Skip to main content
main-content

Über dieses Buch

This volume documents the proceedings of the International Symposium on Adhesive Joints: Formation, Characteristics and Testing held under the auspices of the Division of Polymer Mater­ ials:Science and Engineering of the American Chemical Society in Kansas City, MO, September 12-17, 1982. There is a myriad of applications (ranging from aerospace to surgery) where adhesives are used to join different materials, and concomitantly the understanding of the behavior of adhesive joints becomes very important. There are many factors which can influence the behavior of adhesive joints, e.g., substrate preparation, in­ terfacial aspects, joint design, mode of stress, external environ­ ment, etc., and in order to understand the joint behavior in a holistic manner, one must take due cognizance of all these germane factors. So this symposium was planned to address not only how to make acceptable bonds but their characterization, durability and testing were also accorded due consideration.

Inhaltsverzeichnis

Frontmatter

Substrate and Interfacial Aspects

Frontmatter

The Role of Surface and Bulk Characterization in the Evaluation of Adhesive Joints

Adhesive joints are often evaluated by a mechanical test followed by exposure of a similar joint to elevated temperature, humidity, and/or corrosive atmosphere. After this exposure to a deleterious atmosphere, the bonded joint is usually tested again using the same geometry and conditions. The original load to failure is usually taken as a measure of the quality of the joint and performance during the accelerated test determines how well the joint will hold up in service. In the past only these numerical values were noted and there was little diagnostic work on the failure surfaces. More recently, however, there has been greater emphasis on where and why a failure took place. In order to fully evaluate failure surfaces it is necessary to determine the locus of failure. Often this task requires modern methods of surface characterization, especially when the failure takes place along a weak boundary layer (WBL). These surface characterization probes use beams of ions, electrons or photons and include ISS, SIMS, AES, and XPS. Examples of the use of these techniques along with microscopy are shown for adhesive bonding research on aluminum, titanium and steel.

W. L. Baun

Practical Adhesion Measurement in Adhering Systems: A Phase Boundary Sensitive Test

The strength of adhesive bonds is assessed by means of mechanical tests in which an increasing load is applied until failure occurs. This paper describes a three point flexure test using a single adherend/adhesive structure. Contrary to what occurs with usual tests such as single or double lap shear and peel tests, this configuration is sensitive to the adherend surface chemistry variations and adhesive cure conditions and thus allows evaluation of the properties of the adhesive bond joint interphase. Data are presented which demonstrate the flexure test sensitivity to the effects of surface prebonding treatments on interphasial mechanical properties, such as rigidity modulus. Photoelastic isochromatic fringes are recorded simultaneously with flexure test data as a means to monitor stress distribution, failure initiation and crack propagation. This test shows that the stresses are uniformly distributed along the specimen length and remain parallel to the interfacial region and insure failure totally along the adherend/adhesive interface.

A. A. Roche, M. J. Romand, F. Sidoroff

Effect of a Phenol Formaldehyde Resin of the Adhesion of a Polysulfide Sealant

The results obtained in a study to evaluate the effect of a phenol formaldehyde resin on the adhesion of a polysulphide sealant is presented. The adhesion of the sealant is improved significantly by the incorporation of phenol formaldehyde resin. This is explained on the basis of increased boundary layer strength and the results obtained are compared with the ‘theory of attachment site’. Addition of phenol formaldehyde resin to the sealant caused an increase in cross-link density, ultimate tensile strength, stress at 100% elongation, hardness and a decrease in ultimate elongation and water absorption. The observed property changes are explained on the basis of polymer-filler interactions. Higher concentrations of phenol formaldehyde resin in the sealant caused a decrease in bond strength, cross-link density and mechanical properties. This is attributed to the lesser extent of cure of polysulphide resin.

R. Ramaswamy, P. Sasidharan Achary

Polysulfide-Polyurethane Interfacial Aspects

We studied polyurethane coating aging and determined polyurethane coating/polysulfide sealant interfacial integrity using scanning electron microscopy. A rapid microspecimen hand-pulled peel test was developed that will find application in designing future aircraft sealants. Dynamic mechanical analysis was used to follow polyurethane aging.

A. M. Usmani

Epoxy Resin Wetting of E Glass Single Filaments as it Relates to Shear Strength

The work of adhesion, W, of epoxy resin, water, and methylene iodide on sin e E-glass filaments was measured using a modified Wilhelmy apparatus. The resin used was a bisphenol A epoxy resin, and the fibers were coated with differing concentrations of silanes. Shear strengths were measured usint a single filament adhesion test. It was found that increasing Wa of resin on the fibers correlates with increasing shear strength of dry and water-boiled composites. This result and the results of methylene iodine and water wetting are analyzed in terms of the geometric mean approximation for the work of adhesion, and good agreement is found.

