River Rapids Conference Proceedings of the Society for Experimental Mechanics Series Mechanics of Composite and Multi-functional Materials, Volume 6 Piyush R. Thakre Raman Singh Geoff Slipher Proceedings of the 2017 Annual Conference on Experimental and Applied Mechanics River Publishers
Conference Proceedings of the Society for Experimental Mechanics Series Series Editor Kristin B. Zimmerman, Ph.D. Society for Experimental Mechanics, Inc., Bethel, CT, USA
River Publishers Piyush R. Thakre • Raman Singh • Geoff Slipher Editors Mechanics of Composite and Multi-functional Materials, Volume 6 Proceedings of the 2017 Annual Conference on Experimental and Applied Mechanics
Published, sold and distributed by: River Publishers Broagervej 10 9260 Gistrup Denmark www.riverpublishers.com ISBN 978-87-7004-961-0 (eBook) Conference Proceedings of the Society for Experimental Mechanics An imprint of River Publishers © The Society for Experimental Mechanics, Inc. 2018 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, or reproduction in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Preface Mechanics of Composite and Multifunctional Materials represents one of nine volumes of technical papers presented at the 2017 SEM Annual Conference & Exposition on Experimental and Applied Mechanics organized by the Society for Experimental Mechanics and held in Indianapolis, IN, June 12–15, 2017. The complete Proceedings also includes volumes on: Dynamic Behavior of Materials; Challenges In Mechanics of Time-Dependent Materials; Advancement of Optical Methods in Experimental Mechanics; Mechanics of Biological Systems, Materials and Other Topics in Experimental and Applied Mechanics; Micro-and Nanomechanics; Fracture, Fatigue, Failure and Damage Evolution; Residual Stress, Thermomechanics & Infrared Imaging, Hybrid Techniques and Inverse Problems; and Mechanics of Additive and Advanced Manufacturing. The commercial market for composites continues to expand with a wide range of applications from sporting equipment to aerospace vehicles. This growth has been fueled by new material developments, greater understanding of material behaviors, novel design solutions, and improved manufacturing techniques. The broad range of applications and the associated technical challenges require an increasingly multidisciplinary and collaborative approach between the mechanical, chemical, and physical sciences to sustain and enhance the positive impact of composites on the commercial and military sectors. New materials are being developed from recycled source materials, leading to composites with unique properties and more sustainable sources. Existing materials are also being used in new and critical applications, which require a deeper understanding of material behaviors and failure mechanisms on multiple length and time scales. In addition, the unique properties of composites present many challenges in manufacturing and in joining with other materials. New testing methods must be developed to characterize the novel composite properties, to evaluate application and product life cycle performance, as well as to evaluate impacts and merits of new manufacturing methods. This volume presents early research findings from experimental and computational investigations related to the processing, characterization, and testing of composite, hybrid, and multifunctional materials. Freeport, TX, USA Piyush R. Thakre Tulsa, OK, USA Raman Singh Aberdeen, MD, USA Geoff Slipher v
Contents 1 Scrap-Rubber Based Composites Reinforced with Boron and Alumina........................ 1 A.B. Irez, Jennifer Hay, Ibrahim Miskioglu, and Emin Bayraktar 2 Characterization of Thermoplastic Matrix Composite Joints for the Development of a Computational Framework...................................................... 11 Joseph R. Newkirk, Cassandra M. Degen, and Albert Romkes 3 Experimental Study of Laser Cutting Process of Titanium Aluminium (Ti-Al) Based Composites Designed Through Combined Method of Powder Metallurgy and Thixoforming. . . . . . . . 21 S. Ezeddini, G. Zambelis, E. Bayraktar, I. Miskioglu, and D. Katundi 4 Mechanical Characterization of Epoxy: Scrap Rubber Based Composites Reinforced with Nanoparticles ................................................................. 33 A.B. Irez, I. Miskioglu, and E. Bayraktar 5 Mechanical Characterization of Epoxy – Scrap Rubber Based Composites Reinforced with Nano Graphene............................................................... 45 A.B. Irez, I. Miskioglu, and E. Bayraktar 6 Mechanical Characterization of Epoxy – Scrap Rubber Based Composites Reinforced with Alumina Fibers ............................................................... 59 A.B. Irez, E. Bayraktar, and I. Miskioglu 7 Scaled Composite I-Beams for Subcomponent Testing of Wind Turbine Blades: An Experimental Study............................................................. 71 Mohamad Eydani Asl, Christopher Niezrecki, James Sherwood, and Peter Avitabile 8 Development Analysis of a Stainless Steel Produced by High Energy Milling Using Chips and the Addition of Vanadium Carbide....................................... 79 C.S.P. Mendonc¸a, F. Gatamorta, M.M. Junqueira, L.R. Silveira, J.H.F. Gomes, M.L.N.M. Melo, and G. Silva 9 Design of Magnetic Aluminium (A356) Based Composites through Combined 2 Method of Sinter + Forging 3............................................................... 89 D. Katundi, L.P. Ferreira, E. Bayraktar, I. Miskioglu, and M.H. Robert 10 Design of Low Composites from Recycled Copper + Aluminium Chips for Tribological Applications ......................................................... 101 F. Gatamorta, E. Bayraktar, I. Miskioglu, D. Katundi, and M.H. Robert 11 Liquid Metal Dispersions for Stretchable Electronics ...................................... 111 A.S. Koh, G.A. Slipher, and R.A. Mrozek 12 Laser Cutting of the TiN +Al2O3 Reinforced Aluminium Matrix Composites Through Semisolid Sintering......................................................... 115 Sonia Ezeddini, D. Katundi, Emin Bayraktar, and I. Miskioglu vii
