Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Effect of nanoclay fillers on wood adhesives and particle board properties Xian, Diyan 2012

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2012_fall_xian_diyan.pdf [ 3.47MB ]
Metadata
JSON: 24-1.0072768.json
JSON-LD: 24-1.0072768-ld.json
RDF/XML (Pretty): 24-1.0072768-rdf.xml
RDF/JSON: 24-1.0072768-rdf.json
Turtle: 24-1.0072768-turtle.txt
N-Triples: 24-1.0072768-rdf-ntriples.txt
Original Record: 24-1.0072768-source.json
Full Text
24-1.0072768-fulltext.txt
Citation
24-1.0072768.ris

Full Text

EFFECT OF NANOCLAY FILLERS ON WOOD ADHESIVES AND PARTICLE BOARD PROPERTIES  by  Diyan Xian B.Eng., South China Agriculture University, 2009  A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF APPLIED SCIENCE  in  The Faculty of Graduate Studies  (Forestry)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  April 2012  © Diyan Xian, 2012  Abstract The objective of this research is to investigate the effect of nanoclay additions to particleboard resins on the properties of particleboard made with those resins. Two nanoclays, Cloisite30B, a modified nanoclay and Nanofil16, an unmodified clay, were blended with the two resins used to produce particleboard: Urea Formaldehyde (UF) and Melamine Formaldehyde (MF). Coupling agent was added to nanoclays to facilitate clay dispersion into the resin. X-ray diffraction tests showed that mechanical mixing was sufficient to exfoliate Closite30B into both resin types and enable the intercalation of Nanofil116/resin mixtures.  Addition of nanoclays and coupling agents had small to severe adverse effects on resin curing: Cloisite30B slightly delayed the curing process of both UF and MF resin and reduced the reaction heat of curing, and the addition of coupling agent together with Closite30B further compounded this effect. Nanofil116 significantly delayed the curing reaction of both resins and decreased the heat of reaction. The coupling agent had a significant further detrimental effect on the resin cure.  In order to test whether nanoclays had a positive or negative effect on the adhesive strength of UF and MF resins, the shear strength of clay-modified resin were tested and compared with that of unadulterated resin. Regardless of whether coupling agent was used, the clay-modified UF resin had lower bonding strength than pure UF resin. In contrast, three kind of clay-modified MF resin had higher bonding strength then pure MF resin.  Based on these findings those MF resins which have higher shear strength were blended with furnish to fabricate particle board using different clay loading rates. Most clay treatments had no significant effect on particleboard physical or mechanical properties. The only significant ii  improvement was for internal bond strength which increased when using either 2% Closite30B or Nanofil116 with or without coupling agent. Higher clay loading rates tended to decrease board strength properties. In conclusion, the modified Closite30B nanoclay and the unmodified Nanofil116 nanoclay had only a minor effect on improving UF and MF resin strength and the particle board properties.  iii  Tables of contents Abstract ........................................................................................................................................... ii Tables of contents .......................................................................................................................... iv List of tables ................................................................................................................................... vi List of figures ................................................................................................................................ vii Acknowledgements ........................................................................................................................ ix 1 Introduction .................................................................................................................................. 1 1.1 Background ........................................................................................................................... 1 1.2 Rationale ............................................................................................................................... 3 1.3 Hypothesis............................................................................................................................. 3 1.4 Approach ............................................................................................................................... 4 1.5 Structure of thesis ................................................................................................................. 5 2 Literature review .......................................................................................................................... 6 2.1 Overview ............................................................................................................................... 6 2.2 Nanoclay fillers and resins .................................................................................................... 7 2.2.1 Structure of nanoclay ..................................................................................................... 7 2.2.2 Thermoplastic resin and thermosetting resin ................................................................. 9 2.2.3 Hybrid morphology of nanoclay-resin composites ...................................................... 11 2.3 Nanoclay dispersion methods ............................................................................................. 13 2.4 Wood nanoclay composites processing .............................................................................. 15 2.5 Characterization nanoclay-resin hybrid structure ............................................................... 17 2.5.1 X-ray diffraction .......................................................................................................... 18 2.5.2 Microscopy techniques ................................................................................................ 19 2.5.3 Thermo analysis ........................................................................................................... 22 2.6 Nanoclay reinforcement mechanism................................................................................... 28 2.7 Characterization of properties of polymer-nanoclay-wood composites ............................. 29 2.7.1 Dimensional stability of clay-reinforced nanocomposites (thickness swell and water absorption) ............................................................................................................................ 29 2.7.2 Mechanical properties of polymer-nanoclay and wood -nanoclay composite............. 30 iv  2.8 Summary ............................................................................................................................. 32 3 Materials and methods ............................................................................................................... 35 3.1 Materials ............................................................................................................................. 35 3.2 Preparation of resin and nanoclay mixtures ........................................................................ 36 3.3 Evaluation of bulk resin properties ..................................................................................... 37 3.3.1 X-ray diffraction test .................................................................................................... 38 3.3.2 DSC test of resin and clay mixture .............................................................................. 38 3.3.3 Dynamic mechanical analysis (DMA) test .................................................................. 38 3.3.4 Lap-Shear test - automatic bonding evaluation system (ABES) ................................. 39 3.4 Manufacture of particle boards and evaluation of the effect of clay loading on board properties................................................................................................................................... 40 3.4.1 Three layer particle board manufacturing process ....................................................... 41 3.4.2 Test of mechanical properties of large boards ............................................................. 44 3.4.3 Board properties testing ............................................................................................... 44 4 Results and discussion ............................................................................................................... 50 4.1 XRD analysis ...................................................................................................................... 50 4.2 DSC analysis of the curing process .................................................................................... 57 4.3 DMA Test ........................................................................................................................... 64 4.4 Lap-shear test - automatic bonding evaluation system (ABES) approach ......................... 68 4.5 Particle board properties test results ................................................................................... 70 4.5.1 MOR/MOE test ............................................................................................................ 70 4.5.2 Edge screw withdrawal (SWR) test ............................................................................. 72 4.5.3 Internal bonding (IB) test ............................................................................................. 73 4.5.4 Thickness swelling test ................................................................................................ 74 4.6 Conclusion and general discussion ..................................................................................... 76 5 Comments and future work........................................................................................................ 78 References ..................................................................................................................................... 80 Appendix A: Statistical analysis result of DSC test...................................................................... 87 Appendix B: Statistical analysis result of Lap-shear test............................................................ 103 Appendix C: Statistical analysis result of PB properties test ...................................................... 106 v  List of tables Table 3.1: Nominal properties of the UF and MF resins. ............................................................. 35 Table 3.2: Nanoclays supplied by Southern Clay Ltd. ................................................................. 36 Table 3.3: Experimental parameters and response variables for the large board production. ...... 41 Table 3.4: Hot press cycles parameters......................................................................................... 44 Table 4.1: 2 θ and d-space values from XRD patterns (Some of the samples had no intensity peak and this is denoted by the ‘np’ entry in the table). ........................................................................ 56 Table 4.2: Mean values of 3 measurements for T onset , ΔH, T peak ................................................. 63  vi  List of figures Figure 2.1: Structure of layer montmorillonite (Adapted from Pavlidou and Papaspyrides, 2008 used with permission from Elsevier) .............................................................................................. 8 Figure 2.2: Schematic representation of (A) urea-formaldehyde (UF) and (B) melamineformaldehyde (MF) resin systems (Adapted fromYoung No and Kim, 2005 with permisstion). 10 Figure 2.3: Scheme of different types of composite arising from the interaction of layered silicates and polymers: (a) phase separated microcomposite; (b) intercalated nanocomposite, and (c) exfoliated nanocomposite ( Adapted from Alexandre and Dubois, 2000 with permission). .. 12 Figure 2.4: Flow chart of wood nanoclay polymer composite process (Adapted from Lü and Zhao, 2004 with permission) ........................................................................................................ 16 Figure 2.5: Crack initiation process of clay-filled epoxy resin ( Adapted fromWang et al., 2005 with permission)............................................................................................................................ 21 Figure 2.6: Schematic of DSC curve (Hon, 2000). ....................................................................... 23 Figure 2.7: Tan δ, storage (E') and loss (E'') modulus for a sample of PVA (Hatakeyama and Quinn, 1999). ................................................................................................................................ 25 Figure 2.8: DMA result of a resole resin. (Pilato, et al., 2010) .................................................... 27 Figure 3.1: The Labmaster TS-2010 mechanical mixer: (a) the mixing head and (b) the impellor agitator. ......................................................................................................................................... 37 Figure 3.2: (a) air-operated clipper, (b) applying resin, (c) automatic bond evaluation system. . 40 Figure 3.3: (left) Drais particleboard batch-blender and (right) Pathex press. ............................. 42 Figure 3.4: Cutting pattern of 25" by 25" particle board .............................................................. 45 Figure 3.5: Apparatus of center-point loading flexural test. ......................................................... 46 Figure 3.6: The assembly for edge screw withdrawal test ............................................................ 47 Figure 3.7: The assembly of internal bonding (IB) test ................................................................ 48 Figure 3.8: Thickness swell samples in the tank........................................................................... 49 Figure 4.1: Typical XRD patterns of Cloisite30B, Cloisite30B + UF resin. ................................ 51 Figure 4.2: Typical XRD patterns of Cloisite30B, Cloisite30B and coupling agent + UF resin.. 51 Figure 4.3: Typical XRD patterns of Nanofil116 and Nanofil116 + UF resin. ............................ 52 Figure 4.4: Typical XRD patterns of Nanofil116, Nanofil116 and coupling agent + UF resin. .. 52 Figure 4.5: Typical XRD patterns of Cloisite30B, Cloisite30B + MF resin. ............................... 53 Figure 4.6: Typical XRD patterns of Cloisite30B, Cloisite30B and coupling agent +MF resin. . 53 Figure 4.7: Typical XRD patterns of Nanofil116, Nanofil116 + MF resin. ................................. 54 vii  Figure 4.8: Typical XRD patterns of Nanofil116, Nanofil116 and coupling agent + MF resin. .. 54 Figure 4.9: Typical heat flow curves of different UF resins with different loading of Closiste30B. ....................................................................................................................................................... 58 Figure 4.10: Typical heat flow curves of different UF resins with different loading of Closiste30B and coupling agent.................................................................................................... 58 Figure 4.11: Typical heat flow curves of different UF resins with different loading of Nanofil116. ................................................................................................................................... 59 Figure 4.12: Typical heat flow curves of different UF resins with different loading of Nanofil116 and coupling agent. ....................................................................................................................... 59 Figure 4.13: Typical heat flow curves of different MF resins with different loading of Closiste30B. .................................................................................................................................. 61 Figure 4.14: Typical heat flow curves of different MF resins with different loading of Closiste30B and coupling agent.................................................................................................... 61 Figure 4.15: Typical heat flow curves of different MF resins with different loading of Nanofil116. ................................................................................................................................... 62 Figure 4.16: Typical heat flow curves of different MF resins with different loading of Nanofil116 and coupling agent. ....................................................................................................................... 62 Figure 4.17: Typical storage modulus and loss modulus of UF resin and clay added UF resins. 65 Figure 4.18: Typical storage modulus and loss modulus of MF resin and clay added MF resins.66 Figure 4.19: Shear strength of UF and clay added UF resins. ...................................................... 69 Figure 4.20: Shear strength of MF and clay added MF resins. ..................................................... 69 Figure 4.21: Average modulus of rupture for particleboards bonded with different MF resin + clay mixes. .................................................................................................................................... 71 Figure 4.22: Average modulus of elasticity for particleboards bonded with different MF resin + clay mixes. .................................................................................................................................... 71 Figure 4.23: Test values of the screw withdrawal test .................................................................. 72 Figure 4.24: Internal bond strength values of different treatments ............................................... 73 Figure 4.25: Water absorption and thickness swelling rate by different treatments ..................... 74  viii  Acknowledgements I owe a lot of thanks to many people who accompanied me through this journey. First and foremost, I would like to express my sincere appreciation to my supervisor Dr. Gregory D. Smith for his remarkable support, advice, understanding, and encouragement during the research, especially when failures and frustration came. Without that I would have been lost. I would also like to acknowledge the helpful advice from my committee members, Dr. John Kadla and Dr. Taraneh Sowlati.  My heartfelt thanks to the current and former members of the wood composite group: Chao Zhang, Emanuel Sackey, Jorn Dettmer, Kate Semple, Solace Sam-Brew, Shayesteh Haghdan, Ying-Li Tasi and Xuelian Zhang for their teaching of operating laboratory equipment, their help in the preparation of sample and thesis and sharing their useful experience with me.  Appreciation is also extended to Jennifer Braun, Feng-Cheng (Aries) Chang of BioMaterials Chemistry group , George Lee of the Timber Engineering Group, Vincent Leung and Lawrence Gunther from Advanced Wood Processing Center for their help in running samples and setting up the equipment.  I owe a special thanks to all my friends in Faculty of Forestry for their advice and dedication to my study and life. Finally, I would like to express my sincerest gratitude and appreciation to my parents for their constant encouragement, patience, and support in both good and difficult times.  ix  1 Introduction 1.1 Background Particleboard (PB) is a commonly used panel product made from hammer-milled wood particles that has a relatively low cost of production compared with Medium fiber board (MDF). It has a smooth surface that can be easily laminated or painted. As a result, it is widely used in furniture applications such as desks, shelves, and cabinets (Wong, 2008).  The price of PB has declined slightly over the past 5 years(Douglas Clark, 2011). However, the production cost per cubic meter of PB has been increasing since 2008 and is expected to increase further in 2011and 2012 (RISI, 2011). Since the price of PB has not changed greatly and production costs continue to increase, this places significant pressure on PB manufacturers to remain profitable and as a result several plants have closed over the past few years (Pepke, 2010).  The objective of this research is to investigate whether using nanoclays as fillers in PB resins can improve the panel properties, and if so, can the amount of resin used to make the PB be decreased while maintaining board properties. The raw material costs for manufacturing PB include wood furnish, resin, and wax, and resin is the most costly. Thus it is necessary to find ways of reducing resin costs while maintaining adequate board properties. Usually, increasing the resin content of the board will result in improved panel properties. One possibility for reducing resin costs is to add a low-cost filler to the resin thereby reducing the total amount of resin required for board production (Shi, Qiu, and Zheng, 2004). Additives such as fillers, curing agents, and coupling agents, could reinforce the resin and improve composite properties (Giannelis, Krishnamoorti, and Manias, 1999). Powdered fillers such as finely ground wood 1  flour, nut shells, and rice hulls have been mixed with wood adhesives in an attempt to reduce over penetration of adhesive into wood or better reinforced panels (Nishizawa et al , 1982). Mineral fillers are also low cost additives and have been shown to be able to reduce resin usage while maintaining board properties (Shi et al., 2004). Nanomaterials have least one dimension in the nanometer (10-9 m) range, and have been proven to be excellent fillers for wood resins and other polymers (Pavlidou and Papaspyrides, 2008). Montmorillonite (MMT) clays are naturally nanomaterials silicate minerals when it dispersed in various polymer matrices (Alexandre and Dubois, 2000). MMT is more widely available and significantly lower in cost than other nanomaterials such as carbon nanotubes or nanoaluminum particles. The price of MMTs typically is ranging from $US2.0 to $3.50 per pound.  The surface area of MMT platelets becomes very large once the clay stacks are dispersed and can interact with the resin and improve its mechanical properties (Giannelis, et al., 1999; Hetzer and Dekee, 2008). MMT has shown some promise in improving the strength and mechanical properties of various resin matrices. Most research to date has focused on blending processes and clay dispersion in thermoplastic resins and wood plastic composites (WPC). For example, Hetzer and De Kee (2008) reported that adding 2-10% MMT to polyamide-6, polypropylene, and polyethylene high polymers improved their strength, elastic modulus, flame and heat resistance, and water resistance. Zerda et al., (2001) found that small amounts of modified MMT can improve the mechanical strength of the epoxy resin when using a MMT content of 3-12 %.  