Elisabeth J. Berger, Yona Eckstein

Factors Influencing Behavior of Adhesive Joints

Frontmatter

Influence of Surface Roughness on Mechanical Properties Of Joints

Mechanical properties of joints are dependent on the geometry of surface defects caused during manufacture of the adherends. This paper presents experimental results on the influence of the roughness, due to grinding and sand blasting, on tensile properties of joints. We studied specimens made of low carbon steel rods, bonded by an epoxy resin containing mineral fillers. The surfaces were prepared with various roughness conditions: 2 grindings (coarse and fine), 8 grindings with sand blasting (combination of 2 grindings and 4 diameters of sand particles), and one shot blasting. The roughness parameters were measured, particularly the total depth and the average spacing of roughness. These measurements showed that the prepared surfaces had well defined and reproducible roughness. Electric strain gauges at specific points on the metallic parts were used; these provided measurements of local strains (at ±1 μm/m) during the loading. The variation of these strains represents indirectly the joint behavior: the elastic limit, the appearance of micro-cracks and their propagation up to failure.

Y. Gilibert, G. Verchery

The Effects of Surface Conditioning of AISI-304 Stainless Steel on the Interfacial Properties of Alloy/Epoxy Composite Structure; Adherend Surface Characterization Using X-Ray Emission Spectroscopy (LEEIXS)

AISI-304 stainless steel adherend-epoxy bonded specimens were subjected to a three point flexure test. This test used single interface type specimens. The measurements were conducted in order to: (i) investigate the effects of adherend surface reproducibility and metal surface conditioning on joint strength, (ii) determine the sensitivity and repeatability of the test, (iii) gain a better idea of the range of applicability of the test.

F. Gaillard, A. A. Roche, M. J. Romand

Surface Characterization of Anodic Oxides on Aluminium Alloys by Means of Surface Potential Difference, Surface Impedance and Surface Morphology

To obtain acceptable adhesion and durability of adhesive bonded structures, anodizing of aluminium alloys in chromic acid or phosphoric acid is commonly used in the aircraft industry. In order to find a non-destructive testing technique to inspect pretreated parts and to gain further understanding, we used some simple techniques to characterize surfaces. Using Transmission Electron Microscopy (TEM) as a tool to characterize surface morphology showed that porous oxides gave the best adhesion quality. The Surface Potential Difference (SPD) measurement was found to be very sensitive to adsorbed dipole layers on surfaces. Surface Impedance (Z) measurements were introduced for determination of the sealing quality of anodic oxides, but can also be used to characterize non-sealed oxide layers. On the basis of experimental data obtained, the effects of several variables in the anodizing process on SPD and Z measurements and TEM morphology are discussed. Also the effects of ageing of anodic oxides by dipping in ambient temperature water are shown.

A. Kwakernaak, R. Exalto, H. A. van Hoof

Investigation into the Effect of Surface Treatment on the Wettability and the Bondability of Low Surface Energy Materials

An experimental effort has been undertaken to examine the effect of surface treatment on various low surface energy thermoplastic materials to promote wettability and bondability of these substrates. Changes in wettability were followed by contact angle measurements. These measurements were correlated with the bondability of treated surfaces using a usual twopart epoxy adhesive. Air-plasma cleaning is a more effective treatment than the usual chemical-acid etching in increasing the surface energy of polymers to make them more wettable and bondable, except for polypropylene and for polytetraf luoroethylene. This work has enabled several satisfactory solutions for bonding these substrates by selecting suitable surface treatments and adhesives for each.

J. P. Jeandrau

The Effect of Moisture on the Dimensional Stability of Adhesively Bonded Joints

Torsion shear creep tests were made on two bisphenol A-type epoxy adhesives as a function of temperature, humidity, and applied stress. The effects of varying the filler content of the two systems were also determined. The relationship between creep rate and stress was logarithmic. Although the presence of moisture had a much greater effect in the unfilled epoxy systems than in the filled systems, the relationship between humidity effect and filler content was not a simple one. The results are discussed in terms of the residual stresses generated by temperature changes and by moisture absorption. The non-monotonic nature of the creep rate-stress relationship previously shown to exist in a number of filled epoxy systems is explained in terms of the combined effects of thermal and moisture-induced stresses.

E. J. Hughes, J. Boutilier, J. L. Rutherford

The Dimensional Stability of Epoxy Adhesive Joints

An interferometric technique which employs Moiré patterns has been used to study the deformation that occurs when an adhesive joint undergoes water uptake. It has been demonstrated that very large swelling stresses can be generated as a result of the inhomogeneous swelling. The magnitude and position of these stresses can be obtained by application of equations developed from beam and elasticity theory. A consequence of this analysis is the prediction of damped normal displacement waves in the cover slip in regions of negligible water concentration. This is observed experimentally. Measurements are reported on the irreversibility of dimensional changes when an adhesive joint is subjected to uptake and subsequent removal of the water responsible for swelling — on the assumption that the epoxy based adhesive is free from shear stresses during the cure — and on the influence of dissolved inorganic salts upon the swelling behaviour.

