<%BANNER%>

Engineering Polyethylene Surface Energy for Adhesion

Permanent Link: http://ufdc.ufl.edu/UFE0042455/00001

Material Information

Title: Engineering Polyethylene Surface Energy for Adhesion
Physical Description: 1 online resource (144 p.)
Language: english
Creator: JACKSON,DAVID C
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: GRAFTING -- POLYETHYLENE -- SURFACE -- ZWITTERIONIC
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: One would be hard pressed to live an entire day without coming into contact with polyethylene. It is in the plumbing, automobiles, refrigerators and homes of almost everyone in modern society. Worldwide, an estimated 80 million metric tons are produced each year. While polyethylene?s versatile properties make it commonplace, its inert surface limits its effectiveness in many applications. From relatively trivial issues such as printing to the life and death matter of ballistic protection, the poor surface adhesion to polyethylene is a major issue. It is by altering this surface that the full potential of the material can be realized. Sometimes this can be done using a simple flame or acid treatment. In other cases, more advanced surface modification methods are required. This work is focused on modification of polyethylene surfaces, by a grafting process, to control adhesion. Grafted samples were characterized using attenuated total reflection Fourier transformed infrared spectroscopy (ATR-FTIR), sessile drop contact angle and x-ray photoelectron spectroscopy (XPS). ATR-FTIR was used quantify graft content and chemistry, such as carbonyl content. Contact angle measurements were used to calculate surface energy using the Owens?Wendt?Kaelble (OWK) and Lifshitz-van der Waals (LW) methods. XPS quantified the amount of new surface elements, such as sulfur or zinc. Results showed a significant increase of adhesion with epoxy of human osteoblasts (hOBs) and a significant decrease in adhesion of algae. Adhesion was enhanced between polyethylene and epoxy, e.g. fiber pull-out strength increased 15%, via a copolymer graft of acrylic acid and glycidyl methacrylate. A poly (glycidyl methacrylate) (PGMA) graft on polyethylene increased adhesion of gelatin, which supported a 106% raise in attachment hOBs. Adhesion to polyethylene was inhibited by use of either a poly (sulfobetaine methacrylate) (PSBMA) zwitterionic graft or poly (acrylic acid) (PAA) graft complexed with zinc. An assay with Ulva zoospores indicated 92% and 88% inhibition for the PSBMA and PAA respectively compared to polyethylene without graft.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by DAVID C JACKSON.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Brennan, Anthony B.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0042455:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042455/00001

Material Information

Title: Engineering Polyethylene Surface Energy for Adhesion
Physical Description: 1 online resource (144 p.)
Language: english
Creator: JACKSON,DAVID C
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: GRAFTING -- POLYETHYLENE -- SURFACE -- ZWITTERIONIC
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: One would be hard pressed to live an entire day without coming into contact with polyethylene. It is in the plumbing, automobiles, refrigerators and homes of almost everyone in modern society. Worldwide, an estimated 80 million metric tons are produced each year. While polyethylene?s versatile properties make it commonplace, its inert surface limits its effectiveness in many applications. From relatively trivial issues such as printing to the life and death matter of ballistic protection, the poor surface adhesion to polyethylene is a major issue. It is by altering this surface that the full potential of the material can be realized. Sometimes this can be done using a simple flame or acid treatment. In other cases, more advanced surface modification methods are required. This work is focused on modification of polyethylene surfaces, by a grafting process, to control adhesion. Grafted samples were characterized using attenuated total reflection Fourier transformed infrared spectroscopy (ATR-FTIR), sessile drop contact angle and x-ray photoelectron spectroscopy (XPS). ATR-FTIR was used quantify graft content and chemistry, such as carbonyl content. Contact angle measurements were used to calculate surface energy using the Owens?Wendt?Kaelble (OWK) and Lifshitz-van der Waals (LW) methods. XPS quantified the amount of new surface elements, such as sulfur or zinc. Results showed a significant increase of adhesion with epoxy of human osteoblasts (hOBs) and a significant decrease in adhesion of algae. Adhesion was enhanced between polyethylene and epoxy, e.g. fiber pull-out strength increased 15%, via a copolymer graft of acrylic acid and glycidyl methacrylate. A poly (glycidyl methacrylate) (PGMA) graft on polyethylene increased adhesion of gelatin, which supported a 106% raise in attachment hOBs. Adhesion to polyethylene was inhibited by use of either a poly (sulfobetaine methacrylate) (PSBMA) zwitterionic graft or poly (acrylic acid) (PAA) graft complexed with zinc. An assay with Ulva zoospores indicated 92% and 88% inhibition for the PSBMA and PAA respectively compared to polyethylene without graft.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by DAVID C JACKSON.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Brennan, Anthony B.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0042455:00001


This item has the following downloads:


Full Text

PAGE 1

1 ENGINEERING POLYETHYLENE SURFACE ENERGY FOR ADHESION By DAVID CARNABY JACKSON A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCT OR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011

PAGE 2

2 2011 David Carnaby Jackson

PAGE 3

3 For m y grandfathers Clark W. Carnaby Jr. and Ralph C. Jackson : two great men that I admire and emulate

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my advisor, Dr. Tony Brennan, for all his assistance. It was his help and support that made this possible. I would also like to thank my committee members: Dr. Christopher Batich, Dr. Laurie Gower, Dr. John Mecholsky and Dr. Stephen A. Miller for their assistance. Their th oughts and contributions have helped to make my work better. Other faculty and staff were extremely helpful in my time at Florida. Jennifer Wrighton was invaluable as a secretary and friend. Dr. Cliff Wilson and Dr. Leslie Wilson were indispensible in my early years as a graduate student, both for their assistance in the lab and their friendship Gary Scheiffele helped me establish proper infrared spectroscopy methods. Dr. Thomas Wright and Dr. Steven C. Ghivizzani were extremely valuable in helping me develop my ideas on orthopedic sutures. Dr. Josephine Allen made my cell studies possible. Dr. James Callow and Dr. Maureen Callow were essential for my work on marine fouling of polyethylene. I can say without exception that my all of fellow graduate st udents in the Brennan group have been dependable hard workers and good friends. Dr. James Schumacher, Dr. Iris Schumacher, Dr. Amin Elachchabi, and Kenneth Chung were all very helpful in my early years in the group. Julian Sheats, James Seliga, Joshua Low itz and Scott Cooper all contributed greatly to my work with fiber reinforced composites. Dr. Chelsea Magin, Dr. Christopher Long Angel Ejiasi Sean Royston and Kimberly Struck were extremely helpful in my work in marine fouling. Jiun Jeng Chen has been a vital contributor to my work on surface grafting, particularly in calculating surface energy. Joe Decker and Adwoa Baah Dwomoh have been ve ry friendly and helpful in their year here.

PAGE 5

5 In my more challenging times, my strength and confidence have been bu oyed by my family, my friends and my faith. My parents, Jim and Cathy Jackson, have always been a source of love and support. My brother Steve and his wife Ardi are wonderfully supportive and entertaining. My extended family, including the Carnabys, the Reichs, the Baseys, the Horgans, the Dickinsons, the Gilchrists and the Dereszynski s, have always been a great source of pride and joy. My friends, such as Brian and Stephanie Siebert, Micah Abresch, Eric Arvidson, Nathaniel Kish, Jeremy Geiger, Kyle Tay lor, Sean Bryan, Dave Krause, Liz Hanak and are exceptional people and I cherish th eir company I would also like to thank Pastors John Roth and Dan Prugh and the entire community at First Lutheran Church in Gainesville for the friendship and support they offered.

