\. 9...: r, «t .. each-n4. 3.2. i u... a, w I, I I? I . r 51:33:53. 3% V .11. 1. 3!. v 15.31:?) n. 2. t! o, 3.53:1...9. 7.31:. .22.. .. :..?.S.¥..f3t. in... . y. it. 2.13:3. L! 2-25.... :h! v.1. .9... nail; i Vk . ‘. L1; .23. cu - iv .. :\ 721.. (w. £.D £32?» l— TE UNIVERSITY LIBRARI S ’imitu’i'iiil‘immm i an it 3 1293 multluzafl This is to certify that the thesis entitled Surface sulfonation of polypropylene resin to improve the mechanical properties of wood fiber/polypropylene composites presented by ‘ Sudawan Supachokouychai has been accepted towards fulfillment of the requirements for MASTER degree in PACKAGING alK/adfi‘é ack R. Giacin \J ajor professor November 27 1995 Q7 £74 Date I _.4440—""\ Dr. Susan E. Selke 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution LIBRARY Michigan State University PLACE II RETURN BOX to romovo thin chockout from your rocord. TO AVOID FINES Mum on or bdoro duo duo. DATE DUE DATE DUE DATE DUE W0 L [fig— DE' 11—!— MSU Io An Affirm-tin Action/EM OpportunIty Institution SURFACE SULFONATION OF POLYPROPYLENE RESIN TO IMPROVE THE MECHANICAL PROPERTIES OF WOOD FIBER/POLYPROPYLENE COMPOSITES BY Sudawan Supachokouychai A THESIS Submitted to Michigan State university in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE School of Packaging 1995 ABSTRACT SURFACE SULFONATION OF POLYPROPYLENE RESIN TO IMPROVE THE MECHANICAL PROPERTIES OF WOOD FIBER/ POLYPROPYLENE COMPOSITES BY Sudawan Supachokouychai The effect of surface sulfonation on enhancing the mechanical properties of wood fiber/surface sulfonated polypropylene composites was investigated. The polypropylene (PP) resin in a powdered form was sulfonated for various reaction times (0, 5, 8, 10, and 20 minutes), and then utilized as a matrix phase in the fabricating composites with wood fibers (40% w/w) . The mechanical properties of the respective composites, to include tensile properties, flexural properties, and Izod inmact strength, were determined. The maximum sulfonation level achieved on the PP resin was considered low, resulting in a minimal enhancement in interfacial interaction between the wood fiber and the polymer matrix phase. While statistically significant enhancement in mechanical properties was observed with increased sulfonation levels, the results were not considered of commercial utility. No change in tensile strength of wood fiber/polypropylene composites was observed following storage up to 9 weeks at 35 °C and 90% RH, even though significant levels of water vapor were sorbed. To my parents, Kongdej and Sompit iii ACKNOWLEDQENTS I would like to express my sincere gratitude and respect to my major advisor, Dr. Jack R. Giacin, for his great guidance, assistance, and encouragement. I would like to thank my co-advisor, Dr. Susan E. Selke, for her valuable advice and comment given on this project through its completion . I am grateful to Dr. Indrek Wichman, Department of Mechanical Engineering, for serving on my committee and for his recomendations . I would like to offer my special thanks to Mike Rich and Brian Rook for their advice and great assistance on the extruder, Instron machine, sulfonation system, and other equipment at the Composite Materials and Structures Center. My thanks are also extended to Bob for his useful help on using the equipment at Packaging Building and his humor, and to Mark Sanderson from Montell, Inc., Lansing, Michigan, supplying polypropylene resin for this study. Finally, I would take this opportunity to thank the Center for Food and Pharmaceutical Packaging Research, for financial support of this project. iv TABLE OF CONTENTS Page LIST OF TABLES....................................... vii LIST OF FIGURES...................................... INTRODUCTION......................................... LITERATURE REVIEW.................................... 1.Composite Materials............................. 1.1 Introduction............... ................. Q U" 01 01 H X 1.2 Prediction of Properties.................... 1.3 Interface and Interphase.................... 19 2.8ulfonation.................... ...... ........... 22 2.1 Introduction.............. ..... ..... ........ 22 2.2 Sulfonation Reactions....................... 24 3.Review of Prior Research........ ..... ........... 28 MATERIALS AND METHODS................................ 38 1.Materials............. ............ . .......... ... 38 1.1 Matrix...................................... 38 1.2 Reinforcing Filler..................... ..... 4O 2.Methods......................................... 42 2.1 Sulfonation Treatment....................... 42 2.2 Sample Preparation.......................... 45 2.3 Mechanical Testing.......................... 48 2.4 water Sorption Studies...................... 51 2.5 Density Measurement......................... 52 2.6 Statistical Analysis............. ........ ... 52 RESULTS AND DISCUSSION............................... 1.8urface Characteristic........................... 2.Density of Composites............................ 3.Tensile Properties..................... ..... ..... 4.Plexural Properties.............................. 5.Izod Impact Strength............................. 6.Water Sorption Studies....... ................ .... SUMMARY AND CONCLUSIONS.............................. RECOMMENDATIONS FOR FURTHER RESEARCH................. APPENDICES................................ ........... APPENDIX A......................................... APPENDIX 3......................................... APPENDIX C......................................... APPENDIX D......................................... APPENDIX E......................................... BIBLIOGRAPHY......................................... vi 53 53 61 63 75 82 87 93 96 98 98 102 105 112 118 151 LIST OF TABLES Table Page 1 General Properties of Pro-fax 6501.............. 40 2 Composition of Composites and Materials by weight Percent.................................. 46 3 Atomic Concentration for ansulfonated and Sulfonated PP Resins by BSCA Analysis........... 55 4 Relative Atomic Ratios of Sulfonated PP Resins.. 55 5 Comparison of Sulfur Content Determined by ESCA Analysis and Elemental Analysis as a Function of Sulfonation Time. ................... 56 Results of Density (g/cc).................. ..... 62 Results of Tensile Strength at Break (MPa)...... 69 Results of Percent Elongation at Break.......... 71 WGQQ Results of Modulus of Elasticity (MPa).......... 73 10 Results of Flexural Strength (MPa).............. 78 11 Results of Flexural Modulus (MPa)........ ....... 80 12 Results of Izod.Impact Strength.(J/m)........... 85 13 Tensile Strength Data (MPa) of Samples Stored under Bumidified Conditions, at 35°C and 90% RH, for 0, 3, 5, 7, and 9 weeks..................... 91 14 Properties of PP Resin Determined by Modulated Differential Scanning Calorimeter............... 100 15 Data of Density (g/cc).......................... 105 16 Data of Tensile Strength at Break (MPa). ........ 106 vii 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Data Data Data Data Data Data of of of of of of Periods DEE! of Percent Elongation at Break.... ......... Modulus of Elasticity (MPa)....... ..... . Flexural Strength (MPa)................. Flexural Modulus (MPa).................. Izod Impact Strength (J/m).............. ‘Weight (gram) Measured at Different of Storage Time......................... Tensile Strength (MPa) Measured After Different Periods of Storage Time............... One-way Analysis of Variance of Density Values.. One-way Analysis of Variance of Tensile Strength at Break Data, for Lengthwise Direction......... One-way Analysis of Variance of Tensile Strength at Break Data, for Crosswise Direction. ...... ... One-way Analysis of variance of Tensile Strength at Break Data in Lengthwise Direction Vs. era-"1.. Dir.ct1°n00000000000......OOOOOOOOO... One-Why Analysis of Variance of Percent Elongation at Break Data, for Lengthwise Dir.cti°n.........OOOOOOOOOOOO0.00.0.0...0....O. One-way Analysis of Variance of Percent Elongation at Break Data, for Crosswise Dir.cti°n00000OOOOOOOOOOOOOOOOOOOO0.00.......... One-way Analysis of Variance of Percent Elongation at Break Data in Lengthwise Direction Vs. Crosswise Direction... ....... ..... One-way Analysis of Variance of Modulus of Elasticity Data, for Lengthwise Direction....... Oneeway Analysis of Variance of Mbdulus of Elasticity Data, for Crosswise Direction........ viii 107 108 109 110 111 112 117 118 119 120 121 123 124 125 127 128 33 34 35 36 37 38 39 40 41 42 43 44 One-way Analysis of Variance of Modulus of Elasticity Data in Lengthwise Direction Vs. Crosswise Direction............................. One-way Analysis of Variance of Flexural Strength Data, for Lengthwise Direction......... One-way Analysis of Variance of Flexural Strength Data, for Crosswise Direction.......... One-way Analysis of Variance of Flexural Strength Data in Lengthwise Direction Vs. Crosswise Direction....................... ...... One-way Analysis of Variance of Flexural Modulus Data, for Lengthwise Direction.......... Oneeway Analysis of Variance of Flexural Modulus Data, for Crosswise Direction..... ...... One-way Analysis of Variance of Flexural Modulus Data in Lengthwise Direction Vs. Crosswise Direction............................. One-way Analysis of Variance of Izod Impact Strength Data, for Lengthwise Direction ......... One4way Analysis of Variance of Izod Impact Strength Data, for Crosswise Direction.......... Oneeway Analysis of Variance of Izod Impact Strength Data in Lengthwise Direction Vs. Crosswise Direction............................. One-way Analysis of Variance for Tensile Strength Data Compared between Conditioned Samples, at a Period of Storage Time ........... . .......... .. One-way Analysis of Variance for Tensile Strength Data Compared between Different Periods of Time, for a Composite Material........................ ix 129 131 132 133 135 136 137 139 140 141 143 LIST OF FIGURES Figure Page 1 Sulfonation Reaction of PE...................... 25 2 Sulfonation Reaction of PP. . . . . . . . . . . ........... 27 3 Repeating Unit of PP Structure. . ................ 39 4 Cellulose Molecule ....... . . . . ..... . . . ........... 41 5 Schematic Diagram of Sulfonation System. . ....... 44 6 Theoretical Molecular Structure of Sulfonated PP 54 7 Atomic Percent Sulfur Concentration of PP Resins As a Function of Sulfonation Time. . . . . . . . 58 8 Tensile Strength at Break (MPa) ......... . . . ..... 70 9 Percent Elongation at Break. .................... 72 10 Modulus of Elasticity (MPa) 74 11 ler‘l strwgth (ma) 0 O O O O O O O O O O O O O ........... 79 12 Flexural Modulus (MPa) . . . . . . . . . . . . .............. 81 13 Izod Impact Strength (J/m) . . . . . . . . . . . . ....... . . . 86 14 Weight Increase (96 w/w) As a Function of Storage Time (days), at 35 °C and 90% RH ........ 89 15 Tensile Strength of Samples Stored under Humidified Conditions, at 35 °C and 90% RH, for 0, 3, 5, 7, and9weeks..................... 92 16 MDSC Curves for PP Resin ........................ 101 17 ESCA Analysis for Nonsulfonated PP Resin. . . . . . . . 102 18 ESCA Analysis for 5-min Sulfonated PP Resin. . . . . 103 19 ESCA Analysis for 8-min Sulfonated PP Resin. . . . . 103 20 ESCA Analysis for 10-min Sulfonated PP Resin. .. . 104 21 ESCA Analysis for 20-min Sulfonated PP Resin.. . . 104 xi INTRODUCTION Composite materials are generally composed of one or more dispersed phases (reinforcing structure) enclosed in a continuous phase or matrix and are classified as particulate or fibrous, based on the geometry of the dispersed phase. The type of reinforcing material is very important, since the properties of the composite are strongly related to the properties and quantities of the components, as well as their chemical and physical interactions. The reinforcing agent should provide maximum improvement of desired physical properties, be inexpensive and readily available, have good dispersion and wetting characteristics, and be available in controlled particle sizes, among other desired requirements. Wood fibers, as a reinforcing filler for thermoplastic composites, have gained a significant amount of attention because of the many advantages they offer. Apart from their relatively low cost, such fillers have low density, low equipment abrasion, no health hazard, high strength-to- weight ratio and are easily renewable. The filler being investigated in this study, therefore, is Aspen Hardwood Fiber. In considering polymer/wood fiber composites, the main drawback involves the hydrophilic character of the wood- based filler surface, adversely affecting the interfacial interaction with the hydrophobic polymer phase. To date, a number of studies have focused on the inclusion of dispersants and coupling agents with wood fiber/plastic composites, and the effectiveness of the additives in enhancing the interfacial interaction between the dispersed and matrix phases and thus, the mechanical properties of the composites. The polymer matrices investigated include: (1) high density polyethylene, (ii) polypropylene, and (iii) a recycled mul ti - layer polypropyl ene/adhesive/ ethylene -vinyl alcohol copolymer container resin. The inclusion of modifiers in high density polyethylene based composites was found to enhance the mechanical properties of the resultant composites by improving fiber/polymer matrix adhesion (Selke et al., 1989 and Childress, 1991). Two additives which showed promising resultants were maleic anhydride modified polypropylene, and ionomer modified polyethylene. The multi- component composite (iii) was found to have properties superior to those of a composite formed with polypropylene alone (Simpson, 1991) . This was thought to be due to improved fiber adhesion, resulting from the polar functionality of the adhesive and ethylene-vinyl alcohol copolymer components. Sulfonation chemistry offers a new approach to chemically and structurally modifying the surface of polymers (Walles, 1989; Walles, 1973; and Walles, 1971). Since the sulfonation process introduces sulfonate groups along the polymer backbone, through a displacement reaction with hydrogen atoms , virtually any polymer , except for f luoro- chloropolymers and some silicones, can be sulfonated. Further, the sulfonation process itself is not surface limited, i.e. the process can be extended under diffusion control below the surface up to depths of a micron or more. Thus, modification of not only the surface but the surface region is possible. In principle, this makes it possible to modify the surface of polymers, independent of their chemical composition, and can be applied to wood fiber/ polymer composites, resulting in enhanced compatibility and a concurrent improvement in adhesion and therefore mechanical properties. Baraguchi (1993) evaluated the effect of surface sulfonation of high density polyethylene (BDPE) on the mechanical properties of BDPE/wood fiber composites, including tensile, flexural and impact properties. It was found that a longer exposure time in sulfonation of HDPE resin and an increased surface area of the resin (i.e. powder form) resulted in an increased level of sulfonation. The extent of sulfonation achieved, however, was quite low and did not modify the dispersive and polar characteristics of the polymer to a level which resulted in enhanced interfacial interaction between the EDPE and wood fiber with a concurrent increase in mechanical properties. Consequently, further studies were proposed by Haraguchi, designed to increase the sulfonation level of HDPE with a corresponding modification of surface energies of the polymeric matrix. In contrast to polyethylene, the sulfonation of polypropylene was found to readily modify the surface energy properties of the polymer surface, since the presence of tertiary carbons on the polymer molecule provides active sites for 803 insertion (Wangwiwatsilp, 1993) . In the current study, surface sulfonation of polypropylene (PP) is being carried out in order to determine the effect of sulfonation on the chemical structure of the polymer surface region, and its effect on the mechanical properties of wood fiber/surface sulfonated PP composites. Therefore, the primary objectives of the study include: 1) Determination of the density and distribution of sulfonate groups on the surface of polypropylene following surface sulfonation. 2) Determination of the effect of sulfonate group concentration on the mechanical properties of wood fiber/ surface sulfonated polypropylene composites. LITERATURE REVIEW 1. Composite Materials 1.1. Introduction Several definitions of composites have been given in the literature. Since the term 'composite' refers to something made up of two or more distinct parts, a material having two or more distinct constituent materials or phases may be considered a composite material (Agarwal and Broutman, 1980). Composites may be separated into two basic forms, namely: (1) composite materials: and (2) composite structures. The latter are characterized by a discontinuous matrix, i.e. sandwich structures and coated materials, whereas composite materials are comprised of a dispersed filler embedded 'within a continuous matrix (Richardson, 1987). Furthermore, the composite concept can be related to either the mdcroscale or macroscale (Richardson, 1977). The microscopic composite materials may not be generally regarded as composites in a strict application. In this study, a composite material has been defined as a macroscopic combination of two or more materials, as separate phases and combined to form desired structures so as to take advantage of certain desirable properties of each component (Grayson, 1983). For instance, fibers as a discontinuous phase are embedded within a continuous phase in fibrous composites. Typically the discontinuous phase has higher strength and stiffness than the continuous (matrix) phase does. There generally must be a substantial volume fraction («40% or more) of discontinuous phase (such as fibers) in order to provide reinforcement. The small cross sections potentially minimize flaws on the fiber structure, and thus fibers display much higher strength along their length than the bulk material. In order for fibers to be widely used for structural or nonstructural purposes and due to their small cross section, they need a binding material such as a matrix. Two main functions of a matrix are: (1) to bind the reinforcements and hold them in place; and (2) to deform and distribute the stress to the fibrous constituents under an applied load (Schwartz, 1992). The matrix also serves to separate fibers from contacting each other. Since the fibers are likely to be brittle, the matrix additionally serves to protect the fiber surfaces against abrasion or environmental attack, both of which can lead to fracture. In composites, therefore, the strong and stiff reinforcing fibers contribute high tensile and flexural properties. On the other hand, the defamation of the matrix at crack tips absorbs energy and reduces stress concentration (Schwartz, 1992) . Composites can be classified into two groups: Particulate Composites and Fibrous Composites, based on the shape of the discontinuous phase (Agarwal and Broutman, 1980) . Reinforcements in particulate composites are in the form of 'particles', which can be in various shapes, such as spheres, rods, flakes, and irregular shapes, with approximately equal dimensions. Fibrous composites, or so- called fiber-reinforced composites, are reinforced with the reinforcement fibers having a length much greater than their cross-sectional axes. Because of the shape of the reinforcing phase, fibers effectively improve fracture resistance of the matrix. In particulate composites, the particles are normally added for cost reduction, rather than for reinforcement, or they may be used for other purposes, such as for reducing shrinkage, for increasing surface hardness, etc. The relative hardness of the particles places constraints on the matrix deformation between the particles and the matrix, thereby improving the stiffness of the composites, but not potentially the strength (Agarwal and Broutman, 1980). On the other hand, composites with reinforcing fibers gain both strength and stiffness. Two forms of the reinforcing fibers used for fibrous composites are continuous (or long) fibers and discontinuous (or short) fibers. Continuous-fiber composites, containing long fibers, are very strong in the direction of the fiber axis but are weak in the transverse direction. Continuous fibers bear stress equally at all points along their length and are primarily the load-bearing component in the load direction (Agarwal and Broutman, 1980) . For short or discontinuous fiber composites, fibers are classified as having an aspect (length-to-diameter) ratio between 10-1000 (Richardson, 1977). Besides the fiber strength, the length of the fibers also greatly affects the mechanical properties of the composites. Further, the transmission of stress imposed on the matrix to the fibers via interfacial interaction becomes crucial. If improved bonding at the interface is achieved, for a given system, the contribution of fibers to the composite mechanical properties can be maximized. Therefore, the properties of both the matrix and fibers, as well as the fiber-matrix interface, are very important to the composite properties. 