1 L AYERED SILICATE PARTICLES FILLED POLYMER NANOCOMPOSITE FOR BARRIER APPLICATION By JINWOO KWAK A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR TH E DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010
2 2010 Jinwoo Kwak
3 To my father, mother and my lovely wife for all devoted love and sacrifices they m ade toward my education
4 ACKNOWLEDGMENTS First of all, I am grateful to the late, Dr. Charles L. Beatty my advisor for his endless support I would like to thank Dr. Matthew Siobhan, from SCF processing Inc., for her exceptional guidance and generous support throughout my doctoral studies. I will never forg et her help and kindness. And, I would like to express my thanks to Dr. Bruce A. Welt for providing valuable comments and help in the progress of my research. I would like to extend my thanks to Dr. Valentin Craciun for his time and effort toward me as my co chair. My thanks also go es to Dr. Christopher Batich and Dr. Hassan El Shall for serving on my committee. I would also like to express my gratitude to Sungwan Jeon for his thankful endeavor and assistance as a coworker. I also appreciate Tom Gasset, A niket Selarka, Myong Hwa Lee and Dr. Junghun Jang for helping me in several ways. I would like to thank the faculty and staff in Department of Materials Science and Engineering at the University of Florida The financial support of the Kraft Foods, Inc a nd the US Army Natick Soldier RD&E Center during my Ph.D. study is gratefully acknowledged. This research was possible with their enduring support and kind cooperation. Finally, I wish to thank all of my friends both near and far who have always encouraged me throughout my life.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4 LIST OF FIGURES .........................................................................................................................9 LIST O F ABBREVIATIONS ........................................................................................................15 NOMENCLATURE ......................................................................................................................19 ABSTRACT ...................................................................................................................................23 CHAPTER 1 INTRODUCTION ..................................................................................................................25 Polymer Clay Nanocomposite ................................................................................................26 Three Conventional Microstructures ...............................................................................26 Preparation of Polymer Clay Nanocomposite .................................................................27 The Fundamental Concept for Barrier Solutions ....................................................................28 Barrier Mechanism in a Polymer F ilm ............................................................................28 Basic Barrier Model for Polymer Clay Nanocomposite Materials .................................31 2 LITERATURE REVIEW ON MICROSTRUCTURE OF POLYM ER CLAY NANOCOMPOSITE MATERIALS FOR BARRIER APPLICATIONS ..............................37 Microstructures of Polymer Clay Nanocomposite .................................................................37 Microstructures in Practical Nanocomposite Materials ..................................................37 Roles of organic modifier for intercalation and exfoliation .....................................37 Effects of process ing parameters on intercalation and exfoliation ..........................42 Formation of various microstructures ......................................................................46 Interplay between clay orientation and segmenta l motion of polymer chains .........53 Effects of Microstructures on Physical Properties ..........................................................55 Mechanical property .................................................................................................55 Thermal property ......................................................................................................57 Optical property ........................................................................................................58 Barrier property ........................................................................................................59 Effects of Microstructure on Barrier Property ........................................................................59 Extended Barrier Model for Polymer Clay Nanocomposite ...........................................59 Practical Polymer Clay Nanocomposite for Barrier Applications ..................................66 The effects of microstructure of clay platelets on barrier properties .......................66 The effects of processing and selection of matrix on barrier properties ..................68 The barrier properties based on geometry of clay platelets ......................................73 Different types of barrier structure ...........................................................................75
6 3 POLYMER CLAY NANOCOMPOSITE COATINGS ON NONPOLAR POLYOLEFIN SUBSTRATE TO ENHANCE BARRIER PROPERTY ............................101 Introduction ...........................................................................................................................101 Experiments ..........................................................................................................................104 Materials ........................................................................................................................104 Step 1: Preparation of Polymer Clay Nanocomposite Solutions ...................................104 Step 2: Atmospheric Pressure Plasma Treatment ..........................................................105 Step 3: Coating Samples after APP Treatment ..............................................................105 Step 4: Drying PCN Coated PP specimens ...................................................................105 Characterizations ...........................................................................................................105 Results and Discussion .........................................................................................................106 Surface Modification by Atmospheric Pressure Plasma Treatment ..............................106 Clay Exfoliation .............................................................................................................108 Barrier Properties ...........................................................................................................109 4 E FFECT OF pH ON MICROSTRUCTURE OF LAYERED SILICATE AND BARRIER PROPERTIES IN NANOCOMPOSITE COATING SYSTEM S ......................124 Introduction ...........................................................................................................................124 Experiments ..........................................................................................................................127 Preparation of Polymer Clay Nanocomposite Solutions at Various pHs ......................127 Coating Samples After Atmospheric Pressure Plasma Treatment ................................128 Characterizations and Analysis .....................................................................................128 Results and Discussion .........................................................................................................129 A House of Cards Structure ...........................................................................................129 Effects of pH on the Clay Platelet Microstructure ........................................................130 Barrier Properties ...........................................................................................................133 5 THEORETICAL ESTIMATION OF BARRIER PROPERTIES OF POLYMER CLAY NANOCOMPOSITE CONTAINING TWO CLAY TYPES "PC2N" .................................149 Introduction ...........................................................................................................................149 Experiments ..........................................................................................................................150 Preparation of PC2N Films ...........................................................................................150 Characterizations ...........................................................................................................151 Discussion of Model and Results .........................................................................................151 Derivation of Barrier Equation for PC2N .............................................................................154 6 EFFECT OF POLY(ACRYLIC ACID) ON BARRIER PROPERTIES OF HIGHLY FILLED NANOCOMPOSITE FILMS .................................................................................164 Introduction ...........................................................................................................................164 Experiments ..........................................................................................................................165 Preparation of Highly Filled Nanocomposite Films ......................................................165 Characterizations ...........................................................................................................166 Results and Di scussion .........................................................................................................167
7 Variables Affecting Barrier Properties ..........................................................................167 Roles of Poly(vinyl) Butyral Layer ...............................................................................170 APPENDIX 7 CONCLUSIONS ..................................................................................................................188 A CLASSIFICATION OF COMMONLY USED LAYERED SILICATES ...........................194 B COMMERCIALLY AVAILABLE ORGANOCLAYS AND ORGANIC MOLECULES USED FOR THE MODIFICATION OF THE SURFACE OF NANOCLAYS ..................195 LIST OF REFERENCES .............................................................................................................196 BI OGRAPHICAL SKETCH .......................................................................................................208
8 LIST OF TABLES Table page 21 Four different formulations to prepare polymer clay composites (Demirkol et al., 2007) ..................................................................................................................................78 22 Average spherulites radius (Rsph) and Tc crystallized from the melt as a cooling rate of 1o C/min (Yu et al., 2003) ...............................................................................................78 23 Oxygen permeability of HDPE/clay nanocomposites at different clay concentrations and ultrasonic amplitudes (Swain and Isayev, 2007) .........................................................79 24 Processing conditions for blown film (Thellen et al., 2005) ..............................................79 31 Specific data of PV A samples ..........................................................................................111 32 Compositions of PCN solutions used for coating ............................................................111 33 Specific conditions of APP treatment ..............................................................................112 34 OTR results of PVA coated PP specimens ......................................................................112 41 Sp ecific conditions for the preparation of each specimen ...............................................135 42 EDS results for white unknown particles dispersed in clay composite solutions ............135 51 Specific information of all clays used for the preparation of PC2N ................................159 52 Specific experimental conditions for the preparation of composite films .......................159 61 C orresponding relative amounts of clay and PAA for each sample ................................175
9 LIST OF FIGURES Figure page 11 B asic crystal structure of MMT .........................................................................................32 12 E volution of morphologies of PCN from natural MMT to fully exfoliated PCN and corresponding general patterns of XRD; a more realistic microstructure which is a mixture of all structures and a corresponding TEM micrograph .......................................33 13 P rocedures for prepar ing PCNs .........................................................................................34 14 P ermeability mech anism of gas molecules ........................................................................35 15 Volume temperature relationship for amorphous polymer ............................................... 36 16 Effects of microstructures of clay p latelets in polymer on barrier properties of PCNs. ....36 21 Phase diagrams of polymer gr= 0.2, Ngr = 5 (Ginzburg et al., 2000) ....80 22 Expectation of arrangements of modifiers in the intergallery regions ...............................80 23 WAXRD patterns of the organically modified nanoclays and their nanocomposites and correspondin g TEM micrographs ...............................................................................81 24 WAXRD Pattern of HDPE/MMT Nanofil 15 and HDPE/HDPE g MA/MMT Nanofil 15 nanocomposites ................................................................................................81 25 WAXRD Pat terns of SAN/MMT and EVOH/MMT ........................................................82 26 Illustration showing distribution of organic MMT layers in a cross section of nylon66/organic MMT nanocomposites. (Yu et al., 2003) ........................................................82 27 Illustration of peeling mechanism for the exfoliation process (Borse and Kamal, 2009) .................................................................................................................................82 28 Shear induced orientation of platelets with the flow (x), gradient (y) and vorticity (z) directions (Lin Gibson et al., 2003) ..................................................................................83 29 TEM micrographs showing PP/clay elongated at 150 C ..................................................83 210 TEM micrographs for EVA/SIOM, EVA/DIOM, EVA/TRIOM ......................................84 211 Illustration of the morphological hierarchy at different length scales extant in the PET/CL30B and TEM micrographs ..................................................................................84 212 Variation of dspacings of the MAPE/CL20A and MAPP/CL20A nanocomposite fibers depending on volume fraction of clay .....................................................................85
10 213 A beam configuration of SAXS measurement and resulting SAXS photographs of MAPE/CL20A nanocomposite fibers and MAPP/20A nanocomposite fibers ..................85 214 Y and Z beam configuration of SANS measurement and resulting SANS patterns of both configurations of films consisting of PEO/LRD with lower aspect ratio or PEO/CNa with higher aspect ratio .....................................................................................86 215 Schematic diagram to represent th e procedure for the preparation of a cured epoxy clay fabric film composite with with exfoliated clay nanolayers at the outer surfaces of the clay film ...................................................................................................................86 216 TEM of the nanocomposite viewed along x showing layeredsilicates oriented orthogonally to the BCP lamellae ......................................................................................87 217 Schematic picture to represent the fundamental concept for the use of anionic polymer as a binder additive to achieve a clay film of densely packed structures ............87 218 TEM image of a cross section of the saponite film with 20 wt% of carboxymethyl cellulose sodium salt and corresponding schematic mic rostructure, ................................ 87 219 TEM image of the film with a polymer loading of 20 wt%, showing the c axis direction (crosssection) ......................................................................................................88 220 Schematic representation of the internal architecture of the PVA/MTM nanocomposite (Podsiadlo et al., 2007) ............................................................................88 221 DSC crystallization exotherms plots of PE gMAn and I.44PA/PE gMan nanoc omposites and XRD patterns for PE gMAn and I.44PA/PE gMAn systems ........88 222 Optical micrographs and Hv patterns ................................................................................89 223 TEM imag es of PUCCN with 5% C (a and b) and PULCN with 3% L (c and d), (I); AFM images of PUCCN and PULCN, (II) (Mishra et al., 2008) ......................................89 224 Bright field TEM images of the extruded polystyrene/montmorillonite pellet and corresponding in situ WAXD patterns ...............................................................................90 225 TGA result for normal PVA/organoclay PCN and crosslinked PET/CL30B PCN ...........90 226 T hree common particles geometries used in models ........................................................90 227 T heoretical effects of volume fraction on relative permeabilities .....................................91 228 Theoretical effects of volume fraction on relative permeabilities .....................................92 229 A unit of three silicate platelets representing the arrangement of platelets in whole system of id ealized PCN structure ..................................................................................... 92
11 230 Effect of confinement of clay platelets on polymer chain segment immobility and resulting relative permeability ...........................................................................................92 231 E ffect s of lateral spacing on d spacing and relative permeability at different aspect ratio of platelets; this plot is based on Eq uation (2 12) ( = 0.05) .................................... 93 232 C ompari son of the changes in H spacing as b varies between larger and smaller platelets .............................................................................................................................. 93 233 S chematic figure to depict the variance of S values depending on platelets orientation s .........................................................................................................................94 234 E ffect s of volume fraction on Rp when fully exfoliated structure .....................................94 235 E ffect s of volume fraction on Rp for present barrier models c alculated at aspect ratio of 50 and 1000 ...................................................................................................................95 236 Comparison of prediction for Rp as a function of aspect ratio; all curves were plotted based on the E quation 22, 23, 24, 25, 26, 2 11 and 213 at = 0.05 .........................96 237 Illustration of effective width and its effect on barrier property of PCN ........................... 96 238 OTR chart of a pure PET s heet and PCN sheets (left) and corresponding TEM micrographs of MMT PCN ). .............................................................................................97 239 TEM micrographs to represent the difference in tortuosity (refer to dot lines) between phaseseparated microst ructure of unmodified MMT PCN and exfoliated one of modified MMT PCN ..............................................................................................97 240 O xygen relative permeability of composite depending on clay content (wt%) and TEM micrographs of PU/CL UE400S00 ...........................................................................98 241 TEM micrograph of a compression molded specimen containing 5 wt% of clay filler dispersed in polyesteramide matrix ...................................................................................98 242 Oxygen permeability rate of various blown composite/neat film samples ........................99 243 Chemical structure of poly (vinyl alcohol co vinyl acetateco itaconic acid) where, x, y and z is 97, 2 and 1, res pectively (Grunlan et al., 2004) ................................................99 244 E ffect s of PVA contents on oxygen permeability of neat polymer containing no fillers and effect of clay loadings on oxygen permeability of terpolymer/ CNa compos ites ..........................................................................................................................99 245 C oncept of probability of a barrier cell using Kadanoff cell and Critical volume fraction versus aspect ratio of platelets with S= 0. ..........................................................100 246 TEM micrographs of various PI/clay hybrid films ..........................................................100
12 247 Commercialized structural composite to enhance oxygen barrier property .................... 100 31 SEM micrographs of Laponite JS powder... ....................................................................113 32 A chemical structure of PVA. ..........................................................................................114 33 Schematic mechanism showing the formation of ideally exfoliated PVA/Laponite JS nanocomposite solution through the ordered experimental procedures ..........................114 34 Schamtically illustrated experime nt procedures ..............................................................115 35 Controllable variables when APP treating the surface of i PP. .......................................115 36 SEM micrographs taken at 6000X of untreated surface ..................................................116 37 3D surface and height images taken by AFM (contact mode) on a 20 x 20 m2 of untreated surface ..............................................................................................................117 38 I nfluence of the flow rate of nitrogen gas which is one of the APP parameters; 3D surface images taken by AFM. ........................................................................................118 39 Influence of the WD which is one of the APP parameters ; 3D surface images taken by AFM ............................................................................................................................119 310 Influence of the flow rate of nitrogen gas which is one of the APP parameters on surface roughness. ............................................................................................................120 311 Influence of WD which is one of the APP parameters on surface roughness. ................120 312 WAXRD patterns of the pure Laponite JS powder and several PCNs having exfoliated silicate layers in PV A mat rix. .........................................................................121 313 Difference in WAXRD patterns between LP20PVA PCN samples prepared with and without high shear process ...............................................................................................121 314 WAXRD patterns of several PCN coated surface containing various amount of silicate particles obtained at low angle range from 2 to 10 degree. .................................122 315 Dependence of the content of clay loaded in PCN on OTR values. ................................122 316 Effect of the flow rate of nitrogen gas on OTR values. ...................................................123 317 Effect of the WD on OTR values. ....................................................................................123 41 S tructure of silicate layers, Laponite JS when dispersed in water phase .........................136 42 S chematic mechanism of a house of cards st ructure ....................................................137
13 43 S chematic mechanism of a house of cards structure through the diffusion of LMW anionic PAA chains into a clay structure and expected ideal barrier structure. ..............138 44 Schamtically illustrated experiment procedures ..............................................................139 45 SEM micrographs of Laponite JS clay powder and corresponding clay aerogel ............140 46 TEM micrographs of clay aerogel consisting of three structural features, edge standing, tilted and paralleled as indicated as arrows ......................................................141 47 DRIFT results of composite samples of C5PA73 prepared at pH 2, 4, and 6 .................142 48 Photographs of the states of composite solutions prepared under different pH conditions .........................................................................................................................142 49 Picture of C5PA91IS solution after 2 weeks storage and SEM micrographs of its by product, white particles dispersed in the solution. ...........................................................143 410 E ffects of the pH on the peak positions of XRD results depending on PAA contents. Red lines represent the positions of peaks for pure Laponite JS powder ........................144 411 E ffects of the am ount of PAA on the microstructure of PCN samples prepared at basic condition. ................................................................................................................145 412 TEM micrographs of C5PA73BS and C5PA73AS .........................................................145 413 XRD results of clay compos ite powder and corresponding coated layer by a coating rod ....................................................................................................................................146 414 E ffect of clay platelet microstructure on barrier properties when the same amount of clay was arranged .............................................................................................................147 415 SEM micrograph s of nanocomposite coated surfaces .....................................................147 416 D ependence of permeability on pH and the amount of PAA for highly filled PCN coating system. .................................................................................................................148 51 Schematic diagram showing a filled polymer with (a) one clay type and (b) two different clay types ...........................................................................................................160 52 E ffect of the total volume fraction of clay2 on the relative permeability of PC2N .........161 53 Actual relative permeability of various composite films calculated fr om OTR values measured by MOCON .....................................................................................................161 54 Optical transmittance of composite films at the visible wavelength range ..................... 162 55 Picture s of prepared transparent and translucent composite film samples ......................162
14 56 P arallel arrangement of two nanoclays in an effective volume. ......................................163 61 Schematically presented experiment procedure for the preparation of PVB coated PAA/CNa nanocomposite film ........................................................................................176 62 Schematic diagram of an intuitive curvature test to measure fle xibility of obtained film specimens .................................................................................................................177 63 OTR values of highly filled nanocomposite film specimens of different polymer loading. .............................................................................................................................177 64 Schematic microstructure of clay platelets and PAA chains and corresponding SEM micrographs of prepared film specimen ..........................................................................178 65 SEM micrographs (x 5,000) of cross section areas of obt ained highly filled nanocomposite film specimens. .......................................................................................179 66 X ray diffraction pattern for CN3P0, P10, P30, P50, P70, and P100; (b) a FWHM as a function of polymer loadings ........................................................................................180 67 DSC thermograms of highly filled nanocomposite films with various amounts of polymer ............................................................................................................................181 68 SEM micrographs (x 5,000) of surface areas of obt ained highly filled nanocomposite film specimens. ................................................................................................................182 69 D ependence of RMS on the amount of polymer in highly filled nanocomposite films. .183 610 An expected schematic picture of protruded curved edge of tactoids from the surface of CN3P70.. .....................................................................................................................183 611 3D height contrast images taken by AFM .......................................................................184 612 Power spectral density diagrams of highly filled nanocomposite samples ......................185 613 Curvature of highly filled nanocomposite films with various amounts of polymer and a picture of a flexible film specimen CN3P50) ...............................................................186 614 TGA thermograms of highly filled nanocomposite films with various amounts of polymer ............................................................................................................................187
15 LIST OF ABBREVIATION S AFM Atomic F orce M icroscopy APP Atmospheric Pressure Plasma BCP Block C opolymer BOPP Biaxial ly O riented P oly(propylene) CBMC Carboxymethyl C ellulose CEC Cation E xchange C apacity CED Cohesive E nergy D ensity CL Cloisite (CL15A, CL20A, CL25A, CL30B: refer to Table.21) CNa Cloisite Na DDAB Dimethyl D istearyl A mmonium B romide + DGEBA Diglycidyl E ther of B isphenol A DIAB D ioctadecyldimethyl A mmonium B romide DP Degree of P olymerization DRIFT Diffuse Reflectance Infrared Fourier Transform DS Degree of S aponification DSC Differential S canning C alorimetry DTA D odecyl T rimethyl A mmonium EDS Energy D ispersive X ray S pectroscopy EG Ethylene G lycol EVA Ethylene V inyl A cetate EVOH Ethylene V inyl Alcohol FDA The United States Food and Drug Administrat ion FE SEM Field E mission S canning E lectron M icroscopy FWHM Full W idth at H alf M aximum
16 GA Glutaraldehyde gMAn Maleic A nhydride G rafted GRAS Generally R ecognized A s S afe HDPE High D ensity P olyethylene HMW High M olecular W eight IEP Isoelectric P oint LBL La yer B y L ayer process LMW Low M olecular W eight LP Unmodified Laponite LRD Laponite RD MAPE Maleated P oly(ethylene) MAPP Maleated P oly(propylene) MC M odified C lay MDEA N methyl D iethanol A mine MMA Methyl M ethacrylate MMT Montmorillonite MRE Meal R eady to E at NBR Acrylonitrile B utadiene C opolymer OAC Organic A mmonium C hloride OTR Oxygen T ransmission R ate PAA Poly(acrylic acid) PALS Positron A nnihilation L ifetime S pectroscopy PC2N PCN consisting of two clay types PCL Poly( caprolactone) PCN Polymer C lay N anocomposite
17 PDMS Poly(dimethyl siloxane) PEG Poly(ethylene glycol) PEO Poly(ethylene oxide) PET Poly(ethylene terephthalate) PI Poly(imide) PMMA Poly(methyl methacrylate) PP Poly(propylene) PS Poly(styrene) PSD Power S pectral D ensity PU Polyurethane PUCN Thermoplastic P olyurethane C lay N anocomposite PVA Poly(vinyl alcohol) PVB Poly(vinyl butyral) PVC Poly(vinyl chloride) RH Relative H umidity RMS Root M ean S quare SAN Styrene A crylonitrile C opolymer SANS Small A ngle N eutron S cattering SAXS Small A ngle X ray S cattering SBR Styrene B utadiene R ubber SBS Styrene B utadiene S tyrene scCO2SEM Scanning E lectron M icroscopy Supercritical C arbon D ioxide SIAB Octadecyltrimethyl A mmonium B romide SPA Sodium P olyacrylate
18 TEM Transmission E l ectron M icroscopy TGA Thermal G ravimetric A nalysis TPU Thermoplastic P olyurethane TRIAB T ricetadecylmethyl A mmonium B romide TSE Twin S crew E xtruder VA Vinyl A lcohol VAc Vinyl A cetate WAXRD Wide A ngle X ray D iffractometer WD Working D istance
19 NOMENCLATURE Chapter 1 J diffusive flux of gas through film Q the amount of gas passing through a surface of certain area A surface of area t time D diffusion coefficient for a gas molecule through the film S solubility coefficient C concentration of gas molecules C1C concentration of gas molecules inside film (one side) 2l thickness of the film concentration of gas molecules outside film (counter side) p partial pressure of gas p1p partial pressure of gas inside film (one side) 2P permeability partial pressure of gas outside film (counter side) permachor parameter a a constant based on gas molecules (Equation (17)) s a constant based on polymers (Equation (17)) material density TgV glass transition temperature f diffusion jump length average free volume fraction of a polymer effective jump frequency
20 Chapter 2 gr N grafting density gr volume fraction of fillers (clay particles) grafting length polymer clay pair wise interaction AcA area of an alkylammonium ion of a 2:1 phyllosilicate eT optical transmittance area of a half unit cell of a 2:1 phyllosilicate C the dependent on the the ratio of the refractive index of clay and matrix d dspacing (Equation (21)) fwR weight concentration of clay particle per unit volume cm Rayleigh scattering factor average aspect ratio nm wavelength of incident light refractive index of matrix polymer RpP permeability of polymer clay nanocomposites Relative permeability Po aspect ratio permeability of matrix polymer without nanoclay t thickness of a clay platelet (refer to Figure. 2 26) d diameter of a clay platelet (r efer to Figure. 226) L length of a clay platelet (refer to Figure. 2 26) reduction parameter of permeability 1 polymer chainsegment immobility factor 2 detour factor
21 H face to face distance between two clay platelets (refer to Figure. 2 29) b edge to edge distance between two clay platelets (refer to Figure. 2 29) S orientation parameter angle between direction of orientation and sheet normal (refer to Figure 233). Pc critical value of clay platelets for minimum permeability cL/w degree of exfoliation (length/thickness) critical volume frac tion Chapter 3 Ra R s urface roughness calculated as average roughness max R maximum h eight qL Evaluation length r oot mean square value Z(x) profile height function Chapter 5 Clay 1 Clay type 1 with large aspect ratio Clay 2 Clay type 2 with smaller aspect ratio than Clay 1 L1 lateral length including edge to edge distance of clay 1, ( l1+b1L ) 2 lateral length including edge to edge distance of clay 1, ( l2+b2H ) ) 1 vertical length occupied by Clay 1 per unit volume (h1+w1h ) 1h face to face distance of clay 1 2w face to face distance of clay 2 1w width of clay 1 2l width of clay 2 1l length of clay 1 2 length of clay 2
22 b1b lateral edge to edge distance of clay 1 2N lateral edge to edge distance of clay 2 1 N number of clay 1/unit volume within effective volume of clay 1 2 Tv number of clay 2/unit volume within effective volume of clay 1 total volume of two clay types 1v volume of clay 1 2v volume of clay 2 Ve ff1 V effective volume of clay 1 eff1 T effective volume of clay 2 volume fraction of two clay types 1 volume fraction of clay 1 2 volume fraction of clay 2 polymer volume fractio n of polymer n total number of clay 2 occupy ing in a layer in the unit cell l the number of layers in the unit cell d the shortest tortuous path through unfilled polymer dad lateral tortuous path a2 d tortuous pa th additional lateral tortuous path by clay 2 Chapter 6 k local curvature r radius of curvature
23 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy L AYERED SILICATE PARTICLES FILLED POLYMER NANOCOMPOSITE FOR BARRIER APPLICATION By Jinwoo Kwak May 2010 Chair: Charles L. Beatty Major: Materials Science and Engineering A polymer clay nanocomposite (PCN) system for barrier applications has been studied and various mat erials have been utilized to achieve a better barrier solution. This study aims at making progress in the p ackaging industry toward super barrier materials by tailoring process ing parameters to the creation of new barrier solutions and also by systemizing theoretical background knowledge. Comparisons of several barrier models were made in order to develop a new model and to assist with material selection and experimental design. PCN consisting of synthe tic layered silicate (Laponite) in polyvinyl alcohol ( PV A ) w as applied to atmospheric plasma treated, semi rigid polypropylene (PP) samples. PCN coating bonded to polyolefins offer s a simpler method to achiev e higher barrier properties without modifying commodity materials H ighly filled PCN coating layer s, with parallel arranged dense structures of Laponite w ere studied. An investigation on effect s of pH of poly(acrylic acid) (PAA) on PCN coating system performance suggested that enhanced barrier properties w ere achievable with the aid of positively charged poly(acrylic acid) and a shear coating process es T his work also suggested another PCN coating system that incorporates a dense stratifi ed structure of silicate platelets using a polymer as a linkage molecule
24 B enefits of dispersing smaller nanoparticles between larger ones in a polymer matrix (PC2N) were explored and a barrier model to estimate relative permeability of the PC2N system was proposed. I t was demonstrated that t otal volume fraction of particles determine d barrier properties and this was confir med by experiment No significant d ifference in oxygen transmission rate of three composite films was found. A film containing only Cloisite had a lower oxygen transmission rate than t hat of Laponite and PC2N films because of the relatively larger aspect r atio of Cloisite ; however Laponite a nd PC2N films had better optical transmittance due to the smaller aspect ratio resulting in lower numbers of scattering events Achieving good flexibility and high barrier properties was possible with PAA negatively cha rged organic molecules as linkage molecules and poly(vinyl butyral) (PVB) exterior layers Unlike the observed house of cards structure formed in Laponite aqueous solution, the large aspe ct ratio of Cloisite Na+ ( CNa ) clay platelets resulted in parallel arrangement as observed by scanning electron microscopy ( SEM ) Result showed that the layered structures had vari ous features depending upon the amount of linkage polymer applied. In addition, enhancement of barrier properties and flexibility was achieved after PVB coating. A b arrier model based on Nielsen s detour theory or Beall s filling void theory predicted overall barrier properties experiment s
25 CHAPTER 1 M icrostructure s of clay platelets in polymer matrices ha ve been widely studied in order to better understand relationships between physical structures of polyme r clay nanocomposites (PCN) and resulting properties A variety of applications such as material reinforcement [1 5] gas barrier and membrane separation s [4 15] advanced electr onic devices  and biomedical application s  have shown great potential of PCN materials These studies help to elucidate characteristics of filler particles and polymer s as well as mutual interaction s between components Because of differences between conventional composites with macro sized filler and PCN, new approaches are required to better understand features of nano fillers in polymer chains. S tudy of orientation and dispersion of nanosized inorganic platelets in polymer and effects on physical properties have shown differences that suggest potential for better design s of PCN materials compared to previously proposed systems INTRODUCTION C ommonly used filler particles for PCN are phyllosilicate T he classification of commonly used phyllosilicate for PCN i s listed in Appendix A The basic crystal structure of Montmorilonite (MMT) which is one of the more commonly used silicates under the smectites group is shown in Figure 11. An octahedral sheet is between two tetrahedral silica sheets and oxygen at oms in these sheets are common to both layers. Resulting three layer units are aggregated together showing negative surface charges with positive ions between units that neutraliz e the whole structure The weak bond between the cation and negative surface allows diffusion of water and other polar molecules into this interplanar space, intergallery region inducing an expansion of the mineral structure. In general, separat ion of clay platelets via th is expansion process and insertion of polymer chains into th e intergallery region ha s proven to offer enhancement of phys ical properties such as barrier flam e retarda nt mechanical, optical and
26 membrane properties. U nderstanding mechanism s of changes from micron sized particles to nanosized particles and correlat ions to changes in physical propert ies are key aspects to the nanotechnology required for the preparation of PCN materials. This study ha s made progress in the area of packaging films regarding enhanced gas barrier properties. Through the literature study various practical PCN systems and several barrier models to estimate theoretical permeability have been reviewed. Based on this theoretical background, four barrier s tructures have been developed for study. In this chapter three typical microstructures of PCN, generalized preparation methods for PCN and fundamental mechanism of barrier performance in a polymer film as well as in a nanocomposite structure are reviewed Polymer Clay Nanocomposite D egree of dispersion and dissoc iation of clay platelets by diffusion of polymer chains into interlayer spacing determines the microstructure of PCN. In fact, the term PCN only can be used for the morphologically ideal case, when particles are exfoliated. Exfoliation means that clay nano platelets are uniformly dispersed in the polymer matrix When polymer is inserted into the intergallery of the mineral making a regularly enlarged spacing, an intercalated microstructure is achieved and separation of platelets causes an increase i n surface area, resulting in modified physical properties. However, poor separation leads to local morpholog ies, a phaseseparated micro sized morphology with aggregated stacks dispersed in polymer matrix forming a microcomposite resulting in slightly mo dified properties compared to both the intercalated and exfoliated. Figure 12 illustrates the evolution of morphologies of clay platelets in the polymer matrix from their initial state of aggregated stacks to fully exfoliated one X ray diffractomet ry ( XR D ) and t ransmission e lectron m icroscopy ( TEM ) have been used in combination to identify microstructures. According to Bragg s law, d spacing calculated from a characteristic peak in Three Conventional Microstructures
27 XRD is usually used to determine separation of interspacing due to polymer chains and this is a descriptor used to analy ze intercalated morphology ( blue curve in Figure 1 2). W hen an intercalated morphology is achieved, a shift to lower 2 (2 theta ) angle s of a characteristic peak (a shift from the black curve to the blue curve) is observed. F urther increase in both the dspacing and random orientations of clay platelets caused by diffusion of polymer chains results in an exfoliated morphology. L ack of regularities from platelets results in no peak as shown by the red curve in Figure 12. Small humps may appear instead of strong peaks suggesting the possibility of remain ing tactoids However, discrepancies can arise in analyzing XRD patte rns because a peak can appear from a unit cell of crystallized polymer matrices or from additives added during the process of preparation of PCN Thus interpretation of XRD requires care and consideration about variable s that can affect positions of peaks M ore definitive tools such as TEM are also used to analyze PCN structures. Because of differences in atomic configuration between polymer and platelet, contrasts from different diffractions appear in TEM micrograph as shown in Figure 12. F igure 12 repr esents a general realistic microstructure of PCN material consisting of intercalated and exfoliated region s C lay platelets appear as dark lines and some gradient around the dark lines illustrate interfacial region s between clay and polymer. Even though some differences in final morphologies of PCN may depend on selection of materials and processing parameters, processes for preparation of PCN can be classified into three categories with resulting microstructures s hown in Figure 13. Preparation of Polymer Clay Nanocomposite In solution mixing, w hen clay platelets are added to an aqueous phase, clay powder is in the form of aggregated nano particle stacks Stacks may be separated by hydration of cations C ations may be exchanged with organic modifier s with a lkyl chains in order to modify the hydrophilic nature of inorganic platelets and to open intergallery regions Diffusion of polymer
28 chains into intergallery regions is usually accomplished by high shear forces. For intercalating polymerization, monomer di ffuse s into the intergallery space homogeneously, and then polymerization occur s in the inner and outer regions resulting in intercalated morphology of PCN. In melt process, shear and crystallization significant ly effect on final morphologies of PCNs. The Fundamental Concept for Barrier Solutions P ermeation of a gas molecule through neat polymer s, without filler s is well known. G as molecule s are a dso rbed on the surface and diffuse into and through the interior of a film and finally desorbed from the opposite side of the material. D iffusion of a permeant, such as a gas molecule, conforms to the Ficks law and sorption and desorption of gas molecules follow Henrys law as shown in Figure 14 ( I ) Barrier Mechanism in a Polymer F ilm At Q J S teady state diffusion of g as across a film results in mass flux (J) across the film described by Eq uation (1 1) (1 1) Where, Q is the amount of gas passing through a surface of area (A) normal to the flowing direction and t is duration during which the diffusion occurs. According to Ficks first law, flux, J is also expressed as Eq uation ( 12) l C C D x C D J ) (2 1 ( 12) w here, D is the diffusion coefficient of a gas molecule through the polymer film and C1 and C2 represent the concentration of gas molecules at each sid e of the film, respectively a nd l is the thickness of the polymer film.
