Christophe Ageorges

Advances in fusion bonding techniques for joining thermoplastic matrix composites: a review

Abstract Joining composite materials is an issue because traditional joining technologies are not directly transferable to composite structures. Fusion bonding and the use of thermoplastic films as hot melt adhesives offer an alternative to mechanical fastening and thermosetting adhesive bonding. Fusion bonding technology which originated from the thermoplastic polymer industry has gain a new interest with the introduction of thermoplastic matrix composites (TPC) which are currently regarded as candidates for primary structures. The improvement of thermoplastic polymer matrices, with the introduction of recent chemistries such as PEEK, PEI and PEKEKK. exhibiting increased mechanical performance, service temperature and solvent resistance (for the semi-crystalline systems) also supported the growth of interest for fusion bonding. This review looks at the state of the art of fusion bonding technology and focuses particularly on the three most promising fusion bonding techniques: ultrasonic welding, induction welding and resistance welding. Physical mechanisms involved in the fusion bonding process for modelling purposes are discussed including heat transfer, consolidation and crystallinity aspects. Finally, the application of fusion bonding to joining dissimilar materials, namely thermosetting composites (TSC)/TPC and metal/TPC joints, is reviewed.

Experimental investigation of the resistance welding for thermoplastic-matrix composites. Part I: heating element and heat transfer

A comprehensive experimental investigation of the resistance welding of carbon-fibre and glass-fibre reinforced polyetherimide (PEI) laminates is presented. Lap-shear and double-cantilever-beam specimens were resistance welded by using fabric and unidirectional heating elements. The statistical distribution of the resistance of the heating elements was characterised, and the effects of the temperature on the heating element resistance were evaluated. The influence of the mechanical contact pressure on the contact resistance was also investigated. One of the main process parameters in resistance welding, i.e. the power density, was studied in detail. Heating uniformity in the heating elements was assessed through thermal imaging, allowing for a comparison between fabric and unidirectional heating elements. Temperature histories were measured and compared to those simulated by a three-dimensional transient heat transfer finite-element model. Factors limiting the size of the welded joint, i.e. temperature non-uniformity within the welding area and current leaking to the laminate, were investigated. Proper electrical insulation using a glass-fibre/PEI interlayer between the heating element and the laminate when joining carbon-fibre/PEI laminates effectively eliminated current leaking and enabled large-scale resistance welding.

Experimental investigation of the resistance welding of thermoplastic-matrix composites. Part II: optimum processing window and mechanical performance

Abstract An experimental investigation of the resistance welding of carbon-fibre and glass-fibre reinforced polyetherimide laminates is presented. The optimum resistance welding time based on a criterion of maximum lap shear strength was determined. The time required to achieve intimate contact predicted by a three-dimensional transient finite-element model featuring heat transfer and consolidation correlated well with the optimal welding time. The influence of the welding pressure on lap shear strength was investigated, and the consolidation quality obtained in the welded joint was related to the processing conditions. The extent of flow occurring during welding, or the reduction of thickness of the welded joints, was shown to be related to lap shear strength. Four failure mechanisms, leading to different values of lap shear strength, were identified including interfacial failure, cohesive failure of the heating element, tearing of the heating element and tearing of the laminate. Experimental and numerical processing windows were constructed and correlated well to each other. A comparison between fabric and unidirectional heating elements, in terms of lap shear strength and the interlaminar fracture toughness, GIc, was performed. It was demonstrated that large-scale lap-shear coupons and double cantilever beam specimens can be resistance welded providing that current leaking to the laminate is avoided.

Resistance welding of thermosetting composite/thermoplastic composite joints

Abstract An investigation of the resistance welding between carbon fibre (CF)-reinforced polyetherimide (PEI) and CF-reinforced epoxy laminates is presented. A three-dimensional transient finite element model (FEM) featuring heat transfer, consolidation and thermal degradation was used for simulating the process. A hybrid interlayer made of a glass fibre (GF) fabric essentially impregnated with PEI on one side and with epoxy resin on the other side was produced to provide mechanical interlocking between the thermoplastic (TP) and the thermosetting (TS) systems. The ‘optimal’ resistance welding time based on the maximum lap shear strength (LSS) was determined for three power levels and correlated to the time required to achieve bonding predicted by the FEM. Consolidation quality and failure mechanisms were discussed in relation with processing parameters. Experimental and simulated processing windows were constructed and correlated to each other. However, thermal degradation as predicted by the model did not correlate to a reduction in performance of the joint.

Characteristics of resistance welding of lap shear coupons.: Part II. Consolidation

Abstract A consolidation model based on transient three-dimensional heat transfer for the resistance welding of thermoplastic-matrix composite lap-shear specimens is established. The consolidation occurring in the resistance welding process was studied in terms of intimate contact and autohesion processes. Effects of power level on the time to achieve full intimate contact were determined. The influence of the consolidation pressure on the degree of intimate contact was investigated. Different welding configurations of lap-shear specimens were evaluated, i.e. APC-2 laminate/PEEK film, APC-2 laminate/PEI film and CF–PEI laminate/PEI film. The bonding time was compared with experimental electrified times, and close agreement was obtained. Local thermal degradation in the heating element was discussed for high power levels. A processing window for the CF–PEI/PEI configuration was established, which showed close agreement with that determined experimentally.

Characteristics of resistance welding of lap-shear coupons. Part III. Crystallinity

Abstract A transient three-dimensional heat-transfer, consolidation and crystallinity model for the resistance welding of thermoplastic-matrix composite lap-shear specimens is established. The heat-transfer model assumes orthotropic heat conduction in the composite parts and includes heat losses by radiation as well as natural convection. For the APC-2 laminate/PEEK film welding configuration, three crystallisation kinetics models are compared during the cooling stage of the resistance welding process. Cooling rates are predicted for natural cooling and the total processing time is determined. A coupled crystallisation kinetics/crystal-melting model is developed to predict the final crystallinity level in the welded joint. The effect of power level on the final crystallinity in the joint is investigated. Latent heat due to crystallisation and crystal-melting events is predicted and taken into account in the heat-transfer model. The influence of the environmental temperature on cooling rates and on the final crystallinity level in the joint is discussed. For the CF–PP laminate/PP film welding configuration, final crystallinity levels are predicted and compared with experimental data.

Single fibre transverse debonding: stress analysis of the Broutman test

The paper presents an extended analytical approach for the interfacial transverse stress that is generated by the Broutman test specimen under compression. The analysis is based on the division of the specimen into a bulk region and a near fibre region. Treating separately each region a compound equation for the interfacial stress can be derived. The equation also includes residual thermal stress and fibre anisotropy. A 3D finite element model was used to validate the approach. The calculations are performed for two commonly used material systems (carbon/glass fibre, epoxy resin). A comparison between the finite element results and the analytical solutions indicates that the accuracy of the analytical approach is very good.

Resistance Welding of Metal/Thermoplastic Composite Joints:

An investigation of the resistance welding between carbon fiber (CF) rein-forced polyetherimide (PEI) and aluminium substrates (7075-T6 grade alloy) is presented. A three-dimensional transient finite element model (FEM) featuring heat transfer, consolidation and thermal degradation was used for simulating the process. Two mechanisms are distinguished in the consolidation model: (1) removal of the initial surface profile of the laminate modelled by the establishment of intimate contact between the two substrate surfaces and (2) penetration of the thermoplastic (TP) polymer in the micro-pores of the aluminium oxide surface modelled using a capillary flow model. The “optimal” welding time based on the maximum lap shear strength (LSS) was determined for various power levels and correlated to the bonding time predicted by the FEM. Consolidation quality and failure mechanisms were discussed in relation to processing parameters. The effect of the welding operation on overaging (annealing) of the aluminium alloy also was investigated. Experimental and simulated processing windows were constructed and correlated to each other. However, thermal degradation of PEI as predicted by the model did not correlate to a reduction in performance of the joint.

Simulation of Impulse Resistance Welding for Thermoplastic Matrix Composites

The impulse resistance welding (IRW) process is modelled using a three-dimensional transient finite element model (FEM) featuring heat transfer and consolidation. The welding of single lap joints is simulated for APC-2 laminates, with additional PEEK film inserted at the welding interface. The effects of the power signal on (a) temperature uniformity in the welding interface and (b) processing times are investigated. Two welding configurations are evaluated; in the first one, the heating element outside the welding stack is left in-air, while in the second case, it is embedded in a conductive medium in order to improve temperature homogeneity within the welding interface. Based on criteria of consolidation and thermal degradation, optimum processing windows were simulated.