<%BANNER%>

Rheological and Abrasion Resistant Properties of Transparent Polymer/Silicate Nanocomposite Coatings


PAGE 1

RHEOLOGICAL AND ABRASI ON RESISTANT PROPERTIES OF TRANSPARENT POLYMER/SILICATE NANOC OMPOSITE COATINGS By JENNIFER MARIE BRANDT A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Jennifer Marie Brandt

PAGE 3

This document is dedicated to Michael Sachs, Kristin Brandt, and my parents, David and Marie Brandt.

PAGE 4

iv ACKNOWLEDGMENTS I would like to express my appreciation to my committee chair and advisor, Dr. Abbas Zaman. I would also like to thank Dr. Zamans graduate student, Heather Trotter, and his group of undergraduate students. Speci al thanks go to Betty Arias for her hard work on may aspects of this work. I would like to thank my committee memb ers Dr. Charles Beatty and Dr. Susan Sinnott for their advice and support. I would also like to than k the students and staff of the Particle Engineering Research Center at the University of Florida. I would like to thank the Air Fo rce for their financial support.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS..................................................................................................iv TABLE OF CONTENTS.....................................................................................................v LIST OF TABLES.............................................................................................................vii LIST OF FIGURES..........................................................................................................viii ABSTRACT........................................................................................................................ix 1 INTRODUCTION........................................................................................................1 1.1 Polymer Nanocomposites.......................................................................................1 1.2 Literature Review...................................................................................................4 1.3 Project Background Information............................................................................6 2 RHEOLOGICAL PROPERTIES AND STRUCTURAL DEVELOPMENT..............8 2.1 Introduction.............................................................................................................8 2.2 Materials...............................................................................................................11 2.3 Experimental Procedure........................................................................................13 2.3.1 Sample Preparation.....................................................................................13 2.3.2 Rheological Measurements........................................................................14 2.3.3 Microscopy.................................................................................................15 2.4 Discussion of Results............................................................................................16 2.4.1 Shear Viscosity...........................................................................................16 2.4.1.1 Effect of shear rate on viscosity.......................................................16 2.4.1.2 Effect of solids loading on viscosity................................................21 2.4.2 Viscoelastic Properties a nd Structural Development.................................22 2.4.3 Microscopy.................................................................................................26 3 SURFACE ACTIVATION.........................................................................................30 3.1 Introduction...........................................................................................................30 3.2 Materials...............................................................................................................32 3.3 Experimental Procedure........................................................................................33 3.3.1 Sample Preparation.....................................................................................33

PAGE 6

vi 3.3.2 Fourier Transform Infrar ed Spectroscopy (FTIR)......................................34 3.4 Discussion of Results............................................................................................35 3.4.1 Acid Surface Treatments............................................................................35 3.4.2 Silane Coupling Agent Treatments............................................................39 3.4.3. Film Casting onto Treated Interlayer Substrates.......................................40 4 ABRASION RESISTANT AND ADHESIVE PROPERTIES..................................42 4.1 Introduction...........................................................................................................42 4.2 Materials...............................................................................................................43 4.3 Experimental Procedure........................................................................................44 4.3.1 Sample Preparation.....................................................................................44 4.3.2 Taber Linear Abrader.................................................................................44 4.3.3 Visual Observation of Film Delamination.................................................45 4.4 Discussion of Results............................................................................................46 4.4.1 Multi-Layered Films...................................................................................46 4.4.2 Delamination of Coated Interlayer Samples...............................................50 5 CONCLUSIONS AND FUTURE WORK.................................................................52 5.1 Conclusions...........................................................................................................52 5.2 Future Work..........................................................................................................54 LIST OF REFERENCES...................................................................................................55 BIOGRAPHICAL SKETCH.............................................................................................58

PAGE 7

vii LIST OF TABLES Table page 2-1: Several shear rates corresponding to co mmon tape casting gap sizes (calculated from casting speed/gap size)....................................................................................11 2-2: Chemical composition of Laponite nanoparticles13...................................................12 2-3: Physical properties of Laponite nanoparticles13.........................................................13 2-4: PVA/Laponite dispersion and film characteristics.....................................................17 3-1: Characteristics of Dupont Sent ryGlas Plus ionomer interlayer...............................32 3-2: Absorption frequencies for co mmon bonds found in OMNIC software....................36

PAGE 8

viii LIST OF FIGURES Figure page 1-1: Various phases observed for polymer-layered silicate systems...................................3 1-2: Dimensions of a Laponite particle 13............................................................................7 2-1: Viscosity as a function of shear rate for dispersions of 10wt% Laponite with increasing polymer dosages (25C)..........................................................................17 2-2: Viscosity as a function of shear rate for dispersions of Laponite JS in 2 wt% PVA solution (25C) at different wt% solids: ( ) 0 wt% ( ) 1.3 wt% ( ) 1.96 wt% ( ) 2.91 wt%.............................................................................................................18 2-3: Shear viscosity as a func tion of shear rate and solids loading for dispersions made from 1 wt% PVA Solution (25C)...........................................................................19 2-4: Viscosity as a function of shear rate for dispersions of 4 wt% Laponite prepared in 1 wt% PVA and 3 wt% PVA solutions (25C) ( ) 1 wt% PVA ( ) 3 wt% PVA..........................................................................................................................20 2-5: Relative viscosity (with respect to th e viscosity of the suspending fluid) as a function of Shear rate for dispersions of 4 wt% wt Laponite dispersions prepared in 1 wt% PVA and 3 wt% PVA solutions (25C) ( ) 1 wt% PVA ( ) 3 wt% PVA..........................................................................................................................21 2-6: Viscosity as a function of solids load ing for dispersions of Laponite particles prepared in a 1 wt% PVA solution at two different shear rates ( ) 1 s-1 ( ) 1000 s-1..............................................................................................................................2 2 2-7: Storage modulus vs. a ngular frequency and polymer dosage for dispersions of 10 wt% Laponite particles (25C).................................................................................23 2-8: Storage modulus as a function of angul ar frequency for dispersions of Laponite prepared in 2 wt% PVA solution (25C) ( ) 0 wt% ( ) 1.3 wt% ( ) 1.96 wt% ( ) 2.91 wt%.............................................................................................................24

PAGE 9

ix 2-9: Storage modulus as a function of angul ar frequency for dispersions of Laponite particles prepared in 3 wt% PVA solution (25C) ( ) 0 wt% ( ) 1.96 wt% ( ) 2.91 wt% ( ) 4.3 wt%..............................................................................................25 2-10: Loss modulus as a function of angular frequency for dispersions of Laponite particles prepared in 2 wt% PVA solution (25C) ( ) 0 wt% ( ) 1.3 wt% ( ) 1.96 wt% ( ) 2.91 wt%.............................................................................................26 2-11: Loss modulus as a function of freque ncy for dispersions of Laponite particles prepared in 3 wt% PVA Solution (25C) ( ) 0 wt% ( ) 1.3 wt% ( ) 1.96 wt% ( ) 2.91 wt%.............................................................................................................27 2-12: SEM micrograph of flocculated La ponite particles cast from a 3 wt% PVA dispersions with 50 wt% Laponite mass fraction in dried film................................27 2-13: AFM images of dry films cast from a 3 wt% PVA matrix with final solids loadings of A.) 0 wt% B.) 40 wt% C.) 50 wt% D.) 60 wt%....................................29 3-1: Schematic of Attenuat ed Total Reflection (ATR)30...................................................32 3-2: Infrared spectrum for the surface of an untreated interlayer sample .........................36 3-3: Infrared spectrum for th e surface of an interlayer sample treated with 100% sulfuric acid solution................................................................................................37 3-4: Infrared spectrum comparison for interlay er samples a) without acid treatment b) 10% sulfuric acid treatment c)100% sulfuric acid treatment ..................................38 3-5: Infrared spectrum comparison for interlay er samples a) without acid treatment b) 2:1 sulfuric/nitric acid treatment c) 3:1 sulfuric/nitric acid treatment .....................39 3-6: Infrared spectrum comparison for interl ayer samples a) without silane treatment b) 3-aminopropyl trimethoxysilane treatme nt c) N-(2-aminoethyl)3-aminopropyl trimethoxysilane treatment ......................................................................................40 4-1: Steel rods of different diameters.................................................................................46 4-2: Average weight loss data with standa rd deviation for abraded glass substrate and substrate coated with 40, 50, and 60 wt% nanocomposite films..............................47 4-3: Average weight loss data with standa rd deviation for abraded glass substrate and substrate coated with two layers of na nocomposite film (x/y: film x coated over film y).......................................................................................................................4 8 4-4: Average weight loss data with standa rd deviation for abraded glass substrate and substrate coated with three layers of nanocomposite film (x/y/z: film x coated over film y coated over film z).................................................................................49

PAGE 10

x 4-5: A comparison of average weight loss da ta with standard deviation for the films which loss less than or equal to the amount of weight as the glass substrate..........50

PAGE 11

xi Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science RHEOLOGICAL AND ABRASI ON RESISTANT PROPERTIES OF TRANSPARENT POLYMER/SILICATE NANOC OMPOSITE COATINGS By Jennifer Marie Brandt May 2005 Chair: Abbas A. Zaman Major Department: Materials Science and Engineering Polymer nanocomposite thin films were cast from colloidal dispersions of polyvinyl alcohol and Laponite JS Rheological studies were performed to determine the effect of shear rate and small amplitude oscillatory shear on the colloidal dispersions. Structural development of the dispersions was determined as a function of polymer concentration and Laponite solids loading. Solu tion casting of the dispersions resulted in transparent coatings with strong adhesion to a glass substrate. Scanning electron microscopy revealed the presence of floccu lated particle clusters in the films. Adhesion of the films to a hydrophobic polymer interlayer was achieved by activating the surface of the polymer using wet-chemical treatments. The treatments included mixtures of sulfuric and nitric ac id and several silane coupling agents. The success of the treatments was determined using FTIR-ATR surface analysis. Highly concentrated acid treatments and several coupling agents were determined to be successful in surface modification of the polymer.

PAGE 12

xii To determine the abrasion resistance of the nanocomposite films, a Taber linear abrader was used to abrade the samples and weight loss was observed. Samples with single, double, and triple coa tings of film with solids loadings of 40, 50, and 60 wt% were cast onto glass sl ides and abraded by the instrument Of these samples, the doublecoated films lost the least weight when comp ared to the glass slide. The films containing high solids loadings had the best perf ormance and the most reproducibility.

PAGE 13

1 CHAPTER 1 INTRODUCTION 1.1 Polymer Nanocomposites A polymer nanocomposite is defined as an or ganic or inorganic phase with at least one dimension on the nanometer scale dispersed in a polymer matrix. Polymer Nanocomposites (PNCs) are used because they have shown to have enhanced material properties including improved barrier propert ies, tensile modulus and strength, flame resistance, abrasion resistance, and re duced shrinkage and residual strength1. Many industries are taking advantag e of nanocomposite technologies because of recent findings regarding the low weight % of clay needed to improve propert ies. This makes it possible to create lightweight films a nd coatings for packaging with increased barrier properties. There is estimated to be a 2 order of magn itude growth of the PNC industry by the year 2005. The following are examples of a few other uses of PNCs in industry:1,2 Fire Resistance the NIST is developing fi re resistant and reduced char coatings for windows and other applications. Asphalt Modification Exxon is developi ng asphalt mixtures with improved mechanical properties using nanoparticles in mixing. Elastomers many tire companies are interested in using PNCs as tire reinforcements. Thermoset Polymers many industries are in terested in films and coatings with improved barrier properties, abrasion resistance, and thermal stability. Polymer nanocomposites can be isotropic or anisotropic and can have a defined structure and orientational or der. Polymer nanocomposites ar e usually grouped into three categories based on the dimensions of th e dispersed phase. Some nanoparticles are isodimensional, or have three dimensions in the nanometer range; these include all spherical, disc-shaped, or clustered particle s. When there are two dimensions in the

PAGE 14

2 nanoscale you can have an elongated disper sed phase such as carbon nanotubes or whiskers, which are commonly used as fillers in nanocomposites. Lastly, if there is one dimension in the nanoscale the long sheets of filler can be stacked to create a layered composite. This is achieved by the swelling of the polymer chains in between the layered sheets. Layered silicates such as Laponite are isodimensional and can be dispersed in a variety of ways.3-5 In layered-silicate systems, the particle s can be arranged in several different structures depending on their method of prepar ation. When the system is phase separated, there are clumps of particles dispersed in a pa rticle matrix with the particle and polymer phases being immiscible. In an intercalated system, the particles are arranged in layers with polymer chains swelling in between the stacked particle galleries, causing them to separate slightly. When these stacked partic les are separated and dispersed throughout the polymer matrix, the system is exfoliated. The exfoliated structure is very desirable when attempting to make a well dispersed system w ith uniform properties. Figure 1-1 is an illustration of the different type of layered-silicate systems.4-5 There are several ways to prepare a polymer-layered silicate nanocomposite system to achieve intercalation or exfoliati on. In situ polymerization can be used to intercalate a system by swelli ng the layered silicate partic les in a liquid monomer and then initiate polymerization to form polymer chains between the particle sheets. In melt intercalation, if a polymer is compatible with the particle su rface, the system is intercalated in the molten st ate during processing. An exfoliated system can also be achieved if the polymer is able to get in between the particle spacing. Layered silicates can also be exfoliated by dispersing the pa rticles in a soluble polymer matrix where

PAGE 15

3 exfoliation occurs due to delami nation of the stacked particle s. This is the most common method for creating an exfoliated system.5 Layered Silicate Particles Polymer Chains Phase Separated System Intercalated System Exfoliated System Layered Silicate Particles Polymer Chains Phase Separated System Intercalated System Exfoliated System Figure 1-1: Various phases observed for polymer-layered silicate systems In a Polyvinyl Alcohol (PVA) system containing Laponite, the particles are dispersed in the PVA matrix and a gel is form ed. When the gel is dried into a film, most of the particles will become intercalated, but steric interactions in the PVA can prevent reaggragation if the molecula r weight is sufficiently high.5 The use of water-soluble polymers in these dispersions is advantageous because of environmental and safety issues which are encountered when using common orga nic solvents. Because of its high aspect ratio, Laponite has the ability to form a stable sol when dispersed in demineralized water at low solids loadings. When mixed with a polymer, a colloidal gel is formed.6-12

PAGE 16

4 1.2 Literature Review Numerous studies have been conducted on the nanocomposite system composed of a polymer matrix with differe nt grades of Laponite as the dispersed phase. Most of the information available on Polymer/Laponite na nocomposites is based on the polyethylene oxide (PEO)/Laponite system. Similar to PVA, PEO is a water soluble polymer that is inexpensive and available in a wide range of molecular weights. When dispersed with surface modified Laponite particles, PEO has been shown to adsorb to the particle surface, creating bridging between the particles. When shear is applied to this mixture, the dispersion will become flocculated and a gelled network will form. These dispersions can also exhibit relaxation be havior when aged for a cert ain amount of time, which is dependent on the concentration of the PEO. A recent study was done on the shear thickening and gelation of the PEO/Laponite system with applied shear.6 In this work, the formation of a shake-gel was observed when a low viscosity sol com posed of low concentrations of PEO and Laponite was vigorously shaken. The phase be havior of the gels, based on visual observation, is dependent on both polymer c oncentration and Laponite solids loading. The gelation mechanism for this system is the bridging between polymer chains when particle surfaces are exposed during the applicat ion of shear. At very high concentrations of PEO, the Laponite surface will become satu rated and a gel will not form. If the PEO concentration is low, the dispersion will re main as a low-viscosity sol, so these boundaries establish a regime where a rigid gel can be formed. The authors performed light scattering experiments to determine the adsorption behavior of PEO onto the Laponite surface. Their data on the surface cove rage proved to be consistent with the phase behavior that was observed.

PAGE 17

5 Another study gave further insight into th e network structure that can be induced by the application of shear to a PEO/Laponite system.7 In this study, th e investigators attempted to quantify the critical shear rate needed for shear-induced reorientation of the Laponite particles so that a viscous gel can be formed. The authors also studied the transitions from liquid like to gel like beha vior that could be observed for these dispersions. Oscillatory shear e xperiments were done to find th is transition, which is the point where the complex moduli intersect. Another study examined the reversible gel behavior of the PEO/Laponite XLG system of low solids loadings.8 Their study showed a composition regime where aggregates can be deformed by applied shea r thus exposing new surface area, which leads to the formation of polymer bridges. This br idging, or flocculation, causes the gelation of the system. When shear is stopped, these polymer bridges can break up causing a relaxation in the storage modulus and reversible gel behavior This study also determined that the reversible sol-gel transition shows time dependent characteristics. In a study done on solid nanocomposites, di spersions of Laponite in PEO were cast onto glass substrates to form very thin films.9 The films were cast from PEO solutions of 0, 2, and 5 weight % with a Laponite mass fraction of 3 weight %. The resulting films were observed to be transparent with good interfacial adhesion. The surface roughness and morphology of the films were characte rized using Atomic Force Microscopy. The AFM images and the RMS roughness calculations showed that the particles are randomly dispersed in the film with polymer chains connecting them. The films containing low concentrations of PEO had this homogeneous structure with partic les being roughly equal in size. The film cast from the 5 wt% PEO solution was observed to be heterogeneous,

PAGE 18

6 with agglomerated Laponite domains separate d by excess PEO. It was also observed that the roughness decreased as the PEO concentrat ion increased because the excess polymer creates a smoother surface. The films containi ng low concentrations of PVA are observed to be exfoliated and randomly oriented w ith a high degree of surface roughness. 1.3 Project Background Information Since the development of polymer nanocomposites, a wide variety of applications for these materials have been identified. Becau se of the improved properties that can be achieved at relatively low cost, stronger coa tings and laminates can be made for military and security purposes. With the recent threat s of terrorism, hurricanes, and forced entry crimes, new technologies in protective build ing materials have been developed. Dupont has developed several glass laminates that provide protection from these dangers using strong, transparent interlayers. For milita ry applications i nvolving bomb and blast protection, an abrasion resistant material is needed for protecti on against sharp glass shards. These small pieces of glass can become deadly projectiles when hit with a strong blast. The strong ionomer interlayers can prev ent large blasts from destroying a building, but a large amount of destructi on and loss of life is caused by these shards breaking off as a result of the blast. This creates a need fo r an adhesive, abrasion resistant coating on the polymer interlayer which will reduce or elim inate this phenomenon while maintaining the abrasion-resistance of glass. The abrasion resi stant coating should be clear, flexible, and have a strong adhesion to the inte rlayer and the glass being used. There are many nanocomposite systems that c ould be used to fabricate this coating. Laponite is a good choice for a filler phase b ecause of its high aspect ratio and its transparency when suspended in water. La ponite particles are water white, which means that they are small enough to be unabl e to scatter light. A diagram of a single

PAGE 19

7 Laponite particle is shown in figure 1-2. Surf ace treated Laponite particles can be mixed with a water-soluble polymer to create a stable dispersion. 25nm 0.92nm 25nm 0.92nm Figure 1-2: Dimensions of a Laponite particle 13 Depending on the polymer being used, a strong transparent film can be cast using a variety of casting methods. Polyvinyl Alcohol is known to form strong, chemical resistant films with abrasion-resistant properties. The mixture of PVA with Laponite is primarily used as a paper coating, but a system using a high molar mass PVA can form transparent films with good tensile properties.5

PAGE 20

8 CHAPTER 2 RHEOLOGICAL PROPERTIES AND STRUCTURAL DEVELOPMENT In the processing of thin, nanocomposite films, it is necessary to consider the rheological properties of the dispersions being used to cast the films. Changes in viscosity and the development of a gelled network can influence the thickness, transparency, and properties of the final film. This chapter invest igates the rheological properties of various PVA/Laponite dispersions and the effect of solids loading and polymer dosage on shear viscosity and structur al development. 2.1 Introduction Colloidal gels are formed when a stable liquid suspension, or sol, of colloidal particles is gelled using a chemical agent. In Laponite suspensions, this gelling agent is usually a filler, binde r, pigment, surfactant, or wetting agent.12,13 When Laponite particles are initially dispersed in water, sodium ions present on the dry particle surface are held between the negatively charged particles by elec trostatic forces. Thes e dispersions exhibit low-viscosity, Newtonian rheology.13 Some Laponite grades have surface treatments that enhance the sol-forming capability of the dispersion and provide electrostatic stabilization of the sol. Laponite JS, which is used in this study, is modified with a tetrasodium pyrophosphate (TSPP) surface treatment. The (P2O4)+ anions create a net negative charge on the particles which cause s the loose layer sodium ions create repulsions between particles.5,13 These stable sols can then be gelled or flocculated with the addition of a chemical agent.

