LOCAL CONTINUOUS FIBER-REINFORCEMENT TAILORED INJECTION MOULDING >>LIGHTWEIGHT POTENTIAL FOR INJECTION MOULDED PARTS<<

Transcription

1 LOCAL CONTINUOUS FIBER-REINFORCEMENT TAILORED INJECTION MOULDING >>LIGHTWEIGHT POTENTIAL FOR INJECTION MOULDED PARTS<< Timo Huber, Frank Henning, Benjamin Hangs, Alexander Roch Fraunhofer Institute for Chemical Technology ICT Abstract Weight reduction of components is becoming increasingly important, for example in automotive applications where significant fuel savings and CO 2 emission reduction can be made. In many cases metal load-bearing parts can be replaced by short or long fiber reinforced thermoplastics. These composites have become an integral part of industrial large scale production especially due to their economical processability and increased functionality. The limited mechanical properties, such as stiffness and impact strength, prohibit the use of injection moulded parts in higher load-bearing applications. Furthermore the viscoelastic behavior of the matrix at high temperature or at permanent load is a disadvantage (tendency to creep). Local continuous fiber reinforcement can overcome these disadvantages by increasing the properties and reducing tendency to creep. The technology of continuous fiber reinforcement with thermoplastic materials includes in principal all pre-impregnated fiber structures like woven and non-woven fabrics, UD-strands, tape layups or winded geometries. To characterize the effect of unidirectional fiber reinforcement in this research work a single pre-impregnated fiber strand was chosen. These strands were then being manufactured to generic specimens in a specialized mould for insert technology with a standard injection moulding process as one of the most important processing technology for thermoplastic polymers. The mechanical behavior of the parts according to the variation of material and process parameters were analyzed with a static tensile test and a dynamicmechanical analysis (DMA). Introduction By analyzing the stress distribution of parts with definite dimensions, the structure can be divided into load-bearing and non load-bearing sections. Lines of stress-flux can be identified which show the load gradient inside the injection moulded part under a specific external load. When using short- or long-fiber pellets for the manufacture of the part all areas of the part are equally reinforced by fibers. Due to the higher density of the fibers compared to the matrix the total weight of the component is increased unnecessarily by the reinforcement of non or partially loaded areas. The tailored placement of reinforcing structures with optimal load-dependent orientation along the application-specific load paths reinforces the component only in the areas where extra strength is required and thus increases the weight only partially. This leads to significant weight savings compared to the processing of fiber-reinforced materials. The manufacturing process will be described for a generic part. Detailed testing results show the mechanical behavior (static and dynamic-mechanical analysis DMA) regarding to material parameters and different load situations of the generic structural parts (e.g. with foamed or short fiber reinforced matrix). Page 1

2 Experimental In this study, a roving made of commingled E glass and polypropylene filaments is used for manufacturing the unidirectional fiber reinforcement (UD-strand). Impregnation and consolidation is done by heating the roving above melting temperature of PP matrix (180 C 230 C / 360 F- 450 F) [1]. The poly-propylene resin used for injection moulding trials is a high performance impact copolymer with an MFR 70 g/10 min (230 C/2.16kg) and solid density of 0.90 g/cm3 [2]. Material variations used for comparative trials, like short or long fiber pellets and foamed polymer matrix were mainly based on this impact modified copolymer resin (Table I). Table I: Materials tested with DMA Material variations Material Description V26 DOW C711-70RNA PP V16 DOW C711-70RNA foamed PP with chemically foamed matrix V27 DOW C711-70RNA strand PP with UD-strand V19 DOW C711-70RNA foamed, strand PP with chemically foamed matrix and UD-strand V28 DOW PP C711-70RNA / DLGF Z 15%LGF V29 DOW PP C711-70RNA/INSPIRE GF3700Z 15%KGF PP with 15 w.-% glass fiber (12mm long fiber pellets) PP with 15 w.-% glass fiber (short fiber pellets) Fiber impregnation In a first step of the production chain the endless fiber reinforcement (commingled PP/GF roving) will be impregnated and consolidated. Therefore the roving is pulled through a heating zone with infrared heater, a consolidating zone and a die. This impregnated strand (with still molten polymer around the fibers) can then be cooled down and cut to length or even be processed to coiled structures (see Fig. 1 and Fig. 2). Page 2