1.2 Prediction of Properties Theoretical models for predicting the mechanical properties of composites are useful and very convenient, where multiple variables, for instance, are studied and experimental approaches have 1 imi tations , i . e . time and cos t . Nevertheless, in some cases, corrections for the derived equations are required because the models are based on various assumptions, all of which are rarely met in actual circumstances . For simplicity, consider a composite system as having two components: a matrix material and continuous fibers that are uniformly dispersed in the matrix. It is assumed that there is a perfect interfacial bonding between the matrix and the fibers, so that both the matrix and fibers will deform equally under a longitudinal load applied to the composite system. According to the rule of mixtures, the tensile strength and modulus of composites can be estimated from equations (1.2.1) and (1.2.2): oc=chf +0-me (1.2.1) ‘where: o = tensile strength V a volume fraction in which subscripts c, fand m refer to the composite material, fiber and matrix, respectively (Agarwal and Broutman, 1980). It is generally assumed that the failure strain of fibers is less than that of a matrix. Under the longitudinal load, failure initiates when the fibers are strained to their fracture strain (Agarwal and Broutman, 1980). As the fiber elements run from end to end in the composite parts, the fiber strength is directly responsible for the strength of the composite. Ec=Efo+Eme (1.2.2) where: E' = elastic modulus (Agarwal and Broutman, 1980) This equation indicates that under loading conditions, the load will be distributed over the matrix and fibers in proportion to their relative cross-sectional areas and elastic moduli (Richardson, 1977). To be able to use high fiber strength proficiently, a larger ratio of fiber modulus to modulus of matrix should be obtained. This is because a higher proportion of the applied 10 load to the composite can be carried by the fiber phase. The increased volume fraction of fibers will improve the composite properties as well, but only within the scope in which the system can retain a very good bonding between the components (Agarwal and Broutman, 1980) . For some applications, it is advantageous to use short-fiber or discontinuous - reinforced composites , instead of continuous-reinforced composites. In this case, the strength of randomly discontinuous-fiber composites is equal in both longitudinal and transverse directions, and usually the short-fiber composites are produced by a cheaper and faster process. However, a sacrifice of a certain level of mechanical performance is inevitable. The elastic modulus of the randomly short-fiber composites, Em, can be estimated using the empirical equation as shown by: Emdom=§EL+gET (1.2.3) where : E L = longitudinal modulus E, a transverse modulus (Agarwal and Broutman, 1980) The longitudinal and transverse moduli must be obtained from the aligned, short-fiber composites having the same fiber aspect ratio and the same fiber volume fraction (Zadorecki et al., 1986). Both moduli of the aligned, short-fiber composites can be predicted from Halpin-Tsai equations, as shown by the following expressions. It is believed that the 11 predictions of these equations are quite accurate, unless the volume fraction of fibers is close to 1 (Agarwal and Broutman, 1980) . E, _1+(21/d)nLV, __4_ (1.2.4) Em l‘IILV/ 1+2 V fl=——k’— (1.2.5) E," l—nTVf where: E /E —1 11L: ( I M) (1.2.6) (El/Em)+2(I/d) E /E -—l -( ’ ...) (1.2.7) T" ’ (E,/E,,)+2 (Agarwal and Broutman, 1980) These Halpin-Tsai equations suggest that the aspect ratio (l/d) will have a significant effect on the longitudinal modulus, but not on the transverse modulus of the aligned, short-fiber composite. Further, it is predicted that the transverse modulus of either short-fiber composites or continuous-fiber composites is the same value. The moduli in both directions, however, are influenced by the fiber volume fraction and the modulus ratio of the fibers and the matrix (Agarwal and Broutman, 1980) . For composites, in addition to the matrix binding and holding the reinforcing fibers and protecting the fibers from handling and environmental hazards, the function of the matrix is also to convey the load to the fibers through the 12 from handling and environmental hazards, the function of the matrix is also to convey the load to the fibers through the fiber ends and small fiber length near the ends. In the case of longwfiber composites, the and effects can be neglected because of their much greater length over the length of which the fibers allow the transfer of stress. In contrast, for the short-fiber composites the end effects must be taken into consideration. The stress transfer for the short fibers is not uniform along the fiber length, in that the fiber ends insignificantly bear stresses, but the stresses acting on the fibers gradually build up while moving from the fiber ends, with the maximum.value at the middle of fiber length. Concurrently, there is the variation of shear stress along the fiber length in the opposite manner (Richardson, 1977). The mechanism. of stress transfer in composites can. be understood by analyzing the force equilibrium of a small element of fiber as follows: (1tr2 )o f + (21rdz)'r = (nr2)(o f + do I) or %=% (1.2.8) where: 07. 2 fiber stress in the axial direction I = interfacial shear stress on the cylindrical fiber-matrix interface r = fiber radius and at = infinitesimal length of fiber (Agarwal and Broutman, 1980) 13 The above relationship suggests that for a fiber of uniform radius, fiber stress will increase with the rate proportional to the interfacial shear stress. This can be integrated to obtain the fiber stress on a cross-sectional distance 2 away from the fiber end: 2 : of=ofo+;Iordz (1.2.9) where: c:f a stress on the fiber and 0 (Agarwal and Broutman, 1980) Regarding the manner of stress distribution, load transfer from matrix to fiber will be attained only if the length of fibers is longer than a certain value, called a load- transfer length. The critical fiber length, 16, considered to be the maximum value of load—transfer length, is the minimum fiber length in which the ultimate strength of fibers, 0',“ , can be achieved, and even the fiber over this length merely supports a stress up to the maximum fiber stress (Agarwal and Broutman, 1980). The critical fiber length can be given by: Q 2’ (1.2.10) a. leaN N d ‘< where: 1: - matrix yield stress in shear . fiber aspect ratio a|~ (Agarwal and Broutman, 1980) 14 The effect of the fiber ends is pronounced and becomes more important with a decrease in the fiber aspect ratio. The fiber reinforcing efficiency will be reduced as the length of fibers decreases. This is because a larger proportion of the total fiber length is not fully loaded (Hull, 1981). Since the stress on the ends of the fibers is always under the maximum fiber stress, the average fiber stress at the critical length will be only (Sm/2. This clearly shows that a much longer fiber length than the critical value will be needed in order for the load-bearing ability of the short- fiber composite to approach that of the continuous composites (Richardson, 1977). The and effects, therefore, result in the lowering of the elastic modulus and strength of short-fiber composites (Agarwal and Broutman, 1980) . In an estimate of the strength of aligned, short-fiber composites, the equation (1.2.1) must be modified. The average fiber stress will replace the tensile strength of the fiber. However, when the fiber length is longer than the critical length, the average fiber stress value becomes close to the maximum fiber stress (Agarwal and Broutman, 1980). The length of fibers in relation to the critical length can affect the fracture characteristics of the composites. The composite strength, as a function of the fiber length, is given as follows: 01] Ty] o =_d—Vf+cmuVln , Ilc (1.2.12) c5cu =oqu/ +omem , I>>lc (1.2.13) where: <3cu a composite ultimate strength ofi, a fiber ultimate strength can = matrix ultimate strength 0' = matrix stress at the fiber fracture strain (Agarwal and Broutman, 1980) These equations define three possible modes of failure, which affect the evaluation of the ultimate strength of short-fiber composites, depending on the fiber length. In the first case, when the fiber length is smaller than the critical length, the composite fracture is governed by the failure of the matrix or at the interface, even with the large size of the applied stress. This is because the maximum fiber stress is lower than the average fiber strength. Secondly, when the fiber is longer than the critical length, the fibers may be loaded to their average strength. For this case, fiber failure will take place when the fiber stress is equal to the ultimate strength of the fibers. For the third case, ‘where the fiber length. is much. greater than the critical length, the behavior of short-fiber composites becomes very similar to that of continuous-fiber composites (Agarwal and Broutman, 1980). In the last two cases, the 16 (Agarwal and Broutman, 1980) . In the last two cases, the fibers are responsible for the failure of the composites. However, equations (1.2.12) and (1.2.13) are valid only if the volume fraction of fibers exceeds a certain minimum value. Otherwise the matrix, instead of the fibers, will support the entire load, even when all the fibers are broken . The fiber ends may induce the building-up of stress. Even at a small load, the presence of stress buildup or stress concentration can result in the separation of fiber ends from the matrix, thereby producing a micro-crack in the matrix. One micro-crack at the fiber end may eventuate several adverse effects. By the propagation of the cracks along the fiber length, the shear stress at the interface may lead to fiber debonding and their separation from the remaining composite. In the other case, the micro-cracks near fiber ends could propagate in the cross direction to other fibers, resulting in immediate composite failure (Agarwal and Broutman, 1980) . During fabrication processes of composites, residual stresses may be inherently built up in the constituents and interface (Agarwal and Broutman, 1980), and the processing can induce fiber breakage (Bigg, 1985) . Either residual stresses or fiber breakage will affect the composite strength . Other parameters affecting the composites' mechanical properties include: (i) the strength of the matrix; (ii) 17 fiber orientation and distribution; and (iii) fiber-matrix interface bonding . The fiber orientation evidently influences the load distribution between fibers and matrix. The degree of off-axis fiber angle results in the reduction of the composite modulus, to some extent, and the tensile strength, to a greater extent (Lee, 1991). To maximize the mechanical properties, the fibers in composites must be parallel to the loading axis. In reality, it is very difficult to control the fiber alignments during fabricating composites, especially with high fiber loading of short fibers. The fibers should also be uniformly dispersed in the matrix. Poor fiber distribution may be a result of the close packing of fibers as well as the limitation of fiber wetting-out by the matrix. Furthermore, the incomplete wetting-out of fibers by the matrix or the presence of volatiles produced during the melt process can lead to the creation of void content (Bull, 1981). Less than 1 96 voids is preferred in a good composite (Agarwal and Broutman, 1980) . The interfacial bonding between fiber and matrix phases is far more important to the behavior of short-fiber composites, compared to that of continuous-fiber composites. Since the fibers are not loaded directly under applied stresses, the stresses are transmitted from the matrix to the fibers through the interface, which contributes to a major portion of the composite strength. The stress transfer mechanism will be less efficient with poor interfacial bonding . 18 Moreover, the interfacial conditions govern the mode of micro-cracks at the fiber surface. The cracks do not propagate along the fiber length within a well-bonded system. The reinforcement efficiency still remains, even with several points of fiber breakage. Besides, a well- bonded interface is a prime factor in achieving the high transverse strength and good environmental performance of composites (Agarwal and Broutman, 1980) . However, the enhancement of the composite modulus is much less influenced by the interfacial strength (Bigg et al., 1988; and Lee, 1991) . Concerning impact properties, there is no well-developed theoretical relationship for this prediction (Bigg et al., 1988 and Lee, 1991) . Although the impact test is widely used as a means of measuring material toughness, the acquired numerical data are significant for qualitative analysis. For example, their use is in quality control, rather than in a quantitative way, such as the use in engineering design (Richardson, 1977). The major factors affecting the impact strength of the composite materials and unreinforced polymers are the testing procedure, rate of impact, shape of the impacting implement, degree and form of crystallinity, and the existence of microdefects in the vicinity of the impact. The impact strength is dependent on the fiber orientation, fiber aspect ratio and interfacial adhesion as well (Bigg et al., 1988). 19 1.3 Interface and Interphase The interface is generally referred to as the interfacial region, or interphase, of the composite system, owing to the difference to some degree from its bulk properties. It is well-known that the characteristics of the interface play a key role in the mechanical performance of the composite materials. As aforementioned, the interface, which is responsible for the stress transmission from matrix to fibers, is a vital contributor to the composite properties. In addition, the interfacial strength is essential for the enhanced environmental performance of the composite. The strength of interfacial bonds must be at least equivalent to that of the matrix, particularly under loading conditions. Otherwise, composite failure such as fracture and delamination at the interface will take place. The quality of the bond is also responsible for the long-term stability of the composite, such as fatigue properties and resistance to hot-wet conditions (Schwartz, 1992) . In the combination of two dissimilar components, the degree of interfacial adhesion between the composite elements may differ from strong chemical bonding to weak frictional forces. This is a design variable which can be regulated by utilizing one or more of the following techniques: 1) modification of fiber; 2) modification of matrix: and 3) inclusion of interfacial-aided additives (Krishnan and Narayan, 1992) . 20 There are five possible mechanisms of adhesion which can occur, either by themselves or in combination, at the fiber- matrix interface of a composite material: (1) mechanical adhesion, (2) adsorption and wetting, (3) interdiffusion, (4) electrostatic attraction, and (5) chemical bonding (Hull, 1981). (1) Mechanical Adhesion. The intimate contact of two surfaces results ' in a mechanical interlocking between the two surfaces. For good mechanical adhesion it is necessary that a molten resin thoroughly wets a rugged fiber surface. The degree of surface irregularity will affect the strength at the interface, since these contact areas increase as the extent of irregular surfaces into which the liquid can penetrate, increases. (2) Adsorption and wetting. A physical attraction, i.e. Van der Waals forces, occurs as a result of the wetting of every pore of solid surfaces by liquids. Strong bonds can be achieved if the entrapped air/ gases or impurities at the fiber surface are avoided. (3) Interdiffusion. A bond is formed between two surfaces by molecular entanglement. The presence of solvents and plasticizers can promote the bonding, while the extent of diffusion and the number of molecules involved will designate the strength of the bond. (4) Electrostatic attraction. The electrostatic forces, i.e. ionic bonding, take place when the contact is presented 21 between two surfaces carrying oppositely charged ions. The strength of the bonds is governed by the charge density. This attraction may aid in the coupling effect, but it is unlikely to be a major contribution to the bond strength of the composite materials. (5) Chemical Bonding. A chemical bond is formed across the interface as a result of the presence of compatible chemical groups on the fiber surface and the matrix surface. The use of coupling agents on glass fibers is one of the examples of reliance on this mechanism, 22 2. Sulfonation 2.1 Introduction The surface composition of polymers can be chemically or mechanically modified to provide new properties, such as surface adhesion, wettability and printability, for the respective polymers. Among various techniques for surface modification, sulfonation provides a series of desirable characteristics, especially being a well-controlled, reproducible process, that is attractive either to laboratory research or to industrial practice (Asthana, 1993) . It has been reported that the sulfonation process creates sulfonates, by introducing polar groups of sulfonic acid on the polymer backbone, with carbon-sulfur bonding between them. Sulfonating reagents in use include, sulfur trioxide (SO3) in the form of gas and liquid, as well as a variety of SO3 complexes, (i.e. pyridine, trimethylamine, trioxane, dioxane, trialkylphosphate) and oleum. Oleum, or fuming sulfuric acid, is among the most popular reagents in commercial use (Gilbert, 1965). Typically, a sulfonated substrate can be any polymer containing either carbon-hydrogen bonds or nitrogen-hydrogen bonds, with exception of pure fluorocarbons and some silicones (Walles, 1989) . Thus, the common engineering polymers such as polyethylene and polypropylene, having low surface energy or non-polar nature, can be treated via a sulfonation process for tailored surface properties. The surface layer of a sulfonated polymer substrate is modified 23 to behave differently from its bulk composition due to the presence of sulfonate groups. This modification results in observed changes in surface properties. Surface sulfonation of polymers has shown its utility in enhancing several surface physicochemical properties, such as adhesion, wettability, barrier properties, dyeability, abrasion resistance, electrical conductance and metallization (Erickson, 1993). For example, the organic vapor permeability of fuel tanks can be considerably reduced by exposing the inner surface of the containers to 803 gas under controlled conditions, with subsequent neutralization ‘with N33 gas (walles, 1989). The findings from wangwiwatsilp (1993) also showed a reduction in the permeability coeffiecient of ethyl acetate and toluene through sulfonated polypropylene films. It was reported by Park (1993) that the surface sulfonation afforded an increase in the polar component of the surface energy and the peel adhesion strength of a polypropylene film. In addition, the study of Fonseda et al.(1985) indicated that the direct sulfonation of polyethylene successfully increased the surface conductivity, the microindentation hardness and the critical surface tension of this polymer. The reported applications of the sulfonation process are at the manufacturing level, including the manufacture of detergents (surface active agent), dye intermediates, ion- exchange resins, sulfonated oils, and other sulfonates of industrial interest (Gilbert, 1965). 24 Although most of the reaction occurs on the polymer surface, sulfonation could extend beyond the surface to the bulk thickness. Walles (1989) indicated that when NH3 gas was used in the neutralization step following the $03 gas phase sulfonation, the penetration of sulfonate (i.e.-SOfNHfl) groups was found to a depth of 20-25 mdcrometers. Further, it was found that the extent of sulfonation, as well as the depth of surface modification achieved, is apt to be manipulated by the concentration of $03 and time of exposure. These two sulfonating parameters are inversely related to each other, so that their combination can be varied to suit a specific purpose (Walles, 1973). 2.2 Sulfonation Reactions Sulfur trioxide has an amphoteric characteristic due to the strongly electron hydrophilic nature of the sulfur atom and the electron-rich nature of the oxygen atom. This amphoteric character explains its behavior as a sulfonating agent, with the atomic sulfur attacking electron-rich (basic) systems, and the atomic oxygen accepting acidic protons. The nature of the sulfonation process is fairly complicated, with a number of studies reported in the literature to elucidate the mechanism of polymer sulfonation. Olsen and Osteraas (1969) studied sulfonated polyethylene (PE) surfaces employing an infrared spectrophotometric technique. The results confirmed the insertion of atomic sulfur on the polyethylene surface, as the presence of sulfonic acid groups on the PE surface was found following sulfonation. It was reported by Walles (1973), that virtually all sulfur 25 atoms were in the form -C-SO3H (alkane sulfonic acid) groups, with much less degree of -C-O-SO3H (hydrogen sulfate) for the sulfonated PE structure. In the study of Ihata (1988), the spectra obtained by infrared, resonance Raman and UV-VIS spectroscopy showed the formation of sulfonic acid functionality during the sulfonation reaction, in accordance with other research. Further, this investigation revealed the possible sulfonation mechanism of polyethylene with gaseous 803 as shown in Figure 1. Firstly, a hydrogen atom on the polymer chain is removed by $03 to give a PE radical. The subsequent mechanism could be either the reaction of PE molecular species with 803 to form sulfonic acid groups, or the removal of sulfurous acid to generate polyene. As the reaction proceeds, longer conjugated polymer chains with sulfonic acid functionality were formed. 'cna'cna-cna' fl' 'cnz'cnz-$B- —'——* -CB,-CH,-(|:H- (1) so,s so,s - -cs=cs- 2 'E C32 I) Figure 1: Sulfonation reaction of PE (Ihata, 1988) 26 In addition, a color change of PE was found as a result of the sulfonation process. Color varied from pale green to dark brown, as the extent of sulfonation increased. A similar result of color change on PE surfaces was reported by Walles (1973) who proposed that the color was induced by various substances resulting from oxidation reactions occurring during the surface modification. For the sulfonation of polypropylene, Asthana (1993) proposed that, like polyethylene, the reaction mechanism involved the insertion of 803 at active sites to yield sulfonic acid functionality, which was followed by the elimination of sulfurous acid groups. Regarding the polypropylene structure, the tertiary carbon atoms of the polymer are considered as tentative active sites for $03 insertion, owing to their high electron density, which would be favored by the hydrophilic reaction with $03. The resulting sulfonic acid groups, however, are likely to desulfonate by the removal of sulfurous acid. The sulfonation reaction of polypropylene is shown in Figure 2. The desulfonation reaction may continue yielding a final product, or intermediates of desulfonation may pursue new reactions. Consequently, it is a complex reaction and various products can be produced, i.e. alkene sulfonic acid, C-sultone, D-sultone, ketones, etc., during the reaction (Asthana, 1993) . 