PAGE 6

6 TABLE OF CONTENTS Page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ .......... 10 LIST OF FIGU RES ................................ ................................ ................................ ........ 11 LIST OF ABBREVIATIONS ................................ ................................ ........................... 15 ABSTRACT ................................ ................................ ................................ ................... 17 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 19 Specific Aim 1: Increase Pullout Strength of Ultra High Molecular Weight Polyethylene (UHMWPE) Fibers by 15% or more via a poly (Acrylic Acid co Glycidyl Methacrylate) (PAAGMA) Graft ................................ .............................. 19 Specific Aim 2: Increase Adhesion of Human Osteoblasts (hOBs) to Polyethylene by 100% or More Using Covalently Bound Gelatin ........................ 20 Specific Aim 3: Reduce the Adhesion of Ulva Zoospores to Low Density Polyethylene Films by 90% or More Using Surface Grafts ................................ .. 20 2 BACKGROUND ................................ ................................ ................................ ...... 22 Adhesi on and Surface Energy ................................ ................................ ................ 22 Surface Spectroscopy ................................ ................................ ............................. 27 Applications of Adhesion to Polyethylene (PE) ................................ ....................... 27 Adhesion in Composites: Ultra High Molecular Weight Polyethylene Fibers .... 28 Adhesion of Cells: High Strength Orthopedic Sutures ................................ ..... 29 Anatomy of the Shoulder ................................ ................................ ............ 29 Surgical Repair ................................ ................................ .......................... 29 Adhesion Between Suture and Living Tissue ................................ ............. 29 Adhesion in the Marine Environment: Biofouling ................................ ............. 32 History of Marine Biofouling Coatings ................................ ........................ 33 No n toxic Anti fouling Surfaces ................................ ................................ .. 33 Surface Topography ................................ ................................ .................. 34 Reducing Fouling of Polyethylene via Surface Grafting ............................. 36 Surface Grafting to Polyethylene ................................ ................................ ............ 36 Plasma Treatments ................................ ................................ .......................... 37 Gamma Radiation Grafting ................................ ................................ ............... 37 Hydroperoxidation Grafting ................................ ................................ ............... 38 Ultraviolet (UV) Photografting ................................ ................................ ........... 38 Summary ................................ ................................ ................................ ................ 41

PAGE 7

7 3 INCREASED PULL OUT STRENGTH OF ULTRA HIGH MOLECULAR WEIGHT POLYETHYLENE FIBERS IN EPOXY MATRIX USING POLY (ACRYLIC ACID CO GLYCIDYL METHACRYLATE) SURFACE GRAFTS ........... 43 Introduction ................................ ................................ ................................ ............. 43 Materials ................................ ................................ ................................ ................. 47 UHMWPE Fiber ................................ ................................ ................................ 47 Mono mer and Initiator ................................ ................................ ....................... 47 Methods ................................ ................................ ................................ .................. 47 Fiber Cleaning ................................ ................................ ................................ .. 47 Fiber Grafting ................................ ................................ ................................ ... 48 Graft Characterization ................................ ................................ ...................... 49 Tensile Tests ................................ ................................ ................................ .... 50 Pull out tests ................................ ................................ ................................ ..... 51 Results ................................ ................................ ................................ .................... 53 Tensile Properties ................................ ................................ ............................ 54 Pull out Test Data ................................ ................................ ............................. 56 Discussion ................................ ................................ ................................ .............. 57 Summary ................................ ................................ ................................ ................ 58 4 INCREASED ADHESION OF GELATIN AND HUMAN OSTEOBLASTS ON TO POLYETHYLENE USING A GL YCIDYL METHACRYLATE SURFACE GRAFT .... 60 Introduction ................................ ................................ ................................ ............. 60 Methods ................................ ................................ ................................ .................. 62 Film and Monomer Preparation ................................ ................................ ........ 62 PGMA Grafting ................................ ................................ ................................ 62 Gelatin Grafting ................................ ................................ ................................ 63 Control Group s and Standards ................................ ................................ ......... 64 Gelatin Desorption ................................ ................................ ............................ 64 Surface Characterization ................................ ................................ .................. 65 Cell Adhesion Assays ................................ ................................ ....................... 66 Results ................................ ................................ ................................ .................... 68 Contact Angle ................................ ................................ ................................ ... 68 ATR FTIR ................................ ................................ ................................ ......... 69 Grafting of PGMA onto LDPE ................................ ................................ .... 69 Binding of Gelatin to LDPE ................................ ................................ ........ 71 Cell Adhesion ................................ ................................ ................................ ... 75 Discussion ................................ ................................ ................................ .............. 77 Conclusions ................................ ................................ ................................ ............ 78 5 REDUCTION OF ULVA SPORE ATTACHMENT ON TO POLYETHYLE NE USING POLY (SULFOBETAINE METHACRYLATE) SURFACE GRAFTS ............ 79 Introduction ................................ ................................ ................................ ............. 79 Methods ................................ ................................ ................................ .................. 81 Sample Preparation ................................ ................................ .......................... 81

PAGE 8

8 Preparing Monomer ................................ ................................ .......................... 83 Grafting Procedure ................................ ................................ ........................... 8 3 Experimental Controls ................................ ................................ ...................... 84 Material Characterization ................................ ................................ .................. 84 Sessile Drop Water Contact Angle ................................ ............................. 84 Dynamic Water Contact Angle ................................ ................................ ... 85 Measurement of Surface Energy ................................ ............................... 86 ATR FTIR ................................ ................................ ................................ ... 87 X ray Photoelectron Spectroscopy ................................ ............................. 87 Ulva Attachment ................................ ................................ ............................... 88 Results ................................ ................................ ................................ .................... 88 Sessile Drop Water Contact Angle ................................ ................................ ... 88 Dynamic Water Contact Angle ................................ ................................ ......... 89 Surface Energy Measurements ................................ ................................ ........ 89 ATR FTIR ................................ ................................ ................................ ......... 90 XPS ................................ ................................ ................................ .................. 92 Ulva Attachment ................................ ................................ ............................... 93 Discussion ................................ ................................ ................................ .............. 93 Graft Characterization ................................ ................................ ...................... 93 Ulva Attachment ................................ ................................ ............................... 94 Concl usions ................................ ................................ ................................ ............ 95 6 REDUCTION OF ULVA ATTACHMENT ONTO POLYETHYLENE SURFACES USING BOUND ZINC VIA AN ACRYLIC ACID SURFACE GRAFT ....................... 97 Introduction ................................ ................................ ................................ ............. 97 Methods ................................ ................................ ................................ .................. 98 Sample Preparation ................................ ................................ .......................... 98 Monomer Preparation ................................ ................................ ....................... 99 Grafting Procedure ................................ ................................ ........................... 99 Binding Zinc to the PAA Surface Graft ................................ ........................... 100 Experimental Controls ................................ ................................ .................... 100 Material Characterization ................................ ................................ ................ 101 Sessile Drop Water Contact Angle ................................ ........................... 101 Dy namic Water Contact Angle ................................ ................................ 101 Measurement of Surface Energy ................................ ............................. 102 ATR FTIR ................................ ................................ ................................ 102 XPS ................................ ................................ ................................ .......... 103 Ulva Attachment ................................ ................................ ............................. 103 Results ................................ ................................ ................................ .................. 104 Static Water Contact Angle ................................ ................................ ............ 104 Dynamic Water Contact Angle ................................ ................................ ....... 104 ATR FTIR ................................ ................................ ................................ ....... 108 XPS ................................ ................................ ................................ ................ 109 Ulva Attachment ................................ ................................ ............................. 110 Discussion ................................ ................................ ................................ ............ 111 Graft Characterization ................................ ................................ .................... 111