J. P. Sargent

Strength Characteristics of Mono and Multiple-Wire Steel to Steel Joints Bonded with an Epoxy Adhesive

Strength characteristics of mono-wire and multiple-wire specimens were investigated as support work for a research project concerned with the pre-stressing (energising) of structural materials. 1,2,3,4,5 The mono-wire joints were composed of cylindrical steel to steel interfaces, the adhesive used being Araldite AY 103 with Hardener HY 991. Compression and tension joints were concentrically constructed and after curing, axially loaded to failure. Nine glue line thicknesses and five bond lengths were selected ranging from 0.001 in to 0.100 in (0.0254 mm to 2.54 mm) and 0.50 in to 2.50 in (12.7 mm to 63.5 mm) respectively. Each test was performed five times. The strongest joints were obtained with the thinnest glue lines and increases up to 0.060 in (1.524 mm) diminished the strength by approximately 32% for both compression and tension specimens. For glue line thicknesses from 0.60 in to 0.100 in (1.524 mm to 2.54. mm) the joint strength was almost constant. Test results displayed a coefficient of variation for the joint strength of between 6.89% and 13.41%.

R. W. Hylands

A Study of Adhesive Joints Between Aliphatic Polyamides and Metals

Injection-moulded adhesive joints of aliphatic polyamides and metals were studied and showed that intermediate thin adhesive films applied from polymer or oligomer solutions onto metals is an efficient means of monitoring the bond strength. Contact events in a metal/thin polymeric film/polyamide system have been investigated and the results are described here. The thickness of intermediate films whose optimum values range from a few tenths to 1–2 μm were found to be a most important parameter that determines the adhesive strength. The bond strength improvement is most obvious during a thermal contact between the intermediate film and metal in air. Adhesive bonds are formed between the metal and the film which are resistant to dynamic stresses and liquids. The bond between polyamide and the intermediate film is usually formed owing to molecular interactions of polar groups in macromolecules at the interface. The joints were found to fail cohesively through the intermediate film. The improved adhesion to metals and cohesive strength of thin intermediate films are believed to result from the fact that thermo-oxidative degradation is inhibited in these during the thermal contact with metal.

S. S. Pesetskii, V. E. Starzynskii, S. V. Shcherbakov

“Honeymoon” Phenolic Fast-Setting Adhesives for Exterior-Grade Finger Joints

All fast-setting adhesive systems tested proved satisfactory for accelerated finger joints production as they show development of strength at ambient temperature rapid enough to allow finger joint machining five to 30 minutes after manufacture, according to the resin system used. The results of accelerated weathering tests showed all the adhesive systems to be weather- and boil-proof. The main advantages obtained were: (i)A considerable decrease in the cost of phenolic adhesives for fingerjointing obtained by the use of tannin extracts, and(ii)improvement in production flow obtained by the decrease of the time delay between the manufacture of the finger joints and their machining or dispatch from an overnight period to approximately ten to 30 minutes at ambient temperature according to the adhesive system used.

A. Pizzi

Ageing of Structural Film Adhesives — Changes in Chemical and Physical Properties and the Effect on Joint Strength

Epoxy-based film adhesives are extensively used in structural aircraft applications but although the one part nature of these materials has many advantages in terms of ease of use, the short shelf-life can be a serious disadvantage, especially when the material spends lengthy times in transit between manufacturer and user. Studies on a number of epoxy and nitrileepoxy adhesives have shown that slow cure, hydrolysis of the resin and specific interactions between components can occur during storage which result in modification of various chemical and physical properties of the uncured adhesives. The relative importance of these reactions depends on the adhesive composition. These modifications are reflected in changes in the strength of joints made with aged adhesives. This paper presents examples of these effects drawn from the results of a number of ageing studies.

C. E. M. Morris, P. J. Pearce, R. G. Davidson

Structural Adhesives Based on Diallyl Phthalate

This paper describes diallyl phthalate resins as a candidate replacement for the moisture sensitive epoxies used in formulating structural adhesives. Polymerized diallyl phthalate had most of the performance properties desired but was excessively brittle. Flexibilizing monomers were evaluated. Some promising copolymers were defined. Diallyl phthalate/epoxy interpenetrating polymer networks were also explored.

A. M. Usmani

Possibilities of Integrating Surface Treatment of Bonding Parts in the Adhesive Bonding Process

The conventional technique of adhesive bonding of metals is characterized in most cases by a time-consuming and expensive surface treatment of bonding parts before applying the adhesive and involves industrial safety and environmental problems.

Klaus Ruhsland

Evaluation and Characterization

Frontmatter

Evaluation of Adhesive Test Methods

Three approaches to bond strength analysis are discussed in this article: those based upon linear elastic stresses, those based upon linear fracture mechanics parameters, and those based upon nonlinear analyses. The first two approaches are then discussed with respect to standard adhesion test methods. The stresses in linear lap shear test specimens are shown analytically and experimentally to be concentrated at the bond termination and to be strongly dependent upon adherend thickness for standard tests. The bond stress intensity factors and energy release rates near the bond termination are given and discussed. Adherend thickness should be at least 4 times larger than that recommended by standard test techniques. The stresses in butt tensile tests are also discussed. It is shown that test specimen alignment problems can lead to low debond forces and excessively large data scatter. Stress intensity factors are also determined for butt tensile tests.