27 cs ia’ so, I ’ -s,so, ’ - s-cs,- ———> —c|:-cs,— : -c=cs- use, (A) on, on, cs, cs, cs, I so, I I -s,so, I I —C=CH- -——> —c=cs-c-cs,— ——-> -c=cs-c=cs- (A) so, Figure 2: Sulfonation reaction of PP (Asthana, 1993) Following sulfonation, it is essential that the 803 groups on the polymer skeleton are neutralized to form a stable species. A. variety' of neutralizing agents can. be ‘used, including aqueous ammonium hydroxide (NH4OH) , ammonia gas (N33), sodium hydroxide (NaOH), etc. If NH4OH is utilized, the hydrogen atom of the sulfonic group will be substituted by a NHp‘ ion (-C-SO3'N'H4+) which is a more stable form. Certain properties, such as barrier properties, can be tailored through the selection of neutralizing agent. This is the result of the nature of the counterion influencing the barrier properties of the sulfonated layer. For example, a substantial decrease in oxygen permeability of sulfonated HDPE was achieved after the cationic exchange of my to Lithium. (Li+) or Sodium. (Na+) ion. Among common metal cations, Li, Na, Cu, Mg, Sr, V, Mn, Co, and Ni can give effective barrier results (walles, 1989). 28 3. Review of Prior Research A number of literature references have focused on the fabrication and properties of cellulose-based composites, particularly with polyolefins. Specific attention has been directed to developing techniques for the improvement of cellulosic composite performance, due to its clear opportunity in competition with other composites, i.e. glass fiber-thermoplastic composites. The following is a review of prior research related to this field. Elason, Kubat and Stromvall (1984) studied the efficiency of cellulosic fillers in common thermoplastics, which included high density polyethylene (HDPE) , polypropylene (PP), normal and impacted-modified polystyrene (PS, SB), and polyamide 6 and 12 (PA6, PA12). The fillers used were wood flour (white spruce), cellulose flour (bleached sulphate, pine) and cellulose fiber (bleached sulphate, birch). Overall, the modulus was improved with increased filler content, whereas the strain at yield and at rupture, and the charpy impact resistance diminished substantially. The tensile stress at yield and at break of HDPE and SB composites were increased to some extent, whereas the others were relatively unaffected. The polyamide and polystyrene composites exhibited unfavorable characteristics, i.e. degradation and discoloration, when being processed. Only polyolefin composites were able to achieve 70% filler content. Two compounding procedures were evaluated, the first by using a kneader, and the second involved a single-screw extruder. The kneader compounding process provided for better homogeneity, and, therefore, higher impact strength 29 resulted. The dimensional stability was also investigated. Of the filled polymers, a decrease in mold shrinkage was displayed, with an increase of filler content. As a filler, cellulose fibers did not exhibit a significant contribution in reinforcement of the resultant composites, despite their strength potential. This was attributed to fiber breakage during the fabrication step and poor adhesion at the polymer-wood fiber interface. Using a compatibilizer to promote the interfacial adhesion in biofiber-polypropylene composites and its effect on mechanical properties were investigated by Krishnan and Narayan (1992) . They used low-density hardwood residue mixed with ground pecan shells (LDHW) as a biobased component for the composites. Two fiber-content levels of 20% and 30% were fabricated. The compatibilizer employed was the graft copolymer of maleated polypropylene with cellulosic parts. Two different processes used in modification were A- and B- modified processes. In the A-modified process, the blend of polypropylene (PP), maleic anhydride (MA) and dicummyl peroxide (DCP) was extruded and pelletized, to produce maleated polypropylene (MAPP) . Then, MAPP, maleic anhydride and LDHW were compounded in the presence of the catalyst. For the B-process, a single run of the PP, maleic anhydride, dicummyl peroxide, LDHW and the catalyst was produced. Both blends with compatibilization showed improved tensile properties over the unmodified composites. Yet, in contrast to composites fabricated by the A-process, the B-modified materials exhibited a lower percent elongation than the virgin polymer. This was thought to be due to crosslink 30 formation by simultaneous reaction of the maleic anhydride with lignocellulosic polymer and polypropylene. Raj and Kokta (1991) examined the use of silane coupling in silane-coated wood fiber/linear low density polyethylene (LLDPE) composites. Three silane coupling agents used were vinyltri (2-methoxy ethoxy) silane (silane A-172), 'y- methacryloxypropyltrimethoxysilane (silane A-174), and 'y- aminopropyltriethoxysilane (silane A-1100). The wood fibers were coated with the different coupling agents before mixing with LLDPE. Loss of tensile strength and increase of modulus were found in the cosposite having untreated wood fiber, with an increase of filler concentration. On the other hand, the use of both silane A-172 and A-l74 resulted in a considerable increase in the mechanical properties of the resultant composites, as compared to the untreated wood fiber composites. The tensile strength was increased with the addition of filler content. The effectiveness of the silane coupling agents to improve bonding at the fiber- polymer interface was evident even at low concentration (<2% by weight of fiber). The authors also indicated the significance in the choice of an initiator (peroxide), which can aid in chemical bond formation between the cellulosic polymer and the coupling agents. The morphology of the polymer-wood fiber composites, as indicated by scanning electron microscopy (SEM), confirmed the above results, where poor dispersion and adhesion of wood fiber with the matrix in the untreated fiber composites, and good bonding at fiber-matrix interface in the pretreated wood composites were shown. Among the three coupling agents evaluated in 31 this study, the silane A-174 performed the best, followed by the silane A-172. This was attributed to their respective functional groups. Raj, Rokta and Daneault (1989) reviewed the effect of fiber treatment on mechanical properties of polypropylene-wood fiber compossite. Two commercial pulps (Tempure 626 and Temalfa-A 6816), and chemithermomechanical aspen were used as reinforcing agents. Various treatments were applied to the fibers. Composites of the fibers with pretreated silane coupling agents (Silane A-172 and A-174) produced decreased tensile strength and elongation. Unlike silane treated composites, the fiber coated with maleated polypropylene, polymethylene polyphonylene isocyanate (PMPPIC) and polypropylene combination gave better tensile strength and modulus with an increased fiber level (0 to 40%). This was attributed to the reactivity of isocyanate with wood fibers in the coated fiber composites. In this study, there was no substantial effect of the initiators, dicummyl peroxide or cummine hydroperoxide (2%) , in promoting bonds at the fiber- matrix interface. It was also found that the shorter fibers with 60-mesh size favorably compared to 20-mesh fibers in the coated fiber composites. Gatenholm et al.(1992) also described studies involving cellulose fiber/polymer composites. The authors found that with an increased fiber loading, the use of maleic anhydride-modified polypropylene (MAPP) as a coupling agent in cellulose-polypropylene composites influenced dramatically an improvement in the material strength. A 32 measurable increase in the ductility was also reported. Different degrees of adhesion in composites were exhibited by the fracture surface images. The outcome from FTIR also revealed the presence of bonding between cellulose and MAPP. Studies involving varying the molecular weight of the coupling agent indicated that the greater the molecular weight of MAPP, the better interfacial adhesion, or the higher tensile strength of the composites. In addition, a prehydrolytic treatment of fibers, resulting in reduction of fiber size, was presented to promote processability and homogeneity of the composite. Lastly, Gatenholm et al. (1992) proposed the use of polyhydroxylbutyrate (PBS) in cellulose composites as a biodegradable composite having good mechanical properties. The structure-property relationships of polypropylene/wood fiber composites were studied by Sain, Rokta and Imbert (1994). Three types of wood fibers were evaluated, including: (i) chemithermomechanical pulp (CTMP); (ii) sawdust: and (iii) explosion pulp. The mechanical and thermal properties, in relation to the morphological structure, were investigated in unmodified and modified composites. The interface modifiers, including maleic anhydride, itaconic anhydride and m-phenylene bismaleimide, were pretreated on PP, and m-phenylene bismaleimide was pretreated on CTMP. The cause of deterioration of mechanical properties with higher fiber loading was indicated by the morphology of the respective composites. Phase separation of fiber and matrix with fiber agglomeration in highly fiber filled, unmodified composites indicated poor dispersion and 33 the presence of micropores. Treatments of maleated PP (MPP), itaconic anhydride modified PP (ITPP), and bismaleimide modified CTMP (BCTMP) were found to improve the mechanical properties, such as strength and toughness of the wood-PP mixture, as compared to those of unmodified composites and unfilled PP. The modified mixtures also exhibited better thermal properties (i.e. decomposition temperature and melting point) and more uniformity in structure over the unmodified counterparts. This was thought to be due to lowering of the interfacial surface energy with the use of modifiers. Recycled polyolefins (95% PE and 5% PP) utilized as a matrix in wood filled composites, were considered by Chtourou, Riedl and Ait-Kadi (1992). The wood fibers in the form of chemithermomechanical pulp (CTMP) consisted of 45% Spruce, 45% Fir and 10% Poplar. The compounding materials were formed by compression and injection molding. Tensile strength at yield and modulus of the composites showed improvement as a function of non-treated fiber content (0 to 30%). The properties of unmodified composites at 10% weight of fiber were compared to those of the acetic anhydride (AA) and phenol-formaldehyde (PF) treated fiber composites, of the same fiber loading. The authors concluded that both fiber treatments generally increased tensile properties, particularly the modulus of the respective composites. This was attributed to the inproved interfacial interaction by the treatment of AA or PF. The extent of the treatment was also of interest. The greatest enhancement of the properties was found at about 12% AA content (or at 12% PF). Both 34 treatments in compression molding gave more favorable properties as compared to those of the same treatments in injection molding, due to higher deformation rates of fibers in injection molding. Furthermore, samples stored under humidified conditions showed less water uptake and relatively high tensile properties for the treated fiber composites, in comarison with the nontreated composite structures. The effect of extreme storage conditions on the mechanical properties of linear low density polyethylene composites was investigated by Kokta, Daneault and Beshay (1986) . The fillers employed in the composites included: (i) grafted chemithermomechanical pulp (CTMP) of aspen: (ii) wood flour; (iii) mica: and (iv) glass fiber. Their mechanical properties, including: absorbed energy (area under stress- strain curve), secant modulus, tensile strength and strain at yield were measured. The four different storage conditions were (1) room temperature; (2) boiling water; (3) 105 °C ; and (4) -40 °C. The mechanical properties of the grafted aspen fiber-filled composites remained relatively unaffected after being exposed to extreme conditions, except for the modulus, tensile strength and strain when measured at -40 °C condition. Overall, grafted CTMP improved polyethylene properties and gave superior mechanical properties to either mica or glass fiber based composites. With respect to water uptake by the composite samples following four-hour immersion in boiling water, the investigators found increased water uptake at higher fiber loading for composites of treated pulp. The treated pulp 35 composite also exhibited better dimensional stability than those of mica or glass fiber filled composites. The analysis of tensile and impact properties of recycled newspaper fiber-filled polypropylene composites was reported by Sanadi et al.(1994). Coupling agents incorporated in the composite system were (i) E-43 : maleic anhydride-grafted PP (MAPP) with MW a 10,000 and 6% maleic anhydride (MA); (ii) G-3002 : MAPP with MW . 40,000 and 6% MA; (iii) BPMA : MAPP with MW 2 100,000 and <0.5% MA; and (iv) AABP : acrylic acid-grafted PP with 6% acrylic acid (AA). Two mixing methods were evaluated which were described as the single stage and two stage methods. The two stage technique, where PP was added later to the mixture of the fibers and coupling agent, exhibited slight improvement in tensile and unnotched impact strength over the properties of samples from the single stage method. An increase in tensile strength was found for the E-43 and G-3002 composites, whereas there were modest increases by BPMA and smaller improvement in the AABP systems, when compared to that of the uncoupled specimens. The ‘level of molecular weight and graft content for MA were found to be the main factors contributing to the mechanical properties of the composites. The unnotched impact strength was improved with inclusion of coupling agents. The values of strengthening efficiencies were still fairly low for the well bonded composites. This was believed to be due to the short fiber lengths in the system. Childress (1991) investigated the effect of additives on mechanical properties of wood- f iber / high densi ty 36 polyethylene composites. The additives used were maleic anhydride modified polypropylene (MAPP) at 1, 3 and 5% weight ratios; ionomer modified polyethylene (Surlyn) at 1, 3 and 5% (wt/wt): and two low molecular weight polypropylene resins (Proflow 1000 and Proflow 3000), each at 5% (wt/wt) loading. The properties evaluated included tensile properties, impact strength, creep and water sorption. Only MAPP functioned as a true coupling agents which enhanced composite properties overall. Generally, Proflow 1000 and 3000 decreased the mechanical properties of the respective composites. Surlyn exhibited a small effect on tensile properties. Water sorption was likely promoted by Surlyn and Proflow 3000. However, the results showed no significant difference from the composites without additives. Simpson (1991) studied the use of recycled multi-layer polypropylene bottle resin with wood fiber in the form of composites. In these studies, 30, 40 and 50% (wt/wt) of fiber content were incorporated into composites of virgin PP and PP Reclaim (PPR). For comparison, samples from the original polymers were produced. Tests were performed to evaluate the tensile properties, flexural modulus, impact strength, creep and water absorption, all in lengthwise and crosswise fiber directions, except for the creep and water absorption tests. The structure of the PP Reclaim was PP/adhesive/EVOH/adhesive/regrind/PP. In general, the test results were favorable in the lengthwise direction. Simpson (1991) concluded that higher mechanical properties were exhibited by PPR-wood fiber composites, as compared to PP- wood fiber composites. Also, the PPR composites afforded 37 longer retention under load values, either at ambient or extreme conditions. This was thought to be the result of contributions of the components in the PP Reclaim structure to interfacial bonding. In addition, an increase in water absorption was found with an increased fiber concentration. MATERIALS AND METHODS 1. Materials 1.1 Matrix The polymer used as a matrix for all composites was the injection molding grade of polypropylene homopolymer (Pro- fax 6501) in a powdered form, which was supplied by Montell U.S.A., Inc., Lansing, Michigan. The reported properties of the polypropylene (PP) are shown in Table 1. The melting point and percent initial crystallinity of the resin were measured by a DSC 2920 Modulated Differential Scanning Calorfmetry' (MDSC), manufactured. by’ TA. Instruments, New Castle, Delaware. The values are 164.2 °C and 34% respectively. The heat transition curves from MDSC are shown in Appendix A . The surface area of the polymer determined by Nitrogen Adsorption at -196 °C using a Micromeritics Pulse Chemisorb Model 2700 apparatus is approximately 0.09 mZ/g. The particle sizes of the resin were determined by the sieve mesh technique. Over 90% of the PP particle sizes were found in the range of 355 to 855 microns. The moisture content of the powdered PP resin is 0.08%. Polypropylene is a linear thermoplastic having propylene monomers as building blocks. The repeating monomer unit of PP structure is shown in Figure 3. Three different types of 38 39 stereochemical configurations are isotactic, syndiotactic and atactic, depending on the catalyst and process of polymerization. The most commonly used foam of PP structure is isotactic (Modern Plastics), which is the highest regular form of PP and is a crystalline polymer. Due to its methyl pendant groups, PP has a relatively high glass transition temperature (Tg) and high melting point (Tm), so that PP is stiffer and stronger but lower in percent crystallinity and ductility, if compared to polyethylene. PP has excellent chemical resistance with the exception of strong oxidizers and. nonpolar solvents. It is also highly resistant to moisture (hydrophobic). The low surface energy results in very low bonding, printing and painting ability. In addition, PP is susceptible to sunlight and heat by an oxidative degradation process (Seymour and Carraher, 1984). ‘Im’ Figure 3: Repeating unit of PP Structure 40 Table 1: General Properties of Pro-fax 6501 (Source: Montell U.S.A., Inc.) Properties Averaged Value Melt flow rate, dg/min 4 Density, g/cm3 0.9 Tensile strength at yield, MPa 35 Elongation at yield, % 12 Flexural modulus, MPa 1700 notched Izod impact strength, J/m at 23 °c 40 at -18 °C <16 1.2 Reinforcing Filler Aspen hardwood fibers in the form of thermomechanical pulp were utilized as a reinforcing filler for all composites in this study. The fibers were obtained from two sources: (1) Canfor Panel and Fibre Division (New"Westmister, B.C., Canada); and (2) Georgia-Pacific Corporation (Phillips, ‘Wisconsin). The fibers from sources 1 and 2 will be referred to Fiber-1 and Fiber-2 later in this study. Wood is a natural polymeric, cellular structure with hydrophilic and polar functionality. Three basic organic constituents of wood are cellulose, hemicellulose and lignin. In general, hardwoods are composed of 41-45 % cellulose, 23-30 % hemicelluloses and 19-28 % lignin by weight (Mullins and. McKnight, 1981). The dimensions of 41 hardwood fibers range from 1.0 to 1.5 m. in length and average 15 microns in diameter (Stamn, 1964). Cellulose (C5H1005)n is a linear chain structure composed of B-D-glucopyranose units ‘with a 1,4- glucosidic linkage (Browning, 1963). The cellulose molecule is illustrated in Figure 4. This macromolecule contains monomer units ranging from 8,000 to 10,000, on average (Dinwoodie, 1989). Cellulose consists of crystalline and amorphous domains. Up to 70% of cellulose is crystallites. Intermolecular and intramolecular bonding of cellulose molecules in crystalline regions is very strong, which makes it hardly accessible to chemical reaction (Mullins and McKnight, 1981) . The amorphous regions are permeable so that the hydroxyl groups are readily attacked by reactant molecules such as water. CHLOE O H n F°__ OH H H H OH _ J. Figure 4: Cellulose Molecule 42 Hemicelluloses are polysaccharides made of various sugars including glucose, galactose, mannose, xylose and arabinose. The hemicellulose molecules are more complex and lower in molecular weight than the cellulose. Hardwood hemicelluloses have a large proportion of xylan. Glucomannan, consisting of glucose and mannan units, is at a level of 3-5 % in hardwood. Hardwood xylan is comprised of a series of xylose units linked and to end as a backbone. Methyglucoronic acid groups and acetyl groups are attached to the backbone, with ratios of 1 and 7 units, respectively, for every 10 xylose units (Mullins and McKnight, 1981) . Lignin is an amorphous polymer composed of hydroxyl- and methoxy-substituted phenyl propane units. It, as the structural support and cement material of plants, is concentrated in the spaces between wood cells and deposited within the matrix of cellulose microfibrils. Lignin in hardwood contains guaiacyl (coniferyl alcohol) and syringyl alcohol units formed as a copolymer of the two alcohols (Dinwoodie, 1989). 2. Methods 2.1 Sulfonation Treatment Sulfonation treatment of PP resin was performed at the Composite Materials and Structures Center (CMSC), Michigan State University. The sulfonation system unit was designed and manufactured by Coalition Technologies, Ltd. (Midland, Michigan). The principal operational components for the sulfonation process include a Sulfur Trioxide Generator and a Rotating Drum Reactor. The sulfur trioxide ($03) gas was 43 generated in the Generator. The rotating drum reactor is the chamber where the polymer substrate is held and the reaction of 803 on the polymer surface takes place. A schematic diagram of the Sulfonation System is illustrated in Figure 4. The operating cycle of the sulfonation process is briefly described below. First, an amount (~1816 grams) of powdered PP resin was charged into the rotating drum reactor and the connection between the generator and the rotating drum was made. Nitrogen (N2) was purged through the chamber for 10 min at a flow rate of 120 cc/min, then a vacuum (10'1 to 10'2 torr) was applied for 5 min, and N; was purged through the rotating drum reactor again for 5 min. Water and other reactants were eliminated or minimized in this step before introducing gaseous 803 to the chamber. A flow of SO3 gas at a concentration of 0.7% (v/v) was forced continuously through the reactor for a period of time (5, 8, 10 and 20 min). Following the indicated reaction time, N2 was purged through the reactor for 10 min to remove residual SO3 gas. The neutralization process was started by introducing ammonia gas (NH3) for 1-2 seconds. The drum reactor was rotated for a few minutes, and the process was ended by purging N2 through the reactor for 5 min. Sousa» coauecouasm no 5503 oauefleoom “n ease: 8:8 883291 45 The surface composition of sulfonated and nonsulfonated resins was characterized by a PHI 5400 ESCA System (Perkin- Elmer Corporation, Physical Electronics Division, Eden Prairie, Minnesota), at the CMSC. In addition, the sulfonated resins with the exposure time of 5, 8 and 10 min were submitted to Galbraith Laboratories, Inc. (Knoxville, Tennessee) for Elemental Analysis. 