PAGE 21

9 Gelling agents usually come in the form of a polymer, electrolyte, or pH modifier. These compounds reduce repulsive particle-par ticle interactions, cau sing thinning of the electrical double layer pres ent in the sol. When th e electrical double layer is sufficiently neutralized, van der Waals attrac tive forces draw the particles together causing a solid-like gel phase to form. The addi tion of polymer to a stabilized colloidal dispersion will change the properties of the dispersion depending on whether the polymer is nonadsorbing or strongly adsorbing. When a nonadsorbing polymer is added to a stable sol it can induce depletion flocculation. This occurs when two particles approach each other to a distance smaller than the radius of gyration of the pol ymer chain, thus the polymer is excluded from the interparticl e zone. The osmotic pressure between the particles is reduced causing an attractiv e force to develop between particles.12,14 As flocculation proceeds, particle pairs ( doublets) become particle triplets and so on until flocs containing many groups of particle s start to appear. If the flocs are densely packed, they would form a close-packed solid structure. Instead, most flocs are open, ramified structures, or fractals, which are self-similar structures which look the same at all magnifications. As a floc grows, the por osity increases, creating a much larger, open structure. The floc can eventually grow to fill up the volume fraction of a sample, forming a percolated network, or gel.14,15 The growth can be diffusion limited, where small flocs form to create larger flocs, or it can be reaction limite d, where the density of the floc grows over time. The flocculation be havior of colloidal gels can effect the rheological behavior of the dispersion. Understanding the rheology of gels is very important when processing these materials. Their response to sh ear rate, shear stress, and strain depend strongly on particle

PAGE 22

10 size and shape as well as th e gel strength and degree of flocculation. The rheological behavior of flocculated gels is difficult to reproduce due to their shear history and time dependent structure.16-23 Two types of tests can be performed on colloidal gels to determine their flow properties, these are steady-shear viscosity and small amplitude oscillatory shear flow measurements. Steady sh ear viscosity is the most widely measured aspect of rheology and can determine wh ether a material is Newtonian or nonNewtonian. Using a parallel disk torsional rheometer, only a small amount of sample is needed to obtain data about the behavior of the material. Small Amplitude Oscillatory Shear (SAOS) experiments are useful in dete rmining the viscoelastic properties of a colloidal gel. SAOS can be used to observe structural development, relaxation behavior, and various transitions for colloidal gels.21-23 Many rheological studies have been conduc ted on a colloidal system containing Laponite of different grades in a matrix of pol yethylene oxide (PEO). In these systems, if the polymer concentration is not enough for full coverage, the PEO is adsorbed onto the bare surfaces of the Laponite particles and a gel forms as a result of bridging flocculation.24 When a strong shear is applied to a PEO/Laponite suspension, the polymer chains can desorb from the particles and r eadsorb onto a different particle, forming a shake gel.6 Modulus relaxation data shows that th ese shake gels ar e reversible and have aging characteristics.8 The mixture of Laponite JS (LJS) with low molecular weight Polyvinyl Alcohol (PVA) has been used in the paper coatings industry to form coherent films with good barrier properties.25 Relatively little is known about the rheological properties of a PVA/Laponite mixture and the effect these prop erties have on film processing. There are

PAGE 23

11 several methods that can be used to cast the PVA/Laponite films and the rheological behavior of the dispersions determines wh ich methods are feasible. A low-shear method would be solution casting, where the dispersion is poured over the substrate and set to dry until the solvent is evaporate d. Another technique would be tape casting, which can go to very high shear rates. Some examples of sh ear rates at a speed of 10inches/minute are shown in Table 2-1. To obtain a very thin film, on the order of 10m, the shear rate would be very high. 2-1: Several shear rates corresponding to comm on tape casting gap sizes (calculated from casting speed/gap size) Gap Size Shear Rate 3mm 1.41 s-1 1mm 4.23 s-1 0.5mm 8.46 s-1 0.1mm 42.33 s-1 0.05mm 84.66 s-1 0.01mm 423.3 s-1 In this study, a high molecular weight P VA is used as the matrix polymer for Laponite JS dispersions which will be used to make thin, optically clear coatings for glass and polymer substrates. 2.2 Materials The polymer used in this study is a pa rtially hydrolyzed grade of Polyvinyl Alcohol. The PVA (Aldrich Chemical Co. cat # 36310-3) has a molecular weight between 126,000-186,000 g/mol. PVA is a neutral, linear polymer synthesized from a vinyl acetate monomer with a chemical formula given by [CH2 CHOH]n. The PVA used in this study has a degree of hydrolysis of 87-89% which is the extent of conversion from polyvinyl acetate to polyvinyl alcohol. The PVA is soluble in water and can easily form

PAGE 24

12 thin films with good tensile st rength properties on glass s ubstrates. The films have good chemical resistance and resist ance to thermal degradation. The clay particle used in this study is Laponite (Southern Clay Products, Inc. Gonzales, TX). Laponite particles are synt hetic layered nanosilicate clays with an empirical formula given by: Na+0.7[(Si8Mg5.5Li0.3)O20(OH)4)]-0.7.13 The Laponite crystals are disk-shaped and are stacked when dry. Th e disks have a diameter of 25 nm and a width of 0.92 nm.13 The unit cell consists of one layer of octahedral magnesium atoms with a small amount of lithium impurities. Th is layer is sandwiched between two layers of tetrahedral silicon atoms. Both laye rs are balanced by 20 oxygen atoms, 2 hydroxyl groups, and sodium ions which are released when the particles ar e dispersed in water.13 Some chemical and physical pr operties of Laponite JS are shown in Tables 2-2 and 2-3. The grade of Laponite used for this study is Laponite JS, which is surface treated with tetrasodium pyrophosphate, an i norganic polyphosphate dispersi ng agent. All grades of Laponite form clear colloidal dispersions when mixed with water. 2-2: Chemical composition of Laponite nanoparticles13 Material Weight % SiO2 50.2 MgO 22.2 Li2O 1.2 Na2O 7.5 PsO5 5.4 F 4.8

PAGE 25

13 2-3: Physical properties of Laponite nanoparticles13 Property Value Appearance Free flowing white powder Bulk Density 950 kg/m3 Surface Area (BET) 300 m2/g Loss on ignition 8.70% pH (2% suspension) 10 Storage Hygroscopic, should be stored under dry conditions 2.3 Experimental Procedure 2.3.1 Sample Preparation The polymer stock solutions are made by dissolving PVA granules in deionized water at room temperature and stirring at moderate speeds for 24 hours with a magnetic mixer. The Laponite dispersions are made by mixing 200 mL of polymer stock solution with a given amount of Laponite (according to final solids lo ading in films) at pH 10. These dispersions are mixed with a high sh ear mixer (Hamilton Beach Commercial Shear Mixer 950) to ensure good mi xing and prevent aggregation. Defoaming agents are used when needed to prevent viscous bubbles fr om forming, which can alter the rheological results and weaken the final films. The polym er/clay dispersions are agitated on a shaker for 48 hours prior to rheological measurement to ensure good mixing. The samples were kept mixing until they were measured to avoid effects of time dependency, which is common for polymer/Laponite dispersions.6-8 The nanocomposite film samples were made by solution casting the PVA/Laponite gel dispersions onto glass petri dishes at 35C. They were left to evaporate in a convection oven for 24 hours, or until they were observed to be completely dry. The

PAGE 26

14 film quality is improved by this slow evaporation process since the clay platelets have more time to assemble under gravitational and osmotic forces16. These nanocomposite films are optically clear with th e Laponite phase trapped in th e thin film resulting in good interfacial adhesion to the glass substrate.9 2.3.2 Rheological Measurements A Paar Physica UDS 200 Rheometer was used to examine the steadyand oscillatory-shear-flow using a parallel plate geometry (plate radius, 2.5cm). Steady shear flow experiments were conducted to determine steady shear viscosity ( ) as a function of shear rate ( ) with shear rates ranging from 1-5000 s-1. Oscillatory shear flow experiments were performed to determine the vi scoelastic properties of the dispersions. The storage (G ) and loss (G ) moduli are related to the real and imaginary components of the complex viscosity, and respectively. With a given angular frequency ( ) and phase shift ( ) they can be calculated as follows: G G tan 2.1 G 2.2 G 2.3 And finally, the complex moduli can be related to the complex viscosity by G i G G i * 2.4 Where 1 i. An angular frequency range of 1-50 Hz was chosen with a fixed strain amplitude of 5% to maintain the measurements in the linear viscoelastic region.26-28 Sample evaporation is prevented by using a circular solvent trap covered with an

PAGE 27

15 aluminum cap, which isolates the sample during measurement. All experiments were performed at a temperature of 25C. There are several sources of error that must be considered when performing rheological measurements. Factors such as humidity, sample size, and ambient temperature can have an effect on the samp le behavior. For the following experiments, each batch of samples had one repeat for ev ery three samples run. This repeat was compared to the original data to verify the reproducibility of the experiment within +/3%. Since Laponite dispersions are known to have a time-dependency,8 all samples were kept on a shaker until use. 2.3.3 Microscopy Scanning Electron Microscopy (SEM) was done on the films to determine the structure of the laponite particles in the cast films. Images were taken using a JEOL JSM6335F Scanning Electron Microscope to exam ine the structure inside a solution cast nanocomposite film with a final solids lo ading of 50 wt% from a Laponite dispersion prepared in a 3% PVA matrix. The film sa mple was mounted onto an aluminum mount and coated with Gold-Palladium to impr ove conductivity. The image was taken at a 2000x magnification and a beam energy of 1kv to prevent destruction to the polymer matrix of the film. Atomic Force Microscopy (AFM) wa s performed to examine the surface morphology of the nanocomposite films. The film s used for AFM were also cast from a 3 wt% PVA matrix with Laponite solids loading of 0, 40, 50, and 60 wt%. The measurements were done at a setpoint of 2 volts a scan rate of 3 Hz, and a scan size of 1 m.

PAGE 28

16 2.4 Discussion of Results 2.4.1 Shear Viscosity 2.4.1.1 Effect of shear rate on viscosity. The steady shear viscosity ( ) as a function of shear rate ( ) was measured in three dispersion systems, one which varied polymer dosage, one which looked at final solids loading in cast films, and one system which varied Laponite solids loading. The effect of shear rate on the steady shear viscosity of dispersions cont aining Laponite nanoparticles di spersed in solutions of PVA at dosages ranging from 1 to 12 mg/(g solids) and a Laponite loading of 10 wt% is shown in Figure 2-1. It can be observed at low pol ymer dosages the Laponite can form a stable sol which exhibits a constant viscosity ( o) and Newtonian behavior in the shear rate range examined. At high polymer dosages, the viscosity of the dispersions decrease with increasing shear rate, i.e., they exhibit shear thinning, or thixotropi c, behavior at high shear rates. These data show that with in creasing polymer dosage th ere is almost four degrees of magnitude increase in viscosity, wh ich can be seen for dosages above 3mg/(g solids). As polymer is added to the Laponite dispersions, they swell between the particles and cause an increase in the zero shear viscos ity. With an increase in applied shear rate, the particles and polymer chains align themse lves in a more desirable flow structure, causing shear thinning in the system. The Laponite solids loadings used in Figur e 2-2 are the actual concentrations of filler that would be used to process nanocomposite films with final solids loadings of 40 wt%, 50 wt%, and 60 wt%. These samples were prepared in a 2 wt% PVA solution as the suspending media and are high enough to pr oduce films of good tensile strength. The characteristics of these films are shown in Table 1. The data in Figure 2 shows the relationship between shear visc osity and shear rate for seve ral calculated Laponite solids

PAGE 29

17 loadings in 2 wt% PVA solution. The dispersi ons containing very low weight fractions of Laponite maintain a constant viscosity ( o) and behave as Newtonian fluids throughout the range examined. The further addition of Laponite causes an increase in the zero shear viscosity ( o) and the onset of shear thinning shifts to lower shear rates. 0.01 0.1 1 10 100 1000 110100100010000 shear rate, 1/s Pas 1mg/g solids 2mg/g solids 3mg/g solids 5mg/g solids 8mg/g solids 12mg/g solids Figure 2-1: Viscosity as a function of shear rate for dispersions of 10wt% Laponite with increasing polymer dosages (25C). Table 2-4: PVA/Laponite dispersion and film characteristics Wt% PVA in water PVA Mass Fraction in dried films Mass fraction Laponite in Solution ( wt%) Laponite mass fraction in dried films 2% 60% 1.3 40% 2% 50% 1.96 50% 2% 40% 2.91 60% 3% 60% 1.96 40% 3% 50% 2.91 50% 3% 40% 4.3 60%

PAGE 30

18 0.001 0.01 0.1 1 10 100 110100100010000shear rate, 1/s Pas Figure 2-2: Viscosity as a function of shear rate for dispersions of Laponite JS in 2 wt% PVA solution (25C) at different wt% solids: ( ) 0 wt% ( ) 1.3 wt% ( ) 1.96 wt% ( ) 2.91 wt%. The shear viscosity data are shown in Figure 2-3 for dispersions of Laponite particles at solids loadings ranging from 0-5 wt% prepared in a 1 wt% PVA stock solution. The viscosity shows changes with both shear rate and volume fraction of the particles. At a fixed weight fraction and at low shear rates, all samples show Newtonian behavior. For these low weight fractions, lower shear rates are not in the linear range of the instrument, and are therefore left out of the data. At moderate shear rates, the viscosity falls monotonically with increasing shear rate and there is no indication of a second plateau at high shear rate s. There is at least three or ders of magnitude change of viscosity as the solids loading is increased A transition from a sh ear rate independent region (Newtonian plateau) to a shear rate de pendent region occurs at low shear rates as the solids loading is increased. Over the shear rate dependent region, the plots of

PAGE 31

19 viscosity as a function of shear rate are approximately linear. Therefore, the linear region of the plots of log as a function of log can be described by the power law model. 0.001 0.01 0.1 1 10 0.1110100100010000shear rate, 1/s Pas 0 wt% solids 1 wt% solids 2 wt% solids 3 wt% solids 4 wt% solids 5 wt% solids Figure 2-3: Shear viscosity as a func tion of shear rate and solids loading for dispersions made from 1 wt% PVA Solution (25C). Figure 2-4 represents the shear viscosity da ta for dispersions of Laponite particles of 4 wt% solids loading dispersed in solutio ns of 1 wt% PVA and 3 wt% PVA solutions respectively. It can be observed that the di spersion containing 3 wt% PVA has a viscosity that is one degree of magnitude larger th an that for the 1 wt% PVA. Both show Newtonian behavior at low shear rates and shear thinning beha vior at high shear rates. Additionally, the onset of shear thinning occurs at the same shear rate for both curves.

PAGE 32

20 0.01 0.1 1 10 100 0.1110100100010000 shear rate, 1/s Pas Figure 2-4: Viscosity as a function of shear rate for dispersions of 4 wt% Laponite prepared in 1 wt% PVA and 3 wt% PVA solutions (25C) ( ) 1 wt% PVA ( ) 3 wt% PVA. The relative viscosity data as a function of shear rate for the above mentioned dispersion are presented in Figure 2-5. The vi scosity data are normalized with respect to the viscosity of 1 wt% and 3 wt% PVA solutions as the suspending media. There is not a significant difference in the relative shear viscosities of the two dispersions. This behavior indicates that increase in the viscosit y is solely due to increase in the viscosity of the suspending media with PVA concentratio n. Since the shape of the curves stays the same when the effect of the polymer is removed, it may be that polymer-particles interactions in the polymer/clay dispersions are not dominating the rheological behavior of the system.

PAGE 33

21 1 10 100 1000 110100100010000 shear rate 1/srelative viscosity Figure 2-5: Relative viscosity (with respect to the vi scosity of the suspending fluid) as a function of Shear rate for dispersions of 4 wt% wt Laponite dispersions prepared in 1 wt% PVA and 3 wt% PVA solutions (25C) ( ) 1 wt% PVA ( ) 3 wt% PVA. 2.4.1.2 Effect of solids loading on viscosity. As with most general polymer/filler systems, it appears that the differences in vi scosity at various volume fractions are more significant at low shear rates. The viscosity as a function of solids loading at shear rates of 1 s-1 and 1000 s-1 for dispersions prepared in 1 wt % PVA solutions is shown in Figure 2-6 and can be observed at low shear rates. Addition of the Laponite filler can cause the viscosity to increase by an order of magnitude At low shear rate, the viscosity of the system is dominated by particle-particle, polymer-particle, and structure of the dispersions. While at high shear rates, hydr odynamic forces are dominant and control the viscosity of the system. From a processing point of view, it is more convenient to process these materials at higher shear rates due to th e significant decrease in viscosity at higher

PAGE 34

22 shear rates. At low shear rates, processing w ould be very slow and can result in defects such as bubbles and cloudy s pots in the final product. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 0246 wt% solids Pas Figure 2-6: Viscosity as a function of solids loading for dispersions of Laponite particles prepared in a 1 wt% PVA solution at two different shear rates ( ) 1 s-1 ( ) 1000 s-1. 2.4.2 Viscoelastic Properties and Structural Development The effect of polymer dosage on the struct ural development of the colloidal gels can be determined using small amplitude oscillatory shear measurements at different dosages. The response of the dispersions, at 10 wt% Laponite, to small-amplitude oscillatory-shear flow is s hown in Figure 2-7. At the lowest polymer dosage (1mg/(g solids)), a low storage modulus can be observed, suggesting that at this concentration, the dispersion behaves as a viscoe lastic liquid, unable to st ore large amounts of elastic energy. In the range of 2-5mg/ (g solids), there is a structural transition between liquid and solid behavior where the value of the storage modulus drops, indicating a breakdown

PAGE 35

23 in the flocculated viscoelastic gel structure. At polymer dosages above 5mg/(g solids), there is a very weak frequency dependence of the storage modulus, so the viscoelastic gel does not break down at high frequencies. This transitional behavior of the dispersions at moderate polymer dosages suggests a dependenc e of the structural breakdown of the gel on the polymer concentration used. The higher the polymer concentration, the less likely it is for a breakdown from solid-like behavior to liquid-like behavior to occur. The interactions between polymer and particles contribute to th e elastic properties of the dispersions, especially since the dispersions ar e initially flocculated. At the frequencies where the breakdowns are occurring, the weak ly bonded percolated network structure is breaking down. 1.00E-07 1.00E-06 1.00E-05 1.00E-04 1.00E-03 1.00E-02 1.00E-01 1.00E+00 1.00E+01 1.00E+02 1.00E+03 1101001000 rad/sG Pa 1mg/g solids 2mg/g solids 3mg/g solids 5mg/g solids 8mg/g solids 12mg/g solids Figure 2-7: Storage modulus vs. a ngular frequency and polymer dosage for dispersions of 10 wt% Laponite particles (25C).

PAGE 36

24 A comparison for the storage modulus data as a function of angular frequency for dispersions of Laponite (at calculated solids loadings) in 2 wt% and 3 wt% PVA matrices are shown in Figures 2-8 and 2-9. In both plots, the dispersion s containing no Laponite show liquid-like behavior throughout the ra nge examined. For the 2 wt% dispersions, there is a breakdown of the gel structure from solid-like beha vior to liquid-like behavior at high frequencies for all Laponite. This occu rs for all three Lapon ite solids loadings, although the onset of liquid-like behavior shifts to high fre quencies as the solids loading is increased. For the 3 wt% PVA disper sions, the storage modulus has a weak dependence on frequency and stays relative ly constant throu gh the range shown, increasing slightly as the frequency is increased. 1.00E-07 1.00E-06 1.00E-05 1.00E-04 1.00E-03 1.00E-02 1.00E-01 1.00E+00 1.00E+01 1.00E+02 101001000 rad/sG Pa Figure 2-8: Storage modulus as a function of angul ar frequency for dispersions of Laponite prepared in 2 wt% PVA solution (25C) ( ) 0 wt% ( ) 1.3 wt% ( ) 1.96 wt% ( ) 2.91 wt%.

PAGE 37

25 1.00E-07 1.00E-06 1.00E-05 1.00E-04 1.00E-03 1.00E-02 1.00E-01 1.00E+00 1.00E+01 1.00E+02 1.00E+03 1101001000 rad/sG Pa Figure 2-9: Storage modulus as a function of angul ar frequency for dispersions of Laponite particles prepared in 3 wt% PVA solution (25C) ( ) 0 wt% ( ) 1.96 wt% ( ) 2.91 wt% ( ) 4.3 wt%. The corresponding loss modulus data for the 2 wt% and 3 wt% PVA samples are shown in Figures 2-10 and 2-11. At both PVA concentrations, the samples containing no Laponite have a steady increase in loss modulus as the angular frequency is increased. Comparing these with the storag e modulus data, the value of loss modulus is consistently higher than that of the storage modulus, s uggesting that the disp ersions containing no Laponite exhibit liquid-like behavior for all frequencies. For the 2 wt% PVA dispersions containing Laponite, the loss modulus stays stead y at low frequencies with the magnitude of the storage modulus being larger than th at of the loss modulu s, showing solid-like behavior. At higher frequencies, there is a sharp increase in loss modulus, and the dispersions go from solid-like behavior, to liquid-like behavior. This phenomenon is not observed for the 3 wt% PVA dispersions because of the increase in polymer

PAGE 38

26 concentration. As the angular frequency is in creased, the loss modulus stays constant at a value below the storage modulus, showing solidlike behavior for all angular frequencies. 1.00E-01 1.00E+00 1.00E+01 1.00E+02 1.00E+03 101001000 rad/sG Pa Figure 2-10: Loss modulus as a function of angular frequency for dispersions of Laponite particles prepared in 2 wt% PVA solution (25C) ( ) 0 wt% ( ) 1.3 wt% ( ) 1.96 wt% ( ) 2.91 wt%. 2.4.3 Microscopy Figure 2-12 is a scanning electron micr oscope image depicting the fractal structure of the Laponite nanoparticles disperse d in a polymer matrix. This floc appears to be 30-40m in diameter, and is frozen in th e film. The floc is not connected with other flocs in the vicinity. It is apparent from this image that this is not a percolated networked structure, which is formed when a fractal network grows to create an interconnected three-dimensional network struct ure. The formation of the fractal network in these films may be due to aggregation over long time scales.15

PAGE 39

27 1.00E-07 1.00E-06 1.00E-05 1.00E-04 1.00E-03 1.00E-02 1.00E-01 1.00E+00 1.00E+01 1.00E+02 1.00E+031101001000 rad/sG Pa Figure 2-11: Loss modulus as a function of freque ncy for dispersions of Laponite particles prepared in 3 wt% PVA Solution (25C) ( ) 0 wt% ( ) 1.3 wt% ( ) 1.96 wt% ( ) 2.91 wt%. Figure 2-12: SEM micrograph of flocculated La ponite particles cast from a 3 wt% PVA dispersions with 50 wt% Laponite mass fraction in dried film.

