Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Process engineering and characterization of composite electrocoatings Chen, Michael Nai-Chia 2004

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

Item Metadata

Download

Media
831-ubc_2004-0401.pdf [ 15.04MB ]
Metadata
JSON: 831-1.0079064.json
JSON-LD: 831-1.0079064-ld.json
RDF/XML (Pretty): 831-1.0079064-rdf.xml
RDF/JSON: 831-1.0079064-rdf.json
Turtle: 831-1.0079064-turtle.txt
N-Triples: 831-1.0079064-rdf-ntriples.txt
Original Record: 831-1.0079064-source.json
Full Text
831-1.0079064-fulltext.txt
Citation
831-1.0079064.ris

Full Text

Process Engineering and Characterization of Composite Electrocoatings  by  Michael Nai-Chia Chen B. A. Sc., Department of Metals and Materials Engineering, The University of British Columbia, Vancouver, Canada, 2002  A THESIS S U B M I T T E D IN PARTIAL FULFILLMENT O F T H E R E Q U I R E M E N T S FOR THE D E G R E E OF MASTER OF APPLIED SCIENCE  THE FACULTY OF GRADUATE STUDIES D E P A R T M E N T OF METALS AND MATERIALS ENGINEERING  W E A C C E P T THIS THESIS A S C O N F O R M I N G TO THE REQUIRED STANDARDS  The University of British Columbia August 2004 © Michael Nai-Chia Chen, 2004  UbcI  w  THE UNIVERSITY OF BRITISH C O L U M B I A  F A C U L T Y OF G R A D U A T E S T U D I E S  Library Authorization  In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  Michael Nai Chia Chen Name of Author  11/08/2004  (please print)  Date (dd/mm/yyyy)  Title of Thesis:  process Engineering and Characterization of Composite Electrocoatings  Degree:  Master of Applied Science  Department of  Metals and Materials Engineering  Year:  2004  The University of British Columbia Vancouver, BC Canada  grad.ubc.ca/forms/?formlD=THS  page 1 of 1  last updated: 11-Aug-04  Abstract Electrocoating (E-coat) has conventionally been used primarily for corrosion protection purposes as a barrier coating beneath the primer of the automobile body. For E-coat to be used as a long term solution for outdoor equipment barrier protection, mechanical properties such as abrasion/wear, scratch and hardness must be improved. In this work we assess the possibility of such improvement of E-coatings by combining them with ceramic particles. The effects of ceramic filler particle size and concentration, along with deposition parameters and processing procedures on the properties of epoxy cathodic barrier coating have been investigated.  The deposition voltage was varied  from 100 to 300 volts with various filler (a-AI 0 ) concentrations of 0 - 30 vol% and 2  3  particle size of 0.3 - 5 |j.m. The effect of the filler was also tested in an electrochemical cell in order to foresee corrosion resistance variations due to different concentrations of filler in the coating. Other possible applications were also investigated to utilize the composite E-coat processing to create a functional coating with transducer like effects. The ceramic filler content was observed to improve mechanical properties such as hardness, scratch resistance and wear. Mechanical properties showed a decline at higher concentration of filler (> 10 vol%) due to lower resin content, which increased filler-to-filler interactions. The thickness of the deposited films increased with increased filler concentration and deposition voltage, with filler concentration having a greater effect on the deposited film thickness. Large filler particle size (3-5 urn) was observed to deteriorate the film's mechanical properties when compared to a smaller filler particle (0.3-0.5 urn). The filler-film interface showed homogeneous bonding and encapsulation of the filler particles, resulting in excellent adhesion of 68-82 M P a along with improved corrosion resistance with increased filler content in the film up to 10 vol% filler. A ii  functional composite film was also deposited by adding piezoelectric ceramic powders within E-coatings. A hysteresis loop was observed during testing showing poling during deposition. Resonance and anti-resonance frequencies were found to be 3.90x10 Hz 7  and 3.80x10 Hz respectively. This demonstrates that the composite E-coat process is a 7  viable method of manufacturing low cost and large area coverage thin film sensors and actuators.  iii  Table of Contents Chapter 1 . Introduction  1  Chapter 2 . Literature Review and Research Objectives  3  2.1 E-coating Process  3  2.1.1 Anodic and Cathodic E-coating Systems  4  2.1.2 Advantages of E-coating  6  2.1.3 E-coating Process Phenomena  8  2.1.4 Film Thickness  9  2.1.5 Heat evolution during E-coating  11  2.1.6 E-coat Paint and Material Selection  12  2.1.7 Substrate Selection and Surface Preparation for E-coating  13  2.1.9 Corrosion Protection by E-coatings  18  2.1.10 Wear of polymer and polymer composites  21  2.2 Piezoelectricity  27  2.3 Piezoelectric Ceramic-Polymer Composite  30  2.4 Objectives  32  Chapter 3 . Experimental Procedure  33  3.1 Materials  33  3.2 Sample Preparation  35  3.3 Characterization the Coatings  37  3.3.1 Microstructure and Thickness Observations  37  3.3.2 Mechanical Properties  38  3.3.3 Dielectric Strength Test  45  3.3.4 Pontentiodynamic Test  46 iv  3.3.5 Piezoelectric Effects  48  Chapter 4 Results and Discussion  52  4.1 Ceramic Filler Powders Characteristics  52  4.2 E-coating Structures  56  4.3 Voltage Dependence of Coating Thickness  62  4.4 Influence of Alumina in Paint Bath on Deposition Process  64  4.4.1 Conductivity Changes  64  4.4.2 Electrical Resistance  66  4.4.3 Thickness of Coatings  69  4.5 Mechanical Properties of E-coatings  71  4.5.1 Adhesion  71  4.5.2 Scratch Resistance  72  4.5.3 Hardness  78  4.5.4 Abrasion Test  79  4.5.5 Filler Effects  85  4.6 Process Modification and Optimization  89  4.6.1 Alumina in Composite E-coating  89  4.7 Alumina Particle Size Effects  98  4.8 Electrochemical Behaviour of Composite E-coats  105  Chapter 5 Summary and Conclusions  108  Chapter 6 Recommended Future Work  110  Chapter 7 References  111  APPENDIX I - Composite E-Coating Thickness  116  APPENDIX II- Composite E-Coating Corrosion Protection Efficiency  118  APPENDIX III- A New Direction: Piezoelectric E-Coatings  120  v  List of Figures  Figure 2-1. Schematic view of overall E-coat process  4  Figure 2-2. Anodic E-coating system, where negative particles are suspended  5  Figure 2-3. Schematic of cathodic E-coat cell during deposition  6  Figure 2-4. Film thickness with time of deposited coatings using constant voltages [26]. 10 Figure 2-5. Heat evolution during E-coating for different surfaces: 1, bare steel; 2, phosphated steel; 3, zinc-plated steel; 4, phosphated zinc-plated steel. U= 300V [12]  11  Figure 2-6. Schematic of typical automotive paint layers with approximate thickness for each layer [15]  14  Figure 2-7. A model of depositon of particles of differently prepared surfaces; (a)nontreated surface, (b) phosphated surface [12]  15  Figure 2-8. Photomicrograph of first layer (wet) E-coated onto the substrate: (a) bare steel; (b) phosphated steel [12]  17  Figure 2-9. Impedance model for a metal coated with organic polymer [32]  20  Figure 2-10. Log of pores resistance (f? ) versus time plot for films deposited at various p  voltages [33]  21  Figure 2-11. Sliding wear test methods; (a) ring-on-ring;(b) face-to-face;(c)pin on flat face;(d) pin on rim;(e)block against ring;(f) pin on flat [40]  22  Figure 2-12. Geometrical factor of surface (a) conformal and (b) counterconformal contacts [40]  23 vi  Figure 2-13. Schematic of different types of hard particle abrasion [40]  24  Figure 2-14. Particle hardness effects on wear volume when comparing hardness H  a  (particle) against H (surface) [43]  25  s  Figure 2-15. S E M of silica particles for rounded (a) and angular (b) shapes [40]  26  Figure 2-16. Wear rate of different types of abrasion and particle sizes of S i C [44]  27  Figure 2-17. Typical crystallized perovskite structure with oxygen atoms at the corner, lead in the centre of each face and a titanium or zirconium molecule at the centre of the cube structure  28  Figure 2-18. Types of connectivity and phases present for composite piezo-ceramics [56]  31  Figure 3-1. Flow chart of the procedure for making alumina suspension and composite bath for deposition  34  Figure 3-2. Schematic of E-coat cell process setup  35  Figure 3-3. Schematics of stud pull adhesion test platform (a) and sample preparation technique (b) [65] Figure 3-4.  39  Schematic of stud placement and failure locations; (a) coating/substrate  interface, (b) cohesive, (a) + (b) mix and (c) pull adhesive  40  Figure 3-5. Schematic of stylometer scratch test platform [65]  41  Figure 3-6. Chipping of coating during scratch testing  42  Figure 3-7. Screen capture of results for scratch testing with failure noted by the acoustic burst (circle); (a) normal force, (b) normal friction, (c) transverse friction and (d) acoustic  43  Figure 3-8. Schematics of Microscale Abrasion Tester [34]  44  Figure 3-9. Load displacement curve relating force to stiffness [65]  45  Figure 3-10. Schematic of breakdown voltage test setup during testing  46  vii  Figure 3-11. Schematic of electrochemical cell for potentiodynamic testing [68]  48  Figure 3-12. Picture of setup for impedance measurements  49  Figure 3-13. Schematics of sample holder jig for two-side coated samples  50  Figure 3-14. Picture of Sensor Tech S S 0 5 polarization meter  51  Figure 4-1. A14 as received AI2O3 powder under S E M  53  Figure 4-2. A16 as received A l 0 powder under S E M  53  Figure 4-3. BM532 as received P Z T powder under S E M  54  2  Figure  4-4.  A16  alumina  3  particle  distribution  after  10  minutes  ultrasonic  deagglomeration Figure  4-5.  A14  55 alumina  particle  distribution  after  10  minutes  ultrasonic  deagglomeration  55  Figure 4-6. BM532 P Z T powder particle size distribution after 10 minutes ultrasonic deagglomeration  56  Figure 4-7. E-coated and cured samples with variable deposition voltages and concentrations of A16 alumina in coating  58  Figure 4-8. S E M image of A16 alumina (12.5 vol%) reinforced composite coating  59  Figure 4-9. P Z T (12.5 vol%) E-coat composite cross-section S E M image  59  Figure 4-10. High magnification of PZT composite E-coat cross-section  60  Figure 4-11. High voltage (300V) and high alumina (30 vol%) in coating cross-section SEM  60  Figure 4-12. Surface morphology of high alumina (30 vol%) and high voltage (300V) deposited coating  61  Figure 4-13. Cross-section S E M image of A14 particle reinforced coating deposited at 200 volts  62  viii  Figure 4-14. Voltage dependence of thickness for the paint during deposition at constant voltage  63  Figure 4-15. Conductivity measurements of composite bath with increased A16 alumina in suspension  65  Figure 4-16. pH measurement with increasing A16 alumina in suspension  66  Figure 4-17. Schematic of test points chosen for dielectric measurements for coated samples of various concentrations of alumina and deposition voltages Figure 4-18. Dielectric strength with increasing A16 alumina in coating  68 68  Figure 4-19. Thickness changes with increase in voltage and concentration of A16 alumina in coating  70  Figure 4-20. Percentage increase of thickness with increased A16 alumina in coatings deposited at a constant voltage of 200 V  70  Figure 4-21. Adhesion strength of tensile stud pull test of coatings deposited at 200V 72 Figure 4-22. Typical pencil hardness test for 150V deposited coating with increasing A16 alumina content  73  Figure 4-23. Scratch test results performed by stylomer for various deposition voltages and A16 alumina concentration in coating  74  Figure 4-24. Scratch test results for 'as received', 4.2 vol% and 10 vol% A16 alumina in coatings at various deposition voltages  75  Figure 4-25. Magnified 10x Stylometer results for (a) 0 vol% A16 alumina in coating and (b) 10.0 vol% A16 alumina in coating  76  Figure 4-26. Scratch test display resulting from a non reinforced film with acoustic burst (circle); (a) normal force, (b) normal friction, (c) transverse friction and (d) acoustic. 77 ix  Figure 4-27. Scratch test display resulting from a 10.0 vol% A16 alumina reinforced film with acoustic burst (circle); (a) normal force, (b) normal friction, (c) transverse friction and (d) acoustic  77  Figure 4-28. Hardness for coatings deposited at 200 volts with varying A16 alumina concentration in coating  78  Figure 4-29. Modulus of elasticity for coatings deposited at 200V with varying concentrations of A16 alumina in coatings  79  Figure 4-30. Wear rate of 12.5 vol% A16 alumina samples where coating was not penetrated  81  Figure 4-31. Profilometry wear bead shape results for 150 volts and 12.5 vol% A16 alumina coating  82  Figure 4-32. Profilometry wear bead shape results for 150 volts and 0 vol% A16 alumina coating  83  Figure 4-33. Profilometry wear bead shape results for 150 volts and 22.4 vol% A16 alumina coating  84  Figure 4-34. Elastic modulus prediction of the composite compared to actual data measured Figure 4-35. Interface S E M of E-coat deposited on A16 alumina layer  86 88  Figure 4-36. High magnification S E M image of interface between E-coat and A16 alumina layer  88  Figure 4-37. Comparison of volume percent concentration of various sample forms. ...90 Figure 4-38. Residue changes for 2.5 cm from the surface of the solution at 4.5 vol% A16 alumina added to paint bath  90  Figure 4-39. Residue changes for 2.5 cm from the bottom of the solution at 4.5 vol% alumina added to paint bath  91  Figure 4-40. Viscosity changes with increase volume percent A16 alumina in solution.... 91 Figure 4-41. Conductivity changes for as received paint bath after each subsequent deposition at 200V  93  Figure 4-42. Current Density of deposition versus time with increasing A16 alumina concentration in composite E-coatings  94  Figure 4-43. Alumina A16 concentration influence on induction time during deposition of film  95  Figure 4-44. Calculated film thickness incorporating influences contributed by A16 alumina compared to actual measured thickness  96  Figure 4-45. Influence of alumina A16 on the deposition efficiency  96  Figure 4-46. Deposition yield with increase alumina A16 in coating  97  Figure 4-47. Influence of alumina A16 on the current density applied during deposition. 97 Figure 4-48. Scratch test comparison of A16 and A14 alumina in coating deposited at various voltages  98  Figure 4-49. Induction time difference between A16 and A14 alumina at various deposition voltages  99  Figure 4-50. Thickness variation when using A14 compared to A16 at various deposition voltages  100  Figure 4-51. Deposition yield comparison between A16 and A14 reinforced coatings 101 Figure 4-52. Comparison of the current density of deposited A16 and A14 coatings 102  xi  Figure 4-53. Coulombs consumed per area covered using A16 and A14 particles at varying deposition voltages  103  Figure 4-54. Deposition efficiency of the overall process for A16 and A14 reinforced coatings  103  Figure 4-55. Breakdown voltage comparison between A14 and A16 coated samples. 104 Figure 4-56. Dielectric Strength differences between A16 and A 1 4 coated samples. .105 Figure 4-57. Electrochemical behaviour of 5 vol% and 10 vol% A16 alumina filler composite coating compared with to 'as received' (0 vol%) coating  xii  106  List of Tables Table 2-1. Paint bath mixing ingredients for resin solution  13  Table 2-2. Cleaner Selection for E-coating process [17]  18  Table 2-3. A list of commonly used ions for substitution in P Z T [50,51].  30  Table 3-1. Composition of "as received" standard stock solution of cathodic epoxy bath. 33 Table 3-2. Bath details of each coating deposition with the alumina suspension (50 wt%) added to constant amount of 150 ml of stock paint Table 3-3. Conditions for potentiodynamic measurements of coated samples  36 47  Table 4-1. Thickness and wear penetration depth of the coatings for various deposition voltages and A16 alumina concentrations  xiii  80  Acknowledgment The author wishes to express his gratitude to his supervisor, Dr. Tom Troczynski for this advice, patience, and encouragement during this project. Thanks are also extended to the faculty, staff, and fellow graduate students in the Department of Metals and Materials Engineering. The assistance of Carmen Oprea, Dr. Eswar Prasad of SensorTech Inc., Collingwood, Ontario, Research  Council, Vancouver,  British  Dr. Sing Yick of  Columbia, A U T 0 2 1  National  N C E Network  of  Excellence, S N Coatings and the Natural Sciences and Engineering Research Council of Canada, is gratefully acknowledged.  