2.2 Sample Preparation 2.2.1 Commanding A Baker Perkins Model ZSK-30, 30 mm, 26:1 co-rotating twin- screw extruder (Werner 8: Pfleiderer Corporation, Ramsey, New Jersey) at the CMSC was used for homogenizing the compounds. The extruder is composed of five heating zones. The temperature of each heating zone was manipulated via a controlling system. By heating and water-cooling procedures, the heating temperature was maintained. The wood fiber and resin at a 2:3 weight ratio were dryblended in a separate container. The mixture was fed through the feeding zone of the extruder by a Weight-Loss-Differential Weigh Feeder (Acrison, Inc., Moonachie, New Jersey) with a control system of MD II 2000 Weigh Feeder Controller (Werner & Pfleiderer Corporation, Ramsey, New Jersey). The compositions of the composite materials fabricated are shown in Table 2. The operating parameters of the extruder for compounding the mixture were as follows: heating temperature range, 135 to 165 °C: screw speed, 100 rpm: feed rate, 6.06 to 9.47 gm/min. The parameters for extruding unfilled polypropylene were 170 °C, 100 rpm and 22.7 to 26.5 gm/min, for heating temperature, screw speed and feed rate, respectively. The 46 percentage of wood fiber used in all composites was 40% by weight. The materials extruded through the die were cut into bars and cooled by air. The weight and length of each bar was approximately 37 gm and 13 cm, respectively. Purging the retained material from the extruder was done before and after compounding, with pure resin. Table 2: Composition of Composites and Materials by Weight- Percent No. Material Code Composition 1 NS 60% Nonsulfonated PP / 40% wood Fiber4I“’ 2 NSn 60% Nonsulfonated PP / 40% wood Fiber-2m 3 85 60% 5-min Sulfonated PP / 40% wood Fiber-1 4 88 60% 8-min Sulfonated PP / 40% Wood Fiber-1 5 810 60% 10—min Sulfonated PP / 40% Wood Fiber-1 6 810n 60% 10-min Sulfonated PP / 40% Wood Fiber-2 7 S20n 60% 20-mdn Sulfonated PP / 40% Wood Fiber-2 8 PP 100% Polypropylene Resin t" 1 and 2 refer to the wood fibers from Canfor Panel and Fibre Division, Co. and Georgia-Pacific Corp., respectively. 2.2.2 Compressimnolding The extruded materials were formed into sheets using a Carver Model M Laboratory Press compression molding machine (Fred 8. Carver, Inc., Menomonee, Wisconsin). Two sizes of 47 frames were used: a 150 by 150 by 2.5 mm frame for the tensile samples; and a 127 by 127 by 3.2 mm frame for the impact and flexural samples. The heating temperature of the two platens was set at 170 °C. The sample bars were sandwiched between metal plates. Polyethylene terephthalate (PET) sheets were placed between the samples and metal plates, both top and bottom sides, in order to prevent sticking to the metal plates and to provide smooth surfaces to the samples. The conpression-molded sample was held between the heating platens for 10 min under pressure, which was gradually increased to 35,000 lbs. The system was then cooled down to 28 °C by circulation of cold water for approximately 20 min. 2.2.3 W By using a mechanical saw (Jarmac Co., Springfield, Illinois), the molded sheets were cut into test specimens, in the fiber (lengthwise) direction and perpendicular to the fiber (crosswise) direction of the composite materials. For the tensile test, the 150 x 150 x 2.5 mm. sheets were cut into 150 by 20 by 2.5 mm pieces, which were then shaped into Dumbbell Type I specimens using a Tensilkut Model 10-13 Specimen Cutter (Tensilkut Engineering Division Sieburg Industries, Inc., Danbury, Connecticut). The dimensions of the test specimens were as follows: total length, 150 mm; overall width, 20 m; width of narrow section, 10 mm; and thickness, 2.5 mm. The molded sheets of 127 x 127 x 3.2 mm size were cut into test specimens for flexural and impact testing. The flexural specimens have the dimensions as follows: length, 127 mm; width, 12.7 mm; and thickness, 3.2 48 m. The impact specimens were cut into dimensions of 62 mm in length, 12.7 mm in width and 3.2 mm in thickness. The specimens were notched using a TMI Notching Cutter (TMI Testing Machines, Inc., Amityville, New York). The notch angle was 45° and the depth of samples, at notch, was 10.16 2.3 Mechanical Testing All specimens were conditioned at 23 °C and 50% RH for at least 40 hours prior to testing. At least seven samples per material per test method were tested at once. 2.3.1 W Tensile strength, percent elongation and modulus of elasticity were measured by an Instron Universal Tensile Tester Model SFM-20 (United Calibration Corporation, Hunting Beach, California) at ambient conditions (23 °C, 50% RH). The ASTM D638-91, Standard Test Methods for Tensile Properties of Plastics (ASTM, 1993), was followed. A laser extensometer was chosen for measuring tensile strength and percent elongation. The test conditions were set as follows: full scale load cell, 1000 lbs; crosshead speed, 0.02 in/min for composites and 1 in/min for original polymer; and gauge length, 2 in. For tensile modulus measurements, a standard extensometer with l-inch gauge length was enployed. The other test conditions were as follows: full scale load cell, 20 lbs; and crosshead speed, 0.02 in/min (2 %/min of the gage length), respectively. 49 Promptly after the individual test, a computer system interfaced to the Instron Universal tester calculated the tensile properties following the equations given below: Tensile Strength (at yield or at break), a where: P' a maximum.load at yield or at break A - original cross-sectional area Percent Elongation (at yield or at break), %EG (L—Lo) x 0 96EI? = 100 where: L a extension at yield or at break L a original gauge length 0 Modulus of Elasticity, E where: AP = difference of stress corresponding to a linear portion of the load- deflection curve AL = corresponding difference in strain 2.3.2 W Sample flexural strength and flexural modulus were determined by an Instron Universal Tensile Tester Model SFM- 20 (United Calibration Corporation, Hunting Beach, California) at ambient conditions (23 °C, 50% RH), following ASTM D790-92, Standard Test Methods for Flexural Properties of Unreinforced and Reinforced Plastics and Electrical 50 Insulating Materials (ASTM, 1993). Test method I, a three- point loading system was utilized. The parameters of the test were set as follows: load cell, 20 lbs; test speed, 0.05 in/min; and span-to-depth ratio, 16:1. The test was terminated upon sample rupture or at 5% axial strain, depending which came first. Immediately after the individual test, a computer system interfaced to the Instron Universal tester calculated the tensile properties following the equations given below: Flexural strength, 8 3PL 2bd2 where: P = load at moment of break L - support span b - width of tested specimens d - depth of tested specimens Flexural Modulus, EB L3 E, = "'3 4bd where: m a slope of the tangent of an initial straight-line portion of the load- deflection curve 2.3.3 W Notched Izod Impact resistance was determined by a TMI Izod Impact Tester, Model 43-02 (TMI Testing Machines, Inc., Amityville, New York) at ambient conditions (23 °C, 50% RH). 51 The 5 ft-lb pendulum weight was used. The testing was conducted in accordance with the ASTM D256-92, Standard Test Methods for Impact Resistance of Plastics and Electrical Insulating Materials (ASTM 1993) . The machine automatically calculated and reported the impact strength value, which is the energy required to break the sample. Mathematically, an impact strength is represented by the area under the stress- strain curve until the rupture point. 2.4 Water Sorption Studies The dumbbell-shape specimens--prepared in the same manner as tensile specimens--of NS, SS, 88, 310 and PP in the fiber direction were utilized in this study. These specimens were stored in a chamber of controlled temperature and relative humidity, at 35 °C and 90 % RH. The samples were weighed at predetermined intervals of 0, 1, 3, 5, 7 and 9 weeks, respectively. The water (if any) on the surface was wiped off with a paper towel and weight measurements were performed immediately. The percentage of weight increase (or water sorption) was calculated by the following equation: Weight Increase, % = _W‘;I_W£ x100 0 where: W - sample weight after storaged W0 - original sample weight After 3, 5, 7 and 9 weeks storage, five to seven samples of each material were withdrawn for tensile testing. The tensile strength was evaluated following the ASTM D638-91 (ASTM, 1993). 52 2.5 Density Measurement Density of all composites was determined using a water displacement method. Long strips of approximately 115 x 12 x 2.5 mm dimensions were cut from the molded sheets. The weight of every sample was measured. A graduated cylinder (25 ml), containing clean water without bubbles, was used for measuring the volume of the samples. Each sample was put in the cylinder individually. The increased level of water in the cylinder due to water displacement was equal to the sample volume. The level of water in the cylinder was read before and after the displacement. The weight divided by the volume value of each sample was reported as a density of the respective sample . 2.6 Statistical Analysis SPSSQ for Windows“ Student Version (Release 6.0.1) Software program was used to perform statistical analyses. The procedure chosen was a one-way analysis of variance with the Tukey-HSD test to determine the statistical significance of the numerical data obtained in this study, at a 95% confidence interval. The analysis was performed on: (i) all mechanical properties--such as tensile properties, flexural properties and impact properties--between composite groups in both lengthwise and crosswise directions, as well as between two fiber directions of each composite material: and (ii) the tensile strength data between material groups for each storage time, and between storage conditions for each material type . RESULTS AND DISCUSSION 1. Surface Characteristic The surface composition of nonsulfonated and sulfonated PP resin samples, determined by Electron Spectroscopy for Chemical Analysis (ESCA), are presented as the percentage of atomic concentration and the relative atomic ratios in Tables 3 Iand. 4, respectively; The sulfur concentrations determined by ESCA and by Elemental Analysis are reported as a function of reaction time in Table 5. The atomic concentrations of carbon, C, oxygen, 0, nitrogen, N, and sulfur, S, are reported for various sulfonationrtime treatments (0 to 20 minutes) of the polypropylene resin. The presence of silicon, Si, was found in the resin of O-min and 5-min sulfonation time (not reported). This was attributed to contamination during sample preparation and handling. The oxygen found in the nonsulfonated resin is thought to be due to oxidation of the resin during processing. Table 4 presents the relative atomic ratios obtained for the respective sulfonated samples, which illustrate the chemical change occurring on the surface during the sulfonation process. The ratios were in accordance with the theoretical molecular structure of the sulfonate group. For instance, with an increased reaction time, the ratios of O/S and N/S remain relatively constant, at approximately 3 to 1 and 1 to 1, respectively. The results support the presence of 53 54 sulfonic acid functional groups on the polymer backbone, and their complete neutralization by NH3 occurring. The ammonium sulfonate group (-SO3'NH4+) on the polymer backbone, as a theoretical outcome of the sulfonation reaction, is depicted in Figure 6. The C/S ratio was found to decrease as the sulfonation time increased. This finding indicated. that higher sulfonation levels on the PP resin were achieved when reaction time was extended. In addition, the C/S ratio approached a constant value, 28 to 1, at 10-min sulfonation time . in _ + a II CH; SO,‘NH, CH; H CH3 . . ./ \/ \/ \/ \ Figure 6: Theoretical Molecular Structure of Sulfonated PP Asthana (1993) proposed that there was a limitation for the sulfonation of polypropylene and that sulfonation beyond the limiting level could result in degradative reactions on the polymer surface, such as chain scission. At the molecular level, chain movement due to the incorporation of SO3"NH,1|,+ 55 Table 3: Atomic Concentration for Nonsulfonated and Sulfonated PP Resins by ESCA.Analysis Percentage Atomic Concentration Sande gmdmmiJ: summun_9 Intmunanl SuLflun_S 0 min 98 1.4 - - 5 min 95 3 0.9 0.8 8 min 93 4.5 0.9 1.2 10 min 85 8 4 3 10 min‘ 85 8 4 3 20 min‘ 83 10 4 3 ' : reduced level of resin charge in the rotating drum reactor Table 4: Relative Atomic Ratios of Sulfonated PP Resins Sample 9L8 QLS M8 0 min - - .. 5 Ida 118.8 3.8 1.1 8 min 77.5 3.8 0.8 10 min 28.3 2.7 1.3 10 mdn‘ 28.3 2.7 1.3 20 min‘ 27.7 3.3 1.3 7 3 reduced level of resin charge in the rotating drum reactor 56 Table 5: Comparison of Sulfur Content Determined by ESCA Analysis and Elemental Analysis, as a function of Sulfonation Time Sample Atomic % Sulfur Total % Sulfur Per Gram (ESCA) of Resin (Elemental) 0-min 0 N/A S-Idn. 0.8 < 0.05 8-min 1.2 < 0.05 lO-min 3 < 0.05 10 min‘ 3 s/a 20 min‘ 3 N/A ‘ : reduced level of resin charge in the rotating drum reactor species was inpeded by the structure of PP and, therefore, did not allow for additional insertion of sulfonate groups beyond the sulfonation limit (Asthana, 1993). The maximum achieved sulfonation level reported by both Asthana (1993) and Wangwiwatsilp (1993) was found to be one sulfonate group per three repeating monomer units when polypropylene films were utilized. The reaction site on the polypropylene was proposed to be at the tertiary carbon, due to the highly electrophilic reaction with the 803 gas (Asthana, 1993). However, in the present study, as shown in Table 4, the highest achieved sulfonation level gave a C/S ratio of 28, which indicated, on average, one sulfur atom per 28 carbon atoms, or approximately 9 propylene monomer units per 57 sulfonate group. The results from Asthana (1993) and Wangwiwatsilp (1993) suggested that even though the extent of sulfonation achieved in the present study approached a constant level, the low sulfonation achieved was not limited by the nature of the polypropylene, but to other external factors, which are discussed in more detail in a later section. As shown graphically in Figure 7, the extent of sulfonation level achieved was found to increase with reaction time. A constant level of sulfur content was reached following a 10- min exposure time. In addition to determining the atomic sulfur content by ESCA, the total sulfur concentration (per gram of resin) determined by elemental analysis for the respective sulfonated resins was also obtained. The total sulfur content achieved was less than 0.05 % by weight, despite the increase in sulfonation level with exposure time. Several attempts at achieving higher levels of sulfonation were carried out, including a reduction of the resin charge to about one third of the original quantity (from 1816 to 681 grams) with reaction times of 10 minutes and 20 minutes, respectively. No further increase in sulfur content was achieved (see Table 5) . Atomic % Sulfur 58 0 I it 0 5 10 15 20 Sulfonation Time (min) Figure 7: Atomic Percent Sulfur Concentration of PP Resins As a Function of Sulfonation Time 59 In comparing the sulfonation of the PP resin and PP films, it is necessary to consider other factors such as the sample surface areas and operational differences for each sample. The PP resin, in the form of a powder, had a surface area of 0.09 mZ/g. The total surface area of the PP resin (681 grams) in the rotating drum was 61.29 mg, whereas the exposed surface area of the PP films (four, 6 x 13 in. films) sulfonated by a batch process (Wangwiwatsilp, 1993) was approximately 0.40 m3. Thus, the ratio of the surface area of PP resin to PP films per run is approximately 153 to l. The extremely large surface area of the resin can result in the depletion of the sulfur trioxide concentration, thereby showing no further formation of sulfonate groups with the additional time of sulfonation. Another reason for the depletion of SO3 content in the rotating drum reactor is related to the clogging of the sulfur trioxide synthesizer at the location where the sulfur dioxide and air inlets were connected to the system. The 803 synthesizer is the unit in which the reaction of SO; and air occurred at 427 °C, in the presence of a catalyst, to produce 803 for the 803 generator. The clogging of the inlet lines was attributed to the presence of a desiccant (CaSO4) at the base of the catalyst bed which was included to remove any moisture from the feed lines prior to catalytic oxidation of the 802 to yield 803 for sulfonation. The clogging resulted in a lower supply rate of $03 to the storage tanks, which in turn lowered the capacity of the 803 generator in supplying 803 for the sulfonation reaction. Sulfonation of the PP resin required a larger quantity of 60 803 due to the extremely high surface area of the resin sample charged into the reactor, as compared to that of the PP films. This problem was discovered at the completion of the present study and may be the root cause of the low levels of sulfonation achieved. Moreover, films used as a substrate can provide total exposure of the film surface to the 803 gas in the sulfonation chamber. In contrast, it is difficult to control the amount of exposed surface for the powdered resin. Through proper agitation in the rotating drum and by controlling the amount of charge weight of the resin in the reactor, together with maintaining a sufficient SO3 level, the sulfonation reaction on the polypropylene resin may be optimized to achieve the maximum expected level . 61 2. Density of Composites The density values were determined for nonsulfonated and sulfonated composites fabricated with wood fibers from two suppliers. At least eight replicate analyses were performed for each sample. The mean and standard deviation for the density data of the respective composites are summarized in Table 6. The detailed data are presented in Appendix C, Table 15. A one-way analysis of variance was performed to determine any significant difference of means, at 95% confidence interval. The detailed analysis is presented in Appendix E, Table 24. The composites from sulfonation times under 10 minutes were approximately equal in density values, with only the density of 20-min sulfonated composites showing statistically significant differences from that of the composites with 0-, 5- and 8-min sulfonation times. The higher observed density value for 20-min sulfonated composite is assumed to be the result of a variation in the composites' homogeneity from point to point in the compounding process. A difference in wood fiber weight fraction as reflected by the composite homogeneity would be expected to affect the density of the materials directly. Alternatively, although no statistically significant increment showed among sulfonated composites with 10 min. sulfonation time and under, there was a trend of an increased density values of the sulfonated composites, as a reaction time increased. The reaction of the polymer with 803 in the sulfonation process resulted in the substitution of sulfonic acid groups for hydrogen atoms. This insertion of SO3H groups in the polymer backbone can 62 result in an increase in weight and therefore the density of the composite system. With a longer reaction time, more sulfonate groups were substituted onto the polymer chain, which is the matrix phase in the composites. The sulfonation effect on the density of the respective composites was, however, kept mdnimal due to the low levels of sulfonation being achieved. Table 6: Results of Density (g/cc) Material Material Mean 8d. Code N8 Nonsulfonated PP/Wood Fiber-1 1.049 0.014 NSn Nonsulfonated PP/Wood Fiber-2 1.052 0.020 85 5min Sulfonated PP/Wood Fiber-1 1.053 0.017 88 8min Sulfonated PP/Wood Fiber-l 1.050 0.011 810 10min Sulfonated PP/Wood Fiber-1 1.064 0.011 810n 10min Sulfonated PP/Wood Fiber-2 1.063 0.015 820n 20min Sulfonated PP/Wood Fiber-2 1.078 0.009 63 3. Tensile Properties For the respective composites, tensile strength and percent elongation at fracture moment, and modulus of elasticity were determined in both lengthwise and crosswise fiber directions. The tensile strength and elongation at yield point, as well as modulus of elasticity, were reported for the pure PP polymer parallel to the extrusion flow. The mean and standard deviation of 12 to 14 replications for tensile properties of the respective composites and PP are summarized in Tables 7, 8 and 9, respectively. The results are also depicted graphically in Figures 8, 9 and 10, where the tensile strength, percent elongation and modulus of elasticity data are presented, respectively. The detailed data of the tensile properties are also presented in Appendix C from Table 16 to Table 18. A one-way analysis of variance of tensile properties of the respective composites was performed to determine any significant differences between means, at a 95% confidence interval. The results of the statistical analysis are shown in Appendix E from Table 25 to Table 33. 