PAGE 40

28 The freely dispersed flocs present in th e SEM image of the PVA/Laponite system interact through non-covalent bonds such as van der W aals attractive forces and hydrostatic interactions. Under high shear ra te or a high frequency, the weakly bonded gel can breakdown, creating a discontinuity in solid-like behavior, as previously discussed. The dimensions and density of th e fractal structures have an effect on the properties of the dispersions a nd the cast films. These propert ies can be determined using computational modeling and mathematical analysis.24 Atomic force microscope images for several different film solids loadings are shown in Figure 2-13. These pictures show that the clay particles are randomly oriented in the film and are connected by polymer.9 A film containing no laponite is shown in Figure 2-13A, this image shows a smooth surf ace with some variation in height but a very small roughness. The images of the f ilms containing Laponite particles show surfaces with some roughness and many height va riations due to partic les at the surface. The 40 wt% film has some where the height decreases, suggesting that the polymer matrix exists between the particles. As the solids loading is increased, there are less areas where the polymer exists and the height increases, which may due to particles aggregating on film surface.

PAGE 41

29 AB CD Figure 2-13: AFM images of dry films cast from a 3 wt% PVA matrix with final solids loadings of A.) 0 wt% B.) 40 wt% C.) 50 wt% D.) 60 wt%.

PAGE 42

30 CHAPTER 3 SURFACE ACTIVATION 3.1 Introduction In order to adhere polymer nanocomposite films onto a polymer substrate, there needs to be good interfacial adhesion be tween the two materials. Many polymer substrates are hydrophobic, or have a high surfac e tension, and it is difficult to adhere a water-based film to that t ype of surface. To form a bond between the film and the substrate, the surface tension must be reduced and the surface of the substrate must be activated so that covalent bonds can fo rm between the film and the surface.29 There are several ways to activate the su rface of a material to create functional groups capable of improving interfacial adhe sion. These methods include wet-chemical treatments, mechanical treatment, exposure to flames, ultraviolet radiation, and ion beams.29-33 Plasma treatment is considered the mo st efficient way to activate a polymer surface for adhesion to metal.29 This technique can provide surface activation without effecting the bulk properties of the material. In a gas plasma process, weak bonds on the surface of a material are replaced by highly reactive carboxyl, carbonyl, and hydroxyl groups. This is achieved by exposing the surface of a material to ionized gas containing ions, atoms, electrons and neutral specie s; this process is done in a vacuum.30 Although there are many advantages to this process, it requires expensive equipment. The most cost effective way to create surface chemical f unctionalization is by using various wetchemical treatments. These treatments work by removing a weak surface boundary layer

PAGE 43

31 31 from the polymer interface, creating cova lent bonds at the surface. Some common materials used for these treatments include various acids and coupling agents. When exposed to a polymer surface, these chemi cals will change the composition of the surface, creating active domains capable of bonding with a hydrophilic interface.32 In order to determine the success of a surface treatment, the chemical composition of the surface must be analyzed before and after the treatment. Fourier Transform Infrared (FTIR) spectrometry is ideal for ev aluating changes on the surface of a material. Using Attenuated Total Reflectance (ATR), a sampling of the material surface can be taken with little sample preparation comp ared to traditional transmission sampling.30,32 For this method, a total intern ally reflecting IR beam will co me in contact with a dense crystal with a high refractive i ndex at a certain angle. When the sample is in direct contact with the infrared beam it will be ab sorbed by an IR absorbing material sample. The adsorption wave will be attenuated in th e sample and the data will be sent to a detector and an infrared spectrum will be generated. A schematic of the mechanism of an ATR sampling is shown in Figure 3-1. Sample spectra collected using ATR are different from those obtained by transmission. Shifts in band intensity and fr equency can result in displacements of peak maximums and intensity.30 Because of these differences, it is difficult to perform quantitative analysis using ATR.

PAGE 44

32 32 IR Beam ATR Crystal Sample in contact with Crystal Detector IR Beam ATR Crystal Sample in contact with Crystal Detector Figure 3-1: Schematic of Attenuat ed Total Reflection (ATR)30 3.2 Materials A SentryGlas Plus Ionomer Interlayer was obtained from the Air Force and was manufactured by Dupont. It is composed of an ethylene/Methacrylic Acid copolymer with trace amounts of sodium salts. The material is non-toxic and stable at normal temperatures. Some of the typical properties of the interlayer are shown in Table 3-1. 3-1: Characteristics of Dupont Sent ryGlas Plus ionomer interlayer Characteristics Metric Unit Metric Value Test Young's Modulus MPa 300 ASTM D5026 Tear Strength MJ/m3 50 ASTM D638 Tensile Strength MPa 34.5 ASTM D638 Elongation % 400 ASTM D638 Density g/cm3 0.95 ASTM D790 Flex. Modulus (73F) MPa 345 ASTM D790 Coefficient of Expansion (20C to 32C) 10-5 cm/cm C 15-Oct ASTM D696

PAGE 45

33 33 The acids used for the surface treatments were sulfuric acid (Fisher Scientific) and nitric acid (Fisher Scientific). Several diffe rent silane coupling agents were used for surface treatments. The silanes used in this study are listed below: 2 (3,4 Epoxycyclohexyl) Et heltriethyl Triethoxysilane 5,6 Epoxyhexyltriethoxysilane 11 (Triethoxysilyl)udecanal 3 Aminopropyl Trimethoxysilane N (2 Aminoethyl) 3 Aminopropyl Trimethoxysilane N (2 Aminoethyl) 3 Aminopropylsilanetriol 25% Bis (2 Hydroxyethyl) 3 Amin opropyl Triethoxysilane, 62% in ethanol Vinyl terminated PDMS fumed silica reinforced Pentamethyl cyclopentasiloxane Some of these silane coupling agents turned the interlayer samples yellow, so not all of them were used as film substrates. 3.3 Experimental Procedure 3.3.1 Sample Preparation Two different chemical treatments were used to activate the surface of the interlayer samples, acid treatments and sila ne coupling agents. The acid treatment was used to oxidize the surface by va rying the concentra tion of a mixture sulfuric and nitric acid. Treatments containing no nitric acid we re also used, but th e sulfuric acid was diluted to different levels. For these treat ments, the interlayer sheets were thoroughly cleaned with deionized water and cut into 1. 5 inch by 3 inch rect angular coupons using a shear cutter. The acid solutions were mixe d on a magnetic stir plate and allowed to equilibrate to room temperature. The in terlayer coupons were dipped into the acid solutions for two minutes using Teflon-coat ed tweezers. After being removed from the solution, the treated samples were dipped in water to clean off any excess acid solution

PAGE 46

34 34 and dried with a towel. This method was used for all acid-treated samples at all concentrations. For the silane coupling agent treatments, a 1ml transfer pipette was used to drop the coupling agent onto the surface of the interlayer coupons. All samples were treated for one minute and then washed with deionized water until all exce ss coupling agent was removed from the surface. Any samples that instantly yellowed on contact with the silane coupling agent were disqualified as surface mo difiers and were no longer used in the experiment. Nanocomposite films were cast onto the tr eated interlayer substrates by solution casting in a convection oven at 25C. The interlayer coupons were put into plastic Petri dishes and 75mL of the disper sions were poured into the di sh to cover the interlayer samples. Dispersions in 1, 2, and 3 wt% pol ymer matrices were prepared with solids loadings corresponding to dry film concen trations of 40, 50, and 60 wt% for each polymer solution. 3.3.2 Fourier Transform Infrared Spectroscopy (FTIR) The FTIR experiments were performe d using a Nicolet MAGNA 760 Bench with Spectra Tech Continum IR Microscope. The samples were analyzed using the IR Microscope with an ATR objec tive slide-on microscope accessory. A silica crystal with a refractive index of 3.4 was used to analyze the sample by contact with the surface. The microscope assembly was cooled with liqui d nitrogen before using the microscope and ambient gases were pumped into the IR ch amber. The samples were loaded onto the microscope platform, surface treated side faci ng the crystal. Before the sample comes in contact with the crysta l, a background spectra is taken to assess the amount of noise in the instrument before sampling. The microsc ope is then pressed onto the sample surface

PAGE 47

35 35 until the Contact AlertTM system signaled the position wher e optimal contact pressure is achieved for repeatable results. A set number of samples are taken until a smooth curve is produced. The single beam spectra from the background and then again from the sample are combined and a total absorbance is determined using the following formula:30 ) log( ) 1 log(Background sample R 3-1 This is also related to sample reflectance (r) by ) log( ) 1 log(r R 3-2 The magnitude of log(1/R) is also proportiona l to the concentration of bonds on a sample surface, which is useful in performing quali tative analysis on the change of a sample surface. 3.4 Discussion of Results 3.4.1 Acid Surface Treatments There were several different acid treatments applied to the interlayer samples to produce a change in the surface properties. The FTIR provided information about changes in the chemical structure of the inte rlayer surfaces. Figure 3-2 is an FTIR curve for the untreated interlayer specimen. The largest peaks in the spectrum correspond to two very large CH2 peaks at 2917 cm-1 and 2850 cm-1. There are also several smaller peaks that also correspond to C-H and CH2 bonds in polyethylene, some of these peaks are represented in Table 3-2 from inform ation obtained on the OMNIC software. The spectral libraries in the ONMIC software matc hed the curves with that of low density polyethylene, which is consistent with the polyethylene matrix of the interlayer material. The second phase of the interl ayer copolymer, methacrylic aci d, was not observed in the

PAGE 48

36 36 spectral libraries. This spectrum was reproducib le when moved to different sections of the sample and when analyzed several times at one spot. Untreated Interlayer -H2O CO2 BLC 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20 0.22Log(1/R) 500 1000 1500 2000 2500 3000 3500 4000 Wavenumbers(cm-1) Untreated Interlayer -H2O CO2 BLC 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20 0.22Log(1/R) 500 1000 1500 2000 2500 3000 3500 4000 Wavenumbers(cm-1) Figure 3-2: Infrared spectrum for the surface of an untreated interlayer sample. Table 3-2: Absorption frequencies for co mmon bonds found in OMNIC software. Bond Assignment frequency range cm-1 CH, CH2, CH3 Stretching 2850-3000 CH bend 675-1000 and 13501470 CH2 wag 1450-1470 C=O stretch 1650-1870 Si-OR stretch 1000-1100 Si-H silane stretch 2100-2360

PAGE 49

37 37 The FTIR plot in Figure 3-3 is for an interlayer sample treated with a 100% concentrated solution of sulfuric acid. This highly concentrated solution caused some yellowing on the surface of th e interlayer, but a change in the surface chemistry is observed. The two CH2 peaks are still presen t in the spectrum, but a new large peak is observed at a wavelength of approximately 1150 cm-1, which corresponds to a carbonyl (C=O) stretch bond frequency. Us ing the spectral libraries on the OMNIC software, this value corresponds to an acid group on the surf ace of the interlayer. It can also be observed that the height of the CH2 peaks at 2917 cm-1 and 2850 cm-1 is less that that observed for the untreated interlayer. This occurs because the new activated groups on the surface of the interlayer in creased in concentration, so the amount of polyethylene observed on the surface will decrease. Sulfuric 100% 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 0.11 0.12 0.13 0.14 0.15Log(1/R) 500 1000 1500 2000 2500 3000 3500 4000 Wavenumbers(cm-1) Sulfuric 100% 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 0.11 0.12 0.13 0.14 0.15Log(1/R) 500 1000 1500 2000 2500 3000 3500 4000 Wavenumbers(cm-1) Figure 3-3: Infrared spectrum for the surface of an interlayer samp le treated with 100% sulfuric acid solution.

PAGE 50

38 38 Several concentrations of the sulfuric acid treatments were used to activate the surface of the interlayer samples. Figure 3-4 is a comparison between the untreated interlayer specimen, a 10% sulfuric acid treat ment, and a 100% sulfuric acid treatment. It can be observed that the 10% sulfuric acid so lution has almost no eff ect on the surface of the interlayer while the 100% sulfuric acid solution has a very large effect. This suggests that the extent of surface ac tivation depends on the concentr ation of the treatment being used. An optimal treatment would be one that sufficiently activates the surface of the interlayer without discoloration. Untreated interlayer 1 H20 C02 BLC -0.00 0.05 0.10 0.15 0.20 Log(1/R) 10% sulfuric 1 BLC H-C 0.00 0.05 0.10 0.15 0.20Log(1/R) 100% sulfuric 2 -0.00 0.05 0.10 0.15Log(1/R) 500 1000 1500 2000 2500 3000 3500 4000 Wavenumbers(cm-1) Untreated interlayer 1 H20 C02 BLC -0.00 0.05 0.10 0.15 0.20 Log(1/R) 10% sulfuric 1 BLC H-C 0.00 0.05 0.10 0.15 0.20Log(1/R) 100% sulfuric 2 -0.00 0.05 0.10 0.15Log(1/R) 500 1000 1500 2000 2500 3000 3500 4000 Wavenumbers(cm-1) Figure 3-4: Infrared spectrum comparison for interlay er samples a) without acid treatment b) 10% sulfuric acid treatment c)100% sulfuric acid treatment A sulfuric acid and nitric acid mixture was also used to treat the surface of the interlayer. This mixture was used to create a strong oxidizing effect on the material surface while preventing discoloration. A comp arison between an untre ated interlayer, a treatment using 2:1 sulfuric acid to nitric acid, and a treatment using a 3:1 sulfuric to

PAGE 51

39 39 nitric acid treatment is shown in Figure 35. The carbonyl stretch bond is still present on the material surface at comparable concentr ations for both samples (concentration is proportional to absorbance). Although there is not a substantial difference in the curves, there was discoloration in th e 3:1 sulfuric acid sample. Untreated -0.00 0.05 0.10 0.15 0.20Log(1/R) Sulfuric/Nitric 2:1 0.00 0.05 0.10 0.15 Log(1/R) Sulfuric/Nitric 3:1 0.00 0.05 0.10 0.15 Log(1/R) 500 1000 1500 2000 2500 3000 3500 4000 Wavenumbers(cm-1) Untreated -0.00 0.05 0.10 0.15 0.20Log(1/R) Sulfuric/Nitric 2:1 0.00 0.05 0.10 0.15 Log(1/R) Sulfuric/Nitric 3:1 0.00 0.05 0.10 0.15 Log(1/R) 500 1000 1500 2000 2500 3000 3500 4000 Wavenumbers(cm-1) Figure 3-5: Infrared spectrum comparison for interlayer samples a) without acid treatment b) 2:1 sulfuric/nitric acid treatment c) 3: 1 sulfuric/nitric acid treatment 3.4.2 Silane Coupling Agent Treatments All of the silane coupling agents listed above have been used to treat the interlayer samples, but only some were successful ba sed on surface discoloration. Some of the coupling agents caused substa ntial yellowing and destruction of the sample. The two successful silane coupling agent treatments were 3-aminopropyl trimethoxysilane and N(2-aminoethyl) 3-aminopropyl trimethoxysilane Figure 3-7 is an FTIR spectrum comparison for the 3-aminopropyl trim ethoxysilane and c N-(2-aminoethyl) 3aminopropyl trimethoxysilane coupling agents. Th e most distinguishing characteristic of

PAGE 52

40 40 these plots is the large peak at 1100 cm-1 that corresponds to a Si-OR stretching bond. Since the height of the absorbance peak is proportional to the c oncentration of bonds on the surface, the sample treated with 3-am inopropyl trimethoxysilane has a large amount of functionalized silane groups on its surface. 500 Untreated -0.00 0.05 0.10 0.15 0.20Log(1/R) 3-aminopropyl TMS -0.00 0.05 0.10 0.15Log(1/R) N-(2-aminoethyl) 3-aminopropyl TMS -0.00 0.05 0.10 0.15 0.20Log(1/R) 1000 1500 2000 2500 3000 3500 4000 Wavenumbers(cm-1) 500 Untreated -0.00 0.05 0.10 0.15 0.20Log(1/R) 3-aminopropyl TMS -0.00 0.05 0.10 0.15Log(1/R) N-(2-aminoethyl) 3-aminopropyl TMS -0.00 0.05 0.10 0.15 0.20Log(1/R) 1000 1500 2000 2500 3000 3500 4000 Wavenumbers(cm-1) Figure 3-6: Infrared spectrum comparison for interlayer samples a) without silane treatment b) 3aminopropyl trimethoxysilane treatmen t c) N-(2-aminoethyl)3-aminopropyl trimethoxysilane treatment The absorbance spectrum for N-(2-aminoethyl) 3-aminopropyl trimethoxysilane. The silane peak is not as pronounced for this coupling ag ent, but the shift in the absorbance peaks suggests that the surface has still been substantially modified. 3.4.3. Film Casting onto Treated Interlayer Substrates The PVA/Laponite dispersions described a bove were coated onto the substrates of interlayer samples treated with a 2:1 sulfur ic/nitric acid treatment and a treatment of 3aminopropyl trimethoxysilane. These treatments were determined to produce the least

PAGE 53

41 41 discoloration to the interlay er surface and the most surf ace change according to FTIRATR qualitative analysis. The surfaces treated with 2:1 sulfuric/nitric acid showed the best film adhesion based on visual inspecti on. The surfaces treated with 3-aminopropyl trimethoxysilane had delamination during the drying process.

PAGE 54

42 CHAPTER 4 ABRASION RESISTANT AND ADHESIVE PROPERTIES 4.1 Introduction When formulating an abrasion resistant coating, the performance of the coating depends on several factors. The polymer mol ecular weight, filler content, drying method, processing temperature, and dispersion rheol ogy can effect the fina l properties of the protective coating. The scratch resistance a nd abrasion resistance of protective films are often confused, but maximizing one of these pr operties may result in a reduction in the other. Abrasion and scratch resistance are rela ted to different material properties. Scratch resistant materials usually have a high har dness and little flexibility. Conversely, abrasion resistant materials depend on the toughness of the coating and are usually flexible and have a high tensile strength.34-36 There are several methods for testing th e abrasion resistance of coatings. There are several standard test me thods for testing abrasion resi stance, ASTM D 4060 being the most common.36 In this method, a Taber Linear Abra ser with a grinding wheel is used to abrade a coating on a surface over a fixed amount of time. The weight loss of the coating is measured as a function of number of cycles. The Taber apparatus can be fitted with many different accessories for measuring ab rasion resistance, scratch resistance, and color transfer. For smaller samples, such as th e samples used in this study, a Taber Linear Abraser (Model 5750) can be used to test small areas and contoured surfaces.

PAGE 55

43 43 The Taber Linear Abraser can be used to analyze wear resistance of a wide variety of materials. The abrasion tests can be programmed to up to 999,999 cycles with operating speeds of 2, 15, 25, 40, and 60 cycles per minute. The initial load on the specimens is 100 grams, which is loaded by the spline shaft. The load can be increased in 250 gram increments using additional weight discs. The free floating arm has operating stroke lengths of 0.5, 2, 3, and 4, whic h can be chosen depending on the size of the specimen. The actual abrading of the surface is done by using a wearaserTM, which is an abradant made of the same calibrase and calibrade material as the grinding wheels that are commonly used. The calibrase material is resilient and consists of rubber and small abrasive particles. The calibrade wearaser is non-resilient and is composed of vitrified stone and small abrasive particles. 4.2 Materials Several colloidal dispersions with different Laponite filler content were used to make the multi-layered nanocomposite films. For the samples cast using the doctor blade, a 3 wt% PVA matrix was used with final La ponite loadings of 40, 50, and 60 wt% after drying. This high concentration of PVA is n eeded for this technique because it will form a gel. A dispersion with a low viscosity w ould move through the feeder too fast, making it difficult to coat the films uniformly. These multi-layered samples were coated onto extra wide glass slides. The interlayer samples used were treated with a 2:1 sulfuric to nitric acid solution. The samples were cut into coupons and were co ated with PVA concentrations of 1 wt%, 2 wt%, and 3 wt% with Laponite solids loadings correspondi ng to a dry film concentration of 40 wt%, 50 wt%, and 60 wt% for each PEO concentration. These samples were solution cast and left in a convection oven at 25C for 24 hours.