G. P. Anderson, K. L. DeVries, G. Sharon

A Critical Appraisal of Dental Adhesion Testing

The various methods used to investigate dental adhesion are reviewed. Concepts from fracture mechanics are used to interpret adhesive and cohesive failure of adhesives used in dentistry and to analyze bond strength testing techniques. The practical strength of an adhesive interface is a complex relationship between the stress distribution, specific adhesive fracture energy and the size and shape of interfacial defects. Bond strength tests are not designed to investigate these separate aspects related to adhesive failure. The problem of defects in dental adhesive failure are unique in that the number, size and shape of defects which result in failure are mainly technique related. However, the investigation of dental adhesive failure can generally be characterized as ignoring the relationship that specific adhesive fracture energy and defects have on the practical strength of adhesive systems. Based on theoretical considerations we propose that dental adhesion testing be reevaluated from a theoretical, experimental and most significantly from a practical viewpoint.

W. J. O’Brien, S. T. Rasmussen

Adhesive Joint Characterization by Ultrasonic Surface and Interface Waves

Methods of measurement of the velocity and attenuation of ultrasonic waves have been used for a long time for investigating structural and mechanical properties of materials. The main purpose of this article is to show that a properly updated ultrasonic method can be successfully used for investigating surface and interface films, in particular, of adhesively bonded structures. Theoretical analysis of the propagation of elastic waves in a multi-layered medium can be used for relating the velocity of surface and interface waves with the elastic properties of the film and substrates. If the elastic properties of the substrates are known, then the viscoelastic properties of films can be determined from measurements of the velocity and attenuation of these waves. The velocity of surface and interface waves is a function of not only the film properties, but also of adhesion of the film to the substrates. This makes the suggested technique potentially promising for the evaluation of adhesion properties of thin surface and interface films.

S. I. Rokhlin

Nondestructive Evaluation of Some Bonded Joints

Free films and lap-shear bonded metal joints were compared with respect to their linear dynamic viscoelastic behavior. Thermosetting and thermoplastic adhesives were investigated. Damping phenomena and relative shear moduli indicated a broader mechanical relaxation occurring at higher temperatures in the bonded assemblies. Triblock copolymer model adhesives were used to illustrate the importance of sample preparation in the evaluation of bonded materials.

T. C. Ward, Margaret Sheridan, D. L. Kotzev

Ultrasonic Assessment of Cure Rate Effects in Bonded Honeycomb Structures

The effect of variations in cure temperature on fillet formation in adhesively bonded honeycomb structures has been defined in terms of fillet length, degree of cure and Tg of the cured neat adhesive. The results are correlated with predicted behavior based on ultrasonic NDE analysis prior to testing. The ultrasonic test method uses advanced pattern recognition analysis of attenuation and velocity data extracted from a multi-frequency Lamb wave propagated along the face sheet of the honeycomb. Laboratory applications of this test method are described.

R. A. Pike, R. S. Williams

The Three-Point Bend Test for Adhesive Joints

Unless a large amount of degradation has occurred in the interfacial region of a metal-to-metal adhesive bond, the T-peel and wedge opening test usually provide only mode I information on the bulk adhesive. The single-lap shear test shows primarily mixed-mode failure. However, the thickness of the adherend must be carefully considered in order not to have a large amount of mode I failure. When these three tests failed consistently to find defects that were incorporated into the interfacial region of test specimens, the three-point bend test geometry was evaluated. This initial study, utilizing the short beam shear test geometry, has shown that several pieces of information may be obtained from the experimental data. The initial portion of the test curve appears quite linear and the slope of the curve seems directly related to the bulk properties of the adhesive. The portion of the curve where a definitive failure occurs appears directly related to interfacial failure of the bonded joint.

N. T. McDevitt, W. L. Baun

Accelerated Aging Procedures for Glue-Wood Bonds and their use for In-Plant Quality Control

Bond degradation in a particulate material like waferboards or oriented strand boards bonded with synthetic adhesives like phenol formaldehyde or urea formaldehyde can be explained as a failure or the breaking of the three dimensional mechanical branching of the cured adhesive due to the stresses generated by shrinking and swelling of wood as a result of moisture changes, as well as due to the chemical degradation of the adhesive which is hydrolytic in nature. In a service environment, the principal factors causing glue-wood degradation are (1) heat, (2) moisture, (3) stresses (generated during shrinkage and swelling), (4) microorganisms, and (5) chemicals.