PAGE 9

9 Sessile Drop Water Contact Angle ................................ ........................... 111 Dynamic Contact Angle ................................ ................................ ............ 112 Surface Energy ................................ ................................ ........................ 112 ATR FTIR ................................ ................................ ................................ 112 XPS ................................ ................................ ................................ .......... 113 Characterization Summary ................................ ................................ ....... 113 Ulva Attachment ................................ ................................ ............................. 113 Conclusions ................................ ................................ ................................ .......... 114 7 FUTURE WORK ................................ ................................ ................................ ... 116 A dhesion to Polyethylene Fibers for Composites ................................ ................. 116 Optimizing Grafting Procedure ................................ ................................ ....... 116 Cross linking GMA using Ethylene Diamine ................................ ................... 116 Adhesion to Polyethylene in the Body ................................ ................................ ... 117 Grafting onto Commercial UHMWPE Sutures ................................ ................ 117 Examining Other Proteins and Growth Factors ................................ .............. 117 Stem Cell Differentiation ................................ ................................ ................. 118 Adhesion to Polyethylene in the Marine Environmen t ................................ ........... 118 Optimizing Graft Kinetics for LDPE PSBMA ................................ ................... 118 Examining the Effectiveness of LDPE PSBMA for Limiting Attachment of Other Marine O rganisms ................................ ................................ ............. 118 Examining Other Polymers for Binding Zinc ................................ ................... 119 Combining Graft with Surface Topography ................................ ..................... 119 8 CREATING PATTERNED POLYETHYLENE USING A MELT PRESSING PROCESS ................................ ................................ ................................ ............ 121 Introduction ................................ ................................ ................................ ........... 121 Methods ................................ ................................ ................................ ................ 123 Sample Production ................................ ................................ ......................... 123 Characterization ................................ ................................ ............................. 125 Results ................................ ................................ ................................ .................. 125 SEM ................................ ................................ ................................ ................ 125 Pattern Fidelity ................................ ................................ ............................... 127 Contact Angle ................................ ................................ ................................ 129 Discussion ................................ ................................ ................................ ............ 130 Summary ................................ ................................ ................................ .............. 130 LIST OF REFERENCES ................................ ................................ ............................. 132 B IOGRAPHICAL SKETCH ................................ ................................ .......................... 143

PAGE 10

10 LIST OF TABLES Table page 3 1 Concentration of AA and GMA monomers and BP photoinitiator in acetone for grafting solution ................................ ................................ ............................. 47 3.2 Tensile modulus, peak load and strain at break for untreated and 3x PAAGMA grafted Dyneema SK78 filaments. ................................ ................. 54 3 3 Comparison of percent pullou t for untreated and 3x PAAGMA grafted Dyneema SK78. ................................ ................................ ............................... 56 4 1 A summary of gelatin desorption treatments. ................................ ..................... 64 5 1 Physical constants for deionized water, diiodomethane and glycerol. These are the liquids used to calculate the energy of the surface[7]. ............................ 86 5 2 Results from dynamic contact angle for untreated LDPE and LDPE PSBMA. A sterisks indicate values where the grafted sample is statistically different from the untreated LDPE. ................................ ................................ ................... 89 5 3 Contact angle data for deionized water (DI) glycerol (GL) and diiodomethane (DM)on bot h LDPE and LDPE PSBMA ................................ .............................. 90 5 4 Surface energy of LDPE and LDPE PSBMA grafted samples as determined using the OWK method. ................................ ................................ ..................... 90 6 1 Results PAA and LDPE PAAZn samples. ................................ ................................ .................... 105 6 2 Contact angle data for deionized water (DI) glycerol (GL) and diiodomethane (DM) for LDPE, LDPE PAA, LDPE PA .................. 106 6 3 Surface energy (in mJ/m 2 ) of LDPE, LDPE PAA, LDPE samples as determined using the OWK and LW methods. .............................. 106 6 4 Atomic concentration of carbon, oxygen and zinc from XPS analysis. ............. 110 A 1 Feature dimensions of positive and recessed Sharklet. ................................ ... 127 A 2 Fidelity of features in positive and recessed SK patterned LDPE using PDMSe molds. ................................ ................................ ................................ .. 128 A 3 Contact angles for smooth and patterned surfaces of LDPE. ........................... 129

PAGE 11

11 LIST OF FIGURES Figure page 2 1 Contact angle of a drop of liquid on a surface, showing tensors from Young Equation ................................ ................................ ................................ ............. 24 2 2 An example of a double row suture for repairing a torn rotator cuff [34]. ........... 30 2 3 Baier Curve which shows fouling as a fu nction o f surface energy [62]. ............. 34 2 4 2.5 SK 2x2 Recessed (dimethyl siloxane) elastomer. ..................... 34 2 5 Activation of benzophenone and the formation of surface radicals on a polyethylene surface. ................................ ................................ .......................... 39 3 1 Illustration of crystalline structures in polyethylene with folded chains (Left) and chain extended configurations (Right). ................................ ........................ 43 3 2 Sketch depicting how a single fiber is made of many macrofilaments and an SEM image of a single fiber where macrofibrils are clearly visible .................... 44 3 3 Chemical structures of AA (left) and GMA (right). ................................ ............... 4 6 3 4 Schematic of the fiber tow during the washing process. ................................ ..... 48 3 5 Path of the Dyneema SK78 fiber during the UV Grafting process. ................... 49 3 6 A series of images illustrating how single filaments a re prepared for tensile tests. ................................ ................................ ................................ .................. 51 3 7 A single filamen t used for mechanical testing. ................................ ................... 52 3 8 ATR FTIR spectra of acetone washed Dyneema (Blue), untreated Dyneema(Red) and 3x PAAGMA grafted Dyneema Fiber (Black). .................. 53 3 9 Ratio of peak areas for untreated Dyneema, washed Dyneema and 3x PAAG MA grafted Dyneema. ................................ ................................ ............ 54 3 10 A stress strain plot for tensile tests of untreated (blue ) and 3x PAAGMA grafted (red). ................................ ................................ ................................ ....... 55 3 11 Examples of filament pull out (A) and fracture (B) during single filament pullout tests. ................................ ................................ ................................ ....... 56 3 12 Peak load for untreated Dyneema and 3x PAAGMA grafted Dyneema which exhibited pullout failure. ................................ ................................ ............ 57

