FRICTION-BASED INJECTION CLINCHING JOINING (F-ICJ) OF GLASS-FIBER- REINFORCED PA66 AND ALUMINUM HYBRID STRUCTURES M. Sônego 1, A.B. Abibe 2, J.F. dos Santos 2, L.B. Canto 1, S.T. Amancio-Filho 2,3,* Helmholtz-Zentrum Geesthacht (HZG), Max-Planck Strasse 1, D-21502, Geesthacht, Germany sergio.amancio@hzg.de 1 PPG-CEM UFSCar; 2 Helmholtz-Zentrum Geesthacht (HZG), 3 Hamburg University of Technology (TUHH) ABSTRACT Friction-based Injection Clinching Joining (F-ICJ) is a new staking-based technique to join hybrid structures. It consists in using frictional heat and force to soften/melt and deform a polymeric stud fitted in a through hole of a perforated joining partner. After consolidation, the deformed stud (stake) will join the partners by creating a staked overlap joint. This work evaluated the effect of two tools designs - flat and conical pin - on the thermal history during processing, joint microstructure, and mechanical strength of hybrid joints of glass-fiber-reinforced PA66 and aluminum 2024. The use of a conical-pin tool resulted in higher heat input and better cavity filling, which promotes improved mechanical anchoring. As a result of this, an increase of 18% in the lap shear force (1038 ± 16 N) for conical-pin tool joints was observed when compared to the joints produced by the flat tool (880 ± 38 N). Keywords: hybrid structures, friction, Friction-based Injection Clinching Joining, composite INTRODUCTION Injection Clinching Joining (ICJ) is a recent joining technology based on staking, injection molding, and mechanical fastening which can be used to join hybrid 8087
secondary and tertiary structures. The technique was developed and patented by the Research Center Helmholtz Zentrum Geesthacht (HZG), in Germany in 2010 (1). Generally, staking-based joining technologies use a thermoplastic component with an integrated stud and another partner material with a through hole. The components are pre-assembled with the stud fitting in the through hole resulting in an overlap joint. In ICJ, thermal energy is used to soften/melt the stud, and a tool is used to form the stud into a stake, and inject the softened/melted polymeric material into cavities present in the through hole. The formation of a stake into the through hole creates mechanical anchoring between the thermoplastic component and its joining partner resulting in a mechanically fasted joint. With the objectives of achieving short joining cycles and high energy efficiency, a variant of ICJ was developed based on frictional heating Friction-based Injection Clinching Joining (F-ICJ). This preliminary work explores the effect of tool geometry on F-ICJ joint properties. Results of thermal history monitoring, joint microstructure, and mechanical testing obtained from hybrid joints of glass-fiber-reinforced PA66 and aluminum 2024 were analyzed in this work. EXPERIMENTAL PROCEDURE A composite of PA66 reinforced with 30 wt% of short glass fiber (ERTALON 66- GF30, Quadrant Plastics, Switzerland) and aluminum alloy AA2024-T351 (Rio Tinto Alcan, Canada) were used in this work. These materials were machined with the geometries showed in Fig. 1. Cavities in shape of chamfer and threads were machined in the through holes at the aluminum parts. Fig. 1: Specimen geometries: (A) parts positioned for joining; and (B) overlap joint for single-lap shear testing. Adapted from (2) with permission. 8088
The joints were produced using non-consumable cylindrical tools with either a flat contact surface, or with a conical pin, fixed in a high-speed friction welding machine (RSM400, Harms & Wende, Germany). A schematic drawing of the steps of this process for both tool geometries is showed in Fig. 2. Fig. 2: Schematic drawing of the F-ICJ process steps using flat tool and conical-pin tool. In a first moment, the polymeric stud is fitted with the through hole of the joining partner, and they are aligned with the tool (Fig. 2A) onto a backing plate. The rotating tool moves towards the joining partners (Fig. 2B), and frictional heat is generated when the tool touches the polymeric stud during the friction phase (Fig. 2C). In the next stage (forging phase Fig. 2D), the rotational speed is decreased and a higher axial pressure is applied to inject the softened/melted polymer in the cavities in the metallic partner s through hole. Cavity filling is needed, since it is responsible for the mechanical anchoring of the joint. During the forging phase, the final shape of the stake is achieved. After a short cooling time, the joint is consolidated (Fig. 2E). In F-ICJ, the mechanical energy owing to tool rotation is converted into heat as a result of the friction between the surfaces of tool and polymer, being this transformation highly efficient (3, 4). In this work, flat and conical-pin tools were used. These tool geometries create different final stake geometries. When a flat tool is 8089
used, a solid stake is formed. On the other hand, a conical-pin tool will create a hollow stake. The F-ICJ process can be divided in friction phase and forging phase. In the friction phase the rotational speed is high and the axial pressure is low. The main controlling parameters for the friction phase are rotational speed of the tool (RS), frictional time (FT), and frictional pressure (FP). These parameters directly influence the heat generation. The following stage is the forging phase, in which the tool rotational speed is decreased until stopping, and the axial pressure is increased. In this phase the controlling parameters are forging time (FoT) and forging pressure (FoP). These two last factors are important to control the injection of the molten polymer into the cavities and for final consolidation of the stake. In this study, the process parameters used for joining were: RS 20000 rpm; FT 1500 ms; FP 0.4 MPa; FoT 1500 ms; and FoP 0.8 MPa. Temperature monitoring was performed with an infrared thermal camera (ImageIR 8800, Infratec GmbH, Germany). To compare the heat input in F-ICJ joints with different tool geometries, the temperature of the stake was measured at the moment of tool retraction. The microstructure of the cross section of the joints was analyzed with an optical microscope (DM IRM, Leica Camera AG, Germany). Through microscopy, the cavity filling of chamfer and threads by the polymer could be evaluated. The mechanical performance of overlap F-ICJ joints was evaluated through single-lap shear tensile testing. The specimens (Fig. 1C) were tested in a universal testing machine (1478, Zwick Roell, Germany) according to ASTM D5961 (5), at a crosshead speed of 2 mm/min at room temperature. Five replicates were used in each condition. RESULTS AND DISCUSSION Fig. 3 presents the results of thermal monitoring from the F-ICJ process. To evaluate the heat input generated by each of the tool geometries, the temperatures measured at the stake head after tool retraction were compared. The measurements were performed in the borders of the stake head as showed in the schematic drawing (Fig. 3A), and the average measured maximal temperatures were compared (Fig. 3B). This region of interest was selected since it encompasses the 8090