xiv  Chapter 1. Introduction A B M W 3-series car is comprised of approximately 7 0 % metallic materials (60% Iron, 8% cast/forged light alloys and 3% non-ferrous)[1]. The necessity for protection against the environment (acid rain, sand and dust, etc) and human introduced corrosion agents such as salt, has led to the emergence of many coating technologies in last 2030 years. By using coatings - specifically painted coatings - the automotive industry has consistently protected its metallic components, using an automated process called electrocoating (E-coat) which has become the norm in industry. The  E-coating  normally  consists of a mixture  formulated in order to suit a specific application.  of polymers,  differentially  The coating is able to provide  protection against salt spray, corrosion, UV contamination and other effects of outdoors exposure [2]. However, most automobiles coatings consist of multiple layered coating, with E-coat being the second layer from the metallic substrate. Each coating would have its own specific use and design, with E-coat being there for the sole purpose of corrosion protection barrier, while the clearcoat (outermost layer) confers mechanical protection.  However, the E-coat method has seen increased use in non-automotive  related fields such as appliances, outdoor equipment and various other metallic artifacts. Therefore, there is a needed to improve the E-coat in order to decrease its susceptibility to erosion and impact. To resolve these problems, it has been proposed in this thesis that the addition of a ceramic reinforced phase be introduced into the polymer matrix, creating a polymer-ceramic composite. The material of choice was 99.8% a-alumina (AI2O3). Alumina was chosen over other ceramics due to its inertness, relative abundance and low cost processing through the Bayer process. Also, the fact that alumina has good mechanical properties such as Vicker's hardness ~1600, 1  Young's modulus ~380 G P a , compressive strength - 3 0 0 0 M P a , and tensile strength ~150-500MPa [3] played an important role in selecting it as the reinforcement phase.  2  Chapter 2. Literature Review and Research Objectives 2.1 E-coating Process Electrocoating, also known as "E-coat", "elpo" and "electrodeposition", is a fully automated coating method based on the principles of plating. It is an electrochemical process by which organic paint solids are deposited on the bare metal [2]. The polymer coatings provide corrosion protection by acting as a barrier layer between the substrate material and the environment [4-6]. The electrocoat system applies a DC charge to a metal part immersed in a bath of oppositely charged paint particles dispersed in a water bath. The paint particles are drawn to the metal part and paint is deposited on the part, forming an even, continuous film over the entire surface, until the coating reaches the desired thickness. At that thickness (typically 20 - 50 ^m), the film insulates the part, the attraction stops and the process is complete. Depending on the polarity of the charge, electrocoating is classified as either anodic or cathodic. A flow chart of the E-coat process can be seen on Figure 2-1; the substrate undergoes several stages of preparation, which include surface cleaning using various alkaline solution, followed by surface preparation (roughening or conversion coating) and then its coated using E-coat. Once the coating is deposited, it follows a curing stage and it is inspected for defects and removal and rework are performed if visual defects are present.  3  Parts to be coated  Cleaning  Paint and Material Selection;  kq-  "3Z Coating Preparation  Surface Preparation  •3  E-Coat Bath and; Rinse  E-Coat Cure  Inspection  Finished; Product  Figure 2-1. Schematic view of overall E-coat process  2.1.1 Anodic and Cathodic E-coating Systems In an anodic electrocoating system (Figure 2-2), the part to be coated is the anode (positively electrical charge); this attracts the negatively charged particles in the paint bath. However, due to the positive charge on the metal substrate, some metallic ions are simultaneously leached into the paint solution: M° - • M  n +  + ne  _  (2-1)  The metal dissolution degrades the performance properties of the anodic system, since the corrosion of the metal may create other negative effects to the metal surface and alter the properties of the coating. The main reason for anodic coatings is that it is  known to offer economical systems that yield excellent color and gloss control coatings [8]. Hydrogen (cathode) and oxygen (anode) are produced in the solution due to the decomposition of water when a voltage is passed through the solution.  ( ) +  pi 11 n  <  Coating Film <  (-)  -  J  Figure 2-2. Anodic E-coating system, where negative particles are suspended.  In a cathodic electrodeposition system, the sample to be coated is negatively charged; this in turn attracts the positively charged particles in the paint bath. A key advantage to using a cathodic system (Figure 2-3) as opposed to an anodic system, is that the reversed polarity from the anodic systems significantly reduces the amount of metal dissolution entering the paint bath; therefore, it reduces the possibility of metal mixing within the cured coating. Cathodic coatings are considered high-performance coatings with excellent corrosion resistance that can be formulated for exterior durability [8]-  5  pi II; n +  Coating Film  (")  •. + ;•:  v  +  Figure 2-3. Schematic of cathodic E-coat cell during deposition.  2.1.2 Advantages of E-coating There are several advantages of using electrocoating over other coating methods such as dip-coating or spray-coating. Electrocoated objects are highly resistant to a corrosive environment. Also, another key property that makes E-coating so effective, is that the paint is able to be deposited on intricate or hard to reach areas, allowing the entire metallic part to be painted. This is due to the to fact that the entire part to be coated is electrically charged evenly throughout, and the particles are able to be deposited freely until the insulation due to the build up of thickness prevents further attraction. E-coating can be applied to intricate enclosed shapes, ranging from a cylinder-like shape to anything more complex. However, the condition required is that  6  the part to be coated must be fully immersed into the paint bath and a constant stir must be present in the bath to allow even distribution of the particles. Another advantage is the efficiency of E-coating. Transfer efficiencies of better than 95% result in reduced paint waste especially when compared with spray applied coatings [2]. The concentrations of solvents are also at a minimum, depending on mixture according to the individual recipe; some are known to use as little as 1-3% volatile organic solvent [2], so these formulation are environmentally friendly, nonflammable and of low viscosity. E-coat process can be easily automated if needed for a large-scale project. There is minimal handling throughout the processing and contact usually only occurs after the paint particles have been air-dried bonded onto the part surface. By then the paint is rigid enough to be handled. During the baking (190-300°C), the coating does not sag, nor will hot vapors cause the paint to wash off from the recessed areas. From a cost perspective, the paint costs are constant (~$15/gallon) and the paint thickness is easily reproducible from job to job due to physical simplicity of the process (in other coating processes, such as powder coat, wet spray, etc. the finished product can vary according to the operator's technique). This in turn reduces overall labor costs, as only one operator is needed for the equipment and very little if any rework is needed. However, there are some key disadvantages of the E-coat process. The paint's operating temperature is relatively low (<300°C). Limited production levels of multiple colors make it a less aesthetically pleasing process. There would be a requirement of multiple tanks for multiple colors. Initial costs capital is much higher than for other types of coating system; therefore, a large production is necessary to outlay the costs. A s mentioned previously, the process allows for complete coverage of the part to be painted, however, if one desires certain areas not to be covered, masking specific areas  to be left uncoated can be costly and time consuming during production [13].  Also,  there is a limit in the mechanical performance of an organic polymer. More specifically, the low hardness and wear resistance values make it unsuitable for certain applications. Another constraint in the E-coat process is the limited thickness that can be produced. Since the coating thickness depends on the applied potential and resistance of the part, once the first coat is applied, the insulating properties of the paint will not allow for further paint to be applied on top. Therefore, another method or coating process will be required to increase the thickness. Understanding the chemistry and physics behind the process will allow the process to be maximized.  2.1.3 E-coating Process Phenomena The  electrodeposition  of  water-dispersed  organic  coatings  is  a  complex  electrochemical process that includes several aspects [7]: 1. the evolution of hydrogen by H 2 O decomposition on a cathode 2 H 0 + 2e" ->H + 2 0 H " 2  (2-2)  2  2. electrophoretic migration of the resin micelles and associated pigment towards the cathode; 3. electrocoagulation of the resin micelles at the cathode surface by neutralization of positively charged groups in the resin with electrochemically generated OH" ions R—NH  + 3  + OH" -> R — N H + H 0 2  2  (2-3)  4. electroosmosis of water in the pores of the deposited films, caused by the potential difference across the film and surface charges on the pore walls; 5. heterogulation, that is the film deposited on the metal substrate; and 6. adagulation, that is further film deposition upon already-deposited layers.  8  2.1.4 Film Thickness For coagulation of the epoxy to occur, there must be suffient concentration of OH" ions at cathodic surface. The minimum concentration or critical concentration of OH" is determined by the specific quantity of electricity that must be passed, for a specific time (induction time t\) at a specific current density for the electrolysis of water to satisfy the minimum requirements [7]. For a static electrolyte, induction time can be calculated by using Sand's equation [8]. ft  =  (COH-)  2  F /DoH-n4f 2  (2-4)  where C o h - is the critical OH" ion concentration, D o h - the diffusion coefficient for hydroxyl ions, F is Faraday's constant and j the current density. E q . (2-4), having mostly constants and measurable variables, can be written in a shorter form t,f = K  (2-5)  where K is a constant. Since some of the OH" ions are consumed according to E q . (2-3), there needs to be an additional term added in order to account for the total OH", and Kubo [9] has proposed the following equation, f/ =K+K / i  1  (2-6)  in which K i is a constant whose numerical value depends on the properties of the resin used. The critical concentration of OH" ions closest to the cathode needed to initiate the coagulation has been estimated by several authors [8,10] to be that which gives a pH = 12. The coating thickness is dependent on the applied voltage. Therefore, film growth under a constant deposition voltage is: dbldt = cj  (2-7)  9  where b is the film thickness and c the coulombic yield (cm /C). If the film has ohmic 3  characteristics, then the relation j = idJ/b  (2-8)  can be used, where U is the deposition voltage (which is evenly distributed along the film) and K is the conductivity of the film. Substituting Eq.(2-7) into Eq.(2-8) and integrating, the relation between film thickness and time becomes: b = (2cKU) (t-tf m  12  (2-9)  Eq. (2-9) shows that film thickness is related to the square root of the deposition time at a constant deposition voltage. Practically, it is observed that the growth stops after < 50 |j.m thick film build-up. This self-control of the film growth is another useful feature of Ecoating process.  20 l  Figure 2-4. Film thickness with time of deposited coatings using constant voltages [26].  10  2.1.5 Heat evolution during E-coating A s previously discussed, the E-coating process is dependent on the conductivity of the part/substrate. A s the thickness and paint build-up increases on the surface of the cathode, the attractive forces between the paint particles and the part decrease due to the increase in surface ohmic resistance. Experiments published [12] show that there is significant heat evolved on the cathode throughout the process. Temperature rises in the first 15 s interval, and has a subsequent decay (Figure 2-5). This is explained by the fact that the current is the largest at the beginning of the process and heat is generated within the high resistance film [12]. The observed temperature-time dependence is a complex function of the increase in the film thicknesses during deposition, thermal conductivities of substrates, paint liquid and current decay functions. The influence of substrate will be discussed in detail in the following section. The possibility of film rupturing (cratering) during deposition is greatly enhanced when the heat is very high, e.g. for zinc-plated steel, Figure 2-5.  time, t/s Figure 2-5. Heat evolution during E-coating for different surfaces: 1, bare steel; 2, phosphated steel; 3, zinc-plated steel; 4, phosphated zinc-plated steel. U= 300V [12]. 11  2.1.6 E-coat Paint and Material Selection The paint and material selected for an E-coating process significantly affects the final product of the E-coat process. Therefore, the proper formulation and process must be used. Resin paint mixtures are similar to a household paint mixture containing resin, pigments, solvents, Dl water and other proprietary additives. A typical E-coat contains 2-6% of a high-boiling, water-soluble solvent such as butyl cellosolve or hexyl cellosolve. Higher concentrations increase the volatile organic compound (VOC) emissions [11]. Pigments within the paint do not undergo any electrical charging as seen on the resins. Resins however, are converted from soluble ionic state to an insoluble neutral form. Epoxy resins are the most commonly used precursors for both cathodic and anodic E-coatings.  For anodic resins, carboxylic groups are chemically  attached to the resin molecules. The resin is mixed with potassium hydroxide, sodium hydroxide, triethyl amine or other amine compounds to solubilize the resin molecules. Cathodic resin is chemically modified so that tertiary amine groups are located along the backbone carbon chains of the molecules. Without the ionizable groups such as the amines to electrically charge the resin, it cannot be used in an E-coat formulation. When treated with dilute acid solubilizers such as acetic acid or formic acid, the tertiary amine groups form the positively-charged particles needed for cathodic E-coating. An example of a cathodic electrodeposition bath preparation [14]: 1. Mix the components Epikote 1004 and 1001 and Ethyl Cellosolve 143 parts, (Table 2-1) and stir until all the ingredients are completely dissolved. 2. After heating to 50°C, add Diethanolamin and Isopropyl Alcohol to the initial components while stirring over a one hour period. Then maintain the mixture at 80-85°C for 3 hours.  12  3. Next, add the Adduct B-10065 and Ethyl Cellusolve (100 parts) over a 30 minute period while stirring. Continue to stir for 1.5 hours to obtain an amino-epoxyisocyanate resin. 4. Using this resin prepared in steps 1-3, with acetic acid ( as a neutralizer), titanium oxide (as a pigment) and carbon black, prepare an aqueous electrodeposition bath. Table 2-1. Paint bath mixing ingredients for resin solution. Epikote 1004  336 parts  Epikote 1001  143 parts  Ethyl Cellosolve  140 parts  Diethanolamin  59 parts  Isopropyl Alcohol  20 parts  Adduct B-10065  202 parts  Ethyl Cellosolve  100 parts  2.1.7 Substrate Selection and Surface Preparation for E-coating E-coating substrates must be conductive enough to allow the cathodically charged paint particles to be attracted and deposited onto the surface. Where exposure to environmental effects is highest, including door panels, hoods, and chassis, Ecoating is usually applied as an intermediate step prior to primer application and after phosphate treatment as shown in Figure 2-5 [15].  