29 Combining Equation ( 11) and ( 12) results in Eq uation ( 13). l At C C D Q ) (2 1 ( 13) Since Henrys law applies for typically low concentrations of gas ; thus C can be expressed as Sp C ( 14) Where S is the solubility coefficient and p is the partial pressure of gas. Therefore, Eq uation ( 13) can be rewritten as the following Eq uation ( 15) l At p p DS Q ) (2 1 ( 15) P ermeability (P) is defined the combined coefficient of D and S (P=DS), also known as DS, where D is the diffusion coefficient, and S is the solubility coefficient, P can be estimated from the following Eq uation ( 16) ) (2 1p p At Ql P ( 16) Therefore, knowing the partial pressures of gas on both sides of the films the amount of gas diffused and dimensions of the film allows us to estimate the permeability of a polymer film sample. Koros  reports that permeability is related to diffusion and solubility as mentioned above (P=DS) and hence permeabili ty P may either decrease or increase with increasing molecular weight of the penetrant depending upon which factor dominates. Therefore, the permeability of a polymer is a function of both the diffusion and solubility coefficients which are mainly determined by the chemical and structural properties, such as polymer polarity and
30 degree of crystallinity. Salame  proposed a comprehensive semi empirical correlation of polymer structure and gas permeability based on the Permachor parameter ( ). P = a s3 w here a and s are constants based on the permeant gas and polymers. And this Permachor parameter is based on the molec ular forces holding the polymer together, cohesive energy density (CED), material density ( ) and average free volume fraction (V ( 17) f 7 5 ln 71 ) (2 fv CED Permachor ) of the polymer. Therefore, the Permachor value of a polymer can be calculated using Eq uation ( 18) which is based on the r atio of amorp hous and crystalline regions  ( 18) Therefore, the choice of a polymer for a high barrier property could be based on the Permachor value. Equation (18) suggests that t he higher Permachor value, the higher the barrier property. Increasing barrier properties of polymer materials can be realized in several ways. A polymer generally consists of crystalline and amorphous region s as shown i n the simple fringed micelle model of semicrystalline polymer of Figure 14 ( II ) Gas molecules diffuse into amorphous regions with lower packing densit ies because crystalline region s effectively block diffusion due to higher packing densit ies S mall inte rmolecular pores which are called excess free volume hole introduced by random entanglements of polymer chains in amorphous regions allow gas molecules to diffuse through the film by a mechanism known as vacancy hopping diffusion  Excess free volu me hole is caused by differences in specific volume s of polymer in different equilibrium states as shown in Figure 15. G eneral ly the temperature called T2 is the point where a real discontinuity in specific volume occurs and T2 is generally Tg 50oC f or a
31 glassy polymer. For instance, i n the case of PE, the average size of free holes is below 1 nm above Tg, with concentration of about 10 1021 26 1 D / cc. The Van der W aals diameter for oxygen is 0.375 nm [28,29] Permeant gas molecule s usually sta y longer in free vo lume holes and occasionally hop into neighboring hole s by the formation of a channel as shown in Figure 15. Therefore, the number and size of holes which is called static free volume determine s the permeability of a polymer Static free volume i ncluded within a polymer is related to solubility (S) and dynamic free volume, and it determines diffusivity (D) with diffusion jump length ( ). This relation regarding diffusivity was reported by the following Equation ( 19)  ( 19) Where, is the effective jump frequency. Increasing crystalli nity is the most effective way to lower permeability of a polymer material. However, flexibility cost, optical properties under certain circumstances often deteriorate in sympathy with increasing crystallinity. Artificially increasing apparent crystallini ty may be accomplished with u niform ly dispers ed and exfoliated clay platelets that are impermeable to gas molecules Neils e n s detour theory  suggested this fundamental concept of enhanc ing barrier properties in PCN materials and based on this model, a variety of studies have been conducted. As shown in Figure 16, the detour path would be increased significantly when a fully exfoliated microstructure (c or d) is achieved when compared to those of others (a and b). In this way, the microstructure of PCN has significant effect s on barrier properties governed by factors such as aspect ratio and orientation of clay platelets. Aspect ratio, volume fraction, orientation and other factors have been involved in approximat ing the actual permeability in various b arrier models. Basic Barrier Model for Polymer Clay Nanocomposite Materials
32 Figure 11. B asic crystal structure of montmorilonite
33 Figure 12. E volution of morphologies of PCN from natural MMT to fully exfoliated PCN and corr esponding general patterns of XRD; a more realistic microstructure wh ich is a mixture of all structures and a corresponding TEM micrograph 
34 Figur e 13. P rocedures for prepar ing PCNs
35 Figure 1 4. P ermeability mechanism of gas molecules through polymer film, I; D iffusion motion of gas molecules in free holes wi thin the amorphous region of the neat polymer, II; M otion of gas molecules to avoid impermeable clay platelets in PCN materials, III.
36 Figure 15. Volume temperature relationship for amorphous polymer (upper) L attice holes in polymer and vacancy hopping diffusion model of gas molecules (lower )  Figure 1 6. Effects of microstru ctures of clay platelets in polymer on barrier properties of PCNs. (a) Phaseseparated (inset figure shows relatively shorter detour path through the neat polymer), (b) interc alated, (c) fully exfoliated and (d) more realistic exfoliated structure (arrow lines represent the diffusion path of gas molecules and the direction of diffusion is perpendicular to the surface of the PCN layer)
37 CHAPTER 2 LITERATURE REVIEW ON MICROSTRUCTURE OF POLYMER CL AY NANOCOMPOSITE MATERI ALS FOR BARRIER APPL ICATIONS Unlike three typical microstructures practical polymer clay nanocomposite ( PCN ) have shown diverse microstructures such as wedged preferentially oriented nematic phase, densely stratified and others A general microstructure of PCN in real composite systems can be described as a mixture of intercalated and exfoliated structures Interplay among polymer matrix, nanoclays organic molecules covering nanoclays and processing parameters lead s to a structure representing the whole composite system. T his chapter reviews variables that affect formation of certain microstructure s of layered silicate particles from the stage of material selection to processing parameters. Microstructures of Polymer Clay Nanocomposite Roles of organic modifier for intercalation and exfoliation Microstructures in Practical Nanocomposite Materials Achieving a complete exfoliation involves matching surface energy between matrix polymer chain s and organic modifier s of nanoclays in order to increase interactions between nanoparticles and polymer. A theoretical phase diagram of PCN was suggested  to explore the role of grafted organic modifier to prove the importance of compatibility and interactions between nanocla y particles and polymer matrix in determining an average microstructure of PCN. Density function theory for the ordering of nanoclay particles and self consistent field approximation for long range order interaction between nanoclay layers covered with cer tain type s of modifier s are used to realize phase behavior of model ed polymer clay mixture s. T he resulting phase diagram is shown in Figure 21. Phasebehaviors of a PCN can var y depending on grafting density ( gr) grafting length ( Ngr) volume fraction ( ) polymer clay pair wise interaction s ( ) and whether it is located in an
38 isotropic or nematic phase. This theoretical development agrees well with recent experimental re sults [7,3437,39] I ncreases in gr a nd NgrO rganic modifier s serve to ( a) increas e intergallery space to prevent platelet platelet co hesion and ( b) match polarity to accelerate favorable interactions be tween polymer and modifier chains It can be deduced that the crucial factor for obtaining high degree of exfoliation at small volume fraction is increased affinity of polymer chains to nanoclay surfaces as a result of the increased intergallery distance. These functions of organic modifier s are highly dependent on arrangement s of modifier molecules which give rise to different particle surface properties. Subsequently, modified surface pr operties have a significant effect on the final morphology during swelling and shear processes. improve miscibility between nanoclays and polymer matrix with lower values Thus, it can be demonstrated ( Figure 2 1) that the equilibrium phase behavior at short grafting length is in the immiscible region. This suggests that there should be a st obtain an exfoliated structure (isotropic or nematic phase) when grafting length is short. A rrangement s of organic modifier s depend on chains characteristics in low charge 2:1 clay minerals. Depending on difference between areas of alkylammonium ion s Ac and areas of a half unit cell of a given 2:1 phyllosilicate Ae arrangement s of organic modifier s in the intergallery region s are determined. As shown in Figure 22 (IV), when Ac is equal to Ae, the monolayer of alkylammonium ions is close packed (F igure 2 2 (II), a). As chain length increases above monolayer coverage, alkylammonium ions adapt bilayer morphology (Figure 2 2 (II), c) resulting in some increases in dsp acing by inserti on of one more layer When the area for closepacked morphology is m ore than twice the area available for each monovalent cation (Ac > 2Ae), the transition from a complete bilayer to a pseudotrimolecular layer occurs and alkylammonium
39 cations will form paraffin type structures in higher charge 2:1 clay minerals (Figu re 2 2 (II), b). When paraffin type microstructure s form tilting angle shows changes that are sensitive to l ayer charge as shown in Figure 22 (III)  M easurement of tilt angle was suggested by Usuki et al  as illustrated in Figure 22 (I). Therefore c ation exchange capacity (CEC) value of clay, amount of organic modifier, length of alkyl chains, and dimension of clay particles are critical to understanding arrangement s of organic modifier s A pproximation of molecular length and the spacing from X r ay diffractometer (XRD) using the equation suggested by Usuki allow s reasonable estimation of average tilting angle and it may be possible to study whether modifier molecules are inclined to the silicate layer surface. A dsorption of modifier onto the layer sur face would be in monolayer if the ratio of the amount of modifier for cation exchange reaction to clays CEC approaches to unity Organic modifier s (also refer red to as surfactant s ) with long alkyl tails have been used for exfoliation and a va riety of commercially available modified nanoclay s (organoclays) can be used depending on the types of processing and matrices. Appendix B lists commonly used commercial organoclays and modifier s used. Organoclays have also been prepared by the cation exchange react ion by many research groups and effects of surfactant molecules on the final morphology of PCN have also been reported. E ffect s of modifier molecules arrangement on characteristics of intergallery region s w ere investigated  R esults from thermal and X RD analys e s of different grades of organoclays provided understanding of the order ed state of grafted alkyl chains. W ater desorption from or through surface s of organoclays decreased order of Nanofil 804 bearing hydroxymethyl group, Nanofil 919 containing one tallow and one benzyl methyl, and ditallow chain attached Nanofil 15 E ndothermic peak area for melting tallow chains of Nanofil 15 was much larger than the other two Nanofil grades. This s uggested that benzyl or
40 hydroxyl group interrupt ed formation of ordered structure s of alkyl chains with higher packing density. The fact that the basal spacing mainly depends on length of alkyl chains attached to surface s of nanoclays was confirmed by the result that wellstacked alkyl chains of Nanofil 15 ha d t he hig her d001Polymer matrices bearing hydrogen bonding such as polyamide and poly(vinyl alcohol) (PVA) usually have a good affinity for surfaces of nanoclays because of the hydrophilic nature of the intergallery zone. When considering two f unctions of the organic modifier for good exfoliation, single long alkyl cation s are required to be exchanged with cations in the gallery to make this space larger without distort ing the polarity of the hydrophilic polymer chains As vinyl acetate (VAc) co ntent increases in e thylene vinyl acetate (EVA), intensity of intercalating peak became larger in XRD pattern s because the higher the polarity caused by VAc groups in backbone chains, the more easily polymer chain s diffuse d into interlayer regions Single alkyl chain modifier molecules could not render enough space for full exfoliation by EVA chains even with higher VAc content resulting in a intercalated morphology  Terpolymer, m odified PVA consisting of 97 wt% of vinyl alcohol (VA), 2 wt% of VAc and 1 wt% of itaconic acid showed a strong interaction with organic modifier molecules because of the stronger hydrogen bonding by COOH on the itaconic acid unit than normal OH bonding  than others in XRD. PCNs based on polyurethane (PU), which has both lower and higher polarity at the soft and hard segment, respectively, revealed various interactions at the interface area between PU and organoclays. Nonpolar hydrocarbon attached to Nanofil 15 filled PCN formed a phase separated morphology because of the poor attractive force between organoclays and polar PU matrix. In this case electrostatic force between nanoclays tend ed to squeeze polymer out, which conf i rmed expectation s based on the function of grafted organic modifier. However, the case of
41 hydroxyethyl attached Nanof il 804 deviated from the expectation in that increased interactions between organoclays bearing polar groups and polar PU result s in steric hindering of platelet platelet cohesion by tethering polymer chains to surface s of clay. Based on results of XRD and t ransmission e lectron m icroscopy (TEM) ( Figure 23) the author concluded that there was little or no reaction wi th the suggested reasons of deactivation of OH groups through hydrogen bonding to the aluminosilicate surface or shielding effect s by long alk yl chains  E ffect s of adding maleic anhydride grafted high density polyethylene (HDPE g MA) as interfacial agent between organoclays and HDPE phases on morphology were i nvestigated as shown in Figure 24  Due to the function of HDPE gMA which i s a polar copolymer used as an interfacial agent, organoclays were dispersed well when using HDPE g MA. On the other hand, TEM image s of HDPE/ Montmorilonite ( MMT ) Nanofil 15 showed microsized agglomerates suggesting that HDPE chains could not enter into the interlayer spacing. Wide Angle (WA) XRD (WAXRD) of HDPE/MMT Nanofil 15 also confirmed this by showing intercalated peaks at around 3.1 degree s suggesting no increase in d spacing caused by insertion of polymer chains. E ffectiveness of interfacial agen t s for better dispersion worked well for polypropylene ( PP) /MAPP/MMT system s as shown in TEM micrographs of Figure 2 4 (II, III) [45,46] Strong interaction energy between polymer chains and organoclays does not guarantee a large amount of polymer penetrat ing into the interlayer spacing This was confirmed by XRD and boundary region between polymer and clay particles in TEM micrographs  While s tyrene acrylonitrile copolymer (SAN)/MMT shows a discrete boundary region with poor interfacial affinity, ethy lene vinyl alcohol (EVOH)/MMT revealed a tightly associated feature. The feature of the quite diffused boundary region represents a stabilized boundary due to strong
42 interaction energy. However, dissociation of organoclays was better developed in SAN/MMT c omposite. More polymer penetrated into the intergallery resulting in wellexfoliated structure s In Figure 2 5, the intercalating peak after 5 min in the case of EVOH/125C composite is much more intense than that of SAN/125C. Therefore, initial diffusion r ate of guest polymer molecules depends on polarity of the polymer. Increasing modifier sometimes decreases ability of polymer molecules, especially those with low molecular weight such as an oligomer, to penetrate into the clay. In other words, oligomer mo lecules prefer to penetrate into the intergallery space of MMT when the surface of this clay is partially covered with a modifier  B onding between delocalized quadpolar of a benzyl group in organic modifier and/or epoxy resin induces strong V an der Waals attractions. This interaction is enhanced by the preferential parallel orientatio n of benzyl groups with surfaces of nanoclays Therefore when epoxy resin is used as polymer matrix, use of modifier containing long alkyl chains and benzyl groups wil l assist exfoliation by increasing dspacing and facilitating penetration of polymer Addition of hydroxyethyl group to the chemical structure of a modifier will further promote tethering of chains to the silicate surface. However, some criteria were requi red for selection of modifier in epoxy/MMT composite system. Long alkyl chain masked OH group may hinder the reaction with epoxy resin and excess number of OH groups were sometimes counterproductive  Effects of process ing parameters on intercalatio n and exfoliation Various processes ing methods including insitu polymerization, solution, and melt process have been suggested to be capable of producing a complete exfoliated system. Among various ways, melt process ing is the most common method and has b een widely used in the industry because of its eas e of operation and flexibilit y Many benefit s can be realized because melt
43 process ing usually utilize s commercially available facilities and materials. Such process es make it possible to vary formulations e asi ly. Four different formulations with different facilities and materials were built as shown in Table 2 1. M icrostructures were examined to investigate effect s of processing parameters  When nonpolar polymer was used such as F1 and F3, intercalation or exfoliation could not be attained by batch mixing or t win s crew e xtruder (TSE) because of poor interaction between clay platelets and polymer Moreover, this drawback cannot be improved by controlling process ing parameters. In this case, u sing polar g roups attached to interfacial agent s is an efficient way to achieve intercalation or exfoliation Dynamic properties such as shear viscosity and elasticity could be used to judge microstructur al state during process ing. A phase separated structure, with pa rticle size larger than 1 m, generally shows decreased shear viscosity and elasticity with increasing mixing time and intensity. However, completely exfoliated or intercalated system s, which can be regarded as colloidal suspensions, show a counter behavior because of the enhanced interactions between matrix and particles Sonication and TSE process es with use of modified poly (dimethyl siloxane) ( PDMS ) with a polar moiety as matrix in F2 showed a typical colloidal dynamic behavior of a well exfoliated micr ostructure. F4 showed some sensitivity to processing conditions. While increasing specific energy input for mixing increased basal spacing, an intercalated structure was found at relatively short mixing times and low specific energy input  Lee et al proved that ultrasonication is most effective for dispersion of nanoparticles in low molecular weight (LMW) polymer matrices  In situ ultrasonication at elevated temperature was successfully used to enhance clay dispersion and exfoliation in LMW PP/ CL20A without any mechanical shear or aid of interfacial agent. Increase in ultrasonic energy
44 promised increased interlayer spacing with enhanced shear thinning behavior as well as incr eased shear viscosity due to formation of gel like structure s This can be explained in the same manner as with other works  such that lager input energy resulted in well exfoliated structure s with randomly dispersed clay platelets consisting of edge to edge and edgeto face contacts. The efficiency of ultrasonication to obtain well exfoliated structure s was shown with TEM for features of much smaller particle size and disp ersed single platelets (Figure 24 (II)). However, ultrasonication was not a n efficient candidate for exfoliated PCN when high molecular weight ( HMW ) PP was used as matrix showing only modest increases in the interlayer spacing. A masterbatch process consisting of two steps insitu intercalative polymerization and melt intercalation, was used to obtain a poly( caprolactone) (PCL)/MMT PCN. Polymerized masterbatches showed intercalated/exfoliated microstructures and these m asterbatches were dispersed into molten PCL and Poly(vinyl choloride) (PVC) matrices. As a result, intercalated/exfoliated PCL or PVC PCNs, which could not be obtained by direct melt blending, were successfully prepared  Highly concentrated PCN ( above 20 wt% ) with an intercalated microstructure was achievable by overcoming challenges of high viscosity using supercritical carbon dioxide (s cCO2) as a solvent for the intercalating p olymerization of methyl methacrylate ( MMA ) ScCO2Injection molding induces different orientations of clay platelets between the region near the flat surface regions and the bulk regions of bar shaped composite samples (Refer to Figure 2 6). Figure 2 6 show s spatial distribution s of clay platelets dispersed in Nylon 66 in a cross section normal to the injection molding direction. While predominantly parallel orientation of platelets to the surface of bar shaped sample were observed near the flat surface, orientation in was effective in distribut ing MMA monomer homogeneously prior to polymerization 
45 the bulk regions rotated themselves about the injection molding direction as indicate d as an curved arrow in Figure 2 6. Difference s in orientation between these region s may effect on crystallization  S election of mixing method and extrusion is a key factor for melt processing to attain uniformly dispersed and dissociated clay platelets. In other words, shear stress and residence time in the extruder ha ve to be considered. Ham aker  m ade it possible to estimate adhesive energy and force between clay platelets during extrusion theoretically  According to this work, unmodified clay requires much higher shear stress because of considerably higher at tractive interaction between clay platelets with shorter interlayer spacing. Small tactoids also requires high shear stress. During melt processing, the fundamental mechanism of size reduction for clay platelets is based on erosion and surface peeling as shown in Figure 27. Peeling angle, determines the stress required for peeling of platelets and there is a critical angle for the initiation of polymer intercalation process. Different clay interlayer characteristics such as aspect ratio dspacing and initial affinity of polymer chains are key factors in determin ing mixing method and estimate s for energy required. Inducing a preferential orientation of clay platelets in a PCN system also requires appropriate shear rates [39,60,61] A critical shear rat e was found for the formation of a macroscopic domain pattern with preferential orientation of clay platelets in the polymer matrix when a PCN was obtained from solution process  Randomly oriente d platelets at equilibrium showed isotropic small angle neutron scattering ( SANS ) pattern s in the x z and y z planes resulting in a physical gel structure by diffusion of polymer chains. When shear rate is above a critical value flow induced by shear rate disrupts the
46 transient physical network formed in equilibrium state and clay platelets move faster than polymer can diffuse thus fragmenting the network into macroscopic domains. M acroscopic domain patterns accompan y flow induced nanoscale heterogeneit ies with structural connectivity in a macroscopically ho mogeneous system. O rientation of clay platelets shows anisotropic SANS pattern proving a shear induced orientation as shown in Figure 28 (II)  Similarly flow induced microstructures with different features were also found in PP with maleic anhydride grafted PP ( PPgMAn ) /MMT PCN obtained by melt extrusion process. In this work, flow induced internal structural change occurred in both shea r and elongational flow and changes were very different from each other in that shear induced change involved a n e xtremely long relaxation time. As shown in Figure 29 (a) and (b), coherent order of the orientation was lower in (b) and this suggested that slow elongation flow leads to formation of a house of cards structure. R esults of this study also showed that st rong straininduced hardening originated f rom the microstructure with preferential perpendicular orientation of clay platelets to the stretching direction  Formation of various microstructures Several morphological features that are different from the three typical microstructures have been observed under various conditions. The physical properties from these microstructures are known to be substantially different from normally well intercalated/exfoliated structure in nano scale. F ully exfoliated stru ctures can be realized when two conditions of sufficient interlayer spacing and strong interaction between polymer chains and nanoclays. A wedged structure can be formed due to lack of interlayer space of nanoclays and/or polarity of polymer chains at cert ain condition. EVA/MMT nanocomposite is an example showing a wedge structure. VA content in EVA polymer chain determines polarity of this polymer. Therefore VA content and three organic modifiers containing different length s of alkyl chain s result ed in dif ferent
47 interlayer gaps that were controlled to investigate the effect on morphological features of prepared PCN materials  The organic modifier s used were o ctadecyltrimethyl ammonium bromide (SIAB), dioctadecyldimethyl ammonium bromide (DIAB) and tric etadecylmethyl ammonium bromide (TRIAB) The higher number of long alkyl chains the modifier possessed, the larger interlayer spacing was obtained as sh own in WAXRD pattern of Figure 210 (II). Cross checking TEM micrographs and a WAXRD pattern of c orresp onding PCN materials shows that partially exfoliated, intercalated, and a mixture of partially intercalated and exfoliated structure s were obtained for EVA/SIOM, EVA/DIOM and EVA/TRIOM, respectively. A partially exfoliated structure referred to as a wedge structure of EVA/SIOM occurred because the basal spacing of SIOM is not enough to insert polymer chains and VA content is also not enough to induce strong interaction between the chains of EVA and SIOM. Therefore only some portion of EVA chains was wedged into the sheets of clay so, the ordered structure of SIOM is damaged s howing an exfoliated feature in WAXRD pattern. When the number of alkyl chains attached to the modifier molecules was enough to increase the basal spacing for EVA chains to enter, in tercalated PCN was obtained as shown in Figure 210 (b). A tactoid, which is also termed a stack, consisting of several sheets stacked faceto face with an interlayer charge and an agglomeration of these tactoids is a micron sized particle which is a m ain component in phase separated structures It is possible to form individually dispersed aggregates of completely delaminated sheet throughout the system. D ifferent type s of exfoliated structure s were introduced in cross linked polyethylene terephthalate (PET)/CL30B nanocomposites system  Two situations of exfoliated structure were observed in PET/CL30B. First a long range ordered exfoliation in which clay sheets are completely delaminated and homogeneously dispersed throughout the matrix ( Figure 2 11 (c) ) S econd a
48 short range ordered exfoliation which contains localized regions of exfoliated sheet dispersed throughout the matrix ( Figure 211 (a) and (b) ) This short range ordered exfoliation is useful to envisage macroscopic physical behaviors. Volume fraction and orientation of nanoclays in polymer are key factors in the formation of various microstructures in PCNs. MAPE/CL20A and MAPP/CL20A with incorporation of various concentrations of nanoclays were prepared to address morpholog ical evolution from high volume fraction condition to dilute condition  Volume fraction of clay is a major consideration in determin ing specific microstructures when other factors are optimized. Volume fraction controlled morphologies were classified into four dis tinct stages as shown in Figure 212. At high concentrations stage IV, silicate particles are so close that V an der Waals force between particles induces strong attractive interactions and particles inevitably adapt the morphologies of phase separated or intercalated structures. Stage III is a dual state that exfoliated and intercalated structures coexist and stage II of intermediate concentration possibly has ordered exfoliated morphologies by dominating steric interactions. Lower concentration of clay pa rticles at stage I eventually render s a disordered exfoliated morphology because the interactions between polymer chai ns and clay particles govern microstructures of this system. It has been emphasized that relation s between microstructures of PCN and t hei r physical properties in nanocomposite field are different from conventional composites. In addition to state of dispersion and degree of aggregation, there is one crucial factor that has significant effects on PCN materials. This factor is the overall pre ferential orientation of clay platelets. The study of MAPE (PP)/CL20 A  also considered the orientational behaviors of clay particles. 2D Small Angle X Ray Scattering ( SAXS ) patterns reminiscent of strong anisotropic features on the equator direction we re found for both MAPE/CL20A and MAPP/CL20A nanocomposite fibers,
49 which indicates that n ormal directions of clay platelets are perpendicular to the shear direction for fiber formation A notable fact is that different anisotropic tendency and orientation parameter ( S) in terms of concentration of clay particles was observed in MAPE/CL20A and MAPP/CL20A. While MAPE/CL20A showed monotonic increase of S as volume fraction of CL20A increased, S decreased when volume fraction wa s above 15 % in MAPP/CL20A nanocom posites This was confirmed by two curves along with two series of SAXS photographs ( Figure 2 13) M onotonic increase of S with increases in clay volume fraction in MAPE/CL20A was explained by the fact that increased possibility of collision between clay particles prevents particles from tumbling freely hence facilitating intercalated structure of higher order ( S = 1). Decrease in S above 15 vol% of clay in MAPP/CL20A was due to tactoids formed in the composite which act as a domain. These tactoids were t umbled instead of completely separated into individual platelets thus lowering S. These tactoids can be regarded as an exfoliated/intercalated structure of a global scale mentioned above. In conclusion, selection of appropriate volume fraction is a key fa ctor in determining the desired microstructure under MAPP or MAPE based nanocomposite systems. General or intuitive response of clay platelet orientation in a polymer is c orientation which is a direction normal to the surface of a composite and bori entation which occurs during strong strain induced hardening. O rientation of platelets shown in MAPE (PP)/CL20A nanocomposite is a typical borientation. A dvent of a general c orientation was discovered in multilayered nanocomposite polymer films prepa red by a layer by layer spreading exfoliated polymer clay solution, which was confirmed by SANS patterns that have relatively more isotropic patterns in the y beam direction and relatively large anisotropy in the z beam co nfiguration ( Figure 214 )  La ponite RD (LRD) particles with a lower aspect ratio and Cloisite Na+ ( CNa ) particles with a higher aspect ratio exfoliated in
50 polyethylene oxide (PEO) were oriented by spreading. Collapsed network structures during drying process es lead s to an increase in concentration of particles in the system and subsequently reduces the rate of relaxation toward their original structure from the oriented structures by spreading. LRD showed s mall amount of anisotropy in film sample s of higher concentrations as shown in ( a d) of Figure 214. It is generally expected that larger particles align more preferentially than the smaller ones at a similar concentration. However, the orientation of LRD was stronger than that for CNa in a spreading process LRD contains more parti cles in a restricted space than CNa at the same volume fraction because of small aspect ratio and lower density of LRD. Closer distance between particles induced more attractive interactions resulting in a stronger network. S trong networks in solution assi sted in good alignment of LRD particles showing a higher degree of orientation during the filmspreading process. D ependence of orientation on volume fraction of LRD indicated that there is a critical concentration at which highly aligned network s occur a s also shown in MAPP/CL20A. Similarly a critical concentration for an intercalated microstructure was observed in poly ( methyl methacrylate) (PMMA)/Cloisite composites prepared by in situ solution polymerization using supercritical carbon dioxide ( S cCO2) as a solvent. Cloisite particles in PMMA formed highly ordered nematic architecture when concentration s reached 40 wt% which is considered a critical limit. Above this critical transition concentration, d spacing decreases and the polymer volume is homoge neously distributed, resulting in fu lly intercalated microstructure  It can be concluded from three cases [39,57,60] that different preferential orientations can be rendered depending on the types of clay as well as the shearing methods T here also ex ists a critical concentration at which highly ordered microstructures and highly concentrated nanocomposites are formed
51 For the purposes of maximiz ing functions for specific applications, the microstructure of PCN materials sometimes displayed unique stru ctures that are completely different from those described so far. Epoxy/Na MMT fabric film composite was successfully obtained by dip coating a heterostructured mixedion clay film with a mixture of diglycidyl ether of bisphenol A ( DGEBA) and Jeffamine Up on curing this system produced unprecedent oxy gen barrier properties. Figure 215 represents the structure of heterostructured mixedion clay film. Heterostructured mixed ion clay film was able to be prepared by a partial cation exchange reaction which is possible due to the diffusionlimited exchange process. Partial cation exchange in outer surface of a clay self supporting film made it possible to insert polymer resin between particles by modifying the swellability of the particles in this region. A den sely stratified unmodified layer and polymer filling of microvoids between organoclays are two key factors to attain superior barrier properties. An orientationally ordered hierarchical nanocomposite for the purpose of mechanical and barrier enhancements w as attained by inserting poly styrene ( PS) functionalized clay particles into the PS domain of styrene butadiene styrene (SBS) tri block copolymer (BCP) through roll casting. Individually dissociated clay sheet s were stabilized by tethering high molecular w eight PS chain s to clay layer s and as relaxation time decreased, individual exfoliated clay layers likely f or m ed a discotic nematic because of large areal dimensions and curved shape of exfoliated clay platelets ( Figure 2 16 (b) ) As exfoliated clay sheets served as a template for the lamellar domains, more layer m isorientation with more defects in microstructure s of SBS C BCP w ere significant when compared to neat SBS BCP  In order to attain super barrier properties with moderate flexibility, new d ensely stratified clay structure s w ere devised [65,66] Unlike highly concentrated nanocomposites w ith the clay
52 content at about 50 wt% or general PCN materials with small loading usually less than 10 wt%, these structures use clay particles as a major com ponent to maximize barrier function with addition of a few binder additives. Densely packed clay platelets with face to face parallel stacking was realized by ionic bonding between anionic pendent groups of polymer chain and the positively charged edges of platelets The attached anionic polymer chains on one edge of a platelet further promoted parallel arrangement as shown in the left picture of Figure 2 17. As a result, the limitations of low flexibility and transparency were overcome. Flexible clay films using carboxymethyl cellulose (CBMC) sodium salt or poly(acrylic acid) (PAA) sodium salt as binder additives were prepared by casting aqueous dispersion of saponite with these binder additives  T he microstructure of CBMC clay film was different from that of PAA clay film in that CBMC molecules entered both the interlayer space and the edge space as shown in Figure 2 18 (b) showing larger d spacing value. Sodium polyacrylate (SPA) was also adapted as a binder to fabricate a flexible saponite clay film  WAXRD results showed that increases in polymer loading lowered stacking faults and lower full width at half maximum (FWHM) values obtained from (001) rocking curves of samples of lower polymer loadings indicated high degree s of preferential c axis orientation. Thus, as polymer loading decrease d, degree of preferred orientation increased with incr easing stacking faults. Figure 2 19 exhibits s chematics of crystal structures and TEM micrographs at different planes  Highly efficient load transfer r equired for higher strength and modulus was realized by layered polymer nanocomposites consisting of PVA and MMT. The microstructure shown in Figure 220 was built by layer by layer (LBL) process of clay platelets. By achieving a high degree of structural organization, number of interactions was maximized promising efficient load
53 trans fer. Scanning e lectron m icroscopy ( SEM ) micrographs in Figure 2 20 show 200300 bilayer films as well as dense coverage of nano platelets and strict planar orientation  In terplay between clay orientation and segmental motion of polymer chains When a PCN material forms, polymer chains and anisotropic nanoplatelets have mutual effects on their behavior. In fact, behaviors of polymer chains and orientational behaviors of clay particles would be varied on a case by case basis Specific types of process es determine crystallization behavior. D ifferent orientation s of clay platelets between the surface and bulk of the PCN ( Figure 26 ) lead to a different crystalline phase. Bulk re gion s contained more perfect phase crystals for Nylon 66 because of relatively lower cooling rates in the bulk. S urface region s showed that the oriented crystallization of Nylon 66 was induced by the oriented clay platelets  PE/modified clays (MC), CL15A and 20A and PP/MC PCNs were obtained by melt processing and orientation during drawing w as analyzed using x ray and Raman spectroscopy. P resence of clay particles in PE and PP matrix reduced the orientational ability of polymer chains in amorphous re gion s during drawing In comparison with corresponding neat polymer drawn by the same drawing ratio, total orientational order as well as crystallites of polymer matrix was lower. However, clay platelets did not have any significant effects on the orientat ional ability of PE and PP crystallites  Clay platelets in copolyamide/CL30B PCN obtained by melt processing acted as a nucleation agent chan ging degree of crystallization and crystalline phases, and phases that formed lamellar with different t hicknesses at different cooling rates  Exfoliated clay platelets in PE g M A n increased polymer crystallization rate s by promoting heterogeneous nucleation and twodimensional crystallite growth following the diffusioncontrolled mechanism. F ast crystallization did not guarantee higher degree of crystallization because the exfoliated clay particles reduced mobility of crystallizable chain segments Decreased intensity in a 110
54 reflection for PE g M A n peaks by insertion of 5 wt% of clay platelets was giv en to prove lowered degree of cr ystallinity ( Figure 221 (right) ) Chain confinement effects were confirmed by inves tigating Tc ( Figure 221 (left)) Depressed Tc for PE gM A n relative to PE is caused by pendant MAn groups. L ower energy was required for th e crystallization process because of more affinity obtained from hydrogen bonding from MAn groups However, increased TcThe concentrati on factor of well exfoliated clay has critical effects on orientation of polymer. A ccording to the calculation of spherulites radius using Hv patterns of CL15A filled HDPE and HDPE g MA n PCNs prepared by a melt compounding and compression molding as shown in Figure 222 and Table 2 2, average spherulites radius decreases in the presence of clay platelets. Spherulites size decreased due to an increased nucleation rate, which was accelerated because clay platelet s act as a nucleation agent for the polymer cry stallization. Thus orientation of PE was determined by nucleation mode and confined nature of spherulitic crystallization process among oriented clay platelets resulting in significantly anisotropic HDPE spherulites  with increasing volume fraction of clay platelets is caused by chain confinement effects. Chains confined by clay platelets required m ore energy to be crystallized. W hile well dispersed particles can increase crystallization rate by donating more available nucleation sites, the degree of crystallization is controlled by chain confinement effects in composite systems  Thermoplastic polyurethane (TPU) clay PCN (PUCN) has different confinement behaviors depending on the types of clay. Preferential confinement of CL20A with soft segment and that of unmodified Laponite (LP) with the hard segment as the more polar part were confirmed by TEM and atomic forc e microscopy ( AFM ) as shown in Figure 223. PUCN with Laponite clay is represented here as PULCN and that with Cloisite is represented as PUC C N in the figure.