3.1. Effect of Fiber Direction The referred to '1engthwise' and 'crosswise' fiber directions were based on the assumption that the extrusion compounding process can provide the preferred alignment of short fibers in the flow direction of the extrudate, which is regarded as the lengthwise direction. For comparison, the direction perpendicular to the flow is regarded as the crosswise direction. From all tensile data observed, the tensile strength of NSn, 810 and 810n; the percent 64 elongation of NS, NSn and 85: and modulus of elasticity of NSn composites showed higher values in the lengthwise direction than in the crosswise direction, while there was no statistically significant difference in the tensile properties between the remaining composites as a function of orientation direction. The fibers parallel to the loading axis generally provide composites with higher tensile properties than do fibers in the crosswise direction, due to the load transfer from the matrix through fiber ends and hence a more efficient contribution of fibers to the composite properties. However, the results indicated that only a low degree of uniform fiber orientation was present in the composites from this experiment. The small extent of preferred alignment was probably accounted for by the pronounced effect of the shear field, during the extrusion process, producing no net change in fiber orientation, rather than the fiber rotation by the elongational field. Also, fibers were shortened because of fiber breakage during the extrusion process. In practice, it is difficult to satisfactorily control the fiber direction during the extrusion process. Fibers tended to be aligned in a random fashion, which produced composites more or less as an isotropic material. Even though statistical differences were found between fiber directions in some cases, composites with a large degree of aligned fibers usually give markedly higher mechanical properties i.e. tensile strength and modulus, for samples aligned parallel to the fiber direction. A possible reason for the somewhat lower properties observed in the crosswise samples is that the sheet formed from the extruded bars did not have good 65 bonding when molded by the compression molding procedure. Similar findings were also presented by Haraguchi (1993) for polyethylene/ wood fiber composites. Another reason for the lower properties observed might be the low homogeneity achieved in the mixing process, resulting in a wide deviation in properties. 3.2. Effect of Sulfonation The tensile properties of both nonsulfonated and sulfonated composites were compared, as a function of sulfonation time, within the composite system for each type of wood fibers used. Thus, the evaluation was focused on two sets of composites: (1) N8, 85, 88 and 810: and (ii) NSn, 810n and S20n. Based on statistical analysis, the observations with respect to tensile strength, percent elongation and modulus of elasticity properties of the composites are summarized as follows : (l) The tensile strength values were: (i) improved for samples 85 and 88 over sample N8 in both the lengthwise and crosswise directions: (ii) increased for sample 810 over sample N8 in the lengthwise direction; (iii) increased for sample 810n over sanple NSn in the crosswise direction: (iv) increased for sample S20n over sample NSn in both directions: (v) equivalent between samples 85 and S8 in the lengthwise direction and between samples 85, S8 and 810 in the crosswise direction: and (vi) equivalent between samples 810n and 820n for both fiber directions. (2) The percent elongation values were: (i) lower for sample 85 than for sample N8 in both the lengthwise and 66 crosswise directions: (ii) lower for samples 88 and 810 than for sample N8 in the lengthwise direction: (iii) lower for sample 810n than for sample NSn in the lengthwise direction; (iv) lower for sample S20n than for sample NSn in both directions: (v) equivalent among samples 85, 88 and 810 in the lengthwise direction and between samples 88 and 810 in the crosswise direction: and (vi) equivalent between samples 810n and 820n in both fiber directions. (3) The modulus of elasticity values were: (1) equivalent among samples NS, 85 and 810 in both the lengthwise and crosswise directions; (ii) higher for sample 88 than for sample 810 in the lengthwise direction; and (iii) equivalent among samples NSn, 810n and 820n in both directions. There was a variation of the resultant mechanical properties of the respective composite structures. Nevertheless, the findings showed that sulfonation had a statistically significant effect on the enhancement of tensile strength, and little or no effect on the improvement of the other tensile properties evaluated for the composites. The lack of significant improvement of the composite mechanical properties by sulfonation of the PP resin was attributed in part to the low sulfonation level achieved. The levels of sulfonation achieved are thought to contribute minimally to improvement of interfacial interaction between fiber and matrix phases, thereby showing a small influence in enhancing the tensile properties of the resultant composites . Hence , with economical and practical considerations in mind, this achievement did not exhibit any apparent benefit at this time in commercial or industrial ap; Iul te ex it ca 8t 8a th 8&2 e1c 67 applications. It should be noted, however, that the sulfonation treatment for the enhancement of composite tensile properties still shows potential. For example, examination of the tensile strength data showed a marginal increase for the sulfonated composites, as compared with the nonsulfonated structures. Since PP has shown its capability to achieve higher sulfonation levels via the sulfonation treatment, as reported by previous investigators (Asthana, 1993 and Wangwiwatsilp, 1993), it is assumed that if the optimization of the sulfonation treatment for PP resin was achieved, the higher level sulfonated PP resin could provide a substantial increase in the mechanical properties of the composites, as a result of the improvement of fiber-matrix interfacial interaction. 3.3. Effect of Fiber Type Fiber-1 and fiber-2 (as described in Materials section) were used as reinforcing fillers for both nonsulfonated and sulfonated composites. To compare the effect of the two fibers on the tensile properties, the following pairs of composite materials were evaluated: (1) samples NS and NSn: and (ii) samples 810 and 810n. Based on statistical analysis, the following conclusions were drawn: (1) In the case of the tensile strength property, there was no statistically significant difference between the NS vs. NSn sample pair and the 810 vs. 810n sample pair, except that the tensile strength of sample NS was higher than that of sample NSn in the crosswise direction; (ii) For percent elongation, a higher percent elongation for sample NSn in both directions was observed, while sample 810n showed a 68 higher value in the crosswise direction over its respective counterparts. For the lengthwise direction, sample 810n showed no significant difference from sample 810; and (iii) For the modulus of elasticity, no statistically significant difference was found between the NS vs. NSn sample pair, and the 810 vs. 810n sample pair, except for the NS vs. NSn samples in the crosswise direction. The properties of wood fibers can vary greatly among species. In addition, within the same species, the fiber characteristics can differ from batch to batch, when different parts of wood are used. The ways in which fibers are processed, and handled, for example, are important to properties of the fibers (Richardson, 1987). Fiber characteristics such as fiber length, fiber length distribution and fiber aspect ratio can greatly affect the properties of short-fiber composites. The load transmission from matrix to fibers is dependent on the fiber aspect ratio. When the fiber aspect ratio is decreased, the ability of stress transfer also decreases (Hull, 1981). From the findings of the present studies, there was little or no significant difference in tensile properties between fiber-1 and fiber-2 composites. Therefore, the effect of the two fiber types on the tensile properties could be neglected. ..u )IS RSI BS: 85 SS 88 38 SN SIC SII Sli SZI 32C PP 69 Table 7: Results of Tensile Strength at Break (MPa) Material Material Fiber Mean 8d. Code Direction N8 Nonsulfonated PP/Wood Fiber-1 Lengthwise 16.29 1.33 NS Nonsulfonated PP/Wood Fiber-1 Crosswise 15.56 1.43 NSn Nonsulfonated PP/Wood Fiber-2 Lengthwise 16.46 0.94 NSn Nonsulfonated PP/Wood Fiber-2 Crosswise 13.35 1.55 85 5min Sulfonated PP/Wood Fiber-1 Lengthwise 19.17 1.51 85 5min Sulfonated PP/Wood Fiber-1 Crosswise 18.61 2.22 88 8min Sulfonated PP/Wood Fiber-1 Lengthwise 19.04 1.50 88 8min Sulfonated PP/Wood Fiber-1 Crosswise 18.01 1.19 810 10min Sulfonated PP/Wood Fiber-1 Lengthwise 18.74 1.81 810 10min Sulfonated PP/Wood Fiber-1 Crosswise 17.13 1.70 810n 10min Sulfonated PP/Wood Fiber-2 Lengthwise 17.84 1.42 810n 10min Sulfonated PP/Wood Fiber-2 Crosswise 16.01 2.36 820n 20min Sulfonated PP/Wood Fiber-2 Lengthwise 19.31 1.13 820n 20min Sulfonated PP/Wood Fiber-2 Crosswise 19.15 1.69 PP Virgin PP 33.32a 0.26 5 Reported as tensile strength at yield 70 35 ,MILE grams-rs (MPa) I Lengthwi se 3 0 l__ L'JCrosswise 25 _ 73—-wwe_ 20 15. 104 NS NSn 85 88 810 810n 820n PP Material Figure 8: Tensile Strength at Break (MPa) 71 Table 8: Results of Percent Elongation at Break Material Material Fiber Mean 8d. Code Direction NS Nonsulfonated PP/Wood Fiber-1 Lengthwise 2.90 1.10 NS Nonsulfonated PP/Wood Fiber-1 Crosswise 1.69 0.74 NSn Nonsulfonated PP/Wood Fiber-2 Lengthwise 4.67 1.04 NSn Nonsulfonated PP/Wood Fiber-2 Crosswise 2.39 0.63 85 5min Sulfonated PP/Wood Fiber-1 Lengthwise 1.43 0.45 85 5min Sulfonated PP/Wood.Fiber-1 Crosswise 1.09 0.29 88 8min Sulfonated PP/Wood Fiber-1 Lengthwise 1.35 0.28 88 8min Sulfonated PP/Wood Fiber-1 Crosswise 1.31 0.31 810 10min Sulfonated PP/Wood Fiber-1 Lengthwise 1.72 0.46 810 10min Sulfonated PP/Wood Fiber-1 Crosswise 1.30 0.61 810n 10min Sulfonated PP/Wood.Fiber-2 Lengthwise 2.26 0.61 810n 10min Sulfonated PP/Wood Fiber-2 Crosswise 1.94 0.52 S20n 20min Sulfonated PP/Wood Fiber-2 Lengthwise 1.75 0.30 820n 20min Sulfonated PP/Wood Fiber-2 Crosswise 1.71 0.35 as Virgin as 8.855 0.36 5 Reported as percent elongation at yield U 72 PERCENT ELONGATION I Lengthwi se DCrosswise NSn 85 88 810 810n S20n Material Figure 9: Percent Elongation at Break PP 73 Table 9: Results of Modulus of Elasticity (MPa) Material Material Fiber Mean 8d. Code Direction N8 Nonsulfonated PP/Wood Fiber-l Lengthwise 3281 340 NS Nonsulfonated PP/Wood Fiber-1 Crosswise 3088 276 NSn Nonsulfonated PP/Wood Fiber-2 Lengthwise 2856 383 NSn Nonsulfonated PP/Wood Fiber-2 Crosswise 2493 469 85 Brain Sulfonated PP/Wood Fiber-1 Lengthwise 3408 594 85 5min Sulfonated PP/Wood Fiber-1 Crosswise 3122 537 88 8min Sulfonated PP/Wood Fiber-1 Lengthwise 3472 413 88 8min Sulfonated PP/Wood Fiber-1 Crosswise 3487 496 810 10min Sulfonated PP/Wood Fiber-1 Lengthwise 3005 376 810 10min Sulfonated PP/Wood Fiber-1 Crosswise 3034 517 810n 10min Sulfonated PP/Wood Fiber-2 Lengthwise 2780 149 810n 10min Sulfonated PP/Wood Fiber-2 Crosswise 2845 315 820n 20min Sulfonated PP/Wood Fiber-2 Lengthwise 2927 435 S20n 20min Sulfonated PP/Wood Fiber-2 Crosswise 2824 383 99 Virgin as 1725 176 4500 74 MDDULUS OF ELASTICITY (MP8) 4000 « 3500 I Lengthwi se U Crosswise 3000 . 2500 -I 2000 . 1500 4 1000i 500 1 MS N83 85 S8 810 S1011 8201! PP Material Figure 10: Modulus of Elasticity (MPa) 75 4. Flexural Properties Flexural strength and flexural modulus were determined in both lengthwise and crosswise fiber directions for the respective composites. The flexural modulus value for pure polypropylene was determined in the direction parallel to the extrusion flow. The mean and standard deviation of 12 to 16 replications for flexural properties of the respective composites and. PP 'were determined. and are tabulated. in Tables 10 and 11. The results are also illustrated graphically in the histograms shown in Figures 11 and 12, where the flexural strength and flexural modulus data are plotted, respectively. The detailed data of the flexural properties are also presented in Appendix C in Tables 19 and 20. A one-way analysis of variance of flexural properties of the respective composites was performed to determine any significance between means, at a 95% confidence interval. The results of the statistical analysis are summarized in Appendix E in Tables 34 to 39, respectively. 4.1 Effect of Fiber Direction None of the composites exhibited a statistically significant difference in flexural strength between lengthwise and crosswise directions, except for samples NSn and 810, which showed higher flexural strength in the lengthwise direction. For flexural modulus, samples NS, NSn, 810 and 810n, in the lengthwise direction, exhibited higher values than those in the crosswise direction, whereas the other composites gave equivalent flexural modulus values in both directions. These findings suggested that the orientation of fibers did not seem to be predominant in the direction of extrusion flow, 76 but were more likely in a random fashion. The differences observed between the values in the crosswise and lengthwise directions may be due in part to the lack of bonding between the extruded bars during the compression molding step to form the test sheets, as well as to the nonhomogeneity of the compounding step. 4.2 Effect of Sulfonation The flexural properties of both nonsulfonated and sulfonated composites were compared as a function of sulfonation time, within the composite system for the two wood fiber types used. Thus, the evaluation was focused on two sets of composites: (1) samples NS, 85, S8 and 810; and (ii) samples NSn, 810n and S20n. Based on statistical analysis, the observations, with respect to flexural strength and flexural modulus of the conposites, are summarized as follows: (i) For flexural strength, sample 810 displayed a higher value than sample N8 in the lengthwise fiber direction; (ii) In the crosswise direction, samples 810n and S20n showed higher flexural strength than sample NSn; (iii) No significant difference between samples 85, 88 and 810 was observed in either direction: (iv) For flexural modulus in the crosswise direction, samples 810n and S20n exhibited higher values than sample NSn; and (v) Sample 85 gave a higher modulus value than samples 810 and N8 in the crosswise direction. The other samples showed no statistically significant difference in flexural property values within the group. The poor interfacial adhesion between fiber and matrix was not likely to be overcome by the low extent of sulfonation achieved for the sulfonated PP samples. Thus, the properties 77 of the sulfonated composites were comparable to the nonsulfonated materials. However, it should be noted that a marginal increase in flexural strength. was observed. for samples 810, 810n and 820n, respectively. These findings provide supportive evidence of the potential effect of sulfonation on the mechanical properties of polymer/wood fiber composites, if a higher degree of sulfonation of PP is achieved. 4.3 Effect of Fiber Type To provide a comparison of the effect of fiber-1 and fiber-2 on the flexural properties, the following pairs of composite materials were evaluated: (i) samples NS and NSn; and (ii) samples 810 and 810n. Based on statistical analysis, the conclusions drawn are as follows: For flexural strength, no significant difference was found between samples NS vs. NSn and samples 810 vs. 810n in both the lengthwise and crosswise directions. For flexural modulus, only sample NS afforded higher values than NSn in both directions. The results showed little or no influence of fiber type on the composites' flexural properties. 78 Table 10: Results of Flexural Strength (MPa) Material Material Fiber Mean 8d . Code Direction N8 Nonsul fonated PP/Wood Fiber-1 Lengthwise 38 . 54 3 . 26 N8 Nonsul fonated PP/Wood Fiber-1 Crosswise 37 . 36 4 . 16 NSn Nonsul fonated PP/Wood Fiber-2 Lengthwise 40 . 40 2 . 72 NSn Nonsulfonated PP/Wood Fiber-2 Crosswise 34 . 65 4 . 31 85 5min Sulfonated PP/Wood Fiber-1 Lengthwise 41 . 74 2 . 54 85 5min Sulfonated PP/Wood Fiber-1 Crosswise 39 . 67 3 . 39 S8 8min Sul fonated PP/Wood Fiber-1 Lengthwise 40 . 06 3 . 76 88 8min Sulfonated PP/Wood Fiber-1 Crosswise 40 . 27 2 . 64 810 10min Sul fonated PP/Wood Fiber-1 Lengthwise 42 . 58 3 . 58 810 10min Sul fonated PP/Wood Fiber-1 Crosswise 39 . 36 3 . 46 810n 10min Sul fonated PP/Wood Fiber-2 Lengthwise 42 . 49 3 . 37 810n 10min Sul fonated PP/Wood Fiber-2 Crosswise 41 . 03 3 . 74 820n 20min Sulfonated PP/Wood Fiber-2 Lengthwise 41 . 51 2 . 58 820n 20min Sulfonated PP/Wood Fiber-2 Crosswise 43 . 28 4 . 03 55 50 45 79 FLEXURAL STRENGTH (MP8) I Lengthwi se D Crosswise NS NSn 85 88 S10 810n Material Figure 11: Flexural Strength (MPa) S20n 80 Table 11: Results of Flexural Modulus (MPa) Material Material Fiber Mean Sd. Code Direction N8 Nonsulfonated PP/Wood Fiber-1 Lengthwise 3017 205 N8 Nonsulfonated PP/Wood Fiber-1 Crosswise 2719 246 NSn Nonsulfonated PP/Wood Fiber-2 Lengthwise 2610 318 NSn Nonsulfonated PP/Wood Fiber-2 Crosswise 2374 143 85 5min Sulfonated PP/Wood Fiber-1 Lengthwise 2995 112 85 5min Sulfonated PP/Wood Fiber-1 Crosswise 3096 165 88 8min Sulfonated PP/Wood Fiber-1 Lengthwise 2927 202 88 8min Sulfonated PP/Wood Fiber-1 Crosswise 2900 222 810 10min Sulfonated PP/Wood Fiber-1 Lengthwise 2990 164 810 10min Sulfonated PP/Wood Fiber-1 Crosswise 2823 161 810n 10min Sulfonated PP/Wood Fiber-2 Lengthwise 2824 170 810n 10min Sulfonated PP/Wood Fiber-2 Crosswise 2707 141 820n 20min Sulfonated PP/Wood Fiber-2 Lengthwise 2720 192 820n 20min Sulfonated PP/Wood Fiber-2 Crosswise 2800 194 PP Virgin PP 1454 140 81 4000 FLEXURAL MODULUS (MP8) ILengthwise 3 5 00 _ UCrosswise 3000i 2500 . 2000 . 1500. —— e— 1000 1 i— 500 . NS NSn 85 S8 810 810n S20n PP Material Figure 12: Flexural Modulus (MPa) 82 5. Izod Impact Strength Izod impact strength was determined in both the lengthwise and crosswise fiber directions for the respective composites, and also determined for pure PP. The mean and standard deviation of 16 replications for impact strength data of the respective composites and pure PP are summarized in Table 12, and are presented graphically in Figure 13. The detailed results of the impact strength test are summarized in Appendix C, Table 21. A one-way analysis of variance of impact strength of the respective composites was performed to determine any statistically significant difference between means, at a 95% confidence interval. The results of statistical analysis are shown in Appendix E, Tables 40 and 42. 5.1 Effect of Fiber Direction The impact strength values of sample N8 in the crosswise direction and samples 88, 810 and 820n in the lengthwise direction were statistically higher than the impact strength values for the sample measured in the other fiber direction. The impact strength values of 85 and 810n did not display any significant difference between lengthwise and crosswise directions. A slight difference in the impact properties between the two fiber directions may be due to the fact that, as discussed earlier, a preferred alignment in the direction of the extrudate flow may occur to a small degree, such that the fibers were oriented equally in both the direction parallel and perpendicular to the extrudate flow. 83 5.2 Effect of Sulfonation The impact strength of nonsulfonated and sulfonated composites was compared within the composite system fabricated from wood fiber-1. For conposites fabricated with wood fiber-2, only impact values for sulfonated composites were considered, since no values for the nonsulfonated composite were determined. Thus, the evaluations were on the following groups: (1) samples NS, 85, 88 and 810: and (ii) samples 810n and 820n. The results of statistical analysis, are as follows: (1) sample 810 gave a higher impact strength value in the lengthwise direction than samples NS, 85 and 88; (ii) no statistical difference was found between samples NS, 85 and S8 in the lengthwise direction; (iii) samples NS and 810 gave higher impact values than samples 85 and S8 in the crosswise direction: (iv) sample 810n gave higher impact values than sample S20n in both directions. The findings showed that the effect of sulfonation on inpact strength was minimal for the sulfonated composite structures. This was thought to be due in part to the low sulfonation levels achieved. Still, it was found that samples 810 and 810n showed a positive effect of sulfonation. The polar characteristics of the sulfonated PP resulted in an enhancement in interfacial interaction between the polymer and the wood fibers. This yielded an increase in the mechanical properties of the composite structures. Moreover, the impact strength values were also influenced by the geometry of the tested specimens and the notch size and radii. These factors could result in variation from one sample to another, and could account for the difference between samples 810n and 820n. During the impact, 84 deformation of the ductile matrix and its ability to absorb energy is interrupted by the discontinuity of the fiber phase. Thus, the fibrous composites would exhibit lower impact resistance than the unfilled polymer. However, the impact strength of PP polymer was not as high as expected (see Table 1). This could result from a change in the crystalline fraction of the polymer, during heat processing (extrusion). It is well known that the hydrogen at the tertiary carbon of PP is susceptible to degradative oxidation reactions, such as chain scission. especially when exposed to heat. 5.3 Effect of Fiber Type To compare the effect of fiber-1 and fiber-2, the pair of composite materials evaluated were 810 vs. 810n. Based on statistical analysis, the results showed no statistically significant difference between the impact strength values for samples 810 and 810n in either the lengthwise or crosswise directions. The results indicated that the impact strength was independent of the fiber type. 85 Table 12: Results of Izod Impact Strength (J/m) Material Material Fiber Mean 8d . Code Direction N8 Nonsul fonated PP/Wood Fiber- 1 Lengthwise 20 . 65 2 . 03 N8 Nonsul fonated PP/Wood Fiber-1 Crosswise 22 . 54 0 . 86 NSn Nonsul fonated PP/Wood Fiber-2 Lengthwise -5 -5 NSn Nonsul fonated PP/Wood Fiber-2 Crosswise -5 -5 85 5min Sul fonated PP/Wood Fiber-1 Lengthwise 20 . 68 1 . 31 85 5min 8ul fonated PP/Wood Fiber-1 Crosswise 20 . 45 1 . 61 88 8min Sul fonated PP/Wood Fiber-1 Lengthwise 21 . 60 2 . 18 88 8min Sul fonated PP/Wood Fiber-1 Crosswise 19 . 81 0 . 93 810 10min Sul fonated PP/Wood Fiber-1 Lengthwise 26 . 67 3 . 58 810 10min Sulfonated PP/Wood Fiber-1 Crosswise 24 . 14 2 . 73 810n 10min Sul fonated PP/Wood Fiber-2 Lengthwise 24 . 53 2 . 52 810n 10min Sul fonated PP/Wood Fiber-2 Crosswise 25 . 40 1 . 31 820n 20min Sul fonated PP/Wood Fiber-2 Lengthwise 21 . 35 1 . 81 820n 20min Sul fonated PP/Wood Fiber-2 Crosswise 19 . 28 1 . 59 PP Virgin PP 20 . 49 3 . 15 5 No reported data, due to the errors in the experiment. 86 IZOD IMPACT STRENGTH (J/m) 35 Fl Lengthwi se 1 30 .L___-’ DCrosswise i——— fl #*fi 25 ..