13  Surfaco  CLEARCOAT (~1.6-2.0milsor -40-50micr ons)  <CC/BCIntcrface  BASECOAT (~0.8milsor ~20micr ons)  PRIMER (~1.2miIsor ~25micr ons)  PRIMER (~1.2milsor -25micr ons)  ELECTROCOAT ,(~lmiIor ~25microns) ZINCPHOSPHATE SA TREATMENT /^j  PLASTIC SUBSTRATE  METAL : • (STEEL SUBSTRATE)  Figure 2-6. Schematic of typical automotive paint layers with approximate thickness for each layer [15]. Other possible substrates of choice are aluminum alloys, titanium alloys, zinc alloys and magnesium alloys. However, the user must carefully study the surface effects during the entire process, accounting for heat evolution, pH of bath, oxidation paths and how the surface quality will affect the deposition properties. The results of such a study [12] show that 4 of the more popular automotive materials are: bare steel, phosphated steel, zinc-plated steel, and phosphated zinc-plated steel. Applying a conversion coat to the substrate influences the surface conductivity of the part. From the results shown in the study [12], a phosphated surface is almost insulating, except for the pores between the phosphate crystals (Figure 2-6). Deposition starts to occur first within the pores and as OH" ions diffuse out, paint growth occurs over the phosphate. This produces a non-homogeneous coating layer in which the pores are not open, that is they do not penetrate to the metal surface. The wet film structure can be used to predict the final film quality.  14  — PAINT  (a).  PAINT - PHOSPHATE STEEL  (b) Figure 2-7. A model of depositon of particles of differently prepared surfaces; (a)nontreated surface, (b) phosphated surface [12].  A s shown in Figure 2-6, the bare steel has a very compact first E-coat layer while the phophated steel surfaces are more porous. A similar conclusion can be reached for the zinc-plated steel and the phosphated zinc-plated steel. It was also concluded that the resistivity of the films is not dependent on the film thickness, but rather on the first initially deposited layer. A s shown in Figure 2-7 the bare steel has a more compact initial layer than the other surface treated sample. Also, another study [16] of surface pretreatment effects on aluminum alloys 2024-T3 and Alclad 2024-T3 showed that pretreatment has a significant influence on paint properties. It was concluded that, by monitoring the nature of the oxide layer on aluminum alloys, the adhesion performance could be improved by an order of magnitude when compared to native surfaces. The study also concludes that conversion coatings, which are developed for improvement of corrosion performance as well as adhesion, could be eliminated by a simple process like alkaline cleaning. Common practice is to employ alkaline cleaner to remove dirt and oils found on industrial metal parts in manufacturing.  15  Aluminum objects generally receive a conversion coating prior to electrocoat and can be electrocoated simultaneously with ferrous parts [11]. However, the composition or chemistry of the metal is one of the key limiting factors in cleaner choice. The cleaner must be chosen so as to be compatible with the metal being processed. In multimetal cleaning lines, nonferrous metals are typically the limiting factors. With these metals it is important to choose a cleaner that either does not attack or overetch the metal; however, a controllable attack is desirable. Most aluminum and zinc alloys with slightly different alloy content can vary widely in their ability to withstand either alkaline or acidic cleaner attack. In some cases, where minute etch is desirable, slightly more or less is unacceptable [17]. A classification system was proposed to make cleaner choice easier (Table 2-2 [17]: 1. Ferrous or Iron Bearing: Cold-rolled steel, hot-rolled steel, stainless steel, and ferrous castings. 2. Nonferrous:  Aluminum sheet, coil, castings, extrusions, zinc castings,  galvanized, terneplate, and zinc plated. 3. Yellow Metals: Copper and brass. 4. Mixed Metals: Combination of the above. 5. Composites: Mixtures of metals with other materials.  16  Figure 2-8. Photomicrograph of first layer (wet) E-coated onto the substrate: (a) bare steel; (b) phosphated steel [12].  17  Table 2-2. Cleaner Selection for E-coating process [17]. Type of Cleaner  Description  High alkalinity  Noncaustic sensitive metal processing  PH= 11-13.5  only. Ferrous, stainless, and yellow metals. Composites  High alkalinity (buffered)  Multimetal lines ferrous/nonferrous  pH=10.5-12.5 Low alkalinity  Nonferrous, aluminum, zinc, also  PH= 6-9  effluent sensitive or restricted cleaning operations  High-foam surfactants  Static tank cleaning, immersion systems, and spray wand nonrecirculating systems  Controlled-foam surfactants  Agitated immersion ultrasonic, turbulant  Low-foam surfactants  Spray cleaning and recirculated systems  High temperature  Aged and oxidized soils of a waxy  140°F and above  nature, heavy accumulation and short process contact line  Medium temperature  Controlled solids. Light to heavy  120-140°F  accumulations. Short to medium contact line  Low temperature  Light to medium soils. Typical for  90-120°F  medium contact line  18  2.1.9 Corrosion Protection by E-coatings When a barrier such as an E-coat is deposited on a substrate, the main mode of resistance is via inhibition and physical obstruction against fluid penetration. The properties which allow organic coatings to behave as a corrosion barrier are governed by several other complex mechanisms. The performance of the coating system can depend on these factors: (a) the dielectric properties of the coating; (b) the adhesion of the coating to the given substrate; (c) amount of oxygen and water absorbed by the coating; (d) permeability of ions through the coating; (e) inhibitors and other compounds addition to the paint bath; (f) surface treatment of the metallic substrate's surface; and (g) reaction behavior during corrosion at the metal-coating interface during absorption of oxygen and water [4,18-22]. It has been previously shown [23-26] that the most important causes for corrosion in organic coatings are the oxygen and water uptake, and ion penetration through the coating. If there are imperfections or defects present in the coating, the defects will act as channels for corrosive ions to attack the metal surface at a rapid rate, hence localized corrosion can happen. In order to study the effect of corrosion on Ecoated metallic substrates such as bare steel, phosphated steel, aluminum alloys and conversion coated aluminum alloys, the impedance technique has been developed [2730]. The impedance model is shown in Figure 2-8, where R  n  electrolyte, R  p  is the resistance of  is the pore resistance due to electrolyte penetration, R is the metal t  charge-transfer resistance, C is the capacitance of the intact coating layer and Cd is the c  electrical double-layer at the metal/solution interface.  19  R _n_ AAA—  Rp -AAAA-  Figure 2-9. Impedance model for a metal coated with organic polymer [32]. Miskovic-Stankovic et al. [31, 32] have suggested that diffusion of electrolyte (3% NaCI for studies) into the polymer-coating occurs in two steps. Firstly, there is absorption of pure water into the micropores of the E-coating at a rate governed by Fick's law. The second step is related to water infiltration through macropores, where contact  between  electrochemical  the  electrolyte  reactions  begin  and the at  the  metallic  metal  surface  surface.  can  be seen,  However,  it  was  and also  demonstrated that step number two is the dominant factor for corrosion through A . C . impedance measurements. This is due to fact that conduction is enhanced rapidly when the electrolyte comes in contact with the metallic surface after breaching the coating through macropores. But if the electrolyte is in contact with coating as in step number one, the conduction variance is negligible, and the net result is swelling of the polymer. The polymer chains in micropores hinder mobility of ions and water molecules, which is not seen in macropores. The coatings are usually noted to be a rather inert and dielectric material by nature. Drazic et al. [33] have also shown that R will decrease with time when a 200V p  coated steel substrate is placed in 3% NaCI.  This gives the conclusion that ionic 20  conductivity is increasing and the coating is deteriorating in its protective capacity due to electrolyte penetration through the pores. A further look at a logarithmic R versus time p  shown in Figure 2-9, shows that the initial pore resistance, R °, is almost equivalent for p  all ranges in voltage. Therefore electrodeposition voltage does not significantly affect the coating properties prior to contact with corrosive agents.  Figure 2-10. Log of pores resistance (R ) versus time plot for films deposited at various voltages [33]. p  Another fact that can be seen is the rapid decrease in R for lower voltage (<200V) p  deposited samples and the almost stable R value for high voltage (>200V) depositions. p  2.1.10 Wear of polymer and polymer composites Polymers and polymer composites have been the choice for many applications ranging from biomaterials (joints, bio-coatings, etc) to machine parts and tools. This is due to their unique properties of being resistant to corrosion, low density and relatively high toughness, along with good finish and low coefficient of friction [34]. However, 21  polymers are frequently subjected to abrasive and sliding wear which deteriorates the performance of the material. Sliding wear involves two surfaces sliding on of each other, either with or without lubrication. There are several different techniques for the sliding wear test. A s shown in Figure 2-10. The geometry of the wear contact surface is also critical to the results as the amount of wear, as sometimes a pin or block might conform to the equivalent surface geometry or be of counterconformal (Figure 2-11).  c  Figure 2-11. Sliding wear test methods; (a) ring-on-ring;(b) face-to-face;(c)pin on flat face;(d) pin on rim;(e)block against ring;(f) pin on flat [40].  In the "abrasive wear" or "erosion" test, the material is to be removed or displaced by the hard particles on the surface of contact. There are three distinct types of abrasive wear: two-body abrasion, three-body abrasion and erosion, Figure 2-12. The two-body abrasion particles are stationary and cause protuberance on the counterface, while the three-body wear particles are mobile and roll or slide between the surfaces. 22  Two-body abrasion may also occur due to the difference in hardness and roughness of two contacting surfaces.  (a)  (b)  Figure 2-12. Geometrical factor of surface (a) conformal and (b) counterconformal contacts [40].  Erosion wear occurs when hard particles impact the surface via gas streams or flowing liquid (slurry) carriers. This type of wear usually occurs in lubricating oils, when hard particle contaminants entering the system are re-circulated at various flow speeds.  23  The abrasion significantly depends on the hardness, shape and size of the hard particles used. If the ratio between the hardness of the abrasive particle, H , and the a  surface hardness, H , is less than ~ 1 , the wear rate decreases due to slow volume of s  extraction (Figure 2-13). The wear rate increases significantly with increased particle hardness.  (a)  Two-body abrasion  (b)  Three-body abrasion  ^  (c)  ^  ^  ^  ^  ^  ^  ^  ^  Erosion  Figure 2-13. Schematic of different types of hard particle abrasion [40].  24  Figure 2-14. Particle hardness effects on wear volume when comparing hardness H (particle) against H (surface) [43].  a  s  The shape of the particle influences the point of contact, with angular (sharp) shaped particles being more abrasive than rounded particles (Figure 2-14). Shape of the particle can be described in terms of a two-dimensional projection of the particle via optical microscopy. A roundness factor, F, can be used to compare the ratio between the actual area, A, of the projection and that of a circle with equal perimeter, P [40].  25  (2-10) If F = 1, then the projection would be considered to be a circle; a smaller ratio would indicate a separation away from circular shape. Particles typically can range in size from 5 to 500  and the effects have been correlated for all three types of abrasive  wear and all three show identical trend of rapid increase in wear rate followed by a plateau (Figure 2-15).  Figure 2-15. S E M of silica particles for rounded (a) and angular (b) shapes [40].  26  ~  1.4  E  o  x 4  Speed mm s  Copper  1  Erosion 0.12 x 10  6  H2.0  1.2  O)  ^ 3„ 0 5  |  1.0  2  0.8  co  E E. CD  to  CD  i  g © 0.6  ro  CD  to to  •I  1 o g  CO  g  0.4  CD > CO  0.21- co L.  JO CO >%  CD  L-  •o o 00 £  50  -Q  100  150  200  250  0  Abrasive particle size (urn)  Figure 2-16. Wear rate of different types of abrasion and particle sizes of S i C [44].  The results of particle size effect can be influenced by the material's behaviour driving abrasion. For the material shown in Figure 2-13, plastic flow was present during the testing, while a more brittle material will most likely fracture upon impact.  It also  demonstrates that larger particle have much higher abrasion damage capabilities, which is why it is normal to filter out the large contaminants within a system (e.g. lubrication oil) to reduce wear of parts.  2.2 Piezoelectricity Piezoelectricity  is  a  phenomenon  which  is  only  demonstrated  by  non-  centrosymmetric crystals whereby an electric polarization or charge is induced in the material upon the exertion of a stress. This property can also be reversed, in which a measurable strain is developed within the material upon the presence of a proportional electric field. The former effect is typically used in dynamic pressure sensors, the latter property is commonly used in actuators [45]. Depending on the material used and the 27  fabrication and design, the frequency range where piezoelectric materials can detect changes in force or motion can be from below 1 Hz to above several MHz. Conversely, displacements in the urn range can be precisely measured, as can force changes from mNtokN. Most piezoelectric ceramic sensor formulations are based on P Z T (trademark of Clevite Corporation), with solid-solution composition of ~52-54 mole% lead zirconate (PbZr0 ) and 46-48 mole% Lead Titanate (PbTi0 ). 3  3  P Z T exhibits a crystallized  perovskite structure, like most ferroelectric materials, as shown in Figure 2-16. Lead atoms are positioned at the corners of the unit cell and oxygen at the face centers. The titanium or zirconium ions are located at the centre of the unit cell.  0  Ti,Zr  Figure 2-17. Typical crystallized perovskite structure with oxygen atoms at the corner, lead in the centre of each face and a titanium or zirconium molecule at the centre of the cube structure. 28  Polycrystalline ferroelectric ceramics such as P Z T initially contain randomly oriented domains/regions. These domains form upon cooling through the Curie temperature, to minimize the total elastic energy in the ceramic [46]. This randomness means that the initial ferroelectric material does not exhibit piezoelectric effects. However, piezoelectric effects can be induced by the application of a static electric field larger than the saturation field, but smaller than the breakdown voltage field at elevated temperatures below ferroelectric Curie point, where the domains are easily aligned. When the electric field is removed, some of the more highly strained domains will tend to revert to their natural state (depolarize), but the majority will remain aligned (remnant polarization) resulting in a permanently poled material. Depending on the overall composition, a titanium-rich P Z T system has sizable tetragonal elongation along the [001] and a large spontaneous polarization in the same direction. For a zirconium-rich PZT system, the rhombohedral ferroelectric state is favored and distortion occurs along the [111]. During poling of the material, phase changes between the rhombohedral and tetragonal phases can be seen [47]. Commonly, in order to enhance piezoelectric properties, a donor dopant (Table 2-3) is added. This causes the crystal structure to increase in metal/cation vacancies, enhancing domain reorientation. Therefore, two categories of P Z T can be distinguished. Depending on the dopant activity, donor or acceptor, "soft" and "hard" P Z T are tagged. A soft PZT will be designated as 5A and 5H; it uses a donor dopant which leads to large piezoelectric coefficients {d , d 31  33  and d ), large permittivity, high electrical losses, large 15  electrochemical coupling factors, very high electrical resistance, low mechanical quality factors, and low coercive field [48]. Hard PZT's (designated 4 and 8) with acceptor dopants cause oxygen vacancies. These vacancies pin domain walls with the  29  spontaneous polarization within a domain [49]. They are then characterized to have low piezoelectric coefficients, low permittivity, low losses, low electrical resistivity, high Q , m  and a high coercive field [50]. Table 2-3. A list of commonly used ions for substitution in P Z T [50,51]. Type Pb-site donors (Ti-Zr)-site donors Pb-site acceptors (Ti-Zr)-site acceptors  Dopants La , Bi , Nb , Sb , T h N b , T a , S b , W** K , Na , Rb F e , A l , l n , C r , Co \ Ga , M n , Mn , Mg , C u Sr* , C a * , B a * (for Pb* ), S n (forTi or Z r ) Cr, U a +  3+  b+  +  3+  a +  a+  +  +  a +  4 +  b +  +  +  a+  2 +  Isovalent substitution  3+  5 +  6  2 +  +  6+  3+  2 +  +  4 +  4+  4+  Multivalent ions  IsoValent substitutions tend to reduce Curie temperature and hence increase the room temperature permittitivity [52]. Multivalent ions, which are used to substitute for either T i o r Z r , reduce age effects [53]. 