thermomechanically affected zone of the polymer, and can give an estimate (or indication) of the heat input at the polymer-tool interface. The flash material expelled during process (Fig. 3C) achieved temperatures above 300 C (Fig. 3D). These temperatures are high enough to melt the PA66 matrix, whose crystalline melting point is 255 C. Fig. 3: Evaluation of heat input at F-ICJ joints: (A) schematic drawing of the measurement regions for solid stake (flat tool) and hollow stake (conical-pin tool); (B) comparison of the average maximal temperatures for each tool geometry; (C) schematic drawing of the flash material expelled during processing; (D) maximal temperature of the expelled flash material. Average maximal stake head temperatures after tool release were lower for the flat tool (43 ± 6 C) in comparison to the conical-pin tool (57 ± 2 C). The heat input difference between the two conditions is a result of the contrasting shearing profiles induced by these tool geometries. This difference can result in distinctive volumes of thermally affected polymer. With higher temperatures the viscosity of the molten 8091
polymer decreases, which can contribute for a more efficient cavity filling in ICJ joints. Cavity filling from chamfer and threads was analyzed through macro- and micrographs of the cross sections of the F-ICJ joints produced with the flat and conical-pin tool. Fig. 4 shows surface views of stakes formed using a flat tool (Fig. 4A) and a conical-pin tool (Fig. 4B). The flat tool created a solid stake which could not fill the cavities in the metallic partner. The conical-pin tool created a hollow stake, which completely fill the chamfer cavity. Fig. 4: Surface view of stake heads produced with (A) flat tool; and (B) conical-pin tool. Micrographs of the polymer-metal interfaces at the cross section of joints produced with (C) flat tool; and (D) conical-pin tool; for simplicity only halves of the symmetric stake joints are depicted. Fig. 4 also shows micrographs of the polymer-metal interfaces at the cross sections of the joints produced with the flat tool (Fig. 4C) and conical-pin tool (Fig. 4D). None of the tool geometries were able to induce filling of the threaded cavities with the process parameter combination used. This is a result of the short joining cycles from F-ICJ adopted in this exploratory study, allied with low thermal conductivity of the polymer. These conditions prevent uniform heat diffusion through the stake volume, so that the polymer close to the threads cannot flow into the cavities because of its higher viscosity. To overcome this limitation, new designs for the polymeric stud, metallic cavities, or tool geometry must be developed, seeking a more efficient heat distribution over the stake volume. 8092
In general, the conical-pin tool achieve better chamfer cavity filling than the flat tool, since it penetrates at the center of the polymeric stud and then can soften a larger amount of polymer. The polymer in the center of the stud is melted and pushed to the surroundings of the stud, effectively filling the chamfer cavity. Besides, thermal measurements show evidence that the conical-pin tool profile was more efficient in frictional heat generation, achieving higher process temperatures. The mechanical properties of F-ICJ joints were evaluated through single-lap shear testing. Typical mechanical behavior of joints produced with flat tool and conical-pin tool are showed in Fig. 5. Fig. 5: Examples of the mechanical behavior under single-lap shear testing for F-ICJ joints produced with flat tool and conical-pin tool. The mechanical behavior of solid stake and hollow stake joints was similar. However, the joints produced with a conical-pin tool (hollow stake) achieved values of ultimate lap shear force 18% higher than flat tool (solid stake) joints. This better 8093
performance is a result of a slight more efficient chamfer cavity filling induced by the conical-pin tool, which leads to better mechanical anchoring. CONCLUSION This study evaluated the effects of two F-ICJ tool geometries (flat tool and conical-pin tool) on the heat input, microstructure, and mechanical strength of glassfiber-reinforced PA66 and aluminum AA2024 overlap joints. The conical-pin tool configuration generate more frictional heat within the evaluated range of joining parameters, decreasing the viscosity of the molten polymer, which facilitates the flow of the molten polymer into the cavities of the through hole of the metallic partner. Because of this improved mechanical anchoring, hollow stake joints produced by conical-pin tool could achieve an increase of 18% in ultimate lap shear force (1038 ± 16 N) compared to the solid stake joints produced with a flat tool (879 ± 28 N). ACKNOWLEDGEMENTS The authors would like to acknowledge the financial support of the Helmholtz Association through the Young Investigator Group, Advanced Polymer Metal Hybrid Structures (Grant No. VH-NG- 626). REFERENCES 1. S.T. AMANCIO-FILHO, J.F. DOS SANTOS, M. BEYER; 11/607,159[7,780,432 B2]; 2010. 2. A.B. ABIBE, Departamento de Engenharia de Materiais, Universidade Federal de São Carlos, São Carlos - Brasil, 2011. 3. R. SHAEFER, RWTH-Aachen, Alemanha, 1971. 4. D. SHOEBER, Technische Hochschule Karl-Marx-Stadt, Alemanha, 1986. 5. A. INTERNATIONAL; D5961: Standard Test Method for Bearing Response of Polymer Matrix Composite Laminates; 2010. 8094