PAGE 56

44 44 Two different wearasers were used to ab rade the samples using the Taber Linear Abrader. The CS-10 Calibrase Wearaser was used to abra de the films mounted on the glass slides. The H-18 Calibrade Wearaser was used to abrade the single-layered films on the interlayer samples. 4.3 Experimental Procedure 4.3.1 Sample Preparation A doctor blade (Richard E. Mistler TTC-1000) was used for tape casting the multilayered films onto the glass slides. Sufficiently gelled dispersions were loaded into a stationary feeder apparatus that is connected to a moving belt made of silicone coated Mylar. The samples were cast at a speed of 10 inches/minute, which roughly corresponds to a shear rate of 1.41 s-1. The films were cast onto wide glass slides and left to dry overnight with air circulating over them. After the first coating, some of the samples were coated again to create a double layer, and so me from that group were coated again to form a triple layer. Interlayer samples treated with a 2:1 sulf uric/nitric acid mixture were coated with nanocomposite films to analyze delamination. Dispersions in 1, 2, and 3 wt% polymer matrices were prepared to cast films with solids loadings corresponding to dry film concentrations of 40, 50, and 60 wt% for each polymer solution. The films were cast in a convection oven at 25C. 4.3.2 Taber Linear Abrader The experimental methods used for the Taber Linear Abraser are based off of ASTM D 4060. The uncoated and coated glass sl ides were tested as a function of number of cycles in increments of 10, goi ng from 10-60 cycles using a calibrade wearaser. The speed was kept constant at 25 cycles per minute with an opera ting stroke of 1 inch and a

PAGE 57

45 45 constant load of 100 grams. The samples we re secured to the workspace using a metal clamp with cushioning in between to prevent sample breakage. A small brush was used to prevent the accumulation of abraded partic les during testing, the samples were continually monitored. The sample weight loss was measured in between each cycle increment. The method for testing the coated interlayer samples was slightly different due to their tendency to delaminate from the interl ayer surface. For this experiment, a calibrase wearaser was used to abrade the samples for 50 cycles at a stroke length of 1 inch. The load was increased from 100g to 850g in increments of 250 grams using the additional weight discs. If no delami nation was observed at the hi ghest load, the speed was increased form 25 cycles/minute to 40 cycles /minute and then to 60cycles/minute. The sample is considered successful if no de lamination is observed up to this point. 4.3.3 Visual observation of Film Delamination To obtain a better understanding of th e delamination behavior of the nanocomposite films from the interlayer s ubstrate, a visual observation test was conducted on coated interlayer samples. The samples were solution cast with films made from dispersions in 1 wt%, 2wt%, and 3wt% PVA matrices. The films had final solids loadings of 40 wt%, 50 wt%, and 60 wt% for each polymer concentration. The films were cast onto interlayer substrates cut into 1 x 2 coupons and surface treated with a 2:1 sulfuric/nitric acid treatment. To visually obs erve the delamination of the films from the interlayer substrate, the samples were bent over steel washers of di fferent diameters. A schematic of the washers are shown in Figure 4-1.

PAGE 58

46 46 1.9 cm1.59 cm0.95 cm 1.27 cm Figure 4-1: Steel rods of different diameters Each sample was bent over the rods, starting from the largest diamter, and the delamination behavior of the films was recorded. 4.4 Discussion of Results 4.4.1 Multi-Layered Films The weight loss data for samples coated w ith one layer of nanocomposite film are shown in Figure 4-2. The samples are compared to the weight loss of the glass substrate. The sample containing 40 wt% Laponite in the dried film had a weight loss comparable to that of the glass slide. The films containi ng higher solids loadings had a much larger weight loss, almost triple that of the glass s lide. For the coated samples, the clay platelets will be randomly oriented in th e film and connected by polymer.9 The films containing higher solids loading would have less conn ecting polymer, creating a rougher surface. This rough surface may be easier to abrade than the 40 wt% surface, which would have more polymer connecting the particles. A nother observation was that the standard deviation of the film-coated samples were much larger than the sta ndard deviation for the glass slide.

PAGE 59

47 47 0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 40%50%60%Glass Slide Solids Loading, wt%Average Weight Loss (g) 0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 40%50%60%Glass Slide Solids Loading, wt%Average Weight Loss (g) Figure 4-2: Average weight loss data with standa rd deviation for abraded glass substrate and substrate coated with 40, 50, and 60 wt% nanocomposite films The weight loss results for the double-coate d samples are shown in figure 4-3. In this case, all of the samples performed better than or equal to the glass substrate at resisting weight loss by surface abrasion. C ontrary to the single-coating results, the samples containing high solids loadings of La ponite such as 60/50 (60 wt% coated onto 50 wt%), and 50/60 lost the least amount of we ight. Additionally, the samples which lost the least amount of weight also have a standa rd deviation that is much smaller than the samples that lost a larger amount of weight. The samples which have a 40 wt% on the bottom had very large standard deviations which may be the result of low surface roughness creating a decreased in terfacial adhesion to th e second layer of film.

PAGE 60

48 48 0 0.001 0.002 0.003 0.004 0.005 0.006 60/4050/4060/5040/5040/6050/60Glass Slide Treatment, wt% LaponiteAverage Weight Loss (g) Figure 4-3: Average weight loss data with standa rd deviation for abraded glass substrate and substrate coated with two layers of nanocomposite film (x/y: film x coated over film y). Figure 4-4 shows the weight loss results for the triple coated samples. Some of these samples loss less weight than the glass substrate, but several did not perform as well. The samples that had the 40 wt% Laponi te film as the top layer (40/50/60 and 40/60/50) lost more weight than the glass and had large standa rd deviations. As with the double coated samples, this could be due to a decrease in interfacial adhesion between the film layers. The samples with the 60 wt% La ponite film as the top layer performed well but also have high standard deviations. The e rror in this data is very large, making it difficult to draw conclusions due to an inability to reproduce the results.

PAGE 61

49 49 0 0.001 0.002 0.003 0.004 0.005 0.006 0.007 50/60/4060/50/4060/40/5040/60/5050/40/6040/50/60Glass Slide Treatment, wt% LaponiteAverage Weight Loss (g) 0 0.001 0.002 0.003 0.004 0.005 0.006 0.007 50/60/4060/50/4060/40/5040/60/5050/40/6040/50/60Glass Slide Treatment, wt% LaponiteAverage Weight Loss (g) Figure 4-4: Average weight loss data with standa rd deviation for abraded glass substrate and substrate coated with three layers of nanocomposite film (x/y/z: film x coated over film y coated over film z) A comparison of the films which loss less th an or equal to the amount of weight as the glass substrate are shown in Figure 45. All of these coatings are sufficient in providing abrasion resistance that is better than or equal to that of a typical glass specimen. Again it can be seen that the most successful coatings had the smallest standard deviation. When compared to th e triple-coated samples, the double-coated samples with films of high solids loading not only lose less weight th an the glass slide, but the results are reproducible.

PAGE 62

50 50 0 0.001 0.002 0.003 0.004 0.005 0.006 0.0076 0/4 0 5 0/ 40 60/50 4 0/50 4 0/ 60 50/60 50/ 60 /4 0 6 0/ 50/ 40 60/40/5 0 Gl a s s S l ideTreatment, wt% LaponiteAverage Weight Loss (g) 0 0.001 0.002 0.003 0.004 0.005 0.006 0.0076 0/4 0 5 0/ 40 60/50 4 0/50 4 0/ 60 50/60 50/ 60 /4 0 6 0/ 50/ 40 60/40/5 0 Gl a s s S l ideTreatment, wt% LaponiteAverage Weight Loss (g) Figure 4-5: A comparison of average weight loss da ta with standard deviation for the films which loss less than or equal to the am ount of weight as the glass substrate 4.4.2 Delamination of Coated Interlayer Samples Upon visual observation, the coated inte rlayer samples had good adhesion with the nanocomposite film. When abraded by the calibrase wearaser at different loads and speeds as described earlier, none of the films delaminated from the interlayer surface. Although the Taber instrument was at its ma ximum capacity, the films were not rubbed from the surface and there was very little h aze on the transparent film. Other forms of adhesion testing, such as dyna mic mechanical analysis (DMA ) or and Instron peel test would provide more insight into the interfaci al strength between the films and the treated interlayer substrate. Visual observation of delamination of the PVA/Laponite films from the interlayer substrate was achieved with the steel rod test. All of the film samples prepared

PAGE 63

51 51 in a 1 wt% PVA matrix did not delaminate wh en bent over the rods. Even the smallest diameters did not succeed in changing the appearance of the film or adhesion to the substrate. The films prepared in 2 wt% PVA dispersions did not fracture from the interlayer surface, but they did delaminate. Th e film with the lowest solids loading (40 wt%) was not effected by the larger diameter s, but at 1.27cm the film/interlayer interface turned white, suggesting the film was peeli ng from the surface. The 50 and 60 wt% films had this same behavior on the 1.9cm rod. Th e films cast in a 3 wt% PVA matrix fractured on the largest diameter rod. When bent, they fractured along the middle of the sample and continued to peel back if bent further. Larger rods would be needed to define the point where fracture occurs for these films.

PAGE 64

52 CHAPTER 5 CONCLUSIONS AND FUTURE WORK 5.1 Conclusions When Laponite JS is dispersed in water, it can form a stable sol which exhibits Newtonian behavior. When polymer is presen t, the interactions between the polymer chains and the particles cause a change in th e rheological properties of the dispersions. As polymer dosage is increased in a Laponite dispersion, particle flocculation causes a weak colloidal gel to form and the viscosit y increases. Shear thinning behavior can be seen for these colloidal gels at high sh ear rates, noting a breakdown and particle alignment in the weakly flocculated network st ructure. SEM analysis shows that the flocs present in the dispersions are open, ramifi ed structures that are evenly dispersed throughout the dispersions. The flocs are not interconnected, which makes it easier for structural breakdown to occur at high sh ear rates and high angular frequencies. The processing conditions for films made from 2 wt% and 3 wt% PVA solutions include Newtonian-like behavior at low shear rates and shear thinning behavi or at high shear rates. For the 2 wt% PVA dispersions, th e weakly flocculated gel structure can be broken down at high frequencies; this does not occur for dispersions made in a 3 wt% PVA matrix. For low shear processing tec hniques like solution casting, a 2 wt% PVA matrix is ideal; the 3 wt% PVA matrix would work for high shear techniques such as tape casting and spin coating. The processing conditions of the colloidal gels are determined as a function of both solids loading and polymer concentration. An increase in solids loading at fixed

PAGE 65

53 53 PVA concentration causes a shift in zero shear viscosity and an increase in the shear rate dependent region of the steadyshear viscosity curves. The effect of the suspending fluid is crucial in the processing of the coatings due to the large increase in viscosity with increasing PVA concentration. Thin, transparent films ca n be cast from the PVA/Laponite dispersions. These films exhibit good adhesion to glass slid es, but cannot be cast onto a hydrophobic polymer interlayer. To activate the surface of the interlayer, and reduce the surface tension, several surface treatments were us ed. Using FTIR-ATR, it can be observed that several of these treatments can successfully activ ate the interlayer surface. A 100% sulfuric acid treatment will activate the surface, but will create substantial yellowing. A mixture of sulfuric and nitric acid in a ra tio of 2:1 will also activate the surface but will not cause discoloration. This treatment s howed a substantial improvement in adhesion between the cast films and the interlayer s ubstrate. Silane coupling agents will also successfully modify the interl ayer surface, but it was obs erved that the interfacial adhesion to the nanocomposite films was not improved. To determine the abrasion resistance of cast nanocomposite films on glass slides, a Taber Linear Abrader was used determine weight loss as a function of number of cycles. Coatings with one, two, and three laye rs of film at different solids loading were tested for weight loss agai nst an uncoated glass slide specimen. The single coated samples did not perform as well as the glass s lide, i.e. they lost more weight than the glass. The double coated samples performed be tter than the glass slide at every solids loading, with the highest solids loadings ( 50/60 and 60/50) losing the most weight. With the triple coated samples, only some lost less weight than the glass slide. In conclusion,

PAGE 66

54 54 the optimal coating for the glass to minimi ze weight loss and maximize efficiency is a double-layered film of high solids loading. 5.2 Future Work The field of polymer-layered silicate nanocomposites is very large and has a broad range of applications. Further study of the PVA/Lapon ite JS system and systems with similar properties is useful to assess their use in th e films and coatings industry. For use as an abrasion resistant film, more mechan ical and thermal analysis of the material must be performed. Adhesion is a large con cern for this application and maximizing this property would be useful. The following are suggestions for possible future work. To further determine the abrasion resistan t properties of the films, mechanical studies can be conducted on in terlayer samples coated with various PVA/Laponite films. These studies would include Instr on tensile testing and dynamic mechanical analysis. Since tensile stre ngth is proportional to abrasion resistance, an optimal formula can be determined. To continue rheological studies on the PVA/Laponite system, an in-depth study on the structural development of th e dispersions can be conducted. Perform surface activation of the polymer interlayer surface using RF plasma treatment. This treatment is more expe nsive than wet-chemical treatments and requires special equipment, but may be mo re effective in creating an activated polymer surface. When used as a laminated glass, the Se ntryGlas ionomer is heat treated and becomes a harder, more glass-like material. This glassy polymer is adhered to window glass to improve the properties of the window. Analysis can be done on the surface of heat treated interlayer us ing contact angle measurements and FTIRATR. UV-VIS experiments can be performed on gl ass, film-coated glass, and abraded materials to determine the transparency of the films compared to regular window glass. The effect of abrasion on the tr ansparency of the glass can also be determined using this technique. Using different materials for both the polym er matrix and the nanosilicate can be used to continue this investigation. Some possible polymers would be Poly Methyl Methacrylate (PMMA), Poly Ethyl ene Oxide (PEO), Polycarbonate (PC), and polyvinyl acetate (PVA). Also, different grades of Laponite can be used as the nanosilicate filler.

PAGE 67

55 LIST OF REFERENCES 1. A.R.Vaia.; R. Krishnamoorti, 2001. Po lymer nanocomposites: introduction. ACS Symposium Series 804. Materi als and Manufacturing Director ate, Air Force Research Library, 2001. 2. J. Collister, 2001. Commercialization of polymer nanocomposites. ACS Symposium Series 804. Edison Polymer Innovation Corporation. 3. S. Bandyopadhyay; A. J. Hseih. E. P. Giannells. 2001. PMMA Nanocomposites synthesized by emulsion polymerizati on. ACS Symposium Series 804. 4. B. Wetzel; F. Hauper; K. Friedrich; M. Zh ang; M. Rong. Impact and wear resistance of polymer nanocomposites at low filler cont ent. Polymer Engineering and Science, September 2002. 5. M. Alexandre, P. Dubois. Polymer-layered silicate nanocomposites: preparation, properties, and uses of a new class of ma terials. Materials Science and Engineering 28 (2000) 1-63. 6. J. Zebrowski, V. Prasad, W. Zhang, L. M. Walker, D.A. Weitz. Shake-gels: shearinduced gelation of Laponite-P EO mixtures. Colloids and Surfaces A: Physicochem. Eng. Aspects 213 (2003) 189. 7. S. Lin-Gibson, H. Kim, G. Schmidt, C. C. Han, E.K. Hobbie, J. Shear-induced structure in polymer-clay nanocomposite so lutions. Colloid Interface Sci. 274 (2004) 515. 8. D. C. Pozzo, L.M. Walker. Reversible sh ear gelation of polym er-clay dispersions. Colloids and Surfaces 240 (2004) 187-198. 9. V. Ferreiro, G. Schmidt, C. Han, A. Kari m. Dispersion and nucleating effects of clay fillers in nanocomposite polymer films. Polymer Nanocomposites: Synthesis, Characterization, and Modeling (2002) 177. 10. A. Nelson, T. Cosgrove. A small-angle neutron scattering study of adsorbed poly (ethylene oxide) on Laponite. Langmuir 20 (2003) 2298. 11. G. Schmidt, M. Malwitz. Properties of pol ymer-nanoparticle composites. Curr. Opin. Colloid Interface Sci. 8 (2003) 103.

PAGE 68

56 56 12. Ronald G. Larson. The structure and rheol ogy of complex fluids. Oxford University Press, New York, 1999. 13. Rockwood Additives Ltd., Cheshire, Unite d Kingdom. Laponite Technical Bulletin L204/01g 14. R. Krishnamoorti, K. Yurekli. Rheology of polymer layered silicate nanocomposites. Curr. Opin. Colloid Interface Sci. 6 (2001) 464. 15. J. Labanda, J. Llorens. Rheology of La ponite colloidal dispersions modified by sodium polyacrylates. Colloids and Surf aces A: Physiochem. Eng. Aspects 249 (2004) 127-129. 16. M. Malwitz, A. Dundigalla, V. Ferreiro, P. Butler, M.C. Henk, G. Schmidt. Layered structures of shear-oriented and multilayered PEO/silicate nanocomposite films. Phys. Chem. 6 (2004) 2977. 17. J. Ren, B. Casanueva, C.A. Mitchell, R. Krishnamoorti. Disorientation kinetics of aligned polymer layered s ilicate nanocomposites. Macromolecules 36 (2003) 4188. 18. Z. Shen, J. Chen, H. Zou, J. Yun. Rheology of colloidal nanosized BaTiO3 suspension with ammonium salt of polyacrylic acid as a dispersant. Colloids and Surfaces 244 (2004) 61-66. 19. M. Sjoberg, L. Bergstrom, A. Larsson, E. Sjostrom. The effect of polymer and surfactant adsorption on the colloidal stability and rheo logy of kaolin dispersions. Colloids and Surfaces 159 (1999) 197-208. 20. J. A. Lee, M. Kontopoulou, J. S. Pare nt. Time and shear dependent rheology of maleated polyethylene and its nanocomposites. Polymer (2004) 1-6. 21. S. Granick, H. Hu, G.A. Carson. Na norheology of confined polymer melts. 2. nonlinear shear response at strongly adso rbing surfaces. Langmuir 10 (1994) 3867. 22. A. Mukhopadhyay, S. Granick. Microand nanorheology. Curr. Opin. Colloid Interface Sci. 6 (2001) 423. 23. J. Bergenholtz. Theory of rheology of colloidal dispersions. Curr. Opin. Colloid Interface Sci. 6 (2001) 484 24. P. Mongondry, T. Nicolai, J. Tassin. Influence of pyrophosphate or polyethylene oxide on the aggregation and gelation of a queous laponite dispersions. J. Colloid Interface Sci. 275 (2004) 191.

PAGE 69

57 57 25. S. Carra, A. Sliepcevich, A. Canevarolo, S. Carra. Grafting and adsorption of poly (vinyl) alcohol in vinyl acetate emulsi on polymerization. Polymer 46 (2005) 13791384. 26. A. A. Zaman, N. Delorme. Effect of polymer bridging on rheological properties of dispersions of charged silica particles in the presence of low-molecular-weight physically absorbed poly (ethylene oxi de). Rheol Acta 41 (2002) 408-417. 27. A. A. Zaman, M. Bjelopavlic, B.M. Moudg il. Effect of adsorbed polyethylene oxide on the rheology of colloidal silica suspensi ons. J. Colloid Interface Sci. 226 (2000) 290-298. 28. R. Chanamai, D. J. McClements. Creaming stability of flocculated monodisperse oilin-water emulsions. J. Colloid Interface Sci. 255 (2000) 214-218 29. H. Anderlean, S. Petit, P. Laurens, P. Marc us, F. Arefi-Khonsari. Effects of different laser and plasma treatments on the inte rface and adherence between evaporated aluminum and polyethylene terepthalate films: X-ray photoemission and adhesion studies. Applied Surface Science 243 (2005) 304-318. 30. Jasco, Inc., Easton, MD. IR Application No te 02-03. Qualitative analysis of powdered solids with FTIR-ATR. 31. A.K. Pal, R.K. Roy, S.K. Randal, S. Gupta, B. Deb. Electrodeposited carbon nanotube thin films. Thin Solid Films 476 (2005) 288-294 32. Y.H. Wang, R. Kumar, X. Zhou, J.S. Pa n, J.W. Chai. Effect of oxygen plasma treatment on low dielectric constant carbondoped silicon oxide thin films. Thin Solid Films 473 (2005) 132-136. 33. J.E. Klemberg-Sapieha, L. Martinu, N.L. S Yamasaki, C.W. Lantman. Tailoring the adhesion of optical films on polymethyl-methacrylate by plasma-induced surface stabilization. Thin Solid Films 476 (2005) 101-107 34. D. H. Jeong, U. Erb, K. T. Aust, G. Pa lumbo. The relationship between hardness and abrasive wear resistance of electrodeposite d nanococrystalline Ni-P coatings. Scripta Materiala 48 (2003) 1067-1072 35. S. Suzuki, E. Ando. Abrasion of thin f ilms deposited onto glass by the Taber test. Thin Solid Films 340 (1999) 194-200 36. L. Dulany, 2002. Scratchand abrasion-resi stant coatings with energy-curable resin samples. UCB Chemicals Corp., Smyrna, GA.