4+  4+  2.3 Piezoelectric Ceramic-Polymer Composite Piezoelectric ceramics have very low sensitivity to hydrostatic pressures because of their directionality of poling. So, the idea of incorporating piezoceramics into composites is said to bring the best aspects of each component in the composite, while minimizing the poorest features of each individual part. Some of the key advantages of composite are higher hydrostatic sensitivity, low dielectric constant, low density (good acoustic impedence matching with water and human tissues), high mechanical damping and mechanical flexibility [60, 61]. Commonly, composite piezoceramics will consist of two phases - a stiff piezoceramic and a soft polymer. A notation system has been derived in order to describe the all possible geometries/connections the two phases can have. It consists of notations given as 0-0, 1-0, 2-0, 3-0, 1-1, 2-1, 3-1, 2-2, 3-2, or 3-3, 30  where the first number denotes the physical connectivity of the active phase and the second number depicts the physical connectivity of the passive phase [62, 63]. Figure 2-17 shows the schematic of the more commonly used composites. The purpose of the polymer is to absorb stress in the transverse and longitudinal directions during hydrostatic applications. The properties of piezo-polymer are dependent on many factors, such as the properties of the dispersed phase powder and method of preparation.  particles in a polymer (0-3)  PZT spheres in a polymer (1-3)  laminated composite (2-2)  transverse poled composite (2-2)  perforated composite: (3-D  honeycomb composite^ (3-1 S)  diced composite d-3)  honey joiiib composite (3 1 P)  PZT rods in a polymer (1-3)  d-j | honeycomb (3-1)  3 ; perforated composite (3-2)  replamine composite (3-3)  BURPS composite.. "(3-3)  ladder structure (3-3)  Figure 2-18. Types of connectivity and phases present for composite piezo-ceramics [56]  31  2.4 Objectives For E-coatings to be used as a long term solution for outdoor  equipment  protection, the coatings mechanical properties such as abrasion/wear, scratch and hardness must be improved. This thesis assesses the possibility of such improvement of E-coatings by combining them with ceramic particles. The objective of this project was to study and characterize the properties of novel composite ceramic E-coating with various types and concentrations of the filler, and for deposition voltages ranging from 100 to 300 volts D C . This was done in order to (i) increase mechanical properties of the E-coat, which was normally used solely as a corrosion protection barrier and (ii) impart new functional properties of the coating, such as piezoelectric properties. The fillers were chosen to be calcined alumina ((X-AI2O3) of grades A14 and A16 (Alcoa, USA), for mechanical  properties  improvement,  and  BM532  PZT  powder  (SensorTech,  Collingwood, O N , Canada) for the functional composite E-coatings. For the case of alumina-filled composites, key mechanical properties such as scratch resistance, hardness, adhesion, and wear along with electrochemical behaviour were investigated, as a function of deposition bath conditions. The process engineering of the coatings were performed at U B C , whereas the functional E-coatings basic piezoelectric characteristics of impedance and polarization were determined collaborating industry (SensorTech).  32  by  Chapter 3. Experimental Procedure 3.1 Materials Commercially available cathodic epoxy Powercron® 590 paste was obtained from P P G Industry. This paste was diluted to create a 15 wt% solid suspension in deionized water. The composition of 3.785 litres (1 gallon) of the bath is shown in Table 3-1. Table 3-1. Composition of "as received" standard stock solution of cathodic epoxy bath. Component  Quantity (grams)  Deionized Water  2216.7  Resin Paste (36.2 wt% solids)  1583.3  Catalyst (E6165)  19.0  This suspension was magnetically stirred at all times and kept at room temperature to prevent premature curing. The standard resin and bath were not altered nor studied due to the complexity of the mixture. It was prepared and used as specified in the instructions given by P P G Industries [64]. Calcined alumina powder, <x-Al203, obtained from Alcoa Industrial Chemicals, USA, was used as the filler for wear enhancements of the composite E-coat.  Two  different grades of alumina powder were used, A16 of particle size 0.5[im and A14 of particle size 3um. The alumina was magnetically stirred in deionized water to create a 50 wt% suspension and stirred for a period of 24 hours prior to use (Figure 3-1). This was followed by a 10 minute ultrasonic deagglomeration using a Horiba Ultrasonic Disruptor Probe, prior to adding it to the epoxy solution. The aqueous suspension was favoured over direct powder addition, in order to preserve as much as possible water to resin ratio and maintain a homogeneous suspension. If dry alumina powder was added directly, the total solids to water ratio would be significantly disrupted and would hinder 33  coating deposition. Various percentages of alumina were used to a maximum of - 3 0 vol% in the composite solution.  Alumina Power Dl Water  i  Magnetic Stir 24 hours  50wt% Suspension |  Ultrasonic Deagglomeration 10 mins  50wt% Suspension Epoxy Bath |  Magnetic Stir 24 hours  Composite Bath Figure 3-1. Flow chart of the procedure for making alumina suspension and composite bath for deposition. In the latter stage of the project Lead-Zirconate-Titanate (PZT) B M 532 powder produced by Sensor Tech Inc. Collingwood, O N , was used as the piezoelectric ceramic filler. The percentages of each constituent were proprietary  and unavailable for  publication. Again the powder was added in the form of a water suspension of 50 wt% PZT in deionized (Dl) water. The substrates used consisted of 1.5" by 1" (2.54cm x 3.81cm) stainless steel coupons. The particle size distribution of the 'as received' powders were determined using a Horiba C A P A - 7 0 0 particle distribution analyzer, Scanning Electron Microscopy (SEM) at 20 kV and 15mm working distance was performed using a Hitachi S3000N system  34  with Quartz X O n e Energy Dispersive X-ray (EDX) chemical analyzer to study crosssection  of  coated  specimens and  powder  morphology.  The  time  required  to  deagglomerate would be determined by these tests.  3.2 Sample Preparation To study the effects of ceramic filler on the mechanical properties of the polymer coating, various concentrations of alumina were added to the E-coat bath, ranging from 0 vol% to a maximum of ~30 vol% in solution. These suspensions were then coated onto substrate using various voltages in order to study the influence of the alumina on the coating process. Voltage ranged from 100 volts to a maximum 300 volts with 50 volts increments. The stainless steel substrates were cleansed in alkali soap and dried as recommended by Gruss [17].  Figure 3-2 shows a schematic of the cathodic  electrochemical cell setup for all tests.  Figure 3-2. Schematic of E-coat cell process setup.  35  The substrate is denoted as the negatively (cathode) charged electrode, which is immersed into the bath. The D C power was supplied by a Xantrex X H R 300-3.5 unit with maximum output voltage of 300 volts and 3.5 amperes. The deposition cell (300 ml Pyrex beaker) and two counter electrodes (anodes) were 4" x 1.5" panels of stainless steel, placed in parallel to the cathode at 20 mm distance from each side. The composite solution was magnetically stirred to maintain a homogenous mixture and the stirring stopped immediately prior to application of DC voltage. Once the cathode was in place, the DC voltage was applied until the current reached zero amperes on the digital display of the power source, and for 60 seconds thereafter, to ensure full coverage. The sample was then rinsed with Dl water and allowed to air dry for 30 minutes. The samples  were  cured  at  180°C  for 20 minutes,  according  to  manufacturer's  recommendations. The coated samples were then sheared and polished (SiC wheels and 6(am and  1luti  diamond) to characterize the coating. The data acquisition computer  system plotted the current flow against time of deposition. Table 3-2. Bath details of each coating deposition with the alumina suspension (50 wt%) added to constant amount of 150 ml of stock paint. Bath Name A s received  Mass alumina added to bath (g) 0  A  5  B  10  C  15  D  20  E  40  F  60  Concentration of alumina in composite bath (vol%) 0.000 0.631 1.263 1.894 2.525 5.051 7.576  36  3.3 Characterization the Coatings 3.3.1 Microstructure and Thickness Observations Cross-sectioned samples were prepared in order to observe the homogeneity and dispersion of alumina particles in the polymer matrix. The sheared samples were vacuum mounted in low viscosity resin (Cold Cure parts A and B at 2:1, Industrial formulators, Inc. Burnaby, B C Canada). The mounted sample were subsequently polished using S i C from grit size 50 to 1200. This was followed by 6 and 1 jam diamond polish. The samples were then viewed under the scanning electron microscope using variable pressure mode, which does not require the sample to be conductive. The images were acquired using B S E (backscattering electron) and chemical analysis was performed using E D X (Energy Dispersed X-ray). The setting used was a pressure of 20 Pa and 20.0 kV. For S E (secondary electron) images, the samples were carbon coated and full vacuum mode was used along with 20kV.  The film thickness is controlled by the resistivity of the film (Section 2.1.4), and the introduction of alumina as a second phase has the capacity of altering the deposited film thickness due to difference in resistivity of the two materials (~ 8 times) during deposition, altering deposition properties such as thickness and surface finish. Coating thickness was gauged using a Positector 6000, Defelsko Corp., NY, U S A and then reconfirmed using S E M imaging. The instrument is based on electromagnetic induction and measure the change in magnetic flux density at the surface of a magnetic probe as it is brought near steel. The magnitude of the flux density at the probe surface is directly related to the distance from the steel substrate. By measuring flux density the coating thickness can be determined. The accuracy is set to ±1 |im + 1% for 0-50jim and +2 \im + 1%for> 50^m. 37  3.3.2 Mechanical Properties  The Romulus Universal Tester (Quad Group, Spokane, W A USA) was used to determine adhesion bonding, scratch resistance and micro-hardness for the coatings.  3.3.2.1 Adhesion Test This test consisted of pulling off the attached coating from the substrate via a normal applied force.  Figure 3-3 shows the schematics of the test and the parts  required to perform the tests.  This test used a proprietary epoxy resin (maximum  strength of 72 MPa) that was pre-applied onto equipment-specific aluminum studs manufactured by Quadgroup Inc. These studs were then placed onto the coating surface using an adapter, which aids in keeping the studs at right angles to the surface while the specimens were placed in an oven for one hour at 120°C to allow the stud epoxy to cure and bond onto the surface of the coating. Once cured, the stud was placed into a holder with three-jaw gripper, where it was pulled. There was a stud specific spacer that held the sample, which allowed the stud to be inserted and maintain a normal force. The stud was subsequently pulled off the samples at user set rate (5 Newtons/second) to a maximum force of 200 Newtons. The test ends when either one or more of the bonded structures was detached, or the maximum force was achieved. There are four possible failure modes which include (a) coating adherent failure, (b) coating cohesive failure, (c) a mix of (a) and (b) and (d) pull agent adhesion failure (Figure 3-4).  38  The General Form of Universal Platen Three-J aw Gripper  B  Grip Knob  —  Pull Stack  Test Deck  Pull Down Apparatus  m Clip— Sample Stud Clip-  • Sample  (b) Figure 3-3. Schematics of stud pull adhesion test platform (a) and sample preparation technique (b) [65].  39  Normal  t Stud  Adhesive  Figure 3-4. Schematic of stud placement and failure locations; (a) coating/substrate interface, (b) cohesive, (a) + (b) mix and (c) pull adhesive.  3.3.2.2 Scratch Hardness/Resistance Test The semi-quantitative "pencil" scratch resistance test is considered to be a common industrial test to compare relative coating hardness [66]. This was done by using different grades of sharpened HB system pencils with B9 being the softest, H9 being the hardest and HB being the middle grade. The surface of the coating was scratched at right angles to the tip of the pencil. If a pencil mark was left on the surface of the coating, the coating was considered to be softer than the pencil. On the other hand, if no mark was left, then the coating was harder than the pencil. The coating would be given a pencil grade based on the pencil prior to leaving a mark. This technique however became an inconsistent method of testing.  The sharpness of the  pencil varies from test to test even if sharpened prior to use, also the manually applied force varied from test to test. It was decided therefore that the quantitative scratch test 40  should be done using a mechanized standard testing procedure. The Romulus stylometer platform consisted of a spherical diamond tipped stylus with a constant speed travel stage moving perpendicular to the stylus (Figure 3-5).  Transducer  Stylus.  • E3E1  \ I BStage:  Stage Motor  TI'T) U  Figure 3-5. Schematic of stylometer scratch test platform [65]. An acoustic transducer, mounted on the stylus, was used to identify the point of microcracking initiation, penetration through the film and/or failure of the substrate. Acoustic energy, applied force and tangential (drag) force are plotted as a function of travel distance, as well as effective friction. There are three possible failure modes, (1) Micro-cracking, when the transverse, or diamond drag stress reaches the tensile strength of a non-brittle coating, a microcrack is created normal to the travel direction; bursts of acoustic energy are released as the coating rebounds. (2) Coating separation: the coating is poorly adherent and large areas of coating are pulled away; for better adherent coatings (shown in Figure 3-6), there will be a substantial number of micro-cracks generated before coating failure. (3)  41  Substrate Failure: the compressive strength of the substrate was exceeded, and portions of the substrate surface were crushed. This condition results in high energy noise bursts and normally increases the transverse force and effective friction. When micro-cracking of the coating starts, the acoustic output increases. One should be observant of the "critical event" which was the earliest point at which coating removal is observed and its force level is the measure of adherence. Conclusions can then be drawn on the basis of subsequent microscopic (100 time magnification) confirmation of the failure point.  Figure 3-6. Chipping of coating during scratch testing.  42  Force vs. Distance 90.0- ,  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 1.0  —  1  45.0- 11 '.• ..  •  •• -  '  "  ]  J  350-  ••: -.• . V • '  •\  -Q3  .  -08  ......  -07  1  g30.0- I  ; Mb)  -06  ^  g 250£ 20.0-  -as -0.4  150-  -03  100-  -02  50-  -0.1  003  250  iaco  5.03  12*50  15.00  Scralch Distance (mm)  Figure 3-7. Screen capture of results for scratch testing with failure noted by the acoustic burst (circle); (a) normal force, (b) normal friction, (c) transverse friction and (d) acoustic.  