THE DESIGN OF A THERMOPLASTIC CF COMPOSITE FOR LOW PRESSURE MOLDING

Transcription

1 THE DESIGN OF A THERMOPLASTIC CF COMPOSITE FOR LOW PRESSURE MOLDING Takeshi Ishikawa, Masao Tomioka, Masahiro Osuka Advanced Composites Research Group, Toyohashi Research Laboratories, MITSUBISHI RAYON CO., LTD. Abstract The design of thermoplastic carbon fiber composites that induce high moldability is presented. The composite design, which has a controlled fiber length and orientation, is established by regularly plied unidirectional prepreg sheets, which have designed slits that cut the continuous fibers. Because the flowability of the composite is improved by the slit pattern, the molding process requires a relatively low pressure due to optimizing the slit design, even though the composite has a high fiber volume fraction. The composite material can be shaped not only by compression molding but also through several types of thermoforming processes. Because this type of thermoplastic composite has the advantage of having a short cycle time, this composite is preferable for use in the high-volume production of vehicular components. The structural performance, which is also controlled by the slit pattern, can be estimated using a novel FEM model. Therefore, we can design thermoplastic CF composites that have tailor-made properties, such as flowability, mechanical performance, and balance. Introduction Many industrial sectors desire to reduce greenhouse gas emission due to concerns regarding global warming. Energy conservation is a method that industries have used to reduce emissions. For automotive industries, decreasing fuel consumption is a primary issue to be addressed from the viewpoint of environmentally friendly product development. Weight reduction is one route to improve fuel consumption, and the replacement of steel with carbon fiber-reinforced plastics (CFRPs) is a powerful technique used to minimize the weight of vehicular components. CFRPs have been a recent focus because of their outstanding strength and modulus, given their weight, compared with conventional steel. However, the long molding time sometimes prevents CFRPs from being used in automobile assembly lines. For example, resin transfer molding (RTM) and improved techniques, such as VARTM (1), require improvements in the cycle time to catch up to the line speed. To overcome this disadvantage, prepreg compression molding (PCM) was proposed (2) for applications in the field of thermoset resin composites. Another technique to improve the cycle time is the adoption of thermoplastics, such as matrix resin. Carbon fiber-reinforced thermoplastics (CFRTPs) have not only a good mechanical performance but a short molding time and no curing time compared with thermoset resins. In addition to the balance of structural and flow performance, CFRTPs have the advantage of being easily recyclable. Because most applications for use in vehicular components require the component to be formed into complex geometries, such as a rib structure, the flowability of the composite material is essential. Page 1

2 Short carbon fiber composites were proposed due to their excellent flowability and the complicated 3D geometry that can be achieved by using injection moldings (3-5). However, the moderate mechanical strength of short carbon fiber composites is unfavorable, and the flowability worsens when the volume fraction of the carbon fiber is increased. Conversely, the LFT-D process has been developed, in which compound thermoplastics with long fibers (rovings) are created and are then molded in one operation (6). In the case of glass fibers created using this process demonstrate good mechanical properties and productivity; however, few applications using carbon fibers have been reported. Recently, non-woven carbon tissue reinforced thermoplastics (NWCTs) which consist of randomly dispersed mono-fiber in the matrix resin have been presented (7). These sheets are produced using a method similar to the process used to produce paper. The relatively long fiber length and superior dispersion of the carbon fiber result in excellent mechanical strength, even at low volume fractions of fiber. However, the low flowability caused by the dispersion requires a large pressure in the compression moldings. Therefore, to balance the mechanical properties and moldability, chopped carbon prepreg based thermoplastic composites (CPTCs), which are produced by cutting a unidirectional (UD) prepreg into pieces and randomly distributing these pieces, has been developed (8). To the best of the authors knowledge, the weak point of this composite material is the fluctuation of the mechanical properties. The randomness of the distribution cannot be achieved in all parts of large products. In this paper, we propose a composite design with a controlled fiber length and in which the orientation is established by regularly plied UD prepreg sheets, which have slits designed to cut the continuous fibers. This CFRTP is named slit carbon prepreg based thermoplastic composite (SPTC). However, this type of composite design was already proposed in a Japanese patent (9, 10), and several papers have primarily discussed the mechanical properties limiting the thermoset matrix resins (11-14). Therefore, we focus on the slit pattern s effect on the moldability of the thermoplastic matrix resin. Furthermore, we also establish an original FEM model to simulate the mechanical properties of the SPTC. We discuss the mechanical properties of the SPTC in addition to the flowability of the SPTC. Preparation of the SPTC Experimental setups The carbon fiber TR50S-15L (Mitsubishi Rayon Co. Ltd.) was opened to an area weight of 72 g/m 2 and was sandwiched by a cast film of 40 m thickness that consisted of maleic-anhydride-grafted polypropylene, which was produced by the Mitsubishi Chemical Corp. This laminated sheet was passed through a roll gap under a specific pressure and was heated to 260ºC and a UD prepreg volume fraction of 33% was produced. The slits were inserted into the UD prepreg using a ZUND G3 L-2500 cutting plotter to cut the continuous fibers into a specific length. The design of the slits was achieved by controlling the slit angle and the fiber length (slit interval), as observed in Fig. 1 and Table 1. The slit UD prepregs were plied in a quasi-isotropic manner [(0/45/90/-45)s] 2 and were then consolidated using a press molding machine at a force of 50 kn and at a temperature of 210ºC for 7 minutes. The size of the mold cavity, which is a square that is 300 mm on each side, was the same as the slit UD prepreg; therefore, no deformation Page 2