13 Optimization of Laser Cutting Parameters for Tailored Behaviour of Scrap (Ti6242 + Ti) Based Composites Through Semisolid Sintering.......................................... 131 Sonia Ezeddini, Emin Bayraktar, I. Miskioglu, and D. Katundi 14 Studying Effect of CO2 Laser Cutting Parameters of Titanium Alloy on Heat Affected Zone and Kerf Width Using the Taguchi Method......................................... 143 B. El Aoud, M. Boujelbene, E. Bayraktar, S. Ben Salem, and I. Miskioglu 15 Fatigue Characterization of In-Situ Self-Healing Dental Composites .......................... 151 D.H. Kafagy, S.S. Khajotia, and M.W. Keller 16 Effect of Process Induced Stresses on Measurement of FRP Strain Energy Release Rates .......... 157 Brian T. Werner, Stacy M. Nelson, and Timothy M. Briggs 17 Characterization of UV Degraded Carbon Fiber-Matrix Interphase Using AFM Indentation. . . . . . . 175 Kunal Mishra, Libin K. Babu, and Raman Singh 18 A Study on Mechanical Properties of Treated Sisal Polyester Composites ...................... 179 G.L. Easwara Prasad, B.S. Keerthi Gowda, and R. Velmurugan 19 Strain-Rate-Dependent Failure Criteria for Composite Laminates: Application of the Northwestern Failure Theory to Multiple Material Systems ............................ 187 Joseph D. Schaefer, Brian T. Werner, and Isaac M. Daniel 20 Progressive Failure Analysis of Multi-Directional Composite Laminates Based on the Strain-Rate-Dependent Northwestern Failure Theory................................ 197 Joseph D. Schaefer, Brian T. Werner, and Isaac M. Daniel 21 Experimental Mechanics for Multifunctional Composites and Next Generation UAVs ............. 215 Jeffery W. Baur, Darren J. Hartl, Geoffrey J. Frank, Gregory Huff, Keith A. Slinker, Corey Kondash, W. Joshua Kennedy, and Gregory J. Ehlert viii Contents
Chapter 1 Scrap-Rubber Based Composites Reinforced with Boron and Alumina A.B. Irez, Jennifer Hay, Ibrahim Miskioglu, and Emin Bayraktar Abstract Composites made of reinforced scrap rubber are generating interest in transportation industries due to their unique combination of high strength, low density, and limitless availability. Whether the vehicle is an airplane, truck, car, or boat, weight reduction leads directly to reduced fuel consumption and operating costs. In the present study, different composite formulations are prepared by means of a low-cost production process. In this process, fresh scrap rubber is combined with varying amounts of boron and alumina, with the aim of optimizing strength through control of both composition and constituent bonding. Mechanical properties are evaluated by impact testing, bend testing, and nanoindentation. Microstructure is analyzed by scanning-electron microscopy (SEM). Keywords Recycled composites • Ceramic reinforcements • High-speed nanoindentation, SEM 1.1 Introduction Polymer-matrix composites are used increasingly for engineering applications. Among the numerous polymers that can be used as matrix material, epoxy resins are very popular. Epoxy- resin-based composites are used in many engineering applications, especially in the aeronautical and automotive industries. Superior advantages of epoxy include high specific strength and stiffness, chemical resistance, ease of processing, environmental stability and relatively low cost [1]. However, the brittleness of epoxy limits its usage in some areas. This low fracture resistance arises from highly cross-linked network structure. Therefore, for a few decades, engineers have been working on the improvement of the fracture toughness of epoxies in order to widen their applications [2, 3]. Generally, secondary-phase particles in the epoxy matrix can be incorporated to toughen epoxy resins. Secondary-phase particles can be dispersed in the matrix and they can either be soft fillers such as thermoplastic particles, rubber, or rigid fillers such as silica beads, titania, or alumina [4]. In order to make epoxy-resin-based composites resistant to high mechanical and tribological loads, reinforcements of the matrix with certain fillers are absolutely necessary. In the literature, certain fillers are chosen for the sake of simplicity in processing. However, new composites (lightweight and high-performance) need special fillers in the matrix in order to increase ductility and stiffness of the matrix materials. In reality, scrap rubber particles are ideal, because they are easy to incorporate and have the effect of increasing the toughness of epoxy-based composites. However, scrap rubber in its virgin form is not compatible with epoxy due to poor adhesion between the rubber and epoxy. Surface treatment of the rubber is needed in order to improve its adhesion to the matrix. In the literature [5–14], some of the methods proposed for chemical surface treatments, such as plasma-surface modification and other de-vulcanizations, are time consuming and expensive for the recycling process. In this research, surface treatment of the recycled scrap rubber was achieved by a simple chemical treatment with silane. Hence, the low-cost epoxy-resin-based composites were designed with scrap rubber chemically treated with silane with the addition of boron, alumina and glass bubbles. After researching the Jennifer Hay and Ibrahim Miskioglu contributed equally to this work. A.B. Irez • E. Bayraktar Supmeca-Paris, School of Mechanical and Manufacturing Engineering, Saint-Ouen, France J. Hay (*) Nanomechanics, Inc., 105 Meco Lane, Oak Ridge, TN 37830, USA e-mail: jenny.hay@nanomechanicsinc.com I. Miskioglu (*) Michigan Technological University ME-EM Department, Houghton, MI, USA e-mail: imiski@mtu.edu #Springer International Publishing AG 2018 P.R. Thakre et al. (eds.), Mechanics of Composite and Multi-functional Materials, Volume 6, Conference Proceedings of the Society for Experimental Mechanics Series, DOI 10.1007/978-3-319-63408-1_1 1
compatibility of various reinforcements with the epoxy matrix, we found that the most suitable recipe for aeronautical applications included the addition of fine-scrap-rubber powders (after simple chemical treatments with silane), and then additions of boron, alumina and glass. Further, alumina has good thermal conductivity, inertness to most acids and alkalis, high adsorption capacity, thermal stability and electrical insulation, and so on. Also, it is inexpensive, non-toxic and highly abrasive [15–17]. This work presents a method for processing scrap-rubber powder with epoxy resin (EpometF (SiO2 filled), Buehler USA), reinforced with alumina, boron, and glass, to create novel composites in an economic way. The main objective of this research was to determine the mechanical and tribological properties of these composites. Properties of constituent materials were measured by means of nanoindentation. Impact behaviour was investigated by means of drop-weight testing. Scanning electron microscopy (SEM) was used to study the microstructure of these composites. 1.2 Experimental Conditions 1.2.1 Materials Processing A new design of epoxy-based composite, reinforced with fine scrap-rubber powders, boron, alumina and glass bubbles, was prepared in several steps: 1. The fresh scrap rubber was milled with a fast-rotating toothed-wheel mill to obtain fine rubber powder. 2. Very fine solid-epoxy resin and the scrap-rubber powder (10 wt%) were mixed and chemically treated by using toluene and acrylic acid and vinyltriethoxysilane (2%). After treatment, the mixture was dried in the oven to eliminate any trace of the treatment chemicals. The mixture of epoxy resin and rubber powder was then milled 2 h to obtain a homogenous compound then heated at 80 C for 2 h and this mixture was used as matrix. 3. Dry boron and alumina powders (micro-scale particles) were mixed together in predefined ratios and the boron-alumina mixture was heated at 80 C for 2 h to prevent it from absorbing any moisture. 4. The rubber-epoxy and boron-alumina mixtures were combined in a blender and milled for 4 h. 5. Glass bubbles (GB-hollow glass microspheres manufactured by 3 M with a density of 0.227 g/cm3, specified as S38HSS & K1) were added and mixed to obtain homogenous distribution of the constituents (ultrasound) for an additional 2 h. 6. The specimens were then manufactured by hot compacting (double uniaxial action) under a pressure of 70 MPa at a temperature of 180 C for 30 min. All of the specimens (30 mm in diameter) were cooled slowly. 7. Specimens were cured isothermally at 80 C for 48 h. General compositions of all the composites manufactured are given in the Table 1.1 with specified weight ratios for each constituent. In Fig. 1.1 two different specimen sizes are shown after manufacturing by hot compacting. 1.2.2 Mechanical Tests and Microstructure of the Compositions Dynamic compression tests (drop weight tests) were carried out using a universal drop weight test device. A standard conical punch was released from a height of 900 mm. Impact behavior of the manufactured composites were observed over the test specimens with the help of force-time curves. General microstructures in the transversal direction of two compositions are shown in the Fig. 1.2. All of the compositions have shown a considerably homogenous distribution of the reinforcements in the structure. Some small local agglomerations of reinforcement particles are observed in the structure, which may indicate the need for longer and more aggressive mixing. Also, relatively large black rubber particles are seen in the microstructure. 2 A.B. Irez et al.
1.2.3 Damage Analysis by Means of Scratch Test and 3D Optical Roughness Meter Wear testing quantifies the tribological behavior of the composites. Each wear test involved sliding a 2.5 mm diameter zirconia probe over a 1.5 to 2 mm long track at a frequency of 15 Hz under a load of 30 N for either 50,000 cycles or 100,000 cycles. The damage zone was investigated by a 3D optical-surface scanner. The resistance to scratch deformation (damage) was quantified in terms of scratch depth and wear volume. 1.2.4 Nanoindentation to Measure Constituent Properties Nanoindentation was used to measure the mechanical properties of the various constituents of EBAL 4. Nanoindentation was performed with an iNano fit with a Berkovich indenter (Nanomechanics, Inc., Oak Ridge, TN). EBAL 4 was tested with a new high-speed indentation technique in order to mechanically characterize all materials within a square area of 0.3 mm 0.3 mm; results are presented as highly resolved contour maps and spectra of properties. Within the selected area, an array of 60 60 indents was performed (a total of 3600 indents). Each indent had a peak load of 2mN and individual indents were separated by 5 microns. Indentations were performed at a rate of approximately 1 indent per second; thus, all 3600 indents were completed in about 1 h. 1.3 Results and Discussions 1.3.1 Dynamic Compression (Impact) Testing As mentioned earlier, reinforcing materials were incorporated into the epoxy matrix with the aim of improving its toughness. Figure 1.3 shows the impact force as a function of time for each composite. Further, the additional boron incorporated into Table 1.1 Composition of the epoxy-rubber based composites Epoxy-rubber based composition Reinforcements (wt%) Boron Alumina Glass bubbles EBAL 1 5 5 – EBAL 2 5 5 5 EBAL 3 10 5 5 EBAL 4 5 10 5 Fig. 1.1 Macrograph of the sample after compacting and post curing (a) d ¼50mm(b) d ¼30mm 1 Scrap-Rubber Based Composites Reinforced with Boron and Alumina 3
EBAL 3 seemed to weaken the resulting composite as evidenced by the reduced maximum impact force, relative to EBAL 2 and 4. However, the hollow glass bubbles had a toughening effect, which is consistent with previous studies [18]. 1.3.2 Wear Testing Figure 1.4 shows the profilometry scan for EBAL 2 following a 50,000-pass wear test. The wear tracks from all tests were scanned in the same way, and the depth and volume were measured; these results are provided in Fig. 1.5. High interfacial shear stress is the primary cause of tribological damage [19]. However, fluctuation in stress is also a contributing factor – ahead of the probe, the stress is compressive, but behind the probe, it is tensile. Based on these results, it seems that additional boron provides a greater resistance to wear damage than additional alumina. There is one more interesting observation from these tests. As shown in Fig. 1.6, it seems that some toughening particles were pulled out during wear testing. Such dislodged particles may have the effect of accelerating wear in an uneven and unpredictable way. In Fig. 1.5 damaged areas were given as a function of lost volume during wear tests. By increasing the number of the cycles, surface damage increases as expected. On the other hand, it is difficult to establish a correlation about wear resistance depending on the reinforcement elements. Fig. 1.2 Microstructure of EBAL specimens (a) EBAL 1, (b) EBAL 2, (c) EBAL 3, (d) EBAL 4 4 A.B. Irez et al.
1.3.3 Nanoindentation Figure 1.7 shows the microscopic image for the area of EBAL 4 selected for high-speed indentation. The same set of 3600 measurements of Young’s modulus and hardness are presented both as contour plots (Figs. 1.8 and 1.9) and as histograms (Figs. 1.10 and 1.11). Each indentation manifests as one pixel in the contour plot and as one relative count in the histogram plot. The contour plots of Young’s modulus and hardness give a quick impression of the mechanical inhomogeneity of these materials: The rubber appears purple and the reinforcing ceramics appear red. The mechanical influence of the reinforcing Fig. 1.3 Impact behavior of EBAL composites by dynamic compression test Fig. 1.4 Three dimensional damage traces obtained in the direction of width and length for the specimen EBAL I for 50 k cycles 1 Scrap-Rubber Based Composites Reinforced with Boron and Alumina 5
Fig. 1.5 Lost volume after 50 k and 100 k cycles of wear testing Fig. 1.6 Damage trace over the specimen EBAL- I 50 k cycles Fig. 1.7 Surface of EBAL 4 subject to high-speed indentation array. Area is 0.3 mm on a side 6 A.B. Irez et al.
particle does not extend beyond its boundaries. In other words, the presence of the reinforcing particle does not seem to mechanically alter nearby rubber or epoxy, as evidenced by the sharp contrast in color at material boundaries. The histograms quickly show which materials are most influential in the tested area. There are clear peaks for the epoxy (E ¼ 11.0GPa, H ¼ 0.59GPa) and glass (E ¼ 58.2GPa, H ¼ 8.14GPa). However, these peaks are not sharp: most measurements fall between them, thus indicating a good mixture of materials with varied properties. Finally, although the ceramic particles draw the eye in the contour plots, their influence hardly registers in the histograms. As development of these materials continues, such histograms may be a useful mechanical characterization for predicting performance. Fig. 1.8 Modulus contour plot for EBAL 4 Fig. 1.9 Hardness contour plot for EBAL 4 1 Scrap-Rubber Based Composites Reinforced with Boron and Alumina 7
1.4 Conclusions In this study, hollow glass spheres were used to see their effect of damping. According to the drop weight test, they increased the maximum force values over dynamic compression curves. In other words, they improved the energy absorbing capacity of the composite. Also hollow glass spheres tend to promote decrease in density. From these perspectives, their use can be generalized after a detailed study. In addition, the positive effect of boron particles on the surface wear resistance was seen after a comparative study. Results were found promising and this situation can be supported in the light of longer tests. The histograms obtained as the result of the nanoindentation testing show which materials are most influential in the tested area. These histograms may be useful mechanical characterization for predicting performance. References 1. May, C.A.: Epoxy Resins-Chemistry and Technology, 2nd edn. Marcel Dekker/Wiley, New York (1988) 2. Wetzel, B., Rosso, P., Haupert, F., Friedrich, K.: Epoxy nanocomposites–fracture and toughening mechanisms. Eng. Fract. Mech. 73(16), 2375–2398 (2006) 3. Garg, A.C., Mai, Y.: Failure mechanisms in toughened epoxy resins-a review. Compos. Sci. Technol. 31(3), 179–223 (1988) 4. Lee, J., Yee, A.: Inorganic particle toughening I: micro-mechanical deformations in the fracture of glass bead filled epoxies. Polymer. 42(2), 577–588 (2001) 5. Manzione, L.T., Gillham, J.K.: Rubber-modified epoxies. I. transitions and morphology. J. Appl. Polym. Sci. 16, 889–905 (1981) Fig. 1.10 Histogram of 3600 modulus measurements by nanoindentation on EBAL 4 Fig. 1.11 Histogram of 3600 hardness measurements by nanoindentation on EBAL 4 8 A.B. Irez et al.
6. Nakamura, Y., Yamaguchi, M., Okubo, M., Matsumoto, T.: Effects of particle size on mechanical and impact properties of epoxy resin filled with spherical silica. J. Appl. Polym. Sci. 45(7), 1281–1289 (1992) 7. Okazaki, M., Murota, M., Kawaguchi, Y., Tsubokawa, N.: Curing of epoxy resin by ultrafine silica modified by grafting of hyperbranched polyamidoamine using dendrimer synthesis methodology. J. Appl. Polym. Sci. 80(4), 573–579 (2001) 8. Wetzel, B., Haupert, F., Zhang, M.Q.: Epoxy nanocomposites with high mechanical and tribological performance. Compos. Sci. Technol. 63 (14), 2055–2067 (2003) 9. Zee, R.H., Huang, Y.H., Chen, J.J., Jang, B.Z.: Properties and processing characteristics of dielectric-filled epoxy resins. Polym. Compos. 10 (4), 205–214 (1989) 10. Jin Kim, D., Hyun Kang, P., Chang Nho, Y.: Characterization of mechanical properties of γAl2O3 dispersed epoxy resin cured by γ-ray radiation. J. Appl. Polym. Sci. 91, 1898–1903 (2004) 11. Arayasantiparb, D., McKnight, S., Libera, M.: Compositional variation within the epoxy/adherend interphase. J. Adhes. Sci. Technol. 15(12), 1463–1484 (2001) 12. Nielsen, L.E., Landel, R.F.: Mechanical Properties of Polymers and Composites, 2nd edn. Marcel Dekker, Inc., New York (1994) 13. Rothon, R. (ed.): Particulate-Filled Polymer Composites. Longman Scientific and Technical, London (1995) 14. Rothon, R.N.: Adv. Polym. Sci. 139, 67–107 (1999) 15. Mirjalili, F., Mohamad, H., Chuah, L.: Preparation of nano-scale α-Al2O3 powder by the sol–gel method. Ceram-Silika´ty. 55, 378–383 (2011) 16. Amirshaghaghi, A., Kokabi, M.: Tailoring size of α-Al2O3 nanopowders via polymeric gel–net method. Iran. Polym. J. 19, 615–624 (2010) 17. Zhang, Z., Lei, H.: Preparation of α-alumina/polymethacrylic acid composite abrasive and its CMP performance on glass substrate. Microelectron. Eng. 85, 714–720 (2008) 18. Keivani, M., et al.: Synergistic toughening in ternary silica/hollow glass spheres/epoxy nanocomposites. Polym.-Plast. Technol. Eng. 54(6), 581–593 (2015) 19. Zaimova, D., et al.: Manufacturing and damage analysis of epoxy resin-reinforced scraprubber composites for aeronautical applications. In: Experimental Mechanics of Composite, Hybrid, and Multifunctional Materials, vol. 6, pp. 65–76. Springer International Publishing, Cham Heidelberg\New York\Dordrecht\London (2014) 1 Scrap-Rubber Based Composites Reinforced with Boron and Alumina 9
Chapter 2 Characterization of Thermoplastic Matrix Composite Joints for the Development of a Computational Framework Joseph R. Newkirk, Cassandra M. Degen, and Albert Romkes Abstract Many industries, notably the automotive and aerospace industries, are now utilizing thermoplastic matrix composites (TPMCs) for their improved strength and stiffness properties compared to pure thermoplastic polymers, as well as their manufacturability compared to traditional thermoset matrix composites. The increase in the utilization of TPMCs ushers in the need for the development and characterization of joining methods for these materials. A widely used technique for joining thermoplastics is ultrasonic spot welding (USSW). During USSW, high frequency, low amplitude vibrations are applied through an ultrasonic horn resting on the polymer surface. The vibrations induce frictional heat, producing a solid state joint between polymer sheets. Advantages such as short weld cycle time, fewer moving components and reproducibility make this technique attractive for automation and industrial use. Prior work showed USSW as a feasible, repeatable joining method for a polycarbonate matrix filled with chopped glass fibers. The mechanical properties required for full characterization of the TPMC used in this work were not provided by the manufacturer. As such, the constitutive behavior of both as-received and USSW thermoplastic composite material (polypropylene matrix filled with 30 wt% chopped glass fibers) was characterized. The fiber orientation and distribution in TPMCs has a direct impact on constitutive behavior. To characterize these qualities, optical techniques such as scanning electron microscopy (SEM) and micro computed tomography (micro-CT) were employed. Digital image correlation (DIC) was used to acquire full field strain measurements from the composite material under different loading scenarios. Because the constitutive behavior of polymers is greatly dependent on temperature, temperature measurements during the USSW process and measurement of mechanical properties as a function of temperature will be conducted through infrared (IR) imaging and dynamic mechanical analysis (DMA), respectively. Following the calibration of the constitutive model for the polypropylene matrix TPMC, the mechanical and thermal properties will be used to develop a computational framework for the purpose of predicting the structural response of a composite joint under various loadings. Keywords Composite joining • Ultrasonic spot welding • X-ray micro computed tomography • Thermoplastic matrix composites • Fiber orientation 2.1 Introduction With growing regulations in many industries, especially in the case of emissions, the demand for lightweight strong materials is increasing rapidly. As a result of this, many of these industries are turning to composite materials [1]. The increased usage of these materials drives the need for development of both the manufacturing and joining processes applied to them. Composite materials are generally joined using mechanical fasteners, adhesive bonding, or fusion welding processes. Each of these methods brings unique advantages and disadvantages. Mechanical fasteners are a common joining method, used traditionally with isotropic materials. Anisotropic composite materials are more sensitive to holes and cutouts which shear fibers and create large stress concentrations. Adhesive bonding eliminates the stress concentration associated with mechanical fasteners, but also requires more time intensive preparation work, involves unavoidable curing cycles, and the quality of the bond is dependent on the adherend materials [2]. Fusion welding relies on the melting of the materials to be joined, intermolecular diffusion across the joint, and re-solidification of the matrix material. Fusion bonding techniques generally eliminate, or at least mitigate, stress concentrations and result in joint efficiency which can approach the bulk properties of the joined materials [2]. J.R. Newkirk • C.M. Degen (*) • A. Romkes Department of Mechanical Engineering, South Dakota School of Mines and Technology, 501 E. Saint Joseph St., Rapid City, SD 57701, USA e-mail: cassandra.degen@sdsmt.edu #Springer International Publishing AG 2018 P.R. Thakre et al. (eds.), Mechanics of Composite and Multi-functional Materials, Volume 6, Conference Proceedings of the Society for Experimental Mechanics Series, DOI 10.1007/978-3-319-63408-1_2 11
Historically, composites have been used in applications where cost is not a major design consideration. These composites have traditionally been thermosetting matrix composites (TSPCs), but the inherent limitations of TSPCs, most notably the inability to be joined by fusion welding and high manufacturing cost, have led to a surge in the usage of thermoplastic matrix composites (TPMCs) [2]. TPMCs have a number of advantages over TSPCs: higher fracture toughness, better resistance to wet environments, shorter processing cycles which lower manufacturing cost, recyclability [3]. Of consequence for this work is their ability to be melted and re-solidified which makes TPMCs suitable candidates for fusion welding processes. Fusion welding can be broken into two categories: bulk heating, and frictional heating. Bulk heating involves common processes like autoclaving and compression molding, frictional heating includes spin welding, vibrational welding and ultrasonic welding [2]. The technique studied in this work is ultrasonic spot welding (USSW). In this joining process, the energy transferred to a thermoplastic material subjected to ultrasonic vibrations is dissipated through intermolecular friction, leading to heating of the material and localized melting. An ultrasonic welder typically consists of six main components: a base-plate, pneumatic press, control system, transducer, booster, and welding horn [4]. The ultrasonic spot welder used in this work is a 20 kHz Dukane iQ series Ultrasonic pneumatic press 43Q-220 (Rapid City, SD, USA) (Fig. 2.1). When joining materials through USSW, a large range of weld qualities can be obtained by varying weld parameters. Taking this into consideration, a study was performed using varying weld parameters. The parameters used in the ultrasonic welding process were correlated to the quality of the welds created evaluated by lap shear testing in order to select suitable weld parameters. In general, the mechanical behavior of composite materials differs based upon the loading direction. The behavior of the TPMC used in this work was initially evaluated along two perpendicular loading directions. The mechanical response of the material under tensile loading was evaluated with respect to a fiber orientation that was assumed to be caused by the rolling step in the processing of the TPMC sheets. The orientation of the fibers in a material is of consequence when evaluating mechanical properties, particularly for short fiber composites, a common type of TPMC. Changes in fiber orientation are related to a number of factors, such as the geometrical properties of the fibers, viscoelastic behavior of the matrix material, and the change in shape of the material produced by the processing operation [5]. With processes varying between manufacturers, a means of characterizing fiber orientation is necessary. To evaluate the fiber orientation in the TPMC used in this work, the optical methods of x-ray micro computed tomography (μCT) and scanning electron microscopy (SEM) were employed. These methods were used to characterize the fiber orientation of a TPMC before and after ultrasonic spot welding of the material. Fig. 2.1 20 kHz Dukane iQ series ultrasonic pneumatic press 43Q-220 used for this work 12 J.R. Newkirk et al.
The end goal of this work is to couple the experimental and optical characterization of ultrasonic spot welds in a chopped fiber-thermoplastic matrix composite with computational efforts to produce a physics-based model to guide the design of composite joints. The experimental work discussed here will guide the computational efforts, which will first focus on the development of a mathematical model that accurately describes the multi-physics of the welding process. This model will be based on principles of continuum mechanics and therefore will address both the mechanics and thermodynamics of the welding process, as well as the coupling between these. The development of this model will help guide the design of composite joints created with USSW. This predictive model will be used to derive optimal welding parameters based upon weld application and material properties. This will shorten the time necessary to apply USSW in different applications proving useful for many industries concerned with the manufacturing of novel, light-weight materials. 2.2 Mechanical Characterization of Composite Material The use of USSW on polymer matrix composites has been studied and characterized previously [6]. In this work, initial weld characterizations were performed on pure polypropylene matrix material, followed by further characterization of polypropylene matrix composite materials welds. The composite used in this work was ESP 105 CC, a polypropylene matrix TPMC, produced by RTP Company (Winona, MN, USA). The ESP 105 CC composite is composed of 30 wt% chopped glass fibers and was provided in 1.52 mm thick extruded sheets [7]. The mechanical properties of this material provided by the manufacturer can be found in Table 2.1. The ultrasonic spot welding process varies based upon the input parameter settings used during the weld. As such, a study was necessary to investigate the effects of weld parameters. The parameters that were varied in this study were amplitude, time and weld pressure. In the interest of limiting the use of the composite material, the study was performed on pure polypropylene material. Since the composite material is composed of 70 wt% pure polypropylene, the response to the welding parameters was assumed to be similar to that of plain polypropylene sheeting. The quality of the welds resulting was characterized both optically, by observing bond area, and mechanically, by measuring peak load during lap shear testing. The criteria used to characterize the welds were total bonding area (shear area), and the peak load under lap shear. The bonding area was measured optically prior to lap shear testing using the open source image analysis software ImageJ. In the preparation of the lap shear samples, tabs were adhered to the ends of each test specimen to avoid the production of a bending moment during testing and the samples were tested at a rate of 1.3 mm/min as outlined in ASTM 3164-03 [8]. Samples with a broad range of weld parameters were investigated in order to isolate the parameters with the greatest impact on the quality of the weld. It was found that these parameters were the weld time and weld amplitude. The failure mode exhibited by the samples was almost immediate fracture, at the base of the weld, when the maximum load was reached. The location of the failure in the samples is most likely due to bending stresses, and subsequent stress concentrations that are inherent in lap shear testing, which are also highlighted by Degen et al. [6]. The results of the optical and mechanical characterization of plain polypropylene sheeting are shown in Table 2.2. From this study, the following welding parameters were found to create welds with the largest bond area and highest peak load during lap shear testing in pure polypropylene sheeting: an amplitude of 24μm, a weld time of 2 s and a weld pressure of 483 kPa. The chosen parameters are not immediately apparent upon inspection of Table 2.2, but these values were found to create the largest shear weld areas while exhibiting the least standard deviation and therefore creating the most repeatable welds. Upon finding suitable welding parameters for the pure polypropylene sheeting, the same parameters were used to weld the polypropylene matrix composite, ESP 105 CC. As a reference, a representative 25.4 mm 25.4 mm USSW sample of the composite material is shown in Fig. 2.2. Five lap shear samples of both the plain polypropylene and composite materials were welded using the above parameters and evaluated in terms of bond area, maximum load during lap shear, and maximum shear stress. From the results shown in Fig. 2.3, it is apparent that the pure polypropylene samples exhibited better performance in terms of both average shear area and average failure load in lap shear than their composite material counterparts. Table2.1 Mechanical properties of ESP 105 CC [7] Property Value (Metric units) Young’s modulus 6205MPa Ultimate tensile strength 55MPa Elongation at break 1.5–2.5% Impact strength (IZOD, unnotched) 0.32 J/mm 2 Characterization of Thermoplastic Matrix Composite Joints for the Development of a Computational Framework 13
The maximum load measured in lap shear for the composite samples was directly correlated to the smaller measured shear area. When normalizing the average load by corresponding shear area, the results for the shear stress at failure are more closely aligned for both materials, with the composite showing slightly higher maximum stress. The results of the lap shear testing indicated that the welding parameters found to create suitable welds in the pure polypropylene sheeting may not have been suitable for creating welds in the polypropylene TPMC. The change in the apparent viscosity of the polymer melt present with the addition of the glass fibers in the polypropylene composite was one noted difference in the joining of the two different materials. This effect will be further studied for ESP 105 CC. Rolled sheets of thermoplastic matrix composite often show a directional fiber orientation in the rolling direction. In fact, the manufacturing process is often manipulated to achieve desired fiber orientation to achieve optimal strength properties in the loading direction. Short fiber composites manufactured with conventional molding techniques possess axial mechanical properties approaching those obtained from continuous-fiber composites of equal fiber volume fraction [9]. Taking this into consideration, the mechanical response of the polypropylene composite when loaded both parallel to, and perpendicular to the orientation of the fibers must be characterized. To achieve an initial measure of this behavior, dog bone tensile samples were prepared and tested according to ASTM D638-08 [10]. Samples were cut out in perpendicular and parallel orientations relative to the observed fiber orientation. Representative results of the behavior observed in this testing are shown in Fig. 2.4. All samples tested exhibited the same brittle failure mode. A higher maximum tensile load was achieved in the samples with the loading direction parallel to the fiber orientation when compared to with the loading direction oriented perpendicular to the fiber orientation. This loading behavior indicates that the material is roughly orthotropic in nature. Fig. 2.2 Representative ESP 105 CC USSW sample Table 2.2 Welding parameter combinations and results of both optical and mechanical analysis of welds Amplitude Time Weld pressure Energy Area Peak load (Lap Shear) m s kPa J mm2 N 24 1.50 345 168 142 903 30 161 82 459 24 2.00 192 112 671 30 189 90 510 24 1.50 414 160 121 743 30 161 100 511 24 2.00 195 141 722 30 189 93 567 24 1.50 483 177 144 825 30 158 94 637 24 2.00 208 141 863 30 196 111 588 14 J.R. Newkirk et al.
The material properties of ESP 105 CC must be characterized, in order to provide the necessary input parameters for the development of a computational model. Several approaches have been explored to characterize these properties. The first method employed was nanoindentation, a method in which the load and displacement of a Berkovich indenter are correlated to the material properties of a sample [11]. Nanoindentation allows for the investigation of the mechanical properties of a material at a submicron scale [11]. This capability is intriguing when applied to composite materials as the properties of the matrix and the fiber can be evaluated separately, in addition to the bulk properties of the material. This method relies on Fig. 2.4 Stress-strain diagrams for ESP 105 CC composite tensile samples tested perpendicular and parallel to the fiber orientation tested at 5.08 mm/min Fig. 2.3 Average Shear area and average maximum load in lap shear (a) and average shear stress (b) for pure polypropylene and ESP 105 CC composite lap shear Samples 2 Characterization of Thermoplastic Matrix Composite Joints for the Development of a Computational Framework 15
estimating the contact area of the indenter under load, which can lead to difficulties when applied to polymers. These difficulties are a result of the load resolution of the nanoindenter which is often too coarse for use with thermoplastics. Softer materials, like thermoplastics deform more than traditional materials in the loading range typically used in this method, and this leads to poor correlation in the curve fitting methods used with this technique [12]. Another method that experiences limitations when applied to polymers is the use of strain gauges. When working with thin samples of composite materials, the material being evaluated is often more compliant than the strain gauge being used. This results in a phenomena known as the “reinforcement effect” [13]. As a result of the reinforcement effect, strain gauges measure lower strains compared with the strains experienced locally by the test sample which can lead to significant errors in determination of strain and elastic modulus [13]. To mitigate the inherent issues involved with evaluating the material properties of polymers, a non-contact evaluation method was employed. This method is digital image correlation (DIC) [14]. DIC works by applying a speckle pattern to a sample and comparing digital photographs of that sample at different stages of deformation. By tracking blocks of pixels, the system can measure surface displacement and build up full field 2D and 3D deformation vector fields and strain maps. In order to achieve effective correlation, the pixel blocks utilized need to be random and unique with a good range of contrast and intensity levels [14]. In this work, a pattern was applied to samples using spray paint, with the nozzle at a distance from the sample that allows for a sparse amount of coverage. This method was used to evaluate the full field stress and strain exhibited in ESP 105 CC. Initial DIC analysis was performed using GOM Correlate 2-D analysis software (Braunschweig, Germany). Early testing has produced values of extension, stress, and strain that correlate well with values calculated through traditional means. This correlation allows for material properties to be extrapolated from this testing with some certainty, aiding in characterizing mechanical properties, like Poisson’s ratio, that are difficult to measure using the methods outlined above. 