PAGE 12

12 4 1 Chemical structure of glycidyl methacrylate monomer. ................................ ...... 61 4 2 Grafting setup for grafting PGMA to LDPE film. ................................ .................. 63 4 3 Contact angle of LDPE PGMA samples a t different irradiation times. ................ 68 4 4 Representative spectra of LDPE PGMA grafted samples at 180 s econds. ..................... 69 4 5 Representative spectra of LDPE PGMA at 180 seconds of grafting time (A), L DPE (B) and PGMA standard (C). ................................ ................................ .... 69 4 6 Carbonyl Ratio of LDPE PGMA a s a function of grafting time. .......................... 70 4 7 Comparison of LDPE PGMA Gel (Red) LDPE Gel (Blue) and pure gelatin (Black). ................................ ................................ ................................ ............... 72 4 8 Representative ATR FTIR spectra of LDPE PGMA Gel and LDPE Gel samples before desorption treatment. ................................ ................................ 72 4 9 Representative ATR FTIR spectra of LDPE PGMA Gel samples befor e and after desorption treatment. ................................ ................................ .................. 73 4 10 Representative ATR FTIR spectra of LDPE Gel samples before and after desorption treatment. ................................ ................................ .......................... 73 4 11 Representative ATR FTIR spectra of LDPE PGMA Gel and LDPE Gel samples after desorption treatment. ................................ ................................ ... 74 4 12 Gelatin content for LDPE PGMA Gel and LDPE Gel bef ore and after gelatin removal. ................................ ................................ ................................ .............. 74 4 13 Percent of gelatin conte nt retained after desorption. ................................ .......... 75 4 14 Results of 10 minute adhesion study using human mesenchymal stem cells. ... 76 4 15 Results of 10 minute adhesion study using human osteoblasts. ........................ 76 5 1 Chemical structure of an SBMA monomer unit ................................ .................. 80 5 2 An illustration of the second step of the sample making process. ...................... 82 5 3 Advancing ( A ) and receding ( R ) contact angles of a water drop on untreated LDP E. ................................ ................................ ................................ 85 5 4 Sessile drop contact angle of LDPE and LDPE PSBMA. ................................ .. 89 5 5 ATR FTIR spe ctra of SMBA grafted LDPE and untreated control. ..................... 91

PAGE 13

13 5 6 Quantitative ATR FTIR data showing the carbonyl ratio (R c ) of LD PE and LDPE PSBMA. ................................ ................................ ................................ .. 91 5 7 XPS spectra of LDPE PSBMA LDPE. Su lfur and nitrogen peaks indicate the presence of graft. All values are reported in atomic percent. ............................. 92 5 8 Attachment of Ulva Zoospores to PDMSe LDPE and LDPE PSBMA sample s. ................................ ................................ ................................ ............ 93 6 1 Zinc forming cross links between carboxylic acid groups. ................................ .. 98 6 2 PE PAA and LDPE PAAZn samples. ................................ ................................ ................................ ........... 105 6 3 Average surface energy for LDPE, LDPE PAA, LDPE samples as determined by the OWK method. ................................ ................. 107 6 4 Average surface energy for LDPE, LDPE PAA, LDPE samples as d etermined by the LW method. ................................ ..................... 107 6 5 Representative ATR FTIR spectrum of LDPE PAAZn, LDPE PAA and ................................ ................................ ................................ ........... 108 6 6 Quantitative ATR FTIR data sho wing the carbonyl ratio (R c ) of LDPE PAA LDPE ................................ ................................ ............. 108 6 7 ................................ ................................ ............. 109 6 8 XPS spectrum of LDPE PAAZn. ................................ ................................ ...... 110 6 9 Zoospore attachment to PDMSe, LDPE, LDPE PAAZn, LDPE PAA and ................................ ................................ ............................ 111 7 1 ................................ ................................ ......... 120 7 2 Surface grafting of PSBMA covers up topography. ................................ .......... 120 A 1 Incomplete (Cassie Baxter) and complete (Wenzel) wetting of a rough surface. ................................ ................................ ................................ ............. 123 A 2 Creating patterned LDPE using a silicon wafer. ................................ ............... 123 A 3 Creating patterned LDPE using a PDMSe mold. ................................ .............. 124 A 4 SEM micrographs of LDPE patterned us ing a silicon wafer as a mold. ............ 126 A 5 Positive (Top down (A) and edge view (B)) and negative (Top down (C) and edge view (D)) SK patterns made with PDMSe molds. ................................ .... 127

PAGE 14

14 A 6 Optical micrograph of recessed SK in LDPE. ................................ ................... 128 A 7 Representative water drops on surfaces of smooth LDPE (A), +SK (B), SK (C) and +SK made from PDMSe (D). ................................ ............................... 129

PAGE 15

15 LIST OF ABBREVIATION S AA Acryl ic acid AIBN Azobisisobutyronitrile AMPS 2 Acrylamido 2 methylpropane Sulfonic Acid ANOVA Analysis of variance ASW Artificial sea water ATR FTIR Attenuated total reflectance Fourier transformed infrared spectroscopy ATRP Atom transfer radical polymerization BP Benzophenone DI Deionized w ater DMSO Dimethyl sulfoxide DNA Deoxyribonucleic acid ERI Engineered roughness index GMA Glycidyl methacrylate HMDS Hexamethyldisilazane hMSCs Human mesenchymal stem c ells hOB Human o steoblasts HPX Hydroperoxidation LDPE Low d ensity p olyethylene LDPE Gel LDPE with physisorbed gelatin coating. LDPE PAA Low density polyethylene grafted with poly (acrylic acid) LDPE PAAZn Low density polyethylene grafted with po ly (acrylic acid ) with bound zinc

PAGE 16

16 LDPE PGMA Low density polyethylene grafted with poly (glycidyl methacrylate) LDPE PGMA Gel Low density polyethylene grafted with poly (glycidyl methacrylate) grafted with gelatin LDPE PSBMA Low density polyethyle ne grafted with poly (sulfobetaine methacrylate) LW Lifshitz van der Waals acid/base method for measuring surface energy OWK Owen Wendt Kaelble method for measuring surface energy PAA Poly (acrylic acid) PAAGMA Acrylic acid co glycidyl meth acrylate copolymer PAAZn Poly (acrylic acid) with bound zinc PDMSe Poly (dimethyl siloxane) elastomer PE Polyethylene PEG Polyethylene glycol PET Polyethylene terephthalate PGMA Poly (glycidyl methacrylate) PSBMA Poly (sulfobetaine met hacrylate) SBMA Sulfobetaine methacrylate SEM Scanning electron microscopy SK TBT Tribulyltin (IUPAC name: Tributylstannane ) UHMWPE Ultra high molecular weight polyethylene UV Ultraviolet light XPS X ray photoelectron spectroscopy

PAGE 17

17 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ENGINEERING POLYETHYLENE SURFACE ENERGY FOR ADHESION USING SURFACE GRAFTS By David Carnaby Jackson March 2011 Chair: Anthony B. Brennan Major: Materials Science and Engineering One would be hard pressed to live an entire day without coming into contact with polyethylene. It is in the plumbing, automobil es, refrigerators and homes of almost everyone in modern society. Worldwide, an estimated 80 million metric tons are produced each year. While versatile properties make it commonplace, its inert surface limits its effectiveness in many app lications. From relatively trivial issues such as printing to the life and death matter of ballistic protection, the poor surface adhesion to polyethylene is a major issue. It is by altering this surface that the full potential of the material can be rea lized. Sometimes this can be done using a simple flame or acid treatment. In other cases, more advanced surface modification methods are required. This work is focused on modification of polyethylene surfaces by a grafting process to control adhesion Grafted samples were characterized using attenuated total reflection Fourier transformed infrared spectroscopy (ATR FTIR), sessile drop contact angle and x ray photoelectron spectroscopy (XPS). ATR FTIR was used quantify graft content and chemistry, such as carbonyl content. Contact angle measurements were used to calculate surface energy using the Owens Wendt Kaelble