Only a small amount of work had been done on using nanoclay-filled adhesives to make wood panels. Ashori and Nourbakhsh (2009), found that the mechanical properties (MOR and MOE) of MDF increased with clay content over the 2 to 6% range. Wang, et al. (2008) used nanoclay2  filled adhesives to fabricate experimental OSB, MDF, plywood and PB but found no beneficial effects of using nanoclay-modified resins. Further studies are needed in in this area, i.e. the application of MMT nanoclays to fabricated wood adhesive and wood composites, in particular, particleboard.  1.2 Rationale  Based on the literature review, it was concluded that the addition of MMT to particleboard resin has the potential to improve its properties. The phyllosilicates structure of montmorillonite should facilitate the platelet separation process and make it easier to exfoliate the nano-sized clay layers to reinforce resins used in wood panels (Wang et al., 2008).  The potential for cost savings if resin can be partially substituted with nano-clay, even to a small extent, are substantial. At current resin prices if 0.5% MMT clay was added to a binder system permitted the resin content to be reduced by 1%, say from 8% to 7%, then for a plant with an annual production capacity 200,000 m3 the approximate savings in resin cost alone would be $660,000 CAD per plant per year.  1.3 Hypothesis  The hypothesis for this study is: the addition of small amounts of MMT nanoclay to the resin used to make PB will improve the physical and mechanical properties of PB.  The goal of this work is to determine if MMTs are compatible with UF and MF resins and whether resin-clay mixtures can improve the adhesive strength of these resins, and in doing so,  3  improve the properties of PB made from the same amount of MMT-fortified resins or maintain the PB properties using less modified resin.  1.4 Approach  The aim of this study is to reduce the amount of wood adhesive, and by extension PB production cost, to meet minimum property requirements through the addition of MMT nanoclays to the resin before blending with wood furnish. At the very least, replacing a portion of the resin with nanoclay should not reduce board properties.  The first phase of the work is a preliminary study to determine the effect of adding different types and amounts of nanoclay to several candidate resins and to characterize the ability of the clay to disperse in those resins.  In the second phase of the work, thermomechanical properties of the resin are assessed using Dynamic Mechanical Analysis and the lap shear strength testing of bonded wooden veneers is used as a screening test for determining which clay resin combinations are likely to lead to properties improvement in PB.  The final phase will examine the effect of selected clay-resin mixes (determined from the wood veneer lap shear tests) on the physical and mechanical properties of laboratory-fabricated PB.  4  1.5 Structure of thesis  The structure of this work is presented in the following:  Chapter 1. Introduction: This chapter introduces basic information on PB and discusses how board properties are affected by board resin content and possible solutions for reduce production cost while maintaining or improving panel properties. Chapter 2. Literature review: This chapter was the review of pervious work of nanoclay, resins, processing method of nanoclay-resin wood composite. The techniques for evaluate the clay dispersion, clay reinforced mechanism, and the effect of the nanoclay on wood composite are covered. Chapter 3. Materials and methods: Introduction of the raw materials including adhesives, wood furnish, coupling agent and nanoclay. The processing method for each tests were also detailed in this chapter. Chapter 4. Results and discussion: In this chapter the results from the various tests and the effect of each treatment on resin and panel properties are identified and discussed. Chapter 5. Comments and future work: The final conclusions are summarized and a few possible directions for further investigation described.  5  2 Literature review 2.1 Overview This review examines the enhancement of the physical and mechanical properties the addition of thermoplastic and thermosetting resins when they are mixed with nanoclays. The review focusses mainly on phyllosilicates nanoclays, specifically modified montmorillonite (MMT). The following topics are reviewed:  1. The structure of nanoclay fillers and resin matrices; 2. Methods for incorporating nanoclay into thermoplastic and thermosetting resins for properties enhancement; 3. Characterization of the resin- nanoclay hybrid structure; 4. Nanoclay reinforcement mechanism; 5. Instances of nanoclays used in wood composite and their properties. The structure of nanoclay and the morphology of clays dispersed in the resin matrix are introduced first since the properties of nanoclay-reinforced material depends on the clay platelet size, aspect ratio, processing method and other factors such as clay distribution quality. Analytical methods used to characterize clay morphology and clay dispersion in polymer matrices include X-ray Diffraction (XRD), Transmission Electron Microscopy (TEM) and Scanning Electron Microscopy (SEM). These can determine whether a nanoclay hybrid structure has been obtained. If so then further thermo analysis of the mixtures can be made to determine how the clay interacts with the polymer to affect the chain structure, crystallize rate, curing process and thermo mechanical properties(Pavlidou and Papaspyrides, 2008). Clay effects on  6  polymer structure and curing can be quantified using Differential Scanning Calorimetry (DSC) and Dynamic Mechanical Analysis (DMA).  The effects of a nanoclay on the physical and mechanical properties of nanoclay-resin composites and nanoclay-resin-wood composites are also covered. Most of the applications of nanoclay have been in the field of wood plastic composites and so these are also included in this review.  2.2 Nanoclay fillers and resins 2.2.1 Structure of nanoclay Any particle that has at least one dimension in the nanometer size range (1-100 nm) is deemed to be a nanoparticle. Such particles can be divided into three different categories according to the number of nano-scale dimensions. Iso-dimensional nanoparticles, such as spherical silica, have three dimensions in nanometers. Carbon nanotubes or cellulose whiskers are two dimensional nanoparticles, whereby the cross section is in nano-scale and the third dimension (length) can extend to micrometer or even millimeter size ranges. The third kind of nanoparticle has only one dimension in the nano-scale. These usually take the form of thin sheets that are several nanometers thick. Some synthetic and natural crystals, and layered silicate clays can be classified into the third kind of nano-particle (Sinha Ray and Okamoto, 2003; Lebaron, et al., 1999)  This review is mainly focused on phyllosilicates clays, of which the clay layer thickness is in the nano range. The structure of montmorillonite and hectorite and spaonite place them among the common types of phyllosilicates. As shown in Figure 2.1, the structure consists of 1 nm thick, two-dimensional negatively charged silicate layers with exchangeable cations between the layers  7  (Alexandre and Dubois, 2000). Phyllosilicate clays contain cations such as Li, Na, Rb, Cs, which make them strongly hydrophilic. These exchangeable cations facilitate the modification of phyllosilicate because they can be easily replaced by other cations. In order to make the phyllosilicates more compatible with polymers, the hydrated cations are exchanged with organic cations or cationic surfactants such as alkylammonium (Giannelis, et al., 1999). Organicmodified phyllosilicates have a lower surface energy and allow the silicate layers to interact with the polymers. Usually, the organic modifiers with longer molecule chains are more effective at expanding the interlayer space and separating layers into single sheets (Lebaron et al., 1999).  Figure 2.1: Structure of layer montmorillonite (Adapted from Pavlidou and Papaspyrides, 2008 used with permission from Elsevier)  8  2.2.2 Thermoplastic resin and thermosetting resin Nanoclay as a filler or additive is usually added in to a polymer (referred to as the matrix) in order to enhance the properties of the polymer. A polymer is a very large molecule which is comprised of repeating units and those units connected with each other to form long chains which can be linear, branching or cross-linked (Edwards, 2004). There are two main types of polymers used in wood composite industry: thermoplastics and thermosetting resins.  Thermoplastics are usually linear polymers which may change in structure as temperature changes, such as glass transition, crystallization and melting (Kulshreshtha, and Vasile, 2002). Polylactic acid (PLA), Polyvinyl chloride (PVC) are the most commonly used thermoplastic resins for fabricating wood plastic composites(Pavlidou and Papaspyrides, 2008).  Thermosetting resins are three-dimensional cross-linked networks which are hard, infusible and insoluble after curing. Thermosetting resin is more difficult to characterize than a thermoplastic resin because it remains stable after curing (Hon, 2003). Polypropylene (PP), Polyethylene (PE), Urea formaldehyde (UF), melamine formaldehyde (MF), phenol formaldehyde (PF) resins are the predominant thermosetting resins used as wood adhesives in the production of hot pressed wood composites.  Thermosetting resins are usually a mixture of low molecular weight condensates, intermediates generated by primary addition reactions, and monomers which are all soluble in water. These low molecular weight condensates will further react at higher temperatures and form the final cross-linked, rigid network (Pizzi & Mittal, 1994). The condensation process of UF resin and MF resin synthesis process are shown in Figure 2. 3.  9  (A)  (B)  Figure 2.2: Schematic representation of (A) urea-formaldehyde (UF) and (B) melamineformaldehyde (MF) resin systems (Adapted fromYoung No and Kim, 2005 with permisstion).  The condensation process of MF resin is similar to that of UF, however its network is denser and more cross-linked that the UF network. The stiffness and hardness of MF resin is highest among the know polymers (Doyle, Hagstrand, and Manson, 2003). MF resin is more expensive compared to UF resin and it is often used in conjunction with cheaper UF resin and other  10  modifiers to reduce cost, and reduce the rigidity of the MF resin network. Approaches include incorporating urea and other substituted modifiers to reduce the crosslinking density.  2.2.3 Hybrid morphology of nanoclay-resin composites There are three main types of nanoclay dispersions in a resin matrix that affect its properties and by extension, the bulk properties of clay modified wood composites (Alexandre and Dubois, 2000):  1. Phase separated: The nanoclay particles mix with the resin uniformly throughout the resin matrix as shown in Figure 2.4a. The clay platelets do not separate and the structure is still classified as a micro-composite or aggregate.  2. Interlayer structure: This involves more intimate mixing whereby the polymer is able to penetrate between the clay layers but not fully separate them (Figure 2.4b). The thickness of the resin between the clay layers ranges from a few nm to a few microns. This state is known as intercalation, an important feature being that the thickness of the polymer layer between clay platelets is relatively uniform and the plates remain parallel to each other, and the polymer chains are able to enter the interlayers of the clay and interact with the clay sheets. Nano size silicate clay layers remain distributed almost in parallel direction and the gaps between clay layers are increase from a few nm to μm (Figure 2.3 b).  3. Exfoliated structure (sometimes referred to as a delaminated structure). In this state the clay platelets fully separate and disperse evenly at random angles throughout the polymer matrix (Figure 2.4 c). The distances between the clay platelets become so large that they are no longer aligned together. 11  When the nano-thickness clay platelets fully disperse into the polymer matrix a true nanocomposite is obtained (Giannelis, et al., 1999b,Hetzer and Dekee, 2008, Lü and Zhao, 2004). When nanoclay in a resin is either intercalated or exfoliated, it can significantly enhance the mechanical properties of the clay-filled matrix(Sinha Ray & Okamoto, 2003). This is because in these two hybrid structures the thin layers of nanoclay separate in to the polymer with a high aspect ratio which is between 10:1 and 1000: 1.  Figure 2.3: Scheme of different types of composite arising from the interaction of layered silicates and polymers: (a) phase separated microcomposite; (b) intercalated nanocomposite, and (c) exfoliated nanocomposite ( Adapted from Alexandre and Dubois, 2000 with permission).  12  2.3 Nanoclay dispersion methods The method used to disperse nanoclay into resin will greatly affected by the clay hybrid structure and the properties of clay-filled resin. In order to obtain the desired intercalated or exfoliated hybrid structure, various mixing methods had been developed to separate and disperse the clay into the resin. Ribbon mixing, tumbler mixing, high shear mixing, and even manual shaking have been used to disperse clay into resin (Wang, et al., 2008). Mechanical mixing method is the simplest method for blending the clay and liquid resin together.  Some grinding instruments have also been developed for the purpose of shear mixing silicate nanoclay into resin. A comparative analysis of the effect of different milling and grinding such as bread mill, ball mill, three roll mill, and high speed mixing on nanoclay dispersion into a UV coating was carried out by Landry et al., (2008). TEM results showed that three roll milling and bread mill treatments achieved better clay distribution and clay exfoliation. However, milling and grinding methods can have a negative effect on organic modified clays in that the shearing force scan damage the organic modified group on the clay sheet surface during grinding, resulting in reduced mechanical properties in wood composites made with clay- resin mixtures (Cai et al., 2010; Landry, et al., 2008).  Ultrasonic homogenization has also been employed to obtain an even clay dispersion in liquid resin. In a study by Dean et al. (2007), high shear mixing, ultrasonic bath techniques, and cell disruptor horn sonication were compared. Better clay dispersion was observed in the horn sonication and bath sonication treatment than for the high shear mixing method. However, the heat produced by continued ultrasonic vibration can lead to clay aggregation (Lin et al., 2005). Therefore, careful attention to the mixing temperature and mixing time is necessary to avoid  13  overheating. To get a better mixing result, Chowdhury et al. (2006) used a pulse cycle and water bath to control the mix temperature at around 40˚C to 50˚C when applying the ultrasonic technique. Discontinuous ultrasonic vibration is another alternative processing technique for distributing nanoclay into liquid resin as it reduces overheating and clay aggregation (Lin, et al., 2005).  Compared to all the other clay mixing methods, mechanical mixing has been shown to be the simplest and lowest cost method for blending nanoclay and liquid resin. According to Wang et al. (2008), the mixing method used needs only to result in uniform dispersion of phyllosilicate clay into the resin. It is reported that simple mechanical mixing was sufficient to completely exfoliate nanoclay into UF resin (Lei et al., 2008) and produce intercalation with epoxy resin (Adam et al., 2001) and even exfoliation in epoxy resin (Lan and Pinnavaia, 1994).  The ideal mixing temperature for thermosetting resin should be around room temperature (around 20˚C), because thermosetting resins will start curing or setting at higher temperatures. There is no fixed blending time for mechanical mixing; blending time can vary from 5 min to more than 60 min, as long as the clay is dispersed into the resin uniformly (Lei et al., 2008). Blending speed usually ranges from 500 rpm to as high as 3050 rpm (Cai et al., 2007), a medium mixing speed, 800 or 1000 rpm, is preferred (Wang et al., 2008).  When separating nanoclay into thermoplastic resin, a higher blending temperature is preferred. Melt blending is the most common method used to mix nanoclay with thermoplastic resin, at a temperature that is high enough to give the resin adequate viscosity for further processing, such as exfoliation (Lee et al., 2005 , Lei et al., 2007). An analysis comparing different processing parameters on nanoclay dispersion in polyolefin was carried out by optimizing the mixing 14  parameters. Up to 30 to 40 min mixing time was required for clay delamination. A higher mixing speed, 110 rpm, was significantly better than the low speed, 35 rpm. However, higher concentration of clay made it more difficult for the clay to become completely exfoliated in the polymer (Lee, 2008). In summary, higher mixing speeds and longer blending times can greatly improve clay dispersion.  In addition to the optimization of mixing parameters, the use of compatibilizers, such as coupling agents, can also aid in exfoliating the silicate clay (Kim et al., 2003). It is known that long chain organic modifiers, such as alkyl ammonium, will enlarge the interlayer distance of nanoclay therefore facilitating the exfoliation of clay (Labidi et al., 2010). Coupling agents together with organic modifiers also help the clay platelets exfoliate because the coupling agent has the similar effect on hydrophilic nanoclay ( Han et al., 2008). Commonly used coupling agents are maleated polypropylene (MAPP) , silane coupling agent, and are usually used at less than 10% of the matrix mass or based on the amount of clay additive (Zhao et al., 2006, Nourbakhsh and Ashori, 2009).  2.4 Wood nanoclay composites processing  As illustrated in Figure 2. 3, there are two different methods for producing wood nanoclay composites ( Lü and Zhao, 2004). In the case of wood-plastic composites wood furnish can be compounded with polymer, nanoclay and other additives in one step. The two steps method is to prepare the nanoclay and polymer mixture then incorporated the mixture with wood furniture or solid wood to form the wood nanoclay composite. In the case of hot pressed wood panels where resin is used as a binder in small quantities and the resin is pre-mixed with nanoclay and then blended with wood furnish in a secondary step (Lü et al., 2006). 15  Figure 2.4: Flow chart of wood nanoclay polymer composite process (Adapted from Lü and Zhao, 2004 with permission) The process for fabricating nanoclay- reinforced wood plastic composites (WPC) is similar to conventional WPC or other thermoplastic polymers. Compounding thermoplastic resin and nanoclay is a one-step melt blending process and the mix is extruded and cooled in the designed shape of the end product. In the two step process, nanoclay is first dispersed into the resin or molten polymer matrix and wood flour is then melt blended into the clay resin mixture. An alternative one step process for fabricating wood-plastic composites is to blend nanoclay with the polymer and wood flour or wood fiber simultaneously during grinding, batch mixing, or compounding using a twin-screw or single-screw compounder. The mixture is then consolidated into a nano-composite by hot injection modeling, extrusion, or pressing (Faruk and Matuana, 2008).  For wood composite panels that are bonded with thermosetting resin using veneer, particles, strands, or fibers, the processing methods are similar to those for conventional board manufacturing whereby the binder is the nanoclay-resin mixture. The only requirement for the  16  mixing method is that it is sufficient to intercalate or exfoliate the clay into the resin (Wang et al., 2008).  Blending the nanoclay and adhesive is the first step in producing nanoclay-modified wood composites, whereby the blend of adhesive and clay is further mixed with wood furnish. After applying the resin, the furnish is form into a mat then hot pressed into wood panels during hot pressing process ( Lei et al., 2008, Wang, et al., 2008).  Other studies have investigated incorporating nanoclays into solid wood. A common approach for fabricating solid wood-nanoclay-polymer composites is to fill the cell lumens of solid wood with a clay resin mixture (Lü et al., 2006). In this case the oven dried solid wood was placed under vacuum to remove air, then dipped into liquid resin-clay mix under atmospheric pressure (or higher pressure) and evacuated again after absorption. The wood pieces which impregnated with resin were then air dried in fume hood for one day (Cai et al., 2008).  2.5 Characterization nanoclay-resin hybrid structure  There are several techniques used to investigate the distribution of nanoclay resin mixtures and the effect of the nanoclay on resin properties. X-ray diffraction (XRD) is the main technique used to provide information on the degree of clay agglomeration and exfoliation. Transmission electron microscopy (TEM) and scanning electron microscopy (SEM) are also useful for visualizing the homogeneity of clay distribution in the matrix by providing direct observation of clay platelets or agglomerates in the hybrid structure.  The addition of nanoclay into thermoplastic resin will affect the chain mobility of polymer and the fillers share the stress with polymer matrix (Sui et al., 2007). The presence of a nanoclay will 17  reduce the crystallinity of thermoplastic resin, increase the free volume and therefore change the glass transition temperature. The change in degree of crystallinity can be detected by XRD and DSC, and glass transition temperature can also tested using DSC and Dynamic mechanical analysis (DMA).  2.5.1 X-ray diffraction X-ray diffraction (XRD) has been extensively used for characterizing the microstructure of polymer/layer silicate nanocomposites. XRD is traditionally used to identify and analyze the crystal structure of solid materials. In the silicate layer arrangement of montmorillonite the interlayer distance of the pristine clay and the modified clay is in the range of 1-4 nm (Giannelis, et al., 1999b). The interlayer space between clay layers can be detected by XRD by a peak in the x-ray intensity at a characteristic angle and the inter-platelet distance calculated ( Bragg and Bragg, 1913). According to Bragg’s law, the interlayer spacing (d) in nanoclays and the relative intercalation (RI) of the polymer in nanoclays can be determined using the following equations (Pegoretti, 2007): 𝑛𝜆 = 2𝑑 sin 𝜃  𝑅𝐼 =  (𝑑−𝑑𝑜 ) 𝑑𝑜  × 100  (2. 