3.3.2.3 Microscale Abrasion Test Microscale abrasion test was performed using T E 6 6 Micro-scale Abrasion Tester, Plint and Partners Ltd., Wokingham, UK, (Figure 3-6) at the National Research Counsel at U B C . The abrasive agent consisted of 6jj:m Alcoa alumina suspended in distilled water at 20 wt%. The suspension was fed at a constant rate using a pump and the abrasive solution was fed between the rolling steel ball and the coating surface. The steel ball rotated at a rate of 80 rpm for 1000 cycles and a counter weight of 0.3 Kg was used to maintain constant load on the surface of the coating.  The samples were  subsequently characterized using a interferometric surface imaging system (WYKO NT2000, Veeco Instruments Inc. Woodbury, NY USA)  43  Pivot  Sample and holder  Figure 3-8. Schematics of Microscale Abrasion Tester [34].  3.3.2.4 Hardness Measurement The nano-indentation technique was chosen over micro-indentation due to the relative thinness of the coatings (18-78 ^im). A nano-indenter (Fisherscope H100, Germany) with loads capability ranging from 0.4 mN to 100 mN was used. The maximum depth was chosen to be 2 |im with 50 seconds loading time. From this test the elastic modulus of the coating can be attained directly from the load -displacement plot as in Figure 3-9. The stiffness, S, or Elastic Modulus, is calculated at the transition between load-to-unload were it is considered elastic [65].  44  3.3.3 Dielectric Strength Test Dielectric Strength (DS) is listed in terms of kilovolts per millimetre (kV/mm) taking into account the thickness of the material as breakdown voltage, V , and simply B  calculated as V I thickness. The tests where performed using a Sentry 20 A C / D C Hipot B  Tester (QuadTech, USA). The electrodes where placed directly onto the coating surfaces and breakdown voltages measured and averaged. Figure 3-10 shows a schematic of the breakdown voltage test and setup during testing.  45  3 . 6 0 K\1  OOD Power S u p p l y .  <  E l e c t r o de ^~^Y  *V  <•  .  ; Coating  V  -Holder/Insulator  k-Electrical  -i  Path  ~'  i Substrate  s  I  Figure 3-10. Schematic of breakdown voltage test setup during testing.  3.3.4 Pontentiodynamic Test E-coat has the primary use of a barrier coating against corrosion for metallic substrates. The addition of alumina particles as a secondary phase will alter the corrosion behaviour of the coating. Therefore potentiodynamic test, which is a form of accelerated corrosion test, was used to determine changes in coating barrier protection of the new composite film. The  polarization  behaviour  of  the  composite  was  determined  using  Potentiostat/Galvanostat Model 273A, Princeton Applied Research, U S A , with the measurement hardware SoftCorr III version 3.11 on the Windows™ X P operating system. All experiments were carried out at room temperature and aerated atmosphere, using a saturated calomel reference electrode (SCE) and a platinum counter electrode (Figure 3-11). The samples (working electrodes) were prepared by bonding the noncoated surface of the 0.25" x 0.25" (0.7 x 0.7 mm ) samples to a tail-naked copper wire. 2  46  The welded areas and exposed conducting surfaces were covered using epoxy (Lecoset 7007 cold made by mixing methylmethacrylate  and Leco 100 castable  mounting at 1:1). Furthermore, waterproof TEFLON® tape and acetone diluted stop-off lacquer (50% Microstop, Pyramid Plastic, Inc) insulating paint was used to prevent electrochemical reaction along the non-coated surfaces and to seal the epoxy interfaces. Each sample was then placed in a quartz tube to give the sample enough stiffness to stay in place while the electrolyte was being stirred magnetically when the experiment was in progress. Table 3-3 lists the paramenters used for the experiments. Table 3-3. Conditions for potentiodynamic measurements of coated samples. Electrolyte concentration (NaCI)  1N  Deionized Water pH  5  Measurement Range versus S C E  -600 to +600 mV  Scan Rate  1 mV/s  47  electrode  Figure 3-11. Schematic of electrochemical cell for potentiodynamic testing [68].  3.3.5 Piezoelectric Effects If a new transducer-like coating was to be developed, it required to be characterized for sensitivity, in particular measurements such as strain (d ) coefficient. 33  This was done in order to see if the process may lead in the future to a sensor-like  48  coating. All tests were performed on their equipment by SensorTech Inc., Collingwood, ON.  3.3.5.1 Impedance Measurements Impedance  measurements were taken  using an Agilent 4294A Precision  Impedance Analyzer with a custom made spring loaded jig/holder to keep the sample suspended. Figure 3-12 shows the complete system during testing, and Figure 3-13 shows a magnified picture of the jig used to hold the sample in suspension. The jig consists of a metal base with two Plexiglas stands with spring metallic loaded pins as sample holder.  Figure 3-12. Picture of setup for impedance measurements.  Every piezoelectric material is known to have its unique vibration/frequency at the peak efficiency transfer of an alternating signal would be converted into mechanical energy. This is known as the resonance frequency, f„ or fp>. 49  Holders/ Conductive Pins Plexiglass Stands  !  ^ ^ ^ ^  Sarriple  Figure 3-13. Schematics of sample holder jig for two-side coated samples.  3.3.5.2 Polarization Measurements Polarization  measurements  were  performed  using  a  combination  of  SensorTech's S S 0 5 Polarization meter, T R E K 609E-6 High Voltage (Max. 4000 volts) Amplifier. The sample was held using the custom made jig/holder for two sided coated samples as described in section 3.3.4.1. The polarization of piezoelectric material will show the degree of poling that has taken place. The piezoelectric material must be poled in order to have maximum directional sensitivity, and distinct hysteresis loop will be presented depending on the degree of poling. S o , this test will allow the user to observe whether previous poling has taken place, if a hysteresis loop is observed.  50  Figure 3-14. Picture of Sensor Tech S S 0 5 polarization meter.  Chapter 4 Results and Discussion 4.1 Ceramic Filler Powders Characteristics The powders from Alcoa alumina (A14 and A16) and Sensor Tech Inc. PZT (BM 532) have been characterized with respect to particle size, particle size distribution, and morphology. Since the powders are to be mixed with Dl water, it is important to study the degree of agglomeration and the time required to deagglomerate the particles in suspension, using the ultrasonic dispersive probe.  It is assumed that by increasing  ultrasonic dispersion time the particle distribution should give rise to the particle size ranges for A16, A14 and B M 532, as suggested by the manufacturers of 0.3-0.5um, 35um, and 2-3um, thus ensuring a good suspension of the powder in the E-coat suspension. The powders were studied by S E M to determine particle morphology. Each powder was placed onto a sample holder directly, without using dispersants. This showed the natural state of each powder and presented the requirements for proper dispersion in the suspensions. It can be clearly seen in Figures 4-1 to 4-3, that A14 and BM532 powders have good particle separation and may not requir large amounts of chemical or physical dispersion in water. However, A16 powder appears to be greatly agglomerated in the dry form. Therefore, deagglomeration was required prior to adding the alumina in suspension to the paint bath. Powdered B M 532 was observed to be approximately spherical, with round edges and of similar size (Figure 4-3).  52  Figure 4-3. BM532 as received P Z T powder under S E M .  To determine the time necessary to disperse the three powders, alumina A16 powder having an average particle size of 0.3-0.5um was used as the test sample. The smaller particle, size the more likely it is to agglomerate in a liquid suspension, due to the large surface energy associated with small particles. In the case of larger particles such as A14 (3um) and B M 532 (2|im), the likeliness to observe agglomerates once they have been dispersed according to steps taken for the A 1 6 alumina would be low. Figures 4-4 to 4-6 show the analysis results for particle distribution of each material after 10 minutes of ultrasonic dispersion. A16 still was observed to have agglomerates >0.5 urn, however, over 80% of the particle are within the preferred range (0.3-0.5 jam), which was a substantial improvement over the as received powder (under the S E M which showed the presence of large agglomerates >1 um). A 1 4 and B M 532 show very good distribution after 10 minutes of ultrasonic dispersion. Therefore it was determined 54  that 10 minutes of ultrasonic treatment was sufficient to disperse and deagglomerate the powders in a water suspension.  30  0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  Diameter (^m) Figure 4-4. A16 alumina particle distribution after 10 minutes ultrasonic deagglomeration.  35  00  1.0  2.0  3.0  4.0  5.0  6.0  7.0  Diameter (jim) Figure 4-5. A14 alumina particle distribution after 10 minutes ultrasonic deagglomeration. 55  40 ^  35  ^  30  0 0.0  1.0  2.0  3.0  4.0  Diameter (jim)  5.0  6.0  Figure 4-6. BM532 P Z T powder particle size distribution after 10 minutes ultrasonic deagglomeration.  4.2 E-coating Structures The surface of structure of the A16 filled E-coatings with varying concentration of alumina and deposition voltages are shown in Figure 4-7. The surface of the composite coatings was observed to follow the surface morphology of the substrate forming flat coated surface in most cases. The curing stage also had a large influence on the finished coating surface, due to the flow of the resin during curing. At voltages above 200V crater and pinhole formation increased. Similarly at high concentrations of alumina in coating (30 vol%). The extreme case combing 30 vol% alumina in coating deposited at 300 volts, showed excess materials on the surface and large craters with micron sized pinholes. A cross-section observation of the A16 composite coating viewed under S E M is shown in Figure 4-8. The alumina A16 is dispersed throughout the coating. The  56  morphology for PZT-composite was similar to that of the alumina-based composite, as shown in Figure 4-9 and 4-10. For a high voltage (>200 volts) and high alumina content (> 12.5 vol%), the coatings were very irregular, as observed in the cross-section and surface morphologies shown in Figures 4-11 and 4-12. The holes are 0.3-05 mm large and can be seen immediately after the coating procedure has been completed. The curing stage does not cover nor smooth out the defects, due to the presence of high alumina in the coatings, which hinders the mobility of the resin. The finished composite E-coating has very smooth and glossy surface without showing traces of alumina on the surface when viewed under S E M or optical microscope. However, increased A16 alumina concentration (> 12.5 vol%) and voltage (>200V) dull the coatings appearance and significantly deteriorate the surface. The alumina is observed to be encapsulated beneath the surface, therefore the coating is still compatible with other coating, ie. primer, basecoat and clearcoat.  57  100V  200V  300V  Figure 4-7. E-coated and cured samples with variable deposition voltages and concentrations of A 1 6 alumina in coating.  58  4  SUBSTRATE A16 A l u m i n a  Figure 4-8. S E M image of A16 alumina (12.5 vol%) reinforced composite coating.  SE  WD14.9mm  20 . O k V ° x°600°  ° " 5°0um  Figure 4-9. P Z T (12.5 vol%) E-coat composite cross-section S E M image.  59  x 1 5 Id  L10 0 0  £' 0 k  £ 0 0 .u r 11  Figure 4-11. High voltage (300V) and high alumina (30 vol%) in coating cross-section SEM.  60  Figure 4-12. Surface morphology of high alumina (30 vol%) and high voltage (300V) deposited coating . By comparing the S E M image of an A16 sample deposited at 200 volts (Figure 48) to an A14 filled sample deposited at the same voltage  (Figure 4-13), the  microstructure was observed to be significantly different. The A 1 6 sample showed an even distribution of alumina, while the A14 sample showed patches as large as 20-30 (xm alumina clusters along with low or no reinforcement regions, with only epoxy being present.  61  Figure 4-13. Cross-section S E M image of A14 particle reinforced coating deposited at 200 volts.  4.3 Voltage Dependence of Coating Thickness A s discussed previously in Section 1.1.5, the voltage applied during deposition determined the quality and quantity (thickness) of the electrodeposited coating. The applied voltage induced electrophoretic movement of the charged particle toward the cathode and initialized decomposition of the water at the surface of the cathode. OH" ions forming at the cathode plane increased local pH, which commenced coagulation of the  resin  micelles and  build  up  of the  film.  According to  the  manufacturer  recommendations, time is not a process variable, i.e., each deposition should be performed for at least 90 seconds. For the presented study, time for deposition was the 90 seconds required plus 60 seconds added to assure that full deposition was achieved, to a total of 2.5 minutes. Figure 4-14 shows the relationship between deposition voltage  62  and thickness of the film deposited for the "as received" paint bath. This will be used as the reference data to compare the data acquired from the composite suspensions.  Figure 4-14. Voltage dependence of thickness for the paint during deposition at constant voltage. Figure 4-14 shows that thickness is not linearly dependent on the deposition voltage, but the data appears to fit a parabolic relationship best according to the R  2  fitting using trend line fitting software provided by Microsoft Excel. Deposition voltages above 300V were not performed due to equipment limitations (300 volts and 3.5 Amps), however, once the deposition voltage surpassed 250 volts the film did not appear to cover the entire substrate and pinholes can be seen with the naked eye (Figure 4-17). Deposition voltages below 100 volts would be providing inadequate thickness to the norm of greater than 20 |xm coatings throughout industrial use.  63  4.4 Influence of Alumina in Paint Bath on Deposition Process 4.4.1 Conductivity Changes Changes in conductivity within the solution can significantly alter the deposition conditions. According to the guidelines given by the manufacturer, the conductivity of epoxy bath should be in the range of 900 to 1300 uS/cm. This was fairly consistent throughout the paint bath processing prior to the addition of the alumina suspension. An increase in conductivity would alter the deposition process, more specifically the electrophoretic migration of the charged resin micelles, and coagulation/neutralization of the resin micelles. The changes in conductivity and pH with increasing alumina concentration in the bath are shown in Figure 4-15 and 4-16. The Dl water was tested to have a conductivity of < 10 uS/cm, the 50wt % alumina suspension had a conductivity of 1000 p.S/cm and the as received paint had a conductivity of 1100 (aS/cm. Once the alumina suspension was added, the conductivity measurement showed a substantial increase with the initial addition of alumina suspension to the paint bath. However, after further additions of alumina suspension, the conductivity decreases almost linearly.  64  2000  600 -I 0  1  1  1  1  .  1  1  2  3  4  5  6  1 7  Alumina in Solution (vol%) Figure 4-15. Conductivity measurements of composite bath with increased A16 alumina in suspension.  On the other hand, the pH increased linearly with increased alumina content in the bath. This was expected due to the fact that the initial pH for the as received paint bath and alumina suspensions were 4.6 and 9.8 respectively. Having a weak acid solution and the small addition of a weak base, the neutralization effects are expected to occur. A s observed, the conductivity does not decrease until the concentration of alumina in the solution surpasses > 1.56 vol%, where the weak base (alumina suspension) becomes a significant amount and hydroxyl ions are neutralized by the hydrogen ions in solution, thus decreasing the conductivity and increasing the pH.  65  6.5 6 5.5 A 5^ Q.  4.5 T 4  ^  3.5 H 3 0  2  3  4  5  6  7  Alumina in Solution (vol%) Figure 4-16. pH measurement with increasing A16 alumina in suspension.  