CHAPTER 1 INTRODUCTION Composites exist almost everywhere, from the human body to spaceships. Many materials existing in nature, such as bones in the human body, and wood and bamboo in the forest, derive their superb mechanical properties by combining two or more macroscopic components or constituents which can be distinguished readily by using a microscope. Human bones are naturally occurring fiber reinforced ceramic composites consisting of collagen fibers and a gel-like matrix of calcium phosphate (hydroxyapatite) mineral. The fibers are arranged in a relatively disordered side-toside packing of collagen molecules but in a regular axial structure and are surrounded by the mineral matrix, in the form of small, poorly-crystalline hydroxyapatite crystals, resulting in the strong structure of human bones [1]. Similarly, microscopic examination of wood and bamboo shows a pronounced fibrillar structure. Both of them have many cellulose microfibrils arranged longitudinally and embedded in a lignin matrix, yielding good strength and stiffness properties. Bamboo is especially strong and stiff due to the lower microfibrillar angle with the fiber axis and thicker cell walls [2]. Bamboo, as a result, has been called “nature’s glass fiber” [3]. Most engineering structural materials, like thermoset resin-based composites used in space applications, concrete beams reinforced with steel wire for buildings and metallic alloys for automobiles, are also combinations of two or more phases designed to give better performance and properties, such as strength, stiffness, toughness, thermal resistance, than each single phase material. Mixing two or more distinct materials to synergistically achieve the desired application properties is the fundamental concept of composites. The properties of composites are mostly determined by the properties of the individual constituents as well as the compatibility or adhesion between them. 1 Fiber reinforced composite materials were widely used as early as 4000 B.C. in Egypt for making laminated writing materials from the papyrus plant [4]. There was a rapid growth in the use of fiber reinforced polymer composites for structural engineering applications in the 1970s [3]. This was because of the composites fabricated using fibers such as graphite and glass which could achieve high strengths. In addition, these composites had significantly lower density compared to metals which they were intended to replace. Many kinds of fiber reinforcement, such as strands, woven and non-woven fabrics, yarns, and long and short fiber mats, are commercially available at present. These may be made using synthetic fibers such as carbon, glass, Kevlar®, boron or natural fibers such as cotton, ramie, jute and flax. The use of these reinforcements depends on the desired end properties of applications of the composites. It is common to classify fiber reinforced composites on the basis of the geometry of a representative unit of reinforcement, as presented in Figure 1.1 [5]. In the scheme of this classification, discontinuous short fiber reinforced composites have Fiber reinforced composites (Fibrous composites) Single layer composites (including composites having same orientation and properties in each layer) Multilayered (angle ply) composites Laminates Hybrids Continuous fiber reinforced composites Discontinuous fiber reinforced composites Unidirectional reinforcement Bidirectional Reinforcement (woven reinforcements) Random orientaion Preferred orientation Figure 1.1 Classification of Fiber Reinforced Composites 2 contributed significantly to consumer goods industries because of their ease of fabrication into complex shapes, simple manufacture (mostly by injection molding), and low cost. Micromechanical models and theories of discontinuous short fiber reinforced systems have been developed, which offer convenient solutions to their mechanical behavior and properties [5-7]. On the other hand, although continuous fiber reinforced composites are well understood and modeled, such materials cannot be processed easily for mass production, particularly if different shapes are desired. They are generally confined to products with simple shapes for which the property benefits outweigh the cost penalty [6]. Discontinuous short fiber reinforced composites, therefore, have been used in a variety of applications in industries, such as architectural structures, aircraft, automobiles, ships, sports equipment, and electrical products. In the automotive industry, bulk composite materials, mainly glass fiber thermoplastics, are used on a large scale. The mass production of glass fiber and other synthetic fiber reinforced petroleum based plastics, however, has been causing many environmental problems. These include increase in the number of landfills required, depletion of petroleum reserves, and increase in the amount of CO2 emitted into the atmosphere. In addition, it is apparent that the single most serious problem facing the world today is abnormal weather conditions, caused by the greenhouse effect. The temperature appears to be increasing by a couple of degrees every summer all over the world. As a result, glaciers at the poles are melting with surprising speed. Temperatures have been reaching up to 40°C on some days during the last two summers, and storms have also been harsher than usual in some countries. Global warming, even most skeptics have concluded, is already a reality and a major issue today [8]. To deal with this problem, many developed countries have been enforcing the Kyoto protocol for the past 8 years, mainly to reduce the emissions of carbon dioxide 3 and five other greenhouse gases such as water vapor and ozone. Industry, in many countries, is required to replace conventional materials with environmentally friendly plastics to preserve our environment. Hence the automotive sector, which is a massive user of the materials, has been conducting research on environmentally friendly, sustainable materials [9-11]. The biodegradable materials have been substituted for external and internal components such as underbodies, door trim panels, interior parts commonly made out of conventional composites like glass fiber reinforced plastics (GFRP). It is difficult to discard these composites in an environmentally friendly way or recycle them since the resins used are mostly thermo-set. Even recyclable thermoplastics containing glass fibers are mostly disposed of in landfills, while only a small fraction of them are incinerated. Although a small fraction of the thermoplastic composites are incinerated as solid fuel in energy recycling systems, glass fibers in the composites reduce the net heat in a furnace and thus require more energy, which may damage the furnace. In addition, since glass fibers are more than one-half times heavier than natural fibers, it is obvious that vehicles made of GFRP components are heavier and consume more gasoline than ones that use natural fiber reinforced composites [12, 13]. These characteristics have driven the automotive industry to make significant efforts to develop environmentally friendly reinforced plastics based on natural and fully sustainable, mainly plant derived, materials. Henry Ford, believing that “the most environmentally friendly thing you can do for a car that burns gasoline is to make lighter bodies,” had hoped to shift from steel to lower-weight materials [10]. Reducing the total weight of cars should aid in saving fossil resources because, for instance, nearly half of the mineral oil products consumed in Germany are burned as fuel in cars [14]. Interest in lightweight, strong, and low-cost, natural fibers is rising, and these fibers are poised to replace glass in numerous interior parts. Unlike glass fibers, they 4 release less CO2 as well as NOx and SOx gases and dust and thus do not contribute to worsening of the environment, in fact helping it. Recently, European automotive manufacturers have started using natural fibers such as jute and flax instead of glass fibers [15, 16]. Mercedes Benz set a precedent in 1994 by using jute reinforced plastic for the interior door panels of its E-class vehicles [10]. Such natural fiber reinforced thermoplastics based on polypropylene, polyethylene, etc., which are derived from petroleum, can only be called ‘Semi Green’ composites. They are neither fully biodegradable nor environmentally friendly. They can neither be digested by the natural metabolism nor by the industrial metabolism. Furthermore, it is predicted that, at the current rate of consumption, petroleum will last only for the next 50 years or so [17]. As supplies dwindle, the cost of thermoplastics will undoubtedly increase; thus, it is inevitable that thermoplastic resins have to be replaced eventually by bioderived or plant based resins with comparable mechanical and thermal properties. In order to meet environmental aims fully, both resins and composites need to be fully ‘Green’ to avoid further destruction of the environment. Although the currently available fully ‘Green’ composites, often called biocomposites, do not meet all requirements for automotive components, they will be the next innovational materials in the near future. Furthermore, fully ‘Green’ hybrid composites, which have more than two constituents, would be quite attractive for various products because the composites can be engineered to have the desired properties. This thesis presents the development of fully ‘Green’ nonwoven kenaf fiber mat composites based on modified soy protein isolate and engineered green hybrid composites made through fibrillated bamboo hybridization. These composites are designed to obtain improved mechanical properties by modifying soy protein isolate (SPI) resin and the addition of fibrillated bamboo cellulose. 5 CHAPTER 2 LITERATURE REVIEW 2.1 Fully ‘Green’ Composites There is growing urgency to develop and commercialize new crops, new biobased products and other innovative technology which don’t rely on fossil fuel [18]. There are several advantages of fully green composites over petroleum based composites: low density, low cost of the components, reduced dermal and respiratory irritation, enhanced energy recovery, and biodegradability. Fully green composites (biocomposites) are becoming one of the most important factors in creating a more ecological future. The number of developments in making biocomposites using materials obtained from renewable resources as reinforcements and matrix have been increasing year by year, spurred by the growing seriousness of environmental problems [19]. Reinforcing biofibers, biodegradable polymers and biocomposites have been reviewed by Mohanty et al. [20]. Various natural plant based fibers and fully biodegradable resins, which are truly sustainable and renewable on a yearly basis, are used to fabricate green composites for diverse applications. Manila hemp fibers were incorporated into starch based emulsion-type biodegradable resin, resulting in unidirectional green composites [21]. Their flexural and tensile strength were reported to be 104 MPa and 192 MPa, respectively. Jute fabric composites based on poly(3-hydroxybutyrate-co-8%-3hydroxyvalerate) were fabricated to evaluate the biodegradability [22]. In the degradation study, the weight of these composites decreased by 50% after 150 days in compost burial. Biocomposites using leaf fiber, henequen, and powdered cellulose acetate were injection-molded by Hokens et al. [23]. They reported that cellulose acetate, unlike other biopolymers, showed comparatively better compatibility with natural fibers. 6 A large number of interesting applications made out of biodegradable materials are appearing commercially all over the world [24]. In the US, Canada and Australia, wood fiber based composites for buildings have been under development for some time. In India and South America, jute and sugar cane bagasse fibers are used in low cost housing, and rice husk based composites have been developed recently to make planking/shelving. In Europe, all automotive companies are seriously considering green composites for their components [25]. Recently, FUJITSU Corporation in Japan introduced biodegradable laptop computer casings made out of poly(lactic acid) (PLA), which is derived from corn starch [26]. NEC Corporation in Japan has presented a cell phone made of kenaf fiber reinforced bio plastics based on PLA resin for the first time in the world [27]. In some of these applications, the flexural and impact properties of the composites as well as their tensile properties are usually considered important because of the applied forces experienced during the use. Most green composites of today, however, don’t possess adequate flexural and impact properties compared with conventional composites used in engineering structural applications. Consequently, the industrial usage requires improvements before they can be used on a large scale. Manufacturing techniques for commercial production of green composites are also being developed [28]. These techniques are mostly based on well-established processing methods in composite technology such as compression and injection molding, filament winding techniques, and hand lay-up. Some researchers have focused on the study of the fabrication process for green composites by employing compression and injection molding which are already used for conventional composites [29, 30]. An advantage is that further new equipment does not have to be invented for mass-production of biocomposite materials. When both the manufacturing techniques and the properties of the composites are fully investigated 7 and optimized, industry will be ready to contribute to the preservation of the environment. 2.1.1 Fully ‘Green’ Composites Based on Soy Protein Resin Soy protein resin derived from soybean has long been regarded as a biopolymer candidate. Henry Ford began experimenting with composites based on soybean plastic as early as 1941. Mohanty et al. [31] stated that soy based plastic is one of the best biopolymers, a group which includes polylactides and starch plastics. Soy proteins have been subjected to research to fabricate fully green composites using flax woven fabric, flax yarn and ramie fiber [32, 33]. Chabba et al. [34] fabricated cross-linked soy flour based composites reinforced with flax yarn and characterized their tensile and flexural properties. They found out that the green composites are strong and durable enough for low-load indoor applications. Cheng et al. [35] developed wheat straw particleboard using soybean flour modified with urea and urease inhibitor N-(n-butyl) thiophosphoric triamide for building panels. Liu et al. [36] prepared alkali-treated Indian grass fiber to make soy based biocomposites with a twin-screw extruder and an injection molder. In their research, Indian grass fiber treated with alkali solution significantly increased the tensile strength (60%) and impact strength (30%) as well as the flexural strength (40%). 2.1.2 ‘Green’ Nanoparticle- and Nanofibrous- Composites A particulate composite is defined as a material whose reinforcement may be categorized as particles [5]. Particles, alternatively called fillers, are nonfibrous and basically have no long dimensions with the exception of platelets. Particle fillers, such as ceramic, organic and inorganic particles, are widely used to improve the mechanical, thermal and electrical properties of matrix materials, and increase the abrasion resistance, surface hardness, etc. of the composites. Reducing cost is also, in many cases, the main reason to use the inexpensive particles. As for their mechanical 8 properties, particles in the composites are not generally expected to improve strength significantly. This is because the particles share the load to a much smaller extent than fibers in fiber reinforced composites. On the other hand, the particles can be effective in enhancing the stiffness (modulus) of the composites. As environmental concerns have been growing, green particulate composites, based on biodegradable resins, have been developed, in response, by using organic fillers such as potato starch grain and wood flour, or inorganic fillers such as calcium carbonate [37-39]. Even when inorganic particles, e.g. clay, are used, the composites can be considered ‘Green’ because these particles don’t damage the environment. In recent years, researchers have steadily reduced the particles to nanoscale size with excellent results. Nanoscale particles can be incorporated into biodegradable polymer to make green nanocomposites, a new class of materials. Such organicinorganic nanocomposites have attracted great interest since they exhibit unexpected hybrid properties synergistically derived from the two components [40]. Polylactide and starch based nanocomposites were fabricated by modifying them with layered silicate, organoclay, whose individual layers have thicknesses of the order of 1 nm and very high aspect ratios (10-1000) [41-43]. These nanocomposites exhibited remarkable improvements in the mechanical properties, heat deflection temperature (HDT) and reduced oxygen gas permeability compared with pure matrix resins. Another novel material, a green nanofibril reinforced composite, which has a high performance, is currently being researched. Yano and Nakahara [44] have successfully made starch based biocomposites using kraft pulp microfibrils with a nanometer unit web-like network. The microfibril reinforced composites had a similar stress-strain curve to that of magnesium alloy, and approximately three times higher flexural modulus (12.5 GPa) and strength (310 MPa) than chopped glass fiber reinforced unsaturated polyester composites. Moon et al. [45] fabricated multiwalled 9 carbon nanotube/poly(L-lactic acid) (PLLA) nanocomposites by applying ultrasonic energy in a PLLA solution. They reported that the incorporation of the nanotube as a conductive filler in a PLLA matrix yielded an improvement in Young’s modulus and the electromagnetic wave shielding effectiveness. 2.2 Soy Protein Many different plant proteins, such as soy protein, zein from corn (maize), wheat gluten, potato proteins, and pea proteins, have been utilized in many commercial products including binders in printing inks, coatings for grease-proof paper and plywood adhesive [17]. These proteins are derived from sustainable, yearly renewable resources, and don’t damage our environment. Significant research has been carried out to explore the feasibility of using plant based polymers as an alternative to petroleum based polymers. Research done by Gennadios et al. [46] showed that dry protein films made from corn zein, wheat gluten, and a mixture of wheat gluten/soy protein isolate (2.1:0.9) were very efficient oxygen barriers compared with some non-protein edible coatings and plastic packaging materials. Among these plant based polymers, soy protein, extracted from soy beans, is commercially available worldwide at low cost. In the United States, one of the major producers of soy beans in the world, soy beans are a $17.5 billion market, mainly used for animal feeds [47]. Soy protein films have received considerable attention for food packaging applications since the 1990’s because of their good film-forming ability. Gennadios et al. [48] characterized the tensile properties, water solubility, and water vapor permeability of edible soy protein films modified by heat-curing. They found out that properties of soy protein isolate films can be substantially modified by heat treatment, thus, helping tailor such films to specific packaging applications. 