55 Preferential association of LP with the hard segment was also reflected from the change in soft s egment degradation temperature  Th is work suggested that confinement effects can be varied depending on the characteristics of clay such as chemical composition, hydrophobicity, aggregation tendency, and degree of dispersion in copolymer matrix based composites or polymer s having two more different types of segments such as PU. PS/ Cetyltrimethylammonium bromide exchanged (CT ) MMT composite exhibited a unique behavior during heating process. Initially a preferential orientation that is parallel to sh ear flo w direction was observed at room temperature A s temperature increased to 85 oC intensities of peaks from shear induced ordered structures decreased. After heating to 95 oC, phenyl ring s obtain ed enough transition moment for face to face orientatio n and perpendicular orientation to the clay particles as suggested as in Figure 224 (b). This affected orientation of clay particles resulting in self asse m bled structure s as shown in Figure 2 24 (a) c. Further increasing thermal energy damaged this self assembled structure. Segmental motions of phenyl rings were evidenced by the dichroic ratio obtained from f ourier t ransform i nfrared (FTIR ) s pectroscopy  Mechanical property Effects of Microstructures on Physical Properties Modulus of PCN materials is usually a function of polymer crystallinity, which can be increased by filler reinforcement. Similar to the theory of reinforcement mechanism s for macrocomposite material s the degree of stress transfer to the reinforcement phase governs overall mechanical properties of nanocomposite structures. A nd nanodimensionally dis persed particles usually result in extremely large surface areas exposed to polymer chains strong interactions between polymer and clay leading to better mechanical properties. A dramati c increase in modulus with low volume fraction was achieved in PVA/CNa PCN and HMW
56 copolyamide/CL30B PCN [68,73] Highly stabilized interface evidenced by strong interaction between clay and polymer was found in EVOH/MMT PCN. D ramatic increases of mechanical properties of EVOH/MMT PCN compared to poor mechanical properties of s tyrene a crylonitrile copolymer (SAN)/MMT PCN show ed roles of interfacial interaction for mechanical enhancement  Interfacial agent s improving adhesion between polymer and clay pl ayed a significant role to enhance phase adhesion. As a result, improved stiffness and high elastic modulus w ere attained in HDPE/CNa PCN that is prepared using PE g MAn as an interfacial agent  Three clay types were used to prepare EVOH with VA conte nt of 28% based PCN, and effect s of modifier on mechanical properties showed that better dispersion of platelets in polymer promised higher storage modulus. In addition, modulus increased with increasing clay content for EVOH/modifiedclay PCN N o effect from clay layers on modulus was found when natural MMT was used as filler  Aspect ratio and degree of orientation of clay platelets are two dominating factors for mechanical property improvement of Nylon6/MMT PCN materials which can be both obtained b y large scaled, simple shear process ing When these two factors decreased, modulus and strength of resulting PCN also decreased, while fracture toughness and ductility increased  Degree of orientation of clay platelets was also reported by Zerda. Acco rding to this work, PMMA/organically modified lay ered silicates PCN prepared by s cCO2MMA polymerization was post treated using melt processing. Orientation induced from a post melt process increased tensile modulus up to 220% compared to untreat ed  F or cross linked polymer based PCN, enhancement in mechanical properties from cross linked chains dominates rather than interaction between clay and polymer As a result, moduli decreased with increasing clay content monotonically. Exfoliation at a global s cale also leads to this unique mechanical property which is opposite to normal PCN materials  Recently, PVA/MMT transparent
57 flexible film was developed using LBL process. This structural composite exhibited four times higher strength and nearly one or der of magnitude higher modulus when compared with pure PVA. Highly efficient load transfer process was given as a basis of breakthrough in mechanical properties. Highly efficient hydrogen bonding between clay and polymer referred to as the Velcro effect a nd six membered ring structure between PVA and MMT were reported as two main driving forces for highly efficient load transfer. When glutaraldehyde (GA) was used as a crosslinking agent, much higher modulus comparable to that of Kevlar was observed with exceptional stability under humid condition  Thermal property Thermal stability has been known as one advantage of PCN because clay platelets in PCN act as a barrier that maximize heat insulation and minimize permeability of volatile degr adation products [52,7577,81] It has been noted that thermal decomposition of PCN materials usually shifts toward higher temperature because clay platelets prevent diffusion of oxygen and assist s in formation of char after thermal decomposition [43,58,73,78,79] Ther mal stability increases as the clay content in PCN increases [58,80] Depending on the intrinsic thermal characteristics of polymer or clay, different observations with general functions of clay for thermal stability can be found [63,81] In contrast with the normal phenomenon that onset of degradation is retarded ( left thermal gravimetric analysis ( TGA ) plot of Figure 225) crosslinked PCN system s have show n a different t rend Onset of degradation was hastened as the clay content increased and degradation between 25 and 400 oC occurred at faster rate. There was a monotonic increase in degradation rate with increasing clay content because of the presence of hydroxyl group in the modifier of CL30B providing a supply of oxygen  Thermal stability of PLA based PCN has a unique characteristic in that clay platelets dispersed in PLA matrix hindered degradation of crystalline
58 structure s of PLA hybrid at low temperature and accelerated d eformation at high temperature s  Optical property P roperties of a mat rix such as the degree of crystallization and the size of spherulites, interfacial refractive index difference between mat rix and filler particles, and size of dispersed particles are factors needed to determine optical transmittance of polymer composites  Deng et al. reported that optical transmittance was dependent on size of dispersed particles over the basal spacing of clay in their proposed model for epoxy/clay nanocomposites, ( Equation ( 21) )  m m m C wn R f d Ce T 2 ) 1 (1 ( 21) where T is the optical transmittance of nanocomposites, C is dependent on the ratio of the refractive index of clay and matrix, d is the dspacing, fw is the weight concentration of cl ay particle per unit volume, RC is the average aspect ratio, m is the Rayleigh scattering factor, nm is of clay particles and their size determine optical transmittance of a composite based on this equation. Light scattering by uniformly dispersed particles in polymer that are much smaller than the wavelength of incident visible light will produce Rayleigh scattering. Therefore, it is generally acceptable that the size and dispersion of clay platelets are dominating factors to obtain better optical propertie s [32,73,78] Lower optical clarity at higher clay loading was observed in PVA/MMT PCN because of the strong scattering of MMT  Haze and gloss of Na+saponite particles dispersed PVA hybrid films was not significantly different from those of pure PVA because of small aspect ratio of filler particles 
59 Barrier property PCN materials with high gas barrier properties are attractive for packaging material s. The barrier property of PCN is affected by the arrangement of each inorganic platelet in polyme r as well as aspect ratio of these platelets The length of diffusion path of gas molecules strongly influences the barrier property of a material and this can be extend ed or shorten ed by arranging impermeable platelets in polymer matrix. Therefore, depending on the materials used as a matrix or filler, the type of processing, and various processing variables such as volume fraction, barrier properties of resulting PCN materials are varied. Effects of Microstructure on Barrier Property Barrier properties of PCN can be affected by geometr y and orientation of filler particles. Various barrier models have been proposed to predict the permeability of PCN. Based on the basic assumption that t he direction of di ffusion is normal to the surface relative permeability ( R Extended Barrier Mode l for Polymer Clay Nanocomposite p o pP P R ) can be defined as permeability of PCN film relative to the unfilled film as following Eq uation ( 22) (2 2) Where, P and PoNiels e n a nd Cussler used ribbonshaped filler for their models suggesting the following barrier formulas are permeability of polymer clay nanocomposites and that of matrix polymer without nanoclay, respectively. As suggested by Takahashi et al. [83,84] each model has different particle geometries which then result in different barrier formulas. Figure 2 26 depicts three particles geometries which have been used.
60 2 1 ) 1 ( pR ( 23) 22 1 ) 1 ( pR ( 24) 2) 3 1 ( ) 1 ( pR ( 25) w here is the aspect ratio of a particle, which can be obtained by dividing the width by its thickness and, for the purpose of simplifying comparison between models, thickness, t is set to 1 nm for all models. This assumption is quite reasonable when consideri ng that the actual thickness of clay platelets typically used as filler particles, such as Cloisite and Laponite is about 1 nm (the range of size is from 0.9 to 2 nm)  And is the volume fraction of filler particles in the matrix. Cussler  divided the model into two patterns, regular (Equation ( 24 )) and more general random pattern (Eq uation ( 25 )). Even though random array has lower barrier permeability than regular array, both cases commonly show that the change in barr ier properties will be dominated by the increase in distance traveled for a gas molecule crossing the platelets filled film and the decrease in area available for diffusion as shown in Figure 227. Cussler also explored the efficient size of platelets to e nhance the barrier properties and concluded that a small platelet could act as a shunt, allowing fast diffusion of gas molecules across that part of the film containing one layer of platelets. Therefore, this model has an advantage in that the design of ma terial is possible to optimize the mechanical and optical properties of a barrier film by varying the size of platelets. Models devised by Nielsen  and Cussler did not contain any factor for confinement i n polymer matrix.
61 Gusev  and Fredrickson  both developed barrier models based on diskshaped particles as drawn in Figure 228 (b). The aspect ratio can be calculated by dividing the disk diameter, d by the thickness, t in these two models. Both models focused on the effect of the platelets ge ometries and concentrations on the reduction of permeability of PCN. Gusev established the role of the geometric factor alone with the barrier formula (Eq uation ( 26)) by assuming the platelets were randomly dispersed and nonoverlapped Rational reference point for the understanding of the contribution of molecular level transformation such as crystallization and segmental motions was also provided in this model. ] ) 47 3 / exp[( 1 71 0 pR ( 26) Fredrickson suggested a quantitative framework to estimate the degree of disorder and polydispersity of platelets with the assumption of randomly distributed disk centers of mass with no spatial correlations. Fredrickson also investigated how two levels of concentrations, dilute and semi dilute, affect the barrier formulas and reported the following crossover barrier formula, Eq uation ( 27) which is quite versatile and can be easily extended to more realistic situ ations 2 2)) 2 /( ) 1245 0 1 (( 4 1 x x x Rp ( 27) W here ) 2 / ln( 2 / x As drawn in Figure 228, some differences were observed in effects of volume fraction on Rp as the aspect ratio increased. The changes in Rp were more sensitive to volume fraction of platelets with higher aspect ratio at loading level below 0.05 in Gusev model than Fredrickson model.
62 Xu  evaluated Rp by combining Rp ) ( ) 1 ( PR theory of semi crystalline polym ers with the detour theory Assuming clay platelets as impermeable crystalline domains in semicrystalline polymer matrix based on Klutes theory  Eq uation ( 22) can be also defined as Eq ( 28) where ( 28) and and are volume fraction of clay particles in PCN and the reduction parameter of permeability as a function of due to the nanoclay particles, respectively. Reduction parameter usually varies when the geometry of clay particle and orientation of these particles in polymer matrix are changed. I n general, permeability reduction in PCN arises from two factors, polymer chain segment immobility factor, 1, and detour factor, 2. Total permeability reduction parameter in PCN is a function of these two factors as given in Equation ( 2 9) and this relation made it possible to combine Rp 1 2) ( ) ( theory of semi crystalline polymer with the detour theory. ( 29) The factor of the detour ratio was obtained by calculating the traveling path length of gas molecules when the dime nsion of clay platelets and the distances between these platelets var y Figu re 229 is a unit consisting of three silicate platelets ( the geometry of platelet is a cuboid) which represent the arrangement of platelets in whole system of idealized PCN struct ure. From the de finitions of effective volume and volume fraction suggested by Saunders  H was expressed using L, b, t and volume fraction as shown in Equation ( 210). t b L t L H 2 2) ( ( 210) As a result, d etour based barrier formulas were presented by Xu as the following Eq uation ( 211) and this f ormula can be converted into Equation (2 12) by means of Eq uation ( 210)
63 3 11 2 1 / ) 1 ( L b t L Rp ( 211) 2 3 2 1 12 1 / ) 1 ( H t t L Rp (2 12) Comparing to other barrier formulas, Xu s formula showed more complexity due to the consideration of edge to edge lateral distance and the relation among all dimension and spacing in a fixed volumetric space. Additionally the chain segment immobility f actor added complexity to the equation. Chain segment immobility factor ( 1) impl ying the confinement effect of clay platelet on polymer matrix was contained in this model. When there is no confinement effect, or 1=1, no effects of crystallization on a di ffusion constant of gas molecules would be found. Comparing Rp of the unconfined matrix with that of confined one, PCN with confined matrix has more enhanced barrier properties due to increased crystallinity as shown in Figure 230. As more clay is include d in the system more reduction in both 1 and H are expected, resulting in entropic penalty of polymer confinement. This accelerates the confinement effect of platelets on polymer chains. As a result, a drastically decreased Rp was obtained and this is closer to the actual RpOne remarkable feature of Xu s model is that the effect of lateral distance, b, on H and that R value which is low ered diffusion due to increased crystallinity. p varies depending on the aspect ratio of clay platelets as plotted in Figure 231. Meanw hile, the smaller partic les show somewhat similar tendency of decreaseing in both H and Rp as b increases, the larger particles reveal a slight decrease in both H and Rp showing less dependence on b. T his was explained using the concept of effective volume as described in Figure 232 that when there is the same amount of increase in b for both smaller and larger pa rticles, smaller
64 particles form more layers and subsequently shorten H significantly within the same volumetric space and the same volume fraction. The basic assumption of all models is that the direction of diffusion of a gas molecule is perpendicular to the surface of platelets. However, this assumption causes some significant deviation from the actual permeability of PCN material. To overcome this limitation, Bharadwaj  introduced the orientational order parameter, S, which usually is used to describe an orientational order of a nematic phase in his barrier model. S is defined as 1 cos 3 2 12S (2 13 ) where represents the angle between the direction of preferred orientation and the sheet normal unit vectors (refer to Figure 2 33). This descriptor reflected well the dependence of tortuosity on the orientational order of the platelets in a continuous manner. Fig ure 233 represents how S varies depending on the degree of orientation of platelets schematically. Planar orientation of platelets is a perfect orientation for nematic structure which is regarded as an ideal barrier model. In contrast, any benefit in barrier properties can be achieved by orthogonal orientation of platelets. Tortuosity factor based barrier formula containing orientation order parameters was given by ) 2 1 ( 667 0 1 1 S Rp ( 214) w here, S varies from 1/2 to 1. Figure 234 is a plot of Rp versus volume fraction drawn by Equation (2 14 ) when the orientation is completely random, or S = 0. When the assumption of fully exfoliated structure w as made, there was no significant difference or remarkable feature observed in comparison to other barrier models. The effect of orientation of platelets on Rp as aspect ratio varies was
65 investigated. As a result, small aspect ratio was much more sensitive to S, showing that small rotation of platelets degraded barrier significantly. However, in the case of platelets with larger aspect ratio, enhancement in barrier with a random orientation of platelets is nearly as good as the case where the platelets were aligned perpendicular to the diffusing path (the right curve showing two re d dot lines are positioned similarly). These tendencies were identified from two lower plots in Figure 23 4. As shown in the figure (completely overlapped two solid lines in the ri ght curve), when orthogonal orientation was set for PCN structure, no barrier effects w ere found as expected, regardless of the aspect ratio. It is noteworthy that all models predicted comparable improvement in barrier properties and these models agreed on the fact that the degree of enhancement in barrier properties was higher with higher aspect ratio than with lower aspect ratio. Two curves in Figure 235 represent the case where at higher loading above 0.2 vol% of clay platelets with high aspect ratio t here is no further enhancement and the value converges to 0. Based on these models, the smaller particles can fill more space in the polymer matrix to enhance the barrier properties further in comparison to larger particles. This fact also can be explaine d by the limitations of models in terms of volume fraction and/or aspect ratio. As shown in curves for all model s drawn at the same volume fraction in Figure 236, every model has a certain limited aspect ratio above which it is no longer applicable to est imate barrier properties precisely In other words, the estimation of barrier properties of PCN using high aspect ratio above 1000 can be achieved using Fredrickson, Bharadwaj and Xu s barrier models. Therefore, m odels that can precisely predict the barrie r properties of PCN have to be developed and more variables might be included. However present models have suggested a good guidance for the design of PCN for barrier application with good consistency validated by reported experimental results
66 In general, t he degree of enhancement in barrier performance of the nanocomposite s relative to the neat films is dependent on the microstructures of platelets in the polymer matrix with the intrinsic natu re of impermeable particles such as large aspect ratio and surface area. The general barrier properties of PCN are based on the fundamental theory of physical tortuous paths which are maintained similarly in actual system with some exceptions. As shown pre vious ly, barrier models have shown that the degree of tortuosity is determined by geometric influences such as the extent of exfoliation/intercalation of the platelets and their mean orientation in the polymer matrix Tortuosity was given as two permeability reduction parameters, polymer chain segment immobility factor, Practical Polymer Clay Nanocomposite for Barrier Applications 1, and detour factor, 2 ,The effects of microstructure of clay platelets on barrier properties in barrier model s [97,98] In the real barrier system, these two factors are also meaningful in determining a permeability of a PCN material. M icrostructure o f platelets is a dominating factor when design ing a high barrier material. T he microstructure of practical PCN materials is a broad mixture of stacked particles, intercalated region, global scaled exfoliation, and fully exfoliated region. A dominating micr ostructure in a PCN that can represent the whole structure usually determines the physical parameters such as the barrier properties. Bharadwaj [63,96] suggested that the effective width of silicate layers in his barrier model to describe the effect of exf oliation on barrier properties as drawn in Figure 237. Aggregation increases the effective width and dispers es the same total number of sheet as this aggregated two sheet results in a dramatic decrease in tortuosity. Thus, exfoliation is a crucial factor for obtaining the max imal performance of practical PCN materials for barrier application. A s ignificant improvement in oxygen barrier properties was attained in HMW copolyamide/CL30B nanocomposite films obtained by a melt processing. Uniform dispersion
67 and high degree of exfoliation with a preferential orientation of clay platelets were proved by pronounced shear thinning rheological behavior  Low molecular weight polymer chains were able to be crystallized readily in solution casting process such as P VA/MMT leading to lower permeability  However, for the melt process, high molecular weight has an advantage on permeability since it promot es higher shear stresses, hence favoring the delamination of clay platelets into individual platelets and maximiz ing tortuosity. Confinement of amorphous polymer chains in intergallery region also contributed to lowering permeability because hindered molecular motions restricted the channel formation for gas diffusion  Intergallery modifiers have played cruci al roles in obtaining a desired microstructure by enlarging the intergallery region and accelerating the diffusion of polymer chains into this enlarged gallery space. Thus, the use of intergallery modifier is an efficient way to enhance barrier property of PCN materials [7,42,99101] It was reported that 1wt% of N methyl diethanol amine (MDEA) based organoclay dispersed in PET specimen ( obtained by insitu polymerization ) lowered the permeability by two fold, compared to pure PET. As illustrated in Figure 238, MMT PCN had higher permeability than MDEA PCN. This was explained by the effect of modifier, MDEA in this specific system since the hydroxyl and carboxylic end groups in MDEA reacted with ethylene glycol (EG) in PET resulting in higher affinity of polymer chains to the intergallery region. As a result, while there observed large nanoclay bundles in MMT PCN (a) which are phaseseparated and intercalated regions, more exfoliated and well delaminated clay platelets were found in MDEA PCN (b) promising much longer retention time of oxygen gas molecules in this PCN material. A similar barrier tendency depending on the use of organic modifier was also found in melt blended acrylonitrile butadiene copolymer (NBR) nanocomposites. In contrast with an unmodif ied MMT PCN, modified MMT by dimethyl
68 distearyl ammonium bromide (DDAB) was separated and dispersed well in NBR matrix. A few tens of layers were found in TEM micrographs and such separated layers extended the routes of diffusion significantly as suggested in the TEM micrographs of Figure 239. Increased volume of platelets further decreased the permeability and some deviation from Neilson s theory was caused by the assumption of ideal perpendicular alignment of platelets to the direction of gas diffusion [ 100] The effects of processing and selection of matrix on barrier properties E ffects of shear stress and shear direction on microstructure of clay platelets in the polymer matrix have been reported so far [45,50,58,60,68,95] Finding appropriate shear str ess to exfoliate stacked particles and shear direction to obtain a preferred orientation is the main factor to achieve the ideal microstructure for barrier application with the use of intergallery modifier. Once a proper process is adapted from one of the three typical processes, melt process, solution casting or coating, and in situ polymerization, the next step is to find the optimum processing parameters which largely depend on the characteristics of mater ials. For instance, modified clay platelets that have larger d spacing value require relatively lower shear stress for exfoliation than natural MMT. S ubsequently exfoliated platelets would promise much lower permeability. Therefore, for enhancement of barrier properties of PCN, processing parameters sh ould be considered in the design of practical PCN materials. T he PU/clay nanocomposites were prepared by in situ polymerization  To form an intercalated structure, PEG MDEAPEG (UE400) was used as a modifier and sonication process aided better disper sion of the modified clay platelets. As a result, an intercalated structure with some disorder was attained and their dspacing w as about 2.62.7 nm. In this study, varying clay contents and sonication time resulted in different degree of dispersion and di ssociation of platelets. Fitting the results of gas permeability to a general physical detour theory suggests that
69 the i ncorporation of clay into polymer matrix decreases gas permeability by extending the tortuous path, as shown in Figure 240. PU/CLUE400S 60 had lower gas permeability than PU/CLUE400S00, which means that the degree of d issociation could be increased by the sonication process. (S60 denotes that sonication process was conducted for 60 minutes) PU/CL UE400S60 5 wt% had the lowest gas permeabil ity and its oxygen permeability was 41% lower than that of pure PU Red dot ted lines in the TEM micrographs illustrated the intuitive diffusion path of gas showing that most lines are blocked by uniformly dispersed platelets in PU/CL UE400S60 5 wt% As sh own in this work, a dapting optimum processing parameters depends on the specific materials and is a fundamental consideration to attain a high barrier property. P olyesteramide/nanomer composite films were obtained by melt mixing and subsequent compression molding  Figure 2 41 is a TEM micrograph of a compression molded specimen containing 5 wt% of clay filler. The white arrow represents a void which still exists after compression molding. These microvoids usually deteriorate barrier properties of a com posite film. Transmission rate of oxygen and water vapor decreased as the filler content increased as generally expected Compression molding could contribute to the reduction in permeability modestly by decreasing the void contents in films and increasing crystallinity. Ultrasonic treatment enhanced the intercalation of HDPE/ organoclays nanocomposite system  HDPE/organoclay nanocomposites were obtained using a single screw compounding extruder with the attached ultrasound die operating at various am plitudes. The die pressure and power consumption due to ultrasound were measured at different feed rates of nanocomposites with various clay concentrations ( Table 2 3 ) The microstructures induced by the ultrasound treatment showed an increasing dspacing up to 50% and these structures exhibited enhanced barrier properties such that a reduction in oxygen permeability of
70 nanocomposites was found reduction by 20% at 2.5% clay concentration and a residence time of 21 s. According to this research, the reduction in permeability was achieved regardless of the reduced crystallinity induced by ultrasonic treatment indicating that the effects of extended detour path by incursion of homogeneously dispe rsed filler dominated the permeability coefficient rather than crystallinity of a polymer matrix in this specific condition. There is also a case that there is no significant effect of processing parameters on barrier properties The role of processing pa rameter, specifically the screw speed for melt compounding process of poly(L lactide) ( PLA ) and PLA/MMT, on barrier properties of nanocomposites w as explored by Thellen  Plasticized PLA /organically modified MMT nanocomposites were compounded and blow nfilm processed using a co rotating twin screw extruder at various conditions ( Table 2 4) Figure 2 42 represents effects of screw speed on oxygen permeability of resulting blown film samples. According to results ( Figure 2 42) oxygen permeability of nan ocomposite films w as lower than the pure PLA films in all cases. R eduction in oxygen permeability ranged from 15 to 48%. I t is generally true that the microstructure of platelets is controlled significantly by processing parameters and the optimum microstr ucture of these such as an exfoliation in polymer is one of the crucial factor s to obtain high barrier properties. In this work, however, as screw speed varies at the same feed rate, no significant relation ship between screw speed and oxygen permeability w as observed. In contrast, oxygen permeability of PLA homopolymer was sensitive to processing conditions O ptimiz ing processing parameters wa s more critical when working with the PLA homopolymer than when working with the PLA/MMT nanocomposite. Apparently, variations in screw speed did not affect dispersion of platelets in PLA nor did it influence final properties of blown film nanocomposites.