4.4.2 Electrical Resistance The E-coat film thickness is controlled by the electrical resistance, which increases with the film thickness. The epoxy film was measured to have an average dielectric breakdown voltage of 1.97 kV with the thickness of 24 um. When this is translated in the more common kV/mm units to compare each material's dielectric breakdown voltage, a dielectric strength of - 8 2 kV/mm was calculated, which is substantially higher than that of alumina ceramic with a dielectric strength of 9.9-15.8 kV/mm [50]. Alumina has a high dielectric breakdown voltage when compared to most ceramics considered to be insulators, with the exception of mica (40-80 kV/mm) and boron nitride (36-55 kV/mm). The addition of alumina suspension into the paint bath causes a decrease in dielectric strength of the deposited film (Figure 4-18). Breakdown voltages were obtained at three different points on the coating: first in the centre of one 66  third of the distance from the top, second one in the middle third and lastly center one third from the bottom of the coating (see Figure 4-17) and then the averages were calculated.  From the results shown in Figure 4-18, the initial stages show an  approximately linear decrease in dielectric strength when the concentration of alumina is increased, which agrees with the fact that a lower dielectric strength material was added. However, a change in slope can be seen once ~5 vol% A l 0 2  3  in the coating is  surpassed; this may signify that the coating resistance is now being dictated by the lower dielectric strength material hence the near flattened slope. This effect can be explained by the percolation theory [68]. A composite with two distinct phases with different physical properties (i.e. electrical conductivity) will take the average value of the two phases until sufficient minority phase is present to form a continuous network (percolation limit), and the composite will take the physical properties of the minor phase. For example, if the minor phase has a distinctly smaller electrical conductivity, once the percolation limit is reached, the conductivity will rapidly decrease to the level of the minor phase. This behaviour was observed for the results given in Figure 4-18, where the dielectric strength of the matrix (~82 kV/mm) was decreased to that of the filler material (~20 kV/mm). Cassignol et al [69] and Mamunya et a/.[70] have shown that by filling a non-conductive epoxy with a highly conductive material such as metal powders and conductive polymer polypyrrole, the conductivity of the composite approaches that of the filler material at < 6 vol% filler.  67  Coated  Figure 4-17. Schematic of test points chosen for dielectric measurements for coated samples of various concentrations of alumina and deposition voltages.  100  E  90 1  0 "I 0  '  1  T  1  1  i  1  1  2  4  6  8  10  12  14  Alumina in Coating (vol%) Figure 4-18. Dielectric strength with increasing A16 alumina in coating.  68  4.4.3 Thickness of Coatings A s discussed in Section 2.1.5 there are several factors that influence the resulting thickness of the film.  The initial stages of deposition require a critical  concentration of OH" ions, COH-, to attain pH=12 close to the surface of the cathode, as quantitatively described using Eq. 2-4. Once this critical pH is achieved, coagulation due to neutralization of the cationic species takes place and the film growth period begins. Film thickness has been quantitatively described using E q . 2-9. The conductivity of the coating, K , was increased due to the presence of lower resistance alumina particles in the film, creating not only a lower path of conduction but decreasing the overall film resistance which leads to a increase in thickness of the deposited films (Figure 4-19). This can be seen clearly in Figure 4-20, as the percent thickness increased quite rapidly once ~5 vol% alumina in coating was reached, as explained by the percolation theory, and similarly observed for the dielectric strength testing.  A way of expressing these  changes quantitatively would be to use the rule of mixture (Appendix I) [68] and derive the conductivity for the composite, K . From this, one can derive the relationship C  between thickness, b, and volume fraction epoxy, V ,: e  I I I  b = B(\o-9vey(ry(cy  (4.1)  where T is the induction time variation with alumina, c is the columbic yield and 6 is a constant.  69  80 • 0 vol%  70  • 4.53 vol% 6.08 vol%  _ 60 |^  A  ^ 50 (0 <D •* o 40  x 12.33 vol %  • 10.32 vol %  x  30 20 10 100  150  200  250  300  Deposition Voltage (V) Figure 4-19. Thickness changes with increase in voltage and concentration of A16 alumina in coating.  Alumina concentration (vol%) Figure 4-20. Percentage increase of thickness with increased A16 alumina in coatings deposited at a constant voltage of 200 V.  70  4.5 Mechanical Properties of E-coatings One of the objectives of the presented research was to study the mechanical behaviour of the composite E-coatings. Key industrial tests for polymeric coatings would include pencil testing. More fundamental tests include diamond tip stylus scratch, which simultaneously tests hardness and adhesion of the coating to substrate. Other tests performed are tensile adhesion of the coating to the substrate.  4.5.1 Adhesion The samples were subjected to a tensile force until either the coating was detached from the substrate due to the tensile force, or the proprietary epoxy (Max. strength 72 MPa) detached from the coating surface. For all samples tested, the proprietary bonding epoxy failed at adhesion strengths ranging from 68 M P a to 82 M P a (Figure 4-21) without removing or detaching the coating. It was observed that the use of the tensile method to test adhesion was unable to pull the E-coating of the substrate. The adhesive strength of the film was much greater than that of the supplied bonding intermediate.  71  90 -r  Q. 2 •*•*  c is 75 C  o w  T  5  <  60  0  2  4  6  8  10  12  14  Concentration A l u m i n a (vol%) Figure 4-21. Adhesion strength of tensile stud pull test of coatings deposited at 200V.  4.5.2 Scratch Resistance 4.5.2.1 Pencil Scratch Scratch test was performed using industry standard of pencil scale (Figure 4-22). This showed a large improvement in hardness when alumina was added as a filler. A large increase from H2 pencil hardness to H8 occurred with the addition of only 3.33 vol% alumina in coating. The hardness increased to the maximum H9 after 13.33 vol% in coating was reached and stabilized at that hardness.  72  12  T  10 -  0  5 10 15 Alumina in coating (vol%)  20  Figure 4-22. Typical pencil hardness test for 150V deposited coating with increasing A16 alumina content. A s previously mentioned, the pencil hardness is an industrial method, and it was a quick and easy method to characterize the coating hardness by comparing relative hardness.  4.5.2.2 Stylometer The data obtained from stylometer test are summarized in Figure 4-23 and 4-24. The program allowed the viewing of the entire scratch path with the attached video scope with 10 times magnification (Figure 4-25). The user was able to determine the critical force by interpreting  acoustic signal changes when failure occurred and  validating it with visual inspections via video. Figures 4-26 and 4-27 show a screen capture of the displayed data for as received and reinforced coating tests. The acoustic sound burst/peak for the 'as received' film can be observed to occur right at the initial stages of the test (approx. 3 N and 0.16 mm) and then remains constant, while the reinforced film's burst/peak occurs approximately halfway through the test at approx. 25 73  N and 6.85 mm. This means that the stylus penetrated the unreinforced film without any effort, while the reinforced film required 8 times more force (3 N vs 25 N). This was constant for all 0 vol% E-coatings as no change was observed (Figure 4-24). It can be clearly seen that the presence of alumina in the coating has a substantial effect on scratch resistance. The increase in critical force was very rapid, with multiples of 7-10 times seen with the additions of 4.2 wt% alumina in coating. Depending on the deposition voltage, a peak critical force can be seen, followed by a decrease in force with higher alumina content in coating, likely due to coating defects occurring at high voltage and alumina content (refer to Section 4.2).  80  • 100V 1 1 5 0 V  2  4  A 200V X 2 5 0 V  6  8  X300V  10  12  14  Alumina in coating (vol%) Figure 4-23. Scratch test results performed by stylomer for various deposition voltages and A16 alumina concentration in coating.  74  60  • Q) O  _A  40  re o S 20  • 0 vol% a 4.2 vol% • 10 vol%  o 4 0 100  150  200  250  300  Deposition Voltage (volts) Figure 4-24. Scratch test results for 'as received', 4.2 vol% and 10 vol% A16 alumina coatings at various deposition voltages.  75  (a)  Figure 4-25. Magnified 10x Stylometer results for (a) 0 vol% A16 alumina in coating and (b) 10.0 vol% A16 alumina in coating.  76  Force vs. Distance  Scratch Distance firm)  Figure 4-26. Scratch test display resulting from a non reinforced film with acoustic burst (circle); (a) normal force, (b) normal friction, (c) transverse friction and (d) acoustic.  Force vs Distance  003  2EU  Du:  10.C0 Scratch Distance (irm)  1253  1SC0  Figure 4-27. Scratch test display resulting from a 10.0 vol% A16 alumina reinforced film with acoustic burst (circle); (a) normal force, (b) normal friction, (c) transverse friction and (d) acoustic.  77  4.5.3 Hardness Figures 4-28 and 4-29 shows the results of the hardness and modulus of elasticity measurement results for samples deposited at 200V using A16 alumina as filler material.  Figure 4-28. Hardness for coatings deposited at 200 volts with varying A16 alumina concentration in coating.  A s observed in the resulting plots, the hardness and modulus increased with the amount of alumina present in the matrix. This was expected, as alumina is a much harder material, which will hinder the movements during deformation and stiffen the overall structure. This was observed previously in the scratch testing.  78  Figure 4-29. Modulus of elasticity for coatings deposited at 200V with varying concentrations of A16 alumina in coatings.  4.5.4 Abrasion Test The abrasion/wear resistance tests performed were a means of comparing the effects of concentrations of A16 alumina and deposition voltages. Coatings deposited above 200 volts were not subjected to the tests, due to the irregularities observed, due to rupture and presence of exposed metal. The results of the abrasion test are shown in Table 4-1  as a comparison of penetration  depth  due to the  abrasive  agent,  concentration of alumina in coating, deposition voltage and thickness of the coatings.  79  Table 4-1. Thickness and wear penetration depth of the coatings for various deposition voltages and A16 alumina concentrations.  100V  150V  200V  Concentration (vol%) 0.00 12.50 22.40 0.00 12.50 22.40 0.00 12.50 22.40  Thickness (urn) 18.00 24.00 »50 22.00 25.00 »50 24.00 45.00 »50  Penetration Depth (urn) 24.15 19.03 29.78 24.15 21.10 29.49 24.50 22.30 29.13  The results show that increasing alumina content from 0 to 12.5 vol% increased abrasion resistance by decreasing the penetration depth of the abrasive agent. However, there was an increase in depth of penetration once the alumina content was increased from 12.5 to 22.4 vol%. The results also showed that increasing deposition voltage results in greater penetration depth for 12.5 vol% coating. However, at zero percent alumina, the wear depth was approximately equal for different voltages. The same could be concluded for samples with high concentration (22.4 vol%) of alumina. For zero vol% and 22.4 vol% alumina the coating failed due to penetration to the substrate, because of the limited thickness when the coating was deposited at low voltages. For the high concentration of alumina, the coating was chipped off rather than eroded [40]. The wear volume rate has been calculated for the samples where coating had not been penetrated (12.5 vol% A16 in coating) and is shown in Figure 4-30.  80  0.0014 W 0.0012 c j = 0.0010 £  0.0008  >  0.0004  o E 0.0006 o (0  ^  0.0002 0.0000 100  150  200  Deposition Voltage (V) Figure 4-30. Wear rate of 12.5 vol% A16 alumina samples where coating was not penetrated.  81  X-Profile 10  -  I  '•»'  0  rr  '"~*  j l»  i  (urn) -30  1  1•'  I  0  1  i  * 1 u  1  1  (mm)  1' 1  I 1  1  1.2  2.4  Y-Profile 1  10  :  I-  .  1  0 i  ^.  I ijjy J  (urn)  i! l  -30  0  (mm)  II  1  3  1  1.0  2.0  Figure 4-31. Profilometry wear bead shape results for 150 volts and 12.5 vol% A16 alumina coating.  82  X-Profile  0 (mm)  1.2  2.4  Y-Profile  Figure 4-32. Profilometry wear bead shape results for 150 volts and 0 vol% A16 alumina coating.  83  X-Profile  0 (mm)  1.2  2.4  Y-Profile 10  i  0  . . ' I - :  1  - -  - - -...  .  (H ) -30  J  m  '  0 (mm)  1.0  !  2.0  Figure 4-33. Profilometry wear bead shape results for 150 volts and 22.4 vol% A16 alumina coating.  84  The wear bead profile of the samples coating 12.5 vol% A16 alumina is shown in Figure 4-31. The shape of bead follows that of the rotating steel ball. The shape of other 0 vol% and 22.4 vol% (Figure 4-32 and 4-33) showed a square bead profile as the coating was penetrated.  4.5.5  Filler Effects The alumina particles have shown to improve the coating, hardness, scratch  resistance and microscale abrasion resistance. If one was to take into account the elastic modulus of each constituent, a significant difference in magnitude is seen for the matrix (E =5.21 G P a as measured) and the reinforcing particles (E =~370 G P a ) . By m  f  the use of the rule of mixture calculation, the elastic modulus of the overall composite E can be altered significantly, depending on the volume fraction V of the reinforcing r  phase.  E = E V +E,(l-V ) m  m  (4-2)  m  Eq. (4-3) predicts the modulus of the composite by assuming the particles to be homogeneously dispersed and of cubic geometry [36].  E_ »  E  E +(E -E )V l m  r  m  r  E (E -E )V kl-V h m+  r  m  r  r  If all variables are substituted into the Eqs.4-2 and 4-3 and plotted against actual data, in Figure 4-34, the results show a linear trend for all cases. The calculated the data using Eq. 4-2 shows rapid increase in modulus of the composite with small amounts of reinforcement addition, while the actual data shows slow increase in modulus with a 85  small volume of fillers. Eq. 4-3 on the other hands was seen to follow the same trend as the measured data. The difference is accounted for by the fact that the Eq. 4-2 does not include particle size and shape influence on the composite, or the distribution of the particles. From the S E M images, it can be seen that the dispersion was not perfect throughout the coating, but rather random. Also, the shapes of the fillers were not of cubic geometry, but rather a polygon-like and without uniform arrangement, as assumed when calculated. Therefore, the elastic modulus calculated using Eq. 4-3 can be used to predict the modulus of elasticity of a spherically filled E-coat. On the other hand, the data obtained during testing  have shown that the alumina filler enhanced the  mechanical properties only up to a certain limit.  60  n  1  Alumina in Coating (vol%) Figure 4-34. Elastic modulus prediction of the composite compared to actual data measured.  A closer look at the interface showed a better explanation of the deposition process. In order to be able to observe the particle deposition and compatibility at the 86  interface, a thin layer of alumina was deposited on a stainless steel substrate, and Ecoat (pure paint) was then deposited onto the alumina directly, creating a two layered structure. This method allowed for a larger interface to be studied. Figure 4-35 shows the S E M image of the fracture section of such prepared layered coating. The alumina Ecoat interface can be seen to have full coverage with no observed delamination or voids present at the interface (see Figure 4-36). The epoxy adhered to the alumina particles and adapted to the substrate morphology. Each individual particle that was exposed to the paint bath was observed to be covered completely by the resin, to create a homogenous interface between the two materials. The vol% which  gave the greatest scratch and wear  improvement  was  determined to be - 1 0 vol%. Subsequent increase in concentration of reinforcement saw a decrease in scratch and wear. At higher contents of filler, it has been shown that embrittlement effects due the large quantity of agglomerates with inhomogeneous shape and lower amount of ductile matrix available to bond the particles can occur [38]. These lead to degradation of the mechanical properties, as the particles are now weakly bonded by a much more brittle coating, with less energy absorbing capability, hence the higher depth of wear by the bead and the lower scratch critical force. Wetzel et al. [38] have also shown that a decrease in particle size and maintaining same volume of the fillers can significantly improve the material's behavior, by increasing the contact areas for energy/load transfer by the increased surface area present for interaction..  