PAGE 18

18 (OWK) and Lifshitz van der Waals (LW) methods. XPS quantified the amount of new surface eleme nts, such as sulfur or zinc. Results showed a significant increase of adhesion with epoxy of human osteoblasts (h OB s) and a significant decrease in adhesion of algae. Adhesion was enhanced between polyethylene and epoxy, e.g. fiber pull out strength increased 15%, via a copolymer graft of acrylic a cid and glycidyl methacrylate. A poly (glycidyl methacrylate) (PGMA) graft on polyethylene increased adhesion of gelatin, which supported a 106 % rais e in attachment hOB s. Adhesion to polyethylene was inhibited by use of either a poly (sulfobetaine methac rylate) (PSBMA) zwitterionic graft or poly (acrylic acid) (PAA) graft complexed with zinc. An assay with Ulva zoospores indicated 92% and 88% inhibition for the PSBMA and PAA respectively compared to polyethylene without graft

PAGE 19

19 CHAPTER 1 INTRODUCTION P olyethylene (PE), along with concrete and steel, is one of the most ubiquitous materials in the modern world. An estimated 80 million metric tons of PE are produced each year [ 1 ] Its uses range from the mundane (milk jugs and garbage bags) to high technology (biomedical devices and advanced composites). For such a wide range of application While PE might seem like a simple material, it can have a wide range of properties based on its molecular weight, chain branching and thermal history. This work is focused on adhesion to polyethylene. In some cases such as composites strong adhesion is desirable [ 2 ] Whereas, surfaces exposed to the marine environment are normally designed to inhibit adhesion. Here it is desirable to prevent marine organisms from fouling the surface [ 3 ] This work will look at three different applications: Composites, biomaterials and marine products. T he chemical and thermodynamic methods for influencing adhesion to the surface will be examined in each case This work will show surface energy can be engineered to increase or decrease adhesion to polyethylene as desired. Specific Aim 1: Increase Pullout Strength of Ultra High Molecular Weight Polyethylene (UHMWPE) Fiber s by 1 5 % or m ore via a poly ( Ac r ylic Acid co Glycidyl Methacrylate ) (PAAGMA) Graft It was hypothesized that surface grafts of PAAGMA polymer would improve adhesion by forming a chemical link between the UHMWPE fibers and matrix An inline process was developed for cl eaning and grafting fiber yarns of Dyneema SK78 UHMWPE fibers. Grafts PAAGMA were characterized by quantitative attenuated total reflectance Fourier transformed infrared spectroscopy (ATR FTIR) on Dyneema fibers.

PAGE 20

20 Improved bonding between fiber and epox y was shown by a 40% reduction in pullout failure in pullout tests and a 15% increase in pullout load. Specific Aim 2: Increase Adhesion of Human Osteoblasts (hOB s ) to Polyethylene by 100% or More U sing Covalently Bound Gelatin It was hypothesized that surface grafts of poly ( glycidyl methacrylate ) ( P GMA) would increase the amount of protein irreversibly bound on the surface of polyethylene. LDPE films were grafted with PGMA and characterized with ATR FTIR and sessile drop contact angle. Grafted sample s were exposed to gelatin for 48 hours. T he amount of gelatin bound to the surface was quantified using ATR FTIR both before and after a rigorous solvent treatment The P GMA grafted samples bound statistically more gelatin than the non grafted sample cont rol. An attachment assay showed 106% more hOBs on the grafted samples than the control Specific Aim 3: Reduce the Adhesion of Ulva Zoospores to Low Density Polyethylene Films by 9 0% or More U sing Surface Grafts It was hypothesized that surface grafts of poly ( sulfobetaine methacrylate ) ( P SBMA) could be crea ted using a UV process and that these grafts would greatly reduce the adhesion of Ulva zoospores A protocol for grafting the zwitterionic monomer sulfobetaine methacrylate (SBMA) was developed. The grafted surface was characterized with sessile drop contact angle, ATR FTIR and X ray photoelectron spectroscopy (XPS) Water contact angle decreased significantly after grafting. Surface energy was shown to increase dramatically. ATR FTIR showed a sub stantial amount of PSBMA graft. XPS showed sulfur and nitrogen, which are unique to the graft, were present on the surface. A 9 2 % reduction in settlement of Ulva zoospores was seen on grafted surfaces relative to untreated LDPE

PAGE 21

21 It was also hypothesize d that acrylic acid grafts with bound zinc (PAAZn) would reduce the adhesion of Ulva Zoospores. A method for grafting poly (acrylic acid) to the polyethylene surface and binding zinc was developed. The PAAZn grafted LDPE (LDPE PAAZn) was compared to LDPE grafted with acrylic acid without zinc ( LDPE PAA) control s S amples were characterized using sessile drop contact angle, ATR FTIR and XPS. LDPE PAA LDPE had statistically lower water contact angle than the untreated polyethylene. S amples of LDPE PAA and LDPE PAAZ n did not have statistically different surface energy. The LDPE PAAZn samples reduced zoospore attachment by 85% rel ative to untreated polyethylene but there was no statistical diff erence between LDPE PAAZn and LDPE PAA

PAGE 22

22 CHAPTER 2 BACKGROUND Adhesion and Surface Energy S urface energy is a vital material property A toms or molecules at the surface are missing nearest neighbors therefore surface properties and behaviors are ver y different from those of the bulk [ 4 ] Surface energy, which is sometimes referred to as surface tension, is defined as the change in Gibbs free energy (G) with surface area (A) at constant temperature (T) and pressure (P) The variables i and j are the materials meeting at the surface i.e. corresponds to the surface energy of a liquid with a vapor. Surface energy is fundamentally tied to thermodynamics; therefore it is essential to un derstanding adhesion. Adhesion is a very important property in materials science. Strong adhesive force between surfaces is desirable in applications such as composites [ 2 ] In other applications such as ship hulls, adhesion is to be limited [5] Adhesion is caused by a thermodynamic incentive to lowe r surface energy [ 4 ] Methods for understanding and quantifying surface energy are essential to engineering adhesion in these different applications. The modern scientific understanding of surface energy began with Thomas Young. Young was a prolific scientist and physician who greatly contributed to mechanics (the H e published an essay in 1804 describing the forces influencing a liquid drop on a surface [6] Young describe d how these forces varied across different fluids by examining t he interaction of wine and (2 1)

PAGE 23

23 water with a solid surface He proposed the existence a force sufficiently strong to hold the mass of a drop to the surface. This was the first mention of surface tension as a scientific p roperty S urface energy is calculated by measuring the angle of the triple point formed by interaction of a solid material s urface, a solvent and gas e.g. air There are two methods for evaluating contact angle: Sessile drop and captive air bubble. Sessile drop contact angle applies the liq uid to the surface in air Captive air bubble conversely, applies a gas bubble to the surface in a liquid environment. The angle is measured between the solid liquid and air liquid interface, opening toward the liquid region as shown in figure 2 1 Cont act angle is often done using water but other solvents such as glycerol and diiodomethane are used. The angle at the triple point will be defined by Young Equation (Equation 2 2) Where LV is the sur face energy of the l iquid vapor, E is the equilibrium contact angle, SL is the energy of the solid liquid interface and SV is the energy of the solid vapor interface, which is called the solid surface energy. These terms are tensors acting on the tripl e point as illustrated in figure 2 1 Each tensor is acting away from the triple point because the surface energy acts to reduce the surface area. T he Young Dupr Equation (Equation 2 3) relates the contact angle and surface energy to the Gibbs free e nergy of adhesion. (2 2) ( 2 3)

