joints

Fracture mechanics is related to adhesion theory and the testing of adhesive-bonded joints in the lap-shear configuration. The complexity of the stress field necessitates the strain energy release rate approach, which is followed to derive the relation for a lap-shear sample: G_c = P_c ²/4b (dC/da). G_c is the fracture toughness (critical strain energy release rate), P_c is the breaking or crack instability load, a and b are crack lengths and widths, respectively, and C is the sample compliance for the Tap-shear sample with a crack of these dimensions at each loading edge. It was found that G_c ranged from 1.18 to 1.42 with an average value of 1.34 in.-lb./in.² for epoxy bonded aluminum strips (EPON 934 and Alcald 2024-T3). Evidence, in the form of photoelastic stress patterns, suggesting that crack extension occurs in the opening mode in lap-shear samples is presented and discussed.     

Crack Front Curvature in the Wedge Test

The wedge test, as used for the evaluation of adhesive fracture energy, is usually considered to be a 2D geometry: its simple analysis implies independence of the width of the adhesive joint, b. Recent work has shown this to be an over-simplification, at least in some circumstances, with (hypothesised) anticlastic bending giving rise to curvature of the crack front. As a result, crack front length, a, varies across the joint width leading to ambiguity in the interpretation of results to obtain fracture energy, G _c . This contribution constitutes a more detailed analysis of the geometry of the wedge test (in the particular case of one bending and one rigid substrate), treating the bent member as a plate, rather than as a simple beam. The Kirchhoff-Love plate theory is applied and solved by a perturbation method. Secondary curvature of the beam in the direction normal to the principal curvature results directly from the treatment, and this, in turn, leads to a concave crack front, corroborating the above-mentioned experimental observation.     

Aging and performance of structural film adhesives. III. Effect of humidity on a modern aerospace adhesive

The aging of an uncured, DGEBA-based commercial film adhesive exposed to high humidity at 40°C was monitored by high performance liquid chromatography (HPLC), thermal analysis, solubility, flow, and mechanical testing of joints. DGEBA and brominated DGEBA resins, and representative mixes of these were hydrolyzed in vitro. It is suggested that the major reaction at 40°C is polymerization, which is accelerated by moisture and leads to a crosslinked structure different to that resulting from normal cure. The poorer adhesive performance after aging is mainly the result of reduced resin flow during joint formation, although there is some contribution from the hydrolysis of epoxide groups which leads to lower crosslink density in the cured adhesive. The softening point of the uncured material and T_g of the cured material are useful indicators of the quality of this adhesive which, compared to some earlier formulations, has improved resistance to these conditions.     

Cracking behavior of ZrB2-SiC-Graphite sharp leading edges during thermal shock

Crack initiation and propagation of ZrB₂-SiC-Graphite (ZSG) sharp leading edges (SLEs) subjected to thermal shock were systematically evaluated by the water spraying method followed by a crack dyeing treatment. Distinct differences in the crack patterns among different test conditions were observed, and the cracking behavior of ZSG SLEs (including crack initiation time, crack number and critical failure temperature) was revealed to be strongly dependent on both the cooling rate and the microstructure. The crack propagation during thermal shock could be considered as a quasistatic process (crack speed was lower than 1?cm/s) that needed to be driven by continuous cooling.     

Nanoindentation creep and glass transition temperatures in polymers

The nanoindentation creep behaviour of several different polymers has been investigated. The extent of creep ε is represented by the Chudoba and Richter equation: ε = ε_eln(ε_rt + 1), where t is the loading time and ε_e and ε_r are material constants. Creep was determined in this way for a variety of polymers at T_exper = 301.7 K. Some of the materials studied were far above, some far below and some near their glass transition temperatures T_g. The creep rate ε_r was plotted as a function of y = (T_g ? T_exper); a single curve was obtained in spite of a large variety of chemical structures of the polymers. The ε_r = ε_r(y) diagram can be divided into three regions according to the chain mobility. At large negative y values, the creep rate is high due to the liquid‐like behaviour. At large positive y values in the glassy region, the creep rate is higher than that in the negative y‐value region; the creep mechanism is assigned to material brittleness and crack propagation. In the middle y range there is a minimum of ε_r. These results can be related to glassy and liquid structures represented by Voronoi polyhedra and Delaunay simplices. The latter form clusters; in the glassy material there is a percolative Delaunay cluster of nearly tetrahedral high‐density configurations. The creep mechanism here is related to crack propagation in brittle solids. In the liquid state there is a different percolative Delaunay cluster formed by low‐density configurations, which, as expected, favour high creep rates. Copyright © 2007 Society of Chemical Industry     