R. B. Jathar

Peel-Strength and Energy Dissipation

The effect of energy dissipation on peel-strength is discussed on the basis of mechanics. In the case of T-peeling of specimens consisting of two flexible tapes bonded by elastic adhesive, steady peel-strength is simply given as 2f = Γ, where f is the peeling force per unit width of the tape and Γ is twice the surface energy of the fracture surface. However, in the cases where the energy is dissipated in the adhesive layer or in the adherends, another term w’ representing the energy dissipation is to be added to the peeling force, i.e. 2f = Γ + 2w’. The energy w’ associated with the deformation of the adhesive layer is given by the work done at the tip of the crack per unit length of the adherend during a cycle of the elongation to fracture and subsequent recovery of the adhesive. This energy w’ can be estimated from an area u’ enclosed by the stress-strain hysteresis curve obtained from extention test of the adhesive and the thickness h of the adhesive layer, as 2w’ = hu’.

T. Igarashi

The Estimation of Adhesion in Filled Polymer Systems

Some new methods of estimating the adhesion strength between filler particles and polymer matrix are proposed. All the methods are based on the determination of stresses at which the continuity of samples is disturbed.

Yu. S. Lipatov, T. T. Todosiychuk, P. K. Tsarev, L. M. Sergeeva

Durability or Stability Aspects

Frontmatter

Adhesion and Durability of Metal/Polymer Bonds

This paper reviews the results of a comprehensive investigation made at the author’s laboratories to determine those factors responsible for promoting the integrity and long-term durability of metal/polymer bonds used in the fabrication of aircraft and aerospace structures. Using a multidisciplinary approach and a variety of surface analytical techniques, such as extended resolution scanning electron microscopy (XSEM), X-ray photoelectron spectroscopy (XPS), ellipsometry, and a new technique we call surface behavior diagrams (SBD), we have evolved several important concepts. First, we have determined that the initial integrity of metal/polymer bonds used for structural applications depends critically upon the morphology of the surface oxide on the metal. In the case of the metals studied, Al and Ti, we have observed that certain etching or anodization pretreatment processes produce oxide films on the metal surfaces which, because of their porosity and microscopic roughness, mechanically interlock with the polymer forming much stronger bonds than if the surface were smooth.

J. D. Venables

Failure Mechanisms in the Boundary Layer Zone of Metal/Polymer Systems

A systematic optimization of the long term stability of polymer/metal systems in the boundary zone requires the exact knowledge of the mechanisms producing the so-called adhesional failure particularly under the influence of humidity. Strong and water-stable adhesion, for example, between organic primers and aluminium alloys is obtainable if the metal surface is coated with a porous oxide layer, which can be produced by special etching and anodizing processes. The relatively small molecules of the primers can penetrate practically the whole oxide structure. But despite this, it is easy to demonstrate that very small differences in the parameters of the surface treatment, leading to small differences in the surface morphology, can change the water stability of the boundary zone considerably. The transition zone between the oxides and primers in such systems was investigated by using the wet peel test, transmission-electron microscopy of ultra-thin cross sections from the boundary zone before and after delamination and ESCA on the fracture surfaces.

W. Brockmann, O.-D. Hennemann, H. Kollek

Joint Durability Studies with Abraded, Etched, Coated and Anodized Aluminum Adherends

The effect of surface pretreatment on the aluminum oxide surface is critical and intimately related to the bond permanence of aluminum adhesively bonded joints. It is difficult, however, to distinguish the benefit of different surface treatments in the absence of some weathering test condition since the joints will fail cohesively for many different surface-treating conditions when tested without such exposure. Aluminum alloy 6061-T6 was pretreated by various procedures including sanding, grit blasting, hot acid etching, heat-cured epoxy paint priming, Alodining, and sulfuric acid anodizing. Both room-temperature-curing two-part and one-part heat-curing epoxy adhesives were used to fabricate the test specimens. The weathering evaluation procedures included both accelerated laboratory-type and natural atmospheric exposures. The list of weathering conditions included: immersion in room-temperature water, exposure to continuously condensing humidity at 52°C (125 F), and hot water soak/freeze/thaw cycling; as well as, exposure in an industrial-type atmosphere and a seacoast atmosphere.

J. Dean Minford

Comparative Study of Aluminum Joint Strength and Durability with Varying Thickness, Boehmite-Type Oxide Surfaces

The presence of boehmite oxide on aluminum surfaces has often been associated with weather-durable bonding. In this work, a considerable variance was shown in aluminum joint durability involving such oxide surfaces with varying thicknesses as generated in the absence or presence of electrolyte in boiling water; i. e., deionized or tap water. The changes in oxide were generated starting from a deoxidized surface base by varying the exposure times in boiling water.

J. D. Minford

Epoxy Adhesion to Copper, Part II: Electrochemical Pretreatment

Electrochemical methods of pretreating copper surfaces for adhesive bonding were studied. The objective of the electrochemical pretreatment was to control the oxide layer under the copper-coupling agent complex film, since this layer was found to be the locus of failure of copper/epoxy bonds aged in boiling water. Both cathodic and anodic pretreatments resulted in improved durability of the joint after exposure. Experiments were also performed to understand the mechanism of weakening of copper/epoxy bonds in boiling water.