PAGE 24

24 This is vital to understanding the thermodynamics of adhesion. A lower contact angle correlates with a lower free energy of adhesion. The free energy of adhesion can also be written as: These relations (Equations 2 3 and 2 vital to understanding adhesion and wetting. Figure 2 1: Contact angle of a drop of liquid on a surface, showing tensors from Young Equation Consider the exampl e of a water drop which is a high surface energy liquid wetting a solid surface The SV tensor is stronger for a high surface energy solid, and the triple point is pulled outwar d T h is results in a lower contact angle indicating a low Suc h high surface energy solid materials are defined as hydrophilic Conversely, the LV tensor is dominant for a solid with low surface energy and the triple point is pulled inward limiting wetting. The system has a higher (although still negative) For this reason, water beads up on low energy surfaces. Low energy surface s, those with a water contact angle >90 are defined as hydrophobic. Measuring the surface (2 4)

PAGE 25

25 The proble m in implementing equation directly is that both SV and SL are unknown. The means for substituting for SL fundamental differences between the different methods for calculating surface energy. There are three methods which will be discussed here: t he Zisman method, the Owen Wendt Kaelble (OWK) method and the Lifshitz van der Waals (LW) acid base method. The Zisman method is an empirical model which determines the critical surface energy c of a solid by extrapolating contact angle data. This method assume s that a s LV decreases, SL decreases to zero therefore contact angle measurements for a series of liquids can be extrapolated to where SL is zero. Experimentally, contact angles of several liquids are measured. The cosine of the contact angles are then plott ed and extrapolated to 1 ( =0 when cos =1). The energy value at which the cosine of the contact angle equals one is the critical value c While the Zisman method is effective at characterizing a surface, it has several drawbacks. T his method is based on extrapolation of data; therefore several (4 5) liquids are needed to create an accurate measurement. A Zisman plot using only 2 solutions would be unreliable and is explicitly prohibited by the developers of the method [7] Most importantly, the terms c and SV are similar but not the same ( ) [8] O ther methods are used for directly measuring surface energy. The OWK method substitutes for SL by splitting surface energy into polar ( ) and dispersive ( ) contributions to solve for SV directly. The OWK method assumes the relationship of and to their polar and dispersive components as shown in equations 2 5, 2 6 and 2 7. These values can then be input into equations 2 3 and 2 4 create equation 2 8, which can be solved directly. Values of and are

PAGE 26

26 established in the l iterature for several liquids [7] This equation can be solved for and which can be used to calculate using equation 2 6. Two liquids must be used: one highly polar liquid such as water and one non polar liquid such as diiodomethane. The LW acid b ase method operates on a similar concept, but is more elaborate. This method separates surface energy into Lifshitz van der Waals contributions and acid/base contributions Lifshitz van der Waals contributions include attractive forces from induced, tem porary or permanent dipoles. The acid/base contributions include forces from hydrogen bonding between a proton donor and a proton acceptor a Lewis acid or base respectively. An excellent derivation and description can be found in [7] A brief summary is offered here. There are two basic assumptions in the LW method. First, the so lid surface energy is the sum of the Lifshitz van der Waals interactions ( ) and the Lewis acid/base interactions ( ) as shown in equation 2 9. Second, the acid/base interactions are twice the geometric mean of the acid contributio n ( ) and the base contribution ( ), as shown in equation 2 10. Combining these into equation 2 3 and 2 4 gives the relationship shown in equation 2 11 which can be solved directly. Values of (2 5) (2 6) (2 7) (2 8)

PAGE 27

27 and have been established in the literature for several liquids [7] This makes it possible to solve f or each of the unknowns ( and ) using contact angles from three known liquids. The values for and can then be used in equations 2 9 and 2 10 to calculate Surface Spectroscopy Wh ile contact angle provides valuable information about the surface, other methods can be used to identify specific surface chemistry Attenuated total reflectance Fourier transformed infrared spectroscopy (ATR FTIR) profiles the surface ~1 micron deep and can identify surface groups such as glycidyl and amine groups. These groups can be extremely important to understanding adhesion, as they can react to form a coval ent bond. This can be used either to bind a molecule, such as a protein, to the surface or to bond two phases together. Examples of each of these were performed in this work. Applications of Adhesion to Polyethylene (PE) This work will focus on three applications: Adhesion to PE in fiber reinforced composites, adhesion to PE in high strength o rthopedic sutures and adhesion to PE of organisms in the marine environment. The remainder of this chapter will focus on the relevant backgrounds for each specific case. While these applications may seem very different, they are all connected by the scie nce of adhesion The adhesion properties of polyethylene have been altered in a variety of applications [2, 9 18] The low surface (2 9) (2 10) (2 11)

PAGE 28

28 energy of polyethylene makes it difficult to print ink or even apply adhesive F or this reason, there are a wide variety of surface treatments which have been used on polyethylene to improve its performance Adhesion in Composites: Ultra High Molecular Weight Polyethylene Fibers Ultra high molecular weight polyethylene (UHMWPE) fibe r high strength ( 98 GPa) and low density (0.97 g/ml) makes it an ideal material for a variety of applications, including sail cloth, cut resistant gloves and high strength nautical ropes [19] UHMWPE fiber has higher specific strength than aramid or graphite fibers. Composites from UHMWPE fibers are of particular interest [2, 9, 20 23] I t is possible to make tough, light weight composites for ballistic vests, radar protective domes and winter sports equipment using UHMWPE fibers as a reinforc ing phase T he exceptional mechanical properties of UHWMPE fibers make them sought after for advanced composites but their low surface energy limits their use due to poor adhesion [24 26] Epoxies and polyurethanes are more polar than the fibers result ing in weak bonding between the fiber and the matrix Poor bonding can also lead to thermal instability due to the stress induced by thermal expansion mismatch [27] For these reasons, surface tr eatments to improve adhesion are a necessity. A wide variety of methods have been developed to increase the adhesion between UHMWPE and different matrices [ 2 ] Surface grafts of poly (acrylic acid co glycidyl methacrylate) (PAAGMA) should be able to form a bond with the matrix via the epoxy ring These grafting methods can also be used to improve adhesion between polyethylene and other materials, such as living cells and organisms.

PAGE 29

29 Ad hesion of Cells: High Strength Orthopedic Sutures Anatomy of the Shoulder The rotator cuff is the capsule for the glenohumeral joint. It contains the attachments for the subscapularis, supraspinatus, infraspinatus and teres minor muscles. These muscles co mbine to support the hum erus and play a key role in essential motions of the upper limb such as flexion of the arm, rotation of the arm, and adduction of the arm Injury of these attachments is the most common clinical problem of the shoulder, resulting in more than 4.5 million physician visits per year in the U nited States [28] Surgical repair a s shown in figure 2 2, is often the preferred treatment for a torn rotator cuff. Surgical Repair R otator cuff repair surgeries have an unacceptably high rate of recurrence due to poor adhesion A 2008 study reported that f or severe tears 60% of the repair s are repair [29] This high recurrence rate is a result of many factors, such as patient behavior, severity of the injury and surgical technique [30] Surgeons are making great efforts to evaluate current techniques and develop new, more effective procedures. These procedures include different surgical techniques, different suture anchors and different suture materials [29, 31 41] N ew materials will also improve the surgical outcomes as the current UHWMPE fibers have very poor adhesion. Adhesion Between Suture and Living Tissue Th e most common failure mode is for the suture to pull thorough the tissue [39] This is because, as in composites, adhesion to polyethylene is very limited. One way to improve adh esion is to use natural materials to improve attachment of cells on the material. Natural materials such as collagen,

PAGE 30

30 Figu re 2 2 : An example of a double row suture for repairing a torn rotator cuff [34] Used with permission. matrix (ECM) and provide better cell attachment wit h better compatibility [29, 32] Cells naturally adher e to natural materials using cell integrands resulting in higher attachment Collagen derived materials however, are not as mass producible as synthet ic materials such as UHMWPE A combination of synthetic and natural materials may be an alternative. C oating a synthetic material with a natural one could take advantage of