PAGE 70

58 BIOGRAPHICAL SKETCH The author was born in Havertown, Pe nnsylvania, a small suburb outside of Philadelphia. She obtained her high school di ploma from Cypress Creek High School in Orlando, Florida. She went on to get her b achelors degree in materials science and engineering at the University of Florida in Gainesville, where she was active in the Benton Engineering Council and the Material s Research Society. She continued her education at the University of Florida to pur sue a Master of Scien ce in materials science and engineering. She worked with Dr. A bbas Zaman during undergraduate and graduate studies on colloidal systems and polymer thin films.


Permanent Link: http://ufdc.ufl.edu/UFE0010040/00001

Material Information

Title: Rheological and Abrasion Resistant Properties of Transparent Polymer/Silicate Nanocomposite Coatings
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0010040:00001

Permanent Link: http://ufdc.ufl.edu/UFE0010040/00001

Material Information

Title: Rheological and Abrasion Resistant Properties of Transparent Polymer/Silicate Nanocomposite Coatings
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0010040:00001


This item has the following downloads:


Full Text












THEOLOGICAL AND ABRASION RESISTANT PROPERTIES OF TRANSPARENT
POLYMER/SILICATE NANOCOMPOSITE COATINGS














By

JENNIFER MARIE BRANDT


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2005

































Copyright 2005

by

Jennifer Marie Brandt

































This document is dedicated to Michael Sachs, Kristin Brandt, and my parents, David and
Marie Brandt.















ACKNOWLEDGMENTS

I would like to express my appreciation to my committee chair and advisor, Dr.

Abbas Zaman. I would also like to thank Dr. Zaman's graduate student, Heather Trotter,

and his group of undergraduate students. Special thanks go to Betty Arias for her hard

work on may aspects of this work.

I would like to thank my committee members Dr. Charles Beatty and Dr. Susan

Sinnott for their advice and support.

I would also like to thank the students and staff of the Particle Engineering

Research Center at the University of Florida.

I would like to thank the Air Force for their financial support.
















TABLE OF CONTENTS


page

A C K N O W L E D G M E N T S ..................................................................................................iv

TA B L E O F C O N TEN T S ........................................... ............................................v........

LIST OF TABLES ......................................... ............... vii

LIST OF FIGURES ............................. .. .......... .............................viii

A B S T R A C T ........................................................................................................ ........ .. ix

1 IN TR O D U C T IO N ........ .. ......................................... ..........................................1.

1.1 P olym er N anocom posites ........................................ ....................... ..................
1.2 L literature R review ................ ................ .............................................4....
1.3 Project B background Inform ation ....................... ............................................6...

2 RHEOLOGICAL PROPERTIES AND STRUCTURAL DEVELOPMENT .............. 8

2 .1 In tro d u ctio n ............................................................................................................. 8
2 .2 M ateria ls ............................................................................................................... 1 1
2.3 E xperim ental Procedure........................................ ........................ ............... 13
2 .3.1 Sam ple P reparation ....................................... ...................... ............... 13
2.3.2 R heological M easurem ents ................................................... ................ 14
2 .3.3 M icroscopy ........................................................................................ 15
2 .4 D iscu ssion of R esults......................................... .. ....................... .. .......... ... 16
2.4.1 Shear V iscosity ............................ ...... .................... .. ....... ..... 16
2.4.1.1 Effect of shear rate on viscosity. ................................. ................ 16
2.4.1.2 Effect of solids loading on viscosity ...........................................21
2.4.2 Viscoelastic Properties and Structural Development ..............................22
2.4.3 M icroscopy ............... ............. .............................. 26

3 SU R FA CE A C TIV A TIO N ......................................... ........................ ................ 30

3 .1 In tro d u ctio n ........................................................................................................... 3 0
3 .2 M ateria ls ............................................................................................................... 3 2
3.3 E xperim ental P rocedure......................................... ....................... ................ 33
3.3.1 Sam ple P reparation ....................................... ...................... ................ 33



v









3.3.2 Fourier Transform Infrared Spectroscopy (FTIR)................................34
3.4 D iscu ssion of R esults......................................... .. ....................... .. ........... ..35
3.4.1 A cid Surface T reatm ents ....................................................... ................ 35
3.4.2 Silane Coupling Agent Treatments .......................................................39
3.4.3. Film Casting onto Treated Interlayer Substrates..................................40

4 ABRASION RESISTANT AND ADHESIVE PROPERTIES ................................42

4 .1 In tro d u ctio n ........................................................................................................... 4 2
4 .2 M ateria ls ............................................................................................................... 4 3
4.3 E xperim ental Procedure......................................... ....................... ................ 44
4 .3.1 Sam ple P reparation ....................................... ...................... ................ 44
4.3.2 T aber L inear A broader .......................................... .................. ................ 44
4.3.3 Visual Observation of Film Delamination ...........................................45
4 .4 D iscu ssion of R esults........................................... ......................... ................ 46
4.4.1 M ulti-L ayered Film s................................. .... ................ 46
4.4.2 Delamination of Coated Interlayer Samples..........................................50

5 CONCLUSIONS AND FUTURE WORK ................ ...................................52

5 .1 C o n c lu sio n s ........................................................................................................... 5 2
5 .2 F u tu re W o rk .......................................................................................................... 54

L IST O F R EFE R E N C E S .... ....................................................................... ................ 55

BIO GR APH ICAL SK ETCH .................................................................... ................ 58
















LIST OF TABLES


Table page

2-1: Several shear rates corresponding to common tape casting gap sizes (calculated
from casting speed/gap size) ...................................... ...................... ............... 11

2-2: Chemical composition of Laponite nanoparticles13 ........................12

2-3: Physical properties of Laponite nanoparticles13.................................... ............... 13

2-4: PVA/Laponite dispersion and film characteristics................................ ................ 17

3-1: Characteristics of Dupont SentryGlas Plus ionomer interlayer.............................. 32

3-2: Absorption frequencies for common bonds found in OMNIC software.................36

















LIST OF FIGURES


Figure page

1-1: Various phases observed for polymer-layered silicate systems ..............................3...

1-2: D im tensions of a Laponite particle 13 ........................ ................................7

2-1: Viscosity as a function of shear rate for dispersions of 10wt% Laponite with
increasing polym er dosages (250C )..................................................... ................ 17

2-2: Viscosity as a function of shear rate for dispersions of Laponite JS in 2 wt% PVA
solution (250C) at different wt% solids: (*) 0 wt% (m) 1.3 wt% (A) 1.96 wt%
(+) 2.91 w t%o ............................................................................ ............................... 18

2-3: Shear viscosity as a function of shear rate and solids loading for dispersions made
from 1 w t% PV A Solution (250C) ...................................................... ................ 19

2-4: Viscosity as a function of shear rate for dispersions of 4 wt% Laponite prepared
in 1 wt% PVA and 3 wt% PVA solutions (25C) (+) 1 wt% PVA (m) 3 wt%
P V A ....................................................................................................... ........ .. 2 0

2-5: Relative viscosity (with respect to the viscosity of the suspending fluid) as a
function of Shear rate for dispersions of 4 wt% wt Laponite dispersions prepared
in 1 wt% PVA and 3 wt% PVA solutions (25C) (+) 1 wt% PVA (m) 3 wt%
P V A ....................................................................................................... ........ .. 2 1

2-6: Viscosity as a function of solids loading for dispersions of Laponite particles
prepared in a 1 wt% PVA solution at two different shear rates (+) 1 s-1 (m) 1000
s ............................................................................................................................. 2 2

2-7: Storage modulus vs. angular frequency and polymer dosage for dispersions of 10
w t% L aponite particles (25C ) .................................................................................23

2-8: Storage modulus as a function of angular frequency for dispersions of Laponite
prepared in 2 wt% PVA solution (250C) (*) 0 wt% (m) 1.3 wt% (A) 1.96 wt%
(*) 2.91 w t%o ............................................................................ ............................... 24









2-9: Storage modulus as a function of angular frequency for dispersions of Laponite
particles prepared in 3 wt% PVA solution (250C) (+) 0 wt% (m) 1.96 wt% (A)
2.91 w t% (e) 4.3 w t% ........................... ....................................... 25

2-10: Loss modulus as a function of angular frequency for dispersions of Laponite
particles prepared in 2 wt% PVA solution (250C) (e) 0 wt% (m) 1.3 wt% (A)
1.96 wt% (+) 2.91 wt% ................................................ ................ 26

2-11: Loss modulus as a function of frequency for dispersions of Laponite particles
prepared in 3 wt% PVA Solution (250C) (e) 0 wt% (m) 1.3 wt% (A) 1.96 wt%
(* ) 2 .9 1 w t% ............................................................................................................. 2 7

2-12: SEM micrograph of flocculated Laponite particles cast from a 3 wt% PVA
dispersions with 50 wt% Laponite mass fraction in dried film...............................27

2-13: AFM images of dry films cast from a 3 wt% PVA matrix with final solids
loadings of A.) 0 wt% B.) 40 wt% C.) 50 wt% D.) 60 wt% ...............................29

3-1: Schematic of Attenuated Total Reflection (ATR)30............................32

3-2: Infrared spectrum for the surface of an untreated interlayer sample.......................36

3-3: Infrared spectrum for the surface of an interlayer sample treated with 100%
sulfuric acid solution. ........................... ............................................. 37

3-4: Infrared spectrum comparison for interlayer samples a) without acid treatment b)
10% sulfuric acid treatment c)100% sulfuric acid treatment ..............................38

3-5: Infrared spectrum comparison for interlayer samples a) without acid treatment b)
2:1 sulfuric/nitric acid treatment c) 3:1 sulfuric/nitric acid treatment ..................39

3-6: Infrared spectrum comparison for interlayer samples a) without silane treatment
b) 3-aminopropyl trimethoxysilane treatment c) N-(2-aminoethyl)3-aminopropyl
trim ethoxysilane treatm ent ....................................... ....................... ................ 40

4-1: Steel rods of different diam eters............................................................ ................ 46

4-2: Average weight loss data with standard deviation for abraded glass substrate and
substrate coated with 40, 50, and 60 wt% nanocomposite films...........................47

4-3: Average weight loss data with standard deviation for abraded glass substrate and
substrate coated with two layers of nanocomposite film (x/y: film x coated over
fi lm y ) ................................................................................................................... .... 4 8

4-4: Average weight loss data with standard deviation for abraded glass substrate and
substrate coated with three layers of nanocomposite film (x/y/z: film x coated
over film y coated over film z)............................................................ ................ 49









4-5: A comparison of average weight loss data with standard deviation for the films
which loss less than or equal to the amount of weight as the glass substrate ..........50















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

THEOLOGICAL AND ABRASION RESISTANT PROPERTIES OF TRANSPARENT
POLYMER/SILICATE NANOCOMPOSITE COATINGS

By

Jennifer Marie Brandt

May 2005

Chair: Abbas A. Zaman
Major Department: Materials Science and Engineering

Polymer nanocomposite thin films were cast from colloidal dispersions of

polyvinyl alcohol and Laponite JS. Rheological studies were performed to determine the

effect of shear rate and small amplitude oscillatory shear on the colloidal dispersions.

Structural development of the dispersions was determined as a function of polymer

concentration and Laponite solids loading. Solution casting of the dispersions resulted in

transparent coatings with strong adhesion to a glass substrate. Scanning electron

microscopy revealed the presence of flocculated particle clusters in the films.

Adhesion of the films to a hydrophobic polymer interlayer was achieved by

activating the surface of the polymer using wet-chemical treatments. The treatments

included mixtures of sulfuric and nitric acid and several silane coupling agents. The

success of the treatments was determined using FTIR-ATR surface analysis. Highly

concentrated acid treatments and several coupling agents were determined to be

successful in surface modification of the polymer.









To determine the abrasion resistance of the nanocomposite films, a Taber linear

abrader was used to abrade the samples and weight loss was observed. Samples with

single, double, and triple coatings of film with solids loadings of 40, 50, and 60 wt%

were cast onto glass slides and abraded by the instrument. Of these samples, the double-

coated films lost the least weight when compared to the glass slide. The films containing

high solids loadings had the best performance and the most reproducibility.














CHAPTER 1
INTRODUCTION

1.1 Polymer Nanocomposites

A polymer nanocomposite is defined as an organic or inorganic phase with at least

one dimension on the nanometer scale dispersed in a polymer matrix. Polymer

Nanocomposites (PNC's) are used because they have shown to have enhanced material

properties including improved barrier properties, tensile modulus and strength, flame

resistance, abrasion resistance, and reduced shrinkage and residual strength1. Many

industries are taking advantage of nanocomposite technologies because of recent findings

regarding the low weight % of clay needed to improve properties. This makes it possible

to create lightweight films and coatings for packaging with increased barrier properties.

There is estimated to be a 2 order of magnitude growth of the PNC industry by the year

2005. The following are examples of a few other uses of PNC's in industry:1,2

Fire Resistance the NIST is developing fire resistant and reduced char coatings
for windows and other applications.
Asphalt Modification Exxon is developing asphalt mixtures with improved
mechanical properties using nanoparticles in mixing.
Elastomers many tire companies are interested in using PNC's as tire
reinforcements.
Thermoset Polymers many industries are interested in films and coatings with
improved barrier properties, abrasion resistance, and thermal stability.
Polymer nanocomposites can be isotropic or anisotropic and can have a defined

structure and orientational order. Polymer nanocomposites are usually grouped into three

categories based on the dimensions of the dispersed phase. Some nanoparticles are

isodimensional, or have three dimensions in the nanometer range; these include all

spherical, disc-shaped, or clustered particles. When there are two dimensions in the









nanoscale you can have an elongated dispersed phase such as carbon nanotubes or

whiskers, which are commonly used as fillers in nanocomposites. Lastly, if there is one

dimension in the nanoscale the long sheets of filler can be stacked to create a layered

composite. This is achieved by the swelling of the polymer chains in between the layered

sheets. Layered silicates such as Laponite are isodimensional and can be dispersed in a

variety of ways.3-5

In layered-silicate systems, the particles can be arranged in several different

structures depending on their method of preparation. When the system is phase separated,

there are clumps of particles dispersed in a particle matrix with the particle and polymer

phases being immiscible. In an intercalated system, the particles are arranged in layers

with polymer chains swelling in between the stacked particle galleries, causing them to

separate slightly. When these stacked particles are separated and dispersed throughout the

polymer matrix, the system is exfoliated. The exfoliated structure is very desirable when

attempting to make a well dispersed system with uniform properties. Figure 1-1 is an

illustration of the different type of layered-silicate systems.45

There are several ways to prepare a polymer-layered silicate nanocomposite

system to achieve intercalation or exfoliation. In situ polymerization can be used to

intercalate a system by swelling the layered silicate particles in a liquid monomer and

then initiate polymerization to form polymer chains between the particle sheets. In melt

intercalation, if a polymer is compatible with the particle surface, the system is

intercalated in the molten state during processing. An exfoliated system can also be

achieved if the polymer is able to get in between the particle spacing. Layered silicates

can also be exfoliated by dispersing the particles in a soluble polymer matrix where










exfoliation occurs due to delamination of the stacked particles. This is the most common

method for creating an exfoliated system.5


Phase Separated


Intercalated System


Layered
Silicate
Particles


Polymer
Chains


Exfoliated System


Figure 1-1: Various phases observed for polymer-layered silicate systems

In a Polyvinyl Alcohol (PVA) system containing Laponite, the particles are

dispersed in the PVA matrix and a gel is formed. When the gel is dried into a film, most

of the particles will become intercalated, but steric interactions in the PVA can prevent

reaggragation if the molecular weight is sufficiently high.5 The use of water-soluble

polymers in these dispersions is advantageous because of environmental and safety issues

which are encountered when using common organic solvents. Because of its high aspect

ratio, Laponite has the ability to form a stable sol when dispersed in demineralized water

at low solids loadings. When mixed with a polymer, a colloidal gel is formed.6-12









1.2 Literature Review

Numerous studies have been conducted on the nanocomposite system composed

of a polymer matrix with different grades of Laponite as the dispersed phase. Most of the

information available on Polymer/Laponite nanocomposites is based on the polyethylene

oxide (PEO)/Laponite system. Similar to PVA, PEO is a water soluble polymer that is

inexpensive and available in a wide range of molecular weights. When dispersed with

surface modified Laponite particles, PEO has been shown to adsorb to the particle

surface, creating bridging between the particles. When shear is applied to this mixture,

the dispersion will become flocculated and a gelled network will form. These dispersions

can also exhibit relaxation behavior when aged for a certain amount of time, which is

dependent on the concentration of the PEO.

A recent study was done on the shear thickening and gelation of the

PEO/Laponite system with applied shear.6 In this work, the formation of a "shake-gel"

was observed when a low viscosity sol composed of low concentrations of PEO and

Laponite was vigorously shaken. The phase behavior of the gels, based on visual

observation, is dependent on both polymer concentration and Laponite solids loading.

The gelation mechanism for this system is the bridging between polymer chains when

particle surfaces are exposed during the application of shear. At very high concentrations

of PEO, the Laponite surface will become saturated and a gel will not form. If the PEO

concentration is low, the dispersion will remain as a low-viscosity sol, so these

boundaries establish a regime where a rigid gel can be formed. The authors performed

light scattering experiments to determine the adsorption behavior of PEO onto the

Laponite surface. Their data on the surface coverage proved to be consistent with the

phase behavior that was observed.









Another study gave further insight into the network structure that can be induced

by the application of shear to a PEO/Laponite system. In this study, the investigators

attempted to quantify the critical shear rate needed for shear-induced reorientation of the

Laponite particles so that a viscous gel can be formed. The authors also studied the

transitions from liquid like to gel like behavior that could be observed for these

dispersions. Oscillatory shear experiments were done to find this transition, which is the

point where the complex moduli intersect.

Another study examined the reversible gel behavior of the PEO/Laponite XLG

system of low solids loadings.8 Their study showed a composition regime where

aggregates can be deformed by applied shear thus exposing new surface area, which leads

to the formation of polymer bridges. This bridging, or flocculation, causes the gelation of

the system. When shear is stopped, these polymer bridges can break up causing a

relaxation in the storage modulus and reversible gel behavior. This study also determined

that the reversible sol-gel transition shows time dependent characteristics.

In a study done on solid nanocomposites, dispersions of Laponite in PEO were cast

onto glass substrates to form very thin films.9 The films were cast from PEO solutions of

0, 2, and 5 weight % with a Laponite mass fraction of 3 weight %. The resulting films

were observed to be transparent with good interfacial adhesion. The surface roughness

and morphology of the films were characterized using Atomic Force Microscopy. The

AFM images and the RMS roughness calculations showed that the particles are randomly

dispersed in the film with polymer chains connecting them. The films containing low

concentrations of PEO had this homogeneous structure with particles being roughly equal

in size. The film cast from the 5 wt% PEO solution was observed to be heterogeneous,









with agglomerated Laponite domains separated by excess PEO. It was also observed that

the roughness decreased as the PEO concentration increased because the excess polymer

creates a smoother surface. The films containing low concentrations of PVA are observed

to be exfoliated and randomly oriented with a high degree of surface roughness.

1.3 Project Background Information

Since the development of polymer nanocomposites, a wide variety of applications

for these materials have been identified. Because of the improved properties that can be

achieved at relatively low cost, stronger coatings and laminates can be made for military

and security purposes. With the recent threats of terrorism, hurricanes, and forced entry

crimes, new technologies in protective building materials have been developed. Dupont

has developed several glass laminates that provide protection from these dangers using

strong, transparent interlayers. For military applications involving bomb and blast

protection, an abrasion resistant material is needed for protection against sharp glass

shards. These small pieces of glass can become deadly projectiles when hit with a strong

blast. The strong ionomer interlayers can prevent large blasts from destroying a building,

but a large amount of destruction and loss of life is caused by these shards breaking off as

a result of the blast. This creates a need for an adhesive, abrasion resistant coating on the

polymer interlayer which will reduce or eliminate this phenomenon while maintaining the

abrasion-resistance of glass. The abrasion resistant coating should be clear, flexible, and

have a strong adhesion to the interlayer and the glass being used.

There are many nanocomposite systems that could be used to fabricate this coating.

Laponite is a good choice for a filler phase because of its high aspect ratio and its

transparency when suspended in water. Laponite particles are "water white", which

means that they are small enough to be unable to scatter light. A diagram of a single









Laponite particle is shown in figure 1-2. Surface treated Laponite particles can be mixed

with a water-soluble polymer to create a stable dispersion.

0.92nm







25nm



Figure 1-2: Dimensions of a Laponite particle 13

Depending on the polymer being used, a strong, transparent film can be cast using a

variety of casting methods. Polyvinyl Alcohol is known to form strong, chemical resistant

films with abrasion-resistant properties. The mixture of PVA with Laponite is primarily

used as a paper coating, but a system using a high molar mass PVA can form transparent

films with good tensile properties.5














CHAPTER 2
THEOLOGICAL PROPERTIES AND STRUCTURAL DEVELOPMENT

In the processing of thin, nanocomposite films, it is necessary to consider the

theological properties of the dispersions being used to cast the films. Changes in viscosity

and the development of a gelled network can influence the thickness, transparency, and

properties of the final film. This chapter investigates the theological properties of various

PVA/Laponite dispersions and the effect of solids loading and polymer dosage on shear

viscosity and structural development.