3 Figure 1: Fiber impregnation and consolidation unit Figure 2: Heating zone (left), die with impregnated strand (right) To guarantee a constant impregnation quality in the process it is important to have the optimal parameter settings of the impregnation unit like haul-off speed, the temperature profile of the heating zone and the temperature of the consolidating elements and the die. This will result in a homogenous distribution and wetting of each fiber filament with polymer matrix (see Fig. 3) which is a precondition of optimal reinforcement properties of the UD-strand in the final part [8]. Page 3

4 Figure 3: LM-micrograph of impregnated UD-strands with optimal process parameters (left) and non-optimal impregnation with voids and unconsolidated fiber bundles (right) Specimen for tensile test For static and dynamic tensile testing specimen were used on the basis of EN ISO (see Fig. 4). The specimen were injection moulded on an ARBURG Allrounder 320 C U injection moulding machine with a screw diameter of 30 mm and 600 kn clamping force [3]. Standardized process parameters were used and are listed in Table II. When necessary, small optimizations were applied for the processing of the individual material variation. Table II: Machine settings of injection moulding process Machine settings Parameter Setting Injection speed 40 cm 3 /s Melt temp. 220 C / 428 F Mould temp. 50 C / 122 F Post pressure 300 bar / 10 s Figure 4: Tensile test specimen with 4 mm wall thickness for static and dynamic-mechanical analysis DMA Page 4

5 Dynamic-mechanical analysis (DMA) With this analysis information about the material-specific dynamic viscoelastic properties under sinusoidal, oscillating force like the complex modulus E* can be determined. This complex modulus is made up of the storage module E as the real part and the loss module E as the imaginary part (see equations 1-4) [6]. E * (1) E * E 2 + E 2 (2) E E * * cos δ (3) E E * * sin δ (4) With the phase angle δ, which describes the phase difference between the dynamic stress and the dynamic strain in a viscoelastic material the loss factor tan δ can be deduced as the ratio of loss modulus to storage modulus (equation 5). Tan δ is a measurement of the energy lost expressed in terms of the recoverable energy, and represents mechanical damping or internal friction in a viscoelastic system [5, 7]. tan δ E E (5) Table III: Testing parameters of DMA Testing parameters Parameter Clamp distance Setting 125 mm Static amplitude 0.5 % Dynamic amplitude 0.08 % Test frequency 1 Hz Temperature range -40 C / -40 F up to +100 C / +212 F Page 5

6 Results and Discussion The results of static tensile test concerning the comparison of the reinforcement effect of UDstrand compared to short- and long-fiber pellets as a function of glass fiber content are shown in Fig. 5. Figure 5: Tensile strength and E-Module as a function of glass fiber content Fig. 5 clearly shows the interesting technical potential of unidirectional endless fiber structures in injection moulded parts. Specimens with one UD-strand and a fiber content of 2.3w.-% are almost able to reach comparable tensile strength of specimens with 15w.-% short or long fiber reinforcement. This is due to the optimal position of the UD-strand in relation to the direction of force and optimal coupling of the polymer matrix and the glass fiber. The results of DMA analysis are based on seven specimens for each trial whereas the average value was taken from at least five specimens (dependent on specific standard deviation). The mechanical properties of pure polypropylene (V26) and PP with UD reinforcement (V27) are representatively shown in Table IV. Temp [ C] Aver. V27 E [MPa] Table IV: DMA results of trial no. V26 and V27 Aver. V26 E [MPa] Results V26 and V27 Aver. V27 E [MPa] Aver. V26 E [MPa] Aver. V27 tan δ Aver. V26 tan δ Page 6

7 The level of storage module E of the specimen with UD-strand (V27) in comparison with pure PP (V26) increases clearly as expected and shows similar characteristics. Fig. 6 shows the trend of E and E of V26, V27, V28 (long fiber pellets) and V29 (short fiber pellets) with increasing temperatures. It can be clearly identified that the difference between pure PP and PP with UD reinforcement as well as PP with short and long fiber reinforcement shows a similar curve progression up to the glass transition temperature of the polymer matrix material (at appr. -10 C/-14 F). Beyond this point the difference between pure PP (V26) and UD-strand (V27) increases constantly respectively the properties of UD-strand approaches to KGF and LGF material with significantly higher fiber content. Consequently the effect of the UD reinforcement becomes more visible at higher temperatures. One reason of this effect can be found in the rising of the dominance of the endless glass fiber due to its high elongation module. Figure 6: Storage and loss module In contrast to the storage module the effect of the UD-strand on the loss module is significant lower in comparison with pure PP. The maximum difference is 20%. With respect to applications of UD-strands in structural parts this is a positive result, which can be clearly identified in the comparison of tan δ (see Fig. 7). Before reaching a temperature of around 50 C/122 F all three fiber-reinforced materials show a similar behavior. From this point on the UD-strand shows the phenomena of a decreasing tan δ. Consequently the potential of the local reinforcement especially concerning dimensional stability increases with temperature. Page 7

8 Figure 7: Tan δ of pure PP (V26), PP with UD reinforcement (V27) short (V29) and long fiber pellets (V28) Due to the interesting light weight potential of parts with foamed polymers, the influence of UD reinforcement together with a foamed matrix system was analyzed as well. Therefore the PP was mixed with a chemical blowing agent to generate of foamed structure in the part. The addition of the additive was adjusted to create a weight reduction of 10w.-%. The storage module as well as the loss module decreased as expected whereas the basic curve progression remains constant (see Fig. 8). With the UD reinforcement this loss of strength can be compensated especially at lower temperatures where the foamed matrix shows the worst behavior in comparison to solid PP. At the point of glass transition temperature the curves of foamed PP (V16) and solid PP (V26) as well as with UD reinforcements (V19 & V27) approaches to each other (Fig. 8) and the decrease of stiffness caused by the foamed matrix becomes less with higher temperatures. At the maximum testing temperature of 100 C / 212 F there is no significant difference between foamed and solid polymer system. Page 8

9 Figure 8: Comparison of E and E of specimen with solid and foamed polymer matrix system The basic behavior of UD reinforced parts under dynamic load does not change in terms of solid or foamed polymer structure. Both UD reinforced parts (with solid matrix V27 and foamed matrix V19) show the same phenomenon of a decreasing tan δ at higher temperatures as shown in Fig. 9. It can be observed that the tan δ curve of UD reinforcement with foamed matrix (V19) cross the curve of a solid matrix (V27) at a temperature of around 5 C / 41 F. Figure 9: Tan δ of specimen with solid and foamed structure with and without UD reinforcement Page 9

10 Summary and Next Steps The potential of local UD reinforcements in injection moulded parts was demonstrated. The strongly increasing mechanical properties of thermoplastic parts under dynamic load could be illustrated based on results of a dynamic-mechanical analysis DMA. Next steps will be the development of suitable material models to describe structural properties of local endless fiber reinforced parts and the up-scaling to complex geometries. For structural applications for example in automotive parts it is necessary to test not only at standard conditions but also at low and high temperature, low and high loading rate and the long-term behavior. These results will be the base for developing material models. Concerning manufacturing process there is the main objective to develop a holistic process view from the first part design to the final structural hybrid part geometry with integrated structural analysis and part optimization to find the tailored reinforcement structure and therefor best mechanical behavior of the part (Fig. 10). Figure 10: Procedure of Tailored Injection moulding Acknowledgement The authors thank the Federal Ministry of Education and Research of Germany (BMBF) and the Fraunhofer Society for the advancement of applied research for the financial support of this work. Page 10

11 References 1. N.N., Product Information TWINTEX R PP B, OCV Reinforcements (2010). 2. N.N., Product Information DOW Polypropylene C711-70RNA, Dow Chemical Company (2000). 3. B. Thielicke, M. Grigo, J. Dreissig, Round-Robin-Tests zum Einfluss der Probengeometrie - LFT- Kennwerte aus Zugversuchen, ekonstruktion Oktober 10/ N.N., Bestimmung der Zugeigenschaften Teil 4, EN ISO (1997). 5. G. W. Ehrenstein, G. Riedel, P. Trawiel, Thermal Analysis of Plastics Theory and Practice, Hanser Gardner Publications, Cincinnatti, N.N., Plastics Determination of dynamic mechanical properties Part 1: General principles, ISO :2001 (2001). 7. L.E. Nielsen, R.F. Landel, Mechanical Properties of Polymers and Composites, 2nd edition, CRC Press, Marcel Dekker Inc. New York (1994). 8. G.W. Ehrenstein, Faserverbundwerkstoffe Werkstoffe Verarbeitung Eigenschaften, 2., völlig überarbeitete Auflage, Carl Hanser Verlag München (2006). Page 11