PAGE 31

31 both strengths. Such a material would combine the high strength and production scale of synthetic polymers with the biocompatibility of high cell attachment of ECM derived polymers C ollagen and gelatin are possible choice s for a coating material. Both ha ve been used by a variety of researchers as a tissue scaffold for its mechanical properties and strong tissue bonding [10, 42 47] Mazzocc a et al. claimed to have coated the surface b y soaking the sutures in collagen and gelatin solutions for an extended period of time, [44] The results showed the potential of a collagen coated synthetic m aterial by a significant increase in attachment and growth of osteoblasts and tenocytes The limiting factor of such a processes is the weak bond between the physisorbed protein and the UHMWPE surface. Protein adsorption is a balance of reversible and ir reversible adhesion. Reversible adhesion, where the protein is easily removed, can make up a large fraction of protein adsorbed to a surface [48] Surface grafts have been shown to decrease reversible adhesion, favoring irreversible adhesion. Poly (glycidyl methacrylate) (PGMA) grafts have been shown to be effective for irreversibly binding proteins to a polymer surface [49] Here, as with the earlier composites application the epoxy group forms a chemical bond with the amine grou ps on proteins. This increase s irreversible adhesion. If the proteins are bound to the surface tightly, this should also strengthen the bond between the suture and the surrounding tissue by strengthening the weak link. P GMA is an example of a surface g raft that can be used to increase adhesion, but other surface grafts can reduce adhesion of cells and proteins The adsorption of proteins onto vascular grafts can lead to clotting and restenosi s; therefore there is

PAGE 32

32 considerable research into how to preve nt this Surface g raft s of zwitterionic polymers have been shown to be very effective at reducing adhesion of fibronectin and smooth muscle cells [50 52] Similar grafts can be used to reduce adhesion of marine o rganisms. Adhesion in the Marine Environment: Biofouling B iofouling defined as the undesirable accumulation of organisms, cells and cellular proteins, is an important phenomenon. T he fouling of ships increases the operating costs of merchant and navy sh ips. The accumulation of organisms on the hull increases drag and thereby the amount of energy needed to propel the ship [53] A recent study for US Navy Arleigh Burke class destroyers showed that fouling can increase req uired shaft power at 15 knots by up to 76% [54] Fouling raises the fuel costs of these destroyers $1.15 million per ship per year. It is estimated that a slight reduction of fouling would save the US Navy $19 million a year on this one class of ship s alone The removal of the organisms is also costly requiring dry doc king coating removal and reapplication [5] The fouling of ship hulls can also lead to great distortions of marine ecosystems. Fouling of surfaces can tr ansport native species around the world and introduce them to new environments resulting in serious complications. One example of this is the infiltration of the Great Lakes by the zebra mussel [55] Ships c arried these organisms around the world from the Caspian Sea to North America. T he mussels proliferated, radically changing the ecosystem The introduction of a invasive species caused some native species to go extinct [55]

PAGE 33

33 History of Marine Biofouling Coatings Marine biofouling has been a concern of human beings since early civilization [5 6] Pitch and tar were some of the coatings used to reduce fouling. The Greeks and Romans used copper and lead, which continued to be the materials of choice for centuries. Lead coatings were still in use in the 17 th century [57] C opper could no t be used on modern ships due to an electrochemical reaction with iron and a new anti fouling coating was needed T ributyltin (TBT) became the standard coating around the world by the end of the [56] and was very effective at preventing fouling of multiple species TBT was also very toxic [58] and s ystemic proble ms were reported The most noted example is the oyster farms of Arcachon Bay, France [59] Imposex, where females develop male sex organs and are unable to lay eggs, almost destroyed the local oyster industry of the region TBT was subsequently tightly regulated It was eventually bann ed by the International Maritime Organization in 2008 [58] The ban of TBT reflects the importance of developing non toxic, non r eleasing anti fouling and fouling release coatings which has been a major area of research ever since. Non toxic Anti fouling Surfaces Fouling of marine organisms can be limited via non toxic means such as surface energy. Since the marine environment is complicated and the organisms themselves have different membranes and adhesives, settlement is not a linear function of surface tension. One important model for attachment is the Baier Curve shown in Figure 2 3 [3, 60, 61] F ouling decreases with increasing surface energy for very hydrophobic surfaces until it reaches the foul release zone (22 24 mN/m) F oulin g increases as the surface energy increases beyond this zone, until ~ 60 mN/m where it begins to drop

PAGE 34

34 dramatically M any non toxic coatings are made from silicones (foul release zone) or hydrated materials like poly (ethylene glycol ) ( PEG) (high surface energy zone) for this reason Figu re 2 3 : Baier Curve which shows fouling as a function of surface energy [62] Used with permission. Surface Topography Figure 2 4 : SEM image of +2.8 SK 2x2 ( Left ) and 2.5 SK 2x2 Recessed Sharklet ( Right ) patterns in poly (dimethyl siloxane) elastomer The pattern consists of 2 m wide features with 2 m spacing between them. Features are between 2 and 3 m tall [63] Used with permissi on.

PAGE 35

35 Another non toxic means of preventing fouling is surface topography. Surface topography exists in nature and is very effective at limiting fouling Marine organisms have evolved specific mechanisms to prevent themselves from fouling : crabs, mussels, dogfish and sharks are just a few examples [64] Lithography has been used to mimic these patterns in PDM Se molds [65, 66] P atterns based on shark skin have been very effective. The Sharkl (SK) consists of surface features ~2 m wide and 4, 8, 12 and 16 m long. These features can be either elevated or recessed as shown in figure 2 4 A quantitative parameter called the engineered roughness index (ERI) was developed t o relate empirical data between different surface topographies. The first representation was based on Wenzel roughness ( ) [67] area fraction of feature tops ( and degrees of freedom which refers to the number of directions the organism can move across the surface [68] N ew patterns resulted in a new model, which was based on the number of distinct features ( ) rather than the degrees of freedom [69] This model was later extended to include the size and motility of the fouling organism ( number) and the surface energy of the material ( ) [63] The Reynolds number is used to encompass the effect of motility and size of the organism The latest form of the model is given in equation 2 12. The term m is the slope of the line, encompassing the factors not yet defined by the model. This model, in contrast to the Baier Curve, predicts that settlement will decrease linearly with increasing surface energy. (2 12)

PAGE 36

36 Reducing Fouling of Polyethylene via Surface Grafting Surface grafting can be used to increase the surface energy of PE ( =33 mJ/m 2 ) and reduce fouling Polyethylene can be grafted with different surface chemistries to influence attachment. Both the Baier curve and the ERI pre dict that increasing surface energy should greatly reduce surface adhesion. There are a variety of surface grafts which have been used to alter surface energy and reduce fou ling [70] One example is zwitterionic surface grafts. Surface grafts of zwitterionic polymers such as [2 (Methacryloyloxy) ethyl] dimethyl (3 sulfopropyl) ammonium hydroxide referred to as sulfobetaine methacry late or SBMA have been effective at reducing fouling on glass surfaces [52, 71 74] It is believed that zwitterionic grafts bind water to the surface, making it difficult for the organisms to displace the water a nd settle on the surface. It may also be possible to reduce fouling by binding a n anti foulant. Since the ban of TBT, other paints were developed using copper and zinc ions [75] These are considered less toxic than TBT and can reduce fouling on a surface [76] Surface grafts can be used to bind these metal ions [77] Such a surface would be a low fouling surface similar to those used in these antifouling paints by bind zinc via a surface graft Surface Graf ting to Polyethylene Surface grafting creates polymer chains which are covalently bound to the surface [18] Such grafts are different from coating or spin casting because the grafts are bound to the surface covalently and cannot be removed by a solvent. S urface grafts gene rally represent a permanent change in surface chemistry though surface rearrangement is possible [78] It is also different from chemical treatments s uch as acid etching or flame treatments which do not grow polymer chains on the surface. What follows is a brief summary of different grafting methods.