71 S election of polymer is another consideration for compatibility with clay platelets and dispersion of clay platele ts. Bec ause packing densit ies of polymer chains is an intrinsic characteristic that cannot be modified, selection of polymer should be considered carefully for high barrier applications. P ermeability of s tyrene butadiene rubber (SBR) nanocomposite is mainl y influenced by fractional free volume and tortuous diffusion path effects attributed to the clay platelet morphology  P ositron annihilation lifetime spectroscopy (PALS) suggests a useful method to probe the free volume between polymer chains whi ch contributes to its barrier properties. The study on effect s of free volume on barrier properties of SBR nanocomposites by Stephen et al.  reported that the gas barrier properties of nanoplatelets filled latex membranes were enhanced due to platelet like morphology and high aspect ratio of layered silicates The free volume in the latex membranes decreased when l ayered silicates were embedded and this phenomenon was explained using the confinement of the chain segments. From the perspective of free volume, PVA is a potential ly useful material for preparation of PCN. Because t he large inter and intramolecular cohesive energy resulting from the highly polar hydrogen bonding makes PVA a material with excellent barrier to oxygen modified PVA or EVOH hav e been used as a promising matrix material to achieve high barrier properties. S trong hydrogen bonding and high degree of crystallinity of PVA significantly reduces excess free volume which is used as a source to form channel s for jumping of permeant gas. However, a significant drawback that oxygen permeability is increased in elevated humidity has been an obstacle for the use of PVA as a barrier application. The moisture in atmosphere is absorbed at high porosity sites weakening the cohesive energy of pol ymer by breaking secondary bonding. As a result, oxygen molecules diffuse into the structure and easily increas e permeability. Thus, adding clay platelets to the polymer may be an efficient way to maintain and increase barrier
72 properties of water soluble polymer in humid conditions Yeun et al  enhanced barrier properties by casting PVA/saponite hybrid films onto BOPP and PET. The more clay added in hybrid film, the better barrier properties were attained. Higher amount of clay increased the degree of crystallization due to the role of clay platelets as a nucleation site and increased crystallinity l ed to decreased permeability. Coexistence of intercalated and exfoliated structure in all PVA/CNa hybrid films obtained by a solution casting method, over the full range of clay loading, was observed even though there were some differences found in the relative amount of intercalated region. The mixed structure in these practical composite films also gives rise to enhanced barrier properties  Low compatibility with polymer of k aolinite caused by the large cohesive energy density resulting from the intrinsic formation of hydrogen bonding between consecutive layers was modified by intercalation of dimethylsulfoxide, methanol and octadecylamine As a result, EVOH/kaolinite nanocomposite with wellintercalated/exfoliated structures was obtained by a simple melt blending. Because of the extremely low permachor value of EVOH, oxygen transmission rate of EVOH films is very low As a result, the permeability of al l prepared nanocomposite was below the detection limit of the instrument (i.e. below 105 (cm3 m)/(m2 day atm) ) even at high temperature  Barrier properties of PVA/CNa hybrid films coated on PET substrates using PVA with 8789% of degree of saponific ation (DS), PVA with 99% of DS, and modified PVA respectively, were investigated under various relative humidity ( RH ) conditions  Modified PVA, a terpolymer used in the work consist s of 97 wt% of VA, 2wt% of VAc and 1wt% of itaconic acid as illustra ted in Figure 243. Permeability of terpolymer at 0 %RH showed the lowest value and the higher VAc content contained, the higher permeability observed, as shown in the results of permeability of Figure 244. The author explained this in terms of the hydrog en bonding effect While stronger hydrogen
73 bonding by COOH on itaconic acid unit of terpolymer than OH on vinyl alcohol unit induced larger cohesive energy, lower VA contents reduced the number of hydrogen bonding resulting in lower cohesive energy. 10 w t% of CNa in terpolymer decreased the oxygen permeability below 0.001 cc.mil/m2Three di fferent clay platelets, Hexadecylamine MMT (C16 MMT), dodecyltrimethyl ammonium MMT (DTA MMT), CL25A were dispersed in two different polymer matrices, PU and PLA respectively, by solution intercalation method [80,81] For both cases, permeability reduced linearly with increases in clay content regardless of clay types. However, while C16MMT reduced the permeability significantly when comparing to other two clay types, DTA MMT and CL25A in PU matrix, there was no significant differences in the degree of r eduction of permeability among the three types of clay in PLA matrix. This might be explained by the better compatibility of C16 MMT with PU polymer chains. In other words, C16MMT that has good compatibility with PU reduced the permeability by the reduction of the transport cross section and the increase in tortuous path for gas molecules through the completely dispersed platelets. .day at 55 %RH due to two effects, enhanced cohesive energy caused by stronger hydrogen bonding and extended diffusion path induced by highly exfoliated clay platelets. The barrier properties based on geometry of clay platelets According to the theoretical barrier model, permeability of a PCN material is a function of the volume fraction and the aspect ratio of clay platelet under a general assumption that the flow direction of diffused molecules is perpendicular to the platelets. To understand the actual permeability values, orientational orde r parameter  the concept of lateral spacing and effective volume  random or regular array model  and other factors [91,92] have been developed. To relate the aspect ratio of platelet to barrier properties of PCN, Kadanoff cell was used under the same general assumption as made previously  To construct a barrier cell, at
74 least one vertex should be occupied with a platelet to the direction of permeation path as shown in Figure 245. The critical value of clay content for minimum permeability was obtained from the summation of probabilities that the cell is a barrier by iterative relation. And also, in this work, three main factors affecting barrier properties, aspect ratio (or degree of exfoliation, length/thickness (L/w)), orientational o rder parameter ( S), and dispersion distance were included in barrier probability as shown in Eq uation ( 215) This equation also suggest s that the theoretical prediction of barrier properties conformed well to the experimental critical thresholds as illustrated in the plot of various kinds of platelets of Figure 2 45. c cp L w S 1 2 3 ( 215) w here PcFour different types of clay, hactrite (Raponite RD, Kakuhachi Ind. Co.), saponite ( Smecton SA, Topy Ind. Co.), MMT (Kunipia F, Kunimine Ind. Co.) and synthetic mica (DM clean A, Topy Ind. Co.) were used to prepare polyimide (PI)/clay hybrid films  As a result, synthetic mica with the largest aspect ratio of 1230 nm showed the lowes t permeability. This result conforms to the theoretical expectation from Nielsen models [9093,96] that relative permeability should be smaller as aspect ratio becomes lar ger. represents the critical value of c lay platelets for minimum permeability. This suggest s a good indicator for aspect ratio controlled volume fraction in order to estimate the permeability of PCN materials. Good agree ment was also seen with the result of barrier probability  as sh own in Figure 245. From the comparison between theoretical and experimental results it was concluded that the critical threshold of clay platelets with larger aspect ratio was smaller than platelets with other aspect ratios in reduc ing the oxygen permeab ility. T he state of dispersion of clay types in PI/clay PCN specimens is another challenge to be considered for high barrier
75 solutions. While MMT and synthetic mica hybrid has no peak, hectrite and saponite hybrid showed small peak representing the possibility of aggregated stacks of particles because weak ionic bonding between a protonated amine in the modifier and clay surface resulted in the detachment of organic modifier molecules from the surface during heating process. The state of dispersion and rel ative length of clay mainly determined a range of relative permeability in an order from the large st to the smallest: hectrite, saponite, MMT and synthetic mica  Figure 246 illustrated TEM micrographs for various PI/Clay hybrid films. Cussler insist ed that based on his theoretical model and brief experiments platelets with small aspect ratio could act as a shunt, which allows fast transport across that part of a film containing one layer of platelet  This fact also supports the importance of a spect ratio for the design of barrier materials. Different types of barrier structure It is well known that a preferred orientation in the microstructure of PCN along the shear direction can be induced during processing and there exists a critical concent ration to obtain highly ordered microstructures for highly concentrated nanocomposites [39,57,60,61] Oxygen permeability of PCN is at its lowest value when the orientation of platelets is perpendicular to the direction of diffusion of gas molecules becaus e perpendicularly orientated platelets spontaneously maximize their function to block the diffusion of gas molecules during shear process And it is true that the OTR value of a film decreases to 1/5 and 1/2400 of the pure film, for a 5 wt% and 95 wt% cont ent of inorganic clay platelet a ccording to Nielsen s tortuous path model Based on these two ideal concepts, several researchers have started to study the ideal stratified nematic microstructure of clay platelets for super barrier property [9,65,66] Ebin a  prepared a flexible transparent clay film by casting aqueous dispersion of synthetic saponite as a clay platelet and carboxymethyl cellulose sodium salt and PAA sodium salt as the binder additive s Oxygen transmission rate ( OTR ) of resulting clay film is 0.1 cc (20 m)/m2.day.atm
76 and there have been no polymer films that have comparable gas barrier properties. Similarly high clay volume nanocomposite films were obtained with Na+saponite and SPA as a binder additive. Resulting specimens obtained by the same experimen tal procedure have similar barrier properties  However, gas barrier performance of these clay films was deteriorated under humid condition because hydrophilic clay platelets were swollen and subsequently densely stratified nematic morphology was collapsed Like PVA based PCN films, barrier properties can be damaged under humid conditions. A variety of techniques used to reduce oxygen permeability under high relative humidity have been developed by several research groups and one of the most common meth od used for commercial packaging film is to sandwich the material having high barrier property between two exterior materials which help maintain the barrier property and good looking exterior aesthetics Figure 247 shows the laminate d structure of commer cial packaging products. For instance, the commercial food packaging film KuraristerTM The limitation of clay film can be overcome with epox y clay fabric film composite developed by Triantafyllidis  which is a similar laminating structural composite. The resulting films consist of interior non swellable inorganic clay barrier phase and modified organoclay phase. Cured epoxy exterior layers on both sides protect clay barrier phase inside. Oxygen permeability of these specimens was lowered by 2 3 orders of magnitude from the polymer films and lowered by 3 4 orders of magnitude from the clay films. T he dramatic enhancement in barrier propertie s of clay composite films was independent of both aspect ratio C developed as a laminated structural composite by Kuraray Inc. Japan focused on its properties of easy printing and additional coating. The structure of Meal Ready to Eat (MRE) also consists of several layers to sustain its good barrier properties under various conditions such as radio frequency sterilization, microwave processing and high pressure pasteurization.
77 of the cla y and the nature of the exchang ing cation s at the external surfaces of the film. Rather, barrier properties of resulting films were enhanced by the filling of voids between platelet e dges This contribution factor for high barrier property different from normal PCN material caused some deviation from the extended barrier model based on Nielsen s detour theory. The m ore appropriate barrier model for this material was suggested by Beall  According to Beall s barrier model, the polymer at the interface of the particle filler determines the overall barrier properties of the composite. However, t he flexibility became a trade off for such unprecedented oxygen barrier property.
78 Table 2 1. Four different formulations to prepare polymer clay composites [ 45] Polymer matrix Clay Process F1 HMW PDMS I.44P (4 wt%) 1. Intensive batch mixing, 2. Twin screw extruder F2* LMW PDMS (Silanol terminated) I.44P 1. immediate curing with continuous sonication 2. Twin screw extruder F3 PP I.30E Intensive batch mixing F4 PPMA I.30E 1. Intensive batch mixing, 2. Twin screw extruder *Low viscosity newtonian polymer to minimize Table 2 2. Average spherulites radius (Rsph) and Tc crystallized from the melt as a cooling rate of 1o Sample C/min  R sph T c ( o C) HDPE HDPE/CL15A (5 wt%) HDPE gMA HDPE gMA/CL15A (5 wt%) HDPE g MA/CL15A (10 wt%) 7.6 4.5 7.1 4.4 4.3 122.7 123.3 119.0 123.0 121.0 The average dimensions of the spherulites were determined from the Hv patterns by means of the equation, Rsph m the wavelength of the light in a medium of refractive index n, m the angle between the incident and the sc attered beams corresponding to the maximum intensity.
79 Table 2 3. Oxygen permeability of HDPE/clay nanocomposites at different clay concentrations and ultrasonic amplitudes  sample Clay (%) Amplitude Feeding rate (g/s) Oxygen Permeability (cm3day m-2 ) 1 2 3 4 5 6 7 8 9 10 11 0 2.5 5 10 2.5 2.5 2.5 5 10 2.5 2.5 0 0 0 0 5 7.5 10 10 10 10 10 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.5 0.25 75.9 76.2 76.4 79.7 76.8 80.6 69.7 71.2 73.3 65.9 61.8 Table 2 4. Processing conditions for blow n film  Sample name Screw speed (rpm) Feed rate (g/min) Die temp. ( o C) Neat no.1 Neat no.2 Neat no.3 Neat no.4 Nano no.1 Nano no.2 Nano no.3 Nano no.4 80 110 130 80 80 110 95 80 40.5 40.5 40.5 40.5 40.5 40.5 40.5 40.5 175 175 175 175 165 165 165 165
80 Figure 21. Phase diagrams of polymer gr= 0.2, Ngr= 5 (the lowest value of Ngr suggested in Ref.), Points represent calculated coexistence densities; lines serve as guide to the eye. Dashed lines represent approximate locations of pha se transition boundaries (exact calculation is impossible because one or both coexistence points lie in the region > 0.6). parameter between the surfactant and the polymer  Figure 2 2. Expectation of arrangements of modifiers in the intergallery regions: Schematic diagram for tilt angle measurement of organic modifier, (I)  ; typical arrangements of organic modifier in the intergallery space, (II); dependence of tilt angle of modifier molecules on layer charge density of high charge 2:1 phyllosilicates., (III)  ; the formation of monolayers and bilayers in low charge 2:1 clay minerals depending on Ac and Ae in terms of d spacing, (IV) 
81 Figure 2 3. WAXRD patterns of the orga nically modified nanoclays and their nanocomposites and corresponding TEM micrographs  Figure 24. WAXRD Pattern of (a) HDPE/MMT Nanofil 15 and (b) HDPE/HDPE gMA/MMT Nanofil 15 nanocomposites and corresponding TEM micrographs  (I); a TEM micro graph of PP/MAPP/I 30.E prepared by twin screw extruder  (II); a TEM micrograph of the same nanocomposite with (II) prepared by intensive batch mixing and twin screw extruder  (III)
82 Figure 25. WAXRD Patterns of (a) SAN/MMT and (b) EVOH/MMT at each reaction time and TEM micrograph of the final morphology of corresponding com posites  Figure 26. Illustration showing distribution of organic MMT layers in a cross section of nylon66/organic MMT nanocomposites. The black lines represent the i ntersections of MMT layers, and the background is the polymer matrix  Figure 27. Illustration of peeling mechanism for the exfoliation process 
83 Figure 28. Shear induced orientation of platelets with the flow (x), gradient (y) and vorticity ( z) directions, as indicated is illustrated. A beam direction and general SANS patterns to show the effect of shear on the orientation are added to the figures  Figure 2 9. TEM micrographs showing PP/clay elongated at 150 C with (a) elongation flow rate from 1.0 s Illustration shows the stretching direction and the orientation of clay platelets which is perpendicular to the stretching direction 
84 Figure 2 10. TEM mic rographs for EVA/SIOM, EVA/DIOM, EVA/TRIOM, (I); WAXRD patterns for organoclays (II) and corresponding PCN (III)  (inset figures in each TEM micrograph represent schematic diagrams for each structure) Figure 2 11. Illustration of the morphological hierarchy at different length scales extant in the PET/CL30B showing the dispersion of intercalated/exfoliated aggregates throughout the matrix and the local orientation ordering of the sheets within the aggregates and TEM micrographs of (a) a global scale exfoliation, (b) intercalated/exfoliated sheets at high magnification in exfoliated aggregates and (c) fully exfoliated structure in cross linked PET/CL30B nanocomposites 
85 Figure 2 12. V ariation of dspacings of the MAPE/CL20A and MAPP/CL20A nanocomposite fibers depending on volume fraction of clay; Open circle symbol and open rectangular symbol indicate d spacings of MAPE/20A and MAPP/20A calculated from SAXS data, r espectively. Solid symbols mean d spacing created by shear induced orientation  Figure 2 13. A beam configuration of SAXS measurement and resulting SAXS photographs of MAPE/CL20A nanocomposite fibers with clay concentration of (a) 6, (b) 12, (c) 24, and (d) 42 vol % and MAPP/20A nanocomposite fibers with clay concentration of (e) 6, (f) 12, (g) 24, and (h) 42 vol %, extruded at shear rate of 1000 s 1. Order parameter (S) of MAPE/CL20A and MAPP/CL20A fibers with the concentration of silicates was shown between two series of SAXS photographs 
86 Figure 2 14. Y and Z beam configuration of SANS measurement and resulting SANS patterns of both configurations of films consisting of PEO/LRD with lower aspect ratio or PEO/CNa with higher aspect ratio  Figure 2 15. Schematic diagram to represent the procedure for the preparation of a cured epoxy clay fabric film composite with with exfoliated clay nanolayers at the outer surfaces of the clay film 
87 Figure 2 16. (a) TEM of the nanocomposite viewed along x showing layeredsilicates oriented orthogonally to the BCP lamellae. PS la yers appear light due to OsO4 staining of the PB layers. The high magnification inset shows finely spaced black and white fringes indicative of the intercalated nature of the layered silicates embedded in the PS domains, (b) paralleled oriented BCP lamella e and TEM image viewed along x of the nanocomposite showing layered silicates oriented parallel to the BCP lamellae. The small TEM micrograph is a high magnification image showing the exfoliated nature of the layered silicates, shown as single dark lines, surrounded by the thicker PS domains  Figure 2 17. Schematic picture to represent the fundamental concept for the use of anionic polymer as a binder additive to achieve a clay film of densely packed structures Figure 2 18. (a) TEM image of a cross section of the saponite film with 20 wt% of carboxymethyl cellulose sodium salt and corresponding schematic microstructure, (b) Schematic microstructures of the clay films of saponite/polyacrylic acid sodium salt film 
88 Figure 219. (a) TEM i mage of the film with a polymer loading of 20 wt%, showing the c axis direction (crosssection), (b) Schematic crystal structure of saponite with  face in cb plane, (c) Schematic crystal structure of saponite with  face in ab plane, (d) TEM image of the film with a polymer loading of 20 wt%, showing the abplane (surface) and the corresponding SAED pattern (inset)  Figure 2 20. Schematic representation of the internal architecture of the PVA/MTM nanocomposite (left side) and (A) Cross section of the film, (B) Close up of the cross section showing the separation of layers, (C) Topdown view of a fracture edge of the composite after tensile testing, (D) Top down view of the composites surface  Figure 2 21. DSC crystallization exotherms plots of PE g MAn and I.44PA/PE g Man nanocomposites record ed at 20 oC/min (left) and XRD patterns for PE g MAn and I.44PA/PE g MAn systems depicting 110 and 200 reflections of polyethylene (right) 
89 Figure 222. Optical micrographs and Hv patterns of: ( a) HDPE g MA, (b) HDPE gMA/CL15A (10 wt.%), and (c) HDPE/CL15A (5 wt.%)  Figure 223. TEM images of PUCCN with 5% C (a and b) and PULCN with 3% L (c and d), (I); AFM images of PUCCN (a and b represent height and phase image with 5% C) and PULCN (c and d represent height and phase image with 3% L), (II) 
90 (a) (b) Figure 224. (a) Bright field TEM images of the extruded polystyrene/montmorillonite pellet A: room temperature, B: after heating at 85oC, C: 95oC, D: 110oC and corresponding in situ WAXD patterns between room temperature and 200oC  (b) expected orientation of phenyl ring with respect to clay particle after heating at 95 o C. Figure 225. TGA result for normal PVA/organoclay PCN (left,  ) and crosslinked PET/CL30B PCN (right,  ) Figure 2 26. T hree common particles geometries used in models
91 NielsenVolume Fraction 0.00 0.05 0.10 0.15 0.20 Relative Permeabilty (Rp) 0.0 0.2 0.4 0.6 0.8 1.0 50 100 200 500 1000 2000 Aspect Ratio Cussler Regular arrayVolume Fraction 0.00 0.05 0.10 0.15 0.20 Relative Permeability (Rp) 0.0 0.2 0.4 0.6 0.8 1.0 50 100 200 500 1000 2000 Aspect Ratio Cussler_Random arrayVolume Fraction 0.00 0.05 0.10 0.15 0.20 Relative Permeability (Rp) 0.0 0.2 0.4 0.6 0.8 1.0 50 100 200 500 1000 2000 Aspect Ratio Figure 2 27. T heoretical effects of volume fraction on relative pe rmeabilities, Upper curve calculated from Eq uation ( 23) and lower two curves of the Cussler model were drawn from Equation ( 24) and ( 25 ). (Redrawn from  )
92 GusevVolume Fraction 0.00 0.05 0.10 0.15 0.20 Relative Permeability (Rp) 0.0 0.2 0.4 0.6 0.8 1.0 50 100 200 500 1000 2000 Aspect Ratio FredricksonVolume Fraction 0.00 0.05 0.10 0.15 0.20 Relative Permeability (Rp) 0.0 0.2 0.4 0.6 0.8 1.0 50 100 200 500 1000 2000 Aspect Ratio Figure 2 28. Theoretical effects of volume fraction on relative permeabilities, Left curve calculated from Eq uation ( 26) of Gusev model and right plot of the Fredrickson model were drawn from Eq uation ( 27). (Redrawn from [91,92] ) Figure 229. A unit of three silicate pl atelets representing the arrangement of platelets in whole system of idealized PCN structure (left); effective volume of a clay platelet was represented as dot line (right) Figure 230. Effect of confinement of clay platelets on polymer chainsegment immobility and resulting relative permeability depending on volume fraction of clay, two figures are drawn and calculated based on Equation ( 212 1=1 for left unconfined 1=1.25 (empirical value suggested in  ) for right confined matrix were used. (Redrawn from  )
93 Lateral spacing (b) 0 20 40 60 80 100 d-spacing (H) 0 2 4 6 8 10 12 14 16 18 20 Relative Permeability (Rp) 0.0 0.1 0.2 0.3 0.4 0.5 d-spacing at = 50 d-spacing at =1000 Rp at =50 Rp at =1000 Figure 2 31. E ffect s of lateral spacing on d spacing and relative permea bility at different aspect ratio of platelets; this plot is based on Eq uation (2 12) ( = 0.05) (Redrawn from  ) Figure 2 32. C omparison of the changes in H spacing as b varies between larger and smaller platelets was presented (drawn from the conce pt of  and  ); upper figure represent occupancy of larger particles with effective volume of small spacing of b and when b increases, H has to be decreased to fill up the same volumetric space as shown in middle figure. In the case of smaller partic le, H decreases drastically for the same amount of increases in b as represented in the bottom figure.
94 Figure 2 33. S chematic figure to depict the variance of S values depending on platelets orientations. (Redrawn from  ) Bharadwaj _Fully Exfoliated (S=0)Volume Fraction 0.00 0.05 0.10 0.15 0.20 Relative Permeability (Rp) 0.0 0.2 0.4 0.6 0.8 1.0 50 100 200 500 1000 2000 Aspect Ratio Bharadwaj _Orientation EffectS (Orientation Parameter) -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 Relative Permeability (Rp) 0.0 0.1 0.2 0.3 0.4 0.5 0.6 50 100 200 500 1000 2000 Aspect Ratio BharadwajVolume Fraction 0.00 0.05 0.10 0.15 0.20 Relative Permeability (Rp) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 S = -1/2 at a = 50 S = 0 at a = 50 S = 1 at a = 50 S = -1/2 at a = 2000 S = 0 at a = 2000 S = 1 at a = 2000 Figure 2 34. E ffect s of volume fraction on Rp when fully exfoliated structure (upper), two curves to depict the effect of S on Rp calculated at v olume fraction = 0.05 using E quation ( 214) (Redrawn from  )
95 Figu re 2 35. E ffect s of volume fraction on Rp for present barrier models calculated at aspect ratio of 50 ( upper ) and 1000 (lower )
96 Figure 2 36. Comparison of prediction for Rp as a function of aspect ratio; all cu rves were plotted based on the E quation 22, 23, 24, 25, 2 6, 211, and 213 at = 0.05 Figure 2 37. Illustration of effective width and its effect on barrier property of PCN (redrawn from [63,96] )
97 Figure 2 38. OTR chart of a pure PET sheet and PC N sheets (left) and corresponding TEM micrographs of MMT PCN (a) and MDEA III PCN (b)  Figure 239. TEM micrographs to represent the difference in tortuosity (refer to dot lines) between phase separated microstructure of unmodified MMT PCN and exfo liated one of modified MMT PCN (left) and relative permeability of two PCNs and Neilsons theoretical values (right) 
98 Figure 240. O xygen relative permeability of composite depending on clay content (wt%) and TEM micrographs of PU/CL UE400S00; (a) low and (b) high magnification of 1wt% clay and those of PU/CL UE400S60; (c) low and (d) high magnification of 5wt% clay. (Red dot lines r epresent the diffusion path through the structure of composites)  Figure 241. TEM micrograph of a compression molded specimen containing 5 wt% of clay filler dispers ed in polyesteramide matrix  Sample Clay Content (wt%) R p PU/CL UE400S00 1 3 5 0.827 0.718 0.714 PU/CL UE400S60 1 3 5 0.756 0.639 0.586
99 Figure 2 42. Oxygen permeability rate of various blown composite/neat film samples  Figure 2 43. Chemical structure of poly (vinyl alcohol co vinyl acetateco itaconic acid) where, x, y and z is 97, 2 and 1, respectively  PVA contenets (x)*Relative Humidity (% RH) 0 20 40 60 80 100 O2 Permeability (cc.mil/m2.day) 0 1 6 87-89 (wt%) 99 (wt%) 97 (wt%) Terpolymer/CNaClay loading (wt%) 0 5 10 15 20 O2 permeability (cc.mil/m2.day) 0.00 0.02 0.04 0.06 0.08 0.10 35 %RH 55 %RH Figure 2 44. E ffect s of PVA contents on oxygen permeability of neat polymer containing no fillers (left) and effect of clay loadings on oxygen permeability of terpolymer/CNa composites; x represents the weight percent of vinyl alcohol group as shown in chemical structure in Figure 243 is the product of film thickness and oxygen transmission rate (OTR) measured at 23C (pl otted based on results from  )
100 Figure 2 45. C oncept of probability of a ba rrier cell using Kadanoff cell (left) and Critical volume fraction versus aspect ratio of platelets with S= 0. Thresholds for several typical clay fillers are obtained directly from Eq uation ( 215) based on their aspect ratios  (open symbols). Two sol id symbols indicate the test data of O2 gas permeability in C18montmorillonite (L/W= 150 and c=1.25 ( Ray et al., 2003) ) and polyester clay (L/W =200, and c =1.4  ) nanocomposites (right)  Figure 2 46. TEM micrographs of various PI/clay hybrid films  Figure 2 47. Commercialized structural composite to enhance oxygen barrier property (a) KuraristerTM C developed by Kuraray Inc. Japan; and (b) Composite structure developed for US army MRE case
101 CHAPTER 3 POLYMER CLAY NANOCOM POSITE COATINGS ON NONPOLAR POLYOLEFIN SUBSTRATE TO ENHANCE BARRIER PROPERTY Polymer clay nanocomposite ( PCN ) have been widely studied as property enhancers for polymers for mechanical strength [1,114] thermal stability  and barrier properties [116,117] PCNs are sometimes referred to as hybrids of inorganic fu nctional fillers and are an area of intense research activity. Dispersion of nanoparticles in polymers results in three typical structures that are classified as non intercalated, intercalated and exfoliated as explained in Chapter 1 Exfoliated structure s involve complete separation of parti cles into random arrangements [118,119] Exfoliated structures are necessary to obtain maximum benefits from nanoparticles. Exfoliation requires relatively high shear forces to separate particles in solution and compl icates processing and production of PCNs [112,120] O nce the fully exfoliated structure is obtained, maintaining the exfoliated structure without particle aggregation is another challenge in PCNs. Additionally, obtaining fully exfoliated PCN in nonpolar polyolefin polymers has been difficulties because silicate layers of clay are polar and therefore, incompatible with polyolefin. Significant efforts are underway to overcome this limitation [7,121,122] Introduction In this study, Laponite JS (Southern Clay Products, L ouisville, KY) and polyvinyl alcohol were used to produce polymer nanocomposite coatings as inorganic filler and a hydrophilic polymer matrix respectively Laponite JS is a synthetic layered fluorosilicate modified with an inorganic polyphosphate dispers ing agent. Figure 31 shows SEM micrographs of Laponite JS powder and a schematic structure of Laponite JS. This clay is hectorite, prepared in a reaction between Mg, Li, and Na silicate salts, which results in partially crystalline, monodispersed in size discs, 0.92 nm thick and ca. 25 nm l ong (aspect ratio of 2555), with a bulk density of 950
102 kg/m3 and specific surface area of ca. 300 cm2Po ly(vinyl alcohol) ( PVA ) is a water soluble polymer with linear chains. Figure 3 2 shows the representative chemical structure of general PVA grades Depending on the degree of saponification ( DS ) of PVA, which is the degree of conversion to PVA through the saponification process, the physical properties of PVA varies significantly. PV A is known to be nontoxic and is recognized to be generally recognized as safe (GRAS) by the United States Food and Drug Administration (FDA). PV A is recognized to be a biodegradable polymer and has found uses as a coating material, stabilizer, polarizing film, sizing agent, adhesive, drug delivery system, contact lens material, basic bio material with expanding applications in nano technology [123,124] P V A alone offers excellent oxygen barrier propert ies; however, ethylene vinyl alcohol ( EVOH) is often the preferred polymer used in multilayer packaging film structures. /g  As shown in scanning e lectron m icroscopy ( SEM ) micrographs in Figure 31, Laponite powder exhibited typical agglomerated and grained features of layered silicate particles. It hydrates and disperses in water to give virtually clear and colorless colloidal sols of low viscosity. Therefore, Laponite colloidal solutions of up to 18 wt% are stable for one month. Polymers with e xcellent oxygen barrier properties inherently have high crystallinity and large cohesive energy caused by highly polar hydrogen bonding Nanocomposites involving clay nano particles in PV A have been studied in many laboratories [29,78,125128] While wet, PV A nanocomposites demonstrate excellent particle dispersion in PV A gels. This state rep resents a true nano scale organic inorganic hybrid material. However, drying causes portions clay layers to reaggregate. Steric constraints of PV A impede complete reaggregation of clay, allowing some to remain in the dispersed state 
103 Nano composite preparation techniques are designed to create amorphous domains with uniformly distributed mineral layers, but this approach is limited as the preparation of PV A /clay nanocomposites from solutions is difficult because of reaggregation of clay  To pre vent reaggregation, Yeun, et. A l  prepared PV A /Saponite nanocomposites with various clay concentrations successfully. They reported that oxygen permeability values decreased with increasing clay loading within a 0 10 wt% range, while optical propertie s remained nearly constant. Another major limitation of PV A as a nano composite matrix polymer can be caused by the polar nature of the hydroxyl ( OH) groups that make PV A hydrophilic and soluble in water. Small amounts of water, particularly relative hum idity above about 35%, plasticize PV A resulting in dramatic loss of oxygen barrier properties [130,131] A variety of techniques used to reduce oxygen permeability under high relative humidity have been developed by several research groups. One of the most common is sandwiching the water sensitive polymer between hydrophobic layers such as poly propylene ( PP) and/or polyethylene ( PE ) [24,132] However, laminated structures are expensive and may suffer from poor flexibility. P P is an attractive packaging mate rial because it is readily available low cost and offers good overall mechanical properties. However, PP suffers from low oxygen barrier properties. Improvement of barrier properties of PP without creating expensive laminate structures would greatly enhan ce the usefulness of the polymer. Therefore, this work focused on development of techniques to produce and apply nanobased barrier coatings to PP the substrate Critical to this work was enhancement of bonding between the hydrophibic substrate and the hy drophilic, PV A /Laponite coating. In this work, atmospheric pressure plasma (APP) treatments were used to enhance bonding of coatings.
104 Materials Experiments A P CN solution was made with synthetic layered silicate known as Laponite JS (Southern C lay P roduct s Louisville, KY) To evaluate effects of molecular weight (MW) and degree of saponification on coating adhesion and oxygen barrier, PV A provided by Celanese Company ( Dallas, TX ) and Scientific Polymer Product C o. ( Ontario, NY ) with MW and DS shown in Tabl e 31 were used. Organic ammonia chloride (OAC) ; [2 (Acryloyloxy)ethyl] trimethyl ammonium chloride was used for intergallery modification which can increase the intergallery distance by replacing swollen sodium ion with organic molecules to make it easier for polymer chains to be diffused into this space as shown in Figure 33. (For more explanation, refer to Chapter 1, Introduction) Preparation of PCN solutions was conducted through the four steps as presented schematically in Figure 3 4. OAC buffer solution of 1 wt% of clay was added into a corresponding amount of deionized water at 40 Step 1 : Prepar ation of Polymer Clay Nanocomposite Solutions oC and pH 10 were premixed for 23 minutes with a magnetic stirrer. Laponite JS was added gradually to prevent aggregation. Prepared clay solutions were placed at room temperature for less than a day to obtain good ionexchange reaction in the intergallery of clay platelets. Polymer stock solutions were prepared by dissolving PV A in deionized water at elevated tem perature and stirring with magnetic stirrer for at least 6 hours. Prepared PV A stock solution was added to clay solutions and mixed for 1 hour with a magnetic stirrer and then mixed with a high shear mixer (KadyMillL Kady International, Scarborough, ME) Prepared PCN solutions were placed in a hood at ambient temperature for 310 hours depending on viscosity in order to allow entrained air bubbles to dissipate. Table 3 2 outlines the compositions of all prepared PCN solutions for coating in this study.
105 AP P treatments were performed using a DyneA Myte VCP (Enercon Industries Co., Menomonee Falls, WI). Air and nitrogen gas (about 50/50) were used to create plasma. Figure 3 5 illustrates the four controllable va riables when treating the surfaces These variables are flow rate of nitrogen gas, working distance between the tip and the surface of target specimen, treatment times and belt speed. Table 3 3 lists specific conditions used in this study. S tep 2: Atmospheric Pressure Plasma Treatment Step 3: Coating Samples after APP Treatment Surfaces of rigid PP were spin coated with 20g PCN solution at fixed spincoating parameters involving a twostep process to obtain uniform thickness of coated layer First step was 500 rpm for 1.5 min. Second step was 700 rpm for 30s. The spin coater used was a Model Laurell WS 400B 6NPP/LITE (Laurell Technologies Co., North Wales, PA). Step 4: Drying PCN C oated PP specimens PCN coated PP specimens were dried at ambient temperature in vacuum desiccators to remove possible ai r bubbles for 10 12 hours Morphological effects of APP surface treatments were analyzed using Model JSM 6400 SEM ( JSM6400, Jeol Ltd., Tokyo, Japan) Samples were washed with acetone and then surfaces were sputter coated with gold pal ladium alloy (Au Pd). Micrographs of submicrondetailed surfaces were obtained under ambient conditions using atomic force microscopy (AFM) (Digital Instruments Dimension 3100,Veeco Instruments., Plainview, NY) by contact mode. Oxygen transmission rates of PCN coated samples were measured in accordance with the procedure described in ASTM D 3985 using a Model OX TRAN 2/20MH (Mocon Corporation, Minneapolis, MN). Permeation cell area was 50 cm Characterizations 2. Degree of exfoliation was measured by analyzing changes in inter gallery spacing of clay platelets. The change in spacing was measured
106 using an Wide Angle X Ray Diffractometer ( WAXRD) (Philips XRD APD 3720 powder diffractometer, Philips Electronics, Mahwah, NJ) with a Cu anode ( CuK radiation ( = 1.54056 ; scanning range from 5 to 69.99). Clay powder was mounted on a sample holder with a large cavity and a smooth surface was obtained by pressing the particles with a glass plate. Samples for PCN solution were prepared by spin coating using a silicon wafer as a substrate. The changes in the basal spacing at the small angle range from 2o to 10oResults and Discussion of clay platelets for LP20PVA, LP40PVA, LP50PVA and LP70PVA w ere investigated by using XRD ( Philips MRD X'Pert System Philips Electronics, Mahwah, NJ) with a CuK radiation PCN coated PP specimens were able to be used directly without a sample holder. Surface Modification by Atmospheric Pressure Plasma Treatment Degree of surface modification caused by APP is determined by several important factors such as plasma energy source, which can be classified regarding the excitation mode, exposed area energy density, and the collision time and intensity of the electron energy source  In this research, air and nitrogen gas were used as a plasma source and the flow rate of nitrogen gas and the distance between tip and surface of specimen were controlled to investigate effects of APP treatment on surface modification as well as resulting oxygen barrier property. Resulting SEM micrographs of APP treated and untrea ted PP at 6k magnification are shown in Figure 36 ( a d) Enhanced roughness was observed microscopically depending on the APP treatment parameters. Generally, t he harsher condition such as short working distance (WD) or higher N2 flow rate resulted in mor e rough surfaces. As shown in Figure 36 (b), some pattern on the surface was observed along the treatment direction and this pattern was disappeared as WD was
107 getting closer to the surface. However, no significant effects of APP were observed for WD of 3c m or greater This can be checked by comparing Figure 36 ( c ) and (d) Surface topography of nitrogen plasma PP substrate measured by AFM on a 20 x 20 m2 area is shown in Figure 3 7. Nontreated surfaces in Figure 37 (a) appeared to be flat and smooth w hile APP treated surfaces appear to have surface contours (Figure 37 ( b) ). Figure 3 8 and 39 shows the effects of APP parameters on the surface roughness. These figures are the 3 dimensional height images taken by AFM on a 50 x 50 m2For quantitative analysis of surface roughness of the AFM images, R area of (a) untreat ed surface and from (b) APP treated by App Cond01 to (k) App treated by App Cond10 as listed in Table 33. Apparently more surface contours were observed in (c) (e) and (h) (j) when comparing others. This could be caused by the fact that the harsher AP P conditions make plasma jet to be excited more by obtaining more energy subsequently resulting in better etching effect on the surface. a, Rmax and Root Mean Square (RMS or Rq) values wer e measured Ra and RMS are both representations of surface roughness, but each is calculated differently. Ra is calculated as average roughness (Equation 31) of measured microscopic peaks and valleys. RMS is calculated as the root mean square of measured microscopic peaks and valleys (Equation 32). Each value uses the same peak and valley dimension measurements. A single large peak or flaw within the microscopic surface texture will affect the RMS value more than the Ra value. Rmax dx x z L RL a 0) ( 1 is defined as a maximum height representative of a difference between a highest point and a lowest point  ( 31) 2 1 0 2) ( 1 L qdx x z L R ( 32) Where, L = evaluation length, Z(x) = the profile height function
108 The largest RMS value of 132 nm was obtained at 18 l/min nitrogen (F igure 310). No additional roughness was measured at greater nitrogen flow rates. Therefore, 18 l /min nitrogen was used for APP treatments for this study. The increase in surface roughness by APP treatment is caused by the surface activation involving gra fts of active chemical functions onto the surface  Working distance was the distance between the plasma jet tip and the sample. WD was determined to be critical for effectiveness. Figure 311 shows increased surface roughness at WD of 20 mm. Degree of exfoliation depends on how platelets are modified and how well the intergallery is opened. Because of the small intergallery distance, high shear processes are needed to achieve exfoliation of nanoparticles. Each platelet peel s away fro m a stacked clay layer. WAXRD patterns of pure Laponite JS and PCNs are shown in Figure 312. As a silicate material, Laponite exhibits low angle peaks, much like natural MMT clay. Two high angle peaks were also observed for Laponites JS at about 48 degree s and 62 degrees. Therefore, according to Braggs law (d=n /2sin ), d = 1 x 1.54 / 2 sin (48/2) = 1.89 and dspacing for 62 degree = 1 x 1.54 / 2 sin (62/2) = 1.50 These dspacings should correspond to the lattice spacings of the crystalline unit cells. The two typical peaks of Laponite JS at 48 and 62 degrees disappeared on curves in Figure 312 for exfoliated LP10PVA, LP30PVA and LP50PVA because of the exfoliation of the silicate or high disorder of the clay platelets  However, as shown in Figure 3 13, a sharp strong peak at 48 degrees was shown at the unexfoliated LP20PV A abbreviated as LP20PV A UE, which means that silicate layers were not delaminated nor dispersed. XRD patterns of PCN showed no peak between 40 and 70 degrees and signal inte nsities showed a broad hump between 60 and 63 degree for LP30PV A and LP50PV A in Figure 3 12. This can be explained by Clay Exf oliation
109 exfoliated silicate layers of the clay mineral showing characteristic XRD pattern of amorphous material when platelets are randomly disper sed. This pattern is a characteristic XRD pattern of a clay exfoliated type of polymer clay nanocomposite  The more clay added in PCN, the more distinctive the hump observed by XRD. However, if the clay platelets are fully exfoliated, dspacing canno t be detected by the XRD method because the dspacing would become too large. X ray diffraction patterns of coated samples were also obtained at low angle range from 2 to 10 degree as shown in Figure 314. As shown in this figure, the peaks for (001) plane around 5 to 6 degree were disappeared for LP20PVA, LP40PVA and LP50PVA indicating that exfoliation of clay was achieved for these three samples. However, a broad peak which is shifted to lower degree from the (001) peak position was observed for the sampl e containing higher amount of clay. Partial exfoliation which is a mixture of partially exfoliated and intercalated structures was occurred for LP70PVA because the amount of polymer is not enough to approach the complete exfoliation of all clay re sulting i n some aggregated particles. P VA as a coating material for food packaging materials to reduce oxygen permeability has several advantages in terms of its structural properties as described in Chapter 2 The large inter and intramolecular cohesive energy resulting from the highly polar hydrogen bonding provided PV A with excellent oxygen barrier properties at low relative humidites. Due to the combination of strong hydrogen bonding and high degree of crystallinity, PV A has lower oxygen transmission rate ( OTR ) values when comparing to PET by three orders of magnitude Barrier Properties Clay platelets with large surface area and high aspect ratio were expected to show increased oxygen barrier properties in much the same way as increasing crystallinity of the polymer. Figure 3 15 indicates OTR was a function of clay concentration. Greatest barrier properties as indicated by lowest OTR values were obtained at 50 wt% of clay concentration in
110 the final dried film. Between 20 to 40 wt% of clay concentration, no significant difference in OTR values was observed. Additionally, dramatic increases in OTR values were observed for loadings in excess of 50 wt%. This is due to high degrees of reaggregation of clay resulting in reduced effective surface area per unit concentr ation as well as lower degrees of exfoliation c aused by steric interference [7,44,63] Once high degrees of exfoliation are obtained, oxygen barrier properties are dominated by adhesion strength between PCN coated layer and the surface of substrate. There fore, it was found that quality of surface treatment plays a significant role in determining extent of barrier achieved with PCN barrier coatings. Figures 316 and 317 illustrate effects of parameters of APP treatment on OTR values. Similar tendencies we re observed as a result of surface roughness. Figure 3 16 illustrates effects of nitrogen gas flow rate on the OTR values. Figure 316 shows that flow rate of nitrogen gas did not affect OTR, which was consistently in the range of about 10 cc/m2DS of PV A used for making PCN solut ions was important. L ow DS PVA obtained from (Scientific Polymer Product co.) resulted in relatively poor control barrier coatings (0 wt% clay). High DS PVA obtained from Celanese resulted in good barrier properties with OTR values on the order of 10 cc/m /day. Figur e 317 shows effects of working distance on surface roughness. Working distance was shown to be critical. Poor results were seen when the plasma head was too close or too far away from test specimen. 2/day. Decreasing DS l eads to relatively fewer number of OH groups comparing to that of high DS as well as larger number of bulky side group which remained from precursor Therefore, the combination of strong hydrogen bonding and high degree of crystallinity might be lower resu lting in poor barrier property. M W of PV A did not appear to be significant. OTR values of PVA coated PP specimen were given in Table 3 4.
111 Table 31. Specific data of PV A samples Grade DS MW Company Celvol107 99.3% + 31,000 50,000 Celanese Company Celvol4 25 95.5 96.5% 120,000 150,000 Celanene Company PV A PH 88% 120,000150,000 Scientific Polymer Product Co. 1) High DS: 99.3 +, 2) Fully saponified: 93 98, 3) Intermediate saponified: 9193, 4) Partially saponified: 88 Table 3 2. Compositions of PCN solutions used for coating Sample name PVA grade (MW/DS) Clay (wt%) High shear process LP0PVA 31,000 50,000/99.3%+ 0 O LP10PVA 31,000 50,000/99.3%+ 10 O LP20PVA EX 31,000 50,000/99.3%+ 20 O LP20PVA UE 31,000 50,000/99.3%+ 20 X LP30PVA 31,000 50,000/99 .3%+ 30 O LP40PVA 31,000 50,000/99.3%+ 40 O LP50PVA 31,000 50,000/99.3%+ 50 O LP60PVA 31,000 50,000/99.3%+ 60 O LP70PVA 31,000 50,000/99.3%+ 70 O
112 Table 33. Specific conditions of APP treatment Condition no. N 2 ( l /min) flow rate WD (mm) T reatment times Belt speed APP Cond01 25 20 4 20 APP Cond02 20 20 4 20 APP Cond03 18 20 4 20 APP Cond04 15 20 4 20 APP Cond05 12 20 4 20 APP Cond06 20 30 4 20 APP Cond07 20 25 4 20 APP Cond08 20 20 4 20 APP Cond09 20 10 4 20 APP Cond10 20 5 4 20 Table 3 4. OTR results of PVA coated PP specimens PVA APP condition (refer to Table 3 3) OTR value (cc/m2/day) MW DS (wt%) 150,000 88 8 APP Cond09 119.34 150,000 95 8 APP Cond09 8.05 50,000 99 8 APP Cond09 8.12
113 Figure 3 1. SEM micrographs of Laponite JS powder (a) showing agglomerated feature as indicated as dotted arrows and grained region in a circled region and (b) layered features as shown in a dotted rectangle. ( c) schematic structure of Laponite JS. Laponite JS consists of platelets containing Mg, Li, Si, O and Na cations in intergallery regions. Average aspect ratio of Laponite JS is between 25 and 55 and its bulk density and specific surface area is 950 kg/m3 and ca. 300 cm2 /g respectively
114 Figure 3 2. A chemical structure of PVA. Depnding on x and y, DS of PVA is determined Figure 3 3. Schematic mechanism showing the formation of ideally exfoliated PVA/Laponite JS nanocomposite solution through the ordered experimental procedures
115 Fig ure 3 4. Schamtically illustrated experiment procedures Figure 3 5. Controllable variables when APP treating the surface of i PP
116 Figure 36. SEM micrographs taken at 6000X of (a) untreated surface, (b) APP treated surfaces by APP Cond06, (c) APP Cond01 and (d) APP Cond09
117 Figure 3 7. 3D surface and height images taken by AFM (contact mode) on a 20 x 20 m2 of untreated surface (a) and APP treated surface by APP Cond09 (b)
118 Figure 3 8. Influence of the flow rate of nitrogen gas which is one of the APP parameters; 3D surface images taken by AFM (contact mode) on a 50 x 50 m2 of untreated surface (a) and APP treated surface by APP Cond01 (b), 02 (c), 03 (d), 04 (e) and 05 (f)
119 Figure 39. Influe nce of the WD which is one of the APP parameters 3D surface images taken by AFM (contact mode) on a 50 x 50 m2 of untreated surface (a) and APP treated surface by APP Cond06 (g), 07 (h), 08 (i), 09 (j) and 10 (k)
120 Figure 3 10. Influence of the flow rate of nitrogen gas which is one of the APP parameters on surface roughness Figure 311. Influence of WD which is one of the APP parameters on surface roughness
121 Figure 312. WAXRD patterns of the pure Laponite JS powder and several PCNs having exfoliate d silicate layers in PV A matrix Figure 3 13. Difference in WAXRD patterns between LP20PVA PCN samples prepared with and without high shear process (300) ( 220) With Shear Process Without Shear Process
122 Figure 3 14. WAXRD patterns of several PCN coated surface containing various amount of silicate par ticles obtained at low angle range from 2 to 10 degree Figure 315. Dependence of the content of clay loaded in PCN on OTR values (001)
123 Figure 316. Effect of the flow rate of nitrogen gas on OTR values Figure 317. Effect of the working distance on OTR values
124 E FFECT OF PH ON MICRO STRUCTURE OF LAYERED SILICATE AND BARRIER PROPERTIES IN NANOCOMPOSITE COATING SYST EM S CHAPTER 4 Introduction P roperties of polymer clay nanocomposite ( PCN ) materials can be enhanced by altering the characteristics of filler particles due to competitive interaction s between attractive forces of polymer chain s and silicate layers and electrostatic forces between silicate layers. O ne of three microstructures of silicate platelets in a polymer matrix (phase separated, intercalated, and exfoliated) usually dominates when a composite is formed. E ach microstructure can be tailored by adjusting a variety of factors such as silicate particle types [7,42,45] volume fraction of particles [57,60,69] and the processing techniques used [45,46,56] T he aim of t his study was to optimize barrier properties of PCN films by designing and engineering the microstructure of silicate particles. According to the detour models of permeability  parallel platelets are required to be perpendicular to the permeation direction to achieve superior barrier materials. However, in real nanocomposite system s crystallization characteristics of the polymer [137,138] degree of dispersion / exfoliation [63,69,96] and orientation of clay particles [63,96,139] will hinder ideal barrier routes. To minimize these possible hindrance factors and extend attainable detour length in composite materials, a dense stratification of silicate platelets using a polymer as a b inder material to be attached to the edges of p latelets was devised. The silicate particles used in this study was Laponite JS The concentration of sodium ions over the whole surface range of the Laponite JS platelet s in the intergallery regions was constant prior to addition of platelets into water Initially w hen clay platelets were added to the water, the clay powder form ed the aggregated particle stacks These stacks were separated and swollen by the hydration of sodium ions ( Figure 4 1)
125 Hydrated sodium ions distributed in the intergallery are affected by two forces; electrostatic attraction with a negatively charged platelet surface and osmotic pressure caused by differences in chemical potential between intergallery phase and aqueous phase. T he osmotic pressure applied to sodium ions at the edge of the crystal will be stronger than that at the center region. Furthermore, when two crystals approach, their mutual positive charges repel each other. Therefore hydration of sodium ions and mutual repulsive forces open intergallery regions lowering the osmotic pressure. As a result, sodium ions at platelet center are attracted more to the surface. Therefore this interaction resulted in concentration gradient of sodium ions over the surface of platelets proving that attractive forces of the platelet edge have been weakened and therefore displayed a relatively weaker positive characteristic due to lack of sodium ions ( Figure 42 (a) ) To halt further diffusion of sodium ions and reach a stable equilibrium, addition of polar compounds is required A ddit ion of polar compounds result s in a relatively weaker positive ions being attracted to the negative surfaces of adjacent particles forming a unique structure ( F igure 42 (b) and ( c ) ) As a result of a series of these successive interactions in the system, a house of card s structure consisting of weak ionic bonds is formed. The microstructure of clay platelets formed by obtaining a house of cards structure of dispersed clay platelets in a water clay polar compound system will collapse when water molecules evaporat es with an orientation of clay platelets in  direction D ense stratified arrangement of clay platelets maintaining this orientation and stronger ionic bonding among platelets may be required To maintain this unique collapsed structure and f ill microvoids, a polymer tha h attaches at specific locations as a filler and linker is required. Ogoshi and Chujo  proposed that optically transparent organic inorganic polymer hybrids with anionic polymers could be prepared by controlling pH in an
126 a queous solution. They also reported that the structure of the anionic polymer poly(acrylic acid) ( PAA) could be controlled by varying the pH due to the electric repulsion between polymer chains. Gudeman and Peppas  also found that the swelling ratio of poly(vinyl alcohol) ( PVA) /PAA interpenetrating network can be adjusted by varying pH value in an aqueous solution. In this study PAA wa s used as a filler polymer. Anionic PAA can be prepared by placing PAA chains in a basic solution As a result, PAA chains are extended by mutual electric repulsive force s between chains as the side chains are negatively charged (COO ). These PAA chain s should be short enough to fill the microvoids. The driving force of PAA chains into microvoids is electric attraction ( F igure 43( a ) ) PAA chains are attracted to the center of the platelets as the positive ions at the edge are relatively weak and already bonded with adjacent crystals forming a house of cards structure. Weak shear forces helped the short PAA chains dif fuse into the structure and, with time reorientation of clay platelets remov es shear forces. The d rying process and appropriate degree of external pressure lead s to dense stratified clay structure filled with small amount s of PAA chains. Therefore anioni c PAA attach es two crystals by electrostatic force and the ideal barrier structure may be obtained by forming densely stratified structure ( Figure 4 3 ( b) ) In this study, the microstructure of clay plate lets for better barrier propert ies was devised based on the concept suggested by Ebina  and Tetsuka  The effect of pH on the microstructure of clay particles and barrier properties of resulting composite coated films as well as visual evidence revealed during experimental procedures have been explo red using MOCON, s canning e lectron m icroscopy (SEM) t ransmission e lectron m icroscopy ( TEM ) f ourier t rnasform i nfrared ( FTIR ) spectroscopy and x r ay diffractometer ( XRD )
127 Experiments Preparation of Polymer Clay Nanocomposite Solutions at Various pHs A house of cards structure was identified by investigating the microstructure of clay aerogel. At first, a clay hydrogel was prepared from 15 wt% clay aqueous solution with ammonium chloride and sodium tetraborate as additives after aging for 24 hours. A cla y hydrogel was frozen in 2L glass lyophilization shells at 173K ( 100 oC) and a frozen clay hydrogel was sublimed using a labconco Freeze Dryer 8 Half filled shells were rotated at 30rpm at a temperature of 80 oTo prepare composite coating solution, first of all, low molecular weight PAA (Degree of Polymerization (DP) of 25) wa s dissolved in distilled water and the pH of the mixture wa s controlled using aqueous HCl (1 M) or NaOH (1 M) solution. PAA chains under three different conditions (acidic, basic and intermediate condition) were prepared to compare effect s of pH on the filling void and stratifying clay platelets. An amount of Laponite JS based on weight ratio of clay to PA A in the final dried film was added into deionized water gradually to prevent aggregation. Samples were premixed for 12 minutes with a magnetic stirrer. P olar compound, ammonium chloride (NH C until frozen and t he ice from frozen shel ls was then evacuated for 36 hrs 4Cl) buffer solution, at 1 wt% (according to the amount of clay us ed) wa s added to solution to maintain a house of card structure by lowering osmotic pressure in the intergallery. Prepared PAA stock solution wa s added into the clay mixture and mixe d gently for 1 hour with a magnetic stirrer. A prepared mixture of water clay, polar compound, and PAA wa s degassed in vacuum to minimize structural defects by removing remaining flocculated or aggregated clay impurities. S pecific conditions for preparation of nanocomposite solutions are listed in Table 4 1. A s chematic of ex perimental steps is represented in Figure 4 4.
128 Coating Samples After Atmospheric Pressure Plasma Treatment The optimum conditions for a tmospheric p ressure plasma ( APP ) treatment w ere chosen based on the previous results  These conditions were 18 l/min of nitrogen gas flow rate, 20 mm of working distance, and 4 treatment times. Composite solutions were coated on isotactic (i ) polypropylene ( PP) film after AP P treatment to e nhanc e adhesion. Surfaces of rigid PP were spin coated with 20g coating solution at fixed spin coating parameters involving a twostep process. First step was 500 rpm for 1.5 min. Second step was 700 rpm for 30s. The spin coater was a Model Laurell WS 400B 6NPP/LITE (Laurell Technologies Co., North Wales, PA). Samples were dried at ambient temperature in a vacuum desiccator. Characterizations and A nalysis C hain conformations of anionic PAA in composite material were identified by Diffuse Reflectance Infrared Fourier Transform (DRIFT) Spectroscopy (Thermo Electron Magna 760, Thermo Sc ientific Inc., MA) The structure of the clay film was characterized by WAXRD (Philips XRD APD 3720 powder diffractometer, Philips Electronics, Mahwah, NJ) The change in spacing was measured using an X ray diffractometer with a Cu anode ( CuK radiation = 1.54056 ) with a wide range scanning angle. C lay and clay composite powder were mounted on a sample holder with a large cavity ; a smooth surface wa s obtained by pressing particles with a glass plate. A clay composite sample prepared by rod coating on a silicon wafer was compared with the composite powder to investigate effect s of coating shear direction on orientation of clay platelets. X ray diffraction patterns were obtained at the s mall angle range from 2o to 10o to investigate effects of polymer load ings on microstructure s of clay platelets using WAXRD ( Philips MRD X'Pert System, Philips Electronics, Mahwah, NJ) with a rotation anode and CuK radiation PCN coated PP specimens were able to be used directly without a sample holder.
129 In order to analyze the morphological characteristics of clay aerogel and surfaces of composite films those samples were sputter coated with goldpalladium alloy (Au Pd) and observed using SEM ( JSM 6400, Jeol Ltd., Tokyo, Japan) TEM (TEM 200CX, Jeol Ltd., Tokyo, Japan) micr ographs of clay film showed microstructure of clay platelets. TEM was operated at an accelerating voltage of 200 kV. The dispersion liquid was dropped directly on a copper grid supported with a collodion membrane. Oxygen transmission r ate ( OTR ) values of c oated s pecimens were measured in accordance with the procedure described in ASTM D 3985 using a Model OX TRAN 2/20MH (Mocon Corporation, Minneapolis, MN) to investigate barrier properties Results and Discussion A House of Cards Structure A house of cards structure of Laponite JS clay powder in the aqueous solution was identified by invest igation of the microstructure of clay aerogel Figure 45 shows the changes in morphology from the clay powder to clay aerogel. As shown in SEM micrographs ( Figure 45 (a) and (b) ) the structure of nanoclay powder as received wa s a randomly stacked aggregate while the clay aerogel ( Figure 4 5 (c) and (d)) show ed regular microchannel s in the range of 5 20 m in the bulk. This structure is representation of a house of c ards formation mechanism ( Figure 4 2 and 43 ) M orphological features and formation of house of cards structure have been reported by several researchers [62,89,95,142] Corresponding TEM m icrographs are shown in Figure 46. Aggregated platelets are clumpi ng together because of the van der Waals attraction in dry condition showing a large dark spot containing several dark lines which represent edge stand ing platelets. However, clay platelets were arranged with three features tilted, edge standing, and paral lel ed platelets to the surface through the formation of house of cards structure as illustrated schematically in Figure 4 6 (c) These three orientations result in confined
130 spaces for short PAA chains that are negatively charged to be diffused into microvoid s by ionic attraction toward the center of the surface of platelets rather than platelet edge s Effects of pH on the Clay Platelet Microstructure It has already been reported that PAA changes depend ing on pH in an aqueous solution  PAA chains unde r basic condition can be expanded by electric repulsi ve force according to the dissociation of ionic groups. Therefore, it was necessary to estimate the range of molecular weight s of PAA to obtain a best fit on microvoids which clay microstructure produce s. It was assumed that PAA chains are fully extended under ba sic condition due to electric repulsion among adjacent carboxylate anions, endto end distance ( = nl where n is the number of repeat unit and l is the bond length) can be used to estimat e an approximate molecular weight of PAA that can diffuse into the space of the densely stratified structure of clay platelets resulting from the collapse of the house of cards structure Average surface diameter of Laponite JS is 55 nm and bond length of C C is 0.15 nm and therefore n should be in the range of 1 and 336, which means that low molecular weight or oligomer of PAA is required. The chain conformations of anionic PAA in composite material were identified by DRIFT Spectroscopy as shown in Figu re 47. The peaks at 1550 cm1 and 1710 cm1 are for C = O stretching vibration of the carboxylate anion ( COO-) and carboxylic acid group ( COOH) respectively. In the case of PAA under basic condition, relatively higher peak at 1550 cm1 w ere detected, whi le the much higher peak at 1710 cm1 for PAA under acidic condition w ere observed. U nder acidic condition, short PAA chains will be entangled together rather than extended showing a critical concentration which will not allow PAA chains to diffuse into the clay platelets and this was confirmed by experimental evidenc e Aggregated white small particles were dispersed in a composite gel as shown in F igure 48 in the case of the samples prepared under acidic and isoelectric point (IEP) conditions of PAA after 3 weeks storage.
131 These particles can be formed by two possible reactions. The first potential reaction is salt extraction caused by the sodium cat ions that diffuse from the intergallery region and dissociated chlorine anions from added ammonium chloride a dded. Second reaction can be a phase separation of entangled PAA chains due to the higher intra and intermolecular hydrogen bonding in acidic conditions. To examine which reaction occurred in the system at various pH levels SEM samples w ere prepared by drying th e gel like sample on a glass slide at 70 oTo investigate the specific differences in interspacing of particles amo ng samples depending on pH and the relative amount of PAA in the composite system, three plots were obtained from WAXRD results as presented in Figure 410. These results confirmed the possibility of aggregated PAA particles observed during experiments. A l l samples showed a C for 1 day. The white particles were mounted on the SEM sample holder using a sharp tweezers Some small residual dried clay particles were shown adjacent to these larger particles and there are some coated layers along the surface of these particles ( Figure 49 ) P articles sizes in acidi c condition were larger tha n ones in Isoelectric Point ( IEP ) condition because there were more favorable interactions between PAA chains at lower pH resulting in larger siz es of particles of 400 600 m. Energy dispersive X ray spectroscopy ( EDS ) result (Table 42 and Figure 49) show ed that the white particle portion ha d greater amount s of C and O and relatively small amount s of Si atoms. O ther spots on broken dried clay gel particles and coat ing laye r show ed higher intensity of Si and relatively weaker peaks of C and O. T herefore, phaseseparated PAA particles are aggregated and dispersed in a clay gel. There w ere no particle s in basic conditions of PAA after 6 weeks storage which proves that diffusi on of short PAA chains to the surface of clay platelets had occurred Therefore particles formed as a result of the phase separation of entangled PAA chains caused by effects of pH.
132 similar increase in d spacing regardless of the amount of PAA used. Red lines represent positions of peaks for pure Laponite JS powder  As shown in Figure 410 (a), the interplanar distances for (110) and (300) planes, correspondin g to peak positions of 36.5 and 62 degrees respectively, showed some increases for all pH conditions, while the characteristic peak for Laponite JS at 47.6 degree showed much larger increase under basic conditions U nder acidic conditions no peak for (005) plane at 27.5 degree was found. Among these three samples having the lowest PAA contents, the largest overall increase in interplanar distances at pH 6 for all planes was observed. The peaks for (005) plane decreased as the amount of PAA increased as presented in Figure 410 (b) and (c). M ore PAA content resulted in disappearance of regularity of particle orientation by the insertion of PAA into the intergallery region. In the case of acidic conditions of (b), there wa s a significant increase for (100) plane from 20.4 to 16.3 degree s which means the distance increased from 4.35 to 5.44 according to the Bragg s law. This may be explained by the fact that entangled short PAA chain s inserted into platelets and enlarged the spacing. pH 4 and 7 showed s imilar increases in interplanar distance of most planes. L ess than critical amount of PAA has no effect on interplanar distances and subsequent microstructure of clay platelets. There was no change for (100) under acidic and IEP c onditions when more PAA co ntent were used ( Figure 410 (c)) because more PAA increases possibility of forming phase separated PAA chains resulting in aggregates that are too large to be inserted between platelets. Under basic condition, the similar tendency for increase of each int erplanar spacing is shown for all PAA content except for (005) and (110) planes. To investigate changes in position of (001) peak s depending on the amount of PAA at bas ic condition, X ray diffraction was obtained at lower angle range s from 2 to 10 degree s ( Figure 4 11) As shown in this figure, the peak
133 observed in C5PA91BS disappeared at C5PA73BS and reappeared at C5PA55BS. This might be explained by a larger amount of oriented clay platelets in  direction result ing in a peak for C5PA91BS that contain s more clay However, this peak disappeared as the amount of polymer increased showing similar characteristic peak s with exfoliated samples ( C5PA73BS curve of Figure 4 11) Finally this peak was shifted to a lower degree due to some diffusion of PAA chains into intergallery Because of the relativ ely small aspect ratio compared to natural montmorilonite ( MMT ) or other clay types, it would be easier for the structure of Laponite JS to collaps e by diffusion of polymer. Therefore, randomly collapsed structure of C5PA55BS could result in microvoids and subsequently showing poor barrier properties. Figure 412 shows TEM micrographs of samples prepared under basic condition (a) and acidic conditions (b) of PAA. Figure 49 (a) shows that there wa s no apparent feat ure of edge standing clay platelets or tactoids under basic conditions of PAA, while several black lines in (b) represent edge standing clay tactoids were separated from entangled PAA chains. E ffect s of shear direction on orientation of clay platelets for coating process es w ere shown in Figure 413. Overall intensities of C5PA91BS_c (coated layer by a coating rod) were much lower than those of C5PA91BS_p (clay composite powder) and all peaks except for (005) plane disappeared. This suggests that clay platel ets were oriented in the direction that was parallel to the shear direction as illustrated in Figure 413, which conforms to the relation of shear direction with relative orientation of clay platelets reported [39,60,95] This shows that the coating proces ses with inherent shear direction such as rod coating, doctor blade, or calendaring promise orientation of filler particles in the composite system resulting in better barrier properties Barrier Properties Barrier properties of prepared composite solutio ns with PAAs of different pH values were investigated by measuring OTR of prepared film s from each solution. From our previous work
134  it was reported that significant increases in OTR were observed for loadings in excess of 50 wt% This is due to high degrees of re aggregation of clay as well as lower degrees of exfoliation caused by steric interference. The amount of clay in this composite system is over 50 wt% and, according to our previous results, significant decrease in barrier properties should be observed regardless of pH of polymer used. However, C5PA91BS which contains 90 wt% of clay in the system showed good barrier properties of 9 10 cc/m2Figure 415 shows SEM micrographs of composite coated i PP surfaces. Figure 415 (a) and (b) shows small crack s i n the coating surface and this lowered barrier properties of specimen dramatically by allowing gas molecules to contact to the substrate directly. T he more PAA in the composite solution such as C5PA73 and C5PA55 added, the more cracks on the surface were observed, with associated high OTR value s over 100 cc/m /day which is similar with OTR of composite solution coated specimens having lower clay content with good exfoliation. This lower OTR value could be approached by densely stratified structure s as shown in Figure 414. T he difference in microstructure of clay platelet resulted in different barrier properties at the same amount of clay. 2 /day. Lower pH also has significant effects on barrier properties. The poor interactions between negatively ch arged PAA under lower pH and clay platelets squeezed polymer out from the microvoids and microchannel and result ed in many cracks on the coated surface ( Figure 415 (c) and (d) ). Figure 4 16 illustrates OTR results depending on pH, which conf i rm morphologi cal features of the surfaces of coated layers.
135 Table 4 1. Specific conditions for the preparation of each specimen Table 4 2. EDS results for white unknown particles dispersed in clay composite solutions Element Spot 1 Spot 1 2 Spot 2 Spot 3 Elmt % Atom % Elmt % Atom % Elmt % Atom % Elmt % Atom % C K 15.08 21.97 25.06 35.64 26.24 41.95 4.46* 8.38* O K 47.75 52.23 36.26 38.71 11.00 13.20 12.64 17.83 Na K 2.64 2.01 1.50* 1.11* 3.57 2.98 29.81 29.25 Mg K 23.61 16.99 20.55 14.43 13.22 10.44 15.19 14.10 Si K 10.92 6.80 16.62 10.11 45.97 31.42 37.90 30.44 Total 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 = <2 Sigma #. Sample name Clay (wt%) PAA pH Ratio (by weight) 1 C5PA91AS 5 2 9:1 2 C5PA91IS 5 4 9:1 3 C5PA91BS 5 6 9:1 4 C5PA73AS 5 2 7:3 5 C5PA73IS 5 4 7:3 6 C5PA73BS 5 6 7:3 7 C5PA55AS 5 2 5:5 8 C5PA55IS 5 4 5:5 9 C5PA55BS 5 6 5:5
136 Figure 41. S tructure of silicate layers, Laponite JS when dispersed in water phase
137 (a) (b) (c) Figure 4 2. S chematic mechanism of a house of cards structure (a) initial stage as a powder stat e (b) the formation of concentration gradient of sodium ions in the intergallery regions and the addition of polar compounds and (c) the formation of a house of cards structure
138 (a) (b) Figure 4 3. (a) S chematic mechanism of a house of cards structure through the diffusion of LMW anionic PAA chains into a clay structure and (b) e xpected ideal barrier st ructure
139 Figure 4 4. Schematically illustrated experiment procedures
140 Figure 45. SEM micrographs of (a, b) Laponite JS clay powder and (c, d) corresponding clay aerogel
141 Figure 4 6. TEM micrographs of clay aerogel consisting of t hree structural features, edge standing, tilted and paralleled as indicated as arrows
142 Figure 4 7. DRIFT results of composite samples of C5PA73 prepared at pH 2, 4, and 6 Figure 48. Photographs of the states of composite solutions prepared under different pH conditions; (a) pH 2 (b) pH 4, and (c) pH 6 after 3 weeks storage
143 Figure 4 9. Picture of C5PA91IS solution after 2 weeks storage and SEM micrographs of its by product, white particles dispersed in the solution
144 Figure 410 E ffects of the pH on the peak positions of XRD results depending on PAA contents. T he positions of peaks for pure Laponite JS powder were indicated
145 Figure 4 11. E ffects of the amount of PAA on the changes in basal spacing ((001) peak position) of PCN sam ples prepared at basic condition Figure 412. TEM micrographs of C5PA73BS (a) and C5PA73AS (b)
146 Figure 4 13. XRD results of clay comp osite powder (C5PA91BS_p) and corresponding coated layer by a coating rod (C5PA91BS_c)
147 Figure 4 14. E ffect of clay platelet microstructure on barrier properties when the same amount of clay was arranged (a) as a randomly dispersed state and (b) as a stratified state. Red dot lines show the diffusion path of oxygen gas molecules through the PCN material Figure 415. SEM micrographs of nanocomposite coated surfaces (a), (b) C5PA91BS (c) C5PA91IS (d) C5PA91AS
148 Figure 4 16. D ependence of permeability on pH and the amount of PAA for highly filled PCN coating system
149 THEORETICAL ESTIMAT ION OF BARRIER PROPE RTIES OF POLYMER CLA Y NANOCOMPOSITE CONTAI NING TWO CLAY TYPES "PC2N" CHAPTER 5 Introduction Polymer clay nanocomposite ( PCN ) are attractive for packaging materials because of superior barrier properties such as gas permeability and water vapor tr ansmittance when compared to conventional polymer materials. As explained in previous chapters, barrier properties of PCN are affected significantly by the arrangement and orientation of platelet s in the polymer matrix Gas permeability in PCN is a functio n of the aspect ratio, volume fraction, and relative orientation of anisotropic nanoclay filler in the polymer matrix. Impermeable platelet arrangement and/or aspect ratio will result in the extension or reduction of the diffusion path as suggested by Niel sen  Recently i mprovements in the gas barrier properties of PCN especially at low concentrations of clay platelets (below 5 wt%) have been widely reported and it is important to use model to predict the optimized barrier properties. Several models [ 31,9193,96] which are based on a tortuous path of diffusants have been proposed to relate permeability of PCN to volume fraction, particle dimension and arrangement of filler particles. To date all models predicting the barrier properties of PCN are based on using one nanoclay type. In order to obtain super barrier properties which cannot be prepared by using one nanoclay in PCN, new types of microstructure are required. The author anticipated that b y inserting smallsized clay particles in the intergaller y region of large platelets, the tortuous path, which diffusants need to travel around, is expected to be maximized. Recently, the same type of microstructure consisting of two different clay types was suggested by Stefanescu et al [144,145]. This work proved that the mechanical behavior of PCN could be controlled by adjusting the ratio of Laponite (small aspect ratio) to montmorilonite ( M MT ) (large aspect ratio) in PEO matrix. Stefanescu et al. also
150 proposed an idealized orientation for two clay types. How ever, no model has been proposed to predict the idealized micro structure and subsequent optimized barrier properties for PCN of two clay types. In this study, estimation of relative permeability ( Rp) of PCN consisting of two clay particles having different aspect ratios (PC2N) is modeled by modifying existing tortuosity based models in order to explore the benefit of using two different particle types and compared with experimental RpExperiments results of several PC2N specimens Preparation of PC2N Films I n order t o compare with the barrier model and equation, brief experiments were conducted. Cloisite Na+ (CNa) Cloisite 30B ( C L 30B ) and synthetic layered silicate, Laponite JS (Southern Clay Products, Louisville, KY) were used to prepare a PC2N solution. Ta ble 51 lists the specific information on the clays used. A 7 wt% poly(vinyl alcohol) ( PVA) (Sigma Aldrich co, St. Louis, MO; Mn = 132,000 and degree of saponification = 99%) aqueous solution was prepared as the polymer matrix. Organic ammonium chloride ( O AC ) [2(Acryloyloxy)ethyl] trimethyl ammonium chloride was used for an intergallery modification reaction of the natural clay. Two PCN solutions were prepared following the procedures reported in1622. To prepare a PC2N solution, polymer stock solution wa s prepared by dissolving high molecular weight ( HMW ) PVA in deionized water at elevated temperature and stirring with magnetic stirrer. Laponite powder was added in predetermined amount s of deionized water with vigorous agitation with a mechanical stirrer at room temperature until all powder was well hydrated Aging was required to complete hydration Laponite solution was added to polymer stock solution with agitation. Vigorous agitation with a mechanical stirrer for 10 minutes was conducted to prepare a c ompletely exfoliated Laponite in a PVA matrix. The intergallery of Cloisite platelets was modified with OAC and prepared as an aqueous solution and then added to a PCN solution
151 consisting of PVA and Laponite. Further shearing with a mechanical stirrer was required for the final insertion process. This process was carried out by inserting PVA mixed with the smaller Laponite particles into the intergallery region of Cloisite. All prepared solutions were arranged as films with the size of 4 inch diameter by a casting method using a Teflon petridish. Characterizations O xygen transmission rates (OTR) of prepared film specimens were measured in accordance with the procedure described in ASTM D 3985 using a Model OX TRAN 2/20MH (Mocon Corporation, Minneapolis, MN). Permeation cell area was 50 cm2. Table 52 lists the specific experimental conditions followed to prepare each composite sample and the associated OTR results. RpDiscussion of Model and Results values for samples were calculated by Equation ( 51) with obtained OTR values. Optical trans mittance values were measured using an UV visible photometer with wavelength range of 350 800 nm. PC2N consists of two clay types such as clay 1 a natural MMT with a large aspect ratio and clay 2 a synthetic layered silicate particle with a relatively smaller aspect ratio Figure 51 shows relative dimensions of clays used Synthetic layered silicates are aligned readily as a house of cards structure due to the smaller aspect ratio and their charge interaction along the e lectrical double layers resulting in good exfoliation with relatively weak shear forces [89,118,142] A natural MMT can also be exfoliated or intercalated by the surface modification of nanoclay and subsequent shear processing as reported in many papers [3 9,43,74,118,146] High barrier properties may be achieved by designing the maximum detour length in the microstructure of PC2N composite materials Based on detour theory, Rp of PC2N system is given as Equation ( 51)
152 1 2 2 2 1 12 ) 1 ( 2 H v L L H Reff T P ( 51) Where, L1, L2 and H1 can be defined as lateral length including edge to edge distance, ( l1+b1) and ( l2+b2) of clay 1 and 2 and vertical length (h1+w1Figure 52 represents how the ratio of volume fraction of clay 2 to clay 1 affects R ) occupied by c lay 1 per unit volume respectively. p of PC2N system which is calculated from Equation ( 51) Rp decreases linearly as total volume fraction of filler increases but ther e is no apparent difference among the five ratios showing a completely overlapped line. The effect of increased da caused by an increase in the quantity of clay 2 on Rp might be diminished by the lower quantity of clay 1 platelet in the system. There was n o difference in Rp values when one clay type systems and two clay types systems were compared. Therefore the fact that RpFigure 53 show s R of PC2N can be determined by total volume fraction of two fillers can be confirmed theoretically, the model and the equation establish ed in this paper is important for predicting the relationship between permeability and microstructure of particles in the matrix. p for various composite systems with the ones as shown in Table 52. OTR value of HMW PVA film was 34.5 cc/m2/day and Rp o pP P R values of all composites were calculated using Equation ( 52). ( 52) Where, P and Po are permeability of polymer clay nanocomposites and that of matrix polymer without nanoclay, respectively. P, L, and 30B represent HMW PVA, Laponite JS and Cloisite 30B, respectively. Therefore, PC2N Sample PLCNa consists of HMW PVA, Laponite
153 JS and CNa. The re is no significant change in OTR values when varying the size of particles at the same total fraction of particles for the composite films prepared as expected from theoretical values as shown in Figure 52. Contrary to our expectation, in fact, there wa s no significant effect of inserting smaller nanoclay into the interspacing between larger clay platelets on Rp 2 1: of PCN. However OTR values of composite films containing 30B were lower apparently than those of others composites in Figure 5 3. That is becaus e that larger size particle exhibited better barrier properties as expected in Bharadwaj s model  The reason that both two values for 1:0 and 0:1 of are similar in Figure 52 may be explained the fact that orientation was not in corporated into this model. Also deviations caused by lateral distances, b1 and b2Although extended experiments need to be done with this concern, our preliminary results clearly shows that, t hough higher volume fraction of clay particles le d to lower permeability, the ratio of three components in the PC2N system should be optimized to take advantage of the properties of each clay platelet. For example, while Cloisite gives poor transmittance and relatively heavier weight, it exhibits better barrier properties. Therefore when mixed with Laponite much improved transmittance and reduced weight can be obtained while maintaining similar barrier properties. This fact c ould be confirmed by the results in Figure 54 and 5 5. Generally, properties of a matrix such as the degree of crystallization and the size of spherulites, at lower volume fractions may have affected permeability. Considerations such as the orientation of particles, the possibility of existing aggregated particles in the matrix and actual lateral distances were not considered here Even though there have been limiting factors for accuracy of the model, some consistency between theoretical expectation and real permeability that volume fraction of clay particles in PC2N system is significantly affects permeability rather than the relative size of two particles was observed.
154 the interfacial refractive index difference between matrix and filler particles, and the size of dispersed particles determine the optical transmittance of polymer composites [82,147] The scattering of light by well dispersed particles in matrix which are much smaller than the wavelength of incident visible light will lead to Rayleigh scattering. Therefore Laponite JS with the average aspect ratio of 25 nm was able to produce a near transparent film. However C L 30B with higher aspect ratio can be utilized only for a better barrier material because of its high density and poor optical transmittance. PLCNa which is a PC2N composite shows median transmittance value between those of PL and PNa. Derivation of Barrier Equation for PC2N In order to develop a model to estimate Rp of PC2N, several assumptions were made. First it was assumed that clay platelets have no effect on crystallization of the polymer matrix as assumed in other detour based models The second assumption was that lateral separations (edge to edge distance) b1 and b2 were much smaller than the lengths of particles l1 and l2In order to reduce the complexity of PC2N model, the author also assumed that both clay types maintained a parallel arrangement to the surface of specimen as shown in Figure 51. The direction of diffusion is normal to the surface R respectively as shown in Figure 51. p o pP P R can be defined as permeability of PCN film relative to unfilled film. ( 52) Where, P and Po are permeability of polymer clay nanocomposites and that of matrix polymer without na noclay, respectively. Assuming clay platelets as impermeable crystalline domains in semicrystalline polymer matrix based on Klute s theory  Equation ( 52) can be also defined as Equation ( 53)
155 ) ( ) 1 (T T PR ( 53) Where, T and are total volume fraction of clay particles in PCN and a function for the reduction of permeability due to the nanoclay particles, respectively. As with other researches the authors have made the assumption that the polymer matrix is unconfined. Therefore the permeability is only a function of the volume fraction of filler, i.e. nanoclay, when all the variables are fixed. The total volume of nanoclays, Tv in the polymer matrix is defined as Equation ( 54), 2 1v v vT ( 54) W here 1v is volume of nanoclay type 1 and 2v is that of nanoclay type 2. The authors have assumed as with Bo Xu et al  and Saunder et al  that the shape of a nanoclay platelet is a hexahedron which its length and depth are equal. Therefore total volume of nanoclays is 2 2 2 1 2 1w l w l vT ( 55) And, based on the effective vol ume model suggested by Saunders et al  the effective volume model for this system is modified as illustrated in Figure 56. As shown in Figure 56, the left top quadrant is an effective volume which is dispersed uniformly in the whole system. This uni t space represents the whole composite system, which contains both nanoclay types arranged in layers. Each unit cell is defined by the length of the larger nanoclay platelet. The unit cell cannot be smaller than this dimension. A number of the smaller nanoclay platelets can occupy a layer in the unit cell and the total number of the smaller platelets, n, is dependent of the size of this platelet. T otal volume fraction of nanoclays can be defined as Equation ( 56) 2 2 1 1 2 1 eff Tv N v N ( 56)
156 Where 1 2 are the volume fraction of particles 1 and 2, respectively, and N1 and N2 12 1 polymer Polymer T represent the number of nanoclay 1 and 2 per unit volume within the effective volume of clay 1, respectively. Note that total volume fraction in the composite system can be regarded as Once the fully occupied nanoclay 2 platelets between two larger platelets is regarded as a tactoid which moves together, the volume fraction of clay 2 can be estimated using effective volume of clay 2 instead of the volume of clay 2. The number of clay platelets of type 2 (N2) should be described in a different manner than N1 because effecti ve volume of clay 2 falls within the effective volume of clay 1 ( Veff1 ) ( ) (1 1 2 1 1 1w h b l veff ) by the assumption. Since the effective volumes for clay 1 and clay 2 are calculated from ( 57) ) ( ) (2 2 2 2 2 2w h b l veff ( 58) T otal number of nanoclay per unit volume in the system, NT 2 2 1 1 2 1) (eff Tv v N N N can be written as fol lowing ( 59) Equation ( 510) and ( 511 ) are obtained from Veff and the definition of N1 and N2 1 1 1 1 2 1 1 1 1) ( ) ( 1 1 v w h b l v Neff ( 510) 2 2 1 1 2 1 1 1 2) ( ) (eff effv w h b l nl v nl N ( 511) The tortuous path (d ) per unit volume of diffusant may be e stimated by summing lateral distances that a diffusant must travel due to particles (da) per unit volume relative to the shortest path through unfilled polymer (d) per unit volume The distance of diffusant through the matrix without filler can be calculated from Equation ( 512). The additional lateral distance created
157 when clay 2 particles are embedded into the gap between the layers of clay 1 particles (da2) per unit volume is described in Equation ( 513) and total detour distance (da ) ( ) (1 1 1 1 1 1 1h w v h w N d ) per unit volume cau sed by all fillers in the system may be estimated from the extended term as given in Equation ( 514). Note that the dimension of all tortuous path described here is length/volume and the N is also the number per unit volume. ( 512) 22 2 2 2b l N da ( 513) 2 2 22 2 2 1 1 1 2 1 1 1 1b l N b l N d N b l N da a ( 514) From Equation ( 511 ), nl may be expressed as Equation ( 515) 2 2 2 2 2 2 1 1 2 1 1) ( ) ( ) ( ) (w h b l w h b l nl ( 515) According to the definition of da, maximum values of da promise the longest path distance corresponding lower Rp ) (T values. By the definition of ) ( 2 2 ) ( ) (1 1 2 2 2 1 1 1 1 1 1 'h w b l N b l N h w N d dT ( 516) Therefore Rp 1 2 2 2 1 12 ) 1 ( 2 H v L L H Reff T P of PC2N system is given as Equation ( 517) after combining Equation ( 53) and ( 516) ( 51)
158 Where, L1, L2 and H1 can b e defined as lateral length including edge to edge distance, ( l1+b1) and ( l2+b2) of clay 1 and 2 and vertical length (h1+w1 ) occupied by clay 1 per unit volume respectively.
159 Table 51. Specific information of all clays used for the pre paration of PC2N Cloisite 30B Cloisite Na + Laponite JS d spacing h (nm) 1.85 1.17 0.001 aspect ratio l (nm) 200 1000 75 150 25 avg. aspect ratio* l (nm) 600 112.5 25 w (nm) 1 1 0.92 lateral spacing** b = l/10 (nm) 60 11.25 2.5 D ensity ((g/cm 3 1.9 8 ) 2.86 0.95 *, ** these two values are based on the assumptions of the Rp equation of PC2N Table 52. Specific experimental conditions for the preparation of composite films samples Conc. of PVA ( wt%) 1 ** 2 ** OTR (cc/m 2 /day) P 7 0 0 0 34.500 PL 7 0 0.050 0.050 0.734 PCNa 7 0.050 0 0.050 0.650 P30B 7 0.050 0 0.050 0.488 PLCNa* 7 0.025 0.025 0.050 0.657 PL30B* 7 0.025 0.025 0.050 0.504 *PC2N, **1; Cloisite Na+ or 30B, 2 ; Laponite JS
160 Figure 51. Schematic diagram showing a filled polymer with (a) one clay type and (b) two different clay types (dark rectangle represents clay 1 with the dimension of width of w1 and length of l1 and grey scaled rectangle is clay 2 with th e dimension of width of w2 and length of l2. The separations of each particle are denoted as b1 and b2 for lateral edgeto edge distance and h1 and h2 for face to face distance between the same clay type platelets. The face to face distance between two dif ferent type platelets is denoted as h12. Note that the dot line is the tortuous path for a gas molecule in PCN (P) and thick solid arrow line can be described as the shortest path (d) through unfilled polymer (Po ))
161 Figure 5 2. E ffect of the total v olume fraction of clay2 on the relative permeability of PC2N: obtained by Equation ( 51) from l1= 100 nm, l2= 5 nm, w1 = w2 = 1 nm, b1= l1/10, b2= l2/10, h1= 1 nm, h2 = 0.001 nm Figure 53. Actual relative permeability of various composite films calculated from OTR values measured by MOCON
162 Figure 54. Optical transmittance of composite films at the visible wavelength range (350 800 nm) (A) (B) (C) (D) (E) (F) Figure 5 5. Pictures of prepared transparent and translucent composite film samples: (a) PVA, (b) PL, (c) PLCNa, (d) PCNa, (e) PL30B, and (f) P30B
163 F igure 5 6. P arallel arrangement of two nanoclays in an effective volume: n is the total number of the smaller platelets which can occupy in a layer in the unit cell and l is the number of layers in the unit cell. Other s are defined in the caption of Figure ( 51) in the Discussion Part.
164 EFFECT OF POLY(ACRYL IC ACID) ON BARRIER PROPERTIES OF HIGHLY FILLED NANOCOMPOSITE FILMS CHAPTER 6 Introduction In order to enhance barrier properties, extending tortuous path a pe rmeant molecule should travel is indispensible to polymer clay nanocomposite ( PCN ) This has been approached by dispersing impermeable fillers in polymer. Because extending tortuosity is achievable using completely separated and disper sed nanosized clay fi llers, researches to lower permeabilities of PCN have been focused on obtaining better dispersion and exfoliation of clay platelets in various polymer matrices [46,53,54,148151] However, a PCN in which nanoclay s are completely separated and dispersed has not been satisfied for practical barrier applications Triantafyllidis et al reported a preparation of an epoxy clay fabric film composite with significant oxygen barrier propert y enhancements via a partial ion exchange reaction  Oxygen permeability o f resulting specimens was lowered by 2 to 3 orders of magnitude when compared to a pristine polymer film and by 3 to 4 orders of magnitude when compar ed to a pristine clay film. Because of an intrinsic nature of thermoset epoxy resins it was difficult to obtain a good film flexibility. Achieving film flexibility and super barrier properties can be obtained by designing a new microstructure of clay platelets using negatively charged organic molecules as a binder additive. As introduced in Chapter 2, Ebina e t al  successfully prepared flexi ble transparent clay films with high barrier properties These films were made based on a hypothesis that significant reduction in barrier propert ies can be obtained by maximizing tortuous path. D ue to highly dense self laminating micro structure s of clay platelets, maximized tortuous path and good flexibility w ere able to be realized. One challenge for PCN design proposed by Ebina is that clay films can be readily damaged and its shape can be distorted as a result of min or contact under humid conditions due to swelling of natural montmorilonite ( MMT ) clay by water molecules. W ater content in clay
165 films is important as it will contribute to better flexibility Therefore a critical quantity of water molecules is required to maintain moderate flexibility of the films In this work, existing concept of stratified microstructure reported by Ebina [65,66] was studied and developed to achieve a film for enhanced barrier applications. E ffect s of polymer loading in clay films on ba rrier properties and resulting microstructures of clay platelets w ere investigated. E xterior layers were formed to prevent the swelling of clay films and to promote a better flexibility. Experiments Preparation of Highly Filled Nanocomposite Films Closite Na+The mixture was degassed in vacuum for 1 hour and then poured onto a Teflon caster, dried in hot oven at 60 ( C Na ) a natural MMT (Southern Clay Products, Louisville, KY) powder was dispersed slowly in distilled water using a high shear agitator for 3 hours. Poly(acrylic acid) ( PAA) with average molecular weight (Sigma Aldrich, St. Louis, MO ) of 250,000 was dissolved in distilled water and pH of prepared PAA solution was controlled using aqueous HCl (1 M) or NaOH (1 M) solution. Negatively charged PAA was attained by setting pH of the mixture to basic condition [140,141] These negatively charged PAA binder m olecules will attach to edge s of clay platelets Prepared PAA stock solution was poured into a clay solution and mixed for another 2 hours. Table 61 lists c orresponding relative amounts of clay and PAA for each sample. oC for 15 20 hours. Formed film specimens was detached from the caster and dried at room temperature for several minutes. To prevent swelling of clay films by water molecules, poly( vinyl butyral) (PVB), B60H (Kuraray Inc., Japan) solution dissolved in ethanol with a concentration of 5 wt% was used as an exterior coating Clay films were dip coated in PVB solution and dried overnight at room temperature. T hicknesses of prepared films were measured using a micrometer and compared with scanning electron microscopy ( SEM )
166 micrographs with error range s Figure 6 1 represents schematic o f experiment procedures for prepar ing highly filled PCN film coated with PVB. Characterizations M icrostructures of clay platelets were characterized by x r ay diffractometer ( XRD) ( Philips MRD X'Pert System) using CuK radiation ( = 1.54056 ) with a range of scanning angle from 2 to 10 degree. Transmission electron microscopy ( TEM ) (TEM 200CX, Jeol L td., Tokyo, Japan) also was used to investigate orientation s of clay. An accelerating voltage of 200 kV was used for TEM analysis D ispersion liquid is dropped directly on a copper grid supported with a collodion membrane. Field emission SEM ( FE SEM ) ( JEOL JSM 6335F ) was conducted on fractured crosssection areas of film specimens as well as film surface s Film samples for FE SEM were sputter coated with gold palladium alloy (Au Pd). Submicrondetailed surfaces characteristics were obtained under ambient con ditions using atomic force microscopy ( AFM ) (AFM Dimension 3100, Veeco Co.) 2 areas. O xygen transmission rate ( OTR ) value, which is a steady state rate of transmission of oxygen gas, is widely used as a standard method for estimating barrier propert ies of a material; it is defined as a quantity of oxygen gas passing through a unit area of parallel surfaces of a plast ic film per unit time under conditions of test. OTR of samples were measured using a procedure and instruments (Model OX TRAN 2/20 MH module, MOCON Co.) described in ASTM D398505. Flexibility of obtained film specimen was measured using an intuitive curvature test as shown in Figure 62. In a curvature test, three points, A, B and P on a curved line were fixed and r values, radius of curvature at P from a point o, center of curvature for P were measured. Local curvature of k was obtained by taking a reciprocal value of r. Endothermic heat flows of these composites were measured using Differential Scanning Calorimeter (DSC) (MAS 5800, model DSC 200). All
167 samples were tested in crimped aluminum pans at a heating rate of 25 oC/min under dry nitrogen gas over a temperature range from 25 to 300 oC. D ynamic weight loss tests were conducted by using the same instrument. All tests were performed using sample weights of 1012 mg at a heating rate of 5oC/min over a temperature range from 30 to 250 oResults and Discussion C. Variables Affecting Barrier Properties In general it has been reported that barrier properties of polymeric clay nanocomposit e materials are improved as weight fraction of embedde d fillers increases. However, variances of OTR of highly filled nanocomposite samples resulted in a completely different tendency with that of conventional PCN system where nanoclay loading is usually less than 10 wt%. As observed in Figure 63, barrier properties deteriorated as wt% of polymer decreased. D ensely arranged unique microstructure of clay platelets and filling voids theory suggested by Beall  can be two explanation s f or this counter tendency. D etour paths of oxygen molecules resulting from randomly dispersed filler particles will not be an important fac tor for barrier properties of highly filled nanocomposite materials. Rather, it is important to compare morphological features to analyze r ole s and effect s of polymer loading on barrier prope rties. Morphological feature of clay platelets in nanocomposite film specimens w ere observed using SEM Figure 64 represents a proposed microstructure of clay platelets and PAA chains schematically with SEM micrographs of prepared film samples. As reporte d by Ebina negatively charged PAA chains under basic conditions lead to parallel arrangement by promoting edge to edge attachment among clay platelets as shown in Figure 64. This arrangement is u nlike the house of cards structure formed in Laponite JS aqueous solution as observed in Chapter 4. A f aceto face arrangement of CNa is due to high aspect ratio (Aspect ratio is ca. 75 150 nm  ) CNa clay spontaneously arrang es its platelets in a (001) direction  A rrangements of CNa clay platelets
168 at va rious polymer loadings were confirmed by SEM micrographs of cross section area and XRD analysis. For all SEM micrographs of film specimens, intrinsic sinusoidal morphology of clay platelets was observed. A notable difference is a chevron pattern which was observed in Figure 6 5 (a) This pattern disappeared and debris w as r e duced as PAA loading increased. This indicates that negatively charged PAA chains plays a significant role in determining specimens brittleness affecting therefore barrier properties. C N3P0 containing no polymer showed more winding features, which produce d more microvoids between platelets. In fact, because of highly brittle characteristics of CN3P0 specimen, OTR measurement was not able to be conducted. U se of PAA polymer to organize cl ay platelet by inducing edge to edge arrangement and fill up voids made it possible to improve barrier properties. OTR was lowered when polymer loading exceed ed 23 wt% T he lowest OTR value was recorded at 50 wt% of polymer composition. PAA attached to cl ay platelets edge result ing in more curved features and these features were found in (e) and (f) in Figure 65. When PAA loading exceeded critical amount which is required to induce clay platelets edge to edge attachment, excess PAA might diffus e into microvoids which were resulted from winding feature s of clay platelet microstructure. Some microvoids were filled with these PAA chains resulting in dense structure as shown in CN3P70 (e) and CN3P100 (f) of Figure 6 5. Therefore it could be deduced that enha nced barrier properties of CN3P70 and CN3P100 were caused by densely stratified microstructure of clay platelets and polymer filled voids D ifference of microstructural features was investigated using X ray diffraction as illustrated in Figure 66 (a) T he re was no difference in peak positions for all the samples indicating that the polymer chains were not diffused into the intergallery region between clay platelets. S ta cked clay platelets were bound to each neighbor by ionic bonds with the aid of
169 negativel y charged PAA chains. Full Width at Half Maximum (FWHM) values were measured to compare a degree of orientation as shown in Figure 66 (b). Like that more flatten morphological features of clay platelets were found as polymer loading increased in Figure 6 5 SEM micrographs, FWHM value s decreased suggesting improved orientation was obtained at higher polymer loadings. Decreasing FWHM values with increasing polymer loadings also suggest ed that higher crystallinity was achieved as polymer content was increased This supported OTR results ; lower OTR values w ere obtained at higher polymer loading as oxygen molecules were blocked more. DSC results as shown in Figure 67 also showed the same crystallinity tendency T otal crystallinity of a polymer is a dominating factor which makes it possible to analyze barrier properties to gas molecules among samples comparatively. However, it should be noted that polymer free volume, and diffusion gas type need also to be considered. According to a barrier mechanism, a penetrat ion of oxygen molecules initiates with an adsorption on polymer film surfaces prior to diffusion into films interior. However, if there are some micro or macro sized cracks on film surfaces oxygen molecules will go through the film directly without any adsorption on polymer surface s Therefore it is necessary to compare an amount of polymer and cracks on surface s and to relate their surface characteristics to barrier properties. Surface characteristics were analyzed by performing SEM and AFM. Figure 68 is SEM micrographs of nanocomposite films with various amounts of PAA. Micro sized cracks on surfaces observed in CN3P0 (a) and CN3P10 (b) resulted from lack of polymer ; film surfaces were easily damage d due to low polymer loading. Therefore, OTR values f or samples with high polymer content were lower b y having less probability of a direct diffusion through cracked
170 regions into interior re gion s Excess polymer enhanced barrier properties by filling cracks on the film surface retard ing total diffusion time of gas molecules. Roles of Poly(vinyl) Butyral Layer Clay films have several drawbacks in that these films are so brittle and susceptible to water molecules. Highly filled nanocomposite films which were prepared based on spontaneous arrangement s of clay p latelet also ha d similar drawbacks with those of clay films. E xterior PVB layers were us ed to enable PAA/CNa nanocomposite film s for use as practical barrier application s T wo main roles of PVB layer were to p revent swelling of clay with water molecules an d to improv e flexibility of films. As PVB layer was nonsoluble in water penetration of water molecules was prevented. D egree of physical interlocking which can determine adhesive property between PAA/Clay films and the PVB coating was estimated by examini ng surface roughness. SEM micrographs of surfaces indicated that high polymer loading contributed to a smoother surface of resulting films. However, s urface s became rough when amount of polymer exceeded 33 wt% ( from CN3P30 to CN3P50). A smo other surface re turned at 50wt% (CN3P100) of poly mer loading as shown in (f) of Figure 68. This tendency was confirmed again with RMS data analyzed by AFM as provided in Figure 6 9. A large RMS value of CN3P10 was caused by weakly attached stacked clay platelets and bro ken pieces of this brittle specimen thereby contribut ing to increas ed roughness. The largest root mean square ( RMS ) value was shown in the case of CN3 P70. The largest RMS value of CN3P70 might be due to large aspect ratio of CNa clays showing more curved morphological features. At film surface s some of these clay platelets protruded their surface layer. This effect was also enhanced by large excess in PAA polymer chains which were diffused in microvoids. Increased amount of polymer of CN3P100 lowered RMS value significantly than that of CN3P70. A reason that CN3 P100 has a lower RMS value could be due to protruded edge s of
171 curved platelets of tactoids which were covered by polymer making it smooth er This was drawn schematically in Figure 6 10. To identify excess polymer on surfaces and effects of this excess polymer on roughness specifically, several 3D surface images with two 2D phase contrast images of CN3P10 and CN3P100 and Power spectral density (PSD) were obtained as shown in Figure 611 and 6 12, resp ectively. 3D height contrast images of samples showed a good match with RMS values and SEM surface images. 2D phase contrast image s (f) and (g) illustrated well differences in amount of excess po lymer on surfaces A p olymer phase which is close to viscoela stic materials exhibit s a high phase angle between a cantilever response and that in free air This corresponds to brighter region in a phase contrast image. L ower phase angle is usually observed as darker region which is more elastic This difference in phase contrast image of tapping mode depending on phase angle was shown in (h) Figure 611  Therefore phase image of (g) CN3P100 was much brighter than (f). PSD is a good indicator to investigate correlation length s of rough surface s Lower frequency region exhibits a macro sized roughness and higher frequency is a micro sized roughness which is required to know overall roughness of a sample. A ll samples showed similar tendency for lower frequency regions as shown in Figure 6 12. However, CN3 P70 has hi gher micro roughness than CN3 P100, as excess polymer cover ed micro sized rough surfaces Flexibilities of obtained specimens were tested using an intuitive curvature test. During sample test s no visual delamination of PVB layer from surfaces of nanocompos ite films was observed proving good adhesion between h ighly filled nanocomposite film and PVB layer. As shown in SEM micrographs (Figure 6 4 and 65 ) of fractured surfaces of these specific nanocomposite specimens, delaminated layers were observed. Figure 613 shows flexibilit ies of
172 nanocomposite film specimens before and after PVB coating process. For all samples except CN3P70, flexibilities were improved after PVB coating. Especially, at lower polymer loading, PVB layer s contributed to improving flexibil ity because energy of crack propagation of brittle specimen dissipated into ductile exterior PVB layer. Ductile PVB layer adhered to PAA/CNa nanocomposite film by preventing an initial delamination of brittle clay platelets effectively resulting in much hi gher curvatures. When amount of PAA increased, effect s of PVB layer on flexibility w ere reduced N anocomposite film itself was quite flexible for CN3P10 and CN3P30 and, in fact, they could be fold ed without any cracking This is because amount of water mol ecules existing in PAA linkage molecule s to connect clay platelets with in struc ture of clay nanocomposite film is optimum used for CN3P10 and P30 showing a good flexibility. Therefore, it can be concluded that flexibilit ies of highly loaded nanocomposite f ilms are governed by the extent of hydration of clay platelets as well as the densification of the stratified structure rather than amount of polymer or how the platelets are arranged in this particular case. E ven though brittle characteristic s of clay pla telets w ere improved by addition of PAA chains as shown in SEM micrographs, flexibilities of specimen s w ere not enhanced by increasing the polymer loading PVB coating on both sides of PAA/CNa nanocomposite films c ontribute d to enhanc ed barrier properties as shown in OTR result s in Figure 63. PVB layer s ha ve its own crystalline structure to prevent diffusion of gas molecules and PVB layer s also assist with covering and filling cracks on surfaces of nanocomposite films. Frequently observed Tg of PAA about 102109 oC is completely overlapped with endothermic peak from 50 to 150 oC contributing to typical water desorption of MMT [42,153,154,157] An endothermic peak shown on a curve of CN3P0 which contains no polymer
173 was due to water desorption of MMT. H eat o f fusion for overlapped endothermic peaks for all specimen s can be expressed as Htotal = HPAA + HMMT. HMMT dominated over all peaks of CN3P10, P30 and P50. As polymer content increased, peak area increased as well. PAA inherently shows a high degree of intermolecular hydrogen bonding involving bound water acquired during polymer preparation process [153,154] Therefore, in the cases of CN3P70 and P100, increased degree of intermolecular hydrogen bonding caused by excess amount of PAA resulted in a muc h larger area of heat of fusion. As a result, it can be concluded that when amount of polymer exceeded a certain threshold, endothermic peaks are affected more by intermolecular hydrogen bonding of PAA. Peak position shifted slightly toward lower temperatu re as amount of polymer increased as indicated as an arrow in Figure 67. This phenomenon can be explained by that excess PAA chains which exist on surface s or which are not ass ociated to clay platelets require less energy for an initiation of chain moti ons. Thermal gravimetric analysis ( TGA ) thermograms were measured to confirm effects of water on thermal properties of specimens as demonstrated in Figure 614. W eight loss attributing to water removal of MMT was identified on a curve of CN3P0. W ater molecules contained within PAA polymer chains were bound to clay platelets resulting in retardation of rate and onset of water removal from the composite system. Because small amount PAA polymer chains mainly acts as a linkage molecule to connect between clay p latelets in CN3P10 and P30, water molecules within PAA polymer chains or from hydrated clay platelets were not easily removed causing less weight loss than CN3P0 by 220 oC. F inal weight loss showed much smaller increment from previous one with similar incr ement of polymer as amount of polymer increased. It was more difficult to remove water molecules when PAA content s increased because of a much higher degree of intermolecular hydrogen bonding. In fact, based on flexibility test s the author expected that
174 w ater content s of CN3P10 would be highest while CN3P100 would exhibit poor flexibility due to lower water content. However, results were completely contrary to expectations. Th erefore flexibility is determined by amount of water molecules existing in PAA linkage molecules that connect s clay platelets with in structures of nanocomposite system s and not by amount of water in complete nanocomposite system Even though water content of CN3P100 is the largest, a role of water ascribing to flexibility would be red uced as lower quantity of clay used and added PAA increased brittleness of specimens resulting in poor flexibilities Paralikar et al and Jiang et al. reported that PAA is a typical brittle polymer therefore PAA in composite systems increases brittleness of specimens [155,156] C ritical amount of PAA needed to maintain flexibility has to be determined as exceeding a critical loading will result in a deterioration of macromechanical properties. Large amount of PAA pr edicted lower oxygen transmission rate. I n creased packing density of polymer chains induced by higher degree of intermolecular hydrogen bonding as well as well aligned structures of clay platelets which are flattened showed increased crystallinity as indicated as the largest peak areas in Figure 67. Therefore a barrier property of CN3P100 was enhanced more than any other system It was concluded that optimum range of PAA to achieve good flexibility and barrier property was 30 35 wt%. It is also important to note that by adjusting pH of PAA PAA a lso act as linkage molecules for the clay platelets.
175 T able 61. C orresponding relative amounts of clay and PAA for each sample Samples Conc. of clay (wt%) Amount of PAA (wt%) Thickness 1 Thickness 2 CN3P0 3 0 25 N/A* CN3P10 3 9 28 32 CN3P30 3 23 29 34 CN3P50 3 33 31 33 CN3P70 3 41 33 37 CN3P100 3 50 31 34 1, thickness of clay sample before PVB coating; 2, thickness of clay samples after PVB coating
176 Figure 6 1. Schematically presented experiment procedure for the preparation of PVB coated PAA/CNa nanocomposite film
177 Figure 62. Schematic diagram of an intuitive curvature test to measure flexibility of obtained film specimens Figur e 63. OTR values of highly filled nanocomposite film specimens of different polymer loading. OTR values of PVB coated film specimens were also conducted to estimate the contribution of exterior layer on barrier properties. (RPtoC represent the ratio of polymer to clay; refer to table 6 1 for details)
178 Figure 6 4. Schematic microstructure of clay platelets and PAA chains and corresponding SEM micrographs of prepared film specimen
179 Figure 65. SEM mic rographs (x 5,000) of cross section areas of obtained highly filled nanocomposite film specimens (a) CN3P0, (b) CN3P10, (c) CN3P30, (d) CN3P50, (e) CN3P70 and (f) CN3P100
180 (a) (b) Figure 66. ( a) X ray diffraction pattern s for CN3P0, P10, P30, P50, P70, and P100; (b) a FWHM as a function of polymer loadings
181 Figure 6 7. DSC thermograms of highly filled nanocomposite films with various amounts of polymer
182 Figure 68. SEM micrographs (x 5 ,000) of surface areas of obtained highly filled nanocomposite film specimens (a) CN3P0, (b) CN3P10, (c) CN3P30, (d) CN3P50, (e) CN3P70 and (f) CN3P100
183 Figure 6 9. D ependence of RMS on the amount of polymer in highly fille d nanocomposite films Figure 610. An expected schematic picture of protruded curved edge of tactoids from the surface of CN3P70. Covered region (grey scale) with excess PAA lowered roughness by removing micro roughness on the surface. (Entangled lines and stacked bold lines represent polymer chains and tactoids, respectively)
184 (h) Figure 611. 3D height contrast images taken by AFM (tapping mode) on a 5 x 5 m2 on (a) CN3P10, (b) CN3P30, (c) CN3P50, (d) CN3P70, (e) CN3P100; 2D phase contrast images taken by AFM (tapping mode) on a 5 x 5 m2 on (f) CN3P10, (g) CN3P100; Schematic diagram to illustrate the definition of phase angle shown in ( h) 
185 Figure 612. Power spectral density diagrams of highly filled nanocomposite samples (a) CN3P1 0, (b) CN3P30, (c) CN3P50, (d) CN3P70 and (e) CN3P100
186 Figure 613. Curvature of highly filled nanocomposite films with various amounts of polymer and a picture of a flexible film specimen (CN3P50)
187 Figure 6 14. TGA thermograms of highly filled nanocomposite films with various amounts of polymer
188 CHAPTER 7 The hypothesis and resulting major findings from each section were addressed in this chapter based on the work done so far as following; CONCLUSIONS Chapter 1: T ypical three conventional microstructures (phase separated, intercalated and exfoliated structure) of layered silicate particles in the polymer matrix and their correlations to the barrier property of resulting materials were introduced with the fundamental concept for barrier function of a polymer clay nanocomposite (PCN) material. Chapter 2: Controlling microstructure of filler particles in practical PCN systems is a key factor to achieve an effective barrier material. T his chapter has reviewed variables which have effects on the formation of certain microstructure of layered silicate particle in actual PCN systems from the stage of the materials selection to processing parameters. To maximize physical properties of PCN material, interplay among three main components, polymer m atrix, nanoclays and organic molecules covering these nanoclays and other processing parameters has to be considered. A variety of microstructures of PCN have been reported under various actual conditions. A practical PCN material can be defined as a broad mixture of intercalated and exfoliated structure with different ratios and a dominating structure representing whole composite system s determines physical properties of resulting PCN materials.
189 Several models have been suggested to design a PCN material with better barrier properties. All existing barrier models are based on Nielsen s detour theory and conform well to results of permeability of practical PCN system s And there were two common points for all barrier models. First, barrier property was enh anced with increasing total volume fraction of filler particles. Secondly it was noted that the larger aspect ratio promised better barrier properties. Chapter 3: W ork s in Chapter 3 demonstrate d the potential for PCN coatings to enhance barrier performance of packaging materials PCN, which consists of Laponite in poly(vinyl alcohol) ( PVA) matrix, coated polypropylene (PP) polyolefin after a tmospheric pressure plasma ( APP ) surface treatment showed a good barrier performance. Oxygen transmission rate ( OTR ) value of PP was reduced from 150 to 10 cc/m2 As a result of AFM study, it was found that two optimum conditions for the APP treatments promising the highest level of surface roughnes s and the lowest permeabilities were 18 l /min of nitrogen flow rate and below 20 mm of working distance. /day when coated with a PCN solution containing 50 wt% of filler particle. Through XRD analysis, it was found that an exfoliated structure of PCN sample was obtained after a high shear mixing process However, when filler cont ent increased above 60 wt%, an intercalated feature was detected because of restacked filler particles in limited volumetric space s of polymer matrix.
190 OTR values showed a general barrier tendency that barrier property is enhanced with increasing volume fra ction of filler particle. However, as a result of restacked particles at above 60 wt% diffusion length of gas molecules was reduced resulting in sharp increase in OTR values. Chapter 4: To obtain better barrier properties and to maintain a good barrier pr operties under various conditions, an extended tortuous path models using clay platelets as a major component and polymer as a linkage molecules (Highly filled PCN material) was suggested as a new barrier solution in Chapter 4 Using a unique A house of ca rds structure of Laponite JS and negatively charged poly(acrylic acid) (PAA), m icrostructure s can approach a densely stratified structure s through the face to face attachment of clay platelets. I dentification of a house of cards structure of Laponite JS in an aqueous solution was found by forming a clay aerogel through a freezing and drying process. Transmission e lectron m icroscopy (TEM) micrographs of a clay aerogel showed well three typical morphological features of the microstructure, which are tilted edge standing and paralleled position. The behavior of PAA chain in highly filled PCN coating solution depending on pH was examined by EDS test and, in conclusion, PAA chain in the coating solution prepared under pH level of isoelectric point of PAA was entangled each other with the small amount of embedded clay platelets.
191 PAA in basic condition played a role as a linkage molecule to realize edge to edge contact between clay platelets, the lowest OTR value about 10 cc/m2 A significant amount of cracks on the surface were observed by s canning e lectron m icroscopy (SEM) when PP was coated with PCN coating solution that were prepared at pH 2 and 4, and these cracks resulted in sharp increase in OTR value to 120 cc/m /day was obtained for negatively charged PAA based PCN coated PP. 2Chapter 5: /day. B ar rier model to estimate relative permeability ( Rp Based on Nielsen s detour theory(Nielsen, 1967), Xu s barrier model ( Xu et al., 2006) and a definition of effective volume from Saunder ( Saunders et al., 1999), the barrier formula for PC 2N microstructure was obtained as following equation: ) of PC N containing two different particle types (PC2N) have been proposed in Chapter 5. Initial purpose of this microstructure, PC2N, was t o explore the benefits of dispersing smaller platelets between larger sheets such as Cloisite. 1 2 2 2 1 12 ) 1 ( 2 H v L L H Reff T P w here, L1, L2 and H1 can be defined as lateral length including edge to edge distance, ( l1+b1) and ( l2+b2) of clay 1 and 2 and vertical length (h1+w1 However, contrary to initial expectation s it was concluded based on the comparison with actual experimental results that total volume fraction of filler ) occupied by clay 1 per unit volume respectively. (Refer to Chapter 5 for specific details)
192 particles in a composite material determines barrier properties regardless of specific ratio between two particle types Prepared PC2N composite films were beneficial in terms of their relatively light weight and improved optical transmittance by mixing two clay types due to lower density an d smaller size particle which is advantageous for elastic light scattering s This work suggested that characteristics of each type of clay particles in the PC2N composite can be manipulated and designed to enhance the physical properties while maintaining good barrier properties. Chapter 5: Cloisite Na+ B arrier tendency with polymer loading of prepared highly filled PCN film samples was opposite to the general barrier tendency of normal PCN material based on Nielsen s detour theory This is because tha t different polymer loading resulted in different morphological features varying barrier properties of prepared samples. Morphology of cross sectional area, Crystallinity tendency and microstructure study obtained through SEM, XRD and DSC analysis also conformed well to the barrier tendency with polymer for these samples A verage OTR value of prepared films was 0.5 cc/m platelets form a paralleled oriented microstructure due to their larger aspect ratio and high particle density. Thus it was possible to form a flexible self supporting highly filled PCN film through edge to edge attachment s using negatively charged PAA chains 2/day and the lowest OTR of 0.1 cc/m2/day was found when PCN film containing 50 wt% of polymer loading was coated with exterior PVB layer.
193 T o investigate effect s of surface morphological feature of PCN films on barrier properties, RMS values and PSD diagram were obtained. As a result of these AFM studies, it was concluded that PCN film containing 41 wt% of PAA showed the highest level of surfa ce roughness and this was because the highest level of microroughness caused by protruded curved edges of stacked clay tactoids. Flexible highly filled PCN film is usually prepared by obtaining densely stratified microstructure and sufficient amount of wa ter molecules in the whole composite system. However, flexibilities of these specific samples w ere determined by water molecules existing in PAA linkage molecules and clay platelets themselves. Through SEM micrographs, curvature test and thermal analysis, it was found that 23 33 wt% of PAA loading promised the best flexibility of prepared PCN films. E xterior poly(vinyl butyral) (PVB) layer was able to contribute to better barrier properties and flexibility. PVB exterior layer was not delaminated from PCN f ilm proving that PVB layer adhered well to the PCN film by preventing the initial delamination of brittle clay platelets effectively resulting in an increase in curvature and PVB layer also contributed to enhancing barrier property by preventing direct dif fusion of gas molecules. For example, CN3P30 showed changes in curvature from 0.4 to 1.0 mm1 and in OTR values from 0.8 to 0.5 cc/m2 /day. And optimum range of PAA contents to achieve good flexibility and barrier property at the same time was 30 35 wt%. I hope that these result s will be useful to those who want to do research in the fascinating field of PCN materials.
194 APPENDIX A CLASSIFICATION OF COMMONLY USED LAYERED SILICATES
195 APPENDIX B COMMERCIALLY AVAILAB LE ORGANOCLAYS AND ORGANIC MOLECU LES USED FOR THE MODIFICATION OF THE SURFACE OF NA NOCLAYS tallow represents a mixture of long alkyl homologues with octadecyl the most prominent component. Commercial Name Class Modifier Nanofil 757 Natural MMT (MMT Na + ) NONE Nanofil 919 Modified MMT Dimethyl (Benzylmethyl) tallow* quaternary ammonium Nanofil 804 Modified MMT bis(2 hydroxyethyl) hyd rogenated tallow ammonium Nanofil 32 Modified MMT alkylbenzyldimethylammonium (benzalkonium) Nanofil 15 Modified MMT Dimethyl dihydrogenated tallow ammonium Cloisite Na+ Natural MMT (MMT Na + ) NONE Cloisite 15A Modified MMT dimethyl,dehydrogenated tallo w ammonium Cloisite 20A Modified MMT Dimethyl dihydrogenated tallow ammonium Cloisite 25A Modified MMT dimethyl 2 ethylhexyl (hydrogenated tallow alkyl) ammonium Cloisite 30B Modified MMT methyl tallow bis 2 hydroxyethyl quaternary ammonium Nanomer I.4 4P Modified MMT dimethyldialkyl quaternary ammonium Nanomer I.30E Modified MMT Octadecylamine quaternary ammonium
196 LIST OF REFERENCES 1. F ornes, T.D ., Yoon, P.J., Keskkula, H., and Paul, D.R. (2001). Nylon 6 N anocomposites: T he E ffect of M atrix M olecular W eight Polymer 42: 99299940. 2. Yoo, Y., Shah, R.K., and Paul, D.R. (2007). Fracture B ehavior of N anocomposites B ased on P oly( e thylene co methacrylic acid) I onomers Polymer 48: 48674873. 3. Kim, D.H., Fasulo, P.D., Rodgers, W.R., and Paul, D.R. (2007). Structure and P roperties of P olypropylene B ased N anocomposites: Effect of PP g MA to O rganoclay R atio Polymer 48: 53085323. 4. K ojima, Y., Usuki, A., K awasumi, M., Okada, A., Fukushima, Y., Kurauchi, T., and Kamigaito, O. (1993). Mechanical P roperties of N ylon 6clay Hybrid, J. Mater. Res. 8: 11851189. 5. Podsiadlo, P., Kaushik, A.K., Arruda, E.M., Waas, A.M., Shim, B.S., Xu, J., Nandivada, H., Pumplin, B .G., Lahann, J., Ramamoorthy, A., and Kotov N.A. (2007). Ultrastrong and S tiff L ayered P olymer N anocomposites Science, 318: 8083. 6. Lange, J., and Wyser, Y. (2003). Recent I nnovations in B arrier T echnologies for P lastic P ackaging A R eview Packag. Technol. Sci. 16: 149158. 7. Osman, M.A., Mittal, V., Morbidelli, M., and Suter, U.W. (2003). Polyurethane A dhesive N anocomposites as G as P ermeation B arrier Macromolecules 36 : 98519858. 8. Amberg schwab, S., Hoffmann, M., Bader, H., and Gessler, M. (1998). Inorga nic O rganic P olymers with B arrier P roperties for W ater V apor, O xygen and F lavors J. of Sol Gel Sci. and Technol., 1/2: 141146. 9. Triantafyllidis. K.S., LeBaron, P.C., Park, I., and Pinnavaia, T.J. (2006). Epoxy clay F abric F ilm C omposites with U nprecedente d O xygen B arrier P roperties Chem. Mater., 18: 43934398. 10. Anilkumar, S., Kumaran, M.G., and Thomas, S. (2008). Characterization of EVA/clay N anocomposite M embranes and I ts P ervaporation P erformance J. Phys. Chem. B 112: 40094015. 11. Thomassin, J M., Pagnoulle, C., Caldarella, G., Germain, A., and Jr (2006). Contribution of N anoclays to T he B arrier P roperties of A M odel P roton E xchange M embrane for F uel C ell A pplication J. of Mem. Sci. 270: 5056. 12. Duangkaew, P., and Wootthikanokkhan, J. (2008). Methanol F uel C ell M embranes B ased on S ufonat ed P oly(vinyl alcohol) L ayered S ilicate N anocomposites J. Appl. Polym. Sci., 109: 452458. 13. Sridhar, V., and Tripathy, D.K. (2006). Barrier P roperties of C hlorobutyl N anoclay C omposites J. Appl. Polym. Sci., 101: 36303637
197 14. Choi, W.J., Kim, H.J., Yoon, K.H., Kwon, O.H., and Hwang, C.I. (2006). Preparation and B arrier P roperty of P oly(ethylene terephthalate)/ C lay N anocomposite U sing C lay S upported C atalyst J. Appl. Polym. Sci., 100: 48754879. 15. Brul, B., and Flat, J J. (2006). High B arrier P olyamide/ P olyol efin/ O rganoclay N anocomposites Macromol. Symp., 233: 210216. 16. Baughman, R.H., Zakhidov, A.A., and De Heer W.A. (2002). Carbon N anotubes the R oute T oward A pplications Science, 297: 787792. 17. Moniruzzaman, M., and Winey, K.I. (2006). Polymer N anocomposites C ontaining C arbon N anotubes Macromolecules 39: 51945205. 18. Sandler, J.K.W., Kirk, J.E., Kinloch, I.A., Shaffer, M.S.P., and Windle, A.H. (2003). Ultra L ow E lectrical P ercolation T hreshold in C arbonN anotube E poxy C omposites Polymer 44: 58935899. 19. Brynin g, M.B., Islam, M.F., Kikkawa, J.M., and Yodh, A.G. (2005). Very L ow C onductivity T hreshold in B ulk I sotropic Si ngle W alled C arbon N anotube E poxy C omposites Adv. Mater., 17: 11861191. 20. Kim, Y.S., Wright, J.B., and Grunlan J.C. (2008). Influence of P olymer M odulus on T he P ercolation T hreshold of L atex B ased C omposites Polymer 49: 570578. 21. Alt, V., Bechert, T., Steinrcke, P., Wagener, M., Seidel, P., Dingeldein, E., Domann, E., and Schnettler, R. (2004). An I n V itro A ssessment of T he A ntibacterial P ropert ies and C ytotoxicity of N anoparticulate S ilver B one C ement Biomaterials 25: 43834391. 22. Chen, C.Z., and Cooper, S.L. (2000). Recent A dvances in A ntimicrobial D endrimers Adv. Mater., 12: 843846 23. Damm, C., Mnstedt, H., and R (2008). The A ntimicro bial E fficacy of P olyamide 6/ S ilver N ano and M icrocomposites Mater. Chem. Phys., 108 : 6166. 24. Koros, W. (1990) Barrier P olymers and S tructures: O verview. In: B arrier P olymers and Structure pp 121, John Wiley & Sons Ltd, Chichester UK. 25. Salame, M. (1986). Prediction of G as B arrier P roperties of H igh P olymers Polymer Eng. Sci., 26: 15431546. 26. Prasad, T., and Brown, K. (1995). Permachor M ethod to P redict D iffusivity C oefficients of O rganics in A queous S olutions in F lexible M embrane L i ners Waste Management & Research 13: 4753. 27. Hiltner, A., Liu, R.Y.F., Hu, Y.S., and Bear, E. (2005). Oxygen T ransport as A S olid S tate S tructure P robe for P olymeric M aterials: A R eview J. Polym. Sci. B: Polym. Phys. 43: 10471063.