Jae M. Park, James P. Bell

Hydrothermal Stability of Titanium/Epoxy Adhesive Joints

γ-Aminopropyltriethoxysilane (γ-APS) was shown to be an extremely effective primer for improving the hydrothermal stability of titanium/epoxy adhesive joints when applied to the adherends by adsorption from dilute aqueous solutions prior to adhesive bonding. The breaking strength of lap joints prepared from unprimed adherends decreased slowly during immersion in water at 60°C and was only about 960 psi after 60 days. The breaking strength of joints prepared from adherends primed with γ-APS at pH 10.4 and 8.0 decreased very little during similar hydrothermal aging and was about 1750 psi after 60 days. Joints prepared from adherends primed with γ-APS at pH 5.5 were somewhat less durable and had a breaking strength of about 1180 psi after 60 days in water at 60°C. The performance of γ-APS as a primer was very different for titanium and iron adherends even though the molecular structures of the primer films formed on the two substrates were very similar. As a result, the performance of γ-APS primer films could not be related to the overall molecular structure of the films but was related to acid/base interactions between γ-APS and the oxidized surfaces of titanium and iron.

F. J. Boerio, R. G. Dillingham

A Study on Elastomer/Metal Bonds Applicable in Underwater Sonar Systems

Recent emphasis on the long-term performance of sonar systems requires that investigations be carried out to improve the durability of elastomer-to-metal adhesive bonds in water. A screening test of some commercially available adhesive systems was performed. ASTM-D429 test methods using both conical button specimens and peel-strips were employed. The 90-degree peel test was modified such that the effects of stress corrosion, water salinity and gas content could be examined. Most commercial systems were found to give adequate dry bond strength. Preliminary environmental aging studies also showed that thermal aging did not produce any measurable effect in accelerating the degradation of the elastomer/metal bond. However, chemical attacks from salt water under stress caused great reductions in the bond strength, especially when the oxygen supply was abundant. The weakened bonds were found to show increasingly large portions of fracture failure at the adhesive/metal interface rather than rupture failure of the bulk rubber.

Robert Y. Ting

Stress Analysis and Performance Aspects

Frontmatter

Life Prediction Methodology for Adhesively Bonded Joints

A comprehensive integrated-methodology for adhesive bonded joint life predictions is outlined. The guiding assumption is that the useful life of bonded joints is determined by failure in the adhesive interlayer, and this is the basis for a systematic analysis of information and the techniques required to provide valid predictions. Emphasis is placed on time-dependent fracture mechanics, procedures, including detailed through the-adhesive thickness viscoelastic finite element calculations of the stress-strain distributions, and on analytical methods involving the constitutive relations of the adhesive interlayer. Use is made of instrumented bonded joint data obtained from structural overtest methods under a number of U.S. Air force-sponsored programs. A logical program rationale is outlined which has sufficient generality to apply to any adhesive bonded joint structure including metal and composite adherends, various loading/environmental conditions, which apply to high-performance aircraft. The five main modules of the integrated methodology are: (Definition of) Loads/Environments; (Adhesive) Material Properties; Stress Analysis; Fatigue Mechanisms and Failure Criteria; and Structural Failure Analysis (Life Predictions).

J. Romanko, K. M. Liechti, W. G. Knauss

Stress Analysis of Adhesively Bonded Joints

In this study, the effect of adhesive bondline thickness on the stress distribution in an adhesively bonded joint is examined. A linear variation in bondline stresses through the adhesive thickness is used to model this effect. In this way, an important influence on bond behavior, heretofore neglected or incompletely modeled, is studied. Also presented is a parametric study of the effect of various joint parameters on bondline stress distribution when thickness effects are included.

Ronald A. Kline

The Impact Strength of Adhesive Lap Joints

Structural adhesive joints may be required to withstand high loading rates. The effect of impact loading on the strength and energy absorption of single lap joints has been investigated. A rubber-toughened epoxy adhesive, in bulk form, shows significant changes in mechanical response, when tested in shear over a wide range of strain rates. Despite large changes in the response of the adhesive, the strength of lap joints is only slightly reduced by impact loading. When high strength adherend materials are used, energy absorption is relatively small; however, it is increased by using lower strength adherend materials, as significant plastic deformation can take place in the adherend prior to joint failure.

J. A. Harris, R. D. Adams

The Performance of Adhesive-Bonded Thin-Gauge Sheet Metal Structures with Particular Reference to Box-Section Beams

In an effort to reduce vehicle weight, the automotive industry is developing car body structures made from light-weight materials such as composites, plastics and aluminium alloys. Fabrication of these materials in high volume automotive applications using traditional welding techniques is not feasible and adhesive bonding is now being investigated as a potential assembly method. In order to assess performance characteristics of bonded vehicles, thin-gauge sheet-metal box-section beams have been used to simulate structural details in automotive applications such as car bodies and commercial vehicles. Beams were fabricated from flanged strips by different joining methods to form box-section structures approximately lm. long x 60mm. square. Tests were carried out to determine torsional and flexural rigidity and ultimate flexural strengths, and in the majority of tests, bonded structures gave better characteristics than the equivalent riveted or spot-welded beams. The failures of beams under 3-point bending have been related to buckling of the side webs and further experimental tests have shown that collapse is critically dependant on flange-bend radius. Finite element techniques have been used to analyse stress distribution in the beam section and this confirms the experimental observations of beam collapse.