10 Three types of soy protein products, soy protein isolate (SPI), soy protein concentrate (SPC), and soy flour (SF), are commercially available in the market. About 2 billion bushels (60 lbs/bushel) of soybeans are generally processed each year in the U.S., which produces 91 billion pounds of soy flakes [49]. Soy flakes are defatted mechanically or chemically and are ground to produce soy flour ($0.15/lb) [50]. Soy protein concentrate ($0.70/lb) is obtained from soy flour by leaching out the water/alcohol-soluble sugars. Soy protein isolate ($0.90/lb) is soy protein further extracted from SPC by alkali and reprecipitated by acidification [49]. In addition to the three types of soy proteins, several grades for each product, which are designed to have different functions, can be obtained commercially. The choice from among three kinds of proteins depends on the desired solubility and viscosity in a broad range of food products, such as beverages, bakery goods, soups, and cereal pieces. Table 2.1 shows the composition of the protein products according to the bulletin produced by Archer Daniels Midland (ADM) Company, IL. In Table 2.1, it can be seen that the difference among the proteins is the percent contents of the constitutive proteins, carbohydrates, ash, etc. SPI contains the highest amount (about 90%) of proteins and has been shown to have the best mechanical properties when processed [33, 34, 51]. As a result, SPI has been chosen as a matrix for the fully green composites in this research. Table 2.1 Compositions of Commercial Soy Proteins Component Protein, % Fat (acid hydrolysis), % Ash, % Dietary Fiber, % Carbohydrates, % Moisture, % SPI 90 4-5 5-6 6 SPC 68-72 3-4 5-6 19-20 - 6-9 SF 46-53 3-19 16-18 26-30 8-9 11 2.2.1 Chemical Composition and Fractions of Soy Protein Proteins, polypeptide chains, twist and fold in three dimensions. In polypeptides, 2-amino acids form the monomer units joined by a peptide bond, the amide linkage. The linkage is produced by the condensation reaction of the carboxylic acid function of one amino acid with the amine group in another to make a chain of amides. Soybean protein contains the essential amino acids shown in Table 2.2 [52]. Soy protein consists mainly of the acidic amino acids (aspartic and glutamic acids) and their corresponding amides (asparagine and glutamine), non-polar amino acids (alanine, valine and leucine), basic amino acids (lysine and arginine), uncharged polar amino acid (glycine) and cysteine. Characterization of the soybean protein fractions by their sedimentation constants has been customary for more than 50 years. The four major fractions are termed 2 S, 7 S, 11 S and 15 S. (S stands for Svedberg units. The numerical coefficient is the characteristic sedimentation constant in water at 20ºC.) They were revealed in unltracentrifugal studies on the soluble proteins of soybeans: 22% of 2 S, 37% of 7 S, 31% of 11 S and 7% of 15 S [53]. The 7 S and 11 S are globulins and have complex quaternary structures [54]. The 2 S fraction is composed of low molecular weight polypeptides (in the range of 8,000 to 20,000 Da) and comprises the soybean trypsin inhibitors [55]. The 7 S, a glycoprotein, has a molecular weight of 180,000-210,000 Da and consists of nine subunits [56]. It is further featured in its low sulphur content with only four sulphur atoms in the form of two intramolecular disulphide bridges [54]. The 11 S fraction contains glycinin, the principal protein of soybeans, composed of three acidic and three basic subunits associated through hydrogen bonding and disulfide bonds [52]. The molecular weight of the 11 S is approximately 350,000 Da [57]. The 15 S protein is probably a dimer of glycinin [55]. 12 Table 2.2 Amino Acid Composition of Soybeans Amino acid Isoleucine Leucine Lysine Methionine Cysteine Phenylalanine Tyrosine Threonine Tryptophan Valine Arginine Histidine Alanine Aspartic acid Glutamic acid Glycine Proline Serine Composition (g/16 g nitrogen) 4.54 7.78 6.38 1.26 1.33 4.94 3.14 3.86 1.28 4.80 7.23 2.53 4.26 11.70 18.70 4.18 5.49 5.12 Side group CH(CH3)CH2CH3 CH2CH(CH3) 2 (CH2)4NH2 CH2CH2SCH3 CH2SH H2C H2C CH(CH3)OH H2C OH = N H CH(CH3)2 NH (CH2)3NHCNH2 H2C NH CH3 N CH2COOH CH2CH2COOH H COOH HN H CH2 CH2OH 13 2.2.2 Modification of Soy Protein Isolate Soy protein isolate (SPI) has been studied for uses as engineering applications and packaging products. Commercialization of soy protein films has been working to improve the mechanical properties, which are not yet good enough for such applications [58]. Soy protein has a highly hygroscopic nature because it contains many amino acids with polar groups such as amino, hydroxyl and carboxyl groups. During the curing process, soy protein chains can form several types of crosslinks such as the disulphide bonds between two cystein residues, lysinoalanine and lanthionine crosslinks by linking lysine with cystein and the amide type linkage formed by asparagine and lysine [59]. Many researchers have attempted to enhance the mechanical properties, toughness and water resistance, by chemical and physical modifications of soy proteins. These modifications have introduced new crosslinks into the protein structure in addition to naturally occurring covalent crosslinks, disulfide linkage between cysteine residues [60]. Crosslinking often results in important changes in the chemical, functional and nutritional properties of proteins. Some chemicals, such as glutaraldehydes, calcium salts, acetic anhydride and formaldehyde, have been used to enhance the tensile and puncture strength of SPI films by promoting covalent intermolecular and intramolecular cross-linking of proteins [33, 58, 61]. Park et al. [61] showed that treatment using calcium salts improved the moisture-barrier properties of SPI films because of the presence of calcium bridges between protein constituents in films. Rhim et al. [62] conducted thermal and UV radiation treatments of proteins to improve the mechanical properties. They also reported that the heatinduced cross-links contributed to improvement of hydrophobicity. As the present research showed, soy protein is a versatile material that, through various modifications, can be used in applications where quite demanding 14 performances are expected. In the present research, SPI was modified by blending with Phytagel® water-solution to improve the mechanical (tensile, flexural and impact) properties of kenaf mat composites using the modified SPI resins. Blending methods are becoming an increasingly important portion of the entire plastics industry, 20-40% of the total plastics market by various estimates [63]. This method has been applied to obtain homogeneous SPI suspensions in the present research. 2.2.3 Plasticization of Soy Protein Isolate Soy protein isolate (SPI) has notable potential to impart promising mechanical properties for industrial products by modifying them chemically and physically. Dried pure or modified SPI is, however, very brittle because of its low fracture strain, which creates handling and processing difficulties. Moreover, when SPI is used as a matrix for short fiber reinforced composites, it is highly possible that the resin will fracture before fibers will, resulting in low mechanical properties. These disadvantages have restricted applications of SPI based composites. To overcome the drawbacks of its brittleness, attempts have been made to produce SPI resins by using various plasticizers. Mo and Sun [64] used polyol-based plasticizers such as glycerol, propylene glycol, 1,2-butanediol and 1,3-butanediol and characterized the thermal and mechanical properties and morphology of a plasticized SPI polymer. They have reported that both glycerol and propylene glycol effectively depressed the glass transition temperature by increasing the free volume of the polymer system. In addition, it has found out that these plasticizers had good compatibility with soybean protein because of their small hydrophilic molecules, whereas plastics with 1,3-butanediol gave the highest tensile strength. Polyhydric alcohols such as ethylene glycol, glycerol, propylene glycol, 1,3-propanediol and polyethylene glycol were investigated to evaluate their plasticizing effects on SPI plastic by Wang et al. [65]. They concluded that glycerol is the most suitable 15 plasticizer for soy protein plastic when toxicity is taken into consideration. Glycerol is nontoxic, while propylene glycol has a very low order of toxicity but is more volatile, and ethylene glycol is classified as a hazardous chemical [66]. Glycerol was, therefore, chosen as the plasticizer to reduce the brittleness of the SPI resin and improve its toughness, in this study. 2.3 Gellan and Phytagel® Phytagel®, a commercial brand name of Sigma-Aldrich Biotechnology, is biopolymer gellan produced from bacterial fermentation using Sphingomonas paucimobilis (P8169-Product Information, Sigma-Aldrich Co., St. Louis, MO). It can substitute for agar as a gelling agent and has good thermal stability and high clarity. The properties which greatly facilitate the observation and photography of root growth and determination of the presence of microbes are advantages over agar gels, which are cloudy and opaque [67]. Many researchers have been evaluating the performance of this gellan gum in comparison with agar [68]. The gellan gum has increasingly been used as an agent for plant tissue culture media and microbiological media. Doner et al. [69] investigated the improved solution and gel properties of purified commercial gellan such as Phytagel® and Gel-Gro® (a trademark of ICN Biochemicals, Aurora, OH) to apply to culturing root tissue and to immobilizing viable enzymes and cells. In the absence of divalent cations, Phytagel® forms strong gels preferentially through hydrogen bonding. Lodha and Netravali [70] have showed that incorporating Phytagel® into SPI resin enhanced the tensile properties of the modified SPI resins as well as their thermal stability. Few studies have been done applying biopolymer gel like Phytagel® to engineering green composites. This new idea should yield a novel material. In this research, Phytagel® was used as a SPI resin modifier to improve the 16 tensile properties of SPI resins. In addition, the present research has characterized the effects of Phytagel® on the mechanical (tensile, flexural and impact) properties of kenaf mat reinforced green composites using SPI resins modified with Phytagel®. 2.3.1 Gellan Gum Gellan gum has been used in the food, pharmaceutical and chemical industries due to its novel property of forming thermo-reversible gels when heated and cooled. The gum could be a gelling agent in dental and personal care toiletries and the cosmetic industry [71]. This novel water-soluble polysaccharide has drawn attention since it was successfully produced as a new species of Pseudomonas, renamed as Sphingomonas paucimobilis, on a laboratory scale by Kang et al. [72]. Since the 1800s, other microbial polysaccharides, yielded by various bacteria, have been discovered; xanthan from Xanthomonas campestris, gelrite excreted by Aureomonas elodea, succinoglycan from Rhizobium, etc. [73]. Among these polysaccharides, gellan gum is a commercially important polymer because of its superior properties and wide range of textures obtained by blending it with other gels. It is the most expensive among the food gums because of its low productivity and requirements for difficult and costly downstream processing steps [71]. As the fermentation production is more developed, however, the material will likely become cheaper. CH2OH COOH [glucose] [glucuronic acid] [rhamnose] Figure 2.1 Chemical Structures of Glucose, Glucuronic Acid and Rhamnose 17 Gellan exopolysaccharides like gellan gum, welan gum and rhamsan gum contain linear polymer chains consisting of three kinds of sugars: glucose, glucuronic acid and rhamnose, shown in Figure 2.1. Though these polymers have a similar backbone structure, the differences are in the nature and location of the monosaccharide and disaccharide side chain grouping and in some of the polymers, the presence of L-mannose as an alternative to L-rhamnose in the backbone chain of the polymer [71, 74]. Gellan gum is composed of a linear sequence of tetrasaccharide repeating units of D-glucose (Glcp), D-glucuronic acid (Glcp A) and L-rhamnose (Rhap) in the molar ratios of 2:1:1 as shown in Figure 2.2 [75]. The gum in the solid state adopts a double helix structure with two left-handed, three-fold chains. The structure was confirmed in the presence of cations with scanning tunneling microscopy by Nakajima et al. [76]. - 3-β - D – Glcp – (1 4) -β - D – GlcpA – (1 4) -β - D – Glcp – (1 4) -α - L – Rhap (1- Figure 2.2 Gellan Gum of Linear Sequence of Tetrasaccharide Repeating Units 2.3.2 Phytagel® as a Modifier for Soy Protein Isolate Resin In this research, the water solution of Phytagel® was blended with water-based SPI resin to form an interpenetrating polymer network (IPN) like structure, a crosslinked complex [70]. Sperling [77] defined interpenetrating polymer networks (IPNs) as a combination of two or more polymer networks that are synthesized. IPNs can be held by several types of multicomponent polymer materials, such as a simple polymer blend without any bonding between chains, a graft copolymer and a block copolymer. In addition, there are several different IPNs like sequential IPN, latex IPN and Gradient IPN. Gursel et al. [78] prepared a IPN polymer by photopolymerization of hydroxyethylmethacrylate (HEMA) in the presence of polyhydroxybutyrate-cohydroxyvalerate (PHBV) and discussed the mechanical properties of the IPN polymer 18 compared with those of polyhydroxyethylmethacrylate (PHEMA). It was shown that the failure stress and modulus of the IPN polymer showed much stronger than those of PHEMA homopolymers. Phytagel® is a hydrogel associated with a highly swollen polymer network held together by cross-links or weaker cohesive forces such as hydrogen or ionic associations [71]. It forms strong and stable gels depending on the type and concentration of cations present and on the degree of acetylation. D-glucuronic acid in the repeat unit includes a carboxylic group which promotes strong gelation. In addition, the other sugars, which make up the repeat unit of gellan gum, have many hydroxyl groups, which are capable of reacting with amino and carboxyl groups in SPI polymer by hydrogen bonding [70]. The mixture in which the SPI polymer chains and Phytagel® polymer mingle well together can form a IPN like structure. 2.4 Natural Fibers as Polymer/Resin Reinforcement From an environmental point of view, interest has been growing in investigating the possibility of using natural (plant based) fibers as reinforcements for polymer composites. Various kinds of natural fibers such as abaca, bamboo, flax, hemp, jute, ramie, and sisal have been researched to replace wood fiber and glass fiber as the reinforcements. Pineapple and henequen fibers were used as reinforcements for environmental-friendly green composites based on poly(hydroxybutyrate-cohydroxyvalerate) resin by Luo and Netravali [79, 80]. Like these leaf fibers, natural fibers which are extracted from plant stalks, leaves, seeds, grasses and fruit can be expected to appear in commercial products in the near future. In Table 2.3, the density, mechanical properties, and moisture absorption of E-glass fiber are compared with those of some representative natural fibers [81]. 19 Table 2.3 Properties of Various Natural Fibers and E-galss Fibers Properties Density, d , g/cm3 Tensile strenght, MPa Modulus, E , GPa Specific, E/d Elongation at failure, % Moisture absorption, % E-glass 2.55 2400 73 29 3 - Hemp 1.48 550-900 70 47 1.6 8 Jute 1.46 400-800 10-30 7-21 1.8 12 Fibers Ramie Coir 1.5 1.25 500 220 44 6 29 5 2 15-25 12-17 10 Sisal 1.33 600-700 38 29 2-3 11 Flax 1.4 800-1500 60-80 26-46 1.2-1.6 7 Cotton 1.51 400 12 8 3-10 8-25 It can be seen from these data that the density of glass fiber is over 60% higher than that of natural plant based fibers, and the specific stiffness of glass fiber is comparable to that of some natural fibers. Furthermore, Figure 2.3 gives typical curves of the tensile specific stress versus strain of flax, abaca, cotton and henequen, compared with synthetic polymer fibers such as nylon and polyester [82]. In this figure, it appears that flax and abaca fibers have higher stiffness than nylons and polyesters, and the specific tensile stress of the natural fibers are approximately as high as that of the synthetic fibers such as nylons and polyesters. Figure 2.3 Tensile Stress-strain Curves of Several Fibers: 1, Nylon; 2, Polyester; 3, Flax; 4, Abaca; 5, Cotton; 6, Henequen 20 Similar to reinforcements used in fully green composites, natural fibers can also be incorporated into non-degradable thermoplastic or thermoset resins, which results in semi-green composites. Joseph et al. [83] determined the tensile properties of low density polyethylene composites reinforced with sisal fiber treated chemically by sodium hydroxide, isocyanate, permanganate and peroxide. It was found out that the composites using peroxide treated sisal fibers showed an enhancement in tensile properties due to the peroxide induced grafting. Yuan et al. [84] treated sisal fibers using argon and air-plasma to improve the poor adhesion between the hydrophilic sisal fiber and the hydrophobic polypropylene. The mechanical properties such as tensile strength and modulus, flexural strength and modulus, and the storage modulus of the composites including plasma-treated sisal fibers improved due to the increased surface roughness of the fiber and an increase in oxygen/carbon ratio of sisal fibers, both of which enhanced the fiber/resin interfacial strength. Reinforcing natural fibers, therefore, can be modified by physical and chemical treatments to enhance mechanical properties of semi-green composites. However, for green composites, surface treatments are not needed in most cases because the compatibility of natural fibers and protein and starch based hydrophilic biodegradable resins should be good. These fibers and resins, which contain polar groups, adhere well to each other because of the hydrogen bonds. A promising natural fiber, kenaf stalks, which are comparable in strength to glass, are replacing fiberglass in composite applications [85]. In addition, kenaf fibers are inexpensive compared with glass fibers; kenaf fibers cost ~20-25 US ¢/lb (~44-55 US ¢/kg) as against ~90 ¢/lb (~ US $ 2/kg) for E-glass fiber [86]. In the present study, needle punched kenaf fiber mats were selected as a reinforcement to fabricate fully green composites. As for the compatibility of kenaf fibers and modified SPI resins, 21 the interfacial shear strength of the fiber/matrix interface was measured using the single-filament composite technique [87]. 