PAGE 37

3 7 Plasma Treatments In plasma treatment the substrate is exposed to reactive gas initiated by a fluctu ating magnetic field. The field ionizes the gas to create a reactive species which attack with the substrate Surface groups formed by the reaction are dependent on the gas used. Oxygen gas creates oxygen containing groups like carbonyls and peroxides [20, 79] Nitrogen or ammonia gas creates amine and amide gr oups [17, 80, 81] Argon gas can be used to create surface radials. The radicals induce polymerization of vinyl monomers. A llylamine can be used to create a grafted polymer [82, 83] In each case, plasma treatment has been shown to increase surface energy of the surface Plasma tre atments have the advantage of being very quick. It is possible to create grafted polymer in only a matter of seconds. However unless the power output is controlled, the plasma will etch the polymer surface [84] This etching results in chain scission which decrease s mechanical prop erties [25] T his is an excellent method for treating polyethylene wher e mechanical properties are not of major concern and high volume is needed. Gamma Radiation Grafting The high energy of gamma radiation breaks bonds both at the surface and in the bulk creating radicals. Exposing the irradiated polymer to monomer will r esult in surface graft polymerization This has been demonstrated in the literature for vinyl monomers such as allyl alcohol [85] The bulk radicals are important as well. It is possible to cross link the polymer by exposing it do acetylene gas during ra diation [86, 87] The cross link s can greatly improve mechanical properties such as modulus by limiting chain motion [88] C hain scission during irradiation and residual radicals however, can reduce mechanical stability [89] Furthermore, residual un reacted

PAGE 38

38 radicals can damage the polymer over time. This has been a major complication in the wear of UHMWPE used in knee and hip replacements [90] Hydroperoxidation Grafting Hydroperoxidation (HPX) creates surface grafts by first treating the fibers with a free radical initiator, such as a zobisisobutyronitrile (AIBN) The AIBN extracts hydrogen from the surface, creating surface radicals which react with oxygen to form surface peroxide groups [91] HPX is preferable to plasma because it has less secondary oxidation. More of the surface contains peroxide bonds and carbonyl and other groups [91] These peroxide groups can then be decomposed to create surface graft s via a radical polymerizat i on [15] HPX etches the surface and can damage mechanical properties much like plasma treatment [15] HPX gr ating is also relatively slow, taking several hours to complete. Slow batch process greatly limits its effectiveness for large scale applications. A quicker, continuous process would be preferred. Ultravio let (UV) Photografting Ultraviolet photografting uses a n initiator to create a free radical polymerization. Initiators typically include benzophenone (BP) and other ketones [92] Literature has shown the particular effectiveness of UV initiated photografting using BP as a surface initiator [92 97] The BP molecule is excited by UV light to a diradical state in which it can abstract hydrogen atoms from the surface of a polymer and creating surface radicals. The hydrogenated BP is referred to as a semipinacol radical. In the presence of monomers with a vinyl group, a chain polymerization occurs at the surface radical as shown in figure 2 5

PAGE 39

39 Figure 2 5 : Activation of benzophenone and the formation of surface radicals on a polyethylene surface. The literature has demonstrated the versatility of this method. Polyethylene, polypropylene an d poly (ethylene terephthalate) have all been grafted using this method [98 100] Graft chemistries include monomers such as acrylic acid (AA), acrylamide, glycidyl methacrylate (GMA) and butyl acrylate [100 102] Surface grafting is complete in a matter of minutes using this method with relatively minimal surface damage. For this reason, it is the grafting method used in this work. Graft kinetics is an important consid eration for graft properties, such as the rate of polymerization ( ) A high rate of polymerization is desirable to ensure a high graft yield can be achieved in a relative short amount of time. The rate of polymerization is a function of monomer concentration ([M]), the rate constant for polymerization ( ), the rate constant for termination ( ), the quantum efficiency ( ) and the intensity of the light absorbed ( ), as shown in equation 2 13 [103] The values of k p and k t are constant for a given monomer system. Polymerization rate increases linearly with [M] is a function of [PI] as described by the Beer Lambert Law given in equation 2 14 [103,

PAGE 40

40 104] where a is the absorptivity, b is the thickness of the monomer solution and [PI] is the concentration of photoinitiator. Absorption will increase and reaction rate will increase with the square root of initiator concentration. Increasing reaction rate comes at the expense of kinetic chain length ( ) the average number of monomer molecules consumed by a single radical Kinetic chain len gth serves a representation of molar mass and as shown in equation 2 15, is also a function of [M] and More chains are formed with higher initiator concentration, but there are fewer monomer units per chain Therefore, kinetic chain length decrea ses with the reciprocal square root of Higher monomer concentration increases kinetic chain length linearly I t is necessary to have high monomer concentration and a lower initiator concentration if high molar mass is desired Balancing and is vital to optimizing surface grafting. Larger monomer molecules such as SBMA will have a lower and a much lower polymerization rate. In these cases it might be preferred to increase initiator concentration relative to monomer concentration to insure that the reaction can be completed quickly This would come at the expense of graft molar mass. The concentration of monomer and initiator must be carefull y chosen for a given grafting process. Finally, it is important to limit side reactions such chain transfer and oxygen inhibition. BP radicals are susceptible to oxygen, which will kill potential graft sites. (2 13) (2 14) (2 15)

PAGE 41

41 Monomer solution and the reaction chamber must be purged with nitrogen or argon prior to grafting. Chain initiation on the monomer or the solvent is undesirable. T he carbon hydrogen bond in the monomer or solvent must have higher bond energy than those of the polyethylene surface to favor surface gr afting T here is still however, a competing reaction from the radicals in solution. Monomers which polymerize from radicals in solution form homopolymer polymer which is not grafted to the surface. The creation of homopolymer comes at the expense of graft, so these reactions are to be minimized. This can be done by limiting the thickness of the monomer layer and choosing solvents with limited chain transfer kinetics. A thinner layer of monomer (> 1 mm) favors interaction at the surface relative to th e bulk. It is also possible for the initiator to abstract hydrogen from a graft ed chain creating branched chain. B ranching is only an issu e for high values of [M] but can greatly influence the arrangement of grafts on the surface and their radius of gyra tion [101] Summary The surface of a m aterial is of great importance for adhesion Materials with a high surface energy wet with water easily and are called hydrophilic. Materials with low surface energy repel water and are called hydrophobic. Surface energy is calculated from contact angle measurements with multiple solvents. These thermodynamic factors dictate reversible adhesion but i t is also possible to cause irreversible adhesion via a chemical reaction S urface functional groups which can be identified using ATR FTIR, react to for m a covalent bond Adhesion to polyethylene is an important matter. Polyethylene is ubiquitous ; therefore its applications require many different surface characteristics A dhesion is of

PAGE 42

42 great importance in applications such as composites Since the su rface energy of the polyethylene is quite low, s urface grafts are needed to improve wettability. Grafts which contain epoxide rings, such as GMA, can form a covalent bