Proceedings of the Institution of Mechanical Engineers,Part E: Journal of Process Mechanical Engineering

Lightweight body-in-white is often based on a mix of materials that causes problems in the production process. More specifically, the paint oven process after cathodic dip-coating can lead to damaged adhesive layers due to mismatches in thermal expansion of the materials. The understanding of the curing behavior of the structural adhesive cured in this oven process is of crucial interest to determine the damaging in numerical analyses. Therefore, the curing behavior of a one-component toughened hot curing structural adhesive is modeled using three model-based and a model-free approach as well as experimental data from differential scanning calorimetry. After the parameter identification of the models, the parameters are validated using an oven-process-like temperature profile to compare experimental and numerical data.

The increasingly larger use of adhesive joints in the automotive industry demands a full comprehension of the adhesive behaviour when subjected to dynamic loadings. So far, some authors studied the effect of the strain rate regarding the adhesives performance, usually for the range of approximately 0–10⁵/s; nevertheless, few studies are clear regarding the method used to calculate strain rate, especially when fracture mechanics analysis is the focus of study. Those who present approaches to assess the values of strain rate usually consider a constant value for each test. In this paper, a numerical approach is proposed to assess the strain rate in modes I and II in double cantilever beam and end notch flexure tests, respectively. The results of this study demonstrate that the strain rate in the adhesive bondline along the crack propagation in the double cantilever beam and end notch flexure tests is not constant when loaded at a constant cross-head speed. This finding also helps to justify why experimental R-curves of double cantilever beam tests, when loaded at speeds above quasi-static conditions, do not have a perfect plateau contrasting with those, usually presented by numerical simulations, that do not take into account the effect of the strain rate.

This paper deals with the detection and localization of impurities such as release agent residues and fingerprints on fiber-reinforced plastic surfaces in an inline process step. This information can be used to carry out a locally individual and controlled surface pretreatment using laser radiation in order to improve the adhesion conditions. In this paper, surfaces from polyamide 6 reinforced with glass fiber are considered. To detect the impurities, inline images of the joining surfaces are taken using a laser line triangulation system. These image data are then examined with a specially developed algorithm of digital image processing. The real-time algorithm is based on the method of material-specific threshold values. In addition to real impurities, the algorithm was validated with artificial impurities. The central result of this work is the detection and local impurities of release agent residues and fingerprints down to a detail size of 0.3 mm.

Due to environmental concerns, modern transportation solutions demand drastic reductions of fuel consumption and emissions, which can only be achieved with advanced structures using high-performance lightweight materials. For joining these dissimilar materials, adhesive bonding appears as an optimal solution, since mechanical fastening adds weight to the structure, and welding technology is not easily applicable to reinforced plastics and composites. However, one of the major drawbacks associated with bonded joints is the presence of stress concentrations at the overlap ends, especially in single lap joints. In order to reduce these stress concentrations, several techniques have been developed. One of these is the functionally graded adhesive, in which the adhesive properties gradually vary along the overlap length, leading to a more uniform stresses distribution and improving the joint strength. However, the manufacture of an adhesive layer with properties which gradually vary is complex in practice and so is the creation of numerical models that represent these configurations. In the present work, a numerical model for different-graded distribution of adhesive properties along the overlap was developed, using programmed step functions on finite element analysis background in order to discretize and simplify the continuous properties distribution gradient. Cohesive zone modelling was introduced in the numerical model, enabling it to effectively predict graded joint strength. The model was validated with experimental results of functionally graded joints available in the literature. The numerical model developed presents itself as a powerful tool to predict joint strength for functionally graded joints, without imposing large computational demands.

Joint failure plays a key role in determining structural stability and crash or impact response. Characterizing the joints at high loading rates is challenging as oscillations are often overlaid on the measured data, making interpretation of the results more difficult. This paper builds upon the experimental testing three different mixed-material joints using a split-Hopkinson tension bar. The correction proposed in this work is verified using a finite element model of the entire testing system. The modeling efforts also investigate the differences in a specimen only model and a model including the entire testing system. The failure mechanisms of bolted and bonded joints are investigated, where the substrate stress state is found to play a large role in determining the failure mode for bolted joints. This work lays the foundations needed to investigate the mixed-material bolted and bonded joints in detail.

The use of adhesive bonds has attracted considerable interest from the scientific community. Stepped-lap joints have the advantage of decreasing stress gradients along the bond length, although the outer steps still encounter stress levels above the steps in the inner zone of the joint. One possible way to reduce this stress gradient is to combine this type of joint with the use of two adhesives. This work consists of an experimental and numerical evaluation of stepped-lap dual-adhesive joints between aluminum adherends, for various overlap lengths (L_O), and comparison with stepped-lap single-adhesive joints. The adhesives Araldite® AV138, Araldite® 2015, and Sikaforce® 7752 were evaluated. Numerically, cohesive zone models with a triangular damage law were applied in the joint behavior prediction. The analysis of the results is presented in the form of failure modes, stress analysis, damage variable analysis, load–displacement (P–δ) curves and maximum load (P_m), and energy required to failure (U). It was concluded that, in general, cohesive zone model presented precise predictions. In general, no significant increase in strength was achieved with dual-adhesive joint but, on the other hand, significant energy increases were obtained.

This paper addresses optimization of parameters and measuring setups for in situ computed tomography measurements on marker particle-filled adhesives. The focus of this work was to increase the detail detectability and discriminability between used marker particles and surrounding adhesive compound even for materials with a high X-ray attenuation. Therefore, it is necessary to reduce the effects of artifacts like scattered X-rays and beam hardening, respectively. A key benefit of being able to distinguish different materials clearly improves the correct interpretation of the reconstructed three-dimensional volume significantly, while a reduction of artificial disturbances enhances the possibility to visualize previously overlaid undetected details, e.g. air voids. Performing in situ computed tomography measurements by applying the optimized parameter setups, cavitation formation was observed investigating the particle behavior of specimens modified with glass beads as marker particles under applied load. An achieved bonding enhancement by using a coupling agent as pre-treatment for glass beads was also proven by means of in situ computed tomography. Furthermore, the parameter setups optimized for bulk specimen could be adapted on material combinations, e.g. single-lap shear specimen, by adjusting a few parameters. Additional experiments demonstrate that computed tomography measurements can also be used for analytical purposes, for instance to evaluate the mixing quality of the so-called QUADRO™ or 2C mixers for two-component adhesives.

The increased joint efficiency, distribution of loads and decrease in stress concentrations have led to the increased use of adhesives for structural bonding. However, there are a limited number of techniques for verifying and monitoring the integrity and durability of adhesive bonds. This article studies the potential of estimating the curing and ageing of adhesive bulk samples with embedded fibre Bragg grating sensors through measuring the strain associated with hygroscopic expansion. This is achieved by relating the output of a fibre Bragg grating sensor to the deformation of the structure in which it is embedded. This work considers the possibility of mapping the changing structural resistance to mechanical loading (stiffness) of adhesive bonds as a function of time, under the influence of temperature and moisture as environmental factors. The goal is to map the influence of these environmental factors separately on the one hand, and their combined effect on adhesive bonds, on the other hand. This study subjects several bulk specimens to various environmental ageing loads. The swelling, associated to moisture absorption and that results in mechanical strain, is measured with fibre Bragg grating sensors. The moisture absorption behaviour at different temperatures and environmental relative humidity conditions determined in this way is verified using classical test methods (e.g. differential scanning calorimetry, gravimetric) on multiple fibreless specimens.

This paper provides an insight into the investigation of long thick-layer adhesive façade joint resistance to negative pressure loads, i.e. wind suction. The real structural response to wind actions was simulated with reference to ETAG 034. Each specimen represented a reference façade section. The experiment focused on four different adhesive systems with flexible high strength 1-K polyurethanes and 1-K modified silyl polymers. Several variants of the test assemblies were tested: 1) a test assembly with an adhesive joint when a) the manufacturers’ application instructions for the system were followed and b) the instructions were violated; 2) a test assembly with a mechanical joint. These variants made it possible to compare the properties of both fastening methods, and moreover, to assess the impact of the mounting tape on the properties of the adhesive joint. The comparison of adhesive joints and a mechanical joint proved the greater structural stiffness and stress resistance of bonded assemblies. Monitoring showed that a local failure of the fastening element between the load-bearing frame and supporting structure caused the failure of the bonded assemblies, whereas the specimen with mechanically attached cladding failed due to pull-through of the fasteners. The average failure load of the bonded assembly was 10.88 kPa. In contrast, the failure load of the segment with mechanical fasteners was 10.12 kPa. Even though the difference in maximum pressure loads was only around 7%, the recorded values clearly demonstrate that the weakest part of the whole façade system is the mechanical joint, not the bonded one. Furthermore, the comparison of the results for segments with and without mounting tape showed that tape can have a major impact on the bond strength, since in case of the test specimens without mounting tape, the recorded failure load was a maximum of 30% higher.

Precise adherence to the manufacturer’s instructions and requirements plays an important role in various installation processes. The presented paper deals with the evaluation of the effect of manufacturing imperfections and surface defects on the failure behaviour of flexible adhesive intended for façade application. The failure to comply with the accepted procedures is more common in construction practice than in other sectors of the industry, mostly due to the surrounding conditions and lack of trained supervision. Unfortunately, this may lead to premature failure of adhesively bonded joints and a considerable shortening of the service life of the entire construction. To determine the potential of the risk, five types of artificially applied (a) manufacturing imperfections: (1) application on wet adhesion promoter, (2) application after the expiry of the laying-time, (3) curing of samples at +1℃ (b) surface defects: (4) application on a wet substrate and (5) application on a dirty surface, were suggested. Moreover, the Taguchi L32 orthogonal array design was used to arrange the test setup of all possible combinations. The 1 K polyurethane adhesive was applied in tensile butt joints and single-lap shear joints composed of aluminium alloy and thermally modified wood substrates. The obtained results confirmed that non-observance of the required manufacturing techniques and recommended procedures can have a negative impact on the decrease of the adhesively bonded joint strength and deformation behaviour. Surprisingly, the most critical was not the combination of all suggested types of imperfections and defects. The performed one-way ANOVA revealed that the most perilous was the combination of types 2 and 4 in the tensile test with 77% joint strength reduction. In the shear test, the most critical was the combination of all types of imperfection and defects which led also to a 77% drop of shear strength.

The use of sustainable hybrid components is an important topic in lightweight automotive applications. Wood being a renewable material, when used in combination with other materials such as technical polymers, offers a high potential for producing hybrid components and the implementation of innovative lightweight automotive materials. The feasibility of wood-based hybrid automotive components strongly depends on the properties of the interface between wood, lignin as a renewable coupling agent, and technical polymers. This paper investigates the macromolecular reactions and the bonding area in biobased epoxy adhesives for a specific influence on the performance of structural automotive wood components. Therefore, a typical bisphenol A diglycidyl ether epoxy adhesive was modified with lignosulphonate to increase the penetration depth. The composites were characterized by thermogravimetric analysis coupled with Fourier-transform infrared spectroscopy to validate the crosslinking of the macromolecules and the thermal stability of the adhesive. In the next step, a layer-by-layer composite was built up with the biobased adhesive and 1 mm beech veneer. The bonding area was characterized by scanning electron microscopy and compression tests.