2.4.1 Kenaf Fiber Kenaf (Hibiscus cannabinus L.) is a member of the Malvaceae family, which includes cotton and okra. It is a herbaceous annual non-wood fiber plant whose origin was in east-central Africa. The crop, which closely resembles jute, has been considered as a potential substitute for jute in the manufacture of cordage products since it was introduced into the United States in the 1940s by the Department of Agriculture (USDA) [88]. Kenaf grows so fast that it can be harvested in 4 months, with high dry matter yield. Moreover, it can be grown under more wide-ranging conditions than jute [89]. The fiber, one of the least expensive bast fibers, is a lightweight, biodegradable, low density, renewable resource that can be obtained from a domestic (U.S.) crop. One more advantage which kenaf fiber offers is good soundabsorption characteristics because of its hollow nature [90]. As a result, during the past decades, many researchers have been working on many different aspects of the fiber: various (physical, mechanical etc.) properties, cultivating, producing and recycling process, and the potential markets. Kenaf stem consists of two distinct fibers, bast and core with a percentage makeup of about 35 and 65%, respectively [91]. Bast and core fibers possess different features: bast fibers are longer and have thicker cell walls and a smaller lumen than the core fibers [92]. The chemical compositions of the two fiber fractions are given in Table 2.4 [93]. Based on their properties, the two fibers have different applications: high purity bast fibers can be used in the manufacture of high quality paper products (writing paper, cigarette paper, tea bags), and high purity core can be used in the manufacture of sorbents, particle boards, laboratory animal bedding, and other high- 22 value products [91]. The development of technologies to separate these fibers is very important in the kenaf industry. Table 2.4 Chemical Compositions of Kenaf Bark and Core Fibers Component Lignin, % Holocellulose*, % Alpha-Cellulose, % Extractives, % Bark 11.8 81.1 51.0 2.8 Core 18.3 71.6 34.9 4.8 * Holocellulose includes cellulose and hemicellulose It has been reported that kenaf fibers which were grown in Athens, GA, measure 2.45 mm in length and 12 µm in diameter [94]. Shah et al. [89] stated that the fiber properties and dimensions can vary due to location, climate, variety and stages of maturity. In this research, the morphology of kenaf mats and fibers and the tensile properties of the fibers were characterized. The traditional use of kenaf is seen in the manufacture of sacking, cordage, ropes, fishing nets, etc. During the past decades, researchers have been seeking other potential markets for kenaf to take advantage from an environmental point of view. The success of a new crop venture needs market development. In 1992, Jobes and Dicks [95] stated that an immediate potential market would be newsprint, poultry litter and forage. In the pulp and paper industries, kenaf has been expected to serve as an alternate non-wood fiber because it can help in reducing the deforestation worldwide and have a favorable impact on the economies of many developing and developed countries [96]. The feasibility of producing medium-density particleboard from whole stalk kenaf was researched by Webber et al. [97]. Currently, commercial interest in kenaf fibers among natural non-wood fibers is growing in the fields of textiles and automobiles. 23 2.4.2 Nonwoven Kenaf Mats Kenaf is produced abundantly in the form of fibers, nonwoven mats and fabric in the United States. Among these forms, nonwoven kenaf mats are presently limited to use for commercial products though only a few have been successful, for example, grass mats and erosion mats [98]. This is because mechanically harvested (raw) kenaf fibers, which are coarse and brittle, are difficult to process through conventional textile and nonwoven equipment [99]. This is a common issue with natural bast fibers. Baldwin et al. [100] demonstrated that nonwoven 100% kenaf bast fiber mats could be used as a growth medium for the establishment of some warm-season and cool-season grass species. Ramaswamy et al. [101] have attempted to develop nonwoven kenaf substrates that can lead to high-value and high-volume end-products: furniture, kitchen cabinets, fixtures, wall-coverings, displays, and various other products. To make kenaf mats as successful alternative materials, the processing has to be developed using conventional equipment in nonwoven industries. Successful development of useful and novel nonwovens that include a high percentage of kenaf could result in the increased utilization of kenaf fiber, thereby enhancing markets for U.S. farmers [102]. The nonwoven techniques aim to produce mats or fabrics more economically than the traditional means, namely, weaving and knitting. The study by de Guzman et al. [103] was undertaken to explore a new material for novel textile products from abaca, kenaf and pineapple fibers using nonwoven techniques, specifically, needlepunching and adhesive-bonding processes. One more advantageous aspect for nonwoven mats is that they can be tailored by blending with another fiber to achieve the desired mat properties. Tao et al. [104] prepared nonwoven mats containing 100% kenaf or kenaf/cotton blends. They showed that adding cotton fibers into kenaf mat increases mat strength and oil retention capacity, indicating that the blended mats have 24 potential application in the prevention of soil erosion, weed control, and cleanup of waste liquids. 2.4.3 Kenaf Fiber (Mat) Composites Many attempts have been made to develop thermoplastic resin based kenaf fiber composites with various properties suitable for many different applications for the last decade. Highly filled 85 percent kenaf fiber composites based on polypropylene were studied and compared with commercially available formaldehydebased wood composites by Sanadi et al. [105]. Their study showed that the flexural properties of the kenaf composites exceed those of the wood composites. It concluded that highly filled agro-based fiber thermoplastic composites, which could be used for furniture, automotive and building applications, are an attractive material for countries that hope to develop crops for new uses to save precious forest resources. Maldas et al. [106] determined the mechanical properties and dimensional stability of kenaf fiber filled recycled polyethylene composites at room temperature and after immersion in boiling water. In their study, the performance of the composites was improved by adding coupling agents. Thermoplastic resin composites using mats can be mass-produced by thermocompression molding as effectively and inexpensively as by injection molding, which is applied to make short fiber reinforced composites. Magurno [107] discussed many thermoplastic resin based composites using vegetable fibers including kenaf or jute fabricated by the compression molding process for automotive interior components. Successful development of kenaf mat composites in the automotive industry would be environmentally benign and could generate a noticeable expansion in kenaf cultivation. In contrast with thermoplastic resin, biodegradable resins have been scarcely employed to fabricate fully environmentally friendly green composites using kenaf 25 fiber or nonwoven kenaf mats. Nishino et al. [108] made kenaf reinforced biodegradable composites using poly-L-lactic acid (PLLA). They have reported that the mechanical and thermal properties of kenaf/PLLA composites were competitive with other high performance biodegradable polymer composites. However, as it has been pointed out, PLLA is expensive, and thus its cost performance is not attractive. In the present research, kenaf fiber mats and SPI resin, whose costs are low, were used to fabricate totally green composites. 2.5 Hybrid ‘Green’ Composites Hybrid composites can be defined as materials which contain the reinforcement of more than two distinct fibers or fiber forms in a single matrix. Chou [109] categorized hybrid composites into three types depending upon the arrangements of fibers and pre-preg layers, as shown in Figure 2.4. In Figure 2.4, open circles and filled circles show two different types of reinforcements, and the space between them is the matrix. The first type is described as intermingled or intraply (Fig. 2.4a), in which the different fiber materials are intimately mixed together and infiltrated with a matrix simultaneously. The second type is interlaminated or interply composites (Fig. 2.4b), which are made by bonding together separate laminae each containing just one type of fiber in a matrix. The third category is termed as interwoven (Fig. 2.4c), composed of fabric reinforcements where each fabric contains more than one type of fiber. The composites are very attractive to the material engineering community for the following reasons. They can be designed to achieve better or more balanced properties than normal composites with two phases, one type of fiber and resin. In addition, the cost can be reduced by adding less expensive fibers such as glass and aramid. 26 (a) (b) Fabric reinforcements (c) Figure 2.4 Types of Hybrid Composites: (a) intermingled; (b) interlaminated; (c) interwoven 27 A hybrid green composite is the latest research topic in the green composite field. Although comingled composite materials containing a natural fiber and synthetic fiber based on thermoplastics or thermoset plastics have been researched for some decades, hybrid composites fabricated with fully biodegradable materials have not been studied widely. Tajvidi [110] discussed the static and dynamic mechanical properties of a natural fiber hybrid composite containing kenaf fibers and wood flour as the reinforcements and polypropylene as the polymer matrix. Many researchers have hybridized semi-green composites and glass fiber to evaluate the effect of glass fibers on their mechanical properties [111-114]. In the present research, hybrid green composites were developed by using fibrillated bamboo fibers as the second resin reinforcing phase to fabricate interlaminated hybrid composites. This research investigated the feasibility of fibrillated bamboo fibers to improve the mechanical (tensile, flexural and impact) properties of kenaf mat based composites using modified SPI/ Phytagel® resins. 2.6 Bamboo Fiber Today, environmental problems have been increasingly drawing our notice. Many engineers have worked on natural resources which could replace synthetic materials to prevent depletion of petroleum. Without any preventive measures, more manufacture and disposal of synthetic materials will further aggravate the already existing environmental problems. Bamboo, one of the most promising resources, can be a excellent substitution of synthetic materials because it is not only abundantly available in the world but also a sustainable and natural resource. 2.6.1 Introduction of Bamboo Bamboo, belonging to the grass family, takes only 3-4 months to grow to its full height and maturity. It has high rigidity and tensile strength, which are derived 28 from its longitudinally aligned fibers, as well as its low density [115]. Average mechanical and physical properties of natural bamboo are shown in Table 2.5 [116]. Furthermore, the fracture toughness of bamboo (56.8 MPa·m1/2) has been reported to be higher than that of Al-alloy (33 MPa·m1/2) [117]. Table 2.5 Various Properties of Natural Bamboo Fiber Volume Fraction, % 29.2 Density, g/cm3 0.66 Tensile Strength, MPa 206.2 Tensile Modulus, GPa 20.1 Flexural Strength, MPa 210.3 Flexural Compressive Modulus, Strength, GPa MPa 13.1 78.7 Because of the excellent properties, bamboo has been traditionally used to make living facilities, scaffolds, flooring, tools, and adornments. It is, however, very difficult to process bamboo into many kinds of shapes. As a result, the demand for the industrial applications is so low that farmers have been losing interest in handling or harvesting bamboo groves (forests) with care, and the number of them has also been decreasing. This has led most bamboo groves to expand wildly, invading adjacent forests, in Japan, whereas some bamboo groves have disappeared due to the construction of houses, roads and golf courses [118]. To solve the problem of expanding groves is to broaden the range of the practical uses of bamboo. If the demand increases, the groves will be well managed and will prosper. In order to achieve this, refinements of the processes through which bamboo products are produced from raw bamboo are essential. When bamboo can be utilized industrially, the demand for it will grow, saving both the bamboo groves and the environment. 2.6.2 Fibrillated Micro Bamboo Fibers Bamboo or its fibers can be used to make floors and walls of houses, kimonos and towels. Recently, engineers have devoted much attention and efforts to research 29 on the bamboo fiber reinforced polymer composites into the industrial applications [119, 120]. A steam explosion method was applied to extract bamboo bundles from raw bamboo [121, 122]. In the method, raw bamboo is first allowed to stay in a chamber at high temperature (around 170°C) and pressure (0.8 MPa) for some time. Afterward, the bamboo is given impact force by opening up the valve. By repeating these two processes several times, bamboo fiber bundles can be extracted easily. A steam explosion method is very effective and useful for mass-production of bamboo fiber bundles. Incorporating the refined bamboo fibers with small diameters into composites is expected to generate high mechanical properties. This is because the interface area between the fibers and the matrix increases significantly, and lignin, which is a hydrophobic substance contained in cellulose fibers, is removed during the refinement. Fibrillated bamboo fibers have been extracted successfully by milling steam-exploded raw bamboo [123, 124]. They were almost separated into single fibers. Research done by Takahashi et al. [125] showed, by principle component analysis, that lignin in the steam-exploded bamboo fibers can be removed effectively using an alkali treatment. The treated fibers improved the bending strength of the composites compared with that of untreated bamboo fiber composites. Preliminary studies in our research group have shown the tensile properties of soy protein concentrate resin improve significantly by reinforcing it with a small fraction of micro fibrillated bamboo fibers. 30 CHAPTER 3 EXPERIMENTAL PROCEDURES 3.1 Materials Soy protein isolate (SPI) powder, PRO-FAM® 974, was obtained from Archer Daniels Midland Co., Decatur, IL. SPI has been reported to contain 90% protein, 4% fat (acid hydolysis) and 5% ash. Analytical grade glycerol and sodium hydroxide (NaOH) were purchased from Fisher Scientific, Pittsburgh, PA, and were used without further treatment. Phytagel®, used as a modifier, was purchased from Sigma-Aldrich Co., St Louis, MO. A roll of carded and needle-punched nonwoven kenaf mat was supplied by Flexform Technology Co., Elkhart, IN. Fibrillated bamboo fiber (FBF) in the form of water based slurry was provided by Doshisha University, Kyoto, Japan. 3.2 Processing and Modification of Soy Protein Isolate (SPI) Resin 3.2.1 Preparation of a SPI Resin Sheet SPI resin sheet was prepared using a casting method as described below [51]. Before the sheet was made, a Teflon®-coated glass plate was prepared. A square piece of Teflon® sheet was cut off the desirable size. The square sheet was attached to a glass plate, and 2 cm on all four sides were folded vertically to form a box. Such Teflon®-coated glass plate was put aside. Analytical grade sodium hydroxide pellets were dissolved into distilled water to prepare a 1M NaOH solution. First, the desired amounts of SPI powder and glycerol were weighed separately and mixed with 12 times (by wt. of SPI powder) distilled water in a beaker. The amount of glycerol was based on the weight of the SPI powder. The mixture was stirred in a water bath at 75ºC for 40 min to obtain a homogeneous SPI suspension. After 20 min of stirring, the desired amount of NaOH solution was added to obtain required pH values. The amounts of the NaOH solution were measured to adjust the 31 desired pH values (7, 9 and 11) of a pure SPI suspension by monitoring with an electronic pH-meter (model-145, Corning Inc., Corning, NY) at room temperature beforehand. This stir-heated process is called pre-curing. The pre-cured SPI suspension was used to fabricate kenaf mat composites. The pre-cured suspension was poured onto the Teflon®-coated glass plate to form a sheet. The glass plate with the pre-cured SPI resin was kept in an air circulating oven at 50ºC and dried for 24 hrs. The dried SPI sheet was peeled off the plate. The sheet, sandwiched with aluminum plates, was cured by hot-pressing at 120ºC and 2.1 MPa for 20 min. The hot-pressing was carried out on a Carver Hydraulic hot press (model 3891-4PROA00, Carver INC., Wabash, IN). 3.2.2 Modification of SPI Resin SPI resin was modified with Phytagel® to improve the mechanical properties of the resin. During the dispersing process of SPI powder, Phytagel® was dissolved into a separate beaker with 30 times (by wt. of Phytagel® powder) distilled water in a water bath at about 90ºC. After being dissolved completely, the Phytagel® solution was mixed with the SPI suspension, which was already stirred for 40 min. The SPI/Phytagel® mixture was stirred in a water bath at 75ºC for 20 min (pre-curing). The pre-cured SPI-Phytagel® resin was used for composite fabrication. The same process for drying and curing, as described in section 3.1.1, was subsequently followed to obtain a cured, modified SPI/Phytagel® resin sheet. 3.3 Composite Fabrication 3.3.1 Nonwoven Kenaf Fiber Mat Composites Green composites based on nonwoven kenaf fiber mats were fabricated using SPI resin modified with Phytagel® by compression molding. Kenaf fiber mat composites with pure SPI resin were prepared for comparison with the composites 32 with modified SPI/Phytagel® resins. To prepare the composite specimens, pieces of kenaf mat of desired rectangular size were cut from the roll. The pieces were each divided by hand into two layers of half the thickness. The pre-cured resin was poured onto each layer separately. It was ensured that these layers were completely impregnated with the resin. The impregnated mats were then dried at 50ºC for over 24 hrs in an air circulating oven. A small amount of the same resin was pasted on one side of the top and bottom layers and both sides of the middle layer by hand. After being dried in an oven at 50ºC for approximately 30 min, the three layers were hot-pressed together (cured) under the following sequential conditions. 1. 80ºC for 10 min at 0.27 MPa 2. 100ºC for 10 min at 0.27 MPa 3. 100ºC for 10 min at 3.5 MPa 4. 120ºC for 10 min at 7.0 MPa The hot-press platters opened for a moment to release water vapor between the first and the second step, and between the second and the third step. After curing, a laminate structure of kenaf fiber mat composites was obtained. The ratios of resin and fiber in the composites were calculated by weighing the dry half kenaf fiber mat before the process and the dry composites afterwards. 3.3.2 Hybrid Composites with Fibrillated Bamboo Fiber (FBF) Sheets Fibrillated bamboo fiber (FBF) sheets were incorporated into nonwoven kenaf mat composites using SPI resins modified with 20% Phytagel®, resulting in engineered hybrid interlaminated green composites. The proportion of kenaf fibers, FBF and modified SPI resin in hybrid green composites was 4:1:5 by weight. Kenaf fiber mat halves were obtained by separating the whole mat (through the thickness) which was cut into the desired size from the roll as described previously. 