1) (2. 2)  where n is the integer number of wavelength (n = 1); λ is the wavelength of X-ray; d is the actual interlayer or d-spacing of the clay in the matrix; θ is the diffraction angle corresponding to a specific intensity peak, and d 0 is the d-spacing of the sheets in the pristine clay or organic modified clay. Usually, the results of XRD analysis are 2-dimensinal XRD patterns showing the 2θ angle (twice the diffraction angle) on the horizontal axis and the vertical axis representing the 18  intensity of the X rays. The distance between clay lattices will result in an intensity peak which appears at a specified angle.  From the Bragg Law, the lower peak diffraction angles indicate larger distances between the interlayers of nanoclay. If the nanoclay is totally exfoliated in the resin none of the clay platelets are aligned with each other and the diffraction angle may be too small to be detected. If the nanoclay is intercalated with the resin, the clay platelets are still parallel to each other and just further apart than in the pristine clay, the intensity peaks will shift to a lower angle.  In addition to characterizing the clay dispersion, XRD can also provide other information about the hybrid resin such as crystallite size of thermoplastic resin (Vaia and Liu, 2002). The crystallinity of thermoplastic resin shows up as intensity peaks in XRD patterns. In the case of thermosetting resins, which are generally amorphous, non-crystalline solids there are no sharp intensity peaks in the XRD pattern, instead, it shows up as smooth, rounded wide peaks.  The degree of clay intercalation is quantified at the lower X-ray angles while information on the crystalline phases in the polymer usually appears in the higher range of diffraction angles (Sarrazin et al., 2005). For the range 18˚-30˚, the X-ray intensity peak of clay-added HDPE and clay and wood flour reinforced HDPE were lower than that of pure HDPE, suggesting that the addition of clay and wood flour decreased the crystallite of HDPE in the clay hybrid (Lei et al., 2007).  2.5.2 Microscopy techniques Transmission Electron Microscopy (TEM) can provide direct visual evidence of clay dispersion (Morgan and Gilman, 2002). The silicate layers in thin sliced sections show-up as dark lines in 19  TEM images, while the light areas correspond to resin (Deka and Maji, 2010). When the clay is totally exfoliated, the platelets are visible as dark lines in different orientations. Intercalated clay shows up as parallel layers of alternating dark and light bars (Giannelis et al., 1999a). Clay aggregation can also be easily observed in TEM images. TEM was used by Landry et al., (2008) to evaluate which clay dispersion process (bead milling, ball milling, three roll milling, and high speed mixing) achieved better clay distribution, showing that three-roll milling and bead milling lead to better dispersion than bead milling and high shear mixing treatment.  Vermogen et al., ( 2005) used statistical analysis of TEM images to evaluate the clay dispersion in clay-plastic composites which were prepared using different screws in a twin screw extruder. The clay platelet length, thickness and aspect ratio, inter-platelet distance and the amount of clay aggregation were quantified and statistically analyzed. Based on these results the effect of screw profile on the final clay morphology was assessed.  The main issue with TEM and SEM techniques is the very small volumes of materials examined may not be representative of the whole volume, and many different samples are required in order to develop a more comprehensive view of the bulk of the material. A solution to examining a large number of samples is to use microscopy techniques in conjunction with XRD.  Microscopy is not only useful for direct observation of the clay dispersion, it can also examine the effect of clay on the crystallization and the facture toughness of the resin. Using SEM examination, nano-sized spherulite crystals were easily observed in pure PP, but the size of spherulites was reduced with the addition of nanoclay, suggesting that the presence of the clay interfered with the growth of spherulites (Perrin-Sarazin et al., 2005). This effect was even more pronounced when a coupling agent was added (MA330k). 20  The facture behavior of clay reinforced resin has also been examined by SEM. Typically, the fracture surface of neat resin is smooth with very few cracks are appearing on the surface indicating a typical brittle facture. A cracking trail is formed when two secondary crack fronts come together (Wang et al., 2005). The micro cracks start between the between clay particles and weakly bonded layers then grow and extend further into the matrix when further load is applied. The path taken by the crack can be quite tortuous with the presence of clay, as shown in Figure 2.6. The crack will absorb more energy during growth than the more brittle, smoother cracks. Therefore the sub fracture surface area will increase and share more stress. At low clay content, sample showed minimal fracture surface roughness which isolated to small regions because the clay plays a part in reducing the stress concentration and therefore makes the matrix more resistant to the crack growth (Chen et al., 2003). At a higher clay concentration, this mechanism works on a smaller scale because the distance between clay platelets is decreased. Similar behavior in unsaturated polyester-clay nanocomposites is also reported (Adam, 2001).  Figure 2.5: Crack initiation process of clay-filled epoxy resin ( Adapted fromWang et al., 2005 with permission). 21  One possible reason for clay reinforcement is that the clays share the stress with the resin matrix and prevent the crack to growth. However, this mechanism is as not as effective to gain strong improvement in fracture toughness.  2.5.3 Thermo analysis As was alluded to earlier, once nanoclay is added to the resin matrix the polymer’s crystallization and chain mobility can change. Thermoanalysis instruments such as Differential scanning calorimeter (DSC) and Dynamic mechanical analysis (DMA) can be used to detect these changes and characterize the clay’s effect on resin on several parameters: i.e., glass transition temperature (T g ), crystallization temperatures (T c ), heat of fusion, and entropy of fusion(ΔH) (Menczel and Prime, 2008).  The result of a DSC experiment is a plot of heat flow versus temperature or time. Examples of exothermic peaks (a heat producing event) on the curve, an example of which is shown in Figure 2.7, include crystallization and oxidation reactions, while an endothermic event (heat absorbing) includes phenomena such as melting and decomposition (Hon, 2003).  22  Figure 2.6 was the schematic of DSC curve of transition enthalpy as the temperature changes. The image is not given here according to the Canadian copyright law. Please find the original image in the following reference:  Hon, D. . (2003). Analysis of Adhesives. In K. L. Pizzi, Antonio; Mittal (Ed.), Hand book of wood Adhesive technology (2nd ed.). (pp. 129-150).M. Dekker.  Figure 2.6: Schematic of DSC curve (Hon, 2000). The enthalpy of transition, ΔH, can be expressed using the following equation: 𝛥𝐻 = 𝐾𝐴  (2. 3)  where K is the calorimetric constant and A is the area under the curve. The degree of crystallinity, χ c , of a thermoplastic polymer composite can be computed using the R  following equation:  𝛘𝐜 =  ∆𝐇𝐞𝐱𝐩 ∆𝐇  𝟏  × 𝐖 × 𝟏𝟎𝟎% (2. 4) 𝐟  where ΔH exp is the measured heat of crystallization, ΔH is the heat of crystallization for a 100% crystalline polymer (e.g. ΔH for pure HDPE is 293 J/g), and W f is the weight fraction of thermoplastic resin in the composite.  Lei et al., (2007) found that the addition of pine flour or clay will lower the crystallization rate of HDPE. However, the use of a compatibilizer, maleated polyethylene (MAPE), increased the 23  crystallization of the HDPE in the HDPE -clay-pine composite. The nanoclay acted as an nucleating agent for PP and reducing the crystallization rate provided sufficient time for clay diffusion and produced a more intercalated arrangement (Maiti, Nam, and Okamoto, 2002).  In most cases, the curing peak temperature and heat of reaction are used to quantify uncured thermosetting resins. Liquid samples are usually sealed in high volume pans or tested in pressure DSC to minimize the confounding effect of water evaporation. The cure peak temperature indicates the maximum curing rate (Sichina and Manager, 2000). Becker et al., (2003) found that the presence of nanoclay reduced the resin curing rate as evidenced by a broadening of the exothermic peak and the peak shift to a lower temperature. Other studies suggest that nanoclay does not strongly affect the onset temperature and the peak temperature of the cure reaction (Hussain et al., 2007) ( Ton-That et al., 2004).  MMT-modified UF resin cures faster than pure UF resin with hardener, as evidenced by earlier onset and higher exothermic peak temperature of than UF resin (Lei et al., 2008). The DSC curve of nanoclay and hardener-added UF resin also displayed a wider and lower exothermic peak than that of pure UF with hardener.  Nanoclays may also influence the resin at the molecular level by changing the molecular motion and chain flexibility. Such changes are difficult to detect using DSC but can be measured using DMA (Hatakeyama and Quinn, 1999). A typical DMA scan for a thermoplastic resin is shown in Figure 2.8.  24  Figure 2.7 was not given here according to the Canadian copyright law. It was the curves of tan δ, storage (E') and loss (E'') modulus for a sample of PVA in DMA test. Please find the original image in the following reference: Hatakeyama, T., & Quinn, F. X. (1999). Thermal Analysis Fundamentals and Applications to Polymer Science. (T. Hatakeyama & F.X. Quinn, Eds.)Recherche (2nd ed., p. 131). Toko,Japan: John wiley& sons.  Figure 2.7: Tan δ, storage (E') and loss (E'') modulus for a sample of PVA (Hatakeyama and Quinn, 1999).  For thermoplastic resins the storage modulus, E', is the dynamic elastic response of the sample. The loss modulus, E'', is the dynamic plastic response of samples, and tan δ is the ratio of loss modulus/storage modulus. Usually, the peak of tan δ or E'' are defined as T g . As can be seen from Figure 2.8, E' is approximately constant from before T g , after which point it abruptly decreases. Below T g there is no slippage between adjacent molecules because energy is much less as chain movement cannot occur whereas above T g slippage occurs and the ability of the sample to store elastic. The peak of loss modulus or the tan δ shows that the sample passed through the glass transition temperature (Groenewoud, 2001).  Most studies have found enhancement of storage modulus with the addition of nanoclay. Different studies have found various different clay loadings whereby there is maximum improvement in storage modulus. Nanoclay-reinforced polyethylene (PE) had a higher storage modulus than neat PE, and the storage modulus and loss modulus increase with clay content (Lee 25  et al., 2005). Similar results were obtained for HDPE-clay mixtures, indicating higher stiffness of clay-reinforced polymer and reduced mobility of the polymer chains between the nanoclay layers.  Chowdhury et al., (2006) used pre-cured epoxy resin to impregnate a woven carbon fabric which was tested using a DMA single cantilever beam mode. Storage modulus reached the highest value at the 2% nanoclay (in this case Nanomer® I-28E) loading rate while at 3% clay loading storage modulus was lower than that of the neat resin. Miyagawa, et al. (2004) found that the storage modulus of anhydride cured epoxy/clay composite increased with increasing clay content. A 13% increase in storage modulus was obtained with only 2.5 wt% clay addition. These improvements may have resulted from the high aspect ratio and the interfacial adhesion between epoxy resin and nanoclay (Kotsilkova and Pissis, 2007).  Glass transition temperature (T g ) can be related to molecular weight, cross-linking density, free volume density, and strength of the interface layer between the nanoclay platelets (Hussain et al., 2007). If the T g of a nanoclay-filled resin increases to a higher temperature, then the clay has improved the resin’s thermomechanical properties. However, Awad (2009) found the nanocomposite’s T g was not significantly different from the pure polymer. The T g of a thermosetting adhesive is dependent on the degree of cure and water content (Lapique, 2002) and therefore T g is not as reliable for characterizing thermosetting adhesives.  Mostly, DMA is applied in detecting the curing process of thermosetting resin. In contrast to thermoplastic resin the storage modulus increases as the temperature increases and reaches the peak temperature when the resin was completely cured and forms a highly cross-linked network (Figure 2.9) (Pilato et al., 2010). The onset of curing and end of curing is determined by the 26  storage modulus, the different of the maximum and minimum E′ (ΔE′) represent the rigidity of the resin network (Park and Kim, 2008).  Figure 2.8 was not given here according to the Canadian copyright law. It was the storage (E') and loss (E'') modulus curve of resole resin in DMA test. The storage modulus increased as the temperature increased. Please find the original images in the following reference: Pilato David Nagy, Ellen, L. V. (2010). Phenolic Resins: A Century of Progress Analyses/Testing (pp. 93-135).  Figure 2.8: DMA result of a resole resin. (Pilato, et al., 2010)  Preparation of DMA samples for thermosetting resin’s is different from that of a thermoplastic resin, especially for studying the curing process of thermosetting resin (ASTM, 2008). Various methods had been developed for preparing thermosetting resin DMA samples. One involves placing resin between two plywood pieces as a sandwich sample (Lei et al., 2008), another is to impregnate glass fiber (Mequanint and Sanderson, 2003), carbon fiber (Mequanint and Sanderson, 2003 Kim et al., 1991) or other supporting materials with resin to form a thin film.  The storage modulus of nanoclay-filled MUF resin first decreased with the temperature from 30°C to around 80°C, but increased as heating process continued and reached a plateau. With the presence of nanoclay filler, MUF increased in both the storage modulus and tan δ (Cai et al., 2010). In general, the storage modulus of thermosetting resin (derived from E′ in the rubbery  27  plateau) represents the rigidity of the resin network (density of crosslinking) and increases with nanoclay loading (Park and Kim, 2008, He and Riedl, 2003).  2.6 Nanoclay reinforcement mechanism  It is believed that some chemical bonding between the resin polymers with and nanoclay platelet surfaces may occur. Fourier transform spectroscopy (FTIR) and Nuclear magnetic resonance (NMR) have been employed to investigate the chemical interaction between the nanoclay and resin. Lü et al, (2006) modified PF resin with organic MMT and then impregnated the mixture into solid wood to produce a wood-MMT nanocomposite. The FTIR spectrum of this composite showed −OH vibrating adsorption, which indicates strong linking between PF and MMT had generated by their oxygen atoms. Stronger −C−O vibrating adsorption indicated that more chemical combinations had been built-up between organic clay and the wood (Lü et al., 2006). Han, et al., (2008) has also verified enhanced chemical bonding between organic clay, coupling agent and pMDI resin using both FTIR and NMR.  The large aspect ratio (the ratio of length and width) and large specific surface area of nanoclay platelets is a major contributor to reinforce the polymer matrix. The addition of larger aspect ratio platelets results in a greater increase in the elastic modulus (stiffness) for clay-reinforced resins (Miyagawa et al., 2004). The intercalated or exfoliated hybrid structure of nanoclay-filled resins also improves the fracture toughness of resin. Nanoclay was shown to have larger interfacial adherent surface to the resin (Siddiqui et al., 2007).  The nanoclay platelets hamper the propagation of cracks in a hybrid resin as evidenced by the rough fracture surface of nanoclay-resin composites compared with fracture surface of the same  28  resin containing no nanoclay. As the crack pass through the clay region they impinge on the platelets and later branches. As the clay content increase, the distance between clay platelets decrease creating a more tortuous route for crack propagation (Adam et al., 2001). The major toughening mechanism of nanoclay in resin matrices is that cracks branch into more microcracks to yield more fracture surface areas (Wang et al., 2005). At low clay loading, from 2 to 5%, fracture toughness increased with clay content (Siddiqui et al., 2007).  2.7 Characterization of properties of polymer-nanoclay-wood composites  There are more reports in the literature on the effectiveness of various nanoclays to improve the properties of wood-plastic composites than there are for conventional wood composites such as particleboard. This section briefly covers the studies for these two types of composites.  2.7.1 Dimensional stability of clay-reinforced nanocomposites (thickness swell and water absorption) The dimensional stability of wood-nanocomposites or nanoclay-polymer mixes are reported to be significantly improved with the incorporation of nanoclays. One property that was unaffected by the presence of the nanoclay was the fire retardant of polymers (Zhang et al., 2009, Wang et al., 2007).  Most of studies of nanoclay-added -wood composites found that thickness swelling decreased. Deka and Maji (2010) reported that for clay contents up to 10% in wood flour/ HDPE composites, thickness swell decreased with increasing clay content. For example, adding 2% nanoclay into HDPE-wood flour composites reduced thickness swelling by 41%; higher clay contents further reduced swelling. This may be because the exfoliated clay produces a longer 29  moisture diffusion path. In terms of water absorption Yeh (2007) found that water absorption decreased between 10% to 40% (Yeh, 2007) . The addition of the coupling agent, maleic anhydride grafted polyethylene (MAPE), into PP/wood flour composites further improved the dimensional stability as the coupling agent reduced the chain movement by increasing chain cross linking in the polymer (Sheshmani, Ashori, and Hamzeh, 2010). The thickness swelling rate of PP/bagasse composites decreased with increasing nanoclay content up to 8% (Amir Nourbakhsh and Ashori, 2009).  Cai et al., (2007) studied the water absorption and thickness swelling of clay-resin impregnated solid wood. MUF treated wood showed lower water absorption and thickness values than solid wood, MUF/nanoclay treated wood had significantly lower values than MUF treated wood and solid wood. The nanofiller, Claytone®APA, turned out to be more effective than Cloisite® 30B and Cloisite® Na+.  Thickness swelling of nanoclay-reinforced MDF (Medium Density Fiberboard) has also been investigated. The thickness swelling decreased significantly as the clay content increased. It is believed that voids in the composite and the lumens of fibers were filled with nanoclay which prevented the penetration of water by capillary action into the deeper parts of composite .  2.7.2 Mechanical properties of polymer-nanoclay and wood -nanoclay composite Flexural strength and tensile strength are key properties used to evaluate the performance of wood-based nanocomposites. Studies have shown that the addition of nanoclay influences the flexural properties of both polymer and wood based composites. Clay loading rate has the  30  stronger effect on strength properties, but other factors such as clay type, blending method and coupling agent also play a role.  Faruk and Matuana, (2008) tested five types of Cloisite nanoclays in modified HDPE wood flour composites. They were able to enhance polymer properties using the melt blending process whereas dry blending of ingredients was not effective and that the Closite10A was more effective than other clays. The addition of a coupling agent resulted in higher MOR, MOE and tensile strengths compared with Cloisite10A alone. Lei et al., (2008) found that 1% clay addition to HDPE improved its tensile and flexural strength by 24.2% and 19.6% respectively. Higher clay content reduced the extent of strength increase. The flexural and tensile modulus of PPwood flour- nanoclay composite were increased at 3% loading of organic-modified montmorillonite (OMMT) but were reduced at a higher 6% loading rate. Both of the clay loading rates, 3% and 6%, decreased the impact strength of the composite (Kord, Hemmasi, and Ghasemi, 2010).  Nanoclay has been shown to influence the mechanical properties of wood-based panels. In a study by Ashori (2009), Cloisite Na+ was mixed with dried UF-resinated MDF wood fiber. In the 2% to 6% clay content range, MOR and MOE increased as the clay content increased, but the effect was reduced at 8% clay loading. Higher clay concentration can lead to clay aggregation, reduced bonding strength of the adhesive and reduced board properties (Ashori and Nourbakhsh, 2009).  Internal bond, IB, strength refers to the tensile strength perpendicular to the surface of the panel. According to Ashori and Nourbakhsh (2009), IB strength of MDF increased for clay contents from 2 to 8% and reached a maximum value of 0.6 MPa at 4%. Lei et al. (2008) exfoliated 31  different percentages of nanoclay into UF resin by mechanical stirring and made particle board bonded with the clay-resin mix. They found that in 2-8% range, higher clay loading rate improved the IB strength more evidently.  Wang et al. (2008) used UF resin containing 1 or 2% nanoclay to produce Oriented Strand Board (OSB). They found that IB strength increased by 28% for a 1% clay addition while the IB strength of the samples made with a 2% addition only increased by 11%. However there were no significant changes in the other board properties and in some cases decreased slightly. It was shown that substituting 1% clay in liquid PF resin resulted in the same board properties as pure PF resin (Wang et al., 2008).  