198 28. Dhoot, S.N. (2004). Sorption and T ransport of G ases and O rganic V apors in P oly(ethylene terephthalate) The University of Texas at Austin, Dissertation of Doctor of Philosophy 29. Jang, J., and Lee, D.K. (2004). Oxygen B arrier P roperties of B iaxially O riented P olypropylene/ P olyvinyl A lcohol B lend F ilms Polymer 45: 15991607. 30. Petropoulos, J.H., Paul, D.R., and (1994). In P olymeric G as Separation M embranes CRC Press, Boca Raton, FL. 31. Nielsen, L.E. (1967). Models for T he Permeability of Filled Polymer Systems J Macromol. Sci. (Chem) A 1: 929942. 32. Yu, Y H., Lin, C Y., Yeh, J M., and Lin, W H. (2003). Preparation and P roperties of P oly(vinyl alcohol) C lay N anocomposite M aterials Polymer 44: 35533560. 33. Ginzburg, V.V., Singh, C., and Balazs, A.C. (2000). Theoretical P hase D iagrams of P oly mer/ C lay C omposites: T he R ole of G rafted O rganic M odifiers Macromolecules 33: 10891099. 34. Vaia, R.A., Jandt, K.D., Kramer, E.J., and Giannelis, E. P. (1996). Microstructural E volution of M elt I ntercalated P olymer O rganically M odified L ayered S ilicates N an ocomposites Chem. Mater., 8: 26282635. 35. Vaia, R.A., and Wagner, H.D. (2004). Frameworks for N anocomposites Materials Today Nov : 3237. 36. Illett, S.M., Orrock, A., Poon, W.C.K., and Pusey, P.N. (1995). Phase B ehavior of A M odel C olloid P olymer M ixture Phy s. Rev. E 51 : 13441352. 37. Dijkstra, M., and van Roij, R. (1997). Entropy D riven D emixing in B inary H ard C ore M ixtures: from H ard S pherocylinders T owards H ard S pheres Phys. Rev. E 56: 55945602. 38. Dijkstra, M., Hansen, J.P., and Madden, P. A. (1997). Statis tic M odel for T he S tructure and G elation of S mectites C lay S uspensions Phys. Rev. E 55: 30443053. 39. Malwitz, M.M., LinGibson, S., Hobbie, E.K., Butler, P.D., and Schmidt, G. (2003). Orientation of P latelets in M ultilayered N anocomposite P olymer F ilms J Polym. Sci. B: Polym. Phys. 41: 32373248. 40. Mermut, A.R., and Lagaly, G. (2001). Baseline S tudies of T he C lay M inerals S ociety S ource C lays: Layer C harge D etermination and C haracteristics of T hose M inerals C ontaining 2:1 L ayers Clays and Clay Minerals 49(5) : 393397. 41. Usuki, A., Kawasumi, M., Kojima, Y., Okada, A., Kurauchi, T., and Kamigaito, O. (1993). Swelling B ehavior of M ontmorilonite C ation E amino A cids by C aprolactam J. Mater. Res. 8: 11741178.
199 42. Picard, E., Gauthier, H., Grard, J. F., and Espuche, E. (2007). Influence of T he I ntercalated C ations on T he S urface E nergy of M ontmorillonites: Consequences for T he M orphology and G as B arrier P roperties of P olyethylene/ M ontmorillonites N anocomposites J. Colloid and Interface Sci., 307: 364376. 43. Zhang, W., Chen, D., Zhao, Q., and Fang, Y. (2003). Effects of D ifferent K inds of C lay and D ifferent V inyl A cetate C ontent on T he M orphology and P roperties of EVA/ C lay N anocomposites Polymer 44: 79537961. 44. Grunlan J.C., Grigorian, A., Hamilton, C.B., and Mehrabi, A.R. (2004). Effects of C lay C oncentration on T he O xygen P ermeability and O ptical P roperties of A M odified P oly(vinyl alcohol) J. Appl. Polym. Sci., 93: 11021109. 45. Demirkol, E.A., and Kalyon, D.M. (2007). Batch and C ontinuous P rocessing of P olymer L ayered O rganoclay N anocomposites J. Appl. Polym. Sci., 104: 13911398. 46. Lee, E.C., Mielewski, D.F., and Baird, R.J. (2004). Exfoliation and D ispersion E nhancement in P olypropylene N anocomposites by I nsitu M elt P hase U ltrasonication Polym. En g. And Sci., 44 : 17731782. 47. Lee, S.S., Hur, M.H., Yang, H., Lim, S., and Kim, J. (2006). Effect of I nterfacial A ttraction on I ntercalation in P olymer/ C lay N anocomposites J. Appl. Polym. Sci., 101: 27492753. 48. Gerasin V.A., Bakhov, F.N., Merekalova, N.D., K orolev, Yu.M., Zubova, T.L., and Antipov, E.M. (2005). Influence of T he S tructure of A M odifier L ayer on T he C ompatibility of P olymers with A M odified M ontmorilonite J Eng. Phys. And Thermophys., 78: 871877. 49. Osman, M.A., Mittal, V., Morbidelli, M., and S uter, U.W. (2004). Epoxy L ayered S ilicate N anocomposites and T heir G as P ermeation P r operties Macromolecules 37: 72507257. 50. Borse, N.K., and Kamal, M.R. (2009). Estimation of S tresses R equired for E xfoliation of C lay P articles in P olymer N anocomposites P olym. Eng. And Sci., 10: 641650. 51. Wagener, R., and Reisinger, T.J.G. (2003). A R heological M ethod to C ompare T he D egree of E xfoliation of N anocomposites Polymer 44: 75137518. 52. Chang, J.H., Seo, B.S., and Hwang, D.H. (2002). An exfoliation of organoclay i n thermotropic liquid crystalline polyester nanocomposites Polymer 43: 29692974. 53. Park, J.H., and Jana, S.C. (2003). Mechanism of E xfoliation of N anoclay P articles in E poxy C lay N anocomposites Macromolecules 36 : 27582768. 54. Fu, X., and Qutubuddin, S. (2 001). Polymer C lay N anocomposites: E xfoliation of O rganophilic M ontmorilonite N anoclays in P olystyrene Polymer 42: 807813.
200 55. Kwak, J.W., Welt, B., and Beatty, C. (2009). Enhancement of O xygen B arrier with P olymer C lay N anocomposite C oatings on P olypropyle ne T reated with A tmospheric P ressure P lasma J Appl. Packag. Res., 3: 3956. 56. Lepoittevin, B., Pantoustier, N., Devalckenaere, M., Alexandre, M., Calberg, C., Jr R., Henrist, C., Rulmont, A., and Dubois, P. (2003). Polymer/ L ayered S ilicate N anocomposit es by C ombinded I ntercalative P olymerization and M elt I ntercalation: A M asterbatch P rocess Polymer 44: 20332040. 57. Zerda, A.S., Caskey, T.C., and Lesser, A.J. (2003). Highly C oncentrated, I ntercalated S ilicate N anocomposites: S ynthesis and C haracterizatio n, Macromolecules 36: 16031608. 58. Yu,. Z.Z., Yang, M., Zhang, Q., Zhao, C., and Mai, Y.W. (2003). Dispersion and D istribution of O rganically M odified M ontmorilonite in Nylon 66 M atrix J. Polym. Sci. B: Polym. Phys. 41: 12341243. 59. Hamaker, H.C. (1937). Th e LondonVan der Waals A ttraction B etween S pherical P articles Physica 4: 10581072. 60. Koo, C.M., Kim, S.O., and Chung, I.J. (2003). Study on M orphology E volution, O rientation B ehavior, and A nisotropic P hase F ormation of H ighly F illed P olymer L ayered S ilica te N anocomposites Macromolecules 36: 27482757. 61. Lin Gibson, S., Schmidt, G., Kim, H., Han, C.C., and Hobbie, E.K. (2003). Shear I nduced M esostructure in N anoplatelets P olymer N etwork J. Chem. Phys., 119: 80808083. 62. Okamoto, M., Nam, P.H., Maiti, P., Kot aka, T., Hasegawa, N., and Usuki, A. (2001). A H ouse of C ards S tructure in P olypropylene/ C lay N anocomposites U nder E longational F low Nanoletters 1: 295298. 63. Bharadwaj, R.K., Mehrabi, A.R., Hamilton, C., Trujillo, C., Murga, M., Fan, R., Chavira, A., and Thompson, A.K. (2002). Structure P roperty R elationships in C ross L inked P olyester C lay N anocomposites Polymer 43: 36993705. 64. Ha, Y.H., Kwon, Y., Breiner, T., Chan, E.P., Tzianetopoulou, T., Cohen, R.E., Boyce, M.C., and Thomas, E.L. (2005). An O rientatio nally O rdered H ierarchical E xfoliated C lay B lock C opolymer N anocomposite Macromolecules 38: 51705179. 65. Ebina, T., and Mizukami, F. (2007). Flexible T ransparent C lay F ilms with H eat R esistant and H igh G as B arrier P roperties Adv. Mater., 19: 24502453. 66. Te tsuka, H., Ebina, T., Nanjo, H., and Mizykami, F. (2007). Highly T ransparent F lexible C lay F ilms M odified with O rganic P olymer: S tructural C haracterization and I ntercalation P roperties J. Mater. Chem., 17: 35453550.
201 67. Sagitova, E.A., Prokhorov, K.A., Nikolaeva, G.Yu., Pashinin, P.P., Gerasin, V.A., Guseva, M.A., and Antipov, E.M. (2008). Quantitative C haracterization of T he O rientation of M acromolecules in I ntercalated N anocomposites of P olyolefins/ L ayered S ilicates by R aman S pectroscopy Laser Phy sics 18: 868881. 68. Russo, G.M., Simon, G.P., and Incarnato, L. (2006). Correlation B etween R heological, M echanical, and B arrier P roperties in N ew C opolyamide B ased N anocomposite F ilms Macromolecules 39: 38553864. 69. Gopakumar, T.G., Lee, J.A., Kontopoulou, M., and Parent, J.S. (2002). Influence of C lay E xfoliation on T he P hysical P roperties of M ontmorilonite/P olyethylene C omposites Polymer 43: 54835491. 70. Famulari, A., Arosio, P., Filippi, S., Marazzato, C., Magagnini, P., Minkova, L., and Meille, S.V. (20 07). Clay I nduced P referred O rientation in P olyethylene/ C ompatibilized C lay N anocomposites J Macromol. Sci. B: phys., 46: 355371. 71. Mishra, A.K., Nando, G.B., and Chattopadhyay, S. (2008). Exploring P referential A ssociation of Laponite and Cloisite with S oft and H ard S egments in TPU C lay N anocomposite P repared by S olution M ixing T echnique J. Polym. Sci. B: Polym. Phys., 46: 23412354. 72. Chen, G., Liu, S., Zhang, S., and Qi, Z. (2000). Self A ssembly in A P olystyrene/ M ontmorilonite N anocomposite Macromol. Rapid Commun., 21: 746749. 73. Strawhecker, K.E., and Manias, E. (2000). Structure and P roperties of P oly(vinyl alcohol)/Na+74. Weon, J.I., and Sue, H.J. (2005). Effects of C lay O rientation and A spect R a tio on M echanical B ehavior of N ylon 6 N anocomposite Polymer 46: 63256334 M ontmorilonite N anocomposites Chem. Mater., 12: 29432949. 75. Fischer, H.R., Gielgen, L.H., and Koster, T.P.M. (1999). Nanocomposites from P olymers and L ayered M inerals Acta Polym. 50: 122126. 76. Agag, T., and Takeichi, T. (2000). Polybenzox azine M ontmorillonite H ybrid N anocomposites: S ynthesis and C haracterization Polymer 41: 70837090. 77. Fornes, T.D., Yoon, P.J., Hunter, D.L., Keskkula, H., and Paul, D.R. (2002). Effect of O rganoclay S tructure on N ylon 6 N anocomposite M orphology and P ropert ies Polymer 43: 59155933. 78. Yeun, J.H., Bang, G.S., Park, B.J., Ham, S.K., and Chang, J.H. (2006). Poly(vinyl alcohol) N anocomposite F ilms: T hermooptical P roperties, M orphology, and G as P ermeability J. Appl. Polym. Sci., 101: 591596. 79. Sorrentino, A., Tor tora, M., and Vittoria, V. (2006). Diffusion B ehavior in P olymer C lay N anocomposites J. Polym. Sci. B: Polym. Phys., 44: 265274.
202 80. Chang, J.H., and An, Y.U. (2002). Nanocomposites of P olyurethane with V arious O rganoclays: T hermomechanical P roperties, M orphology, and G as P ermeability J. Polym. Sci. B: Polym. Phys., 40: 670677. 81. Chang, J.H., An, Y.U., and Sur, G.S. (2003). Poly(lactic acid) N anocomposites with V arious O rganoclays. I. Thermomechanical P roperties, M orphology, and G as P ermeability J. Polym. Sc i. B: Polym. Phys. 41: 94103. 82. Deng, Y., Gu, A., and Fang, Z. (2004). The E ffect of M orphology on T he O ptical P roperties of T ransparent E poxy/ M ontmorilonite C omposite Polym. Int., 53: 8591. 83. Takahashi, S., Goldberg, H.A., Feeney, C.A., Karim, D.P., Farre ll, M., OLeary, K., and Paul, D.R. (2006). Gas B arrier P roperties of B utyl R ubber/ V ermiculite N anocomposite C oatings Polymer 47: 30833093. 84. Takahashi, S., and Paul, D.R. (2006). Gas P ermeation in P oly(ether imide) N anocomposite M embranes B ased on S urfac e T reated S ilica. Part 1: without C hemical C oupling to M atrix Polymer 47: 75197534. 85. Property material data, Cloisite Technical Data Sheets Link. Web address http://www.matweb.com/search/GetMatlsByTradename.aspx?navletter=C&tn=Cloisite %C2%AE 86. Dundigall a, A., LinGibson, S., Ferreiro, V., Malwitz, M.M., and Schmidt, G. (2005). Unusual M ultilayered S tructures in P oly(ethylene oxide)/Laponite Nanocomposite F ilms Macromol. Rapid Commun., 26: 143149. 87. Negrete Herrera, N., Putaux, J L., and Bourgeat Lami, E. (2006). Synthesis of P olymer/Laponite C omposite L atex P articles V ia E mulsion P olymerization U sing S ilylated and C ation E xchanged Laponite C lay P latelets Progr. in Solid State Chem., 34: 121137. 88. Kubies, D., Jrme, R., and Grandjean, J. (2002). Surfactant M olecules I ntercalated in Laponite as S tudies by C and Si Mas NMR Langmuir 18: 61596163. 89. Mongondry, P., Tassin, J.F., and Nicolai, T. (2005). Revised S tate D iagram of Laponite D ispersions J. Colloid and Interface Sci., 283: 397405. 90. Cussler, E.L. (2007). Diffusion B arrier, Diffusion Fundamentals 6:72.172.12. 91. Gusev, A.A., and Lusti, H.R. (2001). Rational D esign of N anocomposites for B arrier A pplications Adv. Mater., 13: 16411643. 92. Fredrickson, G.H., and Bicerano, J. (1999). Barrier P roperties of O rien ted D isk C omposites J. Chem. Phys., 110: 21812188. 93. Xu, B., Zheng, Q., Song, Y., and Shangguan, Y. (2006). Calculating B arrier P roperties of P olymer/ C lay N anocomposites: E ffects of C lay L ayers Polymer 47: 29042910.
203 94. Klute, C.H. (1959). Diffusion of S mal l M olecules in S emicrystalline P olymers: W ater in P olyethylene J. Appl. Polym. Sci., 1: 340350. 95. Saunders, J.M., Goodwin, J.W., Richardson, R.M., and Vincent, B. (1999). A S mallA ngle S cattering S tudy of T he S tructure of A queous Laponite D ispersions J. P hys. Chem. B 103: 92119218. 96. Bharadwaj, R.K. (2001). Modeling T he B arrier P roperties of P olymer L ayered S ilicate N anocomposites Macromolecules 34: 91899192. 97. Balazs, A.C., Singh, C., and Zhulina, E. (1998). Modeling T he I nteractions B etween P olymers and C lay S urfaces T hrough S elf C onsistent F ield T heory Macromolecules 31: 83708381. 98. Kuznetsov, D.V., and Balazs, A.C. (2000). Scaling T heory for E nd F unctionalized P olymers C onfined B etween T wo S urfaces: P redictions for F abricating P olymer/ C lay N anocomposi tes J. Chem. Phys., 112: 43654375. 99. Krook, M., Albertsson, A.C., Gedde, U.W., and Hedenqvist, M.S. (2002). Barrier and M echanical P roperties of M ontmorilonite/P olyesteramide N anocomposites Polym. Eng. And Sci., 42: 12381246. 100. Nah, C., Ryo, H.J., Kim, W.D., and Choi, S.S. (2002). Barrier P roperty of C lay/ A crylonitrile B utadiene C opolymer N anocomposite Polym. Adv. Technol., 13: 649652. 101. Kim, S.H., and Kim, S.C. (2007). Synthesis and P roperties of P oly(ethylene terephthalate)/ C lay N anocomposites by I n S itu P olymerization J. Appl. Polym. Sci., 103: 12621271. 102. Choi, W.J., Kim, S.H., Kim, Y.J., and Kim, S.C. (2004). Synthesis of C hain E xtended O rganifier and P roperties of P olyurethane/ C lay N anocomposites Polymer 45: 60456057. 103. Swain, S.K., and Isayev, A.I. (2007). Effect of U ltrasound on HDPE/ C lay N anocomposites: R heology, S tructure and P roperties Polymer 48: 281289. 104. Thellen, C., Orroth, C., Froio, D., Ziegler, D., Lucciarini, J., Farrell, R., DSouza, N.A., and Ratto, J.A. (2005). Influence of M ontmorilonite L ayered S ilicate on P lasticized P oly(L lactide) B lown F ilms Polymer 46: 1171611727. 105. Wang, Z.F., Wang, B., Qi, N., Zhang, H.F., and Zhang, L.Q. (2005). Influence of F illers on F ree V olume and G as B arrier P roperties in S tyrene B utadiene R ubber S tudie d by P ositrons Polymer 46: 719724.
204 106. Wang, Y.Q., Wu, Y.P., Zhang, H.F., Zhang, L.Q., Wang, B., and Wang, Z.F. (2004). Free V olume of M ontmorilonite/ S tyrene B utadiene R ubber N anocomposites E stimated by P ositron A nnihilation L ifetime S pectroscopy Macromol. Rapid Commun., 25: 1977319778. 107. Stephen, R., Ranganathaiah, C., Varghese, S., Joseph, K., and Thomas, S. (2006). Gas T ransport T hrough N ano and M icro C omposites of N atural R ubber (NR) and T heir B lends with C arboxylated S tyrene B utadiene R ubber (XSBR) L atex M embranes Polymer 47: 858870. 108. Cabedo, L., Gimnez, E., Lagaron, J.M., Gavara, R., and Saura, J.J. (2004). Development of EVOH K aolinite N anocomposites Polymer 45: 52335238. 109. Lu, C., and Mai, Y.W. (2005). Influence of A spect R atio on B arrier P roperti es of P olymer C lay N anocomposites Phys. Rev. Lett. 95: 088303(1) 008303(4) 110. Yano, K., Usuki, A., and Okada, A. (1997). Synthesis and P roperties of P olyimide C lay H ybrid F ilms J. Polym. Sci. A: Polym. Chem., 35 : 22892294. 111. Lape, N.K., Nuxoll, E.E., and C ussler, E.L. (2004). Polydisperse F lakes in B arrier F ilms J. of Mem. Sci., 236: 2937. 112. Beall, G.W. (2000). In Polymer Clay Nanocomposites ; Pinnavaia, T.J., and Beall, G.W. (e ds ), John Wiley & Sons Ltd, New York 113. Ray, S.S., Okamoto, K., and Okamoto, M. (20 03). Structure P roperty R elationship in B iodegradable P oly(butylenes succinate)/ L ayered S ilicate N anocomposites Macromolecules 36: 23552367. 114. Hambir, S., Bulakh, N., Kodgire, P., Kalgaonkar, R., and Jog, J.P. (2001). PP/Clay N anocomposites: A St udy of C r ystallization and D ynamic M echanical B ehavior J. Polym. Sci. B: Polym. Phys., 39: 446450. 115. Gilman, J.W. (1999). Flammability and T hermal S tability S tudies of P olymer L ayered S ilicate ( C lay) N anocomposites Appl. Clay Sci., 15: 3149. 116. Kojima, Y., Usuki, A., Kawasumi, M., Okada, A., Karauchi, T., and Kamigaito, O. (1993). Sorption of W ater in N ylon 6clay H ybrid J. Appl. Polym. Sci., 49: 12591264. 117. Messersmith, P.B., and Giannelis, E.P. (2003). Synthesis and B arrier P roperties of P oly( caprolactone) L ayere d S ilicate N anocomposites J. Polym. Sci. A: Polym. Chem., 33: 10471057. 118. Utracki, L.A., Sepehr, M., and Boccaleri, E. (2007). Review, Synthetic, L ayered N anoparticles for P olymeric N anocomposites (PNCs) Polym. Adv. Tech., 18: 137. 119. Ray, S., Quek, S.Y., E asteal, A., and Chen, X.D. (2006). The P otential U se of P olymer C lay N anocomposites in F ood P ackaging Intl. J. Food Eng., 2: 511.
205 120. Xu, W.B., Bao, S.P., and He, P.S. (2002). Intercalation and E xfoliation B ehavior of E poxy R esin/ Cu ring A gent/ M ontmorilonite N anocomposite. J. Appl. Polym. Sci., 84: 842849. 121. Solarski, S., Benali, S., Rochery, M., Devaux, E., Alexandre, M., Monteverde, F., and Dubois, P. (2005). Synthesis of A P olyurethane/ C lay N anocomposite U sed as C oating: I nteractions B etween T he C ounterions of C lay and T he I socyanate and I ncidence on T he N anocomposite S tructure J. Appl. Polym. Sci., 95: 238244. 122. Kato, M., Matsushita, M., and Fukumori, K. (2004). Development of A N ew P roduction M ethod for A P olypropylene C lay N anocomposite Polym. Eng. Sci., 44; 12051211. 123. Lyoo, W.S., and Ha, W.S. (1996). Structure and P roperties of M icrofibrillar P oly(vinyl alcohol) F ibres P repared by S aponification U nder S hearing F orce of P oly(vinyl pivalate) Polymer 37; 31213129. 124. Timofiychuk, O.A., Prudnicova, T.I., Mishurov, D.O., Avramenko, V.L., and Shevcovam R.G. (2009). Decomposition of Biodegradable Films Developed on The Basis of Poly(vinyl alcohol) in The Natural Environment, Phys, Chem and Technol., 7, 6167. 125. Cendoya, I., Lpez, D., Alegra, A., and Mijangos, C. (2001). Dynamic M echanical and D ielectrical P roperties of P oly(vinyl alcohol) and P oly(vinyl alcohol) B ased N anocomposites J. Polym. Sci. B: Polym. Phys., 39: 19681975. 126. Nakane, K., Yamashita, T., Iwakura, K., and Suzuki, F. (1999). Properties and S tructu re of P oly(vinyl alcohol)/ S ilica C omposites J. Appl. Polym. Sci., 74; 133138. 127. Suzuki, F., Nakane ,K., and Pias, J.S. (1996). Formation of A C ompatible C omposite of S ilica/P oly(vinyl alcohol) T hrough T he S ol G el P rocess and A C alcinated P roduct of T he C om posite J. Mater. Sci., 31; 13351340. 128. Lagaly, G. (1986). Smectitic Clays as Ionic Macromolecules Elsevier London, UK. 129. Lagaly, G. (1999). Introduction: F rom C lay M ineral P olymer I nteractions to C lay M ineral P olymer N anocomposites Appl. Clay Sci., 15: 19. 130. Tsai, B.C., and Jenkins, B.J. (1988). Effect of R etorting on T he B arrier P roperties of EVOH, J. Plast. Film Sheet 4: 6371. 131. Iwanami, T., and Hirai, Y. (1983). Ethylene V inyl A lcohol R esins for G as B arrier M aterial Tappi J., 66: 8590. 132. Schaper, E. (1989). Computer A ided B arrier D esign Y ields R etortable F ood P ackage, Modern Plastics 66: 9499. 133. Tendero, C., Tixier, C., Tristant, P., Desmaison, J., and Leprince, P. (2006). Atmospheric P ressure P lasma: A R eview Spectrochimica Acta Part B., 61: 230.
206 134. Singh N.L., Qureshi, A., Mukherjee, S., Rakshit, A.K., Tripathi, A., and Avasthi, D.K. (2007). Surface Modification of Polymeric Blends by Nitrogen Plasma Implantation, Surface & Coating Tech. 201: 82788281. 135. Maiti, M., and Bhowmick, A.K. (2006). New I nsights into R ubber C lay N anocomposites by AFM I maging. 47: 61566166. 136. Ishida, H., Campbell, S., and Blackwell, J. (2001). General A pproach to N anocomposite P reparation Chem. Mater., 12: 12601267. 137. Maiti, P., Nam, P.H., Hasegawa, M.ON., and Usuki, A. (2002). Inf luence of C rystallization on I ntercalation, M orphology, and M echanical P roperties of P olypropylene/ C lay N anocomposites Macromolecules 35: 20422049. 138. Wan, W., Yu, D., Xie, Y., Guo, X., Zhou, W., and Cao, J. (2006). Effects of N anoparticle T reatment on T he C rystallization B ehavior and M echanical P roperties of P olypropylene/ C alcium C arbonate N anocomposites J. Appl. Polym. Sci., 102: 34803488. 139. Lusti, H.R., Gusev, A.A., and Guseva, O. (2004). The I nfluence of P latelet D isorientation on T he B arrier P roperties of C omposites: A N umerical S tudy Modelling Simul. Mater. Sci. Eng., 12: 12011207. 140. Ogoshi, T., and Chujo, Y. (2005). Synthesis of A nionic P olymer S ilica H ybrids by C ontrolling pH in A n A queous S olution, Mater. Chem., 15: 315322. 141. Gudeman, L.F., and Peppa s, N.A. (1995). pH S ensitive M embranes from PVA/PAA I nterpenetrating N etweok J. of membrane sci., 107: 239248. 142. Cummins, H.Z. (2007). Liquid, G lass, G el: T he P hases of C olloidal Laponite J. NonCrystalline Solids 353: 38913905. 143. Daniel, L.M., Frost, R.L ., and Zhu, H.Y. (2008) Edge M odification of L aponite with D imethyloctylmethoxysilane J. Colloid and Interface Sci., 321: 302309. 144. Stefanescu, E.A., Stefanescu, C., Daly, W.H., Schmidt, G., and Negulescu, I.I. (2008). Hybrid P olymer C lay N anocomposites: A M echanical S tudy on G els and M ultilayered F ilms Polymer 49: 3785 3794. 145. Stefanescu, E.A., Daly, W.H., and Negulescu, I.I. (2008). Hybrid P olymer/ C lay N anocomposites: E ffect of C lay S ize on T he S tructure of M ultilayered F ilms Macromol. Mater. Eng., 293: 651656 146. Subramaniyan, A.K., and Sun, C.T. (2006). Enhancing C ompressive S trength of U nidirectional P olymeric C omposites U sing N anoclay Composites: Part A., 37: 22572268.
207 147. Wang, S., Zhang, Y., Ren, W., Zhang, Y., and Lin, H. (2005). Morphology, M echanical and O ptical P roperties of T ransparent BR/ C lay N anocomposites Polymer Testing 24: 766774. 148. Cui, L., Hunter, D.L., Yoon, P.J., and Paul, D.R. (2008). Effect of O rganoclay P urity and D egradation on N anocomposite P erformance, Part 2: Morphology and P roperti es of N anocomposites Polymer 49: 37623769. 149. Yoo, Y., and Paul, D.R. (2008). Effect of O rganoclay S tructure on M orphology and P roperties of N anocomposites B ased on A n A morphous P olyamide Polymer 49: 37953804. 150. Paul, D.R., and Robeson, L.M. (2008). Polym er N anotechnology: Nanocomposites Polymer 49: 31873204. 151. Dasari, A., Yu, Z. Z., Mai, Y.W., Hu, G.H., and Varlet, J. (2005). Clay E xfoliation and O rganic M odification on W ear of N ylon 6 N anocomposites P rocessed by D ifferent R outes Composites Sci. Tech., 65: 23142328. 152. Bhushan, B., and Qi, J. (2003). Phase C ontrast I maging of N anocomposites and M olecularly T hick L ubricant F ilms in M agnetic M edia, Nanotechnology 14: 886895. 153. Maurer, J.J., Eustace, D.J., and Ratcliffe, C.T. (1987). Thermal C haracterizatio n of P oly(acrylic acid) Macromolecules 20: 196202. 154. Caykara, T., and Guven, O. (1998). Effect of P reparation M ethods on T hermal P roperties of P oly(acrylic acid)/ S ilica C omposites J. Appl. Polym. Sci., 70: 891895. 155. Jiang, S.H., Huang, Y., and Shen, J.R. (2003). Effects of C holesteric O rder on M echanical P roperties of (E CE)C/PAA C omposites J. Appl. Polym. Sci., 57 : 493498. 156. Paralikar, S.A., Simonsen, J., and Lombardi., J. (2008). Poly(v i nyl alcohol)/C ellulose N anocrystal B arrier M embranes J. Memb. Sci. 320: 248258. 157. Osman, M.A., Ploetze, M., and Skrabal, P. (2004). Structure and P roperties of A lkylammonium M onolayers S elf A ssembled on M ontmorillonite P latelets J. Phys. Chem. B., 108: 25802588.
208 BIOGRAPHICAL SKETCH Jinwoo Kwak was born in Daegu Ko rea, in January 1977. He received his bachelors degree in textile and polymer chemistry from Yeungnam University Korea in 2003. After that, he joined advanced polymer lab at the same university and finished courseworks for master s degree there. His research interest in Korea was mainly focused on the preparation of high molecular weight polymer materials with various morphological features. In August 2006, he was admitted to the graduate school of the U niversity of Florida in the D epartment of M aterials Science and E ngineering, where he received his master s degree in May 2008. The research throughout his master and doctoral courses was focused on polymer clay nanocomposites.