87  Figure 4-35. Interface S E M of E-coat deposited on A16 alumina layer.  Figure 4-36. High magnification S E M image of interface between E-coat and A16 alumina layer.  88  4.6 Process Modification and Optimization 4.6.1 Alumina in Composite E-coating The alumina content in the coating was found to be higher than the alumina added into the bath. The comparison tests were performed by calcining the scrape-off of the deposited coating and paint bath "cast" samples for each composition, as shown in Table 2-1. These were then compared to that of the alumina added into solution. The results compiled in Figure 4-37 show that the alumina content in the coating was significantly larger than that of the alumina in the bath. The bath cast and scrape-off samples were fairly similar, with differences ranging from 2 to 14%. The bath was also found to need stirring prior to deposition for A16 alumina concentrations higher than ~4.5 vol% in solution. Bath samples were taken 2.5 cm beneath the surface of the bath and 2.5 cm above the bottom of the bath every 5 minutes after stirring had stopped. The samples were dried and calcined and weighted for the alumina and epoxy in the coatings. Figure 4-38 and 4-39 show that A16 alumina in the paint bath does not significantly change the sedimentation of the resin micelles. At concentrations of alumina above 4.5 vol%, sedimentation occurs very fast as seen in Figure 4-40, where viscosity increases by as much as 25% when alumina content increases from 0 to 8 vol% in the bath. Also, a thick deposit was observed at the bottom of the beaker during testing when stirring was stopped.  89  35  0.63  1.26  1.89  2.53  5.05  7.58  A l u m i n a in Bath (vol%) Figure 4-37. Comparison of volume percent concentration of various sample forms.  0  5  10  15  20  25  30  time(min) Figure 4-38. Residue changes for 2.5 cm from the surface of the solution at 4.5 vol% A16 alumina added to paint bath.  90  103  0  5  10  15  20  25  30  time(min) Figure 4-39. Residue changes for 2.5 cm from the bottom of the solution at 4.5 vol% alumina added to paint bath.  30  Alumina in Suspesion (vol%) Figure 4-40. Viscosity changes with increase volume percent A16 alumina in solution.  91  The effects of alumina could also be seen when conductivity was measured in the composite  bath. Section 4.4.1  showed that alumina significantly  increased  conductivity of the bath, followed by a decrease. It was reported that when the alumina suspension was initially added to the Dl water, the concentration of ions in solution significantly increased [35]. The alumina suspension was found to be conductive prior to being added into the paint bath (1000 |xS/cm and pH=9.8). This value of pH would lead to an initial negative charge on the surface of the A16 alumina [35]. However, the paint bath was measured to be of pH= 4.8. This leads to a charge neutralization of the alumina surface when added to the paint bath, decreasing the amount of charged particles in the bath. At the same time the alumina particles will take on a positive surface charge due to the low pH (< 6), and therefore electrophoretic transport to the cathode simultaneously with the resin micelles is possible to create a composite E-coat. Conductivity of the 'as received' paint bath was also observed to decrease after each successive deposition (Figure 4-41), as some charged particles are consumed with each deposition. The presence of the filler phase influenced the deposited coating by increasing the thickness (Figure 4-19) and decreasing the overall dielectric strength of the coating (Figure 4-18). Other key changes observed were the induction time (IT), the time required for the hydroxyl ions to reach the critical concentration, COH-, in order for deposition to begin. The recorded current density versus time data from the deposition process clearly showed a steady shift of the peaks to the right of the time axis, depicting an increase in induction time with increasing alumina in the bath (Figure 4-43). Induction time is read at peak current density in Figure 4-42. The deposition appears to take place in the first second upon current flow, where conductivity and current flow are the 92  highest. After that the film resistance increases with increasing thickness of the coating a decrease in current is observed.  1200  E O  CO  ™  1100  c o o 1000  0  5  10  15  Number of Depositions Figure 4-41. Conductivity changes for as received paint bath after each subsequent deposition at 200V.  For higher alumina content (22 vol% and 30 vol%) the presence of a second peak can be seen at ~ 1 sec. This occurs due to lack of full coverage derived from film rupture and trapped hydrogen gas creating pockets and pinholes. The electrodeposition has been hypothesized to continue through the defective/porous film [26].  93  -—500 ]  time (sec) Figure 4-42. Current Density of deposition versus time with increasing A16 alumina concentration in composite E-coatings.  Usually these defects were healed during the low viscosity/high mobility stage of heat treatment prior to epoxy crosslinking. However, with high concentrations of alumina present, the mobility was hindered, and the deposited structure remains with all visible defects. The alumina has influences in four key processing factors: coating thickness, deposition induction time, volume of epoxy and coulombic yield, according to E q . 4-1. The conductivity of epoxy has been found to be ~10" cm" ohm" [19] and of alumina ~10" cm" ohm" [50]. Based on E q . 4-1, the 8  1  1  7  1  1  conductivity of the coating using the results from the 0 vol% deposition's experimental data, gives a calculated conductivity of 5.75 x 10~ cm" ohm" for the E-coat only film. 8  1  1  The induction time with increasing alumina content is shown in Figure 4-43: there is a linear relationship with increasing alumina. At a constant deposition voltage of 200 94  volts, Figure 4-44 shows the thickness calculated using E q . 4-1, demonstrating the same trend of increase when compared to the actual thickness, knowing that the deposition voltage and conductivity of the bath remained constant. Therefore the alumina content in the film was a major contributor the increasing the thickness of the film. Deposition efficiency (Figure 4-45), which takes into account the amount of mass deposited per coulomb consumed, and initial current density taken at the induction time were both altered by the presence of alumina in the bath. Alumina also influences other deposition properties such as deposition yield (Figure 4-46), and initial current density (Figure 4-47). The deposition yield, which is mass deposited per area coated, increased linearly with the increments of alumina A16.  0.17 o 0 d)  0.16 0.15 -  E 0.14 1— 0.13  -  c  0.12 <• +-» O 0.11 ' 3 "O 0.1 C 0.09 -  o  2  4  6  8  10  12  14  Alumina A16 in Coating (vol%) Figure 4-43. Alumina A16 concentration influence on induction time during deposition of film.  95  50  20 -I  ,  ,  ,  1  ,  ,  1  0  2  4  6  8  10  12  14  Alumina in Coating (vol%) Figure 4-44. Calculated film thickness incorporating influences contributed by A16 alumina compared to actual measured thickness.  250  O  Q.  0  5  10  15  20  25  Alumina A16 in Coating (vol %) Figure 4-45. Influence of alumina A16 on the deposition efficiency.  96  40  5  10  15  20  25  Alumina A16 in Coating (vol %) Figure 4-46. Deposition yield with increase alumina A16 in coating.  Figure 4-47. Influence of alumina A16 on the current density applied during deposition.  97  4.7 Alumina Particle Size Effects The particle size of the reinforcement phase had a significant effect of decreasing the scratch resistance of the composite. The scratch test performed for the A16 composite baths were repeated on A14 composite films with concentration of ~10 vol% alumina in coating, and the results showed a parabolic trend with increasing voltage similarly to that of the A16 alumina; however higher values were obtained with A16 in the bath (Figure 4-48).  In was observed that the increased particle size (0.3 um to 3  l^m) has lowered the force by up to 50%. This was likely due to the decrease in contact surface area between the reinforcement phase and matrix as observed by the crosssection S E M (Figure 4-13), thus creating a weakly bonded structure with fewer physical obstacles in the path of the stylus.  60  25 20 -| 100  '•  1  ,  1  150  200  250  r300  Deposition Voltage (V) Figure 4-48. Scratch test comparison of A16 and A14 alumina in coating deposited at various voltages.  98  The results agree with the study done by Sato [39], which showed a linear decrease in shear modulus with increasing particle size of filler. Also, the presence of larger particles makes the coating behave heterogeneously against abrasive wear, rather than a more homogeneous coating when fine particles were used [40]. The processing parameters have also changed significantly when using A14 particles. The induction time was observed to be erratic, with a linear trend as best fit, while the A16 demonstrated a better fit with gradual decrease in induction time with increasing voltages, as shown in Figure 4-49.  0.15  100  150  200  250  300  Deposition Voltage (V) Figure 4-49. Induction time difference between A16 and A14 alumina at various deposition voltages.  The thicknesses of coating was also affected, by different particle size of alumina as shown in Figure 4-50; the  resulting thickness for A14 containing  E-coats was  approximately half as thin when compared to that of A16 reinforced coating. If the trend 99  was taken into account, the A16 samples had a much larger growth rate than that of A14 filled E-coats. The thickness of the A16 samples approximately double the A14 coated samples at the various deposition voltages.  Figure 4-50. Thickness variation when using A14 compared to A16 at various deposition voltages.  The deposition yield (Figure 4-51) or mass per area covered, was also different, as predicted by the difference in coating thickness. It can be seen that both A16 and A14 alumina in E-coat samples resulted in linear trend with increasing voltage. However, the A16 filled coating showed a much higher deposition yield than that of the A14 filled coatings.  100  Deposition Voltage (volts) Figure 4-51. Deposition yield comparison between A16 and A14 reinforced coatings.  The observation can also be confirmed for the average current density (Figure 4-52) (per area covered). The two characteristics were related, i.e., if more material was to be coated using A14 fillers, a greater amount of current would be required.  It was  observed that A16 filled coatings had higher average current density than that of A14 filled coatings.  101  <r 70 E  <  0  -J  100  1  1  1  1  150  200  250  300  Deposition Voltage (V) Figure 4-52. Comparison of the current density of deposited A16 and A14 coatings.  Other  interesting  deposition  characteristics  observed  were  the  electrical  efficiency at IT (Figure 4-53) (defined as coulombs consumed per area covered) and the deposition efficiency (Figure 4-54) (defined as mass deposited per coulomb consumed). The electrical efficiency was observed to be greater in the A14 deposited samples, where a lesser amount of charge was required to deposit on an equal amount of area. However, it has been previously mentioned that the thickness of coatings containing A14 was smaller and therefore less charges were consumed per area coverage. This was more clearly shown in terms of overall deposition efficiency (Figure 4-54).  102  s  o  100  150  200  250  300  Deposition Voltage (V) Figure 4-53. Coulombs consumed per area covered using A 1 6 and A14 particles at varying deposition voltages.  O 95  100  150  200  250  300  Deposition Voltage (V) Figure 4-54. Deposition efficiency of the overall process for A16 and A 1 4 reinforced coatings.  103  The A14 deposited on average far more material per coulomb consumed, but produced a thinner coating. For the A16 samples, the efficiency decreases almost linearly as more material was deposited, but required more charge to initiate the process. The heterogeneity  of the A14 filled E-coatings was confirmed using the  breakdown voltage test (Figure 4-55).  A variability of ~50% from sampling point to  sampling point was significantly larger than that of the A16 samples, which saw an average difference of ~10%.  >  5  100  150  200  250  300  Deposition Voltages (V) Figure 4-55. Breakdown voltage comparison between A14 and A16 coated samples.  This allowed the formulation of the conclusion that the A14 coated samples were not homogeneously dispersed and a different particle deposition mechanism was present for the larger reinforcement  alumina. A similar result was observed when the  thicknesses of the coatings were taken into account. The dielectric strength (Figure 456) increased with increasing deposition voltage, while the A16 reinforced samples saw 104  a decrease in dielectric strength.  The A14 samples, approached that of the 'as  received' or 0 vol% sample deposited at 200 volts, which gave a result of - 8 0 KV/mm, but the results varied depending on the location of the electrode.  120  n  1  • A16 0  S  • A14  o 100  150  200  250  Deposition Voltage (V)  300  Figure 4-56. Dielectric Strength differences between A16 and A14 coated samples.  4.8 Electrochemical Behaviour of Composite E-coats The corrosion behaviour of the coatings was tested as a means of confirming that the film's application as a barrier coating properties had not been altered. E-coat's main purpose was to create a physical barrier to prevent electrochemical interaction between the environment and the metallic substrate. The addition of ceramic particles, which are generally considered to be inert materials, therefore a prime candidates to be used in corrosive electrolytes, clearly showed the need of an electrochemical study of the properties of the composite E-coat. Figure 4-57 shows the resulting electrochemical  105  behaviour of the composite coatings on the stainless steel substrate in an accelerated corrosion test under 1N NaCI electrolyte condition.  7.0E-01 t= 43(im  6.0E-01  t= 26um  5.0E-01  5vol%  10vol%  4.0E-01  / Ovol%  3.0E-01  c  2.0E-01  o  1.0E-01  o  t= 24um  0.0E+00 -1.0E-01 -2.0E-01 -10  -9  -8  -7  -6  -5  Current (Log(A)) Figure 4-57. Electrochemical behaviour of 5 vol% and 10 vol% A16 alumina filler composite coating compared with to 'as received' (0 vol%) coating. A s seen from the results, the addition of alumina shifts the polarization curves left (more negative) showing improvements in corrosion behaviour by a factor of - 1 0 0 . However, the influence of alumina concentration when increased from ~5 vol% to ~10 vol% does not show any changes either in corrosion potential or corrosion current behavior. It was noted that thickness of the samples used varied with increased alumina concentration, as shown in Figure 4-18. The deposition voltage (200 volts) was constant for all samples. On the other hand, it was clear that increasing the concentration from 5 vol% to 10 vol% alumina did not influence the polarization behaviour of the coating. This could be due to the partial breakdown of the coating during deposition as discussed previously in Section 4.2, leading to the formation of large quantities of localized micro 106  pores lowering the barrier resistance. A detailed calculation on the efficiency of the composite coating when compared to the 0 vol% can be viewed in Appendix II.  107  Chapter 5 Summary and Conclusions  E-coats were deposited on stainless steel substrates while varying concentration and particle size of calcined alumina (a-AI 0 ) filler in the bath. The process created a 2  3  composite E-coating, and experiments were carried out to investigate the mechanical properties (adhesion, scratch resistance, and hardness) of the coatings. The bath properties were investigated in terms of pH and conductivity, in order to determine the modification to the deposition mechanism of the composite bath. The most consistent composite E-coatings were then subjected to a corrosion test using a 1N NaCI electrolyte, to investigate the electrochemical behaviour of the newly deposited coatings, as compared to that of the 'as received' coatings with no ceramic filler. The addition of alumina into the epoxy coatings demonstrated improvements in all mechanical tests performed, including scratch (pencil and diamond stylus), hardness, and abrasive/wear. Adhesion tests were inconclusive, as the supplied bonding compound failed for all compositions before detaching any film. The most consistent coating was determined to be deposited at 200 volts from bath containing ~4.5 vol% A16 alumina. It was observed that outside these settings, the coating performance deteriorated during the deposition and testing. For samples conducted below these setting, the coating was significantly thinner and mechanically weaker, however, good finish and no defects were observed. The conductivity and pH of the bath were significantly altered upon addition of alumina, which subsequently changed the coating mechanism. In instances where large particle fillers (A14, 3-5^m) were used, clusters of filler were clearly observed. These heterogeneous coatings decreasing in scratch resistance by more than half, than that of 108  finer sized filler. However, for the finer powder (A16, 0.3-0.5p.m) more homogeneously dispersed filler was observed in the film resulting in increases in scratch resistance (1000%), increased wear resistance, increased hardness (45%) and stiffness (25%). Coating thickness was also increased by two hundred percent with the addition of alumina filler. The corrosion resistance of the composite coatings was observed to increase the corrosion current (/ rr) by up to one hundred times with the addition of A16 alumina CO  fillers up to 5 vol% in coating. Alumina increased the coating thickness, effectively creating a better barrier protection coating. However, when the concentration of filler in coating was above 5 vol%, the coating thickness doubled, but no increased in corrosion resistance was observed. The increase in A16 alumina filler beyond 10 vol% created defects by hindering the ability of the polymer to self-heal defects (pinholes and ruptures) during curing, which negated the improvements previously observed at 5 vol%. Composite  sensor films  were  also fabricated,  and  they  showed  basic  piezoelectric effects such as resonance frequency and polarization, without the need for poling stage after deposition (Appendix III). It is hypothesized that this was possible due to electric field available to polarize the PZT particles during deposition of the composite film.  109  Chapter 6 Recommended Future Work I. There are several fillers which could be studied, including the possibility of using modified positively charged reinforced particles in solution, to enhance dispersion and simultaneous deposition, thus decreasing the possibility of deposition  and  cluster  deposition.  