In conclusion, the addition of nanoclay can enhance the mechanical properties of nanoclay polymer composites and nanoclay-reinforced wood composite up to a point and then tended to decrease at higher clay contents. Thickness swelling also benefitted from the addition of a small amount of nanoclay to the resin and it is postulated that it filled voids in the furnish and blocked water ingress.  2.8 Summary  Nanoclay-polymer composites, especially platelet nanoclay- reinforced polymers have been extensively studied. There are three types of nanoclay distribution arrangements in the polymer matrix: phase separated, intercalated and exfoliated. Various methods have been used to distribute clay to achieve an intercalated or exfoliated state which is ideal for polymer reinforcement. The methods include, but are not limited to, mechanical mixing, high shear mixing, melt blending (for thermoplastic polymers only), ultrasonic dispersion, and even  32  grinding methods such as ball mill mixing and bead mill mixing. Each mixing method has its advantages and disadvantages. To get uniform clay distribution the appropriate method should be selected according to the material properties. The mixture of clay and resin can be further processed to form a nanoclay-polymer composite or wood-nanoclay-polymer composite. TEM, SEM and XRD have been used to examine the clay dispersion in the polymer or resin matrix. The distance between the phyllosilicate layers can be determined by the X-ray diffraction angle. When clay is intercalated or exfoliated the distances between sheets is greater, it shows in the XRD pattern as the peak of diffraction angle shifting to a lower value. SEM or TEM can further elucidate the clay distribution and is useful for analyzing fracture surfaces and cracking behavior of the nanocomposites.  DSC and DMA tests measure the effect of nanoclay on the thermal behavior of polymers and resins. DSC curves have shown that the clay decreases the degree of crystallinity of thermoplastic polymers and wood plastic composites, making them less brittle. The addition of nanoclay improves the thermo-mechanical properties of thermoplastic resin and influences the curing process of thermosetting resin by altering the cross linking of resin.  Some studies have concluded that nanoclays with high aspect ratio and specific size can be used as fillers to improve the mechanical properties of the polymer matrix. FTIR and NMR analyses show there are extra chemical bonds formed between organic-modified clay and resin which may also contribute to the clay reinforcing mechanism. Organic modifier and coupling agents help separate the clay into resin and facilitate the interaction between clay and resin.  Phyllosilicate nanoclay has been shown to significantly improve the water resistance of nanoclay-wood composites, higher clay content results in reduced water absorption. This is 33  especially so for MDF and wood plastic composites. Most studies that have added nanoclay to conventional wood composites have found that the mechanical properties including bond strength and bending strength are unaffected or improved. The most significant enhancement of properties are for dimensional stability, i.e., reduced thickness swell and water absorption. There are fewer studies on, and less success with adding nanoclays to solid wood.  34  3 Materials and methods 3.1 Materials PB Furnish: The particleboard furnish used in this work was provided by the NewPro Particleboard plant located in Smithers BC. The furnish consisted of spruce (Picea glauca) and pine residues from saw-mills and other facilities in that region.  The particles are produced from saw mill residues, such as hogged mill waste, sawdust, planer shavings, and are distinguished by size: coarse, medium and fine furnish by screening. In this work, the coarse furnish can pass through the 9-mesh screen (mesh opening size is 2.0mm), and the fine furnish can pass through a 32-mesh (mesh opening size is 0.5mm) ( Sackey, et al., 2008). Three layered particleboards were made with coarse furnish in the core layer and fine particle in the face layers. The moisture content of the furnish was approximately 7%.  Veneer: Sliced aspen (Populus tremuloides) veneer, 0.027-inch or 0.7 mm thick was used for lap-shear tests. The moisture content of this veneer was measured to be 8%.  Resins: The resins used in this study were urea formaldehyde (UF) and melamine formaldehyde (MF) resins provided by Momentive Ltd. (previously Hexion Ltd.) (Table 3.1). For some of the experiments a coupling agent, 3-Aminopropyltriethoxysilane (purchased from Alf Aescer) was added to the resins and that mixture used to make PB and the lap-shear specimens.  Table 3.1: Nominal properties of the UF and MF resins. Resin  Name  pH  UF MF  Casco-resin C04SS Casco-Resin HM707  8.1-8.4 9.1-9.5  Solid content (wt. %) 62 57  35  Nanoclays: The nanoclays used in this study were all platelet-based montmorillonite (MMT) clays provided by Southern Clay Ltd. (Austin TX, USA) and are listed in Table 3.2. These clays modified with different quaternary ammonium chlorides. The selected organic modified nanoclay, Cloisite®30B is considered to be organophilic while the pristine montmorillonite Nanofil®116 is hydrophilic. The median particle size of each of these clays was reported to be 13 μm. Table 3.2: Nanoclays supplied by Southern Clay Ltd. Nanoclay  Organic modifier (exchange cation)  Cloisite®30B Methyl, tallow, bis-2-hydroxyethyl, quaternary ammonium Nanofil®116  None  3.2 Preparation of resin and nanoclay mixtures Mixtures of each resin and nanoclay were prepared by measuring out 200 g of resin and the appropriate mass, 4, 8 or 12 g of nanoclay added to the resins and the mixture stirred using a high-shear mechanical stirrer (Lightning Labmaster Model TS-2010, Figure 3.1 a) using a rotation speed of 1000 rpm for 30 min. The 33.2 mm diameter impeller agitator of stirrer is showed in Figure 3.1 b. The actual concentrations of the clays in the resin were 1.96, 3.84, and 5.66 wt/% for the 4, 8, and 12 g clay additions, respectively.  36  (a)  (b)  Figure 3.1: The Labmaster TS-2010 mechanical mixer: (a) the mixing head and (b) the impellor agitator. Batches of 200g liquid resin and clay combination were blended using the mechanical stirrer shown in section 3.2 at a speed of 1000 rpm for 30 min. For the treatments where coupling agent was used, the coupling agent 10% of the clay mass was added into the resin before the clay, mixed for 5 min to allow its hydrolysis. Clay was then added into the resin for further 30 min mixing time.  3.3 Evaluation of bulk resin properties For XRD and DSC tests, two types of platelet nanoclay, Cloisite30B, Nanofil116, were added into UF resin respectively, at a loading of 2%, 4% and 6%, with and without a coupling agent. When coupling agent was used it was added at 10% of the clay mass. Another 4 batches of MF resin containing nanoclays were also prepared with and without coupling agent. DMA and Lapshear test were made only for the 2% clay loading.  37  3.3.1 X-ray diffraction test After mixing using the mechanical stirrer, clay/resin mixtures were cured in a drying oven at 103°C for 24 hours, removed from the oven and cooled. The samples were ground down to powder and mounted in the sample holders of a D8 Focus (Bruker) X-ray diffractometer, and scanned from 3° to15° with a step size of 0.04°, and 0.8 s/step. X-ray radiation was generated by using a 35KV, 40mA Cobalt radiation source.  3.3.2 DSC test of resin and clay mixture Since the curing of the resins may be affected by the clay in the mixtures, the curing of these mixtures was examined using a TA Q1000 Dynamic Scanning Calorimeter (DSC). A high volume pan was placed on an analytical balance and a 10±5 mg sample of either pure resin or the resin-clay mixtures pipetted into the pan and the actual mass of the sample recorded. The O-ring and lid were then placed to cover the pan and the sample then crimped shut using the sample encapsulating press provided with the DSC pan kit. A second reference pan containing no resin was also crimped closed and these pans then placed in the DSC, and calibration was performed using indium standards. Prior to performing a DSC scan, the cell temperature was equilibrated at 20°C and the samples and reference pans heated from 20°C to 200°C at a constant rate of 10°C/min using nitrogen as a purge as at a 50 ml/min flow rate.  3.3.3 Dynamic mechanical analysis (DMA) test For each resin, four treatments were applied for each resin: 2% Cloisite30B addition with or without coupling agent (0.2% based on resin weight or 10% of clay weight), 2% Nanofil116 with and without coupling agent. To detect the curing process, samples were pre-cured by first impregnating resins into filter paper which supports the resins film then heat drying. Filter papers 38  were cut into strips 60 mm long by 12 mm and soaked in the resin mixture for 24 hours to absorb sufficient resin then dried at 80°C for 24 hours to form a thin solid film. Samples were examined using a DMA TA Q800 with a 3-point bending clamp type, scanned from 60°C to 200°C, at a frequency of 1Hz. 0.01% and 0.05% stress applied to the samples respectively.  3.3.4 Lap-Shear test - automatic bonding evaluation system (ABES) DMA tests relevant to the resin’s cohesion strength (Park and Kim, 2008), which evaluate the bulk resin properties. Lap-shear strength can determine the resin’s bonding strength with wood and shear strength of resin. So the resins with the same treatment in DMA were also test in this experiment to further evaluate the effect of clay on resins.  Aspen veneers were cut into pieces 120 mm long by 20 mm wide using a pneumatic clipper (Figure 3.4 a), the veneers that had straight grain and without defects were selected. The overlap length was 5 mm providing an overlap area of 5 mm by 20 mm (The overlap area is 100mm2, 1Mpa =1N/mm2, therefore the strength unit is MPa).  Mixtures of resin and clay were prepared as described previously. The resins were applied to the veneer samples using a small paint brush and the veneer was weighed (Figure 3.2 b). The mass of resin applied to the veneer was in the range of 0.009g-0.01g. The open assembly time for all samples was approximately 1 minute. The Automatic Bond Evaluation System (ABES) has small platens (Figure 3.2 c) that hot press the veneers together and then measure the shear strength of the bond line by pulling the veneers apart. Platen temperature was set at 140°C for UF resin and 160°C for MF resin. The two veneers were placed on the ABES unit and pressed together at a pressure of 1 MPa. Hot pressing time for UF resin was 60s, and 240s for MF resin. 8 replicates for each resin treatment were tested. 39  (a)  (b)  (c)  Figure 3.2: (a) air-operated clipper, (b) applying resin, (c) automatic bond evaluation system.  3.4 Manufacture of particle boards and evaluation of the effect of clay loading on board properties The three clay-resin mixes: Cloisite30B mixed with coupling agent modified MF, Nanofil116 in MF, and Nanofil116 mixed with coupling agent modified MF had higher lap-shear strengths. These mixes were selected as the Clay Type factor for trials in fabricating particleboards to evaluate the effect of clay loading level on board properties. Three clay loading levels, 2%, 4% and 6%, were used.  Three-layer particle boards measuring 25 by 25 by 5/8 inches were made; the experimental parameters are listed in Table 3.3. The mass of each component was calculated using the oven dried furnish mass as the basis for the calculations. There are 3 clay types each with three different loading plus one control treatment containing no clay for a total of 10 treatments; 3 replicate particleboards were made for a total of 30 boards in this phase of the work.  40  Table 3.3: Experimental parameters and response variables for the large board production. Variables: Clay Content Clay type  Replicates Constants: Resin Type Resin Solids Content Board type Board Length Board Width Board Thickness Board Resin Content Board Wax Content Face Furnish Moisture Content Core Furnish Moisture Content Board Moisture Content Shipping Density of Board The Ratio of Face Furnish Responses variables: Internal bond (IB) Screw Withdrawal Resistance (SWR) Thickness Swell (TS) Bending properties (MOR/MOE)  0, 2%, 4%, 6%wt of resin weight Cloisite30B Cloisite30B with10% (clay wt.) coupling agent Nanofil116 Nanofil116 with 10% (clay wt.) coupling agent 3 MF 57 wt% 3 layers 25 inches 25 inches 5/8 inches 10 wt% odw for both face and core layers 1.5wt% odw 7% odw 7% odw 2% odw 45pcf 46% of the total furnish mass Number of samples Number of samples per board per treatment 14 42 8 24 2 6 2 6  3.4.1 Three layer particle board manufacturing process The required amounts of materials were weighted out and transferred into the Drais particleboard batch-blender (Figure 3.3 left). The surface furnish was blended first and due to the small size of the blender, furnish was blended with resin in two batches with the first batch used for the first two replicates and the second batch was blended for the third replicate. The same strategy was used for blending core layer furnish. The Pathex hot press (Figure 3.3 Right) was preheated from room temperature to 180˚C. 41  Resin Pipe Gas Pipe  Resin Pot Balance  Figure 3.3: (left) Drais particleboard batch-blender and (right) Pathex press.  Resin and clay and coupling agent (if applicable) were mixed as described previously and the resin-clay mixture added to a paint pot. The paint-pot was then closed and 30 kPa of pressurized air applied to it. The spray nozzle was then bled until resin began to spray from the nozzle at which point the ball valve on the top of the paint-pot was closed. This step was necessary in order to ensure that all air in the line leading-up to the spray nozzle had been displaced. Emulsified wax was added into to the furnish using a spray bottle before blending with resin to enhance moisture resistance.  The paint pot was then placed on top of a balance accurate to ±1 gm and the balance tared. The Drais blender, shown in Figure 3.3, was turned on, the furnish allowed tumble for 1 minute to help distribute the wax and then the nozzle inserted through a hole in the lid of the blender and the resin sprayed onto the particles. As resin was being sprayed, the mass of the paint pot was monitored and the ball value closed once the correct resin-clay mass had been sprayed onto the furnish. This process took approximately 10 minutes. At this point the spray nozzle was 42  removed, the time noted and the blender left to run for an additional 10 minutes. This was done to ensure that the wood particles were evenly coated with resin. After that the blender was turned off and the furnish was left in blender for a further 10 minutes to allow any fine aerosol droplets to settle out. Then these furnish were ready to form a mat.  A 25” by 25”forming box was place on top of an aluminum caul for forming the mat. The top and bottom layers were fine surface furnish, and the middle layer was composed of the coarse furnish. All three layers were distributed into the forming box consecutively and flattened using a small thick plywood sheet by hand before adding the next layer. The final step was to use the tamper to compress the mat to reduce air gaps and reduce its height before removing the forming box to reveal an even square mat and transfer it into the hot press with another caul on top. The hot pressing schedule for all mats is given in Table 3.4. Pressed boards were cooled to room temperature and then transferred to the conditioning room for 2 weeks to equilibrate the moisture content.  43  Table 3.4: Hot press cycles parameters Proj. Ref. Prod. Ref Press ID Density SEG  LPBNANO Date Particleboard Pane ID Pathex Mat Length 45pcf Thickness Control Set point  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17  fastposn position position position position position position position pressure pressure pressure pressure pressure pressure pressure pressure position  12-13-2010 0M-1 25in 0.625in SEG time  -0.500 in./s 50.00% 38.1mm(1.5 in.) 25.4mm(1.000 in.) 19.05mm (0.75in.) 19.05mm (0.75in.) 15.88mm(0.625in.) 15.88mm(0.625in.) 2.41MPa(349.54psi) 1.72MPa(249.9psi) 1.03MPa(149.6 psi) 0.69MPa(100.2psi) 0.52MPa(74.8psi) 0.34MPa(50.1psi) 0.17MPa(25.1psi) 0Mpa 380mm(15in)  30 s 1s 5s 5s 5s 20 s 5s 400s 10s 10s 10s 10s 10s 10s 10s 20s 20s  Time File Name Mat width Caul thick End Condition  15:18:18 lananopb0m1 25in 0.48in  Position<=80.01mm  Position<=15.88mm Position<=15.88mm Position<=15.88mm Position<=15.88mm Position<=15.88mm Position<=15.88mm Position<=15.88mm Position<=15.88mm  3.4.2 Test of mechanical properties of large boards Each 25” by 25” board was first trimmed to 22” by 22” and then cut into test specimens for MOR/MOE, SWR, TS/WA and IB tests, the dimensions of which are given in Figure 3.4. The dimensions of the SWR specimens was 6" by 3", 6" by 6" for the TS specimens, 2” by 2” for the IB specimens, and 17” by 3” for the MOR/MOE specimens. All specimens were stored in the conditioning room for one week prior to testing. Specimen moisture content were around 7%8%, density was around 0.66-0.69g/cm3.  3.4.3 Board properties testing Specimens for flexure test, screw withdrawal (SWR) resistance test, thickness swelling (TS) test, internal bonding (IB) test were prepared and tested according to ASTM standard D1037-06a. All 44  mechanical tests were carried out on the Sintech 30/D machine using the TestWorks testing control system.  Figure 3.4: Cutting pattern of 25" by 25" particle board For the static bending test, the width and thickness of the samples were measured for calculation of MOR and MOE. The static bending tests were carried out in the center-point loading mode (Figure 3.5) with a 15" span between the two supporting bars. The load was applied to the sample mid-point at a rate of 0.3 in/min (determined according to the standard). The MOR was calculated as follows: and the MOE was decided by the thickness, width of the samples as well as the slope of the deflection curve.  𝐌𝐎𝐑 =  𝟑𝑷𝒎𝒂𝒙𝑳 𝟐𝒃𝒅𝟐  (3. 1) 45  𝐌𝐎𝐄 =  𝐋𝟑  𝟒𝐛𝐝𝟑  ∆𝐏 ∆𝐲  (3. 2)  where: a = area under load-deflection curve to maximum load, lbf·in. (N·m), b = width of specimen measured in dry condition, in. (mm), d = thickness (depth) of specimen measured in dry condition, in. (mm), L= length of span, in. (mm), MOR= modulus of rupture, psi (kPa) MOE = apparent modulus of elasticity, psi (kPa), ∆𝑃  ∆𝑦  = slope of the straight line portion of the load- deflection curve. lbf/in. (N/mm),  P max = maximum load, lbf (N),  Figure 3.5: Apparatus of center-point loading flexural test.  46  For screw withdrawal testing, the screws were supplied by Pro-Fasten Inc., measuring 0.138±0.0003 in. in root diameter and 1 inch in length, with a thread pitch of 16 threads per inch. A 2/3 inch deep lead-hole was drilled into the center of the specimen’s edge surface using an 1/8-inch drill bit. Samples were tested the same day that the screws were inserted. The testing machine (Sintech 30/D Test System) was assembled for SWR testing as shown in Figure 3.6. The specimen was fitted into the holder, and tensile force applied to the screw at a rate of 0.06 in./min and the maximum force in N was recorded by the system. 8 specimens were tested per board for a total of 24 per treatment.  Figure 3.6: The assembly for edge screw withdrawal test  For the IB tests, specimens were first glued to aluminum blocks on both surfaces using hot melt glue and the samples then placed in the test jig in the load-frame (Figure 3.7). A tension load was applied perpendicular to the specimen surface at a uniform rate of 0.05 in/min until failure occurred. The maximum load for every specimen was recorded and then divided by the sample’s cross section area 2” by 2” (which is 50.8mm by 50.8mm). IB test values for treatments were the 47  average of 14 specimens per board or 42 samples per treatment. If failure of the IB specimen occurred at the adhesive layer between the block and sample, the record was discarded  Figure 3.7: The assembly of internal bonding (IB) test  For Thickness Swell/Water Absorption (TS/WA) tests, the mass and four midway thicknesses of each specimen were measured prior immersion in water. Samples were submerged horizontally into the 20˚C water for 24h, and held at a distance of 1inch below the water level (Figure 3.8), leaving sufficient space for sample swelling. After 24h continual submersion, samples were removed from the water and drained for 5 min to remove the residual water on the surface, and the weight and thicknesses at the same four points were measured immediately. Two specimens were tested per board for a total of six per treatment.  48  Figure 3.8: Thickness swell samples in the tank.  Thickness swell (TS) and water absorption (WA) are calculated following the ASTM standards 1037-06a, as follows  𝐖𝐀(%) =  �𝐦𝐭 –𝐦𝟎 � 𝐦𝐨  × 𝟏𝟎𝟎  (3. 3)  where m t is the mass of the sample after immersion (g) and m 0 is the mass of the sample before immersion. 𝐓𝐒(%) =  �𝛅𝐭 –𝛅𝟎 � 𝛅𝐨  × 𝟏𝟎𝟎  (3. 4)  where δ t and δ 0 are the sample thicknesses (mm) after and before the water immersion, R  R  respectively.  49  4 Results and discussion 4.1 XRD analysis X-Ray diffraction is useful for evaluating the degree of clay dispersion in polymer matrices (Ray and Okamoto, 2003). The result of an XRD test is a pattern of X-ray intensity vs the diffraction angle. The distance between the clay layers (d-spacing) can be determined from Bragg’s law if the diffraction angle which corresponds to the intensity peak is known. The mean d-spacings are listed in Table 4.1. Typical XRD patterns of pure nanoclay and the nanoclay in the cured resin matrix are given in Figure 4.1 to Figure 4. 8. Due to the difficulty of computing a mean distribution, the replicate response closest to the mean is shown in the figure; to identify those figures where this was done, those figure captions begin with the word “typical”. The intensity peaks of Cloisite30B disappeared after being dispersed into the UF resin by mechanical mixing. With the addition of a coupling agent, there is also no intensity peak for Cloisite30B. At higher loading (6%) of Cloisite30B, the intensity peak also disappeared, indicating its exfoliation into the UF resin (Figure 4.1). MF resin containing Cloisite30B had no intensity peak for any of the three clay loading levels. The d-space of Nanofil116 in UF resin increased (Figure 4.3, Figure 4.4), which appears in the pattern as the intensity peak of Nanofil116 shifting to a lower 2θ, indicating that the unmodified Nanofil116 were all intercalated with UF resin. The intensity and location of the peak for the clay mix was not changed by clay loading. The intensity peak for Nanofil116-MF mix also appeared at a lower 2θ (Figure 4.7). Adding coupling agent, the intensity peak appeared almost at the same 2θ (Figure 4.8). Higher clay concentration resulted in a higher intensity peak. MF resin was also able to enter the interlayer space of Naonofill116. 