Effect of nitrogen to niobium atomic ratio on superconducting transition temperature of δ-NbN<Subscript><Emphasis Type="Italic">x</Emphasis></Subscript> powders

Powders of cubic niobium nitride δ-NbN _x with a particle size of below 20 μm were prepared by reactive diffusion at T = 1455?1475°C under nitrogen pressures of P ₁(N₂) = 0.1?3 MPa and P ₂(N₂) = 25 MPa. For these powders, the values of the stoichiometric coefficient x, lattice parameter a, and the superconducting transition temperature Tc were measured and the a(x), T _c(x) and T _c(a) functions were analyzed. The T _c values were found to linearly grow with increasing a (decreasing structure imperfection). A maximum value of T _c (15.8 K) corresponded to a maximum value of a (4.3934 Å). Maximain the a(x) and T _c(x) curves were found to correspond to a slightly substoichiometric nitride with x = 0.98. Having synthesized cubic niobium nitrides with 0.892 < x < 1.062, we managed to measure the dependences of a and T _c on x all over the almost entire homogeneity range for δ-NbN _x . Our a(x) and T _c(a) functions were found to reasonably agree with those previously reported for SHS-produced δ-NbN _x powders.     

Relationship between orientation of amorphous chains and modulus in highly oriented polypropylene

By using oriented polypropylene prepared by forced quenching in a zone-drawing-type apparatus, the effect of taut tie molecules on the modulus is studied by measuring the changes of superstructure with an increasing draw ratio and the temperature at which the oriented polypropylene was annealed. Superstructure is analyzed by means of an x-ray method, differential scanning calorimetry, thermal shrinkage, birefringence, and infrared spectrum. Modulus increases with an increasing orientation function of amorphous chains, ?_a, and is decided only by the value of ?_a, so long as the higher value of orientation function of the crystal c axis does not change with the draw ratio or annealing. The taut tie molecules in ultrahigh-modulus polypropylene are loosened by annealing at temperatures below 420 K, but would be incorporated into folded lamellar crystals above the annealing temperature of 420 K. The taut tie molecules does not always have a 3₁ helix conformation.     

Equilibrium solubility of CO2 in rubbery EVA over a wide pressure range: effects of carbonyl group content and crystallinity

The equilibrium CO₂ sorption isotherms and isobars for rubbery EVA containing various amounts of vinyl acetate (VA) over a wide pressure range 10-340 atm and 25-52 °C were investigated. The normalized CO₂ sorption concentration (in cm³ (STP) CO₂/mole VA) isotherms of these polymers as a function of pressure consisted of two distinct regions turning at near P_c. The normalized sorption isotherms in these two distinct regions could be fairly described by two respective power laws: C=K_aP^n_a for above P_c and C=k_bP^n_b for below P_c. The normalized CO₂ sorption isotherms were found to be about the same for four EVA samples having different VA contents for below P_c, suggesting that the sorption process at below P_c may be mainly driven by the presence of carbonyl groups. At above P_c, the degree of crystallinity of the polymer appeared to be a major factor to affect the sorption process, with the higher the degree of crystallinity, the lower the normalized CO₂ sorption concentration in the polymer. The sorption isobars of the polymer as a function of temperature were governed by the interplay of density, viscosity, and diffusivity of CO₂ depending on the pressure studied.     

Impact of geometry on the critical values of the stress intensity factor of adhesively bonded joints

Several studies have dealt with the application of the generalized stress intensity factor (GISF) on the failure load prediction of adhesive joints. However, the effect of geometry on the critical value of the GSIF (H_c) is complex and limits its application. Due to the effect of multiple geometrical features and the limited success in the field of adhesive joints, a statistical analysis is a possibility. This paper investigates the impact of different geometrical features on the H_c in single lap adhesive joints. To achieve this, the statistical response surface methodology (RSM) was used to design the experiments and for the statistical analysis. According to the RSM, 31 arrangements of single lap joints were manufactured and tested. In this analysis, the adhesive thickness, adherend thickness, overlap length and also the free length, each in five different levels, were considered. The effect of linear, quadratic and two-way interactions of the geometrical parameters on the H_c and failure load were also studied. It was shown that H_c is most affected by the overlap length. Variation of H_c in term of the free length is by far higher at lower adhesive thicknesses. Also, the effect of substrate thickness on H_c is more considerable for thinner bondlines. The interactions of overlap length/free length and overlap length/adhesive thickness affect the failure load more considerably than the other studied interactions. The effect of free length on the failure load increases with the bondline thickness, while the effect of substrate thickness is stronger for a lower adhesive thickness.