A. Beevers, A. C. P. Kho

Cyclic Debonding of Adhesively Bonded Composites

To analyze the fatigue behavior of a simple composite-to-composite bonded joint, a combined experimental and analytical study of the cracked-lap-shear specimen subjected to constant-amplitude cyclic loading was undertaken. Two bonded systems were studied: T300/5208 graphite/epoxy adherends bonded with adhesives EC 3445 and with FM-300. For each bonded system, two specimen geometries were tested: (1) a strap adherend of 16 plies bonded to a lap adherend of 8 plies, and (2) a strap adherend of 8 plies bonded to a lap adherend of 16 plies. In all specimens tested, the fatigue failure was in the form of cyclic debonding with some 0°fiber pull-off from the strap adherend. The debond always grew in the region of adhesive that had the highest mode I (peel) loading and that region was close to the adhesive-strap interface. Furthermore, the measured cyclic debond growth rates correlated well with total strain energy release rates GT as well as with its components GI (peel) and GII (shear) for the mixed-mode loading in the present study.

S. Mall, W. S. Johnson, R. A. Everett

Effects of Low Cycle Loading on Shear Stressed Adhesive Bondlines

Adhesively bonded thick adherend model joints were cyclically loaded on squarewave and sinusoidal modes. Durind the cycling loading,the real shear strains of the bondlines and, for the first time, the failure strains were quantitatively measured by use of a special shear strain extensometer. The analysis of bondline strains of a modern structural adhesive indicates dependence on time, shear stress and cycle frequency, and suggests a strengthening of the adhesive in the bondline under cyclic loads. A linear viscoelasticity can be expected only at very low bondline shear stresses. At higher stresses, residual strains build up to a critical level and exceeding this level causes failure of the bondline. Adhesive material properties obtained in the present study provide a guide to allowable shear stresses or strains to design cyclically loaded adhesively bonded structures.

W. Althof

Effect of Scrim Cloth on Adhesively Bonded Joints

This report discusses the effect of the scrim cloth on the fracture behavior of adhesively bonded joints. The thick-adherend model joint specimen was analyzed with the TEXGAP finite element computer code. Three situations were considered: (1) uniform adhesive without inclusion (scrim), (2) adhesive with bonded inclusions, and (3) adhesive with debonded inclusions. The material properties and specimen geometry were those of a model joint previously analyzed without inclusions by General Dynamics/Fort Worth Division (GD/FWD).

E. C. Francis, D. Gutierrez-Lemini

Fracture Aspects

Frontmatter

Mechanical Measurement of Interatomic Bonding Energies at Interfaces

Until recently it has not been possible to relate, quantitatively, the mechanical strength of adhesive joints to the surface chemistry of the system.

E. H. Andrews

Review of Continuum Mechanics Factors in Adhesive Fracture

The possibility of treating adhesive fracture as an engineering analysis problem in continuum mechanics has been emphasized in previous reviews, with applications to designs which utilize bonded surfaces and/or various composite materials. Cost-effective use of this analysis ability will ultimately depend, of course, upon the accuracy of fracture and structural life prediction, as well as the availability of methods to conduct non-destructive examination.

M. L. Williams

Fracture Energetics of Adhesive Joints

In this review, we relate various fracture energetics theories to the types of polymers commonly used in the adhesive joints. The energy-balance concept of Griffith’s fracture criterion is emphasized. The same concept led to the development of various methods of determining the fracture energies (or the failure energies) of both rigid and flexible adhesive joints.

Lieng-Huang Lee

Fracture of Composite-Adhesive-Composite Systems

This program was undertaken to initiate the development of a test method for testing adhesive joints in metal-adhesive-composite systems. The uniform double cantilever beam (UDCB) and the width tapered beam (WTB) specimen geometries were evaluated for measuring Mode I fracture toughness in these systems. The WTB specimen is the preferred geometry in spite of the fact that it is more costly to machine than the UDCB specimen.

E. J. Ripling, J. S. Santner, P. B. Crosley

Characterization of the Fracture Behavior of Adhesive Joints

The desire to use adhesives and composites in structural applications has led to a need for a failure prediction capability for the polymers used in such systems. Unfortunately, this task is greatly complicated by the failure load being dependent not only upon the specimen geometry but also on the previous history of loading, temperature, environment, etc. For the tough, rubber-modified polymers that are of most interest for structural applications the effects of previous history can be dramatic. As a result, predictions based on measurements at a single set of conditions can lead to dangerous over or under estimates of the fracture behavior. In an effort to understand this problem the present work has studied the fracture behavior of various polymer formulations using bulk and adhesive joint specimens tested over a variety of different thermal and loading histories.