--f_—_ — i _ ---- i 14___ 20 4 _ -1 15 I r— t- _ 10 -< I— ~— )— 5 .I _ 0 . e . . NS 85 88 810 810n 820n PP Material Figure 13: Izod Impact Strength (J/m) 87 6. Water Sorption Studies Water vapor sorption by the nonsulfonated and sulfonated composites of fiber-1 was determined and reported as percentage of weight increase versus storage time, as shown in Figure 14. The tensile strength values of the respective composites were determined as a function of storage time and are summarized in Table 13. The results are also illustrated graphically in Figure 15. The detailed data are presented in Appendix D, in Tables 22 and 23, respectively. A one-way analysis of variance was performed to determine any significant difference between the population means of tensile data for each treatment, at a 95% confidence interval. The results of the analysis are sumarized in Appendix E, in Tables 43 and 44, respectively. 6.1 Weight Increase All of the composite structures tested showed a weight increase due to the water uptake as a function of storage time, when exposed to the humidified conditions (35 °C, 90% RH) . Within 7 to 9 weeks, the weight increase reached a steady state or equilibrium level of water sorption. Overall, the maximum amount of water uptake was approximately 2.2 to 2.5% by weight, which is due primarily to water sorption by the wood fibers in the composites. As previously mentioned (see in Materials section), the wood fibers are hydrophilic and hygroscopic in nature, which readily form hydrogen bonds with water molecules and the hydroxyl groups on the fiber chains. Thus, wood fiber composites are susceptible to moisture uptake, and potentially fiber swelling due to water sorption leading to de' 133 81‘: re 811 811 00 CC BL‘ 11C dimensional changes of the material, as well as a deteriorative effect on their mechanical properties. Similar patterns of water sorption between nonsulfonated (NS) and sulfonated (85, 88 and 810) composites were observed, which indicated that there was little effect of sulfonation on water sorption by the respective composites. This may be the result of the achievement of low sulfonation levels on the sulfonated PP resin. Further, the molar concentration of sulfonic acid groups inserted onto the polypropylene, as compared to the abundance of available hydroxyl groups from wood fibers, would tend to have little effect on the moisture sorption characteristic of the sulfonated composites. The variation between the sorption curve of the sulfonated composites (i.e. 85 and S8) and that of the nonsulfonated sample may be attributed to low homogeneity in mixing of the composites, which can result in a variation in composition between one sheet and another. For the pure polypropylene sample, no water sorption occurred because PP, with a hydrophobic attribute, is well resistant to the sorption of water vapor. Weight Increase (% w/w) 89 Storage Time (days) Figure 14: Weight Increase (% w/w) As a Function of Storage Time (days), at 35 °C and 90% RH 6.2 The C011 90% iii an: of Bi 8t 8)! S] 90 6.2 Tensile Strength The tensile strength values were measured for the respective composites after 0, 3, 5, 7 and 9-week storage in the 35 °C, 90% RH conditioned chamber. Only samples with lengthwise fiber direction were evaluated. Based on statistical analysis, the results obtained were as follows: (i) for each of the composites, tensile strength values remained unaffected when stored up to 9 weeks; and (ii) there was no significant difference in tensile strength between samples stored for the same period of time, with the exception of a small increase in tensile strength values for samples 85 and 810 after 7-weeks storage, and sample 85 after 9-weeks storage. Sulfonation of the matrix polymer generally increases the polar contribution to the polymer surface free energy. Therefore, the sulfonated matrix would enhance its compatibility and interaction with wood fiber and, as a consequence, with moisture. However, the effect of hrumidified conditions on the sulfonated composites was minimal from this study, which could be the result of the low level of sulfonation achieved on the PP matrix polymer. Tab? 91 Table 13: Tensile Strength Data (MPa) of Samples Stored under Humidified Conditions, at 35°C and 90% RH, for 0, 3, 5, 7, and 9 weeks Material Tensile Strength (MPa) at Various Storage Time 0-week 3-week 5-week 7-week 9-week N8 16.29 16.76 16.50 16.15 16.37 85 19.17 18.54 18.55 19.04 18.97 88 19.04 17.99 18.50 17.98 17.61 810 18.74 18.86 18.11 18.22 17.59 PP 33.32 35.44 33.56 35.43 34.41 92 4o.IENSILE_SIRENGTH_JMPa) 35 r_.rr ‘ 00-week ill , 7., 1.7.-. I”. r_.,:- ‘ 83-week 3o.__nnui IS-week Livgini W,,,444fl 74755—— i I I l7-week 25.114, f in, ,,11 ,Ill I9-week 20- 10«fi 1‘ [‘i I‘MIIII‘H NS 85 88 810 PP Material Figure 15: Tensile Strength of Samples Stored under Humidified Conditions, at 35 °c and 90% RH, for o, 3, 5, 7 and 9 weeks SUMMARY AND CONCLUSIONS 1. With an increase in exposure time for the sulfonation reaction, the atomic percent of sulfur content on the polymer surface increased. The maximum sulfonation level achieved was one sulfur atom per 28 carbon atoms, or approximately 9 propylene monomer units per sulfonate group. This extent of sulfonation was considered low, as it was only one third of the sulfonation extent achieved by the sulfonation of polypropylene films. 2. The surface sulfonation level approached a steady state within 10 minutes of sulfonation time. No further increase in sulfonation level was achieved, by reducing the quantity of resin added to the rotating drum reactor or increasing the exposure time. However, the total surface area of the resin sample sulfonated in the rotating drum reactor was approximately 153 times greater than the surface area of polypropylene film samples sulfonated in earlier studies (Asthana, 1993 and Wangwiwatsilp, 1993) . A clog or blockage within the 803 synthesizer system, as found after completion of these studies, could have retarded the production of 803, resulting in a low molar concentration of 803 available for sulfonation process. The marked increase in the surface area of the resin sample, together with the clogging of the 803 synthesizer, may have led to a depletion of 803 gas and a 93 94 concomitant reduction in reaction rate following a 10 minute exposure time . 3. The increase in the polar characteristic of polypropylene resin, at the low sulfonation level achieved, was not sufficient to substantially improve the interfacial interaction between fibers and matrix, and therefore a minimal effect on the enhancement of mechanical properties, to include tensile properties, flexural properties and Izod impact strength, was observed. 4. All composites stored under humidified conditions exhibited water sorption, with the extent of water sorbed increasing with increased storage time up to 7 to 9 weeks, at which time the equilibrium sorption level was achieved. The maximum weight increase was approximately 2.2 to 2.5 % by weight. The sample water sorption was attributed mainly to the presence of the wood fibers, which are hydrophilic in nature. Even though the composites experienced water uptake during storage, no change in the tensile strength of the composite samples was found with an increase in storage time. 5. Polymer surface sulfonation was found to have no statistically significant effect on the water sorption capacity of the composites. Moreover, the physical properties of composites were relatively unaffected by the sulfonation. These findings are assumed to result from the low sulfonation level achieved on the sulfonated PP resin. 95 6. At the sulfonation levels achieved in the present study, there is some evidence indicating that sulfonation had a positive effect on enhancing the mechanical properties of the resultant composite structures. For example, the observed increase in tensile strength and flexural strength of the sulfonated PP/wood-fiber composites as compared to those of their nonsulfonated counterparts. These enhanced mechanical property levels were not considered sufficient to find utility for such modified composites in a commercial or industrial practice, since the maximum achievement of the sulfonation step was still low. However, surface sulfonation of the polymer resin showed potential as a method for enhancing the mechanical properties of the composite system with wood fibers, assuming that a higher sulfonation level of the PP resin, as found with PP film samples, is achieved. RECOMMENDATIONS FOR FURTHER RESEARCH 1. Because of the clogging problem found with the sulfur trioxide synthesizer, and the associated depletion of 803 concentration during the sulfonation reaction on PP resin samples, future studies should include repeating the sulfonation process after the 803 generator has been repaired. The operational parameters that were utilized in this study should be employed. Higher levels of sulfonation should be achieved with higher molar concentrations of 803 being generated by the synthesizer. 2. Since PP has been found to achieve higher sulfonation levels, up to approximately 9 propylene monomer units per one sulfonate group, or three times the level achieved in the present study, the limitation on the sulfonation of PP resin does not result from inherent characteristics of the polymer itself, but from some external factors such as the ratio of the resin surface area to 803 concentration, and a restriction of the total resin sample surface area exposed to the gaseous 803. Techniques or conditions that would control the external parameters of the sulfonation process should be considered in an attempt to optimize the process in achieving the maximum degree of sulfonation on the PP resin. For example, reducing further the amount of resin, or increasing the 803 concentration may be done, in order to decrease the ratio of resin surface area to $03 96 97 concentration. Those should be conducted simultaneously with the use of a well-agitated drum reactor. Therefore, collaboration with engineering personnel and technicians or operators might be necessary to achieve this end. 3. The compounding technique utilized for this study was done by feeding dry blends between PP powder and fibers to the feeding zone of the twin-screw extruder. A disadvantage of this technique is that the fibers are exposed to high shear forces during the melting of the polymer, thus damaging the fibers (Bigg, 1985) . Further, the fibers are exposed to high temperatures for a period of time equal to the polymer residence time. The long residence time of the wood fibers in the extruder can result in fiber degradation and breakage, even though the processing heat was controlled at a lower level than its degradation temperature (~200 °C) . This, in turn, will affect the mechanical properties of composites. Thus, the compounding technique should be done by adding fibers to the extruder at a point where the polymer has already been melted. This technique has been used by current researchers in the wood fibers/thermoplastic composite field and it was found to yield less fiber damage during the compounding process. In order to achieve a sufficient modification of polar characteristics of the polypropylene resin structure and concomitant improvement of the mechanical properties of the composite system with wood fibers, the optimization of the sulfonation process and a careful selection of compounding technique are important steps for further research. APPENDICES APPENDIX A Modulated Differential Scanning Calorimetry Modulated differential scanning calorimetry (MDSC) is a new technique which measures differential heat flow between a sample of a material and inert reference at the same temperature. The temperature modulation is programmed to scan a temperature range at predetermined amplitude and period. Thus, the heat flow is determined as a function of sinusoidal change in temperature and as a function of linear change in temperature when average temperature change is utilized. As a result, MDSC give the same information as conventional differential scanning calorimetry (DSC) does, with additional information related to the material properties. The raw signals from the MDSC method are (i) modulated heat flow: and (ii) modulated heating rate (or a derivative of temperature over time). Throughout the mathematical calculations, three curves of heat flow from the MDSC raw signals are evaluated: 1) Total heat flow. Total heat flow is defined as a sum of all thermal events in the sanple and is acquired by averaging the modulated heat flow. This signal is equivalent to the signal received from the conventional DSC. 98 99 2) Reversing heat flow. Reversing heat flow is the heat capacity component of the total heat flow, and it is calculated by multiplying heat capacity (proportional to the ratio of the heat flow amplitude to heating rate amplitude) with the average heating rate. 3) Nonreversing heat flow. It refers to the kinetic component of the total heat flow and is equal to the total heat flow signal subtracted by the reversing heat flow signal. The transitions occurring in the resultant signal are usually nonreversible at specified conditions, such as relaxation, cold crystallinity, decomposition, and thermoset cure. (Source: TA Instruments Co.) The melting temperature (Tm) and heat of fusion of polypropylene resin were obtained from the total heat flow curve. The percent crystallinity values of the resin were evaluated by the heat of fusion data of the sample examined in relation to the heat of fusion of the PP having 100 % crystallinity, of which the value is 209 J/g (Brandrup and Immergut, 1975) . The melting temperature, heat of fusion and percent crystallinity of the PP resin were summarized in Table 14 . 100 Table 14: Properties of PP Resin Determined by Modulated Differential Scanning Calorimeter Sample No. Melt Temp. Heat of Fusion %Crystallinity (°C) (J/g) 1 164.60 73.11 34.98 2 163.78 69.84 33.42 Mean 164.19 71.48 34.20 Sd. 0.58 2.31 1.10 1131. Figure 16: MDSC Curves for PP Resin Salols: 6501 HOHOPOLYIIER D S C F1 1e: C: PP6501.001 Size: 10.1000 e9 Operator: 89R Method: +/- IC/MIN. SC/NIN 300’C Run Date: l7-Anr-95 10' 06 Consent: 50 ML/MIN N2 ounce. MODULATED oI-I’CIMIN. 0.0 0.2 149.45'C -0'2e 73.11J/0 0.7-'-0.0 4 3 1h \ 3 3 127.39'c 154.50'c 5 ; ; -O 4-4 101.2.J/o . 0 Sac-0.2 " ‘1 g s _. E s O I 4 1» H n 4.! IL I 0 u .. i’ z o -o.6- . ° 3“'°~‘ I " e r L > C I) o I: 4 z 1)- (51.35.: 161.68°C -O 8'4 127.36%: 352753”: 0.1-1I--0.6 36 . sou/o 5 - 055d“) . i .. 166.01'C -1 o 55 60 5'0 ‘ 260 2' Y 00-03 1 1 Teaseraturel (.0 68¢ v1.11 TA ans: 2200 Senate: PP HONOPOLYMER 6501 D S C File: C: 8PR6501.001 Size: “.3000 so Operator: BPR FOR SUEDMIAN Method: MOD. RAMP -60 to 200°C Run Oats: 13-Jul-95 03: 34 Consent: SONL/MIN N2 04 "-0.4 0 2‘ 150.75'C 0.2-+ 124.24'C .5 ‘ 5.47'c seam/g K 2 a up \ O \ 1"0.2 3 0 o- 7.93 C”) \ ; " a— -— . 1 ’_'_%- o I S 4 ’\_——— Mu£—-———fi 1 \I E 0.0-‘5 E _ \. U u. - 521°C 3 .. g -0 2-1 r IP00 2 E 8.37'CII) > I . 149.6I°c 2 > :; “‘~-\.\. ”‘55 C 69.84.1/9 g -0.2~- &’ ‘0.4-‘ \\\ ‘T I I "'"0.2 i I i “ -o 5_ 124.co'c '0‘" 96.31J/o I --.. 153.41'c I] -0 a 00 30 V 6 55 do so 60 ' ‘o -I - I I .) Temperature» (°C) HOS VI - IA TA fins! 2200 “DIE p~oaun~aoo= A A A A A J A A A A I I I I I V I ‘— U APPENDIX B ESCA Analysis mm m ear-ea. amassi- MWhmlfimh-(mlfssteo unmana- Inuux’ms usual: main-eels permane- LISIIeIa. m: usual: I. as: A. A A A A 4L 9 Y V V I V i 4D A A A A A A A A I U U U U i U U U V J A A A A A A A A TUIUT'FI' museums Figure 17: ESCA Analysis for Nonsulfonated PP Resin 102 MINE “DIE 103 “SIM 012/5 ear-ea. amass-Ia FILE: was hlmlsemgilfestedsun unmanne- 42§kcls “SEE 8.838kcls Hammad $1202: mnem- L874kcls. m: 8.me h an l A a 1 L 1 A V I V a ' V V v v 1 r A A A A A A A V r V V V V A A. A J NUAUIONQGIS A AA A A A V r V I V V U T I ~ A A A L l 1 A f V t I A A A A V V V V manna O 1 C Figure 18: ESCA Analysis for 5-min Sulfonated PP Resin new «we ”£86“ Inna-0.9m FIE: was POW!“ * mum Us“. ”EMF” 4.19180]; m 1.13003 mun-meld mamrm Luke’s. m AMIeh h ”I A L A L A A A L J V V r I *r V V I A A A A A r V V V I t V V r V T V V A A A A A A L A AAAAAA A Isivvrsrrru A A. LJ A A A A A T V I I T V V V—T V V unuauebsaeeo: A L L A A A V V V V O r C A I V V V 1' ‘ as ”MEN Figure 19: ESCA Analysis for 8-min Sulfonated PP Resin MDIE KDIE 104 [SCAM VII/Si ”If-'6“ names.“ sin F111;?“ PolmluMSulfmtd I‘lesIn (II toe) “EMFEIC- lfikdt.‘ 1.148de ”MY-383.6809 mazmgmma 8.974de UPSET 8.287de N a" 1. J ‘ 1; I i I t 3_ b ‘P 3 p ‘r 8 ‘b h 7 d)- s «I- II 5 eh -I d)- 4 +- 3 '1)- ‘- 2 a» q- 1 1'- A A A A A J A L o I v v I 17 I f v v v V I” u m n m 8 m ”I. IV Figure 20: ESCA Analysis for 10-min Sulfonated PP Resin [SCAM 5mm “£86“ momma.» menace hlmlumfillfaltd anaemia: alumnus: lake’s. m: 1.311de "SWEAR“. mzmrm 0.9de ms assumes" ‘ A A A A A A A 1 ' V v I s ‘r y ' ' 1 18 ‘ 4h 4b ‘. 9 1b .- 1. ‘p . ‘ I. 7 7 ar- ‘ -r- S " .. . v S ‘ .I- ‘ 1P 4 . t .. 3 « i .. 1_ 2 cu- up 4 «II- A ‘ I " a o J. 4 4. . I“ u can so am 8 mm am. 0V Figure 21: ESCA Analysis for 20-min Sulfonated PP Resin IUPPEDUDIXII: Density Measurement Table 15: Data of Density (g/cc) Material Replications 1 2 3 4 5 NS 1.055 1.069 1.058 1.029 1.040 1.036 1.062 1.041 NSn 1.052 1.018 1.042 1.085 1.045 1.079 1.059 1.046 1.047 85 1.065 1.069 1.044 1.072 1.040 1.063 1.029 1.028 1.070 1.048 88 1.072 1.042 1.053 1.045 1.050 1.048 1.048 1.045 1.063 1.034 810 1.044 1.077 1.071 1.052 1.058 1.057 1.064 1.073 1.075 1.074 810n 1.041 1.075 1.043 1.075 1.055 1.054 1.067 1.083 1.054 1.082 S20n 1.073 1.066 1.089 1.073 1.067 1.078 1.082 1.090 1.085 105 106 Mechanical Properties Table 16: Data of Tensile Strength at Break (MPa) Material Replications 1 2 3 4 5 6 7 N8 (LD) 15.77 16.66 17.33 15.96 17.14 16.86 17.73 14.62 18.04 17.23 13.77 15.02 14.67 17.22 NS (CD) 18.32 15.66 16.27 17.61 15.62 15.21 12.97 15.54 16.54 16.13 14.56 15.11 13.49 14.76 NSn (LD) 16.69 15.55 17.83 16.73 16.07 16.41 16.29 16.67 14.96 17.75 16.98 17.26 14.79 NSn (CD) 14.00 11.81 11.31 14.92 13.10 10.41 12.40 14.76 14.13 15.76 14.55 13.00 13.39 85 (LD) 17.42 19.37 17.77 18.67 17.43 19.00 17.38 21.13 18.96 18.79 19.58 21.69 19.28 21.91 85 (CD) 18.73 16.33 21.24 22.04 20.60 21.84 20.14 18.33 18.35 15.25 17.26 17.40 16.91 16.20 88 (LB) 15.32 19.06 18.50 19.11 18.60 18.50 18.18 20.77 17.44 20.22 20.63 20.46 20.63 19.10 88 (on) 18.96 18.02 19.34 16.13 17.50 17.15 16.38 17.90 18.38 20.20 17.34 17.78 17.38 19.68 810 (LD) 14.20 21.13 18.49 21.46 18.69 18.30 20.73 18.60 19.28 17.60 17.23 18.51 18.55 19.64 810 (CD) 17.80 17.78 16.88 16.73 17.82 16.62 16.48 16.38 13.60 15.80 20.14 20.35 17.27 16.10 810n (LD) 18.38 18.04 16.62 17.64 18.00 15.93 17.90 15.70 19.28 20.88 16.68 17.80 19.02 810n (CD) 19.71 14.91 19.24 14.45 12.49 19.40 14.27 13.81 17.13 17.37 15.43 15.77 14.11 S20n (LD) 19.64 19.01 18.94 20.46 20.24 21.22 18.78 19.44 16.56 18.40 19.90 19.32 19.12 S20n (CD) 18.69 20.21 19.09 17.92 18.95 18.91 15.36 17.13 21.40 21.28 19.26 20.64 20.09 PP5 33.90 33.10 33.03 33.61 33.37 33.26 33.46 33.54 33.15 33.03 33.04 33.20 33.28 33.50 5 Reported as tensile strength at yield 107 Table 17: Data of Percent Elongation at Break Material Replications 1 2 3 4 5 6 7 NB (DD) 1.00 3.22 4.59 4.29 1.73 4.35 3.65 3.59 2.68 1.52 2.13 2.72 2.47 2.67 N8 (CD) 1.93 2.45 2.88 1.79 1.53 2.65 1.08 1.42 1.59 1.03 1.62 0.10 1.90 NSn (DD) 4.75 5.66 3.23 5.59 4.04 4.78 5.43 3.05 5.35 4.66 3.91 4.29 6.81 3.84 NSn (CD) 2.73 2.13 1.50 2.45 2.56 2.07 2.22 3.80 2.78 1.72 3.05 2.46 2.60 1.40 85 (DD) 1.20 2.11 1.21 0.92 0.93 1.26 1.13 1.36 1.74 1.30 2.36 1.88 1.15 85 (CD) 0.75 0.73 1.09 1.15 1.40 1.55 1.44 1.33 1.27 0.63 1.09 1.02 0.88 0.93 88 (DD) 0.77 1.07 1.14 1.19 1.32 1.31 1.29 1.54 1.29 1.59 1.59 1.84 1.67 1.23 88 (CD) 1.82 1.21 1.72 1.29 1.18 1.27 0.98 1.68 0.92 1.60 1.10 0.98 0.98 1.60 810 (DD) 0.57 2.12 1.40 1.73 1.49 1.55 1.72 1.85 2.29 1.61 1.54 2.39 1.58 2.23 810 (CD) 1.50 0.89 1.25 1.77 1.15 1.59 0.96 0.83 0.95 2.98 1.52 0.81 0.73 810n (DD) 2.00 1.85 1.89 3.17 2.09 1.41 1.95 3.07 1.67 3.02 2.69 1.52 2.32 3.00 810n (CD) 2.20 1.44 2.01 1.57 1.02 2.45 1.24 2.11 1.60 2.44 2.69 2.35 2.37 1.63 820n (DD) 1.29 2.07 1.66 2.11 1.66 1.85 1.84 2.03 1.87 1.03 1.65 1.83 2.02 1.64 820n (CD) 1.87 2.23 1.45 1.26 1.59 1.93 0.89 1.88 1.63 2.03 1.84 2.02 1.54 1.84 PP‘ 9.36 7.99 9.03 9.08 8.97 8.97 8.74 8.84 8.93 8.93 9.17 8.36 8.51 9.06 ‘fineported as percent elongtion at yield 108 Table 18: Data of Modulus of Elasticity (MPa) Iaterial Replications 1 2 3 4 5 6 7 88 (DD) 2930 2618 3241 3354 3912 3478 3048 3724 2921 3437 3389 3058 3415 3409 88 (CD) 2921 2839 3143 3461 3268 3102 2982 2655 3336 2981 3249 3519 2682 NSn (DD) 2664 3295 2695 3479 2862 3590 2198 2770 2566 3084 2752 2617 2834 2576 NSn (CD) 3059 2212 3312 1846 2313 2230 2740 2406 2095 1992 3123 2835 2247 85 (DD) 3779 4061 3736 3547 4385 3930 2848 2807 2788 3068 2706 2760 3889 85 (CD) 2795 2723 3229 3526 3652 2966 2832 2132 4356 3232 2927 2917 3302 88 (DD) 3385 3163 3154 3287 3969 3221 2963 3452 4346 3771 2853 3802 3510 3732 88 (CD) 3912 4304 3945 4006 4059 3301 2824 2921 3478 3093 3215 3269 3006 810 (DD) 3254 3225 3763 3047 3257 2977 3410 2496 2737 2805 2545 2441 2972 3142 810 (CD) 2755 3336 3265 3621 3476 3626 3525 2839 2106 2495 2588 2440 3371 810n (DD) 2981 2758 2686 2795 2657 2661 3056 2762 2717 2553 2753 3011 2748 810n (CD) 2595 3138 2877 2761 2067 2963 2985 2832 3306 2597 3130 2597 3130 2849 820n (DD) 2615 3128 2142 3562 3189 2624 2737 2493 2730 3607 3165 2755 3299 820n (CD) 3456 2205 3467 2479 2852 2647 2821 2282 3072 2730 2783 2884 3030 PP 2215 1705 1624 1717 1788 1684 1697 1616 1933 1669 1672 1735 1649 1449 Table 19: Data of Flexural Strength (MPa) 109 laterial Replications 1 2 3 4 5 6 7 8 88 (DD) 39.01 39.32 34.21 41.39 44.52 34.38 40.93 35.90 42.86 40.38 36.69 38.13 36.49 42.07 36.15 34.25 N8 (CD) 33.25 31.00 31.41 34.48 33.12 36.51 36.91 35.62 43.08 44.23 41.86 39.32 42.75 36.88 38.34 38.97 NSn (DD) 38.34 40.70 38.33 41.02 40.65 39.06 40.03 45.84 43.93 36.47 42.75 38.35 39.34 44.69 40.23 36.72 NSn (CD) 43.26 26.59 37.01 35.75 36.96 30.40 27.60 40.33 35.99 36.85 31.57 35.30 35.52 32.05 35.87 33.39 85 (DD) 43.71 41.82 39.71 37.81 46.00 38.43 42.66 40.84 40.10 41.98 42.36 45.42 85 (CD) 38.04 42.53 39.29 36.89 40.55 38.56 41.80 35.10 43.27 34.73 45.32 33.65 41.62 39.74 39.82 43.84 88 (DD) 32.58 39.83 39.40 38.26 38.52 36.47 34.47 35.46 43.00 43.99 43.47 42.34 42.60 42.67 44.60 43.21 88 (CD) 37.02 41.22 42.45 42.29 40.20 35.74 33.64 43.73 41.91 40.83 40.69 40.82 40.84 40.71 40.00 42.26 810 (DD) 44.51 39.80 43.36 45.78 39.94 45.71 37.58 41.25 49.44 44.01 45.33 38.60 43.43 38.15 46.27 38.10 810 (CD) 36.85 42.47 42.09 41.25 43.85 43.60 38.58 39.26 32.68 38.93 43.33 33.35 39.32 36.55 40.84 36.78 810n (DD) 47.35 37.16 38.10 44.30 45.79 41.91 40.52 44.55 45.16 38.43 44.43 39.29 46.22 41.65 45.99 38.96 810n (CD) 35.98 29.74 39.41 40.09 44.49 41.96 40.45 41.08 41.96 43.23 43.81 43.81 41.91 40.52 