33 The desired amount of water based slurry of FBF was dispersed into at least 5 times (by wt. of FBF slurry) distilled water in a beaker by stirring with a magnet bar. FBF sheets were obtained by using equipment containing a filtering flask and a vacuum pump as shown in Figure 3.1. Vacuum pump Steel plate with holes for vacuum suction Figure 3.1 Equipment to Make a FBF Sheet To obtain the FBF sheet, the bamboo fiber dispersion was poured onto a sheet of filter paper set up on a filter paper set up on the steel plate with the holes. The water was removed by pulling into a filtering flask using a vacuum pump. The damp bamboo fiber sheet with the filter paper was dried at 70ºC in an air circulating oven for 48 hrs. A FBF sheet with the thickness of approximately 0.15 mm was obtained by peeling it off the filter paper. To fabricate the hybrid composites, 5 pieces (3 kenaf fiber half mats and 2 FBF sheets) were used. Pre-cured modified SPI/Phytagel® resin was impregnated into all these kenaf fiber half mats. The mats with the resin were oven-dried at 50ºC for 24 hrs. A small amount of the same resin was pasted on one side of the top and bottom 34 layers and both sides of the middle layer by hand. Similarly, a small amount of the resin was pasted both sides of the two FBF sheets. The five layers were dried in the oven at 50ºC for approximately 30 min. Each of the bamboo fiber sheets were sandwiched between the layers of the kenaf mats with the resin. The same curing process described in section 3.3.1 was followed to obtain hybrid interlaminated green composites based on kenaf mats and FBF sheets. Figure 3.2 exhibit the schematic of laminate structures of kenaf mat composites and the hybrid composites using FBF sheets. A kenaf mat composite layer A FBF sheet (Kenaf mat composite) Hybrid kenaf mat composite with FBF sheets Figure 3.2 Schematic of Laminate Structures of Kenaf Mat Composites and the Hybrid Composites Using FBF Sheets 3.4 Characterization Techniques 3.4.1 Tensile Testing of Resin Sheets The tensile properties of cured SPI resin sheets and modified SPI/Phytagel® resin sheets were determined according to ASTM D 882-02. Resin strips of 10 mm width and 126 mm length were cut from sheets that were conditioned at 21°C and 65% relative humidity (RH) for 72 hrs before testing. The tensile tests were conducted using a universal Instron testing machine, (model-5566, Instron Co., Canton, MA) at a strain rate of 1 min-1 and a gauge length of 50 mm. Tensile fracture strain was computed based on the gauge length and the distance the crosshead traveled 35 until a specimen fractured. A minimum of 7 specimens were tested for each batch to obtain an average value. 3.4.2 Measurement of Moisture Content The moisture content of SPI resin was measured by using a moisture/volatiles tester (model-SAS, C. W. Brabender Instruments, Inc., South Hackensack, NJ), as per ASTM D 1576-90. The temperature of the tester during the measurement was 105°C. The 5g samples cut from the SPI resin sheets were kept in the tester for 24 hrs after being conditioned for 72 hrs at the ASTM D 1776-98 conditions of 21°C and 65% RH. The moisture content, based on this gravimetric test, was measured directly from the instrument. 3.4.3 Tensile Testing of Single Fibers Tensile characterization of kenaf fibers extracted from the mat was performed on a universal Instron testing machine, (model-5566, Instron Co., Canton, MA), according to ASTM D 3822-01. The gauge length of 10 mm and a strain rate of 0.1 min-1 was chosen. The corresponding crosshead speed was 1 mm/min. Figure 3.3 shows the schematic of the kenaf fiber tensile test specimen. The fiber was mounted on a paper tab cut into a rectangular shape shown in Figure 3.3. The fiber was then glued at two points at a distance equal to the gauge length. The diameter of the fibers was measured at three different locations by an optical microscope, (model-Continuum, Thermo Nicolet Co., Madison, WI). All of the specimens were conditioned for 24 hrs in a room, controlled at the ASTM D 1776-98 conditions of 21°C and 65% RH, prior to testing. At least 65 fibers selected randomly per each batch were tested to obtain the distribution. 36 Kenaf fiber Gauge length Glue Paper Figure 3.3 Schematic of a Tensile Specimen of Kenaf Fiber 3.4.4 Single Fiber Composite (SFC) Technique To determine the effect of Phytagel® content on the interfacial properties between a kenaf fiber and modified SPI resins, the single fiber composite (SFC) technique was performed. The SFC technique has been studied in depth by Netravali et al. [126, 127]. Herrera-franco and Drzal [128] reviewed the SFC technique along with several other techniques for fiber/resin interface characterization. The SFC technique is very useful in not only measuring the fiber/resin interfacial shear strength (IFSS) but also providing information about the failure mode at the fiber/resin interface within the specimen. Figure 3.4 presents the schematic representation of the single fiber fragmentation process [128]. When the applied strain increases, the embedded fiber in a SFC specimen breaks repeatedly at points where the fiber strength, σ f , has been reached. Further strain applied to the specimen results in repetition of this fragmentation process until the remaining fiber lengths become so short that shear stress transfer along their lengths can no longer build up enough tensile stresses to cause any further failures. This maximum final fragmentation length of the fiber is referred to as the critical length, lc . As shown in Figure 3.4, the shorter critical fragment length has the higher tensile stress of the according fragment, which produces the higher IFSS value. 37 Fiber (SFC) Figure 3.4 Schematic Representation of the Single Fiber Fragmentation Process To fabricate a dogbone shaped SFC specimen, a single kenaf fiber was randomly extracted from the kenaf mat. Two long polyethylene filaments were attached to each end of the single kenaf fiber using Krazy® glue to extend its length. After drying the glue, the kenaf fiber along with the polyethylene filaments was mounted in a silicone rubber mold. The pre-cured SPI or modified SPI/Phytagel® resin was slowly poured into the mold. The mold was then placed in an oven and left at 50°C for 24 hrs. For curing, the dogbone shaped SFC specimen taken out from the mold was hot-pressed at 120°C and very low pressure for 5 min. The dimensions of the specimen are shown in Figure 3.5. The diameters of the kenaf fiber in the composite were measured at 5 different locations by an optical microscope (modelContinuum, Thermo Nicolet Co., Madison, WI), and the average value was calculated. 38 Glue points Gauge length 30 mm 25 mm 60 mm Polyethylene filament (thickness = about 0.4 mm) 107 mm Single kenaf fiber Figure 3.5 Dimensions of a Single Fiber Composite Specimen To carry out the SFC test, the specimen was strained at a strain rate of 0.001 min-1 on the universal Instron testing machine (model-5566, Instron Co., Canton, MA) until it broke. The length of the kenaf fiber fragments in the strained specimen was measured under a microscope (model-BX 51: Polarizing Microscope, Olympus America Inc., Melville, NY). Some SFC specimens were removed from the Instron prior to breaking to observe the fragmentation process and to confirm that fragmentation was completed at the time the specimen broke. About 5 specimens per each batch were tested to obtain the average IFSS value. The IFSS was calculated using the following equation which was employed to analyze the microfracture of a fibrous composite by Kelly and Tyson [129, 130]. τ = dσ fl 2lc (3.1) where τ is the average shear stress at the fiber/matrix interface, d is the fiber diameter, σ fl is the fiber failure stress at the average critical length, lc . This equation assumes that the shear stress is uniform along the entire length of the fiber. To obtain the value of σ fl , single kenaf fibers were tensile tested with various gauge lengths (5, 10 and 15 mm) as described in section 3.4.3. The test results produced the relationship between the fracture stress of fibers and the gauge length. The average fracture stress values were computed from their Weibull distribution. From the linear fit function in a log- log plot, the values of σ fl were calculated. 39 3.4.5 Microdrop Technique The microdrop technique has been shown to be useful as well as the SFC technique to measure fiber-matrix interfacial shear strength (IFSS) [131-133]. The compatibility of fiber and matrix is an key factor to determine the mechanical properties of composites. The schematic of this test is shown in Figure 3.6. In this study, E-glass fiber was used. An universal Instron testing machine, (model-5566, Instron Co., Canton, MA) was used to conduct the tests at a crosshead speed of 5 mm/min. Shear stress between the fiber and the resin microdrop was generated by setting a microvice just above the bead and pulling up the fiber. Specimens were prepared by attaching a bead of pre-cured pure SPI resin or modified SPI/Phytagel® resins onto a fiber which had been already mounted on a paper tab as shown in Figure 3.3. These specimens were cured for 5 min at 120°C in an air circulating oven before testing. The IFSS value was computed from the following equation (3.2): IFSS = π F ×d × l (3.2) where F is the force required to debond the microdrop, d is the fiber diameter, and l is the embedded length. At least 20 specimens were tested for each batch to obtain the average number. The diameter of the fibers and the embedded length were measured by an optical microscope, (model-Continuum, Thermo Nicolet Co., Madison, WI) after curing the specimens. Load, F Fiber Microvise Microdrop Figure 3.6 Schematic of the Microdrop Technique 40 3.4.6 Tensile Testing of Composites The tensile testing of kenaf mat composites and the hybrid composites with fibrillated bamboo fibers, based on pure SPI resin or modified SPI/Phytagel® resins, was performed using the Instron testing machine (model-5566, Instron Co., Canton, MA). The testing was conducted as per ASTM D 3039. The green composite specimens were machined into a rectangular shape, as shown in Figure 3.7. Rectangular wooden tabs were attached to the ends of each specimen using epoxy resin to improve grip during the testing. The specimens with the tabs were tested at the crosshead speed of 2 mm/min with a gauge length of 50 mm resulting in a strain rate of 0.04 min-1. At least 7 specimens for each batch were conditioned at 21°C and 65% RH for 72 hrs and tested to obtain average values. Wooden tabs 25.4 mm 50 mm 126.2 mm (thickness = 3.9 ~ 5.2 mm) Figure 3.7 Dimensions of a Composite Specimen Used for Tensile Testing 3.4.7 Bending (Flexural) Testing of Composites The flexural properties, i.e., flexural stress and chord modulus, of kenaf fiber mat composites and hybrid composites with fibrillated bamboo fibers, based on pure SPI resin or modified SPI/Phytagel® resins, were characterized using the three point bending test as per ASTM D 790-02. A universal Instron testing machine (model5566, Instron Co., Canton, MA) was utilized to carry out the bending tests. The specimens were machined into a rectangular shape of 12.7 mm in width, and the length was calculated based on the thickness of the specimen. The specimens should be long enough to allow for an overhang on each end of at least 10% of the support 41 span. The span length of 16 times thickness of the specimens was set up as per the ASTM procedure. The thickness of the specimens was between 3.9 and 5.2 mm. The specimens were tested at a crosshead speed of 1 mm/min. Chord modulus was determined based on two stress points of 0.25% and 0.75% of the flexural strain. At least 7 specimens were tested to obtain average values. All specimens were conditioned at 21°C and 65% RH for 72 hrs before testing. 3.4.8 Izod Impact Testing of Composites Izod impact tests were conducted to measure the impact strength of kenaf fiber mat composites and the hybrid composites with fibrillated bamboo fibers, based on pure SPI resin or modified SPI/Phytagel® resins. The testing was performed as per ASTM D 256-02. The specimens were machined into the dimensions shown in Figure 3.8 and notched using a cutter according to the ASTM standard. All of the specimens were impacted with 21.7 J of the pendulum capacity using a cantilever beam impact tester (model-POE2000TM, GRC International Inc., Santa Barbara, CA). The tester was adjusted to obtain 3.46 m/s of the impact velocity. At least 7 specimens for each batch were tested to obtain average values after conditioning at 21°C and 65% RH for 72 hrs. 45º 12.7 mm 10 mm 50 mm (thickness = 3.9 ~ 5.2 mm) Figure 3.8 Dimensions of a Composite Specimen Used for Impact Testing 3.4.9 Scanning Electron Microscopy (SEM) The morphology of nonwoven kenaf fiber mats and kenaf fibers was studied using SEM (440, Leica Cambridge, Ltd., Cambridge, UK). The microstructures of pure SPI and modified SPI/Phytagel® resin specimens fractured in tensile testing and 42 kenaf fiber mat composite specimens fractured in impact testing were characterized using the SEM to evaluate their fracture behavior. The fractured specimens were mounted on the Cambridge® standard aluminum specimen mounts (pin type) with double-sided electrically conductive adhesive carbon tape (SPI Supplies, West Chester, PA). To coat the specimens with 60% gold and 40% palladium, a sputter coater (Desk II , Denton Vacuum, Moorestown, NJ) was utilized for 30 seconds at a current of 45 mA. The SEM was operated at a low accelerating voltage of 5 kV to avoid charging. 3.5 Data Analysis 3.5.1 Weibull Analysis of Kenaf Fiber Strength and Modulus In order to obtain mean fracture strength and Young’s modulus of kenaf fibers, a statistical analysis using the following classical two-parameter Weibull distribution was carried out: F (σ ) = 1− exp[−(σ /σ 0 )ρ ] σ ≥ 0 (3.3) where σ is the fracture stress, σ 0 is the scale parameter and ρ is the shape parameter. The scale parameter is the 62nd percentile of the strength or modulus value, and the shape parameter indicates the variability in strength or modulus. This analysis was based on the following assumptions (Weakest link rule). • Fibers fail in the same manner across different stress levels. • Fibers have flaws like cracks, notches or other defects. • These flaws are randomly distributed. • Magnitudes of these flaws are random. • Fibers break at their weakest point. The scale and the shape parameters were determined from the Weibull plot of the fracture stress values of individual fibers. The mean, standard deviation (SD), 43 coefficient of variance (CV) and the correlation coefficient were calculated using equations (3.4), (3.5), (3.6) and (3.7), respectively: µ = σ 0Γ(1+1/ ρ) SD 2 = σ 2 0 [Γ (1 + 2/ ρ)− (Γ(1 + 1/ ρ )) 2 ] CV = µ SD N ∑ (xi − x)( yi − y) R = i=1 NN ∑ ∑(xi − x)2 ⋅ ( yi − y)2 i=1 i=1 (3.4) (3.5) (3.6) (3.7) where µ and R are the mean and the correlation coefficient of the Weibull probability distribution function, Γ is the gamma function, and yi and xi are as follows: yi = ln{− ln[1− F (σ i )]} (3.8) xi = ln(σ i ) (3.9) 3.5.2 Statistical Analysis Using Tukey’s W Procedure To make multiple comparisons among a set of k population means in an analysis of variance, Tukey’s W procedure was conducted on a completely randomized experimental design. The procedure, using the Studentized range distribution, has been developed to avoid a high probability of declaring at least one pair of means significantly different when running multiple comparisons [134]. Before using this procedure, it was confirmed that the F test for treatments showed a significant difference. The error rate is controlled on an experimentwise basis at a level approximately equal to the α-level for the F test. For the tensile properties of the SPI resins, a total of 7 specimens were tested for each treatment which is pH value of 7, 9, 11. A total of 7 specimens of the composites containing modified SPI/Phytagel® resins were tested for each treatment with 0, 10, 20 and 40% Phytagel® to determine their mechanical properties. For each treatment, the statistical results of the mechanical, e.g. tensile, flexural and impact, 44 properties are shown in Figures 4.1, 4.3, 4.16, 4.17 and 4.18, respectively. In these figures, it should be noted that the plots indicate the means, and the plots with same letters and same subscript numbers, e.g. a0, b0, c0, a10 and b10, etc. do not show a statistically significant difference at α=0.05. The subscript indicates the percentages of glycerol or Phytagel® content by weight of SPI powder. 45 CHAPTER 4 RESULTS AND DISCUSSION 4.1 Preliminary Modification of Soy Protein Isolate (SPI) Resin Glycerol and sodium hydroxide were incorporated into the SPI resin to obtain better properties. As discussed in section 2.2.3, glycerol was used as a plasticizer. Sodium hydroxide was employed to adjust the pH value of SPI resin. The degree of denaturation of SPI, which has globular structures, depends on pH values. Soy protein has an isoelectric point at about pH 4.5 to 5. At that point, the soy protein is insoluble in water. However, at pH values away from the isoelectric point, the protein molecules open up and become more soluble. The tensile properties and the moisture content of SPI resin are affected by its structural changes caused by denaturation. The effects of pH value of the SPI solution and glycerol content on the properties and the moisture content were, therefore, evaluated before modifying SPI resin with Phytagel®. In addition, to optimize the pH value for SPI resin modification using Phytagel®, pure SPI resin was compared with SPI resin modified with 20% Phytagel® (by wt. of the SPI powder) at various pH values in terms of tensile strength and Young’s modulus. 4.1.1 Effect of pH Value and Glycerol on Tensile Properties and Moisture Absorption of Cured SPI Resin The effect of pH value and glycerol content on the tensile properties and the moisture content of SPI resins are shown in Table 4.1 and Figure 4.1. It is clear from Figures 4.1 (a)-(c) that, for every pH condition, the fracture stress and Young’s modulus of the cured SPI resin decreased, and the fracture strain and moisture content increased with increasing the content of glycerol. This confirms that glycerol acts as a plasticizer for SPI resin, as discussed by other researchers [32, 33]. 46 Table 4.1 Effect of pH Value and Glycerol Content on the Tensile Properties and Moisture Content of SPI Resin (a) 10% Glycerol Content Fracture Stress, Fracture Strain, Young's Modulus, pH Value MPa % MPa 7 25.0 (3.1)* 12.2 (7.1) 1150 (97) 9 22.6 (2.1) 26.9 (10.9) 1091 (82) 11 19.1 (0.9) 75.7 (12.5) 844 (46) * Numbers in parentheses represent standard deviation. Moisture Content, % 10.8 12.0 12.2 (b) 20% Glycerol Content Fracture Stress, Fracture Strain, Young's Modulus, Moisture Content, pH Value MPa % MPa % 7 9.9 (0.4)* 84.2 (15.9) 353 (21) 9 12.0 (0.3) 157.0 (12.8) 348 (28) 11 11.9 (1.0) 176.6 (25.3) 319 (21) * Numbers in parentheses represent standard deviation. 13.8 13.8 13.8 (c) 30% Glycerol Content Fracture Stress, Fracture Strain, Young's Modulus, pH Value MPa % MPa 7 7.5 (1.0)* 152.0 (43.4) 183 (18) 9 7.5 (0.6) 175.4 (23.1) 175 (16) 11 8.7 (0.5) 223.8 (11.9) 157 (8) * Numbers in parentheses represent standard deviation. Moisture Content, % 16.7 17.7 17.2 (d) 40% Glycerol Content Fracture Stress, Fracture Strain, Young's Modulus, pH Value MPa % MPa 7 5.0 (0.6)* 156.9 (32.0) 86 (7) 9 5.7 (0.7) 189.6 (28.9) 94 (8) 11 6.0 (0.5) 233.4 (17.4) 78 (5) * Numbers in parentheses represent standard deviation. Moisture Content, % 21.0 20.9 20.3 47 Figure 4.1 Effect of pH Values and Glycerol on the Tensile Properties and the Moisture Contents of SPI Resins: (a) Fracture Stress, (b) Fracture Strain, (c) Young’s Modulus and (d) Moisture Content Note: Any two means with the same letter (a-c) by the plots don’t show statistically significant difference at α = .05 according to Tukey’s W procedure. The subscripts of the letters (a-c) indicate the percentage of glycerol content (10-40%) in the SPI resin. 