50  120000 Cloisite30B UF+2%Cloisite30B UF+4%Cloisite30B UF+6%Cloisite30B  Intensity(Counts)  100000 80000 60000 40000 20000 0 3  4  5  6  8  9  2θ(˚)  10  11  12  13  15  Figure 4.1: Typical XRD patterns of Cloisite30B, Cloisite30B + UF resin. 120000 Cloisite30B UF+2%Cloisite30B+coupling agent UF+4%Cloisite30B+coupling agent UF+6%Cloisite30B+coupling agent  Intensity(Counts)  100000 80000 60000 40000 20000 0 3  4  5  6  8  9  10  11  12  13  15  2θ(˚)  Figure 4.2: Typical XRD patterns of Cloisite30B, Cloisite30B and coupling agent + UF resin.  51  16000 Nanofill116 UF+2%Nanofill116 UF+4%Nanofill116 UF+6%Nanofill116  14000 12000  Intensity(conts)  10000 8000 6000 4000 2000 0 3  4  5  6  8  9 2θ(˚)  10  11  12  13  15  Figure 4.3: Typical XRD patterns of Nanofil116 and Nanofil116 + UF resin. 14000  Nanofill116 UF+2%Nanofill116+Coupling agent UF+4%Nanofill116+Coupling agent UF+6%Nanofill116+Coupling agent  12000  Intensity(conts)  10000 8000 6000 4000 2000 0 3  4  5  6  8  9 10 2θ(˚)  11  12  13  15  Figure 4.4: Typical XRD patterns of Nanofil116, Nanofil116 and coupling agent + UF resin.  52  140000 Cloisite30B  120000  Intensity(Counts)  MF+2%Cloisite30B MF+4%Cloisite30B  100000  MF+6%Cloisite30B  80000 60000 40000 20000 0 3  4  5  6  8  9  2θ(˚)  10  11  12  13  15  Figure 4.5: Typical XRD patterns of Cloisite30B, Cloisite30B + MF resin. 120000  Cloisite30B MF+2%Cloisite30B+coupling agent  100000  MF+4%Cloisite30B+coupling agent  Intensity(Counts)  MF+6%Cloisite30B+coupling agent 80000  60000  40000  20000  0 3  4  5  6  8  9  2θ(˚)  10  11  12  13  15  Figure 4.6: Typical XRD patterns of Cloisite30B, Cloisite30B and coupling agent +MF resin. 53  14000  Nanofill116 MF+2%Nanofill116  12000  MF+4%Nanofill116 MF+6%Nanofill116  Intensity(conts)  10000 8000 6000 4000 2000 0 3  4  5  6  8  9 2θ(˚)  10  11  12  13  15  Figure 4.7: Typical XRD patterns of Nanofil116, Nanofil116 + MF resin. 18000  Nanofill116 MF+2%Nanofill116+Coupling agent  16000  MF+4%Nanofill116+Coupling agent  14000  MF+6%Nanofill116+Coupling agent  Intensity(conts)  12000 10000 8000 6000 4000 2000 0 3  4  5  6  8  9 2θ(˚)  10  11  12  13  15  Figure 4.8: Typical XRD patterns of Nanofil116, Nanofil116 and coupling agent + MF resin. 54  The coupling agent has one end compatible with the polymer and the other end reacts/interacts or glues better to the filler; it is supposed to have similar effects as the organic modifier, helping extend distance between the clay platelets (Han et al., 2008). The addition of the coupling agent did not show significant effects on the XRD patterns, i.e. the separation of clay platelets. The intensity peak positions were almost at the same 2θ for Nanofil116 + resin and the Nanofil116 + resin + coupling agent. The coupling agent did not aid the separation of unmodified clay either, ie the intensity peak did not shift to a lower 2θ with the addition of coupling agent.  The d-spacing of Nanofil116 was larger in UF resin than that in MF, and the intensity peak in of Nanofil116 + MF mixes was lower than that of the Nanofil116 + UF. This may be because the monomers of UF resin are smaller than those of MF resin and can therefore more easily enter the interstitial spaces between the clay platelets resulting in enhanced separation. The XRD results showed that the mechanical mixing method was able to exfoliate the organic-modified nanoclay Cloisite30B into both UF and MF resin, and enlarged the interlayer spacing of unmodified clay when it was dispersed into UF and MF resin.  55  Table 4.1: 2 θ and d-space values from XRD patterns (Some of the samples had no intensity peak and this is denoted by the ‘np’ entry in the table). Clay or clay added resins Cloisite30B UF+ 2%Cloisite30B UF+ 4%Cloisite30B UF+ 6%Cloisite30B UF+ 4%Cloisite30B+Coupling agent UF+ 6%Cloisite30B+Coupling agent UF+ 6%Cloisite30B+Coupling agent MF+ 2%Cloisite30B MF+ 4%Cloisite30B MF+ 6%Cloisite30B MF+ 4%Cloisite30B+Coupling agent MF+ 6%Cloisite30B+Coupling agent MF+ 2%Cloisite30B+Coupling agent Nanofil116 UF+ 2% Nanofil116 UF+ 4% Nanofil116 UF+ 6% Nanofil116 UF+ 2% Nanofil116+Coupling agent UF+ 4% Nanofil116+Coupling agent UF+ 6% Nanofil116+Coupling agent MF+ 2% Nanofil116 MF+ 4% Nanofil116 MF+ 6% Nanofil116 MF+ 2% Nanofil116+Coupling agent MF+ 4% Nanofil116+Coupling agent MF+ 6% Nanofil116+Coupling agent  2θ( ˚ ) 6.226 np np np np np np np np np np np np 8.155 5.844 5.799 5.833 5.869 5.886 5.702 6.076 6.101 6.276 6.152 6.243 6.188  d-space(A) 16.663 np np np np np np np np np np np np 12.579 17.546 17.682 17.579 17.489 17.421 17.983 16.877 16.809 16.342 16.668 16.425 16.342  56  4.2 DSC analysis of the curing process The curing of a thermosetting resin such as UF and MF is an irreversible exothermic reaction which shows an exothermic peak in the DSC curve. The enthalpy of transition (ΔH) based on peak area and the peak temperature (T peak ) is the point at which the reaction rate is the fastest (Ton-That, et al., 2004). In this work, all of the liquid resins and resin-clay mixes were heated from B-stage monomers and cured in an amorphous state. Glass transition was observed in all of the curing processes for all resins and was not affected by nanoclay addition.  The mean test values were listed in Table 4.2. In order to get a representative comparison of each treatment, the response curves closest to the mean value were used. To signify this to readers, the word “Typical” has been placed at the beginning of each figure caption. Adding Cloisite30B to UF resin had little effect on ΔH and onset temperature compared with pure UF. Adding nanoclay means that the peak temperature was slightly delayed, and with a higher curing temperature (Figure 4.9 to Figure 4.12). Similar results were obtained when coupling agent and Cloiste30B applied together.  The onset temperature and the peak temperature of UF resins increased also when Nanofil116 was added, which means the presence of clay delayed the curing reaction. As the clay loading increased, the area under the curing peak decreased (Table 4.2). ΔH decreased significantly as the clay content increased, and this effect was further exacerbated when coupling agent was also added. These suggest that Nanofil116 negatively affected curing and decreased the crosslinking density of UF resin network.  57  ––––––– –––– ––––– · ––– – –  0.2  UF_Feb 27 2012.005 UF+2-30B_Feb 27 2012.002 UF+4-30B_Feb 27 2012.002 UF+6-30B_Feb 27 2012.002  142.01°C  Heat Flow (W/g)  143.40°C  0.0 143.25°C 104.48°C 43.97J/g 139.21°C  104.31°C 41.48J/g  -0.2  104.49°C 44.35J/g  105.32°C 45.10J/g  -0.4  20  40  60  80  Exo Up  100  120  140  160  180  200  Universal V4.7A TA Instruments  Temperature (°C)  Figure 4.9: Typical heat flow curves of different UF resins with different loading of Closiste30B.  0.2  ––––––– –––– ––––– · ––– – –  UF_Feb 27 2012.005 UF+2-CA30B_Feb 28 2012.002 UF+4-CA30B_Feb 28 2012.002 UF+6-CA30B_Feb 29 2012.002  144.79°C  0.0  144.37°C  Heat Flow (W/g)  142.90°C 105.02°C 106.01°C35.22J/g 35.82J/g  -0.2  139.21°C  105.90°C 40.44J/g  105.32°C 45.10J/g  -0.4  -0.6 Exo Up  20  40  60  80  100  120  Temperature (°C)  140  160  180  200  Universal V4.7A TA Instruments  Figure 4.10: Typical heat flow curves of different UF resins with different loading of Closiste30B and coupling agent. 58  ––––––– –––– ––––– · ––– – –  0.2  UF_Feb 27 2012.005 UF+6-116_Feb 28 2012.002 UF+2-116_Feb 27 2012.002 UF+4-116_Feb 28 2012.002  148.90°C  Heat Flow (W/g)  148.80°C  0.0 149.25°C  139.21°C  115.87°C 31.24J/g  -0.2 118.33°C 35.33J/g  113.32°C 35.05J/g  105.32°C 45.10J/g  -0.4  20 Exo Up  40  60  80  100  120  Temperature (°C)  140  160  180  200  Universal V4.7A TA Instruments  Figure 4.11: Typical heat flow curves of different UF resins with different loading of Nanofil116.  Figure 4.12: Typical heat flow curves of different UF resins with different loading of Nanofil116 and coupling agent.  59  For the MF resin (Figures 4. 13 to 4. 16) the onset temperature of pure MF resin is around 125 ˚C and curing peak temperature is around 147˚C. In contrast to the UF resin, adding Cloisite30B at different loadings did not change the ΔH, or the onset and peak temperatures. However using a coupling agent as well had a deleterious effect on the curing of the clay-resin mixes. The ΔH of the MF resin containing Cloisite30B and coupling agent was decreased compared to the pure MF, which means less reaction heat had been generated by this mix. Nanofil116 decreased the ΔH of MF resin and delayed the curing temperature. Increasing clay concentration led to a lower ΔH and higher peak temperature, and the effect was further exacerbated by the addition of coupling agent.  The organic modifier of Cloisite30B is hydrophilic and likely to interact with the resin monomers (Giannelis, 1996). This interaction may not be as strong as the bonding of the resin so it slightly decreased the curing. While the unmodified clay Nanofil116 found it difficult to interact with the resin and probably blocked the connection of the resin network. The use of coupling agent may further extent the gaps between original resin chain connection therefore it further reduced the curing reaction heat (Table 4. 2) (The Tukey analysis of the treatments are listed in Appendix A pages 87-94).  In summary, low loading of Cloisite30B 2% 30B did not have a strong effect on the curing reactions of UF resin and MF resin. Higher concentrations of Cloisiste30B and the presence of coupling agent adversely affected the curing of both resin types. The Nanofil116 significantly reduced the reaction heat and delayed the curing reaction of both resin types, and the addition of coupling agent further compounded this effect.  60  0.2  ––––––– –––– ––––– · ––– – –  MF_Feb 13 2012.003 MF+2-30B_Feb 13 2012.001 MF+6-30B_Feb 13 2012.001 MF+4-30B_Feb 13 2012.002  0.0 145.83°C  Heat Flow (W/g)  145.15°C 146.71°C 147.61°C  -0.2 119.89°C 41.21J/g 118.14°C 40.59J/g 121.85°C 37.29J/g  -0.4 126.21°C 43.37J/g  -0.6 Exo Up  20  40  60  80  100  120  Temperature (°C)  140  160  180  200  Universal V4.7A TA Instruments  Figure 4.13: Typical heat flow curves of different MF resins with different loading of Closiste30B.  Figure 4.14: Typical heat flow curves of different MF resins with different loading of Closiste30B and coupling agent.  61  ––––––– –––– ––––– · ––– – –  0.2  MF_Feb 13 2012.003 MF+2-116_Feb 10 2012.002 MF+4-116_Feb 13 2012.002 MF+6-116_Feb 14 2012.002  158.91°C 159.26°C  Heat Flow (W/g)  0.0 154.39°C  147.61°C 136.21°C 32.97J/g 137.46°C 32.89J/g  -0.2  129.31°C 38.88J/g  -0.4 126.21°C 43.37J/g  20  40  60  80  Exo Up  100  120  140  160  180  200  Universal V4.7A TA Instruments  Temperature (°C)  Figure 4.15: Typical heat flow curves of different MF resins with different loading of Nanofil116.  ––––––– –––– ––––– · ––– – –  0.2  MF_Feb 13 2012.003 MF+2-CA116_Feb 14 2012.002 MF+4-CA116_Feb 14 2012.001 MF+6-CA116_Feb 15 2012.003  157.83°C 159.23°C  Heat Flow (W/g)  0.0  154.06°C  134.52°C147.61°C 30.77J/g 136.83°C 27.05J/g  -0.2  128.79°C 40.62J/g  -0.4 126.21°C 43.37J/g  20 Exo Up  40  60  80  100  120  Temperature (°C)  140  160  180  200  Universal V4.7A TA Instruments  Figure 4.16: Typical heat flow curves of different MF resins with different loading of Nanofil116 and coupling agent. 62  Table 4.2: Mean values of 3 measurements for T onse t , ΔH, T peak ΔH(J/g)  UF UF+2%30B UF+4%30B UF+6%30B UF+2%30B Coupling agent UF+4%30B Coupling agent UF+6%30B Coupling agent UF+2%116 UF+4%116 UF+6%116 UF+2%116 Coupling agent UF+4%116 Coupling agent UF+6%116 Coupling agent  Onset Temperature T onset (˚C) 105.28 104.91 104.84 104.21 105.73 106.03 106.33 114.06 116.07 115.13 112.68 115.18 119.44  43.82 42.75 41.11 42.17 41.06 38.99 38.57 36.29 34.68 33.83 28.72 28.12 19.28  Peak Temperature T peak (˚C) 139.49 142.43 142.58 144.32 142.49 144.43 144.13 150.40 148.27 147.70 150.73 154.99 154.43  MF MF+2%30B MF+4%30B MF+6%30B MF+2%30B Coupling agent MF+4%30B Coupling agent MF+6%30B Coupling agent MF+2%116 MF+4%116 MF+6%116 MF+2%116 Coupling agent MF+4%116 Coupling agent MF+6%116 Coupling agent  124.15 119.19 121.10 118.31 119.52 118.95 119.38 130.53 134.27 137.92 130.05 137.17 134.82  42.46 38.47 37.22 40.07 34.86 39.91 33.18 36.57 35.75 32.39 37.55 30.90 26.43  146.91 121.10 145.22 147.20 148.74 148.44 147.75 154.35 157.51 159.61 153.62 159.23 158.07  Resins  63  4.3 DMA Test Usually, DMA tests are used to detect the glass transition temperature of thermoplastics and their composites. However, the glass transition temperature (T g ) of thermosetting resins depends on the degree of curing of the resin (Louis et al., 2010, Hon 2003). Therefore, in this study, T g is not used as the basis for quantifying and comparing the resins. Instead, the DMA test was used to study the curing process and compare the final rigidity of thermosetting resins and clay mixtures. Also, the results presented in the figures are typical curves from replicates for representative comparison. The storage modulus increased as temperature increased, and the difference between the maximum and minimum E' (ΔE') represents the rigidity of the resin network (He and Riedl, 2003; Park and Kim, 2008). The minimum point is the gel point of the resin, which marks the onset of curing. For UF resin, E’ decreased at first as the temperature increased, because of the softening of the pre cured UF resin. Once the storage modulus reaches the minimum point, it increases again before reaching its maximum value as the polymerization reaction continues, forming a cross-linked molecular network (Kim et al., 2006). The storage modulus increased during the curing process and reached the maximum value when the curing reaction was complete. The UF resins containing Nanoclay and coupling agent had a higher ΔE' than the control resin alone and nanoclay + UF resin mixtures (Figure 4.17), indicating that their rigidity was higher than that of pure UF resins. The storage modulus of Nanofil116 + UF resin mix was higher than Cloisite30B + UF, and this might be because the Nanofil116 intercalated with UF and Closisite30B exfoliated in UF. The intercalated structure resulted in a stiffer UF resin than the exfoliated structure did. ΔE' of pure UF resin was lower than the other resin - clay mixes, which means it was less rigid.  64  500  ––––––– –––– ––––– · ––– – – ––– –––  UF+Cloisite30B UF+Nanofil116+Coupling agent UF+Cloisite30B+Coupling agent UF+Nanofil116 UF  Storage Modulus (MPa)  400  300  200  100  0  40  60  80  100  120  140  160  200  ––––––– –– –– – ––– ––– –––– ––––– ·  180  200  Universal V4.7A TA Instruments  Temperature (°C)  UF+Closisite30B UF+Nanofill116 UF+Nanofill116+Coupling agent UF UF+Closisite30B+Coupling agent  Loss Modulus (MPa)  150  100  50  0  40  60  80  100  120  Temperature (°C)  140  160  180  200  Universal V4.7A TA Instruments  Figure 4.17: Typical storage modulus and loss modulus of UF resin and clay added UF resins. 65  1000000  ––––––– –––– –– –– – ––– ––– ––––– ·  MF+Closite30B MF+Nanofill116+coupling agent MF+Nanofill116 MF MF+Closite30B+coupling agent  Storage Modulus (MPa)  800000  600000  400000  200000  0  40  60  80  100  120  140  160  250000  ––––––– –––– –– –– – ––– ––– ––––– ·  180  200  Universal V4.7A TA Instruments  Temperature (°C)  MF+Closiste30B MF+Nanofill116+coupling agent MF+Nanofill116 MF MF+Closiste30B+coupling agent  Loss Modulus (MPa)  200000  150000  100000  50000  0  40  60  80  100  120  Temperature (°C)  140  160  180  200  Universal V4.7A TA Instruments  Figure 4.18: Typical storage modulus and loss modulus of MF resin and clay added MF resins.  66  For the MF resin, the storage modulus did not show a sharp increase until 160˚C, which also indicates curing of resin and increasing rigidity. Storage modulus of clay + MF resin mixes increased at a higher temperature of 170˚C which also indicates a delay of curing. In contrast to the UF resin, the storage modulus of pure MF resin was higher than all of the clay+ MF mixtures, which indicates rigidity and lower crosslinking density in resin containing clay. 30B+MF mixes had a lower ΔE' than the 116 + MF mixes, further indicating that the exfoliated organic clay results in more gaps and a more loose resin network connection than unmodified clay did. 30B+coupling agent had the lowest E’ and it is likely that the coupling agent further exacerbated the extent of gaps between the resin network and further reduced cross-linking.  The storage modulus (E’) of MF resin was much higher than that of UF resin due to the higher crosslink density and greater rigidity of the MF structure, as mentioned earlier. It might also be because the original resin network of urea is lower than the ring structure of melamine (Kim, Sumin 2006). The density and stiffness of pure MF resin is among the highest of the thermoset resins, and several attempts have been made to improve the chain flexibility to provide a more flexible structure and reduce material cost. (Dolye M, et al., 2003). Our results show that the addition of clay and coupling agent can slightly increase the rigidity of UF resin, but have the potential to decrease the high crosslinking density of pure MF resin, improving its flexibility.  67  4.4 Lap-shear test - automatic bonding evaluation system (ABES) approach The average lap-shear strengths for each resin and resin-clay mix are shown in Figure 4.19 and Figure 4.20. The average shear strength of glue lines bonded with nanoclay/resin mixtures were slightly lower than those of pure UF resin. Bond strength was further reduced with the addition of coupling agent. However, the treatments were not significantly different from the pure resin controls at p=0.05 (refer to Appendix B for Tukey analysis, page:103-105). For MF resin, most of the treatments had higher average shear strengths than pure MF resin except for the Cloisite30B treated MF resin. However these effects were not significant. The only significant difference at p=0.05 was between MF + Cloisite30B (low) and MF+116 with or without coupling agent. Adding coupling agent seems to have improved the bond strength of MF containing Cloisite30B.  UF resin has a loose resin network, adding nanoclays further reduced the connection of resin, therefore the shear strength decreased. The clay addition reduced the high crosslink MF resin network and likely acted as buffer for its brittle facture. That may be the reason for the slight improvement in shear strength. The lap-shear tests suggest that adding nanoclay to either UF or MF resin will not have a detrimental effect on its ability to bond wood elements. In the case of UF resin it may not even have a beneficial effect. The clay and MF resin treatments (MF+ 30B + coupling agent, MF+ 116, MF+ 30B + coupling agent) that yielded higher average shear strength values have the potential to improve particle board properties.  68  5 Shear Strength(Mpa)  Mean±95%CI, N=8 4 3 2 1 0 UF  UF+116  UF+116+CA Treatment  UF+30B  UF+30B+CA  Figure 4.19: Shear strength of UF and clay added UF resins.  6 Mean±95%CI, N=8 Shear strength(MPa)  5 4 3 2 1 0 MF  MF+30B  MF+30B+CA Treatment  MF+116  MF+116+CA  Figure 4.20: Shear strength of MF and clay added MF resins.  69  4.5 Particle board properties test results Particleboard strength properties tests include MOR/MOE (Figure 4.21 and Figure 4.22), Edge SWR (Figure 4.23), Internal Bond (Figure 4.24), and Thickness Swell/Water Absorption (Figure 4.25). In the graphs, M0 denotes the control group, i.e. boards bonded with pure MF resin. 2bc, 4bc, 6bc, are the nanoclay treatments with 2%, 4%, 6% Colisite30B with coupling agent. “f” stands for Nanofil116, and “fc” denotes Nanofil116 with coupling agent. Mean values are shown for the specified number of test specimens (N) in the upper right corner of the graphs, and the error bars indicate the 95% confident interval (CI). All results were analyzed using the TukeKramer pairwise means comparison test in SAS (See Appendix C pages:106-113), at a 5% significance level.  4.5.1 MOR/MOE test As shown in Figure 4.21 and Figure 4.22, the variance within a group was large such that there were no significant different between treatments (Appendix C pages:106-107). The addition of Colisite30B with coupling agent or Nanofil116 only did not change the bending strength and elastic modulus greatly. In the treatments with Nanofil116 and coupling agent, higher clay loading decreased MOR and lower MOE. 6% Nanofil116 with coupling agent resulted in lower MOR than the control. When clay loading was higher than 4% clay, the average MOE was also lower than control. Therefore, the effect of nanclay on PB’s bending strength was minor or even negative.  70  12.0 Mean±95%CI, N=6  MOR(Mpa)  10.0 8.0 6.0 4.0 2.0 0.0 m0  2bc  4bc  6bc  2f 4f Treatment  6f  2fc  4fc  6fc  Figure 4.21: Average modulus of rupture for particleboards bonded with different MF resin + clay mixes.  3.0  Mean±95%CI, N=6  MOE(Gpa)  2.5 2.0 1.5 1.0 0.5 0.0 m0  2bc  4bc  6bc  2f 4f Treatment  6f  2fc  4fc  6fc  Figure 4.22: Average modulus of elasticity for particleboards bonded with different MF resin + clay mixes.  71  4.5.2 Edge screw withdrawal (SWR) test SWR (Figure 4. 23), was largely unaffected by resin-clay mixes, except for the Nanofil116 with coupling agent (i.e. the fc). 2% Nanofil116 with coupling agent resulted in a higher average SWR than the control (Appendix C page:108), and subsequently reduced significantly at higher concentrations of clay. A trend was observed for all treatments with only Nanofil116 where higher clay loading decreased SWR values, but not significantly. The edge SWR were said related to bond strength(Semple and Smith, 2005), so it is likely that the result of these two tests were similar.  1400 Mean±95%CI, N=12 1200  SWR(N)  1000 800 600 400 200 0 m0  2bc  4bc  6bc  2f 4f Treatment  6f  2fc  4fc  6fc  Figure 4.23: Test values of the screw withdrawal test  72  4.5.3 Internal bonding (IB) test The results for IB strength were similar to those for SWR. Most resin clay mixes increased IB slightly, some to a significant extent, e.g. 2% Cloisite30B with coupling agent and 2% nanofil116 with coupling agent (Appendix C pages:109-111) This result is consistent with the lap shear strength that 2%Cloisite30B and Nanofil116+coupling agent increased the average bond strength. There was also a trend that the IB value declined as the clay loading increased, and the decrease was significant for 6% Nanofil116 with coupling agent. During testing failure in some of the samples occurred in the bond line between the test specimen and the metal block, the test values for these samples were not taken into consideration during data analysis.  1.4  Mean±95%CI  IB Strength(Mpa)  1.2 1 0.8 0.6 0.4 0.2 0 m0  2bc  4bc  6bc  2f  4f  6f  2fc  4fc  6fc  Treatment  Figure 4.24: Internal bond strength values of different treatments  73  4.5.4 Thickness swelling test Most of clay-resin mixes produced a slight improvement in the water resistance properties of particleboard, particularly TS (Figure4.25). However none of the treatments were significantly different (Appendix C pages:112-113). For the three treatments, the average thickness swelling rate tended to decrease as the clay loading increased but not significant different. This was similar with previous studies on adding nanoclay to MDF and solid wood( Ashori and Nourbakhsh, 2009; Cai et al., 2010) . The decrease in the thickness swelling rate might be because the clay filled some of the voids in thecomposite and the lumens preventing the penetration of water (Ashori and Nourbakhsh, 2009). In the end, neither the clay type nor the clay loading had a strong impact on thickness swell and water absorption rate, that is, the dimensional stability of the particleboard was not greatly improved by the addition of nanoclay.  0.12  Mean±95%CI, N=6  Increase rate(%)  0.1 0.08 0.06  Thickness swelling  0.04  Water absortion  0.02 0 mo  2bc  4bc  6bc  2fc 4fc Treatment  6fc  2f  4f  6f  Figure 4.25: Water absorption and thickness swelling rate by different treatments  74  Over all, the addition of nanoclay and coupling agent to MF did not produce a dramatic improvement in particle board properties, in fact, strength properties (MOR, MOE, SWR and IB) decreased at higher clay loading. These results are consistent with those of the lap shear test conducted on the resin, when 2% clay loading was applied the mean strength was increased but not significantly so. When higher clay amounts were applied the mechanical properties decreased, this can also be predict by the DSC results where less ΔH was generated at higher clay loading, indicating that less effective bonding was generated between wood and the resin, resin and resin.  Colisite30B with coupling agent had no effect on the bending strength and the screw withdrawal properties, but was able to slightly improve the water resistance properties. Interestingly adding up to 6% of this clay did not significantly reduce board properties, suggesting that higher amount of this clay can be used to replace the resin in particle board production in terms of resin cost.  Nanofil116 also had no significant effect on board properties, except when used at higher concentrations, whereby strength properties (MOR, MOE, SWR and IB) were reduced. Nano116+MF with or without coupling agent also reduced thickness swelling to a small extent; however the change was not statistically significant.  MF resin containing 2% Nanofil116+coupling agent improved some of the mechanical properties, including SWR and IB strength however properties decreased as clay loading increased. When 6% clay was used, most mechanical properties were significantly lower than the controls bonded with pure MF resin. Nanofil116 with coupling agent, regardless of concentration, was able to decrease the average thickness swell rate of particleboard however which was still not significant in statistic. 75  In conclusion, the addition of nanoclays (with or without coupling agent) mostly had no statistically significant effects on the mechanical properties of the particleboard, with the exception of 2% Nanofil116 and coupling agent which resulted in a significant improvement in SWR and 2% Colisite30B-coupling agent treatment significantly improved internal bond strength. This suggests that adding very small amounts of nanoclays to resin can potentially improve mechanical strength. However, increasing clay addition (especially Nanofil116 with coupling agent) had a deleterious effect on board properties.  4.6 Conclusion and general discussion This work demonstrated the possibility of using nanoclay as fillers and reinforcement for two common thermosetting resins used in the manufacture of particleboard. Mechanical mixing using a high speed stirrer was shown to be effective at dispersing nanoclays into liquid UF and MF resin. Based on evidence from XRD, organic-modified nanoclay is believed to have been exfoliated (i.e. complete separation and even distribution of platelets) in UF resin and MF resin. The unmodified nanoclay is believed to have been intercalated (i.e. opening up the gaps between platelets) within the resin matrix.  The addition of nanoclays to UF did not change the heat of reaction (peak curing temperature and ΔH) in curing to any large extent whereas the enthalpy of MF resin-clay mixes was reduced compared with controls, suggesting that less chemical bonds were generated when nanoclays were added.  The curing process of both UF resin and MF resin were slightly delayed by the presence of modified nanoclay (30B) and the heat of curing reaction was decreased. The addition of coupling agent into the clay-resin mix further compounded this effect. Adding unmodified nanoclay (116) 76  significantly delayed the curing reaction of both resins and decreased the heat of reaction in curing, and again the addition of coupling agent further exacerbated this effect.  Adding nanoclays to UF resin had no, or a slightly reducing, effect on its capacity to bond wood elements. On the other hand the bonding strength of wood veneers cured with MF resin containing the organic nanoclay 30B and coupling agent, or 116 with or without coupling agent was improved but the effect was not statistically significant. Nevertheless MF resin was selected as the binder for testing the effect of nanoclay and coupling agent addition on the fabrication of particleboard.  Adding nanoclays and coupling agents mostly had no significant effects on the strength properties of particleboard, although adding small quantities (2%) appears to have some beneficial effect on board properties (SWR and IB). This is also consistent with the result of the shear strength. That may be because small amounts of clay filler in MF resin can share the facture stress and delay the cracking (Adam, 2001). Using higher concentrations of nanoclay, board properties were decreased, particularly in the case of Nanofil116 with coupling agent. This may be explained by the DSC results that higher clay content results in less effective curing reaction therefore the resin is weaker in bonding.  In summary, the nanoclay fillers did not significantly enhance the wood adhesive strength or the particle board properties. On the other hand, 2% nanoclay addition in wood adhesive did not significantly compromise the resin properties and particleboard properties, this method can still be applied in actual particleboard production to reduce cost. Greater amount of clay can be added into resin to lower the production cost as long as the board properties are maintain.  77  5 Comments and future work In this work the effects of adding nanoclays and coupling agents to UF and MF wood adhesives was assessed. At low clay loading (2%) there appears to be some positive effects on MF resin bond strength, however since board properties deteriorate with higher clay loadings less clay filled MF resin can be used for making particle board to verify whether lower amount of this resin can maintain the board properties. Even though the amount of nanoclay that could be supplemented into the resin is restricted to a low content, nanoclay fillers are able to replace some of the resin in particle board and reduce production cost.  Fortunately, simple mechanical mixing can effectively disperse the nanoclay in the resin matrix. Other mixing methods can also be applied to disperse nanoclay into resins. If not limited by the cost and time, the ultrasonic dispersion method is also possible to get ideal clay distribution in resin matrix (Lin, et al., 2005). However, in the actual application of particleboard production in plants, the mixing instrument should sufficiently disperse the clay in greater amount of resin at a lower cost and minimum time.  Visual techniques for characterizing materials including TEM (Transmitting Electron Microscopy) and SEM (Scanning Electron Microscopy) can be used to investigate the clay dispersion and the cracking process of resins containing nanoclay. A useful further investigation would be to conduct blended furnish open time experiments to determine whether the presence of nanoclays can inhibit the penetration of resin into the wood element subsurface, which results in starvation of the glue line and reduced inter particle bond strength. Another useful investigation would be to test the air and water permeability of particle board bonded with resinnanoclay mixtures to further investigate the effects on void space, connectivity and ease of water 78  ingress. This would help explain why particleboard bonded with resin-clay mixes is more resistant to thickness swelling than the controls.  Due to confidentiality agreement with the resin supply company, there is no chemical analysis available for the commercial UF and MF resins used in this study. It would be desirable to undertake further tests on the chemical interaction between nanoclay and resin, using techniques such as Fourier transform spectroscopy (FTIR) and Nuclear magnetic resonance (NMR) to examine whether any extra or stronger chemical bonding takes place between the clay platelets and the cured resin polymer that may help further explain the reinforcement mechanism of nanoclays within the resin matrix.  79  References ASTM. (2008). Standard Test Method for Plastics : Dynamic Mechanical Properties : Cure Behavior 1. Annual Book of ASTM Standards. doi:10.1520/D4473-08.2 Adam S. Zerda, A. J. L. (2001). Intercalated clay nanocomposites: Morphology, mechanics, and fracture behavior. Journal of Polymer Science Part B: Polymer Physics, 39(11), 1137-1146. Alexandre, M., & Dubois, P. (2000). Polymer-layered silicate nanocomposites: preparation, properties and uses of a new class of materials. Materials Science and Engineering: R: Reports, 28(1-2), 1-63. doi:10.1016/S0927-796X(00)00012-7 Ashori, A., & Nourbakhsh, A. (2009). Effects of Nanoclay as a Reinforcement Filler on the Physical and Mechanical Properties of Wood-based Composite. Journal of Composite Materials, 43(18), 1869-1875. doi:10.1177/0021998309340936 Becker, O., Cheng, Y.-B., Varley, R. J., & Simon, G. P. (2003). Layered Silicate Nanocomposites Based on Various High-Functionality Epoxy Resins: The Influence of Cure Temperature on Morphology, Mechanical Properties, and Free Volume. Macromolecules, 36(5), 1616-1625. doi:10.1021/ma0213448 Bragg, W. H., & Bragg, W. L. (1913). The Reflection of X-rays by Crystals. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 88(605), 428-438. doi:10.1098/rspa.1913.0040 Cai, X., Riedl, B., Wan, H., Zhang, S. Y., & Wang, X.-M. (2010). A study on the curing and viscoelastic characteristics of melamine–urea–formaldehyde resin in the presence of aluminium silicate nanoclays. Composites Part A: Applied Science and Manufacturing, 41(5), 604-611. doi:DOI: 10.1016/j.compositesa.2010.01.007 Cai, X., Riedl, B., Zhang, S. Y., & Wan, H. (2007). Effects of nanofillers on water resistance and dimensional stability of solid wood modified by melamine-urea-formaldehyde resin. Wood and Fiber Science, 39(2), 307-318. CRB, Departement des Sciences du Bois et de la Foret, Universite Laval, Sainte-Foy, QC G1K 7P4, Canada: Allen Press Inc. Cai, X., Riedl, B., Zhang, S. Y., & Wan, H. (2008). The impact of the nature of nanofillers on the performance of wood polymer nanocomposites. Composites Part A: Applied Science and Manufacturing, 39(5), 727-737. doi:10.1016/j.compositesa.2008.02.004 Chen, L., Wong, S.-C., & Pisharath, S. (2003). Fracture properties of nanoclay-filled polypropylene. Journal of Applied Polymer Science, 88(14), 3298-3305. Wiley Subscription Services, Inc., A Wiley Company. Retrieved from http://dx.doi.org/10.1002/app.12153 Chowdhury, F. H., Hosur, M. V., & Jeelani, S. (2006). Studies on the flexural and thermomechanical properties of woven carbon/nanoclay-epoxy laminates. Internal stress 80  and thermo-mechanical behavior in multi-component materials systems, TMS Annual Meeting, 2004, 421(1-2), 298-306. doi:DOI: 10.1016/j.msea.2006.01.074 Dean, K., Krstina, J., Tian, W., & Varley, R. J. (2007). Effect of Ultrasonic Dispersion Methods on Thermal and Mechanical Properties of Organoclay Epoxy Nanocomposites. Macromolecular Materials and Engineering, 292(4), 415-427. doi:10.1002/mame.200600435 Deka, B. K., & Maji, T. K. (2010). Effect of coupling agent and nanoclay on properties of HDPE, LDPE, PP, PVC blend and Phargamites karka nanocomposite. Composites Science and Technology, 70(12), 1755-1761. Elsevier Ltd. doi:10.1016/j.compscitech.2010.07.010 Douglas Clark. (2011). Forestry Product annual market review 2010-2011. Review Literature And Arts Of The Americas (pp. 1-153). Geneva. Doyle, M., Hagstrand, P.-0., & Manson, J.-A. E. (2003). Influence of chemical composition on the rheological behavior of condensation reaction resins. Polymer Engineering & Science, 43(2), 297-305. doi:10.1002/pen.10025 Edwards, K. (2004). Handbook of polymer blends and composites (4 Volumes). Materials (Vol. 25, pp. 263-264). iSmithers Rapra Publishing. doi:10.1016/j.matdes.2003.09.012 Faruk, O., & Matuana, L. (2008). Nanoclay reinforced HDPE as a matrix for wood-plastic composites. Composites Science and Technology, 68(9), 2073-2077. doi:10.1016/j.compscitech.2008.03.004 Giannelis, E. P., Krishnamoorti, R., & Manias, E. (1999). Polymer-Silicate Nanocomposites : Model Systems for Confined Polymers and Polymer Brushes. Advances in Polymer Science, 138, 106-147. Groenewoud, W. M. (2001). Dynamic mechanical analysis. Characterisation of Polymers by Thermal Analysis (pp. 94-122). Amsterdam: Elsevier Science B.V. doi:DOI: 10.1016/B978044450604-7/50005-4 Han, M. S., Kim, Y. H., Han, S. J., Choi, S. J., Kim, S. B., & Kim, W. N. (2008). Effects of a Silane Coupling Agent on the Exfoliation of Organoclay Layers in Polyurethane / Organoclay Nanocomposite Foams ? Polymer, 110, 376-386. doi:10.1002/app Hatakeyama, T., & Quinn, F. X. (1999). Thermal Analysis Fundamentals and Applications to Polymer Science. (T. Hatakeyama & F.X. Quinn, Eds.)Recherche (2nd ed., p. 131). Toko,Japan: John wiley& sons. Hetzer, M., & Dekee, D. (2008). Wood/polymer/nanoclay composites, environmentally friendly sustainable technology: A review. Chemical Engineering Research and Design, 86(10), 1083-1093. doi:10.1016/j.cherd.2008.05.003  81  Hon, D. . (2003). Analysis of Adhesives. In K. L. Pizzi, Antonio; Mittal (Ed.), Hand book of wood Adheisve technology (2nd ed.). M. Dekker. Hussain, F., Chen, J., & Hojjati, M. (2007). Epoxy-silicate nanocomposites: Cure monitoring and characterization. Materials Science and Engineering: A, 445-446, 467-476. doi:10.1016/j.msea.2006.09.071 Kim, J.-tae, Lee, D.-yeoul, Oh, T.-su, & Lee, D.-ho. (2003). Characteristics of Nitrile – Butadiene Rubber Layered Silicate Nanocomposites with Silane Coupling Agent. Journal of Applied Polymer Science, 89, 2633–2640. Kim, M. G., Nieh, W. L. S., & Meacham, R. M. (1991). Study on the curing of phenolformaldehyde resol resins by dynamic mechanical analysis. Industrial & Engineering Chemistry Research, 30(4), 798-803. doi:10.1021/ie00052a027 Kord, B., Hemmasi, A. H., & Ghasemi, I. (2010). Properties of PP/wood flour/organomodified montmorillonite nanocomposites. Wood Science and Technology, 45(1), 111-119. doi:10.1007/s00226-010-0309-7 Kotsilkova, R., & Pissis, P. (2007). Thermoset nanocomposites for engineering applications. Shawbury: Smithers Rapra. Kulshreshtha,A. K, & Vasile, C. (2002). Handbook of polymer blends and composites,. (C. V. A. K. Kulshreshtha, Ed.). Rapra Technonogy Limited. Labidi, S., Azema, N., Perrin, D., & Lopez-Cuesta, J.-M. (2010). Organo-modified montmorillonite/poly(ɛ-caprolactone) nanocomposites prepared by melt intercalation in a twin-screw extruder. Polymer Degradation and Stability, 95(3), 382-388. doi:10.1016/j.polymdegradstab.2009.11.013 Lan, T., & Pinnavaia, T. J. (1994). Clay-Reinforced Epoxy Nanocomposites. Chemistry of Materials, 6(12), 2216-2219. American Chemical Society. Retrieved from http://dx.doi.org/10.1021/cm00048a006 Landry, V., Riedl, B., & Blanchet, P. (2008). Nanoclay dispersion effects on UV coatings curing. Progress in Organic Coatings, 62(4), 400-408. doi:10.1016/j.porgcoat.2008.02.010 Lapique, F. (2002). Curing effects on viscosity and mechanical properties of a commercial epoxy resin adhesive. International Journal of Adhesion and Adhesives, 22(4), 337-346. doi:10.1016/S0143-7496(02)00013-1 Lebaron, P. C., Wang, Z., & Pinnavaia, T. J. (1999). Polymer-layered silicate nanocomposites : an overview. Applied Clay Science, 15, 11-29. Lee, J., Jung, D., Hong, C., Rhee, K., & Advani, S. (2005). Properties of polyethylene-layered silicate nanocomposites prepared by melt intercalation with a PP-g-MA compatibilizer. 82  Composites Science and Technology, 65(13), 1996-2002. doi:10.1016/j.compscitech.2005.03.015 Lee, Y. H. (2008). Foaming of Wood Flour / Polyolefin / Layered Silicate Composites by. Microscopy. University of Toronto. Lei, H., Du, G., Pizzi, A., & Celzard, A. (2008). Influence of Nanoclay on Urea-Formaldehyde Resins for Wood Adhesives and Its Model. Journal of Applied Polymer Science, 109, 24422451. doi:10.1002/app Lei, Y., Wu, Q., Clemons, C. M., Yao, F., & Xu, Y. (2007). Influence of Nanoclay on Properties of HDPE / Wood Composites. Journal of Applied Polymer Science, 106, 3958–3966. doi:10.1002/app Lin, Qiaojia Liu, Jinghong Rao, Jiuping Yang, G. (2005). Study on the property of nanoSiO2/urea formaldehyde resin. Scientia silvae sinicae (China); Linye Kexue (China),, 41(2), 129-135. Lv, W., & Zhao, G. (2004). Design of wood/montmorillonite (MMT) intercalation nanocomposite. Forestry Studies in China, 6(1), 54-62. Retrieved from http://dx.doi.org/10.1007/s11632-004-0010-8 Lü, W., & Zhao, G. (2004). Design of wood/montmorillonite (MMT) intercalation nanocomposite. Forestry Studies in China, 6(1), 54-62. doi:10.1007/s11632-004-0010-8 Lü, W.-hua, Zhao, G.-jie, & Xue, Z.-hua. (2006). Preparation and characterization of wood/montmorillonite nanocomposites. Forestry Studies in China, 8(1), 35-40. doi:10.1007/s11632-006-0007-6 Maiti, P., Nam, P. H., & Okamoto, M. (2002). Influence of Crystallization on Intercalation , Morphology , and Mechanical Properties of Polypropylene / Clay Nanocomposites. Marcomolecules, 36(6), 2042-2049. Menczel, J. D., & Prime, R. B. (2008). Polymers Thermal Analysis of Fundamentals and Applications. (J. D. MENCZEL & R. B. PRIME, Eds.). Hoboken, New Jersey: A JOHN WILEY & SONS, INC., PUBLICATION. Mequanint, K., & Sanderson, R. (2003). Nano-structure phosphorus-containing polyurethane dispersions: synthesis and crosslinking with melamine formaldehyde resin. Polymer, 44(9), 2631-2639. doi:10.1016/S0032-3861(03)00154-X Miyagawa, H., Rich, M. J., & Drzal, L. T. (2004). Amine-cured epoxy/clay nanocomposites. II. The effect of the nanoclay aspect ratio. Journal of Polymer Science Part B: Polymer Physics, 42(23), 4391-4400. doi:10.1002/polb.20289  83  Morgan, A. B., & Gilman, J. W. (2002). Characterization of Polymer-Layered Silicate ( Clay ) Nanocomposites by Transmission Electron Microscopy and X-Ray Diffraction : A Comparative Study. Inorganic Materials, 87, 1329-1338. Nishizawa, Y., Furukawa, T., Teruo, G., & Hirotsugu, O. (1982). Process for Producing Filler for Adhesive for Bonding Wood. Nourbakhsh, Amir, & Ashori, A. (2009). Influence of Nanoclay and Coupling Agent on the Physical and Mechanical Properties of Polypropylene / Bagasse Nanocomposite. Polymer. doi:10.1002/app Park, B.-dae, & Kim, J.-woo. (2008). Dynamic Mechanical Analysis of Urea – Formaldehyde Resin Adhesives with Different Formaldehyde-to-Urea Molar Ratios. Polymer. doi:10.1002/app Pavlidou, S., & Papaspyrides, C. D. (2008). A reviewon polymer–layered silicate nanocomposites. Progress in Polymer Science, 33(12), 1119-1198. doi:10.1016/j.progpolymsci.2008.07.008 Pegoretti, A. (2007). Tensile mechanical response of polyethylene – clay nanocomposites. eXPRESS Polymer Letters, 1(3), 123-131. doi:10.3144/expresspolymlett.2007.21 Pepke, E. (2010). FOREST PRODUCTS ANNUAL MARKET REVIEW 2009-2010. UNECE. Perrin-Sarazin, F., Ton-That, M.-T., Bureau, M. N., & Denault, J. (2005). Micro- and nanostructure in polypropylene/clay nanocomposites. Polymer, 46(25), 11624-11634. doi:DOI: 10.1016/j.polymer.2005.09.076 Pilato David Nagy, Ellen, L. V. (2010). Phenolic Resins: A Century of Progress Analyses/Testing (pp. 93-135). Pizzi, A, & Mittal, K. L. (1994). Advanced wood adhesives technology (1st ed., p. 289). New York, USA: Marcel Dekker, Inc. RISI. (2011). Particleboard and MDF Commentary. Production. Retrieved from www.risi.com Sarrazin, P., Blake, D., Feldman, S., Chipera, S., Vaniman, D., & Bish, D. (2005). Field deployment of a portable X-ray diffraction/X-ray flourescence instrument on Mars analog terrain. Powder Diffraction, 20(2), 128. doi:10.1154/1.1913719 Semple, K. E., & Smith, G. D. (2005). PREDICTION OF INTERNAL BOND STRENGTH IN PARTICLEBOARD FROM SCREW WITHDRAWAL RESISTANCE MODELS. Wood Science and Technology, 38(2), 256-267.  84  Sheshmani, S., Ashori, A., & Hamzeh, Y. (2010). Physical Properties of Polyethylene – Wood Fiber – Clay Nanocomposites. Journal ofAppliedPolymer Science, 118, 3255–3259. doi:10.1002/app Shi, Y., Qiu, J., & Zheng, Z. (2004). Mineralogy Characters of Yunnan Wollastonite and Application of M ineral Filler in Wood-based Panel Industry. China Forestry product inductry, 31(4), 7-10. China Forest Product Industry. Sichina, W. J., & Manager, I. M. (2000). Prediction of Epoxy Cure Properties Using Pyris DSC Scanning Kinetics Software. Manager. Norwalk. Siddiqui, N. A., Woo, R. S. C., Kim, J.-K., Leung, C. C. K., & Munir, A. (2007). Mode I interlaminar fracture behavior and mechanical properties of CFRPs with nanoclay-filled epoxy matrix. Composites Part A: Applied Science and Manufacturing, 38(2), 449-460. doi:DOI: 10.1016/j.compositesa.2006.03.001 Sinha Ray, S., & Okamoto, M. (2003). Polymer/layered silicate nanocomposites: a review from preparation to processing. Progress in Polymer Science, 28(11), 1539-1641. doi:10.1016/j.progpolymsci.2003.08.002 Sui, G., Zhong, W.-H., Fuqua, M. a., & Ulven, C. a. (2007). Crystalline Structure and Properties of Carbon Nanofiber Composites Prepared by Melt Extrusion. Macromolecular Chemistry and Physics, 208(17), 1928-1936. doi:10.1002/macp.200700170 Ton-That, M.-T., Ngo, T.