3 was observed after molding. The thickness of the SPTC was approximately 2.0 mm. Figure 1: Slit design by slit angle and fiber length L (Black lines shows the carbon fiber orientation and the red lines show the slits.) L [mm] Table I: Matrix of experimental and numerical targets [deg] , L , L25 30, L25 45, L25 60, L25 90, L25 Experiment only , L37.5 Measurement of the flowability The flowability is defined as the ratio of the initial thickness to the final thickness after the compression molding process. The conditions of pressure and sample size were determined from the real production process of certain automobile parts using a theoretical scale-down model. The SPTC samples, squares that were 78 mm on each side with a thickness of 2 mm, were plied doubly and preheated at 230 ºC for 5 minutes until a uniform temperature was achieved. After the preheating step, the SPTC was pressed with a force of 33 and 17 kn at a temperature of 145 ºC for 30 seconds. Page 3

4 Measurement of the structural properties Three-point bending tests were conducted using the JIS K7074 standard. The specimen, which was 100 mm long and 25 mm wide, was placed on support spans of 80 mm, and the load was applied to the center by the loading nose, which produced threepoint bending at a rate of 5 mm/min. Numerical Modeling To theoretically investigate the structural performance, a FEM analysis was conducted. The UD prepreg was discretized by a shell element and was cut and placed regularly according to the slit pattern. An adhesion boundary was imposed at the interface between the plies in the thickness direction, and the adhesion strength was defined in a manner to reflect the strength of the matrix resin. The matrix resin inside the slit was expressed by coupling the shell elements in the solid element of the matrix resin. An overview of numerical modeling is shown in Fig. 2. (a) Quasi-isotropic ply of the slit UD prepreg (b) Cross section of the model Figure 2: Overview of numerical modeling Flowability Results and discussion The slit angle s effect on the flowability is shown in Fig. 3, and the large ordinate values demonstrate high flowability. It is clear from Fig. 3 (a) that the small slit angle decreases the flowability, and an improvement is observed when the angle is greater than 30º. The effect of fiber length is shown in Fig. 3 (b), and long fiber lengths result in poor flowability. Conversely, the difference between L25 and L12.5 is relatively small. Therefore, a slit design using an angle greater than 30º and a length less than 25 mm is proposed as ideal for use in the moldable composites. Page 4

5 Flowability Flowability kN 33kN kN 33kN θ15 θ30 θ45 θ60 θ90 Slit angleθ [ ] 2.0 L12.5 L25 L37.5 Fiber length L [mm] (a) Effect of slit angle (b) Effect of fiber length Figure 3: Flowability of the SPTC affected by slit pattern Structural properties The effect of the slit angle on the measured flexural strength and the modulus, together with the numerical results, are shown in Fig. 4. There is a clear dependence of the slit angle on the flexural strength, that is, a smaller angle demonstrates high strength in the case of a slit angle of 5º. The numerical simulation can trace the experimental results accurately except for the 15º case, in which underestimation occurs. The flexural moduli have similar but weak tendencies, and the numerical results demonstrate good agreement with the measured data. The effect of the fiber length on the flexural strength and the moduli are shown in Fig. 5. Both the strength and the modulus have a weak dependency on the fiber length, and longer fiber lengths improve both properties. Adequate numerical estimation can be obtained. Page 5

6 Flexrual strength [MPa] Flexrual modulus [GPa] Flexural strength [MPa] Flexrual modulus [GPa] Measured CAE Measured CAE θ15 θ30 θ45 θ60 θ90 Slit angleθ [ ] 0 θ15 θ30 θ45 θ60 θ90 Slit angleθ [ ] (a) Flexural strength (b) Flexural modulus Figure 4: Structural performance affected by the slit angle Measured CAE Measured CAE L12.5 L25 L37.5 Fiber length L [mm] 0 L12.5 L25 L37.5 Fiber length L [mm] (a) Flexural strength (b) Flexural modulus Figure 5: Structural performance affected by the fiber length Page 6

7 Application The composite design proposed here has excellent flowablity and good mechanical properties. Several applications of our composite design are presented below. Application 1: Hat channel Figure 6: Hat channel Application 2: Rib structure Figure 7: Rib structure Application 3: Floor panel Figure 8: Floor panel Page 7

8 Application 4: Vacuum forming Figure 9: Vacuum forming Summary We have proposed the design of a carbon fiber-reinforced thermoplastic with good flowability as the primary goal. This design strongly depends on the slit pattern, and slit angles greater than 30º induce good flowability. Even though the mechanical properties improved when using a smaller slit angle, we recommend the larger slit angle due to the reasons described above. The composite material that has this slit design can be shaped not only by compression molding but also by several other types of thermoforming processes. Structural parts, such as a rib structure with a complicated 3D geometry, a large panel, and a vacuum formed part, are presented. We can also estimate the mechanical properties of the carbon fiber-reinforced thermoplastic by the original FEM model that was developed in this study; therefore, a slit design that focuses on mechanical properties is proposed. Thus, a tailor-made composite that has a moldability performance that is compatible with its mechanical properties can be achieved. References 1. P. Morgan, Carbon Fibers and Their Composites, Taylor & Francis, (2005). 2. K. Akiyama, Development of Preforming Process in PCM Technology, ACCE Proceeding, (2012) 3. P. T. Curtis, M. G. Bader, & J. E. Bailey, The Stiffness and Strength of a Polyamaide thermoplastics reinforced with glass and Carbon Fibres, J. Matter. Sci., 13 (1978) pp S. Y. Fu, B. Lauke, E. Mäder, X. Hu, & C. Y. Yue, Fracture Resistance of Short-glass-fiberreinforced and Short-Carbon-Fiber-Reinforced Polypropylene under Charpy Impact Load and Its Dependence on Processing, 89-90(1999) pp S. Y. Fu, B. Lauke, E. Mäder, C. Y. Yue, & X. Hu, Tensile Properties of Short-Glass-Fiber- Page 8

9 and Short-Carbon-Fiber-Reinforced Polypropylene Composites, Composites: Part A, 31 (2000) pp M. Garnier, In-line Compounding and Molding of Long-fiber Reinforced Thermoplastics (D- LFT) Insight into a Rapid Growing Technology, ANTEC Technical Paper, (2004) pp M. Hashimoto, T. Okabe, & M. Nishikawa, Development of Thermoplastic Press Sheet with In-Plane Randomly Oriented and Dispersed Carbon Mono-Fibers and Evaluation of the Mechanical Property, Nihon Fukugou Zairyou Gakkaishi, 37 (2011) pp N. Yoshihara, S. Tsuji, & S. Nago, Layered Molded Article of Fiber-Reinforced Thermoplastic Resin, WO2012/133013A1 9. M. Imao, Japanese Patent, Tokukaishou , (1983) 10. H. Sakai, T. Nakakura, S. Kishi, Japanese Patent, Toukaishou , (1988) 11. I. Taketa, T. Okabe, & A. Kitano, A New Compression-Molding Approach Using Unidirectionally Arrayed Chopped Strands, Composites: Part A, 39 (2008) pp I. Taketa, T. Okabe, & A. Kitano, Strength Improvement in Unidirectional Arrayed Chopped Strands with Interlaminar Toughening, Composites: Part A, 40 (2009) pp I. Taketa, N. Sato, A. Kitano, & T. Okabe, Enhancement of Strength and Uniformity in Unidirectionally Arrayed Chopped Strands with Angles Slits, Composites: Part A, 41 (2010) pp I. Taketa, T. Okabe, H. Matsutani, & A. Kitano, Flowability of Unidirectionally Arrayed Chopped Strands in Compression Molding, Composites: Part B, 42 (2011) pp Page 9