48 Fracture Stress, MPa 30 25 20 15 10 5 0 6 300 250 200 150 100 50 0 6 a10 a10 b10 b20 b20 a20 a30 a40 a30 ab40 b30 b40 7 8 9 10 11 12 pH value (a) b40 b30 a40 a40 a30 b20 a30 b20 a20 c10 a10 b10 7 8 9 10 11 12 pH value (b) Fracture Strain, % 49 Young's Modulus, MPa Figure 4.1 (Continued) 1400 1200 1000 800 600 400 200 0 6 a10 a10 b10 a20 a30 ab40 ab20 ab30 b40 b20 b30 a40 7 8 9 10 11 12 pH value (c) 25 20 15 10 : 40% glycerol : 30% glycerol 5 : 20% glycerol : 10% glycerol 0 6 7 8 9 10 11 12 pH value (d) Moisture Content, % 50 Three major theories have been proposed to account for the main effects of plasticizers by Sears and Darby [135]: (i) the lubricity theory ― a plasticizer acts as a lubricant to facilitate movement of resin macromolecules over each other; (ii) the gel theory ― a plasticizer prevents polymer chains from interacting with one another by breaking attachments along the polymer chain and masking the centers of forces such as Van der Waals and hydrogen bonding that have held these polymer chains together; (iii) the free volume theory ― a plasticizer, being a small molecule, increases the number of ends, thus adds a significant amount of free volume, and decreases the glass transition temperature of the polymer. According to all these theories mentioned above, more glycerol in SPI resin makes the resin more ductile. As seen from Figure 4.1 (d), the moisture content of SPI resin increased with the percentage of glycerol content at any pH value. This is because water molecules are preferably attracted to each glycerol molecule, which contains three hydroxyl groups. As can be noticed in Figures 4.1 (a) and (c) and Table 4.1, the fracture stress and Young’s modulus of the resin with 10% glycerol decreased as pH value increased; they dropped significantly at pH 11. It has been reported by Hermansson [54] that dissociation of the 7S globulin in soy protein on the alkaline side of the isoelectric point starts at pH 10, and the 11S globulin dissociates into subunits even at neutral pH. Thus, both 7S and 11S can readily dissociate at pH 11. It is also clear that the moisture content of the SPI resin containing 10% glycerol increased with pH value, as shown in Figure 4.1 (d). Moisture also acts as a plasticizer for SPI resin [59]. The fracture stress and Young’s modulus at pH 11 were, therefore, significantly lower than at pH 7 and 9. On the other hand, for resins containing 20 to 40% glycerol, it seems difficult to tell the differences among pH values in the stress and modulus of the SPI resins, as indicated by the letters (a-b) in Figures 4.1 (a) and (c). The differences may 51 be explained by the trends of the moisture content which are not consistent at each of the glycerol content percentage. As for the fracture strain of the resins, it increased with pH value consistently at every glycerol content, as presented in Figure 4.1 (b). This can be elucidated by the degree of unfolding of the protein chains. Sodium hydroxide (alkaline pH) treatment is known to unfold the protein molecules exposing the polar groups; the denaturation is, therefore, enhanced by increasing the pH to about 11 [55]. More glycerol molecules as well as water tend to attach to the protein chains as the pH value rises from 7 to 11 as the molecules open up and expose polar groups. Both these act as plasticizers causing an increase in the fracture strain of the SPI resins. For further research, in this study, 10% glycerol content was chosen because of its desirable effects on the tensile properties of the resins. At 0% glycerol, the SPI resin is very brittle, and at glycerol content of 20% or higher, the modulus is too low. 4.1.2 Optimization of pH Value for SPI Resin Modification Figures 4.2 (a) and (b) present the tensile fracture strength and Young’s modulus of pure SPI resin and SPI resin modified with 20% Phytagel® (SPI-20PH) at the pH values of 7, 9 and 11. Both the resins contain 10% glycerol by weight of SPI powder, as mentioned previously. The data indicate that the strength and modulus of pure SPI resin decreased with increasing pH value, as discussed in section 4.1.1. For the SPI-20PH resin, the trend remained the same. The focus in this study is on the differences between the pure SPI and SPI-20PH resins. When pure SPI resin was compared with SPI-20PH resin at same pH value, the fracture strength and Young’s modulus of SPI resin showed significant improvement by adding 20% Phytagel®. This mechanism and the effects of Phytagel® on the tensile properties of SPI resin are discussed later in section 4.2. The effects of pH values on the difference between the fracture strength and Young’s modulus of pure SPI resin, 52 Fracture Strength, MPa : Pure SPI Resin : SPI-20PH Resin 50 45 40 35 74.8% 93.8% 106.8% 30 25 20 15 10 5 0 7 9 11 pH value (a) 1600 Young’s Modulus, GPa 1400 35.2% 41.1% 1200 1000 55.8% 800 600 400 200 07 9 11 pH value (b) Figure 4.2 Comparison between Pure SPI Resin and SPI Resin Modified with 20% Phytagel® (SPI-20PH) in (a) Tensile Fracture Strength and (b) Young’s Modulus at the pH Values of 7, 9 and 11 Note: The percentages in the figure represent the increased strength and modulus of SPI resin by modifying with 20% Phytagel® (SPI-20PH) based on pure SPI resin at each pH value. 53 and SPI-20PH are discussed here. The percentages in Figures 4.2 (a) and (b) represent the increased strength and modulus of SPI resin by modifying with 20% Phytagel® based on pure SPI resin at each pH value. For both fracture strength and Young’s modulus, the percentages increased with pH value from 74.8 to 106.8% and from 35.2 to 55.8%, respectively. The effect of Phytagel® on the tensile strength and Young’s modulus of SPI resin prepared at pH 11 is most significant among various pH values studied. This result, therefore, indicated that the degree of unfolding of protein chains at pH 11 was greatest, as expected, and then the protein chains mingled better with the network of Phytagel® at pH 11 than at the other lower pH values. In addition to the physical entanglement of the chains, the increase of the percentages might be induced by the tensile properties of Phytagel® itself at higher pH. It has been reported by Sanderson [75] that the hardness and modulus of gellan gum can increase with the concentration of monovalent ions such as K+ and Na+. These monovalent cations associating with carboxylate groups in every tetrasaccharide repeating-unit (Figure 2.2) promote strong gelation due to the screening of the electrostatic repulsion between the ionized carboxylate groups [76]. Dickinson and Euston [136] have stated that weak attractive interactions may occur between anionic polysaccharides and proteins carrying a net negative charge at neutral pH. Therefore, pH 11 was chosen for further research in this project. 4.2 Modification of Soy Protein Isolate (SPI) Resin Using Phytagel® In this study, Phytagel®, biopolymer gellan gum, was used as a modifier to improve the mechanical properties of SPI resin containing 10% glycerol (by wt. of SPI powder). The pH value of the SPI suspension was adjusted to 11 before blending it with the Phytagel® solution. In order to determine the effect of Phytagel® on the tensile properties of the modified SPI/Phytagel® resins, the percentage of Phytagel® 54 content was set at 0, 10, 20 or 40% based on the weight of the SPI powder. SPI resins containing 0, 10, 20 and 40% Phytagel are termed SPI-0PH, SPI-10PH, SPI-20PH and SPI-40PH, respectively. As mentioned earlier in section 2.3.2, Phytagel® is a strong hydrogel that forms 3-D polymer network held together by cross-links or weaker cohesive forces such as hydrogen or ionic bonds. It can create interpenetrating polymer networks (IPN) like complex with SPI polymer due to hydrogen bonds and physical entanglements of the networks and the polymer chains [70]. The IPN like structure was expected to contribute to enhancing the tensile properties of SPI resin. 4.2.1 Effect of Phytagel® on the Tensile Properties of Modified SPI Resins The effect of Phytagel® on the tensile properties of the modified SPI/Phytagel® resins is shown in Table 4.2 and Figures 4.3 (a)-(d). The fracture stress and Young’s modulus of the resins, as shown in Figures 4.3 (a) and (b), increased significantly from 19.1 MPa to 59.6 MPa and from 844 MPa to 1787 MPa, respectively, as the percentage of Phytagel® content increased from 0% to 40%. At the same time, the fracture strain of the resins decreased from 75.7% to 10.6%. These properties agree with the data obtained by Lodha and Netravali [70]. As expected, these trends are due to the properties (strength, stiffness and brittleness) of Phytagel® itself and the IPN like structure where the SPI and the Phytagel® networks mingle well together. Table 4.2 Tensile Properties of Modified SPI/Phytagel® Resins as a Function of Phytagel® Content Phytagel® Content, Fracture Stress, Young's Modulus, Fracture Strain, % MPa MPa % 0 19.1 (0.9)* 844 (46) 75.7 (12.5) 10 22.3 (1.0) 852 (51) 14.0 (2.9) 20 39.5 (1.4) 1315 (54) 11.8 (1.3) 40 59.6 (3.0) 1787 (130) 10.6 (1.4) * Numbers in parentheses represent standard deviation. Toughness, MPa 14.5 (2.7) 2.6 (0.7) 3.6 (0.5) 4.5 (0.8) 55 Figure 4.3 Tensile Properties of Modified SPI Resins with Various Percentages of Phytagel®: (a) Tensile Stress, (b) Tensile Strain, (c) Young’s Modulus and (d) Tensile Toughness Note: Any two means with the same letter (a-d) by the plots don’t show statistically significant difference at α = .05 according to Tukey’s W procedure. 56 Fracture Stress, MPa 70 60 d 50 40 30 20 a 10 c b 0 0 10 20 30 40 50 Phytagel® Content (by wt. of SPI), % (a) 100 90 80 70 a 60 50 40 30 20 10 0 b b b 0 10 20 30 40 50 Phytagel® Content (by wt. of SPI), % (b) Fracture Strain, % 57 Young's Modulus, MPa 2500 Figure 4.3 (Continued) 2000 1500 b 1000 a a 500 c 0 0 10 20 30 40 50 Phytagel® Content (by wt. of SPI), % (c) 20 18 16 14 a 12 10 8 6 4 2 0 bb b 0 10 20 30 40 50 Phytagel® Content (by wt. of SPI), % (d) Tensile Toughness, MPa 58 Figure 4.3 (d) presents the tensile toughness of SPI resins as a function of Phytagel® content. The toughness of SPI-40PH, 4.5 MPa, was significantly lower than that of SPI-0PH, 14.5 MPa. The toughness was calculated based on the area under the stressstrain curve of the resins. The main reason for the low tensile toughness of SPI-40PH was its lower strain compared with SPI-0PH. When SPI-0PH and SPI-10PH specimens are compared, the tensile fracture strain decreased significantly from 75.7% to 14.0%. Figure 4.4 presents the stressstrain curves of the SPI resins with various percentages of Phytagel®. From the stressstrain curve of SPI-0PH, it is clear that the specimen yielded and continued to deform and absorb energy until fracture. This behavior is typically that of a ductile material. On the other hand, the SPI-10PH specimen broke without yielding according to its stress-strain curve. This difference implies that the hydrogel prevents the polymer chains of soy protein from moving, particularly stretching, in IPN like structures. Compared with the trend observed for the fracture strain, the fracture strengths of SPI-0PH and SPI-10PH didn’t show a big change, and no significant difference in the Young’s modulus was observed as well. This indicates that 10% Phytagel® is not enough to improve the fracture strength and Young’s modulus of SPI resin. This suggests that, at that low concentration, Phytagel® cannot form a crosslinked structure by itself. The fracture stress and Young’s modulus, however, increased steadily after 20% Phytagel®. 4.2.2 Fracture Surfaces of Modified SPI Resins Figures 4.5 (a)-(d) show SEM photomicrographs of the fracture surfaces of tensile tested pure SPI resin and modified SPI/Phytagel® resins. These fracture structures show clearly that the higher the content of the Phytagel®, the greater the roughness of the fracture surfaces. In Figures 4.5 (b), (c) and (d), it can be seen that cracks propagated longitudinally on the surfaces. Furthermore, the cracks on the 59 Stress, MPa 70 60 40% 50 40 20% 30 20 10% 10 0% 0 0 50 100 Strain, % Figure 4.4 Stress-Strain Curves of Modified SPI Resins with Various Percentages of Phytagel® 60 Cracks 100 µm (a) Cracks 100 µm (b) Cracks 100 µm (c) 100 µm (d) Figure 4.5 SEM Photomicrograph of Fracture Surfaces of Tensile Tested SPI Resins Modified with Various Percentages of Phytagel®: (a) 0%, (b) 10%, (c) 20% and (d) 40% 61 surfaces of SPI-40PH appear to be deeper than the ones observed on the surfaces of the other resins. This is perhaps because Phytagel® has highly crosslinked, 3-D polymer network. According to these observations, it is expected that the fracture energy of the SPI resin can be improved by incorporating Phytagel® into it. 4.3 Characterization of Nonwoven Kenaf Fiber Mats and Fibers Needle-punched nonwoven kenaf fiber mats were used as a reinforcement for fully green composites in this research. The properties and structure of the kenaf mat as well as properties of fibers are critical in determining the mechanical properties of the composites. The structure of the mats and fibers, and the tensile properties of the fibers were, therefore, characterized. Moreover, the fibers with the gauge lengths of 5, 10 and 15 mm were tensile tested to study the size (gauge length) effect. These results were applied to determine their tensile strengths at the critical lengths as described later in section 4.4. 4.3.1 Morphology of Kenaf Mats and Fibers Figure 4.6 (a) shows an optical photograph of a rectangular piece of the kenaf mat used in this study. The mat was made up of short kenaf fibers with various diameters and lengths. The mat measured approximately 10 mm in thickness and 165 g/m2 in weight. These characteristics or specifications can be controlled during the process of making nonwoven mats to meet requirements of the applications. A SEM photomicrograph of the mat taken at a higher magnification is presented in Figure 4.6 (b). In the photomicrograph, it can be seen that the fibers are oriented randomly in all directions. Due to the random fiber orientation, composites using the mats are expected to be isotropic. The fiber orientation in the mat is significant since only the fibers in the direction of the stress contribute to the mechanical properties. 62 (a) 100 µm (b) Figure 4.6 (a) Optical Photograph of a Nonwoven Kenaf Mat and (b) SEM Photomicrograph of a Kenaf Mat Showing Individual Fibers 63 The fibers in other directions contribute only partially. As a result, the fiber orientation determines the mechanical properties of any composite based on that mat. Figure 4.7 (a) exhibits the SEM photomicrograph of a fractured end of a tensile tested kenaf fiber. The fiber breaks clear and transverse to the axis, in a straight line characteristic of brittle fracture. In this case, it is because single fibrils strongly bond with one another. The fracture type of kenaf fibers is different from that of synthetic fibers such as Kevlar® and polybenzothiozole fibers. Netravali et al. [87] have shown the fibrillar fractures of these synthetic fibers. The SEM photomicrograph in Figure 4.7 (b) shows that single fibrils in the kenaf fiber are aligned longitudinally. In addition, Figure 4.7 (b) also presents a relatively smooth surface of a kenaf fiber. The surface smoothness should have a negative effect on the interfacial properties between the fiber and a matrix since it precludes physical bonding. 4.3.2 Tensile Properties of Kenaf Fibers The tensile fracture strength and Young’s modulus of kenaf fibers extracted from the mats were measured. The tensile testing was performed on a fiber with 10 mm gauge length at a strain rate of 0.1 min-1. Figures 4.8 (a) and (b) show Weibull distributions of the fracture strength and Young’s modulus, respectively. The mean values of the fracture strength and Young’s modulus calculated using the Weibull parameters were 257 MPa and 14 GPa, respectively. These properties are low compared with many other natural fibers as shown earlier in Table 2.3. It should be noted that the needle-punching process for kenaf mats creates additional flaws in the fibers, particularly where the needles touch the fibers. The Weibull plots for the fracture strength and Young’s modulus in Figures 4.8 (a) and (b) lie in a wide range, which results in shape parameters of 2.28 for the fracture strength distribution and 3.44 for the Young’s modulus distribution. 64 100 µm (a) 10 µm (b) Figure 4.7 Typical SEM Photomicrographs of (a) Fracture End of a Tensile Tested Kenaf Fiber, and (b) A Longitudinal View of a Kenaf Fiber 65 Fracture Probability, % 99.9 90.0 70.0 50.0 20.0 10.0 5.0 2.0 1.0 0.5 0.2 0.1 100 500 1000 Tensile Fracture Strength, MPa (a) Fracture Probability, % 99.9 90.0 70.0 50.0 20.0 10.0 5.0 2.0 1.0 0.5 0.2 0.1 1 10 50 Young’s Modulus, GPa (b) Figure 4.8 Weibull Distributions of the Tensile Properties of Kenaf Fibers with 10 mm Gauge Length: (a) Tensile Frature Strength and (b) Young’s Modulus 66 The shape parameter for the fracture strength of kenaf fibers is lower than that of carbon fibers, about 5 [127]. Carbon fibers have consistent diameter. One reason for the lower shape parameters for the fracture strength and Young’s modulus of kenaf fibers is the non-circular cross sections of fibers, which yield large errors in measuring the diameters. The diameter of the kenaf fibers varied between 30 and 100 µm. Moreover, the differences between kenaf bark and core fibers in fracture strength and Young’s modulus may be another reason for their variability. As discussed in section 2.4.1, the bark and core fibers contain distinctly different structures including walls and lumens. Their chemical compositions such as cellulose, hollocellulose and lignin are also different. Because of these features, their fracture strength and Young’s modulus are expected to vary as well. 4.3.3 Tensile Strength of Kenaf Fibers as a Function of Gauge Lengths The Weibull distribution plots of the tensile strength of kenaf fibers with gauge lengths of 5, 10 and 15 mm are shown in Figure 4.9. The parameters such as shape parameter and scale parameter and factors such as mean, standard deviation, coefficient of variance and correlation coefficient computed from these Weibull distributions are listed in Table 4.3. In Table 4.3, the correlation coefficient of the distribution for each gauge length is more than 0.95. This indicates that the tensile fracture strength values show good fit to Weibull distributions for all these gauge lengths. It can be seen that the mean fracture strength decreased with increasing gauge length. The decrease of the mean fracture strength with increase in the gauge length shows the size effect for the fiber. From 5 mm to 15 mm gauge length, approximately 20% decrease in fiber fracture strength was observed. 67 99.9 Fracture Probability, % 90.0 70.0 50.0 20.0 10.0 5.0 2.0 1.0 0.5 0.2 0.1 100 5 mm 10 mm 15 mm 500 1000 Tensile Fracture Strength, MPa Figure 4.9 Weibull Plots for the Tensile Fracture Strength Distributions of Kenaf Fibers with the Gauge Lengths of 5, 10 and 15 mm 68 Table 4.3 Weibull Parameters and Factors of Tensile Fracture Strength of Kenaf Fibers for Gauge Lengths of 5,10 and 15 mm Gauge Length, Shape Parameter, Scale Parameter, l, ρ mm σ0, MPa 5 2.48 378.1 10 2.28 289.6 15 1.97 228.8 Mean, µ, MPa 335.4 256.5 202.8 Standard Deviation, Coefficient of Variance, Correlation Coefficient, SD , CV, R MPa % 144.5 43.1 0.983 119.0 46.4 0.965 107.7 53.1 0.984 Shape parameters, between about 2.0 and 2.5, of kenaf fibers are low compared with those of synthetic fibers such as glass, graphite and Kevlar® fibers. Netravali et al. [87] have presented the shape parameters: 6.99 for S-2 glass, 4.91 for AS-4 graphite and 13.98 for Kevlar®. The low values of kenaf fibers are characteristic of natural fibers, which vary greatly in fracture strength. In addition, the shape parameter decreased from 2.48 to 1.97, and the coefficient of variance increased from 43.1% to 53.1% as gauge length increased from 5 mm to 15 mm. These trends indicate that the longer gauge length, the bigger variability in fiber fracture strength. Shape parameter is assumed to be constant regardless of change in gauge length based on the weakest link rule. In reality, the assumption may be questionable [127]. Pisanova et al. [137] have discussed a decrease of shape parameters for carbon fibers with increasing gauge length. 4.4 Interfacial Properties between Kenaf Fiber, and Pure or Modified Soy Protein Isolate (SPI) Resin The compatibility of fiber and matrix is a significant factor in determining the mechanical properties of composites. The fiber/matrix interface helps to transfer load from resin to fibers and thus helps to maintain the integrity of the composites. The interfacial shear strength (IFSS) between kenaf fibers, and pure SPI or modified SPI/Phytagel® resins was measured using the single fiber composite (SFC) technique. 