-D., Ding, P., Fang, G., Cole, K. C., & Hoa, S. V. (2004). Epoxy nanocomposites: Analysis and kinetics of cure. Polymer Engineering and Science, 44(6), 1132-1141. doi:10.1002/pen.20106 Vaia, R. a., & Liu, W. (2002). X-ray powder diffraction of polymer/layered silicate nanocomposites: Model and practice. Journal of Polymer Science Part B: Polymer Physics, 40(15), 1590-1600. doi:10.1002/polb.10214 Vermogen, A., Masenelli-Varlot, K., Séguéla, R., Duchet-Rumeau, J., Boucard, S., & Prele, P. (2005). Evaluation of the Structure and Dispersion in Polymer-Layered Silicate Nanocomposites. Macromolecules, 38(23), 9661-9669. doi:10.1021/ma051249+ Wang, Ke, Chen, L., Wu, J., Toh, M. L., He, C., & Yee, A. F. (2005). Epoxy Nanocomposites with Highly Exfoliated Clay : Mechanical Properties and Fracture Mechanisms. Macromolecules, 38(3), 788-800. Wang, S., Qiu, H., Zhou, J., & Wellwood, R. (2008). Phyllosilicate Modifided Resins for Lignocellulousic Fiber Based Composite Panels. Wang, Z.-yu, Han, E.-hou, & Ke, W. (2007). Fire-resistant effect of nanoclay on intumescent nanocomposite coatings. Journal of Applied Polymer Science, 103(3), 1681-1689. Wiley  85  Subscription Services, Inc., A Wiley Company. Retrieved from http://dx.doi.org/10.1002/app.25096 Wong, D. C. (2008). Particleboard performance requirements of secondary wood products manufacturers in Canada. Forestry Productd Journal, 58(3), 34-41. Yeh, S.-K. (2007). POLYPROPYLENE-BASED WOOD-PLASTIC COMPOSITES REINFORCED WITH NANOCLAY. West Virginia University. Young No, B., & Kim, M. G. (2005). Curing of low level melamine-modified urea-formaldehyde particleboard binder resins studied with dynamic mechanical analysis (DMA). Journal of Applied Polymer Science, 97(1), 377-389. doi:10.1002/app.21759 Zerda, A. S., & Lesser, A. J. (2001). Intercalated Clay Nanocomposites : Morphology , Mechanics , and Fracture Behavior. Journal of Polymer Science Part B: Polymer Physics, 39, 1137-1146. Zhang, J., Hereid, J., Hagen, M., Bakirtzis, D., Delichatsios, M. A., Fina, A., Castrovinci, A., et al. (2009). Effects of nanoclay and fire retardants on fire retardancy of a polymer blend of EVA and LDPE. Fire Safety Journal, 44(4), 504-513. doi:10.1016/j.firesaf.2008.10.005 Zhao, Y., Wang, K., Zhu, F., Xue, P., & Jia, M. (2006). Properties of poly(vinyl chloride)/wood flour/montmorillonite composites: Effects of coupling agents and layered silicate. Polymer Degradation and Stability, 91(12), 2874-2883. doi:10.1016/j.polymdegradstab.2006.09.001  86  Appendix A: Statistical analysis result of DSC test This appendix is the test result of the Tukey’s test, grouping the means of board properties tests. The means of which are significantly different will be grouped by different letters or mark. UF resin DSC test Oneway Analysis of ΔH By TREATMENT 50 45  HEAT  40 35 30 25 20 152B  2BC 2F  2FC 4B  4BC 4F  4FC 6B  6BC 6F  6FC UF  TREATMENT  All Pairs Tukey-Kramer 0.05  Excluded Rows 3  Means Comparisons Comparisons for all pairs using Tukey-Kramer HSD q* 3.63404 Abs(Dif)HSD UF 2B 6B 4B 2BC 4BC 6BC 2F 4F 6F 2FC 4FC 6FC  Alpha 0.05  UF  2B  6B  4B  2BC  4BC  6BC  2F  4F  6F  2FC  4FC  6FC  -7.663 -6.593 -6.010 -4.950 -4.900 -2.833 -2.410 -0.133 1.484 2.324 7.437 8.034 16.884  -6.593 -7.663 -7.080 -6.020 -5.970 -3.903 -3.480 -1.203 0.414 1.254 6.367 6.964 15.814  -6.010 -7.080 -7.663 -6.603 -6.553 -4.486 -4.063 -1.786 -0.170 0.670 5.784 6.380 15.230  -4.950 -6.020 -6.603 -7.663 -7.613 -5.546 -5.123 -2.846 -1.230 -0.390 4.724 5.320 14.170  -4.900 -5.970 -6.553 -7.613 -7.663 -5.596 -5.173 -2.896 -1.280 -0.440 4.674 5.270 14.120  -2.833 -3.903 -4.486 -5.546 -5.596 -7.663 -7.240 -4.963 -3.346 -2.506 2.607 3.204 12.054  -2.410 -3.480 -4.063 -5.123 -5.173 -7.240 -7.663 -5.386 -3.770 -2.930 2.184 2.780 11.630  -0.133 -1.203 -1.786 -2.846 -2.896 -4.963 -5.386 -7.663 -6.046 -5.206 -0.093 0.504 9.354  1.484 0.414 -0.170 -1.230 -1.280 -3.346 -3.770 -6.046 -7.663 -6.823 -1.710 -1.113 7.737  2.324 1.254 0.670 -0.390 -0.440 -2.506 -2.930 -5.206 -6.823 -7.663 -2.550 -1.953 6.897  7.437 6.367 5.784 4.724 4.674 2.607 2.184 -0.093 -1.710 -2.550 -7.663 -7.066 1.784  8.034 6.964 6.380 5.320 5.270 3.204 2.780 0.504 -1.113 -1.953 -7.066 -7.663 1.187  16.884 15.814 15.230 14.170 14.120 12.054 11.630 9.354 7.737 6.897 1.784 1.187 -7.663  Positive values show pairs of means that are significantly different.  Level  Mean  87  Level UF 2B 6B 4B 2BC 4BC 6BC 2F 4F 6F 2FC 4FC 6FC  A A A A A A A A  B B B B B B B  C C C C C C C  D D D D  Mean 43.823333 42.753333 42.170000 41.110000 41.060000 38.993333 38.570000 36.293333 34.676667 33.836667 28.723333 28.126667 19.276667  E E E E F  Levels not connected by same letter are significantly different. Level UF 2B 6B 4B 2BC 4BC 6BC 2F UF 4F UF 2B 6F 6B 2B 6B 4B 2BC 4B 2BC 4BC 6BC 4BC UF 6BC 2FC UF 2B 4FC 6B 2F 2B 2F UF 6B 4B 2BC 4F 2B 4B 2BC 4F 6B  Level 6FC 6FC 6FC 6FC 6FC 6FC 6FC 6FC 4FC 6FC 2FC 4FC 6FC 4FC 2FC 2FC 4FC 4FC 2FC 2FC 4FC 4FC 2FC 6F 2FC 6FC 4F 6F 6FC 6F 4FC 4F 2FC 2F 4F 6F 6F 4FC 2F 4F 4F 2FC 2F  Difference Std Err Dif Lower CL 24.54667 23.47667 22.89333 21.83333 21.78333 19.71667 19.29333 17.01667 15.69667 15.40000 15.10000 14.62667 14.56000 14.04333 14.03000 13.44667 12.98333 12.93333 12.38667 12.33667 10.86667 10.44333 10.27000 9.98667 9.84667 9.44667 9.14667 8.91667 8.85000 8.33333 8.16667 8.07667 7.57000 7.53000 7.49333 7.27333 7.22333 6.55000 6.46000 6.43333 6.38333 5.95333 5.87667  2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642  16.8838 15.8138 15.2304 14.1704 14.1204 12.0538 11.6304 9.3538 8.0338 7.7371 7.4371 6.9638 6.8971 6.3804 6.3671 5.7838 5.3204 5.2704 4.7238 4.6738 3.2038 2.7804 2.6071 2.3238 2.1838 1.7838 1.4838 1.2538 1.1871 0.6704 0.5038 0.4138 -0.0929 -0.1329 -0.1696 -0.3896 -0.4396 -1.1129 -1.2029 -1.2296 -1.2796 -1.7096 -1.7862  Upper CL 32.20956 31.13956 30.55622 29.49622 29.44622 27.37956 26.95622 24.67956 23.35956 23.06289 22.76289 22.28956 22.22289 21.70622 21.69289 21.10956 20.64622 20.59622 20.04956 19.99956 18.52956 18.10622 17.93289 17.64956 17.50956 17.10956 16.80956 16.57956 16.51289 15.99622 15.82956 15.73956 15.23289 15.19289 15.15622 14.93622 14.88622 14.21289 14.12289 14.09622 14.04622 13.61622 13.53956  p-Value Difference <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* 0.0001* 0.0001* 0.0002* 0.0002* 0.0013* 0.0021* 0.0026* 0.0036* 0.0043* 0.0069* 0.0097* 0.0126* 0.0136* 0.0243* 0.0292* 0.0322* 0.0551 0.0574 0.0596 0.0746 0.0785 0.1496 0.1622 0.1662 0.1737 0.2501 0.2659  88  Level 6F UF 4BC 6F UF 4B 2BC 6BC 4BC 2B 6BC 2B 6B 6B UF UF 4BC 4B 2BC 2F 6BC 4B 2BC 2B UF 2B 2F 6B UF 6B 4F 2FC 2B 4BC 4B  Level 4FC 6BC 6F 2FC 4BC 2F 2F 6F 4F 6BC 4F 4BC 6BC 4BC 2BC 4B 2F 6BC 6BC 6F 2F 4BC 4BC 2BC 6B 4B 4F 2BC 2B 4B 6F 4FC 6B 6BC 2BC  Difference Std Err Dif Lower CL 5.71000 5.25333 5.15667 5.11333 4.83000 4.81667 4.76667 4.73333 4.31667 4.18333 3.89333 3.76000 3.60000 3.17667 2.76333 2.71333 2.70000 2.54000 2.49000 2.45667 2.27667 2.11667 2.06667 1.69333 1.65333 1.64333 1.61667 1.11000 1.07000 1.06000 0.84000 0.59667 0.58333 0.42333 0.05000  2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642 2.108642  Upper CL  -1.9529 -2.4096 -2.5062 -2.5496 -2.8329 -2.8462 -2.8962 -2.9296 -3.3462 -3.4796 -3.7696 -3.9029 -4.0629 -4.4862 -4.8996 -4.9496 -4.9629 -5.1229 -5.1729 -5.2062 -5.3862 -5.5462 -5.5962 -5.9696 -6.0096 -6.0196 -6.0462 -6.5529 -6.5929 -6.6029 -6.8229 -7.0662 -7.0796 -7.2396 -7.6129  13.37289 12.91622 12.81956 12.77622 12.49289 12.47956 12.42956 12.39622 11.97956 11.84622 11.55622 11.42289 11.26289 10.83956 10.42622 10.37622 10.36289 10.20289 10.15289 10.11956 9.93956 9.77956 9.72956 9.35622 9.31622 9.30622 9.27956 8.77289 8.73289 8.72289 8.50289 8.25956 8.24622 8.08622 7.71289  p-Value Difference 0.3027 0.4189 0.4459 0.4583 0.5418 0.5458 0.5609 0.5710 0.6956 0.7335 0.8097 0.8409 0.8745 0.9417 0.9784 0.9812 0.9819 0.9889 0.9906 0.9916 0.9956 0.9977 0.9982 0.9997 0.9998 0.9998 0.9998 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 . .  Oneway Analysis of PEAKTEMP By TREATMENT  PEAKTEMP  15  10  50  0 2B  2BC 2F  2FC 4B  4BC 4F  4FC 6B  TREATMENT  6BC 6F  6FC UF  All Pairs Tukey-Kramer 0.05  89  Excluded Rows 3  Means Comparisons Comparisons for all pairs using Tukey-Kramer HSD q* 3.63404 Abs(Dif)HSD 6FC 4FC 2FC 2F 4F 6F 6BC 4BC 2BC 2B UF 6B 4B  Alpha 0.05  6FC  4FC  2FC  2F  4F  6F  6BC  4BC  2BC  2B  UF  6B  4B  77.205 75.768 71.502 71.172 69.045 68.465 65.282 64.902 63.258 63.198 60.265 31.762 21.355  75.768 77.205 72.938 72.608 70.482 69.902 66.718 66.338 64.695 64.635 61.702 33.198 22.792  71.502 72.938 77.205 76.875 74.748 74.168 70.985 70.605 68.962 68.902 65.968 37.465 27.058  71.172 72.608 76.875 77.205 75.078 74.498 71.315 70.935 69.292 69.232 66.298 37.795 27.388  69.045 70.482 74.748 75.078 77.205 76.625 73.442 73.062 71.418 71.358 68.425 39.922 29.515  68.465 69.902 74.168 74.498 76.625 77.205 74.022 73.642 71.998 71.938 69.005 40.502 30.095  65.282 66.718 70.985 71.315 73.442 74.022 77.205 76.825 75.182 75.122 72.188 43.685 33.278  64.902 66.338 70.605 70.935 73.062 73.642 76.825 77.205 75.562 75.502 72.568 44.065 33.658  63.258 64.695 68.962 69.292 71.418 71.998 75.182 75.562 77.205 77.145 74.212 45.708 35.302  63.198 64.635 68.902 69.232 71.358 71.938 75.122 75.502 77.145 77.205 74.272 45.768 35.362  60.265 61.702 65.968 66.298 68.425 69.005 72.188 72.568 74.212 74.272 77.205 48.702 38.295  31.762 33.198 37.465 37.795 39.922 40.502 43.685 44.065 45.708 45.768 48.702 77.205 66.798  21.355 22.792 27.058 27.388 29.515 30.095 33.278 33.658 35.302 35.362 38.295 66.798 77.205  Positive values show pairs of means that are significantly different. Level 6FC 4FC 2FC 2F 4F 6F 6BC 4BC 2BC 2B UF 6B 4B  Mean 156.43333 154.99667 150.73000 150.40000 148.27333 147.69333 144.51000 144.13000 142.48667 142.42667 139.49333 110.99000 100.58333  A A A A A A A A A A A A A  Levels not connected by same letter are significantly different. Level 6FC 4FC 2FC  Level 4B 4B 4B  Difference Std Err Dif Lower CL 55.85000 54.41333 50.14667  21.24498 21.24498 21.24498  -21.3551 -22.7918 -27.0584  Upper CL p-Value Difference 133.0551 131.6184 127.3518  0.3426 0.3791 0.4979  90  Level 2F 4F 6F 6FC 4FC 6BC 4BC 2BC 2B 2FC 2F UF 4F 6F 6BC 4BC 2BC 2B UF 6FC 4FC 6FC 6FC 4FC 4FC 6FC 6FC 2FC 2F 4FC 4FC 6B 4F 6FC 2FC 2FC 6F 6FC 2F 2F 4FC 4FC 2FC 2F 2FC 6FC 2F 4F 4F 6FC 6F 6F 6BC 4BC 4FC 4FC 4F 4F 6F 6F  Level 4B 4B 4B 6B 6B 4B 4B 4B 4B 6B 6B 4B 6B 6B 6B 6B 6B 6B 6B UF UF 2B 2BC 2B 2BC 4BC 6BC UF UF 4BC 6BC 4B UF 6F 2B 2BC UF 4F 2B 2BC 6F 4F 4BC 4BC 6BC 2F 6BC 2B 2BC 2FC 2B 2BC UF UF 2F 2FC 4BC 6BC 4BC 6BC  Difference Std Err Dif Lower CL 49.81667 47.69000 47.11000 45.44333 44.00667 43.92667 43.54667 41.90333 41.84333 39.74000 39.41000 38.91000 37.28333 36.70333 33.52000 33.14000 31.49667 31.43667 28.50333 16.94000 15.50333 14.00667 13.94667 12.57000 12.51000 12.30333 11.92333 11.23667 10.90667 10.86667 10.48667 10.40667 8.78000 8.74000 8.30333 8.24333 8.20000 8.16000 7.97333 7.91333 7.30333 6.72333 6.60000 6.27000 6.22000 6.03333 5.89000 5.84667 5.78667 5.70333 5.26667 5.20667 5.01667 4.63667 4.59667 4.26667 4.14333 3.76333 3.56333 3.18333  21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498  -27.3884 -29.5151 -30.0951 -31.7618 -33.1984 -33.2784 -33.6584 -35.3018 -35.3618 -37.4651 -37.7951 -38.2951 -39.9218 -40.5018 -43.6851 -44.0651 -45.7084 -45.7684 -48.7018 -60.2651 -61.7018 -63.1984 -63.2584 -64.6351 -64.6951 -64.9018 -65.2818 -65.9684 -66.2984 -66.3384 -66.7184 -66.7984 -68.4251 -68.4651 -68.9018 -68.9618 -69.0051 -69.0451 -69.2318 -69.2918 -69.9018 -70.4818 -70.6051 -70.9351 -70.9851 -71.1718 -71.3151 -71.3584 -71.4184 -71.5018 -71.9384 -71.9984 -72.1884 -72.5684 -72.6084 -72.9384 -73.0618 -73.4418 -73.6418 -74.0218  Upper CL p-Value Difference 127.0218 124.8951 124.3151 122.6484 121.2118 121.1318 120.7518 119.1084 119.0484 116.9451 116.6151 116.1151 114.4884 113.9084 110.7251 110.3451 108.7018 108.6418 105.7084 94.1451 92.7084 91.2118 91.1518 89.7751 89.7151 89.5084 89.1284 88.4418 88.1118 88.0718 87.6918 87.6118 85.9851 85.9451 85.5084 85.4484 85.4051 85.3651 85.1784 85.1184 84.5084 83.9284 83.8051 83.4751 83.4251 83.2384 83.0951 83.0518 82.9918 82.9084 82.4718 82.4118 82.2218 81.8418 81.8018 81.4718 81.3484 80.9684 80.7684 80.3884  0.5076 0.5710 0.5884 0.6383 0.6806 0.6830 0.6940 0.7403 0.7419 0.7971 0.8052 0.8173 0.8539 0.8659 0.9215 0.9270 0.9476 0.9483 0.9742 0.9998 0.9999 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 . . . . .  91  Level 2FC 2BC 2B 2F 2FC 2F 6BC 6BC 4BC 4BC 6FC 4F 6BC 2FC 2BC  Level 6F UF UF 6F 4F 4F 2B 2BC 2B 2BC 4FC 6F 4BC 2F 2B  Difference Std Err Dif Lower CL 3.03667 2.99333 2.93333 2.70667 2.45667 2.12667 2.08333 2.02333 1.70333 1.64333 1.43667 0.58000 0.38000 0.33000 0.06000  21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498 21.24498  Upper CL p-Value Difference  -74.1684 -74.2118 -74.2718 -74.4984 -74.7484 -75.0784 -75.1218 -75.1818 -75.5018 -75.5618 -75.7684 -76.6251 -76.8251 -76.8751 -77.1451  80.2418 80.1984 80.1384 79.9118 79.6618 79.3318 79.2884 79.2284 78.9084 78.8484 78.6418 77.7851 77.5851 77.5351 77.2651  . . . . . . . . . . . . . . .  Oneway Analysis of T ON SET By TREATMENT 12  ON SET  12  11  11  10 2B  2BC 2F  2FC 4B  4BC 4F  4FC 6B  6BC 6F  TREATMENT  6FC UF  All Pairs Tukey-Kramer 0.05  Excluded Rows 3  Oneway Anova Analysis of Variance Source TREATMENT Error C. Total  DF 12 26 38  Sum of Squares 1090.2421 125.0456 1215.2877  Mean Square 90.8535 4.8094  F Ratio 18.8906  Prob > F <.0001*  Means for Oneway Anova Level 2B 2BC 2F 2FC 4B 4BC 4F 4FC 6B 6BC 6F  Number 3 3 3 3 3 3 3 3 3 3 3  Mean 104.907 105.733 114.060 112.683 104.843 106.033 116.067 115.180 104.210 105.703 115.130  Std Error 1.2662 1.2662 1.2662 1.2662 1.2662 1.2662 1.2662 1.2662 1.2662 1.2662 1.2662  Lower 95% 102.30 103.13 111.46 110.08 102.24 103.43 113.46 112.58 101.61 103.10 112.53  Upper 95% 107.51 108.34 116.66 115.29 107.45 108.64 118.67 117.78 106.81 108.31 117.73  92  Level 6FC UF  Number 3 3  Mean 119.437 105.277  Std Error 1.2662 1.2662  Lower 95% 116.83 102.67  Upper 95% 122.04 107.88  Std Error uses a pooled estimate of error variance  Means Comparisons Comparisons for all pairs using Tukey-Kramer HSD q* 3.63404  Alpha 0.05 4F  4FC  6F  2F  4BC  3.1372 6.5072 5.6205 5.5705 4.5005 3.1238 6.8962 3.5262  2.2505 5.6205 6.5072 6.4572 5.3872 4.0105 2.6395  2.2005 5.5705 6.4572 6.5072 5.4372 4.0605 2.5895  1.1305 4.5005 5.3872 5.4372 6.5072 5.1305 1.5195  2BC  7.1962 3.8262 2.9395 2.8895 1.8195 0.4428  6BC  7.2262 3.8562 2.9695 2.9195 1.8495 0.4728  UF  7.6528 4.2828 3.3962 3.3462 2.2762 0.8995  2B  8.0228 4.6528 3.7662 3.7162 2.6462 1.2695  4B  8.0862 4.7162 3.8295 3.7795 2.7095 1.3328  6B  8.7195 5.3495 4.4628 4.4128 3.3428 1.9662  Abs(Dif)HSD 6FC 4F 4FC 6F 2F 2FC  6FC 6.5072 3.1372 2.2505 2.2005 1.1305 0.2462  2FC  4BC  2BC  6BC  UF  2B  4B  6B  0.2462 6.8962 7.1962 7.2262 7.6528 8.0228 8.0862 8.7195 3.1238 4.0105 4.0605 5.1305 6.5072 0.1428  3.5262 3.8262 3.8562 4.2828 4.6528 4.7162 5.3495 2.6395 2.9395 2.9695 3.3962 3.7662 3.8295 4.4628 2.5895 2.8895 2.9195 3.3462 3.7162 3.7795 4.4128 1.5195 1.8195 1.8495 2.2762 2.6462 2.7095 3.3428 0.1428 0.4428 0.4728 0.8995 1.2695 1.3328 1.9662 6.5072 6.2072 6.1772 5.7505 5.3805 5.3172 4.6838  6.2072 6.5072 6.4772 6.0505 5.6805 5.6172 4.9838  6.1772 6.4772 6.5072 6.0805 5.7105 5.6472 5.0138  5.7505 6.0505 6.0805 6.5072 6.1372 6.0738 5.4405  5.3805 5.6805 5.7105 6.1372 6.5072 6.4438 5.8105  5.3172 5.6172 5.6472 6.0738 6.4438 6.5072 5.8738  4.6838 4.9838 5.0138 5.4405 5.8105 5.8738 6.5072  Positive values show pairs of means that are significantly different.  Level 6FC 4F 4FC 6F 2F 2FC 4BC 2BC 6BC UF 2B 4B 6B  A A A A A  B B B B B C C C C C C C  Mean 119.43667 116.06667 115.18000 115.13000 114.06000 112.68333 106.03333 105.73333 105.70333 105.27667 104.90667 104.84333 104.21000  Levels not connected by same letter are significantly different. Level  Level  Difference Std Err Dif Lower CL  Upper CL  p-Value Difference  93  Level 6FC 6FC 6FC 6FC 6FC 6FC 6FC 4F 4F 4F 4FC 6F 4F 4F 4FC 4F 6F 4FC 6F 4F 4FC 6F 2F 4FC 4FC 6F 6F 2F 2F 4FC 6F 2F 2FC 2F 2F 2F 2FC 2FC 2FC 2FC 2FC 6FC 2FC 6FC 6FC 6FC 4F 6FC 4FC 6F 4F 4BC 2BC 6BC 2F 4BC 4BC 4FC 6F UF  Level 6B 4B 2B UF 6BC 2BC 4BC 6B 4B 2B 6B 6B UF 6BC 4B 2BC 4B 2B 2B 4BC UF UF 6B 6BC 2BC 6BC 2BC 4B 2B 4BC 4BC UF 6B 6BC 2BC 4BC 4B 2B UF 6BC 2BC 2FC 4BC 2F 6F 4FC 2FC 4F 2FC 2FC 2F 6B 6B 6B 2FC 4B 2B 2F 2F 6B  Difference Std Err Dif Lower CL 15.22667 14.59333 14.53000 14.16000 13.73333 13.70333 13.40333 11.85667 11.22333 11.16000 10.97000 10.92000 10.79000 10.36333 10.33667 10.33333 10.28667 10.27333 10.22333 10.03333 9.90333 9.85333 9.85000 9.47667 9.44667 9.42667 9.39667 9.21667 9.15333 9.14667 9.09667 8.78333 8.47333 8.35667 8.32667 8.02667 7.84000 7.77667 7.40667 6.98000 6.95000 6.75333 6.65000 5.37667 4.30667 4.25667 3.38333 3.37000 2.49667 2.44667 2.00667 1.82333 1.52333 1.49333 1.37667 1.19000 1.12667 1.12000 1.07000 1.06667  1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614  8.71950 8.08617 8.02284 7.65284 7.22617 7.19617 6.89617 5.34950 4.71617 4.65284 4.46284 4.41284 4.28284 3.85617 3.82950 3.82617 3.77950 3.76617 3.71617 3.52617 3.39617 3.34617 3.34284 2.96950 2.93950 2.91950 2.88950 2.70950 2.64617 2.63950 2.58950 2.27617 1.96617 1.84950 1.81950 1.51950 1.33284 1.26950 0.89950 0.47284 0.44284 0.24617 0.14284 -1.13050 -2.20050 -2.25050 -3.12383 -3.13716 -4.01050 -4.06050 -4.50050 -4.68383 -4.98383 -5.01383 -5.13050 -5.31716 -5.38050 -5.38716 -5.43716 -5.44050  Upper CL 21.73383 21.10050 21.03716 20.66716 20.24050 20.21050 19.91050 18.36383 17.73050 17.66716 17.47716 17.42716 17.29716 16.87050 16.84383 16.84050 16.79383 16.78050 16.73050 16.54050 16.41050 16.36050 16.35716 15.98383 15.95383 15.93383 15.90383 15.72383 15.66050 15.65383 15.60383 15.29050 14.98050 14.86383 14.83383 14.53383 14.34716 14.28383 13.91383 13.48716 13.45716 13.26050 13.15716 11.88383 10.81383 10.76383 9.89050 9.87716 9.00383 8.95383 8.51383 8.33050 8.03050 8.00050 7.88383 7.69716 7.63383 7.62716 7.57716 7.57383  p-Value Difference <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* 0.0001* 0.0001* 0.0001* 0.0003* 0.0003* 0.0003* 0.0003* 0.0003* 0.0003* 0.0004* 0.0005* 0.0005* 0.0005* 0.0009* 0.0009* 0.0010* 0.0010* 0.0013* 0.0014* 0.0014* 0.0015* 0.0024* 0.0037* 0.0043* 0.0045* 0.0068* 0.0088* 0.0096* 0.0157* 0.0275* 0.0286* 0.0368* 0.0419* 0.1818 0.4703 0.4875 0.7869 0.7909 0.9659 0.9705 0.9939 0.9974 0.9995 0.9996 0.9998 1.0000 1.0000 1.0000 1.0000 1.0000  94  Level 4F 2BC 4F 6BC 2BC 6BC 4BC 2B 4B 2BC UF 6BC UF 4BC 4BC 2B 4FC 2BC  Level 6F 4B 4FC 4B 2B 2B UF 6B 6B UF 4B UF 2B 6BC 2BC 4B 6F 6BC  Difference Std Err Dif Lower CL 0.93667 0.89000 0.88667 0.86000 0.82667 0.79667 0.75667 0.69667 0.63333 0.45667 0.43333 0.42667 0.37000 0.33000 0.30000 0.06333 0.05000 0.03000  1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614 1.790614  Upper CL  -5.57050 -5.61716 -5.62050 -5.64716 -5.68050 -5.71050 -5.75050 -5.81050 -5.87383 -6.05050 -6.07383 -6.08050 -6.13716 -6.17716 -6.20716 -6.44383 -6.45716 -6.47716  7.44383 7.39716 7.39383 7.36716 7.33383 7.30383 7.26383 7.20383 7.14050 6.96383 6.94050 6.93383 6.87716 6.83716 6.80716 6.57050 6.55716 6.53716  p-Value Difference 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 . . . . . .  MF resin DSC test Oneway Analysis of Peak Temperature By TREATMENT  Peak Temperature  16  15  15  14 2B  2BC 2F  2FC 4B  4BC 4F  4FC 6B  6BC 6F  6FC MF  F9  All Pairs Tukey-Kramer 0.05  Missing Rows 3  Means Comparisons Comparisons for all pairs using Tukey-Kramer HSD q* 3.63404 Abs(Dif)HSD 6F 4FC 6FC 4F 2F  Alpha 0.05  6F  4FC  6FC  4F  2F  2FC  2BC  4BC  6BC  -2.628 -2.245 -1.081 -0.521 2.639  -2.245 -2.628 -1.465 -0.905 2.255  -1.081 -1.465 -2.628 -2.068 1.092  -0.521 -0.905 -2.068 -2.628 0.532  2.639 2.255 1.092 0.532 -2.628  3.362 2.979 1.815 1.255 -1.905  8.245 7.862 6.699 6.139 2.979  8.542 8.159 6.995 6.435 3.275  9.232 8.849 7.685 7.125 3.965  MF  2B  4B  6B  9.789 10.079 10.935 11.769 9.405 9.695 10.552 11.385 8.242 8.532 9.389 10.222 7.682 7.972 8.829 9.662 4.522 4.812 5.669 6.502  95  Abs(Dif)HSD 2FC 2BC 4BC 6BC MF 2B 4B 6B  6F  4FC  6FC  4F  2F  2FC  2BC  4BC  6BC  MF  2B  4B  6B  3.362 2.979 1.815 8.245 7.862 6.699 8.542 8.159 6.995 9.232 8.849 7.685 9.789 9.405 8.242 10.079 9.695 8.532 10.935 10.552 9.389 11.769 11.385 10.222  1.255 6.139 6.435 7.125 7.682 7.972 8.829 9.662  -1.905 2.979 3.275 3.965 4.522 4.812 5.669 6.502  -2.628 2.255 2.552 3.242 3.799 4.089 4.945 5.779  2.255 -2.628 -2.331 -1.641 -1.085 -0.795 0.062 0.895  2.552 -2.331 -2.628 -1.938 -1.381 -1.091 -0.235 0.599  3.242 -1.641 -1.938 -2.628 -2.071 -1.781 -0.925 -0.091  3.799 -1.085 -1.381 -2.071 -2.628 -2.338 -1.481 -0.648  4.089 -0.795 -1.091 -1.781 -2.338 -2.628 -1.771 -0.938  4.945 0.062 -0.235 -0.925 -1.481 -1.771 -2.628 -1.795  5.779 0.895 0.599 -0.091 -0.648 -0.938 -1.795 -2.628  Positive values show pairs of means that are significantly different. Level 6F 4FC 6FC 4F 2F 2FC 2BC 4BC 6BC MF 2B 4B 6B  Mean 159.61333 159.23000 158.06667 157.50667 154.34667 153.62333 148.74000 148.44333 147.75333 147.19667 146.90667 146.05000 145.21667  A A A A B B C C C C C  D D D D D  E E E E E  Levels not connected by same letter are significantly different. Level 6F 4FC 6F 4FC 6FC 6F 6F 4FC 4F 4FC 6FC 6F 4FC 4F 6F 6FC 6F 6FC 4FC 4F 4FC 6FC 4F 4F 6FC 6FC 2F 4F 4F  Level 6B 6B 4B 4B 6B 2B MF 2B 6B MF 4B 6BC 6BC 4B 4BC 2B 2BC MF 4BC 2B 2BC 6BC MF 6BC 4BC 2BC 6B 4BC 2BC  Difference 14.39667 14.01333 13.56333 13.18000 12.85000 12.70667 12.41667 12.32333 12.29000 12.03333 12.01667 11.86000 11.47667 11.45667 11.17000 11.16000 10.87333 10.87000 10.78667 10.60000 10.49000 10.31333 10.31000 9.75333 9.62333 9.32667 9.13000 9.06333 8.76667  Std Err Dif Lower CL 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354  11.7688 11.3854 10.9354 10.5521 10.2221 10.0788 9.7888 9.6954 9.6621 9.4054 9.3888 9.2321 8.8488 8.8288 8.5421 8.5321 8.2454 8.2421 8.1588 7.9721 7.8621 7.6854 7.6821 7.1254 6.9954 6.6988 6.5021 6.4354 6.1388  Upper CL 17.02457 16.64124 16.19124 15.80790 15.47790 15.33457 15.04457 14.95124 14.91790 14.66124 14.64457 14.48790 14.10457 14.08457 13.79790 13.78790 13.50124 13.49790 13.41457 13.22790 13.11790 12.94124 12.93790 12.38124 12.25124 11.95457 11.75790 11.69124 11.39457  p-Value Difference <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001*  96  Level 2FC 2F 2FC 2F 2F 2FC 2F 2FC 6F 2F 2FC 4FC 2F 6F 2FC 4FC 2FC 6FC 4F 6FC 2BC 4BC 4F 2BC 6BC 4BC 6F MF 2BC 4FC 6BC 2B 6F 2BC 4BC 4BC 4FC MF 2BC 2B 6BC 4B 2F 4BC 6FC 6BC 6F 2BC MF  Level 6B 4B 4B 2B MF 2B 6BC MF 2FC 4BC 6BC 2FC 2BC 2F 4BC 2F 2BC 2FC 2FC 2F 6B 6B 2F 4B 6B 4B 4F 6B 2B 4F 4B 6B 6FC MF 2B MF 6FC 4B 6BC 4B 2B 6B 2FC 6BC 4F MF 4FC 4BC 2B  Difference 8.40667 8.29667 7.57333 7.44000 7.15000 6.71667 6.59333 6.42667 5.99000 5.90333 5.87000 5.60667 5.60667 5.26667 5.18000 4.88333 4.88333 4.44333 3.88333 3.72000 3.52333 3.22667 3.16000 2.69000 2.53667 2.39333 2.10667 1.98000 1.83333 1.72333 1.70333 1.69000 1.54667 1.54333 1.53667 1.24667 1.16333 1.14667 0.98667 0.85667 0.84667 0.83333 0.72333 0.69000 0.56000 0.55667 0.38333 0.29667 0.29000  Std Err Dif Lower CL 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354 0.7231354  5.7788 5.6688 4.9454 4.8121 4.5221 4.0888 3.9654 3.7988 3.3621 3.2754 3.2421 2.9788 2.9788 2.6388 2.5521 2.2554 2.2554 1.8154 1.2554 1.0921 0.8954 0.5988 0.5321 0.0621 -0.0912 -0.2346 -0.5212 -0.6479 -0.7946 -0.9046 -0.9246 -0.9379 -1.0812 -1.0846 -1.0912 -1.3812 -1.4646 -1.4812 -1.6412 -1.7712 -1.7812 -1.7946 -1.9046 -1.9379 -2.0679 -2.0712 -2.2446 -2.3312 -2.3379  Upper CL 11.03457 10.92457 10.20124 10.06790 9.77790 9.34457 9.22124 9.05457 8.61790 8.53124 8.49790 8.23457 8.23457 7.89457 7.80790 7.51124 7.51124 7.07124 6.51124 6.34790 6.15124 5.85457 5.78790 5.31790 5.16457 5.02124 4.73457 4.60790 4.46124 4.35124 4.33124 4.31790 4.17457 4.17124 4.16457 3.87457 3.79124 3.77457 3.61457 3.48457 3.47457 3.46124 3.35124 3.31790 3.18790 3.18457 3.01124 2.92457 2.91790  p-Value Difference <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* 0.0001* 0.0007* 0.0013* 0.0026* 0.0072* 0.0090* 0.0413* 0.0658 0.0999 0.2138 0.2883 0.3936 0.4838 0.5010 0.5125 0.6384 0.6413 0.6471 0.8674 0.9115 0.9190 0.9708 0.9903 0.9912 0.9923 0.9978 0.9986 0.9998 0.9998 1.0000 1.0000 1.0000  97  Oneway Analysis of ΔH By F9 45  ΔH  40  35  30  252B  2BC 2F  2FC 4B  4BC 4F  4FC 6B  6BC 6F  6FC MF  F9  All Pairs Tukey-Kramer 0.05  Missing Rows 3  Means Comparisons Comparisons for all pairs using Tukey-Kramer HSD q* 3.63404 Abs(Dif)HSD MF 6B 4BC 2B 2FC 4B 2F 4F 2BC 6BC 6F 4FC 6FC  Alpha 0.05  MF  6B  4BC  2B  2FC  4B  2F  4F  2BC  6BC  6F  4FC  6FC  -5.387 -2.997 -2.840 -1.373 -0.477 -0.150 0.507 1.320 2.213 3.897 4.683 6.170 10.643  -2.997 -5.387 -5.230 -3.763 -2.867 -2.540 -1.883 -1.070 -0.177 1.507 2.293 3.780 8.253  -2.840 -5.230 -5.387 -3.920 -3.023 -2.697 -2.040 -1.227 -0.333 1.350 2.137 3.623 8.097  -1.373 -3.763 -3.920 -5.387 -4.490 -4.163 -3.507 -2.693 -1.800 -0.117 0.670 2.157 6.630  -0.477 -2.867 -3.023 -4.490 -5.387 -5.060 -4.403 -3.590 -2.697 -1.013 -0.227 1.260 5.733  -0.150 -2.540 -2.697 -4.163 -5.060 -5.387 -4.730 -3.917 -3.023 -1.340 -0.553 0.933 5.407  0.507 -1.883 -2.040 -3.507 -4.403 -4.730 -5.387 -4.573 -3.680 -1.997 -1.210 0.277 4.750  1.320 -1.070 -1.227 -2.693 -3.590 -3.917 -4.573 -5.387 -4.493 -2.810 -2.023 -0.537 3.937  2.213 -0.177 -0.333 -1.800 -2.697 -3.023 -3.680 -4.493 -5.387 -3.703 -2.917 -1.430 3.043  3.897 1.507 1.350 -0.117 -1.013 -1.340 -1.997 -2.810 -3.703 -5.387 -4.600 -3.113 1.360  4.683 2.293 2.137 0.670 -0.227 -0.553 -1.210 -2.023 -2.917 -4.600 -5.387 -3.900 0.573  6.170 3.780 3.623 2.157 1.260 0.933 0.277 -0.537 -1.430 -3.113 -3.900 -5.387 -0.913  10.643 8.253 8.097 6.630 5.733 5.407 4.750 3.937 3.043 1.360 0.573 -0.913 -5.387  Positive values show pairs of means that are significantly different.  Level MF 6B 4BC 2B 2FC 4B 2F 4F 2BC 6BC 6F 4FC  A A A A A A  B B B B B B B B  C C C C C C C  D D D D D D D  E E E E E  F  Mean 42.460000 40.070000 39.913333 38.446667 37.550000 37.223333 36.566667 35.753333 34.860000 33.176667 32.390000 30.903333  98  Level 6FC  Mean 26.430000  F  Levels not connected by same letter are significantly different. Level MF 6B 4BC 2B MF 2FC 4B 2F MF 4F MF 6B 4BC 2BC 6B MF 2B 4BC 6B 6BC 4BC MF 2FC 4B 2B 6F MF 2F 2B MF 6B 2FC 4BC MF 4F 4B 4FC 2FC 6B 2F 4BC 4B MF 2BC 2B 6B 2F 4F 4BC 6B 2B 4BC 2FC 4F MF  Level 6FC 6FC 6FC 6FC 4FC 6FC 6FC 6FC 6F 6FC 6BC 4FC 4FC 6FC 6F 2BC 4FC 6F 6BC 6FC 6BC 4F 4FC 4FC 6F 6FC 2F 4FC 6BC 4B 2BC 6F 2BC 2FC 4FC 6F 6FC 6BC 4F 6F 4F 6BC 2B 4FC 2BC 2F 6BC 6F 2F 4B 4F 4B 2BC 6BC 4BC  Difference Std Err Dif Lower CL 16.03000 13.64000 13.48333 12.01667 11.55667 11.12000 10.79333 10.13667 10.07000 9.32333 9.28333 9.16667 9.01000 8.43000 7.68000 7.60000 7.54333 7.52333 6.89333 6.74667 6.73667 6.70667 6.64667 6.32000 6.05667 5.96000 5.89333 5.66333 5.27000 5.23667 5.21000 5.16000 5.05333 4.91000 4.85000 4.83333 4.47333 4.37333 4.31667 4.17667 4.16000 4.04667 4.01333 3.95667 3.58667 3.50333 3.39000 3.36333 3.34667 2.84667 2.69333 2.69000 2.69000 2.57667 2.54667  1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270  10.6434 8.2534 8.0967 6.6300 6.1700 5.7334 5.4067 4.7500 4.6834 3.9367 3.8967 3.7800 3.6234 3.0434 2.2934 2.2134 2.1567 2.1367 1.5067 1.3600 1.3500 1.3200 1.2600 0.9334 0.6700 0.5734 0.5067 0.2767 -0.1166 -0.1500 -0.1766 -0.2266 -0.3333 -0.4766 -0.5366 -0.5533 -0.9133 -1.0133 -1.0700 -1.2100 -1.2266 -1.3400 -1.3733 -1.4300 -1.8000 -1.8833 -1.9966 -2.0233 -2.0400 -2.5400 -2.6933 -2.6966 -2.6966 -2.8100 -2.8400  Upper CL 21.41663 19.02663 18.86996 17.40330 16.94330 16.50663 16.17996 15.52330 15.45663 14.70996 14.66996 14.55330 14.39663 13.81663 13.06663 12.98663 12.92996 12.90996 12.27996 12.13330 12.12330 12.09330 12.03330 11.70663 11.44330 11.34663 11.27996 11.04996 10.65663 10.62330 10.59663 10.54663 10.43996 10.29663 10.23663 10.21996 9.85996 9.75996 9.70330 9.56330 9.54663 9.43330 9.39996 9.34330 8.97330 8.88996 8.77663 8.74996 8.73330 8.23330 8.07996 8.07663 8.07663 7.96330 7.93330  p-Value Difference <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* 0.0001* 0.0003* 0.0012* 0.0014* 0.0015* 0.0016* 0.0045* 0.0058* 0.0059* 0.0062* 0.0068* 0.0116* 0.0177* 0.0207* 0.0230* 0.0329* 0.0594 0.0624 0.0649 0.0697 0.0812 0.0993 0.1078 0.1103 0.1767 0.2000 0.2142 0.2525 0.2573 0.2921 0.3029 0.3218 0.4615 0.4960 0.5441 0.5555 0.5627 0.7699 0.8245 0.8256 0.8256 0.8612 0.8699  99  Level 6B 2BC MF 4BC 4B 6BC 2B 2FC 2F 2BC 6B 6F 4B 4BC 2B 2FC 2B 4F 2F 6BC 4B 2FC 6B  Level 2FC 6F 6B 2FC 2BC 4FC 2F 4F 2BC 6BC 2B 4FC 4F 2B 4B 2F 2FC 2BC 4F 6F 2F 4B 4BC  Difference Std Err Dif Lower CL 2.52000 2.47000 2.39000 2.36333 2.36333 2.27333 1.88000 1.79667 1.70667 1.68333 1.62333 1.48667 1.47000 1.46667 1.22333 0.98333 0.89667 0.89333 0.81333 0.78667 0.65667 0.32667 0.15667  1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270 1.482270  Upper CL  -2.8666 -2.9166 -2.9966 -3.0233 -3.0233 -3.1133 -3.5066 -3.5900 -3.6800 -3.7033 -3.7633 -3.9000 -3.9166 -3.9200 -4.1633 -4.4033 -4.4900 -4.4933 -4.5733 -4.6000 -4.7300 -5.0600 -5.2300  7.90663 7.85663 7.77663 7.74996 7.74996 7.65996 7.26663 7.18330 7.09330 7.06996 7.00996 6.87330 6.85663 6.85330 6.60996 6.36996 6.28330 6.27996 6.19996 6.17330 6.04330 5.71330 5.54330  p-Value Difference 0.8774 0.8908 0.9102 0.9162 0.9162 0.9345 0.9832 0.9883 0.9924 0.9932 0.9950 0.9977 0.9980 0.9980 0.9996 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 .  Oneway Analysis of Onset temperature By F9 14  Onset temperature  13 13 12 12 11 112B  2BC 2F  2FC 4B  4BC 4F  4FC 6B  6BC 6F  6FC MF  F9  All Pairs Tukey-Kramer 0.05  Missing Rows 3  Means Comparisons Comparisons for all pairs using Tukey-Kramer HSD q* 3.63404 Abs(Dif)HSD 6F 4FC 6FC 4F 2F 2FC  Alpha 0.05  6F  4FC  6FC  4F  2F  2FC  MF  -7.531 -6.777 -4.431 -3.881 -0.131 0.347  -6.777 -7.531 -5.184 -4.634 -0.884 -0.407  -4.431 -5.184 -7.531 -6.981 -3.231 -2.753  -3.881 -4.634 -6.981 -7.531 -3.781 -3.303  -0.131 -0.884 -3.231 -3.781 -7.531 -7.053  0.347 -0.407 -2.753 -3.303 -7.053 -7.531  6.246 5.493 3.146 2.596 -1.154 -1.631  4B  2BC  6BC  2B  4BC  6B  9.289 10.873 11.013 11.206 11.443 12.083 8.536 10.119 10.259 10.453 10.689 11.329 6.189 7.773 7.913 8.106 8.343 8.983 5.639 7.223 7.363 7.556 7.793 8.433 1.889 3.473 3.613 3.806 4.043 4.683 1.412 2.995 3.135 3.329 3.565 4.205  100  Abs(Dif)HSD MF 4B 2BC 6BC 2B 4BC 6B  6F  4FC  6FC  4F  2F  2FC  MF  4B  2BC  6BC  2B  4BC  6B  6.246 9.289 10.873 11.013 11.206 11.443 12.083  5.493 8.536 10.119 10.259 10.453 10.689 11.329  3.146 6.189 7.773 7.913 8.106 8.343 8.983  2.596 5.639 7.223 7.363 7.556 7.793 8.433  -1.154 1.889 3.473 3.613 3.806 4.043 4.683  -1.631 1.412 2.995 3.135 3.329 3.565 4.205  -7.531 -4.487 -2.904 -2.764 -2.571 -2.334 -1.694  -4.487 -7.531 -5.947 -5.807 -5.614 -5.377 -4.737  -2.904 -5.947 -7.531 -7.391 -7.197 -6.961 -6.321  -2.764 -5.807 -7.391 -7.531 -7.337 -7.101 -6.461  -2.571 -5.614 -7.197 -7.337 -7.531 -7.294 -6.654  -2.334 -5.377 -6.961 -7.101 -7.294 -7.531 -6.891  -1.694 -4.737 -6.321 -6.461 -6.654 -6.891 -7.531  Positive values show pairs of means that are significantly different.  Level 6F 4FC 6FC 4F 2F 2FC MF 4B 2BC 6BC 2B 4BC 6B  A A A A A  B B B B B  C C C  Mean 137.92333 137.17000 134.82333 134.27333 130.52333 130.04600 124.14667 121.10333 119.52000 119.38000 119.18667 118.95000 118.31000  D D D D D D D  Levels not connected by same letter are significantly different. Level 6F 6F 4FC 6F 6F 6F 4FC 4FC 4FC 4FC 6F 6FC 4FC 4F 6FC 6FC 6FC 4F 6FC 4F 4F 4F 6F 6FC 4F 4FC 2F 2FC 2F 2F  Level 6B 4BC 6B 2B 6BC 2BC 4BC 2B 6BC 2BC 4B 6B 4B 6B 4BC 2B 6BC 4BC 2BC 2B 6BC 2BC MF 4B 4B MF 6B 6B 4BC 2B  Difference Std Err Dif Lower CL 19.61333 18.97333 18.86000 18.73667 18.54333 18.40333 18.22000 17.98333 17.79000 17.65000 16.82000 16.51333 16.06667 15.96333 15.87333 15.63667 15.44333 15.32333 15.30333 15.08667 14.89333 14.75333 13.77667 13.72000 13.17000 13.02333 12.21333 11.73600 11.57333 11.33667  2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272  12.0826 11.4426 11.3293 11.2059 11.0126 10.8726 10.6893 10.4526 10.2593 10.1193 9.2893 8.9826 8.5359 8.4326 8.3426 8.1059 7.9126 7.7926 7.7726 7.5559 7.3626 7.2226 6.2459 6.1893 5.6393 5.4926 4.6826 4.2053 4.0426 3.8059  Upper CL 27.14405 26.50405 26.39072 26.26739 26.07405 25.93405 25.75072 25.51405 25.32072 25.18072 24.35072 24.04405 23.59739 23.49405 23.40405 23.16739 22.97405 22.85405 22.83405 22.61739 22.42405 22.28405 21.30739 21.25072 20.70072 20.55405 19.74405 19.26672 19.10405 18.86739  p-Value Difference <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* 0.0002* 0.0004* 0.0004* 0.0006*  101  Level 2F 2FC 2F 2FC 6FC 2FC 2FC 4F 2F 2FC 6F 6F 4FC 4FC 2F 2FC MF MF MF 6FC MF MF 6FC 4F 4F 6F 6F MF 4FC 4B 4FC 4B 4B 4B 4B 2BC 6BC 2B 6F 4BC 2BC 6FC 2F 6BC 2BC 2B 6BC 2BC  Level 6BC 4BC 2BC 2B MF 6BC 2BC MF 4B 4B 2FC 2F 2FC 2F MF MF 6B 4BC 2B 2FC 6BC 2BC 2F 2FC 2F 4F 6FC 4B 4F 6B 6FC 4BC 2B 6BC 2BC 6B 6B 6B 4FC 6B 4BC 4F 2FC 4BC 2B 4BC 2B 6BC  Difference Std Err Dif Lower CL 11.14333 11.09600 11.00333 10.85933 10.67667 10.66600 10.52600 10.12667 9.42000 8.94267 7.87733 7.40000 7.12400 6.64667 6.37667 5.89933 5.83667 5.19667 4.96000 4.77733 4.76667 4.62667 4.30000 4.22733 3.75000 3.65000 3.10000 3.04333 2.89667 2.79333 2.34667 2.15333 1.91667 1.72333 1.58333 1.21000 1.07000 0.87667 0.75333 0.64000 0.57000 0.55000 0.47733 0.43000 0.33333 0.23667 0.19333 0.14000  2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272 2.072272  3.6126 3.5653 3.4726 3.3286 3.1459 3.1353 2.9953 2.5959 1.8893 1.4119 0.3466 -0.1307 -0.4067 -0.8841 -1.1541 -1.6314 -1.6941 -2.3341 -2.5707 -2.7534 -2.7641 -2.9041 -3.2307 -3.3034 -3.7807 -3.8807 -4.4307 -4.4874 -4.6341 -4.7374 -5.1841 -5.3774 -5.6141 -5.8074 -5.9474 -6.3207 -6.4607 -6.6541 -6.7774 -6.8907 -6.9607 -6.9807 -7.0534 -7.1007 -7.1974 -7.2941 -7.3374 -7.3907  Upper CL 18.67405 18.62672 18.53405 18.39005 18.20739 18.19672 18.05672 17.65739 16.95072 16.47339 15.40805 14.93072 14.65472 14.17739 13.90739 13.43005 13.36739 12.72739 12.49072 12.30805 12.29739 12.15739 11.83072 11.75805 11.28072 11.18072 10.63072 10.57405 10.42739 10.32405 9.87739 9.68405 9.44739 9.25405 9.11405 8.74072 8.60072 8.40739 8.28405 8.17072 8.10072 8.08072 8.00805 7.96072 7.86405 7.76739 7.72405 7.67072  p-Value Difference 0.0007* 0.0008* 0.0009* 0.0010* 0.0013* 0.0013* 0.0015* 0.0025* 0.0058* 0.0102* 0.0344* 0.0574 0.0765 0.1227 0.1581 0.2402 0.2530 0.4094 0.4775 0.5325 0.5357 0.5787 0.6784 0.7000 0.8281 0.8509 0.9444 0.9509 0.9652 0.9733 0.9934 0.9969 0.9989 0.9996 0.9998 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 . . . . .  102  Appendix B: Statistical analysis result of Lap-shear test UF resin Lap shear test Oneway Analysis of UF_resin By Sample name 5  UF+30B+CA  UF  3.5  UF+30B  UF+116+CA  4  UF+116  UF_resin  4.5  All Pairs Tukey-Kramer 0.05  Sample name  Means Comparisons Comparisons for all pairs using Tukey-Kramer HSD q* 2.87506  Alpha 0.05 UF -0.48535 -0.26160 -0.07048 -0.01973 0.03690  Abs(Dif)-HSD UF UF+30B UF+30B+CA UF+116 UF+116+CA  UF+30B -0.26160 -0.48535 -0.29423 -0.24348 -0.18685  UF+30B+CA -0.07048 -0.29423 -0.48535 -0.43460 -0.37798  UF+116 -0.01973 -0.24348 -0.43460 -0.48535 -0.42873  UF+116+CA 0.03690 -0.18685 -0.37798 -0.42873 -0.48535  Positive values show pairs of means that are significantly different.  Level UF UF+30B UF+30B+CA UF+116 UF+116+CA  A A A A  Mean 4.3995000 4.1757500 3.9846250 3.9338750 3.8772500  B B B B  Levels not connected by same letter are significantly different. Level  - Level  Difference  UF UF UF UF+30B UF+30B  UF+116+CA UF+116 UF+30B+CA UF+116+CA UF+116  0.5222500 0.4656250 0.4148750 0.2985000 0.2418750  Std Err Dif 0.1688141 0.1688141 0.1688141 0.1688141 0.1688141  Lower CL 0.036899 -0.019726 -0.070476 -0.186851 -0.243476  Upper CL 1.007601 0.950976 0.900226 0.783851 0.727226  p- Difference Value 0.0297* 0.0653 0.1240 0.4076 0.6113  103  Level  - Level  Difference  UF UF+30B UF+30B+CA UF+116 UF+30B+CA  UF+30B UF+30B+CA UF+116+CA UF+116+CA UF+116  0.2237500 0.1911250 0.1073750 0.0566250 0.0507500  Std Err Dif 0.1688141 0.1688141 0.1688141 0.1688141 0.1688141  Lower CL -0.261601 -0.294226 -0.377976 -0.428726 -0.434601  Upper CL 0.709101 0.676476 0.592726 0.541976 0.536101  p- Difference Value 0.6776 0.7885 0.9681 0.9971 0.9981  MF Resin test Oneway Analysis of MF_resin By Sample name  MF+116+CA  MF+116  4  MF  4.5  MF+30B+CA  5  MF+30B  MF_resin  5.5  All Pairs Tukey-Kramer 0.05  Sample name  Means and Std Deviations Level MF MF+116 MF+116+CA MF+30B MF+30B+CA  Number 8 8 8 8 8  Mean 4.58688 4.93588 4.96900 4.30963 4.74988  Std Dev Std Err Mean 0.391489 0.13841 0.394249 0.13939 0.450525 0.15928 0.195046 0.06896 0.460388 0.16277  Lower 95% 4.2596 4.6063 4.5924 4.1466 4.3650  Upper 95% 4.9142 5.2655 5.3456 4.4727 5.1348  Means Comparisons Comparisons for all pairs using Tukey-Kramer HSD q* 2.87506 Abs(Dif)-HSD MF+116+CA MF+116 MF+30B+CA MF MF+30B  Alpha 0.05 MF+116+CA -0.56107 -0.52794 -0.34194 -0.17894 0.09831  MF+116 -0.52794 -0.56107 -0.37507 -0.21207 0.06518  MF+30B+CA -0.34194 -0.37507 -0.56107 -0.39807 -0.12082  MF -0.17894 -0.21207 -0.39807 -0.56107 -0.28382  MF+30B 0.09831 0.06518 -0.12082 -0.28382 -0.56107  Positive values show pairs of means that are significantly different.  104  Level MF+116+CA MF+116 MF+30B+CA MF MF+30B  A A A A  Mean 4.9690000 4.9358750 4.7498750 4.5868750 4.3096250  B B B  Levels not connected by same letter are significantly different. Level  - Level  Difference  MF+116+CA MF+116 MF+30B+CA MF+116+CA MF+116 MF MF+116+CA MF+116 MF+30B+CA MF+116+CA  MF+30B MF+30B MF+30B MF MF MF+30B MF+30B+CA MF+30B+CA MF MF+116  0.6593750 0.6262500 0.4402500 0.3821250 0.3490000 0.2772500 0.2191250 0.1860000 0.1630000 0.0331250  Std Err Dif 0.1951497 0.1951497 0.1951497 0.1951497 0.1951497 0.1951497 0.1951497 0.1951497 0.1951497 0.1951497  Lower CL 0.098307 0.065182 -0.120818 -0.178943 -0.212068 -0.283818 -0.341943 -0.375068 -0.398068 -0.527943  Upper CL 1.220443 1.187318 1.001318 0.943193 0.910068 0.838318 0.780193 0.747068 0.724068 0.594193  p- Difference Value 0.0145* 0.0223* 0.1836 0.3071 0.3962 0.6189 0.7934 0.8739 0.9178 0.9998  105  Appendix C: Statistical analysis result of PB properties test MOR test The SAS System 7 16:38 Tuesday, March 22, 2011 The GLM Procedure Tukey's Studentized Range (HSD) Test for mor NOTE: This test controls the Type I experimentwise error rate, but it generally has a higher Type II error rate than REGWQ.  Alpha 0.05 Error Degrees of Freedom 50 Error Mean Square 2.988619 Critical Value of Studentized Range 4.68143 Minimum Significant Difference 3.304 Means with the same letter are not significantly different. Tukey Grouping  Mean  N  treatment  A A A A A A A A A A A A A A A A A A A  9.1475  6  4f  8.6597  6  6f  8.3298  6  2fc  8.0822  6  2f  8.0417  6  none  7.9257  6  4bc  7.4419  6  2bc  7.3151  6  6bc  6.7991  6  4fc  5.9708  6  6fc  106  MOE test The SAS System 7 16:38 Tuesday, March 22, 2011 The GLM Procedure Tukey's Studentized Range (HSD) Test for mor NOTE: This test controls the Type I experimentwise error rate, but it generally has a higher Type II error rate than REGWQ.  Alpha 0.05 Error Degrees of Freedom 50 Error Mean Square 2.988619 Critical Value of Studentized Range 4.68143 Minimum Significant Difference 3.304 Means with the same letter are not significantly different. Tukey Grouping  Mean  N  treatment  A A A A A A A A A A A A A A A A A A A  2.2786  6  4f  2.2360  6  6f  2.1952  6  none  2.0882  6  4bc  2.0358  6  2f  2.0214  6  6bc  1.9902  6  2bc  1.9158  6  4bc  1.7922  6  4fc  1.5786  6  6fc  107  SWR test The SAS System  14 11:49 Tuesday, March 22, 2011  The GLM Procedure Tukey's Studentized Range (HSD) Test for SWR NOTE: This test controls the Type I experimentwise error rate, but it generally has a higher Type II error rate than REGWQ.  Alpha 0.05 Error Degrees of Freedom 110 Error Mean Square 24739.44 Critical Value of Studentized Range 4.56737 Minimum Significant Difference 207.38  Means with the same letter are not significantly different.  Tukey Grouping  B B B B B B B B B B B B B B B  A A A A A A A  C C C C C C C C C C C C C  Mean  N  treatment  1154.83  12  2fc  1042.50  12  2f  1029.58  12  4f  962.92  12  6f  944.00  12  4bc  941.33  12  4fc  935.58  12  none  914.33  12  2bc  895.58  12  6bc  767.58  12  6fc  108  IB test Tukey's Studentized Range (HSD) Test for IB NOTE: This test controls the Type I experimentwise error rate. Alpha 0.05 Error Degrees of Freedom 363 Error Mean Square 0.029568 Critical Value of Studentized Range 4.50221 Comparisons significant at the 0.05 level are indicated by ***.  treatment Comparison 2fc 2fc 2fc 2fc 2fc 2fc 2fc 2fc 2fc 2bc 2bc 2bc 2bc 2bc 2bc 2bc 2bc 2bc 4f 4f 4f 4f 4f 4f 4f 4f 4f  -  2bc 4f 2f 4fc 4bc 6bc 6f none 6fc 2fc 4f 2f 4fc 4bc 6bc 6f none 6fc 2fc 2bc 2f 4fc 4bc 6bc 6f none 6fc  Difference Between Means 0.09721 0.10804 0.11277 0.12107 0.18731 0.20179 0.21870 0.22665 0.37392 -0.09721 0.01083 0.01556 0.02386 0.09010 0.10458 0.12148 0.12944 0.27671 -0.10804 -0.01083 0.00473 0.01303 0.07927 0.09375 0.11065 0.11861 0.26588  Simultaneous 95% Confidence Limits -0.04057 -0.03051 -0.02290 -0.01915 0.05164 0.06474 0.08091 0.09097 0.23455 -0.23500 -0.11560 -0.10771 -0.10439 -0.03317 -0.02020 -0.00410 0.00617 0.14939 -0.24660 -0.13726 -0.11940 -0.11605 -0.04486 -0.03188 -0.01578 -0.00552 0.13773  0.23500 0.24660 0.24845 0.26129 0.32299 0.33884 0.35648 0.36232 0.51329 0.04057 0.13726 0.13883 0.15211 0.21337 0.22935 0.24707 0.25270 0.40403 0.03051 0.11560 0.12886 0.14211 0.20340 0.21938 0.23708 0.24274 0.39403  *** *** *** *** ***  *** ***  ***  109  The SAS System The GLM Procedure Tukey's Studentized Range (HSD) Test for IB  15  Comparisons significant at the 0.05 level are indicated by ***.  treatment Comparison 2f - 2fc 2f - 2bc 2f - 4f 2f - 4fc 2f - 4bc 2f - 6bc 2f - 6f 2f - none 2f - 6fc 4fc - 2fc 4fc - 2bc 4fc - 4f 4fc - 2f 4fc - 4bc 4fc - 6bc 4fc - 6f 4fc - none 4fc - 6fc 4bc - 2fc 4bc - 2bc 4bc - 4f 4bc - 2f 4bc - 4fc 4bc - 6bc 4bc - 6f 4bc - none 4bc - 6fc 6bc - 2fc 6bc - 2bc 6bc - 4f 6bc - 2f 6bc - 4fc 6bc - 4bc 6bc - 6f 6bc - none 6bc - 6fc  Difference Between Means -0.11277 -0.01556 -0.00473 0.00830 0.07454 0.08902 0.10592 0.11388 0.26115 -0.12107 -0.02386 -0.01303 -0.00830 0.06624 0.08072 0.09762 0.10558 0.25285 -0.18731 -0.09010 -0.07927 -0.07454 -0.06624 0.01448 0.03138 0.03934 0.18661 -0.20179 -0.10458 -0.09375 -0.08902 -0.08072 -0.01448 0.01691 0.02486 0.17213  Simultaneous 95% Confidence Limits -0.24845 0.02290 -0.13883 0.10771 -0.12886 0.11940 -0.11768 0.13428 -0.04636 0.19545 -0.03343 0.21146 -0.01734 0.22919 -0.00703 0.23478 0.13612 0.38618 -0.26129 0.01915 -0.15211 0.10439 -0.14211 0.11605 -0.13428 0.11768 -0.05974 0.19222 -0.04674 0.20818 -0.03062 0.22587 -0.02040 0.23156 0.12290 0.38280 -0.32299 -0.05164 -0.21337 0.03317 -0.20340 0.04486 -0.19545 0.04636 -0.19222 0.05974 -0.10797 0.13692 -0.09189 0.15465 -0.08157 0.16024 0.06157 0.31164 -0.33884 -0.06474 -0.22935 0.02020 -0.21938 0.03188 -0.21146 0.03343 -0.20818 0.04674 -0.13692 0.10797 -0.10787 0.14168 -0.09758 0.14730 0.04561 0.29865  ***  *** ***  *** ***  ***  110  The SAS System 16 12:36 Tuesday, March 22, 2011 The GLM Procedure Tukey's Studentized Range (HSD) Test for IB Comparisons significant at the 0.05 level are indicated by ***.  treatment Comparison 6f 6f 6f 6f 6f 6f 6f 6f 6f none none none none none none none none none 6fc 6fc 6fc 6fc 6fc 6fc 6fc 6fc 6fc  -  2fc 2bc 4f 2f 4fc 4bc 6bc none 6fc 2fc 2bc 4f 2f 4fc 4bc 6bc 6f 6fc 2fc 2bc 4f 2f 4fc 4bc 6bc 6f none  Difference Between Means -0.21870 -0.12148 -0.11065 -0.10592 -0.09762 -0.03138 -0.01691 0.00795 0.15523 -0.22665 -0.12944 -0.11861 -0.11388 -0.10558 -0.03934 -0.02486 -0.00795 0.14727 -0.37392 -0.27671 -0.26588 -0.26115 -0.25285 -0.18661 -0.17213 -0.15523 -0.14727  Simultaneous 95% Confidence Limits -0.35648 -0.24707 -0.23708 -0.22919 -0.22587 -0.15465 -0.14168 -0.11531 0.02791 -0.36232 -0.25270 -0.24274 -0.23478 -0.23156 -0.16024 -0.14730 -0.13122 0.02224 -0.51329 -0.40403 -0.39403 -0.38618 -0.38280 -0.31164 -0.29865 -0.28254 -0.27230  -0.08091 0.00410 0.01578 0.01734 0.03062 0.09189 0.10787 0.13122 0.28254 -0.09097 -0.00617 0.00552 0.00703 0.02040 0.08157 0.09758 0.11531 0.27230 -0.23455 -0.14939 -0.13773 -0.13612 -0.12290 -0.06157 -0.04561 -0.02791 -0.02224  ***  *** *** ***  *** *** *** *** *** *** *** *** *** ***  111  Thickness Swell Test  The GLM Procedure Tukey's Studentized Range (HSD) Test for TS NOTE: This test controls the Type I experimentwise error rate, but it generally has a higher Type II error rate than REGWQ.  Alpha 0.05 Error Degrees of Freedom 50 Error Mean Square 0.000135 Critical Value of Studentized Range 4.68143 Minimum Significant Difference 0.0222  Means with the same letter are not significantly different.  Tukey Grouping  Mean  N  treatment  A A A A A A A A A A A A A A A A A A A  0.072210  6  none  0.061700  6  2bc  0.061149  6  2f  0.061016  6  2fc  0.059963  6  4fc  0.059676  6  4bc  0.058760  6  6bc  0.055812  6  6fc  0.055669  6  4f  0.054413  6  6f  112  Water Absorption test The SAS System The GLM Procedure Tukey's Studentized Range (HSD) Test for WA NOTE: This test controls the Type I experimentwise error rate, but it generally has a higher Type II error rate than REGWQ.  Alpha 0.05 Error Degrees of Freedom 50 Error Mean Square 0.000135 Critical Value of Studentized Range 4.68143 Minimum Significant Difference 0.0222  Means with the same letter are not significantly different.  Tukey Grouping  Mean  N  treatment  A A A A A A A A A A A A A A A A A A A  0.101014  6  6fc  0.100432  6  none  0.100159  6  2bc  0.097921  6  4f  0.095227  6  2fc  0.093698  6  6bc  0.093385  6  4bc  0.087564  6  6f  0.086018  6  2f  0.084932  6  4fc  113  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0072768/manifest

Comment

Related Items