43.80 44.18 820n (DD) 40.29 41.05 37.80 42.19 38.67 44.52 40.11 42.70 39.36 40.59 45.85 39.89 42.81 38.25 44.87 45.13 820n (CD) 44.35 44.49 38.07 44.62 30.70 44.46 43.71 44.34 44.04 45.60 43.77 43.60 48.18 41.58 43.49 47.49 110 Table 20: Data of Flexural Modulus (MPa) Material Replications 1 2 3 4 5 6 7 8 M8 (DD) 2993 3393 2953 2784 3473 3259 2990 2914 3111 2985 2879 2822 3113 2938 2855 2809 M8 (CD) 2482 2486 2288 2623 2826 2586 2488 2582 2828 3217 3050 2857 3029 2705 2643 2811 NSn (DD) 1858 2750 2224 2847 2442 2536 2470 2481 2831 2627 2860 2565 2351 3100 3082 2733 M8n (CD) 2568 2100 2354 2226 2406 2511 2180 2323 2441 2328 2187 2415 2388 2475 2562 2525 85 (DD) 3102 2976 2846 2996 3263 2924 3093 2882 2955 2983 2953 2969 85 (CD) 3092 2958 3403 2766 3013 2955 3177 3280 3064 3029 3150 3093 3269 88 (DD) 2681 2826 2706 2783 3006 2657 2871 2597 3106 3131 3282 3022 2938 3107 3073 3050 88 (CD) 2826 3130 2986 2994 2875 2760 2331 3234 3003 2841 2904 3002 3112 2558 2822 3023 810 (DD) 3196 2905 2824 2858 2981 3253 3166 2805 3188 2934 3236 2872 2905 2943 2997 2768 810 (CD) 2683 2924 2785 2756 2994 2887 2745 2854 2626 2965 2753 2447 3108 2805 2979 2852 810n (DD) 3050 2713 2626 2990 2613 2695 2894 2951 3192 2717 2925 2811 2888 2792 2736 2597 810n (CD) 2433 2434 2778 2691 2840 2905 2821 2722 2722 2828 2732 2715 2766 2498 2638 2793 820n (DD) 2603 2734 2419 2897 2510 2871 2460 2865 2763 3017 2457 2604 2799 2755 2997 2766 820n (CD) 2722 2905 2728 2866 2426 2772 2938 2796 2503 2842 2810 2557 3126 2773 2897 3134 PP 1367 1499 1895 1440 1384 1363 1479 1392 1507 1309 1410 1492 1415 1404 111 Table 21: Data of Izod Impact Strength (J/m) Material Replications 1 2 3 4 5 6 7 8 M8 (DD) 20.02 23.49 20.39 21.67 20.55 19.70 17.88 20.39 20.02 21.46 18.31 19.86 26.32 20.71 18.63 21.03 N8 (CD) 22.05 22.26 21.89 22.21 23.65 23.49 24.45 22.21 22.05 23.49 23.49 22.05 21.89 21.67 22.05 21.67 NSn (DD) 30.64 35.82 30.64 34.22 32.94 30.86 36.41 36.68 31.87 30.86 38.17 32.03 29.84 33.05 34.49 33.58 NSn (CD) 35.00 32.07 32.55 33.00 29.95 37.85 33.05 34.25 36.35 34.98 35.31 37.97 36.50 32.25 38.56 30.06 85 (DD) 21.73 19.22 21.51 21.03 22.42 18.26 21.67 22.26 21.19 19.38 19.54 20.39 21.51 18.63 20.39 21.67 85 (CD) 19.70 23.65 21.51 23.49 21.03 21.67 20.15 19.54 21.67 18.52 18.52 20.23 19.70 19.86 18.31 19.70 88 (DD) 25.52 20.34 24.45 22.42 20.02 24.45 20.34 23.38 18.95 20.39 20.39 20.71 20.39 20.02 24.66 19.11 88 (CD) 19.54 19.54 19.70 19.70 21.19 20.02 17.72 19.54 21.09 20.39 20.71 18.95 20.02 19.27 20.87 18.63 810 (DD) 26.80 19.91 25.95 28.24 25.57 24.29 24.29 21.19 26.16 25.41 25.57 28.72 29.90 34.17 29.42 31.12 810 (CD) 23.12 27.65 29.04 24.29 25.04 24.50 25.25 27.23 25.47 24.08 25.68 22.21 19.11 22.05 19.70 21.89 810n (DD) 21.19 22.05 23.12 20.39 23.12 26.32 27.44 21.67 26.37 25.47 29.42 26.59 25.04 24.45 23.54 26.37 810n (CD) 27.39 26.96 27.17 27.39 24.08 23.65 25.47 24.08 24.17 24.82 25.47 25.47 23.49 25.47 25.63 25.63 S20n (DD) 21.19 25.04 21.19 20.66 24.29 20.82 22.74 19.22 22.58 19.38 21.51 19.54 21.51 23.12 19.91 18.95 S20n (CD) 20.66 18.95 20.82 19.38 19.22 19.22 22.74 20.98 17.46 20.71 18.95 17.46 17.03 17.19 18.95 18.79 PP 19.33 19.11 20.98 30.38 20.82 20.34 18.95 21.51 17.56 17.03 17.56 21.09 19.07 23.12 18.79 22.21 .APPEBHIEK D Water Sorption Studies Table 22: Data of Weight (gram) Measured At Different Periods of Storage Time (1) nonsulfonated Composite (M8) Replications Storage Tine O-week 1-week 3-week S-week 7-week 9-week 1 6.0906 6.1517 6.1937 6.2212 6.2438 6.2432 2 6.0761 6.1340 6.1741 6.2008 6.2213 6.2213 3 6.5440 6.6100 6.6554 6.6862 6.7110 6.7108 4 6.4253 6.5010 6.5515 6.5868 6.6127 6.6110 5 6.2861 6.3417 6.3790 6.4047 6.4245 6.4272 6 6.0410 6.1018 6.1419 6.1693 6.1911 6.1930 7 6.0583 6.1167 6.1563 6.1836 6.2033 6.2052 8 6.2035 6.2662 6.3079 6.3356 6.3578 9 6.1895 6.2443 6.2823 6.3080 6.3276 10 6.3592 6.4298 6.4770 6.5077 6.5330 11 6.2465 6.3164 6.3639 6.3958 6.4204 12 5.9360 5.9926 6.0313 6.0570 6.0771 13 6.2775 6.3379 6.3788 6.4068 6.4285 14 6.0421 6.1025 6.1430 6.1705 6.1920 15 6.4207 6.4909 6.5380 6.5705 16 6.8378 6.9176 6.9741 7.0128 17 6.3910 6.4544 6.4977 6.5272 18 6.2788 6.3383 6.3788 6.4061 19 6.2489 6.3063 6.3448 6.3717 20 6.3366 6.3987 6.4410 6.4704 21 6.2465 6.3219 6.3716 6.4063 22 6.2538 6.3297 6.3786 23 6.2216 6.2950 6.3446 24 6.5055 6.5718 6.6160 25 6.2697 6.3268 6.3662 26 6.2767 6.3338 6.3729 27 6.0933 6.1514 6.1914 112 113 (2) S-ndn Sulfonated Composite (85) Replications Storage Tine O-week 1-week 3-week S-veek 7-week 9-week 1 6.0010 6.0319 6.0678 6.0916 6.1103 6.1110 2 6.0093 6.0637 6.0990 6.1237 6.1427 6.1443 3 6.1869 6.2392 6.2741 6.2979 6.3170 6.3185 4 5.9342 5.9858 6.0190 6.0418 6.0595 6.0604 5 6.0876 6.1419 6.1779 6.2025 6.2218 6.2230 6 6.1608 6.2141 6.2488 6.2735 6.2920 6.2941 7 5.9147 5.9655 5.9991 6.0224 6.0390 6.0407 8 6.2261 6.2855 6.3230 6.3493 6.3693 9 6.0847 6.1413 6.1777 6.2034 6.2221 10 6.2446 6.3051 6.3435 6.3714 6.3921 11 6.4730 6.5381 6.5805 6.6112 6.6339 12 6.0179 6.0713 6.1064 6.1317 6.1504 13 6.0783 6.1380 6.1765 6.2040 6.2242 14 6.0458 6.1048 6.1428 6.1693 6.1892 15 6.3256 6.3807 6.4169 6.4425 16 6.1024 6.1614 6.2013 6.2289 17 6.3000 6.3654 6.4081 6.4388 18 6.5018 6.5591 6.5962 6.6231 19 6.1107 6.1587 6.1900 6.2125 20 6.0800 6.1409 6.1804 6.2075 21 6.0699 6.1265 6.1620 22 6.2542 6.3172 6.3580 23 6.3412 6.4013 6.4402 24 6.3292 6.3889 6.4280 25 6.1843 6.2348 6.2672 26 6.2841 6.3393 6.3759 27 5.9759 6.0225 6.0530 114 (3) 8-nin Sulfonated Composite (88) Replications Storage Tine O-week 1-week 3-week S-week 7-week 9-voek 1 6.2813 6.3409 6.3813 6.4100 6.4316 6.4335 2 5.9230 5.9870 6.0288 6.0577 6.0796 6.0803 3 5.9024 5.9654 6.0059 6.0342 6.0550 6.0560 4 6.2556 6.3134 6.3503 6.3780 6.3982 6.4015 5 6.3635 6.4221 6.4611 6.4875 6.5080 6.5112 6 6.3500 6.4080 6.4468 6.4739 6.4944 6.4979 7 5.9098 5.9587 5.9889 6.0103 6.0266 6.0288 8 6.0521 6.1128 6.1523 6.1800 6.2003 9 6.2403 6.2957 6.3324 6.3590 6.3782 10 6.0763 6.1313 6.1660 6.1923 6.2116 11 6.2480 6.3079 6.3480 6.3758 6.3967 12 6.0398 6.0885 6.1192 6.1412 6.1577 13 6.2629 6.3100 6.3417 6.3640 6.3803 14 6.1415 6.2004 6.2379 6.2655 6.2844 15 6.2633 6.3117 6.3420 6.3647 16 6.2233 6.2853 6.3253 6.3550 17 5.9870 6.0435 6.0804 6.1069 18 6.2974 6.3556 6.3935 6.4216 19 6.1140 6.1681 6.2041 6.2294 20 5.9910 6.0495 6.0878 6.1154 21 6.5195 6.5823 6.6233 6.6526 22 6.1674 6.2272 6.2681 23 6.1217 6.1773 6.2154 24 6.1023 6.1672 6.2088 25 6.3048 6.3633 6.4023 26 6.3557 6.4145 6.4541 27 6.1514 6.2129 6.2538 28 6.0586 6.1116 6.1486 115 (4) lO-nin Sulfonated Composite (810) Replications Storage Tine O-veek 1-week 3-week 5-week 7-week 9-week 1 6.5070 6.5803 6.6248 6.6514 6.6698 6.6687 2 6.3915 6.4615 6.5037 6.5285 6.5463 6.5440 3 6.5896 6.6595 6.7035 6.7291 6.7483 6.7473 4 6.1678 6.2263 6.2632 6.2871 6.3036 6.3050 5 6.4063 6.4796 6.5231 6.5499 6.5692 6.5668 6 6.4013 6.4727 6.5173 6.5489 6.5723 6.5767 7 6.4526 6.5317 6.5764 6.6022 6.6191 6.6159 8 6.0900 6.1473 6.1839 6.2084 6.2256 9 6.6830 6.7527 6.7946 6.8218 6.8407 10 6.8124 6.8890 6.9381 6.9732 6.9995 11 6.3158 6.3897 6.4324 6.4577 6.4745 12 6.1092 6.1679 6.2044 6.2283 6.2461 13 6.4085 6.4784 6.5218 6.5484 6.5659 14 6.5410 6.6166 6.6644 6.6985 6.7236 15 6.2873 6.3660 6.4125 6.4385 16 6.6730 6.7425 6.7861 6.8137 17 6.4914 6.5612 6.6046 6.6304 18 6.4502 6.5129 6.5537 6.5790 19 6.3881 6.4581 6.5018 6.5279 20 6.4860 6.5637 6.6094 6.6364 21 6.5246 6.5962 6.6413 6.6727 22 6.8015 6.8810 6.9314 23 6.2062 6.2689 6.3089 24 6.3235 6.3812 6.4182 25 6.3150 6.3823 6.4254 26 6.3410 6.4133 6.4571 27 6.4272 6.4998 6.5481 28 6.4993 6.5618 6.6015 (5) Polypropylene (PP) 116 Replications Storage Time O-week 1-week 3-week S-week 7-week 9-week 1 4.9072 4.9050 4.9044 4.9052 4.9037 4.9052 2 4.9791 4.9756 4.9755 4.9797 4.9758 4.9781 3 5.0060 5.0025 5.0025 5.0024 5.0029 5.0036 4 5.1055 5.1015 5.1016 5.1041 5.1019 5.1020 5 4.9510 4.9467 4.9466 4.9462 4.9466 4.9465 6 4.7381 4.7328 4.7332 4.7329 4.7328 4.7346 7 4.9540 4.9512 4.9510 4.9508 4.9507 4.9509 8 4.7261 4.7220 4.7223 4.7219 4.7224 9 4.8114 4.8077 4.8080 4.8079 4.8079 10 4.9870 4.9825 4.9824 4.9825 4.9823 11 4.9175 4.9134 4.9133 4.9128 4.9129 12 4.7848 4.7813 4.7808 4.7808 4.7809 13 4.7872 4.7835 4.7830 4.7845 4.7838 14 4.8271 4.8238 4.8228 4.8228 15 4.9742 4.9704 4.9710 4.9698 16 4.6509 4.6470 4.6467 4.6480 17 4.5807 4.5773 4.5773 4.5769 18 4.8224 4.8193 4.8181 4.8175 19 4.6152 4.6120 4.6112 4.6112 20 4.8994 4.8958 4.8958 21 4.6940 4.6898 4.6902 22 4.8678 4.8640 4.8640 23 4.5773 4.5734 4.5735 24 4.6348 4.6315 4.6315 25 4.7038 4.7006 4.7003 117 Table 23: Data of Tensile Strength (MPa) Measured After Different Periods of Storage Time Material Replications 1 2 3 4 5 6 7 After 3-week Storage MS 16.79 18.40 17.23 16.38 15.47 16.71 16.37 85 17.45 18.52 16.78 16.22 21.02 19.58 20.23 88 18.20 19.02 16.41 18.73 16.73 20.25 16.57 810 20.16 18.92 18.46 20.87 15.87 18.85 PP 35.22 36.03 35.65 35.35 34.95 After S-week Storage M8 16.55 13.45 16.13 18.75 15.03 16.58 19.01 85 19.17 18.02 18.83 18.79 19.60 16.91 88 19.66 17.31 18.54 21.14 19.56 16.47 16.83 810 18.76 19.06 15.36 19.27 PP 33.48 33.36 33.67 33.23 34.05 33.56 After 7-week Storage NS 16.78 16.82 17.38 15.42 16.20 15.59 14.90 85 17.69 20.32 18.70 19.38 20.13 19.12 17.95 88 16.63 18.48 16.42 19.23 19.17 810 20.55 16.33 18.21 17.40 19.28 16.74 19.01 PP 35.90 35.29 35.61 35.54 35.20 35.03 After 9-week storage M8 16.20 17.70 17.35 14.27 19.26 14.81 15.02 85 19.06 18.85 19.07 19.43 18.44 88 17.27 16.49 17.13 16.49 19.01 18.29 18.62 810 16.67 17.43 17.42 19.39 17.06 PP 34.70 34.05 34.23 34.28 34.49 34.69 .APPEQHIEX E Statistical Analysis Table 24: One-way Analysis of Variance of Density Values Analysis of Variance Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 6 .0062 .0010 4.9907 .0003 Within Groups 59 .0123 .0002 Total 65 .0185 Group Count Mean Standard Standard 95% Confidence Deviation Error Interval for Mean NS 8 1.0486 .0141 .0050 1.0367 To 1.0604 NSn 9 1.0525 .0202 .0067 1.0369 To 1.0680 85 10 1.0529 .0170 .0054 1.0407 To 1.0650 88 10: 1.0500 .0108 .0034 1.0422 To 1.0577 810 10 1.0644 .0114 .0036 1.0563 To 1.0726 810n 10 1.0629 .0155 .0049 1.0519 To 1.0740 S20n 9 1.0783 .0090 .0030 1.0714 To 1.0852 Total 66 1.0586 .0169 .0021 1.0544 To 1.0627 Multiple Range Tests: Tukey-BSD test with significance level .050 The difference between two means is significant if MEAN(J)-MEAM(I) >- .0102 * RANGE * SQRT(1/M(I) + 1/N(J)) with the following value(s) for RANGE: 4.32 MS 88 NSn SS 810 S20n NS 88 NSn 85 810n 810 S20n * * * (*) Indicates significant differences 118 119 Table 25: One-way Analysis of variance of Tensile Strength at Break Data, for Lengthwise Direction Analysis of Variance Source D.P. Sum of Mean 8 Ratio P Prob. Squares Squares Between Groups 6 134.9586 22.4931 11.3272 .0000 Within Groups 88 174.7470 1.9858 Total 94 309.7056 Group Count Mean Standard Standard 95% Confidence Deviation Error Interval for Mean N8 14 16.2871 1.3274 .3548 15.5207 To 17.0536 NSn 13 16.4600 .9430 .2616 15.8901 To 17.0299 85 14 19.1700 1.5100 .4036 18.2982 To 20.0418 88 14 19.0371 1.5034 .4018 18.1691 TO 19.9052 810 14 18.7436 1.8130 .4845 17.6968 TO 19.7904 810n 13 17.8362 1.4193 .3936 16.9785 TO 18.6938 S20n 13 19.3100 1.1271 .3126 18.6289 TO 19.9911 Total 95 18.1285 1.8151 .1862 17.7588 TO 18.4983 Multiple Range Tests: Tukey-BSD test with significance level .050 The difference between two means is significant if IIAN(J)-MIAN(I) >s .9964 * RANGE * SQRT(1/N(I) + 1/N(J)) with the following value(s) for RANGE: 4.27 NS NSn 810n 810 88 85 S20n N8 NSn 810n 810 * * 33 4 4 35 4 4 820n * * (*) Indicates significant differences Table 26: One-Nay Analysis of Variance of Tensile Strength at Break Data, Analysis of Variance for Crosswise Direction Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 6 324.2102 54.0350 17.2030 .0000 Within Groups 88 276.4098 3.1410 Total 94 600.6200 Group Count Mean Standard Standard 95% Confidence Deviation Error Interval for Mean NS 14 15.5564 1.4299 .3822 14.7308 To 16.3820 NSn 13 13.3492 1.5545 .4311 12.4099 TO 14.2886 85 14 18.6157 2.2196 .5932 17.3341 TO 19.8973 88 14 18.0100 1.1928 .3188 17.3213 To 18.6987 810 14 17.1250 1.6996 .4542 16.1437 TO 18.1063 810n 13 16.0069 2.3563 .6535 14.5830 TO 17.4308 S20n 13 19.1485 1.6859 .4676 18.1297 To 20.1673 Total 95 16.8512 2.5278 .2593 16.3362 TO 17.3661 Multiple Range Tests: Tukey-BSD test with significance level .050 The difference between two means is significant if IIAN(J)-IIAN(I) >- 1.2532 * BANG! * SQRT(1/N(I) + 1/N(J)) with the following value(s) for RANGE: 4.27 810n 810 85 S20n NSn Ns 810n 810 88 85 820n it'd). * (*) Indicates significant differences 121 Table 27: One-Way Analysis of Variance of Tensile Strength at Break Data in Lengthwise Direction Vs. Crosswise Direction (1)Nonsulfonated PP Composite of Fiber-1 (NS) Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 1 3.7376 3.7376 1.9637 .1729 Within Groups 26 49.4876 1.9034 Total 27 53.2252 (2)Nonsu1fonated PP Composite of Fiber-2 (NSn) Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 1 62.8998 62.8998 38.0541 .0000 Within Groups 24 39.6697 1.6529 Total 25 102.5694 (3)5-min Sulfonated PP Composite of Fiber-1 (S5) Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 1 2.1506 2.1506 .5968 .4468 Within Groups 26 93.6897 3.6035 Total 27 95.8404 (4)8-min Sulfonated PP Composite of Fiber-1 (S8) Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 1 7.3852 7.3852 4.0106 .0557 Within Groups 26 47.8767 1.8414 Total 27 55.2618 122 (5)10-mdn.Sulfonated PP Composite of Piber-l (810) Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 1 18.3384 18.3384 5.9389 .0220 Within Groups 26 80.2841 3.0878 Total 27 98.6225 (6)10-mdn.8u1fonated PP Composite of Fiber-2 (810n) Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 1 21.7496 21.7496 5.7490 .0246 Within Groups 24 90.7960 3.7832 Total 25 112.5455 (7)20-min Sulfonated PP Composite of Fiber-2 (820n) Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 1 .1696 .1696 .0825 .7764 Within Groups 24 49.3530 2.0564 Total 25 49.5226 123 Table 28: One-Nay Analysis of Variance of Percent Elongation at Break Data, for Lengthwise Direction Analysis of variance Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 6 115.3145 19.2191 41.0458 .0000 Within Groups 90 42.1411 .4682 Total 96 157.4556 Group Count Mean Standard Standard 95% Confidence Deviation Error' Interval for Mean NS 14 2.9007 1.1034 .2949 2.2636 TO 3.5378 NSn 14 4.6707 1.0367 .2771 4.0722 TO 5.2693 85 13 1.4269 .4524 .1255 1.1535 TO 1.7003 88 14 1.3457 .2776 .0742 1.1854 TO 1.5060 810 14 1.7193 .4626 .1236 1.4522 TO 1.9864 810n 14 2.2607 .6148 .1643 1.9057 TO 2.6157 S20n 14 1.7536 .3026 .0809 1.5789 TO 1.9283 Total 97 2.3058 1.2807 .1300 2.0477 To 2.5639 Multiple Range Tests: Tukey-BSD test with significance level .050 The difference between two means is significant if MEAN(J)-MEAN(I) >- .4839 * RANGE * SQRT(1/N(I) + 1/N(J)) with the following value(s) for RANGE: 4.27 88 85 810 S20n 810n NS NSn 88 85 810 S20n 810n * * ‘8 s s e e "an t t t e t * (*) Indicates significant differences 124 Table 29: One-Way Analysis of Variance of Percent elongation at Break Data, for Crosswise Direction Analysis of Variance Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 6 16.4832 2.7472 10.4036 .0000 Within Groups 89 23.5016 .2641 Total 95 39.9848 Group Count Mean Standard Standard 95% Confidence Deviation Error Interval for Mean NS 13 1.6900 .7346 .2038 1.2461 To 2.1339 NSn 14 2.3907 .6320 .1689 2.0258 To 2.7556 85 14 1.0900 .2849 .0761 .9255 To 1.2545 88 14 1.3093 .3144 .0840 1.1278 T0 1.4908 810 13 1.3023 .6078 .1686 .9350 TO 1.6696 810n 14 1.9371 .5160 .1379 1.6392 To 2.2351 820n 14 1.7143 .3506 .0937 1.5119 To 1.9167 Total 96 1.6363 .6488 .0662 1.5048 TO 1.7677 Multiple Range Tests: Tukey-BSD test with significance level .050 The difference between two means is significant if MlAN(J)-MIAN(I) >- .3634 * RANGE * SQRT(1/N(I) + 1/N(J)) with the following value(s) for RANGE: 4.27 85 810 88 NS S20n 810n NSn 85 810 88 NS * S20n * 810n * * * "an t '8‘ e s * (*) Indicates significant differences 125 Table 30: One-way Analysis of Variance of Percent Elongation at Break Data in Lengthwise Direction Vs. Crosswise Direction (l)ansu1fonated PP Composite of Piber-l (NS) Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 1 9.8808 9.8808 11.0758 .0027 Within Groups 25 22.3027 .8921 Total 26 32.1835 (2)ansulfonated PP Composite of Fiber-2 (NSn) Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 1 36.3888 36.3888 49.3706 .0000 Within Groups 26 19.1634 .7371 Total 27 55.5522 (3)5-min Sulfonated PP Composite of Fiber-1 (85) Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 1 .7652 .7652 5.4478 .0279 Within Groups 25 3.5115 .1405 Total 26 4.2767 (4)8-min Sulfonated PP Composite of Fiber-1 (S8) Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 1 .0093 .0093 .1056 .7478 Within Groups 26 2.2872 .0880 Total 27 2.2965 126 (5)10-min Sulfonated PP Composite of Fiber-1 (810) Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 1 1.1720 1.1720 4.0613 .0547 Within Groups 25 7.2145 .2886 Total 26 8.3865 (6)10-min Sulfonated PP Composite of Fiber-2 (810n) Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 1 .7329 .7329 2.2751 .1435 Within Groups 26 8.3756 .3221 Total 27 9.1085 (7)20-min Sulfonated PP Composite of Piber-2 (820n) Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 1 .0108 .0108 .1008 .7535 Within Groups 26 2.7879 .1072 Total 27 2.7987 127 Table 31: One-Nay Analysis of Variance of Modulus of Elasticity Data, for Lengthwise Direction Analysis of Variance Source D.P. Sum.of Mean Squares P Ratio P Prob. Squares Between Groups 6 6309613.809 1051602.301 6.4980 .0000 Within Groups 88 14241449.41 161834.6524 Total 94 20551063.22 Group Count Mean Standard Standard 95% Confidence Deviation Error Interval for Mean NS 14 3281.0000 340.4608 90.9920 3084.4238 To 3477.5762 NSn 14 2855.8571 382.8729 102.3271 2634.7929 TO 3076.9214 85 13 3408.0000 594.1968 164.8005 3048.9305 TO 3767.0695 88 14 3472.0000 413.4537 110.5002 3233.2789 TO 3710.7211 810 14 3005.0714 375.7557 100.4249 2788.1165 To 3222.0263 810n 13 2779.8462 148.9820 41.3202 2689.8172 To 2869.8751 820n 13 2926.6154 434.7339 120.5735 2663.9083 To 3189.3225 Total 95 3106.1368 467.5771 47.9724 3010.8865 To 3201.3872 Multiple Range Tests: Tukey-BSD test with significance level .050 The difference between two means is significant if MEAN(J)-MEAN(I) >- 284.4597 * RANGE * SQRT(1/N(I) + 1/N(J)) with the following value(s) for RANGE: 4.27 810n. NSn S20n 810 NS 85 88 810n NSn S20n 810 MS * as e e t 8' t e i e (*) Indicates significant differences 128 Table 32: One-Way Analsis of Variance of Modulus of Elasticity Data, for Crosswise Direction Analysis of Variance Source D.P Sum of Squares Mean Squares P Ratio P Prob. Between Groups 6 7449904.231 1241650.?05 6.5022 .0000 Within Groups 85 16231454.20 190958.2847 Total 91 23681358.43 Group Count Mean Standard Standard 95% Confidence Deviation Error Interval for Mean NS 13 3087.5385 275.8069 76.4951 2920.8700 TO 254.2069 NSn 13 2493.0769 469.2742 130.1532 2209.4974 TO 776.6564 85 13 3122.2308 536.7756 130.1532 2797.8606 TO 446.6010 88 13 3487.1538 496.4146 137.6806 3187.1735 To 787.1342 810 13 3034.0769 517.1182 143.4228 2721.5855 To 346.5683 810n 14 2844.7857 315.4241 84.3006 2662.6653 TO 026.9061 S20n 13 2823.6923 382.8621 106.1868 2592.3311 TO 055.0536 Total 92 2983.1304 510.1320 53.1849 2877.4851 TO 088.7758 Multiple Range Tests: Tukey-MSD test with significance level .050 The difference between two means is significant if NIAN(J)-IIAN(I) >s 308.9970 * RANGE * BQRT(1/N(I) + 1/N(J)) with the following value(s) for RANGE: 4.27 S20n 810n 810 85 NSn S20n 810n 810 NS 85 88 ..I' t i t (*) Indicates significant differences 129 Table 33: One-way Analysis of Variance of Mbdulus of Elasticity Data in Lengthwise Direction Vs. Crosswise Direction (1)ansulfonated PP Composite of Piber-l (NS) Source D.P. Sum.of Mean Squares P Ratio P Prob. Squares Between Groups 1 252288.1766 252288.1766 2.6066 .1190 Within Groups 25 2419709.231 96788.3692 Total 26 2671997.407 (2)Nonsulfonated PP Composite of Fiber-2 (NSn) Source D.P. Sum of Mean Squares P Ratio P Prob. Squares Between Groups 1 887145.4367 887145.4367 4.8762 .0366 Within Groups 25 4548310.637 181932.4255 Total 26 5435456.074 (3)5-min Sulfonated PP Composite of Fiber-1 (85) Source D.P. Sum.of Mean Squares P Ratio P Prob. Squares Between Groups 1 530816.3462 530816.3462 1.6557 .2105 Within Groups 25 7694374.308 320598.9295 Total 26 8225190.654 (4)8-min Sulfonated PP Composite of Piber-l (88) Source D.P. Sum.of Mean Squares P Ratio P Prob. Squares Between Groups 1 1547.9373 1547.9373 .0075 .9318 Within Groups 25 5179401.692 207176.0677 Total 26 5180949.630 130 (5)10-min Sulfonated PP Composite of Fiber-1 (810) Source D.P. Sum of Mean Squares P Ratio P Prob. Squares Between Groups 1 5671.1113 5671.1113 .0281 .8682 Within Groups 25 5044435.852 201777.4341 Total 26 5050106.963 (6)10-min Sulfonated PP Composite of Fiber-2 (SlOn) Source D.P. Sum.of Mean Squares P Ratio P Prob. Squares Between Groups 1 28426.6913 28426.6913 .4556 .5059 Within Groups 25 1559748.049 62389.9220 Total 26 1588174.741 (7)20-mdn Sulfonated PP Composite of Fiber—2 (820n) Source D.P. Sum of Mean Squares P Ratio P Prob. Squares Between Groups 1 68855.5385 68855.5385 .4104 .5279 Within Groups 24 4026923.846 167788.4936 Total 25 4095779.385 131 Table 34: One-Nay Analysis of Variance of Plexural Strength Data, for Lengthwise Direction Analysis of Variance Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 6 202.5037 33.7506 3.3503 .0047 Within Groups 101 1017.4535 10.0738 Total 107 1219.9572 Group Count Mean Standard Standard 95% Confidence Deviation Error Interval for Mean NS 16 38.5425 3.2646 .8162 36.8029 To 40.2821 NSn 16 40.4031 2.7242 .6811 38.9515 To 41.8548 85 12 41.7367 2.5386 .7328 40.1237 TO 43.3496 88 16 40.0544 3.7600 .9400 38.0508 To 42.0579 810 16 42.5787 3.5812 .8953 40.6704 TO 44.4871 810n 16 42.4881 3.3751 .8438 40.6896 To 44.2866 S20n 16 41.5050 2.5829 .6457 40.1287 TO 42.8813 Total 108 41.0184 3.3766 .3249 40.3743 TO 41.6625 Multiple Range Tests: Tukey-ESD test with significance level .050 The difference between two means is significant if MEAN(J)-MEAN(I) >- 2.2443 * RANGE * SQRT(1/N(I) + 1/N(J)) with the following value(s) for RANGE: 4.26 NS 88 NSn S20n 85 810n 810 NS 88 NSn S20n 85 810n * 810 * (*) Indicates significant differences 132 'Table 35: One-way Analysis of Variance of Plexural Strength Data, Analysis of Variance for Crosswise Direction Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 6 723.8900 120.6483 8.7489 .0000 Itithin Groups 105 1447.9593 13.7901 Total 111 