D. L. Hunston, A. J. Kinloch, S. J. Shaw, S. S. Wang

Structural Precursors to Fracture in Adhesive Joints

Sructure-property relations in adhesive joints are reviewed for the purpose of explaining the general nature of conditions and structural processes leading to fracture. Reference is made to selected homogeneous materials and adhesive joints to illustrate specific features. The adhesive joint is described as a composite structure composed of adherend and adhesive layers and various chemical and structural transition layers. The nature and dimensions of materials structural features are considered. Traditional and modern methods of analysis are summarized with special emphasis placed on the use of materials characteristic curves to indicate the existence of structural precursors to fracture. These features, in relatively small quantities, severely alter the mechanical properties of the adhesive joint. In only a few instances have structural precursors to fracture been directly observed and identified due to their small dimensions, distribution in the system and limited quantity. It is concluded that identifying the presence of structural precursors to fracture by the use of materials characteristic curves may aid the avoidance of adhesive joint failure.

Wartan A. Jemian

Fracture Toughness of Elastomer Modified Epoxy Adhesives

Elastomer modification is an effective way of toughening glassy epoxy resins. The toughening mechanisms of phase separation and gellation have been studied extensively. This paper reports the effects of carboxyl terminated acrylonitrile rubber on several single component epoxy adhesives, based on Epon 828 and cured with dicyandiamide. Variables include accelerator type and level, cure temperature and elastomer level. There is an optimum amount of rubber which may be added to each system to yield maximum shear strength. Peel strength is maximized in two regions. Rapid gellation of these adhesives may be detrimental to the phase separation and chain extension mechanism and thereby cause lowering of fracture toughness and strength properties.

A. A. Donatelli, C. T. Mooney, J. C. Bolger

A Three-Dimensional Analysis of a Butt Joint with a Flaw

A three-dimensional finite element method is employed to analyze an adhesive butt joint with a through-the-thickness crack in the center of the adhesive layer. The finite element model used accommodates both the extremely small dimensions of the adhesive layer and the larger dimensions of the rest of the region. The crack tip region is modelled by isoparametric quarter-point degenerate brick elements which give the required square root singularity for stresses at the crack tip. The results from a fracture mechanics approach corroborate the earlier experimental results that the strength of an adhesive joint increases with decrease in adhesive layer thickness. It is also observed that the stress intensity factor varies in a non-linear fashion along the crack front.

R. S. Alwar, K. N. Ramachandran Nambisan

The Influence of Layer Thickness and Internal Stresses on the Bond Strength of Metal-to-Ceramic Joints

Sandwich-like joints between Zr, Zr-Nb-Zr or Hf foils of different thickness and rectangular bodies of hot-pressed silicon nitride (HPSN) were fabricated by solid-state bonding at 1150 °C or 1200 °C. Bend bars of 30x5x2 mm were cut from the welded pieces and notches of variable depth a were introduced at the metal-to-ceramic interface. The bond strength was described by critical energy release rates using experimentally determined correction functions and correction functions calculated by means of the FEM method. The effects of layer thickness and crack position on the correction functions of materials joints are discussed from a more general point of view by a comparison with the correction functions of isotropic materials and bimaterials. Measurements of the critical energy release rates of the metal-to-ceramic joints and microstructural investigations indicate that the reduction of internal stresses by microcrack formation and by allotropic transformation of the metal component control together with the metal layer thickness the bond strength of the sandwich-like joints.

W. Diem, G. Elssner, T. Suga, G. Petzow

Bond Strength Characterization of Metal-to-Ceramic and Adhesive Joints by Critical Energy Release Rates

Small three- or four-point bend test specimens of 30 to 60 mm length were used to characterize the bond strength of metal-to-ceramic and adhesive joints by critical energy release rates. The testing procedure involves the determination of correction functions which describe the influence of the geometry and the elastic properties of the joint. The correction functions are obtained by compliance measurements or can be calculated by means of the finite element method. A fracture resistance parameter KC derived from the critical energy release rate is introduced. KC allows a comparison of the bond strength of different joints without considering differences in their elastic properties. Examples of a bond strength determination of glass /epoxy joints, and of the influence of crystal orientation on the strength of sapphire/Cu joints serve to illustrate the utility of this bend test method to determine critical energy release rates of semi-brittle materials joints.

W. Diem, G. Elssner, T. Suga, G. Petzow

Interfacial Properties of Filled Epoxide Resins

The fracture properties of two commercial epoxide resins both unfilled and filled with varying volume fraction of silica, alumina and dolomite particles have been investigated. The variation in the stress intensity factor with the crack velocity was measured using the double torsion test technique. In order to examine the influence of the resin-filler adhesion on the fracture toughness, alumina particles were treated with three silane compounds. In addition the yield stresses and the flexural strengths were measured. In an attempt to si-mulate resin-filler interactions and to measure the fracture energy for interfacial separation a scarf joint was used with a bond angle of 45 °. This geometry produces a mixture of mode I and mode II loading. The combined mode strain energy release rate (G(I,II)c) was determined.

A. C. Moloney, H. H. Kausch, H. R. Stieger

Backmatter

Weitere Informationen