69 To determine the effect of Phytagel® on the strength, the percentages of Phytagel® by weight of SPI powder was set at 0, 10, 20 and 40%. After completing the SFC tests, the specimens were observed using an optical microscope (model-BX 51: Polarizing Microscope, Olympus America Inc., Melville, NY) to measure the fragment lengths of a kenaf fiber in the specimen. At the same time, the fracture of a kenaf fiber in the specimen was evaluated under the microscope. 4.4.1 Determination of the Tensile Strength of Kenaf Fibers at the Critical Fragment Length As mentioned in section 3.4.4, it is necessary to obtain the tensile strength of kenaf fibers at the critical fragment length, lc , in order to compute IFSS, τ , between a fiber and a matrix. To predict the Weibull scale parameter for fiber strength, σ f , at any different gauge length, l , the Weibull shape parameter, ρ , and the scale parameter, σ 0 , at the experimentally tested gauge length, l0 , were used as shown below. ( )σ f = σ 0 l / l0 −1/ ρ (4.1) These Weibull parameters were calculated according to the weakest link rule with statistical independence of fiber segments. Figure 4.10 shows the mean fracture strength, µ , and the standard deviation, SD, of kenaf fibers for various gauge lengths listed in Table 4.3. The mean fiber strength, µ , at the critical length was determined with a subsequent linear fit function [138]. ln(µ ) = − 1 ρ ln(l) + ln{σ 0Γ(1 +1/ ρ )} (4.2) As shown by equation (4.2), a log-log plot of the mean strength versus gauge length produces a straight line with a negative slope of one over the Weibull shape parameter, ρ . From Figure 4.10, the linear fit yielded a shape parameter estimate of ρ = 2.23 . Considering that the shape parameter, ρ , was increasing with decreasing 70 : 5 mm : 10 mm : 15 mm ln (Tensile Strength), MPa 12 10 ρ = 2.23 8 6 4 2 0 1 1.5 2 2.5 3 ln (Gauge Length), mm Figure 4.10 Linear Fit Function of the Mean Tensile Strength of Kenaf Fibers versus the Gauge Length on a Log-log Scale 71 Interfacial Shear Strength, MPa 20 18 16 14 12 10 8 6 4 2 0 0 10 20 40 1 Phytagel® Content (by wt. of SPI), % Figure 4.11 Interfacial Shear Strength Measured by the SFC Technique between a Kenaf Fiber and SPI Resins Modified with Various Phytagel® Content 72 gauge length, ρ = 2.28 obtained for 10 mm gauge length was used to predict the fiber strength at the critical length. The strength calculated using equation (4.2), at the critical fragment length, was in the range of 450 to 750 MPa, depending on the critical lengths measured experimentally for various resins. 4.4.2 Interfacial Shear Strength between Kenaf Fiber, and Pure or Modified SPI Resin Figure 4.11 presents the IFSS values of kenaf fiber/pure SPI or modified SPI/ Phytagel® resins. The IFSS value increased with the percentage of Phytagel® content until 20%. This is a result of the effect of the shrinkage of Phytagel®, a highly swollen hydrogel containing many hydrogen bonds. Phytagel® shrinks a significant amount during the drying process to make modified SPI/Phytagel® sheets or kenaf mat composites using the modified resins. It had been expected that, the higher content of Phytagel® modified SPI/Phytagel® resins have, the greater the resin shrinkage is. The resin shrinkage created forces in the resin which surrounds a fiber, as shown by the arrows in Figure 4.12. Since SPI resins modified with the higher content of Phytagel® can hold a kenaf fiber more strongly due to higher shrinkage, the IFSS between the fiber and the resins increases. The higher IFSS yields shorter critical fragment lengths, as described previously in section 3.4.4. The IFSS value, 10.2 MPa, of kenaf fiber/SPI-40PH, however, was lower than 13.6 MPa obtained for SPI-20PH. The measurement of fiber fragment length of single fiber composites using 40% Phytagel® was tedious because tiny fractures of the fiber were difficult to distinguish. Fiber Resin : force occurred in resin Figure 4.12 Force Created by Resin Shrinkage in Crosssection of a Single Fiber Composite 73 As a result, it was quite possible to miss the fractures and obtain a longer critical length. This is perhaps the major reason why the lower IFSS value was obtained for SPI-40PH. Own et al. [139] have suggested conducting the SFC technique using a matrix that breaks at a three times higher strain than the fiber. Since the strain of the matrix (10.6%) doesn’t exceed three times that of the fiber (less than 4%) in this study, the difficulty in distinguishing the fractures was caused by the brittleness of the resin. In this case, it is possible that the fragmentation was not completed when the fracture of the specimen occurred. This is another reason for the lower IFSS for the SPI-40PH resin. Figures 4.13 (a)-(d) exhibit the typical fractures of kenaf fibers in SFC specimens with the various percentages of Phytagel®. These images were obtained using the optical microscope. The observations of the fractures support the hypothesis that the higher the content of Phytagel®, the higher the IFSS. In Figure 4.13 (a), debonding can be seen in the composites containing SPI-0PH due to the low IFSS. In Figures 4.13 (c) and (d), sharp cracks can be seen in the SFCs with SPI-20PH and SPI40PH, whereas a wider crack accompanied by shear yielding of the SPI-10PH resin in Figure 4.13 (b). These observations are consistent with the fracture modes which were defined based on the degree of interfacial adhesion by Mullin et al. [140] as presented in Figure 4.14 and are stated here: (i) In Figure 4.14 (a), in the case of a weak interface, the initial fiber break is simultaneously accompanied by an interfacial debonding; (ii) in the case of a relatively strong interface but with a matrix that has relatively lower shear than tensile strength capability, the initial fiber break is followed by a butterfly-type crack as seen in Figure 4.14 (b); and (iii) in the case of a strong interface, the initial fiber break is followed by a disk-shaped matrix crack in Figure 4.14 (c). Thus the fiber fractures in the SFCs with SPI-40PH imply strong interfacial adhesion between a kenaf fiber and SPI-40PH. 74 Debonding Wider Crack 100 µm Fiber (a) 100 µm Fiber (b) Sharp Crack Sharp Crack 100 µm Fiber 100 µm Fiber (c) (d) Figure 4.13 Typical Fractures of a Kenaf Fiber Embedded in Modified SPI Resins with Phytagel® Contents: (a) 0%, (b) 10%, (c) 20% and (d) 40% 75 Fiber Fiber Fiber Interfaical debonding (a) Butterfly-type crack (b) Disk-shaped matrix crack (c) Figure 4.14 Three Modes of Fractures which Occur in a Single Fiber Composite during a Tensile Test, Defined by Mullin et al. [137]: (a) weak interface; (b) relatively strong interface ; (c) strong interface 76 4.4.3 Effect of the Shrinkage of Phytagel® on Interfacial Properties To investigate the effect of the shrinkage of Phytagel® on the IFSS between kenaf fibers, and pure SPI or modified SPI/Phytagel® resin as discussed previously, the microbead test using E-glass fibers was conducted to measure the IFSS. In fact, the tensile fracture strength of kenaf fibers was not high enough to conduct the microdrop technique using SPI resins. In order to focus on the shrinkage effect of Phytagel®, glass fibers, which have smooth surface without any treatment and don’t bond chemically with SPI resins, were used. Figure 4.15 exhibits the interfacial strength between a glass fiber, and modified SPI/Phytagel® resin as a function of Phytagel® content. The IFSS between the glass fiber and modified SPI/Phytagel® resin increased slightly with the Phytagel® percent as can be seen in Figure 4.15. These results agree with the observations of the typical fractures of kanaf fibers in SFCs with various Phytagel® contents. As a consequence, it was confirmed that the interfacial adhesion of kenaf fiber/SPI-40PH was strong. 4.5 Mechanical Properties of Nonwoven Kenaf Mat Composites Using Pure or Modified Soy Protein Isolate (SPI) Resin To find the appropriate percentage of Phytagel® for fabricating nonwoven kenaf mat composites using modified SPI/Phytagel® resins, the mechanical properties (tensile, flexural and impact) properties of the kenaf mat composites were determined. The percentage of Phytagel® content was set up to 0, 10, 20 or 40% by weight of SPI powder. To identify the fracture mechanism of the composites, the fracture surfaces of impact tested specimens were observed using a SEM (440, Leica Cambridge, Ltd., Cambridge, UK). The penetration of the modified SPI/Phytagel® resin inside of the kenaf mat composite was evaluated using SEM photomicrographs of the fractured surfaces as well as visual observations. Based on the evaluations, in addition to resin 77 Interfacial Shear Strength, MPa 7 6 5 4 3 2 1 0 0 10 1 20 40 Phytagel® Content (by wt. of SPI), % Figure 4.15 Interfacial Shear Strength Measured by the Microdrop Technique between E-Glass Fiber and SPI Resins Modified with Various Percentages of Phytagel® 78 properties and the interfacial properties between kenaf fibers and modified SPI resins, the mechanical properties of the composites are discussed in this section. It should be stated that three half (thickness) kenaf mats with resin were hot-pressed together, and a resin-rich layer might be found between each reinforcement layer on microscopic examination. However, the laminate composite, fabricated from nonwoven kenaf mats of half thickness with a random fiber orientation, are considered a single-layer composite in this research [88]. In addition to characterizations of the mechanical properties of the composites, the weight fraction of the resin for the composites containing SPI-0PH, SPI-10PH, SPI-20PH and SPI-40PH was measured as mentioned in section 3.3.1. The resin weight fractions of various composites are listed in Table 4.4. The remaining weight is that of kenaf mat. Table 4.4 Resin Weight Fractions Obtained for Kenaf Mat Composites using Various SPI Resins Type of Resin Weight Fraction, % SPI-0PH SPI-10PH SPI-20PH SPI-40PH 55.6 56.5 55.2 49.2 4.5.1 Effect of Phytagel® on the Tensile Properties of Kenaf Mat Composites Figures 4.16 (a) and (b) and Table 4.5 show the tensile fracture strength and Young’s modulus of kenaf mat composites as a function of Phytagel® content. Since the composites containing SPI-40PH resin had large defects inside, they are discussed later in greater detail in section 4.5.4. As can be seen in Figure 4.16 (a), there was no significant increase in the strength of the composites when Phytagel® percent increased from 0% to 20%. 79 Fracture Stress, MPa 40 35 30 ab ab a 25 b 20 15 10 5 0 0 10 20 30 40 50 Phytagel® Content (by wt. of SPI), % (a) 2000 1800 1600 1400 a a a 1200 1000 800 600 400 200 0 a 0 10 20 30 40 50 Phytagel® Content (by wt. of SPI), % (b) Young's Modulus, MPa Figure 4.16 Tensile Properties of Nonwoven Kenaf Mat Composites Using Modified SPI/Phytagel® Resins as a Function of Phytagel® Content: (a) Tensile Stress and (b) Young’s Modulus Note: Any two means with the same letter (a-d) by the plots don’t show statistically a significant difference at α = .05 according to Tukey’s W procedure. 80 Table 4.5 Mechanical Properties of Kenaf Mat Composites Using Modified SPI/Phytagel® Resins as a Function of Phytagel® Content Phytagel® Content, Tensile Fracture Stress, Young's Modulus, Flexural Fracture Stress, Chord Modulus, % MPa MPa MPa MPa 0 30.4 (2.4)* 1538 (112) 36.5 (4.7) 2590 (256) 10 32.8 (3.1) 1660 (156) 45.5 (7.2) 2753 (407) 20 33.4 (1.9) 1699 (79) 58.3 (1.8) 3592 (197) 40 29.5 (2.3) 1619 (102) 43.4 (1.9) 2793 (201) Impact Strength, kJ/m2 4.0 (0.4) 3.7 (0.4) 3.8 (0.2) 6.5 (0.5) * Numbers in parentheses represent standard deviation. To evaluate the effects of the Phytagel® contents from 0% to 20%, on the tensile fracture strength of the composites, the calculated theoretical strengths were compared with the experimental results. In this study, the theoretical values were predicted by using the following “rule of mixtures” modified by Fukuda and Chou [141] as given. σc =σ fVf 1 − l c 2l C0 + σ m (1 − V f ) (4.3) where σ c , σ f and σ m are the fracture stresses of the composites, fibers and SPI resins modified with the various percentages of Phytagel®, respectively, V f denotes the fiber volume fraction, lc is the average critical length, l is the average fiber length and C0 is the orientation factor for random array composites. The detailed process for the prediction of the tensile fracture strength of the composites is discussed in Appendix A. The average experimental results, σ , and the theoretical values, σ c , of the composite fracture strength are listed in Table 4.6. Compared with the predicted values, the experimental results are lower by 16%, 19% and 42% at 0, 10 and 20% Table 4.6 Comparison between Experimental and Theoretical Strengths of Kenaf Mat Composites Containing Different Percentages of Phytagel® Phytagel® Content, % 0 10 20 Experimental Value, σ , MPa 30.4 32.8 33.4 Theoretical Value, σ c , MPa 35.4 38.9 47.3 81 Phytagel® content, respectively. One major factor for the lower experimental strengths is because of the assumptions about the fiber length and the critical length for the rule of mixture. For estimating theoretical values, it was assumed that the fiber length in the mat is uniform and is longer than the critical length, lc . However, in this research, the length of kenaf fibers, extracted from the mat, varied from 0.143 mm to 24.68 mm. It is clear that some kenaf fibers are shorter than the critical length, which are from 0.23 mm to 3.62 mm. The kenaf fibers which are shorter than the critical length cannot hold more tensile stress than the fibers which are longer than the critical lenth. However, fibers even 3 times longer than the critical length have been shown to be ineffective in improving the strength of the composites [131]. As shown previously, the difference between the predicted values and the experimental results of the composite fracture strength at each of the Phytagel® content increases from 16% to 42% Phytagel® content. This might be caused by the brittleness of the modified SPI/Phytagel® resins. The fracture strains of SPI-0PH, SPI10PH and SPI-20PH are 75.7%, 14.0% and 11.8%, respectively, as discussed in section 4.2.1. Since SPI-20PH is much more brittle than SPI-0PH, the composites with SPI-20PH resin contain more fiber breaks which were created by hot-pressing during the fabrication process of the composites. Therefore, although Phytagel® appears to notably improve the fracture strength of the composites in theory, it didn’t in reality as can be seen from the letters of ‘a’ by the plots in Figure 4.16 (a). Figure 4.16 (b) presents the Young’s modulus of kenaf mat composites as a function of Phytagel® content. As seen for the composite fracture strength, adding Phytagel® didn’t improve significantly the modulus of the composites. For random short fiber composites, the predictions of the Young’s modulus values have been made based on the modulus of a fiber and a resin by Fukuda and Kawata [142] as given below. 82 Ec = ClCθ E f V f + Em (1−V f ) (4.4) where Ec , E f and Em are the Young’s moduli of the composites, fibers and resins, respectively, and V f denotes the fiber volume fraction. The factors Cl and Cθ reflect the effects of fiber length and orientation distributions, respectively. In this research, since the moduli of modified SPI/Phytagel® resins were lower by about two orders of magnitude compared with those of kenaf fibers, the second term of the equation (4.4) can be neglected. Therefore, the improvement of the resin modulus by adding Phytagel® didn’t contribute much to enhancing the Young’s modulus of the composites. 4.5.2 Effect of Phytagel® on the Flexural Properties of Kenaf Mat Composites The flexural properties of kenaf mat composites as a function of Phytagel® content are presented in Figures 4.17 (a) and (b) and Table 4.5. Unlike the tensile properties of the kenaf mat composites, their flexural strength and chord modulus improved remarkably as the Phytagel® content increased from 0% to 20%, as can be seen in Figure 4.17. Again, a detailed discussion about the composites with SPI-40PH is presented later in section 4.5.4. During the bending test, the half of the test beam (specimen) above the neutral axis, including the concave surface, is under compressive stress, and the other half of the beam below the neutral axis, including the convex surface is under tensile stress. Although the specimens fail or fracture in the tensile mode at the convex surface, the compressive properties of the material influence the resistance to the fracture. Compressive failure modes are sensitive to the elastic properties of the resin [3]. Therefore, the results of the composite flexural properties indicate that the structure of Phytagel®, which is highly cross-linked, 3-D polymer networks made up of cross-links or ionic bonds such as hydrogen bonds, led to a significant increase in the flexural stress and chord modulus of the composites up to 20% Phytagel®. 83 70 Flexural Stress, MPa 60 c 50 40 a 30 b b 20 10 Chord Modulus, MPa 0 0 10 20 30 40 50 Phytagel® Content (by wt. of SPI), % (a) 4500 4000 3500 b 3000 2500 a a 2000 1500 1000 a 500 0 0 10 20 30 40 50 Phytagel® Content (by wt. of SPI), % (b) Figure 4.17 Flexural Properties of Nonwoven Kenaf Mat Composites Using Modified SPI/Phytagel® Resins as a Function of Phytagel® Content: (a) Flexural Stress and (b) Chord Modulus Note: Any two means with the same letter (a-d) by the plots don’t show statistically a significant difference at α = .05 according to Tukey’s W procedure. 84 4.5.3 Effect of Phytagel® on the Impact Strength of Kenaf Mat Composites The impact strengths of kenaf mat composites using SPI resins modified with various Phytagel® contents are shown in Figure 4.18 and Table 4.5. The differences in impact strengths of the composites containing SPI-0PH, SPI-10PH and SPI-20PH were not significant. However, the impact strength of the composites with SPI-0PH was slightly higher than those of the composites with SPI-10PH and SPI-20PH. As discussed earlier in section 4.4.2, debonding occurred between a kenaf fiber and pure SPI resin, which showed the low interfacial shear strength (IFSS). On the other hand, the single fiber composites with SPI-10PH and SPI-20PH showed a wider crack and a sharp crack, respectively. As a result, the composites with pure SPI resin were able to absorb higher impact energy during fracture propagation than the composites with SPI-10PH and SPI-20PH. The impact strength of the composites with SPI-40PH was significantly (1.6 to 1.7 times) higher than those of the composites with SPI-0PH, SPI-10PH and SPI20PH. This trend was confirmed by the fracture types of the tested specimens shown in Figure 4.19. The impact tested composite specimen with SPI-40PH did not break completely and showed the broken specimen like a hinge. On the other hand, the composites containing SPI-0PH, SPI-10PH and SPI-20PH broke completely into two pieces. Many fibers pull-outs can be seen on the fracture surface of the specimen with SPI-40PH resin as exhibited in Figure 4.19 (b). The higher impact strength of the composites with SPI-40PH might be, therefore, a result of the fibers pulled out as well as the SPI resin toughened by 40% Phytagel® as investigated in section 4.2. Details about the pulled-out fibers are discussed based on the fracture surface analysis of impact tested kenaf composites in section 4.5.4. 85 Impact Strength, kJ/m2 8.0 7.0 6.0 b 5.0 4.0 a a a 3.0 2.0 1.0 0.0 0 10 20 30 40 50 Phytagel® Content (by wt. of SPI), % Figure 4.18 Impact Strengths of Nonwoven Kenaf Mat Composites Using Modified SPI/Phytagel® Resins as a Function of Phytagel® Content Note: Any two means with the same letter (a-d) by the plots don’t show statistically a significant difference at α = .05 according to Tukey’s W procedure. 