2.3 Optical Characterization of Composite Material The orientation of fibers in a composite material directly affects the mechanical properties of the material. This is especially true when working with discontinuous fiber composites. It can be difficult to manufacture TPMCs with a consistent fiber orientation. Practical difficulties arise in using new or existing processes to align fibers with a high degree of precision because of the large number of factors affecting fiber orientation during processing [9]. In the polypropylene matrix composite studied in this work, chopped glass fibers in the material were randomly oriented through the extrusion screw, and some orientation of the fibers was thought to be achieved through the rolling of the material into sheets. No further details of the anisotropic properties of the material were provided by the manufacturer. Taking this into consideration, the fiber orientation of the as-received ESP 105 CC sheeting, as well as the effect of the USSW process on the fiber orientation was characterized and compared to the results of mechanical characterization. The use of micro-computed tomography (μCT) as a fiber orientation characterization technique has been investigated previously by Alemdar et al. [15]. It was found that the μCT could successfully characterize fiber diameter, length and orientation in TPMC. In this work, a Xradia MicroXCT-400 was used for all μCT characterization. In order to establish a baseline fiber orientation, the fiber orientation of the as-received ESP 105 CC was qualitatively characterized. AμCT image of the as-received TPMC is shown in Fig. 2.5. From this image, it was apparent that there was a strong fiber orientation along Fig. 2.5 ESP 105CC as-received μCT image 16 J.R. Newkirk et al.
the x-axis of the sample, indicating the mechanical properties of the ESP 105 CC were roughly orthotropic. The orthotropic nature of the material under axial loading was observed in the study performed during mechanical characterization of the polypropylene composite and correlates well to these μCT findings (Fig. 2.4). From these scans, fiber diameter was approximately 13–15 microns and fiber length was approximately 270 microns. After establishing a baseline of the fiber orientation for the as-received TPMC, the effect of the USSW process on the orientation of the fibers was investigated. This study was performed on 25.4 mm 25.4 mm 1.52 mm thick TPMC samples welded using the parameters chosen during the weld parameter study (Fig. 2.2). Figure 2.6a shows a schematic of the cross-section of the joint of two sheets. The area under investigation in theμCT image of Fig. 2.6b is highlighted with the box in Fig. 2.6a. This area encompasses the region just to the right of the void left where the weld horn was inserted. Here, the orientation of the fibers of each sheet was roughly perpendicular to one another, with the top sheet in Fig. 2.6b having fiber orientation along the x-axis and the bottom sheet having fiber orientation along the y-axis. In the left region of Fig. 2.6b, which was directly in contact with the weld horn, fiber folding in the direction of travel of the horn is present. In addition, the flow of the weld has a distinct, cone-shaped region converging towards the right of the sample at the interface of the two sheets. In this region, there are two distinguishable fiber orientations relative to the flow. At the edge of the flow, the fibers are aligned with the weld flow. In the core of the weld flow, the fibers are oriented perpendicular to the direction of the weld flow. This behavior draws parallels to behavior observed in the study of injection molding of TPMCs [16]. When injection molding short fiber TPMCs, orientation of fibers is caused by differences in velocity between the two ends of a fiber. When the melt flow is homogenous, no change in the orientation of fibers will take place. When the flow is not homogenous, the variation in velocity can be perpendicular or parallel to the flow direction depending on whether the flow is convergent or divergent [16]. This behavior may lead to the differing alignment of fibers at the edge of the weld flow region. It was also apparent that the fiber orientations seen in the two pieces of the welded sample were oriented roughly perpendicular to one another, and this should be taken into consideration when analyzing the changes in fiber orientation throughout the weld affected region of the sample. From this analysis, it can be seen qualitatively that the fiber orientation is more randomly oriented in the region affected by the weld than it is in the rest of the sample and this will behavior will need to be further quantified. The fiber orientation of the ESP 105 CC composite material was further characterized using scanning electron microscopy (SEM). As-received samples were polished along the short axis of the sheet in order to view the apparent fiber orientation throughout the cross section of the sheet. A SEM image of the cross section of as-received ESP 105 CC is shown in Fig. 2.7b. In this cross section, three distinct regions exist in which the fiber orientation is grouped. In Fig. 2.7c, the fibers generally have an orientation in line with the length of the sheet, which is in agreement with the orientation observed throughμCT. There is a region in the middle of the cross section, Fig. 2.7a, which exhibits a much more random fiber orientation than observed in the outer layers, Fig. 2.7c. It is thought that these zones are a product of the rolling manufacturing method. In terms of evaluating the mechanical response of this material under loading, the interaction between these three zones will need to be characterized for the constitutive behavior of the polypropylene to be modeled effectively throughout the entire cross section. Fig. 2.6 Schematic of USSW cross section (a) with μCT field of view outlined in red, and USSW ESP 105 CCμCT image (b) 2 Characterization of Thermoplastic Matrix Composite Joints for the Development of a Computational Framework 17
RkJQdWJsaXNoZXIy MTMzNzEzMQ==