2.1 Introduction

Colloidal gels are formed when a stable liquid suspension, or "sol", of colloidal

particles is "gelled" using a chemical agent. In Laponite suspensions, this gelling agent is

usually a filler, binder, pigment, surfactant, or wetting agent.12'13 When Laponite particles

are initially dispersed in water, sodium ions present on the dry particle surface are held

between the negatively charged particles by electrostatic forces. These dispersions exhibit

low-viscosity, Newtonian rheology.13 Some Laponite grades have surface treatments that

enhance the sol-forming capability of the dispersion and provide electrostatic

stabilization of the sol. Laponite JS, which is used in this study, is modified with a

tetrasodium pyrophosphate (TSPP) surface treatment. The (P204)+ anions create a net

negative charge on the particles which causes the loose layer sodium ions create

repulsions between particles.5'13 These stable sols can then be "gelled" or flocculated with

the addition of a chemical agent.









Gelling agents usually come in the form of a polymer, electrolyte, or pH modifier.

These compounds reduce repulsive particle-particle interactions, causing thinning of the

electrical double layer present in the "sol". When the electrical double layer is

sufficiently neutralized, van der Waals attractive forces draw the particles together

causing a solid-like gel phase to form. The addition of polymer to a stabilized colloidal

dispersion will change the properties of the dispersion depending on whether the polymer

is nonadsorbing or strongly adsorbing. When a nonadsorbing polymer is added to a stable

sol it can induce depletion flocculation. This occurs when two particles approach each

other to a distance smaller than the radius of gyration of the polymer chain, thus the

polymer is excluded from the interparticle zone. The osmotic pressure between the

particles is reduced causing an attractive force to develop between particles.12'14

As flocculation proceeds, particle pairs (doublets) become particle triplets and so

on until flocs containing many groups of particles start to appear. If the flocs are densely

packed, they would form a close-packed solid structure. Instead, most flocs are open,

ramified structures, or fractals, which are self-similar structures which look the same at

all magnifications. As a floc grows, the porosity increases, creating a much larger, open

structure. The floc can eventually grow to fill up the volume fraction of a sample,

forming a percolated network, or gel.14,'15 The growth can be diffusion limited, where

small flocs form to create larger flocs, or it can be reaction limited, where the density of

the floc grows over time. The flocculation behavior of colloidal gels can effect the

rheological behavior of the dispersion.

Understanding the rheology of gels is very important when processing these

materials. Their response to shear rate, shear stress, and strain depend strongly on particle









size and shape as well as the gel strength and degree of flocculation. The rheological

behavior of flocculated gels is difficult to reproduce due to their shear history and time

dependent structure. 16-23 Two types of tests can be performed on colloidal gels to

determine their flow properties, these are steady-shear viscosity and small amplitude

oscillatory shear flow measurements. Steady shear viscosity is the most widely measured

aspect of rheology and can determine whether a material is Newtonian or non-

Newtonian. Using a parallel disk torsional rheometer, only a small amount of sample is

needed to obtain data about the behavior of the material. Small Amplitude Oscillatory

Shear (SAOS) experiments are useful in determining the viscoelastic properties of a

colloidal gel. SAOS can be used to observe structural development, relaxation behavior,

and various transitions for colloidal gels.21-23

Many rheological studies have been conducted on a colloidal system containing

Laponite of different grades in a matrix of polyethylene oxide (PEO). In these systems, if

the polymer concentration is not enough for full coverage, the PEO is adsorbed onto the

bare surfaces of the Laponite particles and a gel forms as a result of bridging

flocculation.24 When a strong shear is applied to a PEO/Laponite suspension, the polymer

chains can desorb from the particles and readsorb onto a different particle, forming a

"shake gel".6 Modulus relaxation data shows that these "shake gels" are reversible and

have aging characteristics.8

The mixture of Laponite JS (LJS) with low molecular weight Polyvinyl Alcohol

(PVA) has been used in the paper coatings industry to form coherent films with good

barrier properties.25 Relatively little is known about the rheological properties of a

PVA/Laponite mixture and the effect these properties have on film processing. There are









several methods that can be used to cast the PVA/Laponite films and the theological

behavior of the dispersions determines which methods are feasible. A low-shear method

would be solution casting, where the dispersion is poured over the substrate and set to dry

until the solvent is evaporated. Another technique would be tape casting, which can go to

very high shear rates. Some examples of shear rates at a speed of 10inches/minute are

shown in Table 2-1. To obtain a very thin film, on the order of 10tm, the shear rate

would be very high.

2-1: Several shear rates corresponding to common tape casting gap sizes (calculated from
casting speed/gap size)
Gap Size Shear Rate
3mm 1.41 s-'
Imm 4.23 s-'
0.5mm 8.46 s-'
0.1mm 42.33 s-
0.05mm 84.66 s-1
0.01mm 423.3 s-'


In this study, a high molecular weight PVA is used as the matrix polymer for

Laponite JS dispersions which will be used to make thin, optically clear coatings for glass

and polymer substrates.

2.2 Materials

The polymer used in this study is a partially hydrolyzed grade of Polyvinyl

Alcohol. The PVA (Aldrich Chemical Co. cat# 36310-3) has a molecular weight between

126,000-186,000 g/mol. PVA is a neutral, linear polymer synthesized from a vinyl

acetate monomer with a chemical formula given by -[CH2 CHOH]n. The PVA used in

this study has a degree of hydrolysis of 87-89%, which is the extent of conversion from

polyvinyl acetate to polyvinyl alcohol. The PVA is soluble in water and can easily form






12


thin films with good tensile strength properties on glass substrates. The films have good

chemical resistance and resistance to thermal degradation.

The clay particle used in this study is Laponite (Southern Clay Products, Inc.

Gonzales, TX). Laponite particles are synthetic layered nanosilicate clays with an

empirical formula given by: Na+o.7[(SisMg5s.Lio.3)O20(OH)4) -07.13 The Laponite crystals

are disk-shaped and are stacked when dry. The disks have a diameter of 25 nm and a

width of 0.92 nm.13 The unit cell consists of one layer of octahedral magnesium atoms

with a small amount of lithium impurities. This layer is sandwiched between two layers

of tetrahedral silicon atoms. Both layers are balanced by 20 oxygen atoms, 2 hydroxyl

groups, and sodium ions which are released when the particles are dispersed in water.13

Some chemical and physical properties of Laponite JS are shown in Tables 2-2 and 2-3.

The grade of Laponite used for this study is Laponite JS, which is surface treated with

tetrasodium pyrophosphate, an inorganic polyphosphate dispersing agent. All grades of

Laponite form clear colloidal dispersions when mixed with water.


2-2: Chemical composition of Laponite nanoparticles13

Weight
Material %

Si02 50.2

MgO 22.2

Li20 1.2

Na20 7.5

Ps05 5.4

F 4.8









2-3: Physical properties of Laponite nanoparticles13

Property Value
Appearance Free flowing white powder
Bulk Density 950 kg/m3
Surface Area
(BET) 300 m2/g
Loss on ignition 8.70%
pH (2%
suspension) 10
Hygroscopic, should be stored under dry
Storage conditions


2.3 Experimental Procedure

2.3.1 Sample Preparation

The polymer stock solutions are made by dissolving PVA granules in deionized

water at room temperature and stirring at moderate speeds for 24 hours with a magnetic

mixer. The Laponite dispersions are made by mixing 200 mL of polymer stock solution

with a given amount of Laponite (according to final solids loading in films) at pH 10.

These dispersions are mixed with a high shear mixer (Hamilton Beach Commercial Shear

Mixer 950) to ensure good mixing and prevent aggregation. Defoaming agents are used

when needed to prevent viscous bubbles from forming, which can alter the rheological

results and weaken the final films. The polymer/clay dispersions are agitated on a shaker

for 48 hours prior to rheological measurement to ensure good mixing. The samples were

kept mixing until they were measured to avoid effects of time dependency, which is

common for polymer/Laponite dispersions.6-8

The nanocomposite film samples were made by solution casting the

PVA/Laponite gel dispersions onto glass petri dishes at 350C. They were left to evaporate

in a convection oven for 24 hours, or until they were observed to be completely dry. The









film quality is improved by this slow evaporation process since the clay platelets have

more time to assemble under gravitational and osmotic forces16. These nanocomposite

films are optically clear with the Laponite phase trapped in the thin film resulting in good

interfacial adhesion to the glass substrate.9

2.3.2 Rheological Measurements

A Paar Physica UDS 200 Rheometer was used to examine the steady- and

oscillatory-shear-flow using a parallel plate geometry (plate radius, 2.5cm). Steady shear

flow experiments were conducted to determine steady shear viscosity (f) as a function of

shear rate (>) with shear rates ranging from 1-5000 s-1. Oscillatory shear flow

experiments were performed to determine the viscoelastic properties of the dispersions.

The storage (G') and loss (G") moduli are related to the real and imaginary components

of the complex viscosity, f' and f" respectively. With a given angular frequency (co) and

phase shift (6) they can be calculated as follows:


tan = G 2.1


77' = -G 2.2
ao

r" = G-- 2.3
Co

And finally, the complex moduli can be related to the complex viscosity by

G" G'
r1 = '- i" = G --i- 2.4
a0) 0

Where i 1= An angular frequency range of 1-50 Hz was chosen with a fixed strain

amplitude of 5% to maintain the measurements in the linear viscoelastic region.26-28

Sample evaporation is prevented by using a circular solvent trap covered with an









aluminum cap, which isolates the sample during measurement. All experiments were

performed at a temperature of 250C.

There are several sources of error that must be considered when performing

theological measurements. Factors such as humidity, sample size, and ambient

temperature can have an effect on the sample behavior. For the following experiments,

each batch of samples had one repeat for every three samples run. This repeat was

compared to the original data to verify the reproducibility of the experiment within +/-

3%. Since Laponite dispersions are known to have a time-dependency,8 all samples were

kept on a shaker until use.

2.3.3 Microscopy

Scanning Electron Microscopy (SEM) was done on the films to determine the

structure of the laponite particles in the cast films. Images were taken using a JEOL JSM-

6335F Scanning Electron Microscope to examine the structure inside a solution cast

nanocomposite film with a final solids loading of 50 wt% from a Laponite dispersion

prepared in a 3% PVA matrix. The film sample was mounted onto an aluminum mount

and coated with Gold-Palladium to improve conductivity. The image was taken at a

2000x magnification and a beam energy of lkv to prevent destruction to the polymer

matrix of the film.

Atomic Force Microscopy (AFM) was performed to examine the surface

morphology of the nanocomposite films. The films used for AFM were also cast from a 3

wt% PVA matrix with Laponite solids loading of 0, 40, 50, and 60 wt%. The

measurements were done at a setpoint of 2 volts, a scan rate of 3 Hz, and a scan size of 1

am.









2.4 Discussion of Results

2.4.1 Shear Viscosity

2.4.1.1 Effect of shear rate on viscosity. The steady shear viscosity (11) as a

function of shear rate (Q) was measured in three dispersion systems, one which varied

polymer dosage, one which looked at final solids loading in cast films, and one system

which varied Laponite solids loading. The effect of shear rate on the steady shear

viscosity of dispersions containing Laponite nanoparticles dispersed in solutions of PVA

at dosages ranging from 1 to 12 mg/(g solids) and a Laponite loading of 10 wt% is shown

in Figure 2-1. It can be observed at low polymer dosages the Laponite can form a stable

sol which exhibits a constant viscosity (Tlo) and Newtonian behavior in the shear rate

range examined. At high polymer dosages, the viscosity of the dispersions decrease with

increasing shear rate, i.e., they exhibit shear thinning, or thixotropic, behavior at high

shear rates. These data show that with increasing polymer dosage there is almost four

degrees of magnitude increase in viscosity, which can be seen for dosages above 3mg/(g

solids). As polymer is added to the Laponite dispersions, they swell between the particles

and cause an increase in the zero shear viscosity. With an increase in applied shear rate,

the particles and polymer chains align themselves in a more desirable flow structure,

causing shear thinning in the system.

The Laponite solids loadings used in Figure 2-2 are the actual concentrations of

filler that would be used to process nanocomposite films with final solids loadings of 40

wt%, 50 wt%, and 60 wt%. These samples were prepared in a 2 wt% PVA solution as the

suspending media and are high enough to produce films of good tensile strength. The

characteristics of these films are shown in Table 1. The data in Figure 2 shows the

relationship between shear viscosity and shear rate for several calculated Laponite solids






17


loadings in 2 wt% PVA solution. The dispersions containing very low weight fractions of

Laponite maintain a constant viscosity (ilo) and behave as Newtonian fluids throughout

the range examined. The further addition of Laponite causes an increase in the zero shear

viscosity (ilo) and the onset of shear thinning shifts to lower shear rates.


*o
S.


'* X X
A" A A A


A AA
+


A A A
nnn


* lmg/g solids
+ 2mg/g solids
A 3mg/g solids
x 5mg/g solids
* 8mg/g solids
* 12mg/g solids


> OOOO


100

shear rate, 1/s


Figure 2-1: Viscosity as a function of shear rate for dispersions of 10wt% Laponite with
increasing polymer dosages (250C).

Table 2-4: PVA/Laponite dispersion and film characteristics

PVA Mass Mass fraction Laponite mass
wt A n Fraction in dried Laponite in Solution fraction in dried
films (wt%) films

2% 60% 1.3 40%
2% 50% 1.96 50%
2% 40% 2.91 60%
3% 60% 1.96 40%
3% 50% 2.91 50%
3% 40% 4.3 60%


1000


100


10


1


0.01


1000


10000












100


10 **** **
AAAAA


SAA A *
A A *





0.01



0.001
1 10 100 1000 10000
shear rate, 1/s

Figure 2-2: Viscosity as a function of shear rate for dispersions of Laponite JS in 2 wt% PVA
solution (250C) at different wt% solids: (e) 0 wt% (m) 1.3 wt% (A) 1.96 wt%
(+) 2.91 wt%.

The shear viscosity data are shown in Figure 2-3 for dispersions of Laponite

particles at solids loadings ranging from 0-5 wt% prepared in a 1 wt% PVA stock

solution. The viscosity shows changes with both shear rate and volume fraction of the

particles. At a fixed weight fraction and at low shear rates, all samples show Newtonian

behavior. For these low weight fractions, lower shear rates are not in the linear range of

the instrument, and are therefore left out of the data. At moderate shear rates, the

viscosity falls monotonically with increasing shear rate and there is no indication of a

second plateau at high shear rates. There is at least three orders of magnitude change of

viscosity as the solids loading is increased. A transition from a shear rate independent

region (Newtonian plateau) to a shear rate dependent region occurs at low shear rates as

the solids loading is increased. Over the shear rate dependent region, the plots of






19


viscosity as a function of shear rate are approximately linear. Therefore, the linear region

of the plots of log fI as a function of log / can be described by the power law model.


10







0.1 0 wt% solids x x
xe

0. 1 wt% solids x X 0
x
A 2 wt% solids x x
A A X
0.01 x 3 wt% solids A A A

x 4 wt% solids 0
5 wt% solids
0.001
0.1 1 10 100 1000 10000
shear rate, 1/s

Figure 2-3: Shear viscosity as a function of shear rate and solids loading for dispersions made
from 1 wt% PVA Solution (25C).

Figure 2-4 represents the shear viscosity data for dispersions of Laponite particles

of 4 wt% solids loading dispersed in solutions of 1 wt% PVA and 3 wt% PVA solutions

respectively. It can be observed that the dispersion containing 3 wt% PVA has a viscosity

that is one degree of magnitude larger than that for the I wt% PVA. Both show

Newtonian behavior at low shear rates and shear thinning behavior at high shear rates.

Additionally, the onset of shear thinning occurs at the same shear rate for both curves.












100



10 0



1



0.1



0.01
0.1 1 10 100 1000 10000
shear rate, 1/s
Figure 2-4: Viscosity as a function of shear rate for dispersions of 4 wt% Laponite prepared in 1
wt% PVA and 3 wt% PVA solutions (25C) (+) 1 wt% PVA (m) 3 wt% PVA.

The relative viscosity data as a function of shear rate for the above mentioned

dispersion are presented in Figure 2-5. The viscosity data are normalized with respect to

the viscosity of 1 wt% and 3 wt% PVA solutions as the suspending media. There is not a

significant difference in the relative shear viscosities of the two dispersions. This

behavior indicates that increase in the viscosity is solely due to increase in the viscosity

of the suspending media with PVA concentration. Since the shape of the curves stays the

same when the effect of the polymer is removed, it may be that polymer-particles

interactions in the polymer/clay dispersions are not dominating the rheological behavior

of the system.










1000




100









1
1 10 100 1000 10000
shear rate 1/s

Figure 2-5: Relative viscosity (with respect to the viscosity of the suspending fluid) as a
function of Shear rate for dispersions of 4 wt% wt Laponite dispersions prepared
in 1 wt% PVA and 3 wt% PVA solutions (25C) (+) 1 wt% PVA (m) 3 wt% PVA.

2.4.1.2 Effect of solids loading on viscosity. As with most general polymer/filler

systems, it appears that the differences in viscosity at various volume fractions are more

significant at low shear rates. The viscosity as a function of solids loading at shear rates

of 1 s-1 and 1000 s-1 for dispersions prepared in 1 wt% PVA solutions is shown in Figure

2-6 and can be observed at low shear rates. Addition of the Laponite filler can cause the

viscosity to increase by an order of magnitude. At low shear rate, the viscosity of the

system is dominated by particle-particle, polymer-particle, and structure of the

dispersions. While at high shear rates, hydrodynamic forces are dominant and control the

viscosity of the system. From a processing point of view, it is more convenient to process

these materials at higher shear rates due to the significant decrease in viscosity at higher









shear rates. At low shear rates, processing would be very slow and can result in defects

such as bubbles and cloudy spots in the final product.




1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2


0 2 4 6
wt% solids

Figure 2-6: Viscosity as a function of solids loading for dispersions of Laponite particles
prepared in a 1 wt% PVA solution at two different shear rates (+) 1 s-1 (m)
1000 s-1

2.4.2 Viscoelastic Properties and Structural Development

The effect of polymer dosage on the structural development of the colloidal gels

can be determined using small amplitude oscillatory shear measurements at different

dosages. The response of the dispersions, at 10 wt% Laponite, to small-amplitude

oscillatory-shear flow is shown in Figure 2-7. At the lowest polymer dosage (lmg/(g

solids)), a low storage modulus can be observed, suggesting that at this concentration, the

dispersion behaves as a viscoelastic liquid, unable to store large amounts of elastic

energy. In the range of 2-5mg/(g solids), there is a structural transition between liquid

and solid behavior where the value of the storage modulus drops, indicating a breakdown









in the flocculated viscoelastic gel structure. At polymer dosages above 5mg/(g solids),

there is a very weak frequency dependence of the storage modulus, so the viscoelastic gel

does not break down at high frequencies. This transitional behavior of the dispersions at

moderate polymer dosages suggests a dependence of the structural breakdown of the gel

on the polymer concentration used. The higher the polymer concentration, the less likely

it is for a breakdown from solid-like behavior to liquid-like behavior to occur. The

interactions between polymer and particles contribute to the elastic properties of the

dispersions, especially since the dispersions are initially flocculated. At the frequencies

where the breakdowns are occurring, the weakly bonded percolated network structure is

breaking down.



1.00E+03 -- -******,***************

1.OOE+02 unm muummu

1.OOE+01
10E+01 xxxxxxxxxxxxxxxxxxxxX
1.OOE+00

1.OOE-01 -AAAAAAA k

O 1.OOE-02 1mg/g solids

Q 1.OOE-03 0 2mg/g solids
A 3mg/g solids
1.OOE-04 x 5mg/g solids 0

1.OOE-05 8mg/g solids^0A**
12mg/g solids ^ 4
1.OOE-06 g$

1.OOE-07
1 10 c, rad/s 100 1000

Figure 2-7: Storage modulus vs. angular frequency and polymer dosage for dispersions of 10
wt% Laponite particles (250C).









A comparison for the storage modulus data as a function of angular frequency for

dispersions of Laponite (at calculated solids loadings) in 2 wt% and 3 wt% PVA matrices

are shown in Figures 2-8 and 2-9. In both plots, the dispersions containing no Laponite

show liquid-like behavior throughout the range examined. For the 2 wt% dispersions,

there is a breakdown of the gel structure from solid-like behavior to liquid-like behavior

at high frequencies for all Laponite. This occurs for all three Laponite solids loadings,

although the onset of liquid-like behavior shifts to high frequencies as the solids loading

is increased. For the 3 wt% PVA dispersions, the storage modulus has a weak

dependence on frequency and stays relatively constant through the range shown,

increasing slightly as the frequency is increased.





1.OOE+02
1.OOE+01
AAAAAAAAAAA A
1.00E+00 Amnnnn A
1.OOE-01
a 1.OOE-02

1.OOE-03
1.OOE-04 lhA

1.OOE-05 g
1.OOE-06 o ***

1.OOE-07
10 100 1000
co, rad/s
Figure 2-8: Storage modulus as a function of angular frequency for dispersions of Laponite
prepared in 2 wt% PVA solution (250C) (*) 0 wt% (m) 1.3 wt% (A) 1.96
wt% (+) 2.91 wt%.












1.00E+03


1.00E+01
1.00E+00
1.00E-01
1.00E-02
1.00E-03
1.00E-04
1.00E-05 -
1.00E-06
1.00E-07
1 10 100 1000

co, rad/s

Figure 2-9: Storage modulus as a function of angular frequency for dispersions of Laponite
particles prepared in 3 wt% PVA solution (250C) (+) 0 wt% (m) 1.96 wt% (A)
2.91 wt% (e) 4.3 wt%.

The corresponding loss modulus data for the 2 wt% and 3 wt% PVA samples are

shown in Figures 2-10 and 2-11. At both PVA concentrations, the samples containing no

Laponite have a steady increase in loss modulus as the angular frequency is increased.

Comparing these with the storage modulus data, the value of loss modulus is consistently

higher than that of the storage modulus, suggesting that the dispersions containing no

Laponite exhibit liquid-like behavior for all frequencies. For the 2 wt% PVA dispersions

containing Laponite, the loss modulus stays steady at low frequencies with the magnitude

of the storage modulus being larger than that of the loss modulus, showing solid-like

behavior. At higher frequencies, there is a sharp increase in loss modulus, and the

dispersions go from solid-like behavior, to liquid-like behavior. This phenomenon is not

observed for the 3 wt% PVA dispersions because of the increase in polymer









concentration. As the angular frequency is increased, the loss modulus stays constant at a

value below the storage modulus, showing solid-like behavior for all angular frequencies.





1.00E+03



1.00E+02 A
Cl

^ 1.00E+01 ,,ol**** ,******* ***



1.00E+00



1.00E-01
10 100 1000
co, rad/s

Figure 2-10: Loss modulus as a function of angular frequency for dispersions of Laponite
particles prepared in 2 wt% PVA solution (250C) (*) 0 wt% (m) 1.3 wt% (A)
1.96 wt% (+) 2.91 wt%.

2.4.3 Microscopy

Figure 2-12 is a scanning electron microscope image depicting the fractal

structure of the Laponite nanoparticles dispersed in a polymer matrix. This floc appears

to be 30-40tm in diameter, and is frozen in the film. The floc is not connected with other

flocs in the vicinity. It is apparent from this image that this is not a percolated networked

structure, which is formed when a fractal network grows to create an interconnected

three-dimensional network structure. The formation of the fractal network in these films

may be due to aggregation over long time scales.15













1.00E+03

1.00E+02

1.00E+01

1.00E+00

1.00E-01

1.00E-02

1.00E-03

1.00E-04

1.00E-05

1.00E-06

1.00E-07


100


1000


to, rad/s


Figure 2-11: Loss modulus as a function of frequency for dispersions of Laponite particles
prepared in 3 wt% PVA Solution (250C) (e) 0 wt% (m) 1.3 wt% (A) 1.96
wt% (+) 2.91 wt%.


Figure 2-12: SEM micrograph of flocculated Laponite particles cast from a 3 wt% PVA
dispersions with 50 wt% Laponite mass fraction in dried film.


gags gegee eSSO 00000 SOS 0000
AAAA AAAAAAAAAAAAAAAAAAAA
* flEE mEEKE *UME *umEE HUE EE*~









The freely dispersed flocs present in the SEM image of the PVA/Laponite system

interact through non-covalent bonds such as van der Waals attractive forces and

hydrostatic interactions. Under high shear rate or a high frequency, the weakly bonded

gel can breakdown, creating a discontinuity in solid-like behavior, as previously

discussed. The dimensions and density of the fractal structures have an effect on the

properties of the dispersions and the cast films. These properties can be determined using

computational modeling and mathematical analysis.24

Atomic force microscope images for several different film solids loadings are

shown in Figure 2-13. These pictures show that the clay particles are randomly oriented

in the film and are connected by polymer.9 A film containing no laponite is shown in

Figure 2-13A, this image shows a smooth surface with some variation in height but a

very small roughness. The images of the films containing Laponite particles show

surfaces with some roughness and many height variations due to particles at the surface.

The 40 wt% film has some where the height decreases, suggesting that the polymer

matrix exists between the particles. As the solids loading is increased, there are less areas

where the polymer exists and the height increases, which may due to particles

aggregating on film surface.
















A B









C D

Figure 2-13: AFM images of dry films cast from a 3 wt% PVA matrix with final solids loadings
of A.) 0 wt% B.) 40 wt% C.) 50 wt% D.) 60 wt%.















CHAPTER 3
SURFACE ACTIVATION

3.1 Introduction

In order to adhere polymer nanocomposite films onto a polymer substrate, there

needs to be good interfacial adhesion between the two materials. Many polymer

substrates are hydrophobic, or have a high surface tension, and it is difficult to adhere a

water-based film to that type of surface. To form a bond between the film and the

substrate, the surface tension must be reduced and the surface of the substrate must be

activated so that covalent bonds can form between the film and the surface.29

There are several ways to activate the surface of a material to create functional

groups capable of improving interfacial adhesion. These methods include wet-chemical

treatments, mechanical treatment, exposure to flames, ultraviolet radiation, and ion

beams.29-33 Plasma treatment is considered the most efficient way to activate a polymer

surface for adhesion to metal.29 This technique can provide surface activation without

effecting the bulk properties of the material. In a gas plasma process, weak bonds on the

surface of a material are replaced by highly reactive carboxyl, carbonyl, and hydroxyl

groups. This is achieved by exposing the surface of a material to ionized gas containing

ions, atoms, electrons and neutral species; this process is done in a vacuum.30 Although

there are many advantages to this process, it requires expensive equipment. The most cost

effective way to create surface chemical functionalization is by using various wet-

chemical treatments. These treatments work by removing a weak surface boundary layer









from the polymer interface, creating covalent bonds at the surface. Some common

materials used for these treatments include various acids and coupling agents. When

exposed to a polymer surface, these chemicals will change the composition of the

surface, creating active domains capable of bonding with a hydrophilic interface.32

In order to determine the success of a surface treatment, the chemical composition

of the surface must be analyzed before and after the treatment. Fourier Transform

Infrared (FTIR) spectrometry is ideal for evaluating changes on the surface of a material.

Using Attenuated Total Reflectance (ATR), a sampling of the material surface can be

taken with little sample preparation compared to traditional transmission sampling.30'32

For this method, a total internally reflecting IR beam will come in contact with a dense

crystal with a high refractive index at a certain angle. When the sample is in direct

contact with the infrared beam it will be absorbed by an IR absorbing material sample.

The adsorption wave will be attenuated in the sample and the data will be sent to a

detector and an infrared spectrum will be generated. A schematic of the mechanism of an

ATR sampling is shown in Figure 3-1. Sample spectra collected using ATR are different

from those obtained by transmission. Shifts in band intensity and frequency can result in

displacements of peak maximums and intensity.30 Because of these differences, it is

difficult to perform quantitative analysis using ATR.









Sample in
contact with
Crystal





7Detector


IR Beam ATR
Crystal

Figure 3-1: Schematic of Attenuated Total Reflection (ATR)30

3.2 Materials

A SentryGlas Plus Ionomer Interlayer was obtained from the Air Force and was

manufactured by Dupont. It is composed of an ethylene/Methacrylic Acid copolymer

with trace amounts of sodium salts. The material is non-toxic and stable at normal

temperatures. Some of the typical properties of the interlayer are shown in Table 3-1.

3-1: Characteristics of Dupont SentryGlas Plus ionomer interlayer


Characteristics Metric Unit Metric Value Test
ASTM D-
Young's Modulus MPa 300 5026
ASTM D-
Tear Strength MJ/m3 50 638
ASTM D-
Tensile Strength MPa 34.5 638
ASTM D-
Elongation % 400 638
ASTM D-
Density g/cm3 0.95 790
Flex. Modulus ASTM D-
(73F) MPa 345 790
Coefficient of
Expansion (- ASTM D-
20C to 32C) 10-5 cm/cm C 15-Oct 696









The acids used for the surface treatments were sulfuric acid (Fisher Scientific) and

nitric acid (Fisher Scientific). Several different silane coupling agents were used for

surface treatments. The silanes used in this study are listed below:

2 (3,4 Epoxycyclohexyl) Etheltriethyl Triethoxysilane
5,6 Epoxyhexyltriethoxysilane
11 (Triethoxysilyl)udecanal
3 Aminopropyl Trimethoxysilane
N (2 Aminoethyl) 3 Aminopropyl Trimethoxysilane
N (2 Aminoethyl) 3 Aminopropylsilanetriol 25%
Bis (2 Hydroxyethyl) 3 Aminopropyl Triethoxysilane, 62% in
ethanol
Vinyl terminated PDMS fumed silica reinforced
Pentamethyl cyclopentasiloxane

Some of these silane coupling agents turned the interlayer samples yellow, so not all of

them were used as film substrates.

3.3 Experimental Procedure

3.3.1 Sample Preparation

Two different chemical treatments were used to activate the surface of the

interlayer samples, acid treatments and silane coupling agents. The acid treatment was

used to oxidize the surface by varying the concentration of a mixture sulfuric and nitric

acid. Treatments containing no nitric acid were also used, but the sulfuric acid was

diluted to different levels. For these treatments, the interlayer sheets were thoroughly

cleaned with deionized water and cut into 1.5 inch by 3 inch rectangular coupons using a

shear cutter. The acid solutions were mixed on a magnetic stir plate and allowed to

equilibrate to room temperature. The interlayer coupons were dipped into the acid

solutions for two minutes using Teflon-coated tweezers. After being removed from the

solution, the treated samples were dipped in water to clean off any excess acid solution









and dried with a towel. This method was used for all acid-treated samples at all

concentrations.

For the silane coupling agent treatments, a lml transfer pipette was used to drop the

coupling agent onto the surface of the interlayer coupons. All samples were treated for

one minute and then washed with deionized water until all excess coupling agent was

removed from the surface. Any samples that instantly yellowed on contact with the silane

coupling agent were disqualified as surface modifiers and were no longer used in the

experiment.

Nanocomposite films were cast onto the treated interlayer substrates by solution

casting in a convection oven at 25C. The interlayer coupons were put into plastic Petri

dishes and 75mL of the dispersions were poured into the dish to cover the interlayer

samples. Dispersions in 1, 2, and 3 wt% polymer matrices were prepared with solids

loadings corresponding to dry film concentrations of 40, 50, and 60 wt% for each

polymer solution.

3.3.2 Fourier Transform Infrared Spectroscopy (FTIR)

The FTIR experiments were performed using a Nicolet MAGNA 760 Bench with

Spectra Tech Continugm IR Microscope. The samples were analyzed using the IR

Microscope with an ATR objective slide-on microscope accessory. A silica crystal with a

refractive index of 3.4 was used to analyze the sample by contact with the surface. The

microscope assembly was cooled with liquid nitrogen before using the microscope and

ambient gases were pumped into the IR chamber. The samples were loaded onto the

microscope platform, surface treated side facing the crystal. Before the sample comes in

contact with the crystal, a background spectra is taken to assess the amount of noise in

the instrument before sampling. The microscope is then pressed onto the sample surface









until the Contact AlertTM system signaled the position where optimal contact pressure is

achieved for repeatable results. A set number of samples are taken until a smooth curve is

produced. The single beam spectra from the background and then again from the sample

are combined and a total absorbance is determined using the following formula:30

1 sample
log() -log( sample 3-1
R Background

This is also related to sample reflectance (r) by


log( ) = -log(r) 3-2
R

The magnitude of log(1/R) is also proportional to the concentration of bonds on a sample

surface, which is useful in performing qualitative analysis on the change of a sample

surface.

3.4 Discussion of Results

3.4.1 Acid Surface Treatments

There were several different acid treatments applied to the interlayer samples to

produce a change in the surface properties. The FTIR provided information about

changes in the chemical structure of the interlayer surfaces. Figure 3-2 is an FTIR curve

for the untreated interlayer specimen. The largest peaks in the spectrum correspond to

two very large CH2 peaks at 2917 cm-1 and 2850 cm-1. There are also several smaller

peaks that also correspond to C-H and CH2 bonds in polyethylene, some of these peaks

are represented in Table 3-2 from information obtained on the OMNIC software. The

spectral libraries in the ONMIC software matched the curves with that of low density

polyethylene, which is consistent with the polyethylene matrix of the interlayer material.

The second phase of the interlayer copolymer, methacrylic acid, was not observed in the









spectral libraries. This spectrum was reproducible when moved to different sections of

the sample and when analyzed several times at one spot.

0.22

0.20 Untreated Interlayer H20 C02 BLC

0.18

0.16

0.14

S0.12

0.10

0.08

0.06

0.04

0.02-

0.00- j
4000 3500 3000 2500 2000 1500 1000 500
Wavenumbers (cm-1)

Figure 3-2: Infrared spectrum for the surface of an untreated interlayer sample.

Table 3-2: Absorption frequencies for common bonds found in OMNIC software.
frequency range
Bond Assignment frequency range
cm1
CH, CH2, Stretching 2850-3000
CH3
675-1000 and 1350-
CH bend 1470
1470
CH2 wag 1450-1470
C=0 stretch 1650-1870
Si-OR stretch 1000-1100
Si-H silane stretch 2100-2360










The FTIR plot in Figure 3-3 is for an interlayer sample treated with a 100%

concentrated solution of sulfuric acid. This highly concentrated solution caused some

yellowing on the surface of the interlayer, but a change in the surface chemistry is

observed. The two CH2 peaks are still present in the spectrum, but a new large peak is

observed at a wavelength of approximately 1150 cm-1, which corresponds to a carbonyl

(C=0) stretch bond frequency. Using the spectral libraries on the OMNIC software, this

value corresponds to an acid group on the surface of the interlayer. It can also be

observed that the height of the CH2 peaks at 2917 cm1 and 2850 cm1 is less that that

observed for the untreated interlayer. This occurs because the new activated groups on

the surface of the interlayer increased in concentration, so the amount of polyethylene

observed on the surface will decrease.

0.15
Sulfuric 100%
0.14
0.13
0.12
0.11
0.10
0.09
S0.08
o 0.07
-J
0.06
0.05
0.04
0.03
0.02
0.01
0.00
4000 3500 3000 2500 2000 1500 1000 500
Wavenumbers (cm-1)

Figure 3-3: Infrared spectrum for the surface of an interlayer sample treated with 100% sulfuric
acid solution.










Several concentrations of the sulfuric acid treatments were used to activate the

surface of the interlayer samples. Figure 3-4 is a comparison between the untreated

interlayer specimen, a 10% sulfuric acid treatment, and a 100% sulfuric acid treatment. It

can be observed that the 10% sulfuric acid solution has almost no effect on the surface of

the interlayer while the 100% sulfuric acid solution has a very large effect. This suggests

that the extent of surface activation depends on the concentration of the treatment being

used. An optimal treatment would be one that sufficiently activates the surface of the

interlayer without discoloration.


0.20 Untreated interlayer 1 -
S H20 C02 BLC
-0.15
o
- 0.10




S0.15
0
J 0.10
0.05
0.00
0.15 100% sulfuric 2
-0.10


-J^ J^ __ -W^
0.05


4000 3500 3000 2500 2000 1500 1000 500
Wavenumbers (cm-1)

Figure 3-4: Infrared spectrum comparison for interlayer samples a) without acid treatment b)
10% sulfuric acid treatment c)100% sulfuric acid treatment.

A sulfuric acid and nitric acid mixture was also used to treat the surface of the

interlayer. This mixture was used to create a strong oxidizing effect on the material

surface while preventing discoloration. A comparison between an untreated interlayer, a

treatment using 2:1 sulfuric acid to nitric acid, and a treatment using a 3:1 sulfuric to










nitric acid treatment is shown in Figure 3-5. The carbonyl stretch bond is still present on

the material surface at comparable concentrations for both samples (concentration is

proportional to absorbance). Although there is not a substantial difference in the curves,

there was discoloration in the 3:1 sulfuric acid sample.

0.20 Untreated



-j
S0.15
o 0.10
0.05
-0.00




0.15 Sulfuric/Nitric 3:1
0.10

0.05



0.00



Wavenumbers (cm-1)

Figure 3-5: Infrared spectrum comparison for interlayer samples a) without acid treatment b) 2:1
sulfuric/nitric acid treatment c) 3:1 sulfuric/nitric acid treatment3:1


3.4.2 Silane Coupling Agent Treatments

All of the silane coupling agents listed above have been used to treat the interlayer

samples, but only some were successful based on surface discoloration. Some of the


coupling agents caused substantial yellowing and destruction of the sample. The two
successful silane coupling agent treatments were 3-aminopropyl trimethoxysilane and N-


(2-aminoethyl) 3-aminopropyl trimethoxysilane. Figure 3-7 is an FTIR spectrum
comparison for the 3-aminopropyl trimethoxysilane and c N-(2-aminoethyl) 3-


aminopropyl trimethoxysilane coupling agents. The most distinguishing characteristic of







40


these plots is the large peak at 1100 cm-1 that corresponds to a Si-OR stretching bond.

Since the height of the absorbance peak is proportional to the concentration of bonds on

the surface, the sample treated with 3-aminopropyl trimethoxysilane has a large amount

of functionalized silane groups on its surface.


0.20 Untreated
0.15
S0.10
-j
0.05
-0.00
0.15 3-aminopropyl TMS

0.10
-j
" 0.05

-0.00
aminopropy trimethoxysilane tr0.20 N-(2-aminoethyl)3-aminopropyl
The absorbance spectrum for N-(2-aminoethyl)0.15 3-aminopropyl TMS
0.10
-j
0.05

4000 3500 3000 2500 2000 1500 1000 500
Wavenumbers (cm-1)

Figure 3-6: Infrared spectrum comparison for interlayer samples a) without silane treatment b) 3-
aminopropyl trimethoxysilane treatment c) N-(2-aminoethyl)3 -aminopropyl
trimethoxysilane treatment.

The absorbance spectrum for N-(2-aminoethyl) 3-aminopropyl trimethoxysilane.

The silane peak is not as pronounced for this coupling agent, but the shift in the

absorbance peaks suggests that the surface has still been substantially modified.

3.4.3. Film Casting onto Treated Interlayer Substrates

The PVA/Laponite dispersions described above were coated onto the substrates of

interlayer samples treated with a 2:1 sulfuric/nitric acid treatment and a treatment of 3-

aminopropyl trimethoxysilane. These treatments were determined to produce the least






41


discoloration to the interlayer surface and the most surface change according to FTIR-

ATR qualitative analysis. The surfaces treated with 2:1 sulfuric/nitric acid showed the

best film adhesion based on visual inspection. The surfaces treated with 3-aminopropyl

trimethoxysilane had delamination during the drying process.















CHAPTER 4
ABRASION RESISTANT AND ADHESIVE PROPERTIES

4.1 Introduction

When formulating an abrasion resistant coating, the performance of the coating

depends on several factors. The polymer molecular weight, filler content, drying method,

processing temperature, and dispersion rheology can effect the final properties of the

protective coating. The scratch resistance and abrasion resistance of protective films are

often confused, but maximizing one of these properties may result in a reduction in the

other. Abrasion and scratch resistance are related to different material properties. Scratch

resistant materials usually have a high hardness and little flexibility. Conversely, abrasion

resistant materials depend on the toughness of the coating and are usually flexible and

have a high tensile strength.34-36

There are several methods for testing the abrasion resistance of coatings. There

are several standard test methods for testing abrasion resistance, ASTM D 4060 being the

most common.36 In this method, a Taber Linear Abraser with a grinding wheel is used to

abrade a coating on a surface over a fixed amount of time. The weight loss of the coating

is measured as a function of number of cycles. The Taber apparatus can be fitted with

many different accessories for measuring abrasion resistance, scratch resistance, and

color transfer. For smaller samples, such as the samples used in this study, a Taber Linear

Abraser (Model 5750) can be used to test small areas and contoured surfaces.









The Taber Linear Abraser can be used to analyze wear resistance of a wide

variety of materials. The abrasion tests can be programmed to up to 999,999 cycles with

operating speeds of 2, 15, 25, 40, and 60 cycles per minute. The initial load on the

specimens is 100 grams, which is loaded by the spline shaft. The load can be increased in

250 gram increments using additional weight discs. The free floating arm has operating

stroke lengths of 0.5", 2", 3", and 4", which can be chosen depending on the size of the

specimen. The actual abrading of the surface is done by using a wearaser which is an

abradant made of the same calibrase and calibrade material as the grinding wheels that

are commonly used. The calibrase material is resilient and consists of rubber and small

abrasive particles. The calibrade wearaser is non-resilient and is composed of vitrified

stone and small abrasive particles.

4.2 Materials

Several colloidal dispersions with different Laponite filler content were used to

make the multi-layered nanocomposite films. For the samples cast using the doctor blade,

a 3 wt% PVA matrix was used with final Laponite loadings of 40, 50, and 60 wt% after

drying. This high concentration of PVA is needed for this technique because it will form

a gel. A dispersion with a low viscosity would move through the feeder too fast, making

it difficult to coat the films uniformly. These multi-layered samples were coated onto

extra wide glass slides.