2171.8493 Group Count Mean Standard Standard 95% Confidence Deviation Error Interval for Mean NS 16 37.3581 4.1618 1.0405 35.1404 TO 39.5758 NSn 16 34.6525 4.3079 1.0770 32.3570 TO 36.9480 .85 16 39.6719 3.3949 .8487 37.8628 To 41.4809 88 16 40.2719 2.6430 .6608 38.8635 To 41.6803 810 16 39.3581 3.4563 .8641 37.5164 To 41.1998 810n 16 41.0262 3.7367 .9342 39.0351 To 43.0174 820n 16 43.2806 4.0289 1.0072 41.1338 To 45.4275 _;Total 112 39.3742 4.4234 .4180 38.5460 TO 40.2024 Multiple Range Tests: Tukey-BSD test with significance level .050 The difference between two means is significant if IIAN(J)-IIAN(I) >- 2.6258 * RANGE * BQRT(1/N(I) + 1/N(J)) with the following value(s) for RANGE: 4.25 ¥ k NSn NS 810 85 88 810n S20n 0.8. i * 810 SS 810n S20n 7*) Indicates significant differences 133 Table 36: One4way Analysis of Variance of Plexural Strength Data in Lengthwise Direction Vs. Crosswise Direction (l)Nonsulfonated PP Composite of Piber-l (NS) Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 1 11.2220 11.2220 .8022 .3776 Within Groups 30 419.6771 13.9892 Total 31 430.8991 (2)Nonsu1fonated PP Composite of Piber-2 (NSn) Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 1 264.5575 264.5575 20.3664 .0001 Within Groups 30 389.6972 12.9899 Total 31 654.2547 (3)5-min Sulfonated PP Composite of Piber-l (S5) Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 1 29.2345 29.2345 3.1181 .0892 Within Groups 26 243.7699 9.3758 Total 27 273.0044 (4)8-min Sulfonated PP Composite of Fiber-1 (S8) Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 1 .3784 .3784 .0358 .8511 Within Groups 30 316.8446 10.5615 Total 31 317.2231 134 (5)10-min Sulfonated PP Composite of Piber-l (810) Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 1 82.9794 82.9794 6.6997 .0147 Within Groups 30 371.5644 12.3855 Total 31 454.5438 (6)10-min Sulfonated PP Composite of Fiber-2 (SlOn) Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 1 17.0966 17.0966 1.3486 .2547 Within Groups .30 380.3144 12.6771 Total 31 397.4110 (7)20-min Sulfonated PP Composite of Piber-2 (820n) Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 1 25.2228 25.2228 2.2026 .1482 Within Groups .30 343.5451 11.4515 Total 31 368.7678 135 Table 37: One-way Analysis of Variance of Flexural Modulus Data, for Lengthwise Direction Analysis of Variance Source D.P. Sum of Mean P Ratio P'Prob. Squares Squares Between Groups 6 2287990.352 381331.7253 8.9892 .0000 Within Groups 101 4284554.167 42421.3284 Total 107 6572544.519 Group Cbunt Mean Standard Standard 95% Confidence Deviation Error Interval for Mean N8 16 3016.9375 204.9502 51.2376 2907.7272 TO 3126.1478 NSn 16 2609.8125 318.2199 79.5550 2440.2451 To 2779.3799 85 12 2995.1667 111.7764 32.2671 2924.1473 TO 3066.1860 S8 16 2927.2500 201.5181 50.3795 2819.8686 TO 3034.6314 810 16 2989.4375 164.2425 41.0606 2901.9189 To 3076.9561 810n 16 2824.3750 170.0944 42.5236 2733.7381 To 2915.0119 S20n 16 2719.8125 191.5436 47.8859 2617.7461 To 2821.8789 Total 108 2864.2963 247.8420 23.8486 2817.0192 To 2911.5734 Multiple Range Tests: Tukey-BSD test with significance level .050 The difference between two means is significant if MEAN(J)-MEAN(I) >- 145.6388 * RANGE * SQRT(1/N(I) 4 1/N(J)) with the following value(s) for RANGE: 4.26 NSn S20n 810n 88 810 85 NS NSn S20n 810n 88 * 810 * * 35 4 4 N8 4 4 (*) Indicates significant differences 136 Table 38: One-Way Analysis of Variance of Plexural Modulus Data, for Crosswise Direction Analysis of Variance Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 6 4318995.674 719832.6123 20.8023 .0000 Within Groups 102 3529551.61l 34603.4472 Total 108 7848547.284 Group Count Mean Standard Standard 95% Confidence Deviation. Error Interval for Mean NS 16 2718.8125 245.7273 61.4318 2587.8737 To 2849.7513 NSn 16 2374.3125 143.0248 35.7562 2298.1000 TO 2450.5250 85 13 3096.0769 165.2682 45.8372 2996.2064 TO 3195.9475 88 16 2900.0625 221.7778 55.4445 2781.8854 TO 3018.2396 810 16 2822.6875 160.8497 40.2124 2736.9767 TO 2908.3983 810n 16 2707.2500 141.3952 35.3488 2631.9058 TO 2782.5942 S20n 16 2799.6875 193.8142 48.4536 2696.4112 TO 2902.9638 Total 109 2765.2661 269.5770 25.8208 2714.0848 To 2816.4473 Multiple Range Tests: Tukey-ESD test with significance level .050 The difference between two means is significant if KEANE!) -IEAN(I) with the following value(s) for RANGE: 4.25 >- 131.5360 * RANGE * SQRT(1/N(I) + 1/N(J)) 810n S20n 810 85 NSn 810n NS S20n 810 88 85 ..‘I'OQ * 4 (*) Indicates significant differences 137 Table 39: One-Nay Analysis of Variance of Plexural Modulus Data in Lengthwise Direction Vs. Crosswise Direction (l)ansu1fonated PP Composite of Piber-l (NS) Source D.Pu Sum.of Mean Squares P Ratio P Prob. Squares Between Groups 1 711028.1250 711028.1250 13.8891 .0008 Within Groups 30 1535797.375 51193.2458 Total 31 2246825.500 (2)Nonsulfonated PP Composite of Fiber-2 (NSn) Source D.Pu Sum.of Mean Squares P Ratio P Prob. Squares Between Groups 1 443682.000 443682.0000 7.2902 .0113 Within Groups 30 1825799.875 60859.9958 Total 31 2269481.875 (3)5-min Sulfonated PP Composite of Piber-l (S5) Source D.P. Sum.of Mean Squares P Ratio P Prob. Squares Between Groups 1 63541.1703 63541.1703 3.1416 .0896 Within Groups 23 465196.5897 20225.9387 Total 24 528737.7600 (4)8-min Sulfonated PP Composite of Fiber-1 (S8) Source D.P. Sum.of Mean Squares P Ratio P Prob. Squares Between Groups 1 5913.2813 5913.2813 .1317 .7192 Within Groups 30 1346923.938 44897.4646 Total 31 1352837.219 138 (5)10-min Sulfonated PP Composite of Fiber-1 (S10) Source D.Pu Sum.of Mean Squares P Ratio P Prob. Squares Between Groups 1 222444.500 222444.5000 8.4182 .0069 Within Groups 30 792723.375 26424.1125 Total 31 1015167.875 (6)10-min Sulfonated PP Composite of Fiber-2 (SlOn) Source D.P. Sum.of Mean Squares P Ratio P Prob. Squares Between Groups 1 109746.1250 109746.1250 4.4863 .0426 Within Groups 30 733870.7500 24462.3583 Total 31 843616.8750 (7)20-min Sulfonated PP Composite of Fiber-2 (S20n) Source D.P. Sum.of Mean Squares P Ratio P Prob. Squares Between Groups 1 51040.125 51040.1250 1.3748 .2502 Within Groups 30 1113793.875 37126.4625 Total 31 1164834.000 139 Table 40: One-way Analysis of Variance of Izod Impact Strength Data, for Lengthwise Direction Analysis of Variance Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 6 2048.9820 341.4970 60.6389 .0000 Within Groups 105 591.3228 5.6316 Total 111 2640.3048 Group Count Mean Standard Standard 95% Confidence Deviation Error Interval for Mean NS 16 20.6519 2.0290 .5073 19.5707 To 21.7331 NSn 16 33.2563 2.5213 .6303 31.9127 To 34.5998 85 16 20.6750 1.3088 .3272 19.9776 To 21.3724 88 16 21.5962 2.1816 .5454 20.4338 To 22.7587 810 16 26.6694 3.5831 .8958 24.7601 To 28.5787 810n 16 24.5344 2.5210 .6302 23.1910 To 25.8777 S20n 16 21.3531 1.8115 .4529 20.3878 To 22.3184 Total 112 24.1052 4.8771 .4608 23.1920 TO 25.0184 Multiple Range Tests: Tukey-BSD test with significance level .050 The difference between two means is significant if MEAN(J)-MEAN(I) >s 1.6780 * RANGE * SQRT(1/N(I) + 1/N(J)) with the following value(s) for RANGE: 4.25 NS 85 S20n 88 810 NSn NS 85 S20n 88 810n 4 '8 4 4 810 4 4 4 4 "an t 4 4 4 4 (*) Indicates significant differences 140 Table 41: One-Way Analysis of Variance of Izod Impact Strength Data, for Crosswise Direction Analysis of Variance Source D.P. Sum of Mean P Ratio P'Prob. Squares Squares Between Groups 6 2611.4955 435.2492 132.2490 .0000 Within Groups 105 345.5691 3.2911 Total 111 2957.0646 Group Count Mean Standard Standard 95% Confidence Deviation. Error Interval for Mean NS 16 22.5356 .8642 .2161 22.0751 To 22.9961 NSn 16 34.3563 2.6780 .6695 32.9292 To 35.7833 85 16 20.4531 1.6114 .4029 19.5945 TO 21.3118 88 16 19.8050 .9286 .2322 19.3102 TO 20.2998 810 16 24.1444 2.7255 .6814 22.6920 TO 25.5967 810n 16 25.3963 1.3078 .3270 24.6994 To 26.0931 820n 16 19.2819 1.5879 .3970 18.4357 To 20.1280 Total 112 23.7104 5.1614 .4877 22.7439 To 24.6768 Multiple Range Tests: Tukey-MSD test with significance level .050 The difference between two means is significant if MEAN(J)-MEAN(I) >- 1.2828 * RANGE * SQRT(1/N(I) + 1/N(J)) with the following value(s) for RANGE: 4.25 S20n 88 85 NS 810 810n NSn S20n 88 85 N8 4 4 4 310 4 4 4 810n * * * "8n 4 4 4 4 4 4 (*) Indicates significant differences 141 Table 42: One-way Analysis of Variance of Izod Impact Strength Data in Lengthwise Direction Vs. Crosswise Direction (1)Nonsulfonated PP Composite of Piber-l (NS) Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 1 28.3881 28.3881 11.6733 .0018 Within Groups 30 72.9562 2.4319 Total 31 101.3444 (2)Nonsulfonated PP Composite of Fiber-2 (NSn) Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 1 9.6800 9.6800 1.4310 .2410 Within Groups 30 202.9304 6.7643 Total 31 212.6103 (3)5-min Sulfonated PP Composite of Fiber-1 (SS) Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 1 .3938 .3938 .1828 .6721 Within Groups 30 64.6447 2.1548 Total 31 65.0386 (4)8-min Sulfonated PP Composite of Piber-l (88) Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 1 25.6686 25.6686 9.1319 .0051 Within Groups 30 84.3258 2.8109 Total 31 109.9944 142 (5)10-min Sulfonated PP Composite of Piber-l (810) Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 1 51.0050 51.0050 5.0334 .0324 Within Groups 30 304.0017 10.1334 Total 31 355.0067 (6)10-mdn Sulfonated PP Composite of Fiber-2 (810n) Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 1 5.9426 5.9426 1.4735 .2343 Within Groups 30 120.9870 4.0329 Total 31 126.9296 (7)20-min Sulfonated PP Composite of Fiber-2 (820n) Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 1 34.3206 34.3206 11.8284 .0017 Within Groups 30 87.0462 2.9015 Total 31 121.3668 143 Table 43: One-Nay Analysis of Variance of Tensile Strength Data Compared between Conditioned Samples, at a Period of Storage Time (1) 3-Week Storage Analysis of Variance Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 3 17 .1378 5.7126 2 .5110 .0839 Within Groups 23 52.3252 2.2750 Total 26 69.4630 Group Count Mean Standard. Standard 95% Confidence Deviation Error Interval for Mean NS 7 16.7643 .9013 .3407 15.9307 To 17.5979 85 7 18.5429 1.8146 .6859 16.8646 To 20.2211 88 7 17.9871 1.4640 .5534 16.6331 To 19.3412 810 6 18.8550 1.7224 .7032 17.0475 To 20.6625 Total 27 18.0070 1.6345 .3146 17.3604 To 18.6536 Multiple Range Tests: Tukey-BSD test with significance level .050 The difference between two means is significant if MEAN(J)-MEAN(I) >- 1.0665 * RANGE * SQRT(l/N(I) + 1/N(J)) with the following value(s) for RANGE: 3.91 - No two groups are significantly different at the .050 level (2) 5-Week Storage Analysis of variance 144 Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 3 18.9657 6.3219 2.2755 .1110 Within Groups 20 55.5645 2.7782 Total 23 74.5302 Group Count Mean Standard Standard 95% Confidence Deviation Error Interval for Mean NS 7 16.5000 1.9567 .7396 14.6904 To 18.3096 85 6 18.5533 .9585 .3913 17.5475 TO 19.5592 88 7 18.5014 1.7208 .6504 16.9100 T0 20.0929 810 4 18.1125 1.8469 .9234 15.1737 To 21.0513 Total 24 17.8658 1.8001 .3674 17.1057 TO 18.6260 Multiple Range Tests: Tukey-ESD test with significance level .050 The difference between two means is significant if NIAN(J)-NEAN(I) >s 1.1786 * RANGE * BQRT(1/N(I) + 1/N(J)) with the following value(s) for RANGE: 3.95 - No two groups are significantly different at the .050 level (3) 7-Week Storage Analysis of Variance 145 Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 3 31.0640 10.3547 7.1320 .0016 Within Groups 22 31.9407 1.4519 Total 25 63.0047 Group Count Mean Standard Standard 95% Confidence Deviation Error Interval for Mean N8 7 16.1557 .8916 .3370 15.3311 T0 16.9803 85 7 19.0414 1.0062 .3803 18.1108 To 19.9720 88 5 17.9860 1.3679 .6117 16.2876 To 19.6844 810 7 18.2171 1.5062 .5693 16.8242 To 19.6101 Total 26 17.8396 1.5875 .3113 17.1984 TO 18.4808 Multiple Range Tests: Tukey-BSD test with significance level .050 The difference between two means is significant if MEAN(J)-MEAN(I) >- .8520 * RANGE * SQRT(1/N(I) + 1/N(J)) with the following value(s) for RANGE: 3.92 N8 88 810 85 N8 88 810 * 85 * (*) Indicates significant differences (4) 9-Week Storage Analysis of Variance 146 Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 3 19.8111 6.6037 4.2577 .0177 Within Groups 20 31.0202 1.5510 Total 23 50.8313 Group Count Mean Standard Standard 95% Confidence Deviation Error Interval for Mean N8 7 16.3729 1.8155 .6862 14.6938 To 18.0519 85 5 18.9700 .3623 .1620 18.5202 To 19.4198 88 7 17.6143 1.0245 .3872 16.6668 TO 18.5618 810 5 17.5940 1.0514 .4702 16.2885 TO 18.8995 Total 24 17.5304 1.4866 .3035 16.9027 TO 18.1582 Multiple Range Tests: Tukey-MSD test with significance level .050 The difference between two means is significant if NEANW’) -NEAN(I) with the following value(s) for RANGE: 3.95 >- .8806 * RANGE * SQRT(l/N(I) + 1/N(J)) 810 88 85 N8 810 88 85 (*) Indicates significant differences 147 Table 44: One-way Analysis of Variance of Tensile Strength Data Compared between Different Periods of Storage Time, (1) NOnsulfonated PP Composite Analysis of Variance for a composite material Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 3 1.3567 .4522 .2072 .8904 Within Groups 24 52.3925 2.1830 Total 27 53.7492 Group Count Mean Standard Standard 95% Confidence Deviation Error Interval for Mean 3-wk 7 16.7643 .9013 .3407 15.9307 To 17.5979 5-wk 7 16.5000 1.9567 .7396 14.6904 To 18.3096 7-wk 7 16.1557 .8916 .3370 15.3311 To 16.9803 9-wk 7 16.3729 1.8155 .6862 14.6938 To 18.0519 Total 28 16.4482 1.4109 .2666 15.9011 To 16.9953 Multiple Range Tests: Tukey-BSD test with significance level .050 The difference between two means is significant if NEAN(J)-NEAN(I) >- 1.0448 * RANGE * SQRT(1/N(I) + 1/N(J)) with the following value(s) for RANGE: 3.90 - No two groups are significantly different at the .050 level (2) 5-min Sulfonated PP Composite Analysis of Variance 148 Source D.P. Sum of Mean P Ratio P Prob. Squares Squares Between Groups 3 1.3585 .4528 .3073 .8198 Within Groups 21 30.9496 1.4738 Total 24 32.3081 Group Count Mean Standard Standard. 95% Confidence Deviation Error Interval for Mean 3-wk 7 18.5429 1.8146 .6859 16.8646 To 20.2211 S-wk 6 18.5533 .9585 .3913 17.5475 TO 19.5592 7-wk 7 19.0414 1.0062 .3803 18.1108 To 19.9720 9-wk 5 18.9700 .3623 .1620 18.5202 To 19.4198 Total 25 18.7704 1.1602 .2320 18.2915 To 19.2493 Multiple Range Tests: Tukey-ESD test with significance level .050 The difference between two means is significant if IEAN(J)-NEAN(I) >- .8584 * RANGE * SQRT(1/N(I) + 1/N(J)) with the following value(s) for RANGE: 3.94 - No two groups are significantly different at the .050 level 149 (3) 8-min Sulfonated PP Composite Analysis of Variance Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 3 2.7873 .9291 .4603 .7128 Within Groups 22 44.4085 2.0186 Total 25 47.1958 Group Count Mean Standard Standard. 95% Confidence Deviation Error Interval for Mean 3-wk 7 17.9871 1.4640 .5534 16.6331 TO 19.3412 5-wk 7 18.5014 1.7208 .6504 16.9100 To 20.0929 7-wk 5 17.9860 1.3679 .6117 16.2876 To 19.6844 9-wk 7 17.6143 1.0245 .3872 16.6668 To 18.5618 Total 26 18.0250 1.3740 .2695 17.4700 To 18.5800 Multiple Range Tests: Tukey-BSD test with significance level .050 The difference between two means is significant if NEAN(J)-NEAN(I) >s 1.0046 * RANGE * SQRT(1/N(I) + 1/N(J)) with the following value(s) for RANGE: 3.92 - No two groups are significantly different at the .050 level 150 (4) 10-min Sulfonated PP Composite Analysis of Variance Source D.P. Sum.of Mean P Ratio P Prob. Squares Squares Between Groups 3 4.4226 1.4742 .6157 .6137 Within Groups 18 43.1001 2.3944 Total 21 47.5227 Group Count Mean Standard Standard 95% Confidence Deviation Error Interval for Mean 3-wk 6 18.8550 1.7224 .7032 17.0475 To 20.6625 S-wk 4 18.1125 1.8469 .9234 15.1737 To 21.0513 7-wk 7 18.2171 1.5062 .5693 16.8242 To 19.6101 9-wk 5 17.5940 1.0514 .4702 16.2885 To 18.8995 Total 22 18.2305 1.5043 .3207 17.5635 To 18.8974 Multiple Range Tests: Tukey-BSD test with significance level .050 The difference between two means is significant if MEAN(J)-MEAN(I) >- 1.0942 * RANGE * SQRT(1/N(I) + l/N(J)) with the following value(s) for RANGE: 3.99 - No two groups are significantly different at the .050 level BIBLIOGRAPHY BIBLIOGRAPHY Agarwal. B.D. and Broutman, L.J.. MAW of_zibor_compooitoo, John Wiley and Sons, Inc., 1980. Asthana, 8., 'Chemical Modification of Polymer Surfaces Using Sulfonation to Improve Adhesion Properties", M.S. Thesis, Michigan State university, East Lansing, MI, 1993. ASTM Standard D256-92, I'Standard Test Methods for Impact Resistance of Plastics and Electrical Insulating Materials", AnnuaLBookaJSTMJLtandards. Philadelphia. PA. pp- 58-74. 1993. ASTM Standard D638-91, "Standard Test Methods for Tensile Properties of Plastics“. We. Philadelphia, PA, pp. 161-173, 1993. ASTM Standard D790-92, I'Standard Test Methods for Plexural Properties of unreinforced Plastics and Electrical Insulating Materials". WWW. Philadelphia, PA. PP. 274-283, 1993. Bigg, D.M., “Principles of Polymer Mixing Technology”, in A. Whelan and J.L. Craft. ed., DexelopmentLInJlastios Toohnology;2, Elsevier Applied Science Publishers Ltd., pp. 175-217, 1985. Bigg, D.M., Biscock, D.P., Preston, J.R., and Bradbury, E.J., “Thermoplastic Matrix Sheet Composites", Polymer Compooitoo, Vol. 9, no. 3, pp. 222-228, June 1988. Brandrup, J. and Immergut, E.H., Polymor_flenobook, 2nd ed., John Wiley and Sons, Inc., 1975. Browning, B.D., Tho_ghomiotzy_of_flooo, Interscience Publishers, 1963. 151 152 Childress, J., "Wood Piber/High Density Polyethylene Composites: Ability of Additives to Enhance Mechanical Properties", M.S. Thesis, Michigan State university, East Lansing, MI, 1991. Chtourou, B., Reidl, B., and Ait-Radi, A., "Reinforcement of Recycled Polyolefins with Wbod Pibers", Journal_ofi WWW. Vol. 11. pp. 372-394. April 1992. Dinwoodie, J.M., - i _ Compooito, The Institute of Metals, 1989. Erickson, B.D., "On the Enhancement of Adhesive Bonding to Polymer and Composite Surfaces Through Gas Phase Sulfonation", M.S. Thesis, Michigan State university, East Lansing, MI, 1993. Ponseda, C., Perena, J.M., Patou, J.G., and Bello, A., "Sulphuric Acid Etching of Polyethylene Surfaces", Journal omnrialLSoience, Vol. 20. pp- 3283-3288. 1985. Gatenholm, P., Pelix, J., Klason, C., and Kubat, J., "Cellulose-Polymer Composites with Improved Properties", WWW. Vol. 7. Plenum Press, NY, 1992. Gilbert, E.E. , 811W. Interscience Publishers, 1965. Grayson, M., W Compononts, Encyclopedia Reprint Series, John Wiley and Sons, Inc., 1983. Haraguchi, P., "The Effect of Surface Sulfonation of High Density Polyethylene (HDPE) on the Mechanical Properties of BDPE/Wood Piber Composites", M.S. Thesis, Michigan State university, East Lansing, MI, 1993. Bull. D.. W. Cambridge university Press, 1981. 153 Ihata, J., "Pormation and Reaction of Polyenesulfonic Acid. I. Reaction of Polyethykene Pilms‘with SO3", Journal_of Polymer_8oionoo: Part A: Polymer Chemistry, Vol. 26, pp. 167-176. 1988. Elason, C, Eubat, J., and Stromvall, B.E., "The Efficiency of Cellulosic Pillers in Common Thermoplastics. Part 1. Pilling Without Processing Aids or Coupling Agents", Intorno 1W, Vol. 10. pp. 159-187. 1984. Kokta, B.V., Daneault, C., and Beshay, A.D., "Use of Grafted Aspen Pibers in Thermoplastic Composites: IV. Effect of Extrume Conditions on Mechanical Properties of Polyethylene Composites", Polymor_§ompooitoo, Vol. 7, No. 5, pp. 337-348, October 1986. Rrishnan, M. and Narayan, R., "Compatibilization of Biomass Pibers‘With Hydrophobic Materials", in R.M. Rowell, T.D. Leufenberg. end ELK. Rewell. ed.. W W. Materials Research Society Symposium Proceedings, Vol. 266, pp. 93- 104, 1992. Lee, 8.14.. W. Vol. 6. VCB Publishers, Inc., 1991. Modern Plastics Magazine, Elastioo_nandbook, McGraw-Hill, Inc. Mullins. LJ- and McKnight. T.S.. W Proportioo_and_nsos, 3rd ed., university of Toronto Press, 1981. Olsen, D.A. and Osteraas, A.J., "Surface Medification of Polyethylene Surfaces. III. Prustrates Multiple Internal Reflection Spectroscopy of Sulfonated Polyethylene Surfaces". WM: Part A-l. vol. '7. pp. 1927-1932. 1969. Park, I., "The Effect of Surface Sulfonation of Polymer Pilms on Dispersion and Non-Dispersion Components of Surface Energy and Peel Adhesion Strength" M.S. Thesis, Michigan State university, East Lansing, MI, 1993. 154 Raj, R.G., Kokta, B.V., and Daneault, C., "Polypropylene- Wbod Piber Composites: Effect of Piber Treatment on Mechanical Properties". W. Vol. 12. pp. 239-250. 1989. Raj, R.G. and Kokta, B.V., "Linear Low-Density Polyethylene Pilled with Silane-Coated Wbod Pibers", in T.D. Vigo and A.P. Turbak, ed., BighzanihrouLMaterialsstosimi WWW. American Chemical Society, pp. 102-113, 1991. Richardson. R.G.-Wu WW. Applied Science Publishers Ltd., London, 1977. Richardson, T., Compooitooi_A_Dooign_Guido, Industrial Press Inc., NY, 1987. Sain, M.M., Kokta, B.V., and Imbert, C., "Structure-Property Relationships of wood Piber-Filled Polypropylene Composite", WW9. Vol. 33. No. 1. pp. 89-104, 1994. Sanadi, A.R., Young, R.A., Clemons, C, and Rowell, R.M., "Recycled Newspaper Pibers as Reinforcing Pillers in Thermoplastics: Part I- Analysis of Tensile and Impact Properties in Po1ypropy1ene". JournalJLReinforssdJlastics ond_§ompooitoo, Vol. 13. pp. 54-67, January 1994. Schwartz. MAL. WWII: 2nd ed.. McGraw-Hill, Inc., 1992. Selke, S., Yam, R., and Nieman, R., "Effects of Additives on Mechanical Properties of wood Piber/High Density Polyethylene Composites". WW. Society of Plastics Engineers, Brook field, CT, pp.1813- 1815, 1989. Seymour, R.B. and Carraher, C.E., Strnotnro;2zoporty RelationshinLiLPnlmrs. Plenum Prose. NY. 1984. Simpson, R., "Composites Materials from Recycled Multi-layer Polypropylene Bottles and wood Pibers", M.S. Thesis, Michigan State university, East Lansing, MI, 1991. 155 Stamn. A.J.. flood_and_Ce11ulose_Soienoe. The Ronald Press Company, NY, 1964. walles,‘W.E., "Resinous Enclosure Members Rendered Impermeable by Sulfonation and Neutralization", USP 3,613,957, 1971. walles, W3E., "Resinous Enclosure Members Rendered Impermeable by sulfonation", USP 3,740,258, 1973. Walles,‘W.E., "Barrier Properties Added to Plastics via Sulfonation and Reductive Metallization", ACS Dallas, April 1989. wangwiwatsilp, R., "The Effect of Surface Sulfonation on Barrier Properties of Polymer Pilme", M.S. Thesis, Michigan State University, East Lansing, MI, 1993. Zadorecki, P., Karnerfors, B., and Lindenfors, S., "Cellulose Pibers as Reinforcement in Composites: Determination of the Stiffness of Cellulose Pibers", Composites_Science_and_Technolog¥. Vol. 27. No. 4. pP- 291- 303. 1986.