86 (a) (b) Figure 4.19 Typical Optical Photograph of Impact Tested Nonwoven Kenaf Mat Composite Specimens Made Using SPI Resins Modified with Phytagel® Content: (a) 0, 10 and 20% and (b) 40% 87 4.5.4 Fracture Surface Analysis of Impact Tested Kenaf Mat Composites Figures 4.20 (a)-(d) present SEM photomicrographs of the fracture surfaces of impact tested kenaf mat composites using SPI resins modified with the various percentages of Phytagel® content. As mentioned previously in section 4.5.3, significant amount of long fibers pulled out was observed on the fracture surfaces of the composites with SPI-40PH shown in Figure 4.20 (d), compared with the other three specimens containing lesser amounts of Phytagel® in the SPI resin. Considering the strong IFSS between kenaf fibers and SPI-40PH, it is highly possible that areas where the fibers were pulled out were resin-poor regions. The regions appeared because of the following two phenomena caused by Phytagel®. First, after drying kenaf mats whose inside was occupied with highly swollen hydrogel with much water, spaces with no resin appeared in the kenaf mats. Second, while kenaf mats with the resin were drying, the thickness caused an uneven distribution of temperature in an oven. The drying rate on both surfaces of the mat was higher than the rate inside. Thus, the dried resin close to both surfaces attracted the wet resin inside which could still move, and then spaces with poor resin inside were produced. In contrast with the trend in impact strengths of the composites, the tensile and flexural properties of the kenaf composites with SPI-40PH resin dropped compared with the composites with SPI-20PH resin. This is because of the physical phenomena induced by Phytagel® as mentioned previousely. The weight fraction of the resin based on the composites with SPI-40PH resin, 49%, was lower than about 55% obtained for the composites with SPI-0PH, SPI-10PH and SPI-20PH resins, as shown in Table 4.4. Therefore, additional Phytagel® used as a modifier for SPI resin, didn’t contribute to improving the tensile and flexural properties of kenaf mat composites. 88 300 µm (a) 300 µm (b) 300 µm 300 µm (c) (d) Figure 4.20 Fracture Surfaces of Impact Tested Nonwoven Kenaf Mat Composites Using Modified SPI/Phytagel® Resins as a Function of Phytagel® Content: (a) 0%, (b) 10%, (c) 20% and (d) 40% 89 4.6 Development of Hybrid Nonwoven Kenaf Mat Composites Using Fibrillated Bamboo Fiber (FBF) Sheets To achieve further improvements in nonwoven kenaf mat composite mechanical properties, fibrillated bamboo fiber (FBF) sheets were incorporated into the kenaf mat composites, resulting in hybrid green composites. Composites made out of FBF have been reported to possess remarkable mechanical properties [125]. Kenaf mat composites, fabricated by hot-pressing three layers together, involved a laminate structure. Due to the structure, in this study, FBF sheets were inserted between the kenaf mat composite layers to improve the mechanical properties of the kenaf mat composites, as described previously in section 3.3.2. In other cases, when the material needs to be stronger in one direction, unidirectional composites using yarn etc. could be sandwiched to reinforce the composite in the desired direction. Based on the mechanical properties of the kenaf mat composites discussed in section 4.5, SPI resin containing 20% Phytagel® (SPI-20PH) was used to fabricate the hybrid green composites in this study. The tensile properties of kenaf mat composites are compared with those of hybrid kenaf composites using FBF sheets in Figure 4.21. The tensile fracture strength and Young’s modulus of the hybrid composites were 37.1 MPa and 2187 MPa, respectively, which represent a 10% and 20% improvement over kenaf mat composites, respectively. Table 4.7 presents the tensile fracture strength and Young’s modulus of FBF sheet, kenaf mat composites and the hybrid composites using FBF sheets. The fracture strength and Young’s modulus of FBF sheet were approximately twice higher than those of kenaf mat composite. The positive effect of FBF sheets on the fracture strength and Young’s modulus on the hybrid composites as seen in Figure 4.21 was because of the high tensile properties of FBF sheet. 90 : Kenaf Mat : Hybrid (Kenaf Mat + FBF Sheets) Fracture Strength, MPa Young’s Modulus, MPa 45 40 35 30 25 20 15 10 5 0 Tensile Strength 2500 2000 1500 1000 500 Young’s Modulus 0 Figure 4.21 Comparison in Tensile Properties of Kenaf Mat Composites and Hybrid Kenaf Mat Composites with Fibrillated Bamboo Fiber Sheets 91 Table 4.7 Comparison in Tensile Fracture Strength and Young’s Modulus of FBF sheet, Kenaf Mat Composites and the Hybrid Composites with FBF sheets Type of Material Fracture Strength, Young Modulus, MPa MPa FBF sheet 65.6 (14.8)* 3055 (903) Kenaf mat composites Hybrid composites with kenaf mat and FBF sheets 33.4 (1.9) 37.1 (2.6) 1699 (79) 2187 (151) * Numbers in parentheses represent standard deviation. Figure 4.22 shows the comparative flexural properties of kenaf mat composites and the hybrid composites using FBF sheets. Contrary to the tensile properties, the difference between kenaf mat based composite and the hybrid composites in flexural properties didn’t appear significant. This can be explained using the following fundamental equation of lamination theory [4]: N M    =    A 0 0 ε o  D κ   (4.5) where {N} and {M } are in-plane forces per unit length and the moments per unit { }length, respectively, ε o and {κ} indicate the midplane strains and the laminate curvatures, respectively, and [ A] and [D] matrices represent the in-plane stiffness and the bending stiffness, respectively, and are defined as follows. ∑[ ]N k [A] = Q (zk − zk−1) k =1 ∑[ ][ ]D = 1 3 N k =1 Q k ( z k3 − z 3 k −1 ) (4.6) (4.7) where the [ A] and [D] matrices are defined as summations over N layers, and [Q ]k is the transformed reduced stiffness of the kth layer corresponding to the z-coordinate. It is noted that all hybrid composites made in this research are symmetric along the midplane of the laminates. It is evident that both of [ A] and [D] matrices are dependent on the layer thickness, tk = (zk − zk−1) . In this study, the thickness of a FBF 92 Fracture Strength, MPa Chord Modulus, MPa : Kenaf Mat : Hybrid (Kenaf Mat + FBF Sheets) 65 60 55 50 45 40 35 30 25 20 15 10 5 0 Flexural Strength Chord Modulus 4500 4000 3500 3000 2500 2000 1500 1000 500 0 Figure 4.22 Comparison in Flexural Properties of Kenaf Mat Composites and Hybrid Kenaf Mat Composites with Fibrillated Bamboo Fiber Sheets 93 sheet, about 0.15 mm, was approximately 100-fold lower than that of one layer of a kenaf mat composite, which was about 1.3 mm. Considering the thickness differences of a FBF sheet and a kenaf mat composite layer, an order of magnitude thicker than FBF sheet and the above lamination theory, the FBF sheets didn’t contribute to enhancing the flexural properties of the hybrid composites. Higher FBF sheet thickness, however, would enhance the hybrid composite properties. The impact strength of the kenaf mat composites and the hybrid composites is exhibited in Figure 4.23. It was found out that the strength of the kenaf mat composites improved 116% by incorporating FBF sheets. Since the FBF sheets were thin, their impact properties could not be measured. However, based on their tensile properties, it can be assumed that the FBF sheets have excellent impact toughness. Overall, the hybrid composites showed potential to improve the mechanical properties for applications in industries such as automotive internal components. Arbelaiz et al. [114] have made polypropylene based hybrid composites reinforced with glass and flax fibers and demonstrated their mechanical (tensile and flexural) properties. They have reported that their tensile strength and modulus, and flexural strength and modulus were approximately 32 MPa, 2000 MPa, 50 MPa, and 3800 MPa, respectively. The ratio of glass and flax fibers in their hybrid composites is 1:3 which is close to the ratio of FBF sheets and kenaf mats (1:4) in the hybrid composites fabricated in this research. The hybrid kenaf mat composites with FBF sheets are comparative or superior to polypropylene based hybrid composites reinforced with glass and flax fibers in terms of tensile and flexural properties. 94 : Kenaf Mat 5 : Hybrid (Kenaf Mat + FBF Sheets) Impact Strength, kJ/m2 4 3 2 1 0 Figure 4.23 Comparison in Impact Streng1th of Kenaf Mat Composites and Hybrid Kenaf Mat Composites with Fibrillated Bamboo Fiber Sheets 95 CHAPTER 5 CONCLUSIONS Fully biodegradable, environmentally friendly ‘Green’ composites based on nonwoven kenaf fiber mats and hybrid ‘Green’ composites by incorporating fibrillated bamboo fiber (FBF) sheets were fabricated using modified soy protein isolate (SPI) resins. Their mechanical properties including tensile, flexural and impact properties were investigated. Kenaf fibers, modified SPI resins and their interfacial properties were also characterized. SPI resin was modified using various Phytagel® contents. The Phytagel® content was varied from 0% to 40%. Based on the results obtained, following conclusions are drawn in this research: 1. The fracture stress and Young’s modulus of SPI resin clearly decreased, and the fracture strain and moisture content increased with increasing the glycerol content. This confirmed the results of the earlier studies that glycerol acts as a plasticizer for SPI resin. 2. When SPI resin properties were compared with the resin modified with 20% Phytagel® (SPI-20PH), the effects of Phytagel® on the tensile strength and modulus of SPI resins prepared at pH 11 were more significant than at the pH values of 7 and 9. 3. The tensile fracture strength and Young’s modulus of the modified SPI resins significantly improved from 19.1 MPa to 59.6 MPa and from 844 MPa to 1787 MPa, respectively, as the Phytagel® content increased from 0% to 40%. In contrast, the tensile fracture strain of the resins decreased from 75.7% to 10.6% with increase in Phytagel® content to 40%. 4. The mean fracture strength and Young’s modulus of kenaf fibers tensile tested at 10 mm gauge length at a 0.1 min-1 strain rate based, on the Weibull distributions, were 257 MPa and 14 GPa, respectively. 96 5. The tensile tests of kenaf fibers with the gauge length of 5, 10 and 15 mm showed a size effect for the fibers in which a 5 mm increase in gauge length yielded approximately 20% decrease in fiber strength. 6. The tensile strength of kenaf fibers at the critical lengths was predicted to be in the range of 450 to 750 MPa. 7. The kenaf fiber/modified SPI/Phytagel® resin interfacial shear strength (IFSS) measured using the single fiber composite (SFC) technique increased with the percentage of Phytagel® content until 20%. The IFSS of the SFC containing SPI resin with 40% Phytagel® (SPI-40PH), however, was lower than that of the SFC with SPI-20PH. The lower IFSS was the result of the resin brittleness rather than the actual bond strength. 8. It was estimated by the microdrop test, using glass fibers, that the interfacial adhesion between a kenaf fiber and SPI-40PH would be strong. 9. The tensile fracture strength and Young’s modulus of nonwoven kenaf fiber mat composites with modified SPI/Phytagel® resins didn’t increase significantly with Phytagel® contents of 0% to 20%. 10. The flexural strength and chord modulus of the kenaf mat composites, however, improved remarkably as Phytagel® content increased from 0% to 20%. 11. The tensile and flexural properties of kenaf mat composites containing SPI40PH dropped from those of the kenaf mat composites with SPI-20PH, indicating additional Phytagel® didn’t enhance their tensile and flexural properties. 12. The impact strength of kenaf mat composites with SPI-40PH was significantly (1.6 to 1.7 times) higher than that of the composites with the other modified resins with lower Phytagel® content: 0, 10 and 20%. 97 13. The tensile strength and modulus of hybrid kenaf mat composites using fibrillated bamboo fiber (FBF) sheets and SPI-20PH were 37.1 MPa and 2187 MPa, respectively, which represent a 10% and 20% improvement over kenaf mat composites using the same resin. 14. The impact strength of kenaf mat composites improved by 116% after incorporating FBF sheets. In this thesis, it is concluded that the mechanical properties of hybrid ‘Green’ kenaf mat composites with FBF sheets are comparable or superior to those of polypropylene based hybrid ‘Semi Green’ composites reinforced with glass and flax fibers. There is significant scope to improve the properties of green composites further by improving design and replacing kenaf fibers with other stronger plant-based fibers. The engineered hybrid composites would be used for industrial applications such as automotive internal components. At the end of their life, they can be easily disposed of or composted without harming the environment, in fact helping it by completing the nature’s carbon cycle. 98 APPENDIX A As discussed in section 4.5.1, the following strength prediction equation of Fukuda and Chou [141] was applied to obtain the theoretical strength of random short fiber composites. σc =σ fVf 1 − l c 2l C0 + σ m (1 − V f ) (A.1) where σ c , σ f and σ m are the fracture stresses of the composites, fibers and SPI resins modified with the various percentages of Phytagel®, respectively, V f denotes the fiber volume fraction, lc is the average critical length, l is the average fiber length and C0 is the orientation factor for random array composites. For simplicity of calculation, it is assumed that the fiber length in the mat is uniform and is larger than the critical length, lc . The data independent of the properties of a matrix are used as follows. σ f = 256.5 MPa, which was presented in section 4.3.2. l = 3.42 mm, which was measured using 293 kenaf fibers, extracted randomly from the mats. C0 = 0.27, which was calculated by Fukuda and Chou [141], considering the two dimensional array model. The parameters dependent on the matrix, such as σ m and lc are shown in Table A.1. Table A.1 Fracture Stresses and Critical Length of Pure SPI Resin and SPI Resins Modified with 10 and 20% Phytagel® Phytagel® Content, % 0 10 20 Fracture Stress, σ m , MPa 19.1 22.3 39.5 Critical Length, l c , mm 1.91 1.57 1.50 99 These data for σ m and lc are same as discussed earlier in sections 4.2.1 and 4.4.1, respectively. In addition, the volume fraction of the fiber, V f , was computed using the following equation: Vf = Wf Wf /df / d f + Wm / dm (A.2) where W f and Wm are the volume fractions of the fiber and the matrix, respectively, and d f and dm denote the densities of the fiber and the matrix, respectively. As mentioned in section 4.5, Wm = 0.55 and then W f = 0.45. For the value of the fiber density, d f , the density (0.93 g/cm3) of kenaf whole stalk (bast and core) was used [143]. The densities of SPI resin and SPI resin modified with 20% Phytagel® were successfully measured by using the density gradient column as per the ASTM D1505- 98e1 procedure by Lodha and Netravali [70]. Based on these densities, the density of SPI resin modified with 10% Phytagel® was calculated. The densities, used for this strength prediction, are tabulated as follows. Table A.2 Densities of Pure SPI Resin and SPI Resins Modified with 10 and 20% Phytagel® Phytagel® Content, % 0 10 20 dm, g/cm3 1.31 1.33 1.34 100 REFERENCES 1. Aspden RM, Skakle JMS. Collagen in human bone, a small-angle scattering study; 2002. 2. Li SH, Zeng QY, Xiao YL, Fu SY, Zhou BL. Biomimicry of bamboo bast fiber with engineering composite materials. Materials Science and Engineering: C 1995;3(2):125-130. 3. Hull D. An introduction to composite materials: the Press Syndicate of the University of Cambridge, 1981. 4. Herakovich CT. Mechanics of Fibrous Composites: John Wiley & Sons, Inc., 1998. 5. Agarwal BD, Broutman LJ. Analysis and performance of fiber composites: John Wiley & Sons, Inc., 1980. 6. De SK, White JR. Short fibre-polymer composites: Woodhead Publishing Limited, 1996. 7. Mlekusch B. Thermoelastic properties of short-fibre-reinforced thermoplastics. Composites Science and Technology 1999;59(6):911-923. 8. Kluger J. By Any Measure, Earth Is at The Tipping Point. Time. 2006 April 3:34-42. 9. Stern B. Jointly Develop Natural Fiber Composites for Auto Industry. 1999 [cited; Available from: http://lists.essential.org/ag-res/msg00011.html 10. Marsh G. Next step for automotive materials. materials today. 2003 April:3643. 11. Prasad PN, Mark JE, Kandil SH, Kafafi ZH. Science and technology of polymers and advanced materials. New York: Plenum Press, 1998. 12. Wambua P, Ivens J, Verpoest I. Natural fibres: can they replace glass in fibre reinforced plastics? Composites Science and Technology 2003;63(9):12591264. 13. Joshi SV, Drzal LT, Mohanty AK, Arora S. Are natural fiber composites environmentally superior to glass fiber reinforced composites? Composites Part A: Applied Science and Manufacturing 2004;35(3):371-376. 14. Dahlke B, Larbig H, Scherzer HD, Poltrock R. Natural fiber reinforced foams based on renewable resources for automotive interior applications (Reprinted 101 from Polyurethanes World Congress 97, Sept 29 Oct 1, 1997). Journal of Cellular Plastics 1998;34(4):361-379. 15. Sherman LM. Natural Fibers: The New Fashion In Automotive Plastics. 1999 [cited; Available from: http://www.plasticstechnology.com/articles/199910fa1.html 16. Brouwer WD. Natural fiber composites: Where can flax compete with glass? SAMPE Journal 2000;36(6):18-23. 17. Stevens ES. Green Plastics : An Introduction to the New Science of Biodegradable Plastics: Princeton University Press, 2002. 18. Mohanty AK, Misra M, Drzal LT. Biocomposites vs. green composites: Opportunities are clearing but challenges are to be made. Abstracts of Papers of the American Chemical Society 2002;224:U40. 19. Netravali AN, Chabba S. Composites get greener. Materials Today 2003;6(4):22-29. 20. Mohanty AK, Misra M, Hinrichsen G. Biofibres, biodegradable polymers and biocomposites: An overview. Macromolecular Materials and Engineering 2000;276(3-4):1-24. 21. Takagi H. Development and Strengthening of Manila Hemp Green Composites. China - Japan Joint Seminar on Green composites; 2005; Shanghai, China; 2005. p. K-1-6. 22. Mohanty AK, Khan MA, Hinrichsen G. Surface modification of jute and its influence on performance of biodegradable jute-fabric/Biopol composites. Composites Science and Technology 2000;60(7):1115-1124. 23. Hokens D, Mohanty AK, Misra M, Drzal LT. Environment-friendly "Green" biodegradable composites from natural fiber and cellulosic plastic. Polyme preprints 2001;42(2):71-72. 24. Baillie C. Eco-Composites. Composites Science and Technology 2003;63(9):1223-1224. 25. Gatenholm P. Cellulosic biocomposites for automotive applications. Abstracts of Papers of the American Chemical Society 2001;221:U186. 26. Your Business In Brief. Plastics Technology. 2005 April. 27. Lin M. NEC Cell Phone - PLA/Kenaf. 2005 [cited; Available from: http://www.vestaldesign.com/vestalblog/2005/12/nec-cell-phone-plakenaf.html 102 28. Herrmann AS, Nickel J, Riedel U. Construction materials based upon biologically renewable resources - from components to finished parts. Polymer Degradation and Stability 1998;59(1-3):251-261. 29. Wollerdorfer M, Bader H. Influence of natural fibres on the mechanical properties of biodegradable polymers. Industrial Crops and Products 1998;8(2):105-112. 30. Mohanty AK, Wibowo A, Misra A, Drzal LT. Effect of process engineering on the performance of natural fiber reinforced cellulose acetate biocomposites. Composites Part a-Applied Science and Manufacturing 2004;35(3):363-370. 31. Mohanty AK, Misra M, Drzal LT. Sustainable bio-composites from renewable resources: Opportunities and challenges in the green materials world. Journal of Polymers and the Environment 2002;10(1-2):19-26. 32. Nam S. Environment-Friendly "Green" Biodegradable Composites using Ramie Fibers and Soy Protein Concentrate (SPC) Polymer [MS]. Ithaca: Cornell University; 2002. 33. Chabba S. Characterization of Environment Friendly 'Green' Composites with Modified Soy Protein Concentrate and Flax Yarn and Fabric [MS]. Ithaca: Cornell University; 2003. 34. Chabba S, Matthews GF, Netravali AN. Green composites using cross-linked soy flour and flax yarns. Green Chemistry 2005;7(8):576-581. 35. Cheng EZ, Sun XZ, Karr GS. Adhesive properties of modified soybean flour in wheat straw particleboard. Composites Part a-Applied Science and Manufacturing 2004;35(3):297-302. 36. Liu WJ, Mohanty AK, Drzal LT, Misra M. Novel biocomposites from native grass and soy based bioplastic: Processing and properties evaluation. Industrial & Engineering Chemistry Research 2005;44(18):7105-7112. 37. Andersen PJ, Kumar A, Hodson SK. Inorganically filled starch based fiber reinforced composite foam materials for food packaging. Materials Research Innovations 1999;3(1):2-8. 38. Fernandes EG, Pietrini M, Chiellini E. Bio-based polymeric composites comprising wood flour as filler. Biomacromolecules 2004;5(4):1200-1205. 39. Piekielna J, Mucha M. Polymer composites containing chitosan and starch as biodegradable fillers. Polimery 1999;44(1):24-29. 40. Park HM, Lee WK, Park CY, Cho WJ, Ha CS. Environmentally friendly polymer hybrids - Part I - Mechanical, thermal, and barrier properties of 103 thermoplastic starch/clay nanocomposites. Journal of Materials Science 2003;38(5):909-915. 41. McGlashan SA, Halley PJ. Preparation and characterisation of biodegradable starch-based nanocomposite materials. Polymer International 2003;52(11):1767-1773. 42. Ray SS, Yamada K, Okamoto M, Fujimoto Y, Ogami A, Ueda K. New polylactide/layered silicate nanocomposites. 5. Designing of materials with desired properties. Polymer 2003;44(21):6633-6646. 43. Ray SS, Yamada K, Okamoto M, Ueda K. New polylactide-layered silicate nanocomposites. 2. Concurrent improvements of material properties, biodegradability and melt rheology. Polymer 2003;44(3):857-866. 44. Yano H, Nakahara S. Bio-composites produced from plant microfiber bundles with a nanometer unit web-like network. Journal of Materials Science 2004;39(5):1635-1638. 45. Moon SI, Jin F, Lee C, Tsutsumi S, Hyon SH. Novel carbon nanotube/poly(Llactic acid) nanocomposites; Their modulus, thermal stability, and electrical conductivity. Macromolecular Symposia 2005;224:287-295. 46. Gennadios A, Weller CL, Testin RF. Temperature Effect on Oxygen Permeability of Edible Protein-based Films. Journal of food science 1993;58(1):212-214. 47. Zhang J, Mungara P, Jane J. Mechanical and thermal properties of extruded soy protein sheets. Polymer 2001;42(6):2569-2578. 48. Gennadios A, Ghorpade VM, Weller CL, Hanna MA. Heat Curing of Soy Protein Films. Transactions of the ASAE 1996;39(2):575-579. 49. Swain SN, Biswal SM, Nanda PK, Nayak PL. Biodegradable soy-based plastics: Opportunities and challenges. Journal of Polymers and the Environment 2004;12(1):35-42. 50. Krinski TL, Hou KC, inventors. Soy protein thickener Patent No. EP 1 054 103 A1, 11/22/2000. 51. Lodha P. Fundamental Approaches to Improving Performance of Soy Protein Isolate Based 'Green' Plastics and Composites [Ph.D]. Ithaca: Cornell University; 2004. 52. Berk Z. Technology of Production of Edible Flours and Protein Products from Soybeans: Food and Agriculture Organization of the United Nations Rome, 1992. 104 53. Wolf WJ, Babcock GE, Smith AK. Purification and Stability Studies of the 11 S Component of Soybean Proteins. Archives of biochemistry and biophysics 1962;99(2):265-274. 54. Hermansson AM. Physico-chemical aspects of soy proteins structure formation. Journal of Texture Studies 1978;9:33-58. 55. Kumar R, Choudhary V, Mishra S, Varma IK, Mattiason B. Adhesives and plastics based on soy protein products. Industrial Crops and Products 2002;16(3):155-172. 56. Wolf WJ. Purification and properties of the proteins. In: Smith AK, Circle SJ, editors. Soybeans: chemistry and technology. Westport, Connecticut: Avi Publishing Company, INC., 1972. p. 93-143. 57. Wolf WJ, Briggs DR. Purification and Characterization of the 11 S Component of Soybean Proteins. Archives of biochemistry and biophysics 1959;85(1):186199. 58. Ghorpade VM, Li H, Gennadios A, Hanna MA. Chemically Modified Soy Protein Films. Transactions of the ASAE 1995;38(6):1805-1808. 59. Lodha P. Green Composites with Soy Protein Isolate Resin and Ramie Fiber [MS]. Ithaca: Cornell University; 2001. 60. Singh H. Modification of food proteins by covalent crosslinking. Trends in Food Science and Technology 1991 August;2:196-200. 61. Park SK, Rhee CO, Bae DH, Hettiarachchy NS. Mechanical Properties and Water-Vapor Permeability of Soy-Protein Films Affected by Calcium Salts and Glucono-δ-lactone. J Agric Food Chem, 2001;49(5):2308-2312. 62. Rhim JW, Gennadios A, Handa A, Weller CL, Hanna MA. Solubility, Tensile, and Color Properties of Modified Soy Protein Isolate Films. Journal of Agricultural and Food Chemistry 2000;48(10):4937-4941. 63. Deanin RD, Manion MA. Compatibilization of Polymer Blends. In: Shonaike GO, Simon GP, editors. POLYMER BLENDS AND ALLOYS. New York and Basel: Marcel Dekker, Inc., 1999. p. 1-21. 64. Mo X, Sun X. Plasticzation of Soy Protein Polymer by Polyol-Based Plasticizers. Journal of the American Oil Chemists' Society 2002;79(2):197202. 65. Wang S, Sue HJ, Jane J. Effects of Polyhydric Alcohols on the Mechanical Properties of Soy Protein Plastics. Journal of Material Science _ Pure Applied Chemistry 1996;A33(5):557-569. 105 66. Monick JA. Alcohols: their chemistry, properties, and manufacture. New York: Reinhold Book Corp., 1968. 67. Becard G, Piche Y. Establishment of Vesicular-Arbuscular Mycorrhiza in Root Organ-Culture - Review and Proposed Methodology. Methods in Microbiology 1992;24:89-108. 68. Shungu D, Valiant M, Tutlane V, Weinberg E, Weissberger B, Koupal L, et al. Gelrite as an Agar Substitute in Bacteriological Media. Applied and Environmental Microbiology 1983;46(4):840-845. 69. Doner LW, Douds DD. Purification of commercial gellan to monovalent cation salts results in acute modification of solution and gel-forming properties. Carbohydrate Research 1995;273(2):225-233. 70. Lodha P, Netravali AN. Characterization of Phytagel® modified soy protein isolate resin and unidirectional flax yarn reinforced "green" composites. Polymer Composites 2005;26(5):647-659. 71. Banik RM, Kanari B, Upadhyay SN. Exopolysaccharide of the gellan family: prospects and potential. World Journal of Microbiology and Biotechnology 2000;16(5):407-414. 72. Kang KS, Veeder GT, Mirrasoul PJ, Kaneko T, Cottrell IW. Agar-Like Polysaccharide Produced by a Pseudomonas Species - Production and Basic Properties. Applied and Environmental Microbiology 1982;43(5):1086-1091. 73. Crescenzi V. Microbial Polysaccharides of Applied Interest - Ongoing Research Activities in Europe. Biotechnology Progress 1995;11(3):251-259. 74. Jansson PE, Lindberg B, Sandford PA. Structural Studies of Gellan Gum, an Extracellular Polysaccharide Elaborated by Pseudomonas-Elodea. Carbohydrate Research 1983;124(1):135-139. 75. Sanderson GR. Gellan Gum. In: Harris P, editor. Food Gels. Bedford, UK: Elsevier science pbuleshers ltd., 1990. p. 201-232. 76. Nakajima K, Ikehara T, Nishi T. Observation of gellan gum by scanning tunneling microscopy. Carbohydrate Polymers 1996;30(2-3):77-81. 77. Sperling LH. Interpenetrating polymer networks: an overview. In: Klempner D, Sperling LH, Utracki LA, editors. Interpenetrating Polymer Networks. Washington, DC: American Chemical Society, 1994. p. 3-38. 78. Gursel I, Balcik C, Arica Y, Akkus O, Akkas N, Hasirci V. Synthesis and mechanical properties of interpenetrating networks of polyhydroxybutyrate-co- 106 hydroxyvalerate and polyhydroxyethyl methacrylate. Biomaterials 1998;19(13):1137-1143. 79. Luo S, Netravali AN. Mechanical and thermal properties of environmentfriendly "green" composites made from pineapple leaf fibers and poly(hydroxybutyrate-co-valerate) resin. Polymer Composites 1999;20(3):367378. 80. Luo S, Netravali AN. Characterization of henequen fibers and the henequen fiber/poly(hydroxybutyrate-co-hydroxyvalerate) interface. Journal of Adhesion Science and Technology 2001;15(4):423-437. 81. Beukers A. Lightness : the inevitable renaissance of minimum energy structures. 2nd ed: Rotterdam : 010 publishers, 1999. 82. Batra SK. Other Long Vegetable Fibers. In: 2nd edition rae, editor. Handbook of fiber chemistry. New York: Marcel Dekker, 1998. p. 549. 83. Joseph K, Thomas S, Pavithran C. Effect of chemical treatment on the tensile properties of short sisal fibre-reinforced polyethylene composites. Polymer 1996;37(23):5139-5149. 84. Yuan XW, Jayaraman K, Bhattacharyya D. Mechanical properties of plasmatreated sisal fibre-reinforced polypropylene composites. Journal of Adhesion Science and Technology 2004;18(9):1027-1045. 85. Valigra L. Tough as soybeans: the long search for plants with optimal material properties starts to pay off. 2005 [cited; Available from: http://www.fiberfutures.org/soybean.html 86. Mohanty AK, Drzal LT, Misra M. Engineered natural fiber reinforced polypropylene composites: influence of surface modifications and novel powder impregnation processing. Journal of Adhesion Science and Technology 2002;16(8):999-1015. 87. Netravali AN, Li ZF, Sachse W, Wu HF. Determination of Fiber Matrix Interfacial Shear-Strength by an Acoustic-Emission Technique. Journal of Materials Science 1991;26(24):6631-6638. 88. Kugler DE. Non-wood Fiber Crops: Commercialization of Kenaf for Newsprint. In: Janick J, Simon JE, editors. First National Symposium on New Crops: Timber Press; 1990. p. 289-292. 89. Shah SMA, Younis MT, Nabi G. Characteristics of kenaf fibers. Part I. Physical characteristics. Pakistan Journal of Scientific and Industrial Research 1980;23(5):213-217. 107 90. Parikh DV. A Quieter Car Ride With Kenaf. Agricultural Research. 2002 July. 91. Columbus PE, Fuller JM. Factors Affecting Kenaf Fiber and Core Separation. In: Sellers TJ, Reichert AN, Columbus PE, Fuller JM, Williams K, editors. Kenaf Properties Processing and Products: Mississippi State University, 1999. p. 83-89. 92. Pande H, Roy DN. Influence of fibre morphology and chemical composition on the papermaking potential of kenaf fibres - A look at what attributes affect tensile strength. Pulp & Paper-Canada 1998;99(11):31-34. 93. Pande H, Roy DN. Recycling Potential of Kenaf Fibers. In: Sellers TJ, Reichert AN, Columbus PE, Fuller JM, Williams K, editors. Kenaf Properties Processing and Products: Mississippi State University, 1999. p. 315-320. 94. Calamari AT, Jr., Tao W, Akin ED. Some Important Physical Characteristics of Kenaf Fiber Bundles and of The Ultimate Kenaf Fibers. In: Sellers TJ, Reichert AN, Columbus PE, Fuller JM, Williams K, editors. Kenaf Properties Processing and Products: Mississippi State University, 1999. p. 187-191. 95. Zhang J, Jobes RA, Dicks MR. Commercial kenaf production: Where are the markets? International kenaf conference; Fresno, CA: International kenaf association; 1993. p. 145-151. 96. Kaldor AF. Kenaf, an Alternate Fiber for the Pulp and Paper Industries in Developing and Developed-Countries. Tappi Journal 1992;75(10):141-145. 97. Webber CLI, Ray CD, Bledsoe RE, Blalock JL. Production Properties of Industrial-grade Kenaf Particleboard. In: Sellers TJ, Reichert AN, Columbus PE, Fuller JM, Williams K, editors. Kenaf Properties Processing and Products: Mississippi State University, 1999. p. 359-365. 98. Fisher G. Kenaf harvesting, processing and products; Ladonia, TX: International Kenaf Association; 1993. p. 8-12. 99. Tao WY, Calamari TA, Crook L. Carding kenaf for nonwovens. Textile Research Journal 1998;68(6):402-406. 100. Baldwin BS, Hensler KL, Goatley JM. Non-woven Kenaf Fiber Mat as a Substrate for Establishing High-quality Soilless Sod. In: Sellers TJ, Reichert AN, Columbus PE, Fuller JM, Williams K, editors. Kenaf Properties Processing and Products: Mississippi State University, 1999. p. 257-263. 101. Ramaswamy GN, Sellers T, Tao W, Crook LG. Kenaf nonwovens as substrates for laminations. Industrial Crops and Products 2003;17(1):1-8. 108 102. Moreau JP, Belberger P, Tao WY. Mechanical Processing of Kenaf for Nonwovens. Tappi Journal 1995;78(2):96-105. 103. de Guzman ZI, Berana NB, Reyes AM, Jr. Abaca, Kenaf and Pineapple Fibers for Nonwoven Fabrics. NSTA Technology Journal 1982;Jan. - Mar.:77-87. 104. Tao WY, Moreau JP, Calamari TA. Kenaf Mats - Properties of Nonwoven Mats from Kenaf Fiber. Tappi Journal 1995;78(8):165-169. 105. Sanadi AR, Hunt JF, Caulfield DF, Kovacsvolgyi G, Destree B. High fiber-low matrix composites: kenaf fiber/polypropylene. Sixth International Conference on Woodfiber-Plastic Composites; Madison, Wisconsin; 2002. p. 121-124. 106. Maldas D, Kokta BV. Composite Molded Products Based on Recycled Thermoplastics and Waste Cellulosics .2. Kenaf Fiber-Recycled Pe Composites. Journal of Reinforced Plastics and Composites 1995;14(5):458470. 107. Magurno A. Vegetable fibres in automotive interior components. Angewandte Makromolekulare Chemie 1999;272:99-107. 108. Nishino T, Hirao K, Kotera M, Nakamae K, Inagaki H. Kenaf reinforced biodegradable composite. Composites Science and Technology 2003;63(9):1281-1286. 109. Chou T. Microstructural design of fiber composites. Cambridge, New York, Oakleigh Victoria: the Press Syndicate of the University of Cambridge, 1992. 110. Mehdi T. Static and dynamic mechanical properties of a kenaf fiber-wood flour/polypropylene hybrid composite. Journal of Applied Polymer Science 2005;98(2):665-672. 111. Lin QF, Zhou XD, Dai G, Bi YM. Some studies on mechanical properties of wood flour/continuous glass mat/polypropylene composite. Journal of Applied Polymer Science 2002;85(3):536-544. 112. Burgueno R, Quagliata MJ, Mohanty AK, Mehta G, Drzal LT, Misra M. Hybrid biofiber-based composites for structural cellular plates. Composites Part a-Applied Science and Manufacturing 2005;36(5):581-593. 113. Burgueno R, Quagliata M, Mehta GM, Mohanty AK, Misra M, Drzal LT. Sustainable cellular biocomposites from natural fibers and unsaturated polyester resin for housing panel applications. Journal of Polymers and the Environment 2005;13(2):139-149. 114. Arbelaiz A, Fernandez B, Cantero G, Llano-Ponte R, Valea A, Mondragon I. Mechanical properties of flax fibre/polypropylene composites. Influence of 109 fibre/matrix modification and glass fibre hybridization. Composites Part aApplied Science and Manufacturing 2005;36(12):1637-1644. 115. Okubo K, Fujii T, Yamamoto Y. Development of bamboo-based polymer composites and their mechanical properties. Composites Part A: Applied Science and Manufacturing 2004;35(3):377-383. 116. Yao W, Li ZJ. Flexural behavior of bamboo - fiber-reinforced mortar laminates. Cement and Concrete Research 2003;33(1):15-19. 117. Amada S, Untao S. Fracture properties of bamboo. Composites Part BEngineering 2001;32(5):449-457. 118. 'Satoyama (= coutryside forest)' is occupied by bamboos? [cited; Available from: http://ss.ffpri.affrc.go.jp/labs/kouho/mori/today-35.html 119. Deshpande AP, Rao MB, Rao CL. Extraction of bamboo fibers and their use as reinforcement in polymeric composites. Journal of Applied Polymer Science 2000;76(1):83-92. 120. Chen X, Guo Q, Mi Y. Bamboo fiber-reinforced polypropylene composites: A study of the mechanical properties. Journal of Applied Polymer Science 1998;69(10):1891-1899. 121. Yamamoto Y, Okubo K, Fujii T. DEVELOPMENT OF ECO-COMPOSITE USING BAMBOO FIBERS - FRPP using Bamboo Fibers Extracted by the Steam Explosion Method -. International Workshop on Green Compo; Japan; 2002. p. 30-34. 122. Tanahashi M, Higuchi T, Kobayashi H, Togamura Y, Shimada M. Characterization and Nutritional Improvement of Steam-Exploded Wood as Ruminant Feed. Cellulose Chemistry and Technology 1985;19(6):687-696. 123. Fujii T, Okubo K, Yamashita N. Development of High Performance Bamboo Composites using Micro Fibrillated Cellulose. Second International Conference on High Performance Structures and Materials; 2004 31 May - 2 June; Ancona, Italy. 124. Okubo K, Fujii T. Improvement of Mechanical Properties of Bamboo Polymer Composites Enhanced with Micro-Fibrillated Cellulose. China - Japan Joint Seminar on Green composites; Shanghai, China; 2005. p. K-22-27. 125. Takahashi N, Okubo K, Fujii T. Development of green composite using cellulose-microfibrils extracted from bamboo without polymer adhesive. Third International Workshop on Green Composites; 2005 March 16-17; Kyoto, Japan; p. 111-115. 110 126. Netravali AN, Schwartz P, Phoenix SL. Study of Interfaces of HighPerformance Glass-Fibers and Dgeba-Based Epoxy-Resins Using Single-FiberComposite Test. Polymer Composites 1989;10(6):385-388. 127. Netravali AN, Henstenburg RB, Phoenix SL, Schwartz P. Interfacial ShearStrength Studies Using the Single-Filament-Composite Test .1. Experiments on Graphite Fibers in Epoxy. Polymer Composites 1989;10(4):226-241. 128. Herrerafranco PJ, Drzal LT. Comparison of Methods for the Measurement of Fiber Matrix Adhesion in Composites. Composites 1992;23(1):2-27. 129. Kelly A, Tyson WR. Tensile Properties of Fibre-Reinforced Metals Copper/Tungsten and Copper/Molybdenum. Journal of the Mechanics and Physics of Solids 1965;13(6):329-350. 130. Kelly A. Interface Effects and Work of Fracture of a Fibrous Composite. Proceedings of the Royal Society of London Series a-Mathematical and Physical Sciences 1970;319(1536):95-116. 131. Lodha P, Netravali AN. Characterization of interfacial and mechanical properties of "green" composites with soy protein isolate and ramie fiber. Journal of Materials Science 2002;37(17):3657-3665. 132. Luo S, Netravali AN. Interfacial and mechanical properties of environmentfriendly "green" composites made from pineapple fibers and poly(hydroxybutyrate-co-valerate) resin. Journal of Materials Science 1999;34(15):3709-3719. 133. Nam S, Netravali AN. Characterization of ramie fiber/soy protein concentrate (SPC) resin interface. Journal of Adhesion Science and Technology 2004;18(9):1063-1076. 134. Tukey JW. Multilayer Significance Procedures - (Preliminary Report). Annals of Mathematical Statistics 1953;24(4):691-692. 135. Sears JK, Darby JR. The Technology of Plasticizers. St. Louis, Missouri: John Wiley and Sons, Inc., 1982. 136. Dickinson E, Euston SR. Stability of Food Emulsions Containing both Protein and Polysaccharide. In: Dickinson E, editor. Food Polymers, Gels and Colloids. England: Royal Society of Chemistry, 1991. p. 132-146. 137. Pisanova EV, Zhandarov SF, Dovgyalo VA. Interfacial Adhesion and Failure Modes in Single Filament Thermoplastic Composites. Polymer Composites 1994;15(2):147-155. 111 138. Phoenix SL, Schwartz P, Robinson HH. Statistics for the Strength and Lifetime in Creep-Rupture of Model Carbon Epoxy Composites. Composites Science and Technology 1988;32(2):81-120. 139. Own SH, Subramanian RV, Saunders SC. A Bimodal Lognormal Model of the Distribution of Strength of Carbon-Fibers - Effects of Electrodeposition of Titanium Di (Dioctyl Pyrophosphate) Oxyacetate. Journal of Materials Science 1986;21(11):3912-3920. 140. Mullin JV, Mazzio VF, Mehan RL. Basic failure mechanisms in advanced composites. NASA Technical Reports 1971;NASw-2093, NASA-CR-124645. 141. Fukuda H, Chou TW. A Probabilistic Theory of the Strength of Short-Fibre Composites with Variable Fiber Length and Orientation. Journal of Materials Science 1982;17(4):1003-1011. 142. Fukuda H, Kawata K. On Young's modulus of short fibre composites. Fibre science and technology 1974;7:207-222. 143. Webber CLI, Bledsoe VK, Bledsoe RE. Kenaf Harvesting and Processing. In: Janick J, Whipkey A, editors. Trends in new crops and new uses. Alexandria, VA: ASHS Press, 2002. p. 340-347. 112