The interlayer samples used were treated with a 2:1 sulfuric to nitric acid solution.

The samples were cut into coupons and were coated with PVA concentrations of 1 wt%,

2 wt%, and 3 wt% with Laponite solids loadings corresponding to a dry film

concentration of 40 wt%, 50 wt%, and 60 wt% for each PEO concentration. These

samples were solution cast and left in a convection oven at 250C for 24 hours.









Two different wearasers were used to abrade the samples using the Taber Linear

Abrader. The CS-10 Calibrase Wearaser was used to abrade the films mounted on the

glass slides. The H-18 Calibrade Wearaser was used to abrade the single-layered films

on the interlayer samples.

4.3 Experimental Procedure

4.3.1 Sample Preparation

A doctor blade (Richard E. Mistler TTC-1000) was used for tape casting the multi-

layered films onto the glass slides. Sufficiently gelled dispersions were loaded into a

stationary feeder apparatus that is connected to a moving belt made of silicone coated

Mylar. The samples were cast at a speed of 10 inches/minute, which roughly corresponds

to a shear rate of 1.41 s-1. The films were cast onto wide glass slides and left to dry

overnight with air circulating over them. After the first coating, some of the samples were

coated again to create a double layer, and some from that group were coated again to

form a triple layer.

Interlayer samples treated with a 2:1 sulfuric/nitric acid mixture were coated with

nanocomposite films to analyze delamination. Dispersions in 1, 2, and 3 wt% polymer

matrices were prepared to cast films with solids loadings corresponding to dry film

concentrations of 40, 50, and 60 wt% for each polymer solution. The films were cast in a

convection oven at 250C.

4.3.2 Taber Linear Abrader

The experimental methods used for the Taber Linear Abraser are based off of

ASTM D 4060. The uncoated and coated glass slides were tested as a function of number

of cycles in increments of 10, going from 10-60 cycles using a calibrade wearaser. The

speed was kept constant at 25 cycles per minute with an operating stroke of 1 inch and a









constant load of 100 grams. The samples were secured to the workspace using a metal

clamp with cushioning in between to prevent sample breakage. A small brush was used to

prevent the accumulation of abraded particles during testing, the samples were

continually monitored. The sample weight loss was measured in between each cycle

increment.

The method for testing the coated interlayer samples was slightly different due to

their tendency to delaminate from the interlayer surface. For this experiment, a calibrase

wearaser was used to abrade the samples for 50 cycles at a stroke length of 1 inch. The

load was increased from 100g to 850g in increments of 250 grams using the additional

weight discs. If no delamination was observed at the highest load, the speed was

increased form 25 cycles/minute to 40 cycles/minute and then to 60cycles/minute. The

sample is considered successful if no delamination is observed up to this point.

4.3.3 Visual observation of Film Delamination

To obtain a better understanding of the delamination behavior of the

nanocomposite films from the interlayer substrate, a visual observation test was

conducted on coated interlayer samples. The samples were solution cast with films made

from dispersions in 1 wt%, 2wt%, and 3wt% PVA matrices. The films had final solids

loadings of 40 wt%, 50 wt%, and 60 wt% for each polymer concentration. The films were

cast onto interlayer substrates cut into 1" x 2" coupons and surface treated with a 2:1

sulfuric/nitric acid treatment. To visually observe the delamination of the films from the

interlayer substrate, the samples were bent over steel washers of different diameters. A

schematic of the washers are shown in Figure 4-1.















1.9 cm 1.59 cm 1.27 cm 0.95 cm

Figure 4-1: Steel rods of different diameters

Each sample was bent over the rods, starting from the largest diameter, and the

delamination behavior of the films was recorded.

4.4 Discussion of Results

4.4.1 Multi-Layered Films


The weight loss data for samples coated with one layer of nanocomposite film are

shown in Figure 4-2. The samples are compared to the weight loss of the glass substrate.

The sample containing 40 wt% Laponite in the dried film had a weight loss comparable

to that of the glass slide. The films containing higher solids loadings had a much larger

weight loss, almost triple that of the glass slide. For the coated samples, the clay platelets

will be randomly oriented in the film and connected by polymer.9 The films containing

higher solids loading would have less connecting polymer, creating a rougher surface.

This rough surface may be easier to abrade than the 40 wt% surface, which would have

more polymer connecting the particles. Another observation was that the standard

deviation of the film-coated samples were much larger than the standard deviation for the

glass slide.












0.016
C 0.014
0.012
,-J
S0.01

0.008 -
S0.006 -
0.004 -
> 0.002 -
0
40% 50% 60% Glass Slide
Solids Loading, wt%


Figure 4-2: Average weight loss data with standard deviation for abraded glass substrate and
substrate coated with 40, 50, and 60 wt% nanocomposite films


The weight loss results for the double-coated samples are shown in figure 4-3. In

this case, all of the samples performed better than or equal to the glass substrate at

resisting weight loss by surface abrasion. Contrary to the single-coating results, the

samples containing high solids loadings of Laponite such as 60/50 (60 wt% coated onto

50 wt%), and 50/60 lost the least amount of weight. Additionally, the samples which lost

the least amount of weight also have a standard deviation that is much smaller than the

samples that lost a larger amount of weight. The samples which have a 40 wt% on the

bottom had very large standard deviations, which may be the result of low surface

roughness creating a decreased interfacial adhesion to the second layer of film.











0.006

0.005 -

0.004 -
,-I

.L 0.003

0.002-

0.001


60/40 50/40 60/50 40/50 40/60 50/60 Gass Slide
Treatment, wt% Laponite



Figure 4-3: Average weight loss data with standard deviation for abraded glass substrate and
substrate coated with two layers of nanocomposite film (x/y: film x coated
over film y).

Figure 4-4 shows the weight loss results for the triple coated samples. Some of

these samples loss less weight than the glass substrate, but several did not perform as

well. The samples that had the 40 wt% Laponite film as the top layer (40/50/60 and

40/60/50) lost more weight than the glass and had large standard deviations. As with the

double coated samples, this could be due to a decrease in interfacial adhesion between the

film layers. The samples with the 60 wt% Laponite film as the top layer performed well

but also have high standard deviations. The error in this data is very large, making it

difficult to draw conclusions due to an inability to reproduce the results.











0.007

0.006

8 0.005 -

-j 0.004 -
'I-
S0.003 -

S0.002 -

0.001

0
50/60/40 60/50/40 60/40/50 40/60/50 50/40/60 40/50/60 Glass
Slide
Treatment, wt% Laponite



Figure 4-4: Average weight loss data with standard deviation for abraded glass substrate and
substrate coated with three layers of nanocomposite film (x/y/z: film x coated
over film y coated over film z)


A comparison of the films which loss less than or equal to the amount of weight

as the glass substrate are shown in Figure 4-5. All of these coatings are sufficient in

providing abrasion resistance that is better than or equal to that of a typical glass

specimen. Again it can be seen that the most successful coatings had the smallest

standard deviation. When compared to the triple-coated samples, the double-coated

samples with films of high solids loading not only lose less weight than the glass slide,

but the results are reproducible.










0.007
o'
0.006
0
-- 0.005

0.004

S0.003

S0.002

.: 0.001

0




Treatment, wt% Laponite


Figure 4-5: A comparison of average weight loss data with standard deviation for the films
which loss less than or equal to the amount of weight as the glass substrate

4.4.2 Delamination of Coated Interlayer Samples

Upon visual observation, the coated interlayer samples had good adhesion with

the nanocomposite film. When abraded by the calibrase wearaser at different loads and

speeds as described earlier, none of the films delaminated from the interlayer surface.

Although the Taber instrument was at its maximum capacity, the films were not rubbed

from the surface and there was very little haze on the transparent film. Other forms of

adhesion testing, such as dynamic mechanical analysis (DMA) or and Instron peel test

would provide more insight into the interfacial strength between the films and the treated

interlayer substrate.

Visual observation of delamination of the PVA/Laponite films from the

interlayer substrate was achieved with the steel rod test. All of the film samples prepared









in a 1 wt% PVA matrix did not delaminate when bent over the rods. Even the smallest

diameters did not succeed in changing the appearance of the film or adhesion to the

substrate. The films prepared in 2 wt% PVA dispersions did not fracture from the

interlayer surface, but they did delaminate. The film with the lowest solids loading (40

wt%) was not effected by the larger diameters, but at 1.27cm the film/interlayer interface

turned white, suggesting the film was peeling from the surface. The 50 and 60 wt% films

had this same behavior on the 1.9cm rod. The films cast in a 3 wt% PVA matrix fractured

on the largest diameter rod. When bent, they fractured along the middle of the sample and

continued to peel back if bent further. Larger rods would be needed to define the point

where fracture occurs for these films.














CHAPTER 5
CONCLUSIONS AND FUTURE WORK

5.1 Conclusions

When Laponite JS is dispersed in water, it can form a stable sol which exhibits

Newtonian behavior. When polymer is present, the interactions between the polymer

chains and the particles cause a change in the rheological properties of the dispersions.

As polymer dosage is increased in a Laponite dispersion, particle flocculation causes a

weak colloidal gel to form and the viscosity increases. Shear thinning behavior can be

seen for these colloidal gels at high shear rates, noting a breakdown and particle

alignment in the weakly flocculated network structure. SEM analysis shows that the flocs

present in the dispersions are open, ramified structures that are evenly dispersed

throughout the dispersions. The flocs are not interconnected, which makes it easier for

structural breakdown to occur at high shear rates and high angular frequencies.

The processing conditions for films made from 2 wt% and 3 wt% PVA solutions

include Newtonian-like behavior at low shear rates and shear thinning behavior at high

shear rates. For the 2 wt% PVA dispersions, the weakly flocculated gel structure can be

broken down at high frequencies; this does not occur for dispersions made in a 3 wt%

PVA matrix. For low shear processing techniques like solution casting, a 2 wt% PVA

matrix is ideal; the 3 wt% PVA matrix would work for high shear techniques such as tape

casting and spin coating.

The processing conditions of the colloidal gels are determined as a function of

both solids loading and polymer concentration. An increase in solids loading at fixed









PVA concentration causes a shift in zero shear viscosity and an increase in the shear rate

dependent region of the steady-shear viscosity curves. The effect of the suspending fluid

is crucial in the processing of the coatings due to the large increase in viscosity with

increasing PVA concentration.

Thin, transparent films can be cast from the PVA/Laponite dispersions. These

films exhibit good adhesion to glass slides, but cannot be cast onto a hydrophobic

polymer interlayer. To activate the surface of the interlayer, and reduce the surface

tension, several surface treatments were used. Using FTIR-ATR, it can be observed that

several of these treatments can successfully activate the interlayer surface. A 100%

sulfuric acid treatment will activate the surface, but will create substantial yellowing. A

mixture of sulfuric and nitric acid in a ratio of 2:1 will also activate the surface but will

not cause discoloration. This treatment showed a substantial improvement in adhesion

between the cast films and the interlayer substrate. Silane coupling agents will also

successfully modify the interlayer surface, but it was observed that the interfacial

adhesion to the nanocomposite films was not improved.

To determine the abrasion resistance of cast nanocomposite films on glass slides,

a Taber Linear Abrader was used determine weight loss as a function of number of

cycles. Coatings with one, two, and three layers of film at different solids loading were

tested for weight loss against an uncoated glass slide specimen. The single coated

samples did not perform as well as the glass slide, i.e. they lost more weight than the

glass. The double coated samples performed better than the glass slide at every solids

loading, with the highest solids loadings (50/60 and 60/50) losing the most weight. With

the triple coated samples, only some lost less weight than the glass slide. In conclusion,









the optimal coating for the glass to minimize weight loss and maximize efficiency is a

double-layered film of high solids loading.

5.2 Future Work

The field of polymer-layered silicate nanocomposites is very large and has a

broad range of applications. Further study of the PVA/Laponite JS system and systems

with similar properties is useful to assess their use in the films and coatings industry. For

use as an abrasion resistant film, more mechanical and thermal analysis of the material

must be performed. Adhesion is a large concern for this application and maximizing this

property would be useful. The following are suggestions for possible future work.

To further determine the abrasion resistant properties of the films, mechanical
studies can be conducted on interlayer samples coated with various PVA/Laponite
films. These studies would include Instron tensile testing and dynamic mechanical
analysis. Since tensile strength is proportional to abrasion resistance, an optimal
formula can be determined.
To continue theological studies on the PVA/Laponite system, an in-depth study
on the structural development of the dispersions can be conducted.
Perform surface activation of the polymer interlayer surface using RF plasma
treatment. This treatment is more expensive than wet-chemical treatments and
requires special equipment, but may be more effective in creating an activated
polymer surface.
When used as a laminated glass, the SentryGlas ionomer is heat treated and
becomes a harder, more glass-like material. This glassy polymer is adhered to
window glass to improve the properties of the window. Analysis can be done on
the surface of heat treated interlayer using contact angle measurements and FTIR-
ATR.
UV-VIS experiments can be performed on glass, film-coated glass, and abraded
materials to determine the transparency of the films compared to regular window
glass. The effect of abrasion on the transparency of the glass can also be
determined using this technique.
Using different materials for both the polymer matrix and the nanosilicate can be
used to continue this investigation. Some possible polymers would be Poly
Methyl Methacrylate (PMMA), Poly Ethylene Oxide (PEO), Polycarbonate (PC),
and polyvinyl acetate (PVA). Also, different grades of Laponite can be used as
the nanosilicate filler.















LIST OF REFERENCES


1. A.R.Vaia.; R. Krishnamoorti, 2001. Polymer nanocomposites: introduction. ACS
Symposium Series 804. Materials and Manufacturing Directorate, Air Force Research
Library, 2001.

2. J. Collister, 2001. Commercialization of polymer nanocomposites. ACS Symposium
Series 804. Edison Polymer Innovation Corporation.

3. S. Bandyopadhyay; A. J. Hseih. E. P. Giannells. 2001. PMMA Nanocomposites
synthesized by emulsion polymerization. ACS Symposium Series 804.

4. B. Wetzel; F. Hauper; K. Friedrich; M. Zhang; M. Rong. Impact and wear resistance
of polymer nanocomposites at low filler content. Polymer Engineering and Science,
September 2002.

5. M. Alexandre, P. Dubois. Polymer-layered silicate nanocomposites: preparation,
properties, and uses of a new class of materials. Materials Science and Engineering
28 (2000) 1-63.

6. J. Zebrowski, V. Prasad, W. Zhang, L.M. Walker, D.A. Weitz. Shake-gels: shear-
induced gelation of Laponite-PEO mixtures. Colloids and Surfaces A: Physicochem.
Eng. Aspects 213 (2003) 189.

7. S. Lin-Gibson, H. Kim, G. Schmidt, C.C. Han, E.K. Hobbie, J. Shear-induced
structure in polymer-clay nanocomposite solutions. Colloid Interface Sci. 274 (2004)
515.

8. D. C. Pozzo, L.M. Walker. Reversible shear gelation of polymer-clay dispersions.
Colloids and Surfaces 240 (2004) 187-198.

9. V. Ferreiro, G. Schmidt, C. Han, A. Karim. Dispersion and nucleating effects of clay
fillers in nanocomposite polymer films. Polymer Nanocomposites: Synthesis,
Characterization, and Modeling (2002) 177.

10. A. Nelson, T. Cosgrove. A small-angle neutron scattering study of adsorbed poly
(ethylene oxide) on Laponite. Langmuir 20 (2003) 2298.

11. G. Schmidt, M. Malwitz. Properties of polymer-nanoparticle composites. Curr. Opin.
Colloid Interface Sci. 8 (2003) 103.









12. Ronald G. Larson. The structure and rheology of complex fluids. Oxford University
Press, New York, 1999.

13. Rockwood Additives Ltd., Cheshire, United Kingdom. Laponite Technical Bulletin
L204/01 g

14. R. Krishnamoorti, K. Yurekli. Rheology of polymer layered silicate nanocomposites.
Curr. Opin. Colloid Interface Sci. 6 (2001) 464.

15. J. Labanda, J. Llorens. Rheology of Laponite colloidal dispersions modified by
sodium polyacrylates. Colloids and Surfaces A: Physiochem. Eng. Aspects 249
(2004) 127-129.

16. M. Malwitz, A. Dundigalla, V. Ferreiro, P. Butler, M.C. Henk, G. Schmidt. Layered
structures of shear-oriented and multilayered PEO/silicate nanocomposite films. Phys.
Chem. 6 (2004) 2977.

17. J. Ren, B. Casanueva, C.A. Mitchell, R. Krishnamoorti. Disorientation kinetics of
aligned polymer layered silicate nanocomposites. Macromolecules 36 (2003) 4188.

18. Z. Shen, J. Chen, H. Zou, J. Yun. Rheology of colloidal nanosized BaTiO3
suspension with ammonium salt of polyacrylic acid as a dispersant. Colloids and
Surfaces 244 (2004) 61-66.

19. M. Sjoberg, L. Bergstrom, A. Larsson, E. Sjostrom. The effect of polymer and
surfactant adsorption on the colloidal stability and rheology of kaolin dispersions.
Colloids and Surfaces 159 (1999) 197-208.

20. J. A. Lee, M. Kontopoulou, J. S. Parent. Time and shear dependent rheology of
maleated polyethylene and its nanocomposites. Polymer (2004) 1-6.

21. S. Granick, H. Hu, G.A. Carson. Nanorheology of confined polymer melts. 2.
nonlinear shear response at strongly adsorbing surfaces. Langmuir 10 (1994) 3867.

22. A. Mukhopadhyay, S. Granick. Micro- and nanorheology. Curr. Opin. Colloid
Interface Sci. 6 (2001) 423.

23. J. Bergenholtz. Theory of rheology of colloidal dispersions. Curr. Opin. Colloid
Interface Sci. 6 (2001) 484

24. P. Mongondry, T. Nicolai, J. Tassin. Influence of pyrophosphate or polyethylene
oxide on the aggregation and gelation of aqueous laponite dispersions. J. Colloid
Interface Sci. 275 (2004) 191.









25. S. Carra, A. Sliepcevich, A. Canevarolo, S. Carra. Grafting and adsorption of poly
(vinyl) alcohol in vinyl acetate emulsion polymerization. Polymer 46 (2005) 1379-
1384.


26. A. A. Zaman, N. Delorme. Effect of polymer bridging on theological properties of
dispersions of charged silica particles in the presence of low-molecular-weight
physically absorbed poly (ethylene oxide). Rheol Acta 41 (2002) 408-417.

27. A. A. Zaman, M. Bjelopavlic, B.M. Moudgil. Effect of adsorbed polyethylene oxide
on the rheology of colloidal silica suspensions. J. Colloid Interface Sci. 226 (2000)
290-298.

28. R. Chanamai, D. J. McClements. Creaming stability of flocculated monodisperse oil-
in-water emulsions. J. Colloid Interface Sci. 255 (2000) 214-218

29. H. Anderlean, S. Petit, P. Laurens, P. Marcus, F. Arefi-Khonsari. Effects of different
laser and plasma treatments on the interface and adherence between evaporated
aluminum and polyethylene terepthalate films: X-ray photoemission and adhesion
studies. Applied Surface Science 243 (2005) 304-318.

30. Jasco, Inc., Easton, MD. IR Application Note 02-03. Qualitative analysis of powdered
solids with FTIR-ATR.

31. A.K. Pal, R.K. Roy, S.K. Randal, S. Gupta, B. Deb. Electrodeposited carbon
nanotube thin films. Thin Solid Films 476 (2005) 288-294

32. Y.H. Wang, R. Kumar, X. Zhou, J.S. Pan, J.W. Chai. Effect of oxygen plasma
treatment on low dielectric constant carbon-doped silicon oxide thin films. Thin Solid
Films 473 (2005) 132-136.

33. J.E. Klemberg-Sapieha, L. Martinu, N.L.S Yamasaki, C.W. Lantman. Tailoring the
adhesion of optical films on polymethyl-methacrylate by plasma-induced surface
stabilization. Thin Solid Films 476 (2005) 101-107

34. D. H. Jeong, U. Erb, K. T. Aust, G. Palumbo. The relationship between hardness and
abrasive wear resistance of electrodeposited nanococrystalline Ni-P coatings. Scripta
Materiala 48 (2003) 1067-1072

35. S. Suzuki, E. Ando. Abrasion of thin films deposited onto glass by the Taber test.
Thin Solid Films 340 (1999) 194-200

36. L. Dulany, 2002. Scratch- and abrasion-resistant coatings with energy-curable resin
samples. UCB Chemicals Corp., Smyrna, GA.















BIOGRAPHICAL SKETCH

The author was born in Havertown, Pennsylvania, a small suburb outside of

Philadelphia. She obtained her high school diploma from Cypress Creek High School in

Orlando, Florida. She went on to get her bachelor's degree in materials science and

engineering at the University of Florida in Gainesville, where she was active in the

Benton Engineering Council and the Materials Research Society. She continued her

education at the University of Florida to pursue a Master of Science in materials science

and engineering. She worked with Dr. Abbas Zaman during undergraduate and graduate

studies on colloidal systems and polymer thin films.