Other  particle  sizes  should  preferential also  be  investigated, with the emphasis on nano-sized particles. Also, other fillers such as metallic or soft polymer particles should be used to compare mechanical properties of the composite coatings. For example, entrapment of iron (or iron oxide) particles in the composite E-coating would allow deposition of magnetic films on non-magnetic substrates.  II. A s mentioned previously, high concentration of alumina fillers tended to hinder the mobility of the epoxy during curing stages, hence the availability of other low viscosity polymers should be investigated, in order to allow higher concentration of filler with fewer defects.  III.  For the case of sensor coatings (Appendix III), larger particle size deposition should be investigated, as it has been shown that larger particle size and thicker films are required for greater sensitivity. However, the P Z T powder is a high density material (7.65 g/cc), and a method of maintaining dispersion in the suspension must be investigated, along with a more conductive coating to deposit a thicker coating.  110  Chapter 7 References 1.  Last First.,"BMW Sustainable Value Report 2002".  2.  U F C Corporation Electrocoating Equipment/'Electrocoat Info".  3.  John W . Halloran, "Calcination", Volume 4, Ceramic and Glasses, Engineering Materials Handbooks, ASM International, 109-114.  4.  U. Rammelt, G. Reinhard, Prog. Org. Coat. 21 (1992) 205.  5.  T. Monetta, F. Bellucci, L. Nicodemo, L. Nicolais, Prog. Org.Coat. 21 (1993) 353.  6.  P.L. Bonora, F. Deflorian. LFedrizzi, Electrochim. Acta, 14 (1996) 1073.  7.  F. Beck, J.O.M. Bockris, Comprehensive  Treatise of Electrochemistry,  Vol.  2, Plenum, New York, 1981, p. 537. 8.  P . E . Pierce, J. Coat. Technol. 53 (1981) 52.  9.  M. Kubo, R. Tagaki and H. Ushieda, Denki Kagaku, 32 (1974) 650.  10.  D.M. Drazic, N.M. Acamovic and O.D. Stojanovic, J. Coat. Techonol.,Q3 (1991)25.  11.  P2IRIS Painting, "E-coat Paint and Material Selection"  12.  N.M. Acamovic, D.M. Drazic, V . B . Miskovic-Stankovic, Prog. Org. Coat. 25 (1995) 293-307  13.  J . J . Oravitz, P P G Industries Inc., Springdale, P a , "Electrocoating"  14.  J.I. Duffy, Noyes Data Corporation, New Jersey,Electrodeposition Processes,  Equipment, and Compositions  1982  15.  K. Adamsons, Prog. Polym. Sci. 25 (2000) 1363-1409  16.  C M . Reddy, R.S. Gaston, C M . Weikart, H.K. Yasuda, Prog. Org. Coat. 33 (1998) 225-231 111  17.  B. Gruss, Pretreatment & Process Inc., Niceville, Fla."Cleaning and surface preparation",  Preatreatment.  18.  J . E . O. Mayne, Br. Corros. J. 5, 106 (1970).  19.  H. Leidheiser, Prog. Org. Coat. 7, 79 (1979).  20.  G . W. Walter, Corros. Sci. 26, 27 (1986).  21.  G . Reinhard, Prog. Org. Coat. 18, 123 (1990).  22.  F. Deflorian, L. Fedrizzi and P. L. Bonora, Corrosion 50, 113 (1994).  23.  J . Parks and H. Leidheiser, Ind. Engng Chem. Prod. Res. Dev. 25, 1 (1986).  24.  F. Bellucci and L. Nicodemo, Corrosion 49, 235 (1993).  25.  E. P. M. van Westing, G. M. Ferrari and J . H. W. de Wit, Corros. Sci. 36, 957 (1994).  26.  V. B. Miskovic-Stankovic, J . Serb. Chem. S o c 67 (5) 2002  27.  F. Mansfeld, M. W. Kendig and S. Tsai, Corrosion 38, 478 (1982).  28.  G . W. Walter, Corros. Sci. 26, 681 (1986).  29.  L. Beaunier. I. Epelboin, J . C. Lestrade and H. Takenouti, Surf. Technol. 4, 237 (1976).  30.  J . D. Scantlebury and K. N. Ho, J. Oil Colour Chem. Assoc. 62, 89 (1979)  31.  V. B. Miskovic-Stankovic, D. M. Drazic and Z. Kacarevic-popovic, Corros. Sci. 38, 9 (1996) 1513-1523  32.  V. B. Miskovic-Stankovic, D. M. Drazic and M. J . Teodorovic, Corros. Sci. 3 7 , 2 (1995)241-252  33.  D. M Drazic and V. B. Miskovic-Stankovic, Corros. Sci. 30, 6/7 (1990) 575582  34.  P.H. Shipway and N.K. Ngao, Wear, 255 (2003) pp742-750 112  35.  I. Zhitomirsky, "Cathodic electro-deposition of ceramic and organocerarmic materials. Fundamental aspects", yAoVances in Colloidal and Interface Science, 97 (2002) 279-317  36.  R.M. Jones, Mechanics of Composite Materials (Scripta Book Company, Washington D.C.1975)  37.  E. Jacquet, F. Trivaudey, D. Varchon, "Calculation of the tranverse modulus of a unidirectional compsite material and of the modulus of an aggregate. Application of the rule of mixtures", Composite Science and Technology, 60 (2000) 345-350  38.  B. Wetzel, F. Haupert, M.Q. Zhang, "Epoxy nanocomposite with high mechanical and tribological performance", Composite Science and Technology, 63 (2003) 2055-2067  39.  K. Sato, " The Mechanical Properties of Filled Polymers", Progress in Organic Coatings, 4 (1976) 271-302  40.  I.M. Hutchings, Tribology: Friction and Wear of Engineering  Materials  (Edward Arnold, Great Britain, 1992) p. 160 41.  A P C International, Piezoelectric Ceramics: Principles and  Applications  (APC International, Ltd., Mackeyville P A , 2002) 42.  Y. Massiani, Progress in the Understanding  and Prevention of Corrosion (  Institute of Materials, London, 1993) p.179 43.  M.A. Moore, Materials in Engineering Applications  44.  A. Misra and I. Finnie, Wear65 (1981) 359-373  45.  R. Maines, Sensors 6, 26 (1989)  46.  R. Barrett and F. Wilcoxon, Sensors, 10, 16 (1993)  47.  R.E. Newham, MRS Bullentin, 22, 20 (1997)  1 (1978), 97-111  48.  H. Jaffe, and D. Berlincourt, P r o a , IEEE, 53,1372 (1965)  49.  D. Damjanovic and R.E. Newnham, J. Intell. Mater. Struct, 3, 190 (1992)  50.  G . H . Haertling, in Ceramic Materials for Electronics,  Processing,  Properties, and Applications, edited by R.C. Buchanan (Marcel Dekkar, New York, 1986). p. 165 51.  Y . X u , Ferroelectric Materials and Their Applications  (North-Holland,  Amsterdam, The Netherlands, 1991), p. 217, 277 52.  B. Jaffe, W.R. Cook, and H. Jaffe, Piezoelectric  Ceramics (Academic  Press, New York, 1971), p. 148 53.  D. Berlincourt, J. Acoust. Soc. Am., 70, 1586 (1981)  54.  R.C. Turner, P.A. Fuirer, R.E. Newnham, and T.R. Shrout, Appl. Acoustic, 41, 299-324 (1994)  55.  T. Ikeda, Fundamentals  of Piezoelectricity  (Oxford University Press, New  York, NY, 1990), p. 209 56.  J . F . Tressler, S. Alkoy and R.E. Newnham, J. Electroceramics,  2:4, 257  (1998) 57.  O.B. Wilson, Introduction to the Theory and Design of Sonar  Transducers  (Peninsula Publishing, Los Altos, C A , 1988), p. 65 58.  S. Alkoy, A. Dogan, A - C . Hladky. P. Langlet, J.K. Cochran, and R.E. Newnham, IEEE Trans. UFFC, 44, 1067 (1997)  59.  K. Rittenmyer, T. Shrout, W.A. Schulze, R.E. Newnham, Ferroelectrics, (1982) p. 189  60.  Safari, R.E. Newnham, L.E. Cross, W.A. Schulze, Ferroelectrics,  41  (1982) p. 197  114  41  61.  R.E. Newnham, D P . Skinner, and L.E. Cross, Mat. Res. Bull., 13, 525 (1978)  62.  T.R. Gururaja, A. Safari, R.E. Newnham, and L.E. Cross, in Ceramics: Properties, Devices and Applications,  Electronic  edited by L.M. Levinson  (Marcel Dekkar, New York, 1988), p. 92 63.  R.C. Twine, Adv. Mater. 4, (1992) p. 819  64.  P P G Formulation Instruction Powercron® 590  65. 66.  http://www.quadgroupinc.com/RomulusManual.pdf R. Barbato, R. Boi and R. Ragazzini, "Determination of micro-indentation hardness of organic coatings", QUALITAL-/Assoc/'az/one  Certificazione  Industriale Alluminio (Novara, Italy) 67.  ASM Handbook, Volume 13A: Corrosion: Fundamentals, Protection^  Testing, and  A S M International 1987) Edited by Stephen D. Cramer and  Bernard S. Covino, Jr 68.  Y - M . Chiang, D.P. Birnie III, and W.D. Kingery, Physical  Ceramics:  Principles for Ceramic Science and Engineering (John Wiley & Sons, Inc. 1997 New York) p. 466-477 69.  C. Cassignol, M. Cavarero, A. Boudet, A. Ricard, "Microstructureconductivity relationship in conducting polypyrrole/epoxy composites" PolymerAO (1999) 1139-1151  70.  Y . P . Mamunya, V.V. Davydenko, P. Pissis, E.V. Lebedev," Electrical and thermal conductivity of polymers filled with metal powders" European Polymer Journal 2% (2002) 1887-1897  71.  Q. Yang and T. Troczynski, " Dispersion of Alumina and Silicon Carbide Powders in Alumina Sol" J. Am. Ceram. Soc, 82 [7] (1999) 1928-1930 115  APPENDIX I - Composite E-Coating Thickness The coating thickness parameters of the E-coat have been modified by the addition of a secondary phase into the system, which has higher conductivity than that of the epoxy. Therefore it is necessary to incorporate the changes to the deposition equation by including the secondary phase, alumina, in order to be able to predict the thickness of the composite E-coat. The E-coating thickness can predicted from E q . A-1 [8]:  1/2  (A-1)  b= thickness (cm) c= coulombic yield (cm /C) K= conductivity of film (1/ cm-ohm) U= deposition voltage (volts) t,= induction time (s) 3  Using the rule of mixture to take into account the conductivity difference between epoxy and alumina to get the overall conductivity of the coating/film, K C ]  K  = V Ke + V | K |  c  e  A  A  (A-2)  V = volume fraction epoxy VAI= volume fraction alumina Ke= conductivity of epoxy (~10" cm" ohm" ) KAI= conductivity of alumina (~10" cm~ ohm~ ) Kc= conductivity of coating E  8  1  7  1  1  1  Further grouping of constants for equation A-1 b = ( 2 L 0 c ( f - f i ) ( ) 1/2 1/2  1/2  1/2  K  (A-3)  Substitute all constants with the constant A A= (2L0 1/2 b=Ac  1 / 2  (f-fi)  1 / 2  ( K ) 1/2  (A-4) (A-5)  116  Substituting equation A - 2 into A - 5 b = A(V  e  Ke +  c"  2  VAIKAI)  (A-6)  trf*  (f-  Recalculating all terms based on epoxy V  A  i = 1 - V  (A-7)  E  Knowing that K M > K , E  K | = A  10K  (A-8)  e  Substituting A - 7 and A - 8 into A - 6 gives rise to the following b = A [ V K e + ( 1 - V ) 1 0 K e ] [t  -  V2  6  e  t]  (A-9)  [cf  V2  2  Further simplification of A - 9 b=A[K (10-9V )] e  e  1 / 2  [f-fi]  1 / 2  [c]  (A-10)  1 / 2  Let  b =  B =  AK  T=  t-t]  B[(10-9V ) e  1  /  (A-11)  2  6  (A-12) (T) (c)]  (A-13)  1/2  Therefore the thickness was influenced by the following parameters: V , T a n d c e  To verify the parameter K f r o m literature, conductivity was calculated using experiments e  results: 200 volts deposition voltage,  0.087  c m / C coulombic yield, 0.024 cm thickness, 3  0.12 seconds induction time and 3 seconds total time.  = 5.75xl0" cw" o/zm~ 8  1  l  (2cU)~2(t-t,)  (A-14)  the calculated conductivity of the 'as received' coating is in the range of that given in the literature. 117  APPENDIX II- Composite E-Coating Corrosion Protection Efficiency Protection Efficiency, P, can be expressed as: P(%) = 100(1 - - ^ - ) I  where /° rrand CO  j  corr  (A-15)  corr  denote the bare and coated substrate [42]. In the case of the  presented study, rather than using the bare stainless steel substrate for  i°com  the 'as  received' corrosion current will be used.  Figure A-1 shows the measurement of the tafel slope in order to obtain the corrosion current from the polarization curves obtained. The resulting data are compiled in Table A-1  -1.6E-01  -I  V  -10  -9  J  ,  1  -8  -7  *  1  r-  -6  -5  Log(A) Figure A-1. Polarization data and corrosion current calculation.  118  Table A - 1 . Data obtained from the polarization curves Sample A s received 5 vol% 10 vol%  Log(A) -6.50E+00 -8.75E+00 -8.95E+00  'corr  (A)  3.16E-07 1.78E-09 1.12E-09  P(%) 99.4 99.6  APPENDIX III- A New Direction: Piezoelectric E-Coatings This study has shown the ability to coat composite E-coat, simultaneously depositing the resin matrix and ceramic reinforcement phase on the substrate, with full coverage of the cathode substrate. The possibilities of using other  reinforcement  materials are significant. A preliminary study that has branched out of this project was the use of the E-coat process to create functional coatings. One variant consisted of creating an electro-mechanical transducer coating with the possibility of converting vibration into electrical energy (or in essence a vibration damping coating), which could be applicable in automobiles body, aerospace components and electronic equipment such as speakers. The coating methodology was similar to that of the alumina, but the high dielectric properties of the resin and the presence of an electric field during deposition would allow the use P Z T embedded into the resin matrix to create an in-situ sensor where  no  extra  poling  steps  would  be  required,  as  in  the  conventional  actuator/transducer processing [41]. The drawback to the process was the inability to apply a thick coating without causing film rupture. However, other forms of aqueous polymer with lower dielectric strength could be used to increase thickness of the piezoceramic-polymer coating. Groundwork has been performed, and promising results have been observed. The coated sample contained - 1 2 vol% P Z T (BM532 from SensorTech, Canada) co-deposited at 150V with commercial E-coat base. The coating shows basic piezoelectric properties when impedance and polarization were tested at SensorTech Inc. The resonance and anti-resonance frequency were to be 3.90x10 Hz 7  and 3.8x10  7  Hz respectively (Figure A-2).  These can then be converted into  electromechanical coupling factor (k ) for a plate [41] of 0.0657. The sample was 31  120  polarized at 130 volts in order to observe if previous poling had occurred during deposition of the coating. If no poling was present no hysteresis loop would have been obtain. The results show a characteristic hysteresis loop (Figure A-3) showing evidence of a poling during the deposition process. The saturation point, P , was found to be s  1.00x10" C / m and the remanent polarization, P , was 4.46x10" C / m . It should be 3  2  4  2  r  noted that  0*33  measurements could not to be obtained, due to the relative thin coating.  More work is required in order to fully understand the process of manufacturing transducers using this technique.  2.E-08 1.E-08  fn  § 5.E-09 C re  T3 0.E+00 Q.  E-5.E-09 -1.E-08 -2.E-08  f  Tl  0.E+00 1.E+07 2.E+07 3.E+07 4.E+07 5.E+07 6.E+07 7.E+07 8.E+07  Frequency (Hz) Figure A-2. Impedance measurement for BM532 P Z T filled polymer E-coat.  121  CM  E o  c o +-» (0 N  • mmm  JS o  1.E-03 1. E-03 8.E-04 6.E-04 4.E-04 2. E-04 O.E+00 -2.E-04 ^.E-04 -6.E-04 -8.E-04  ^—-Start  -3  -  2  -  1  0  1  2  Field Potential(MV/m) Figure A-3. Polarization curve obtained at 130 volts poling for B M 532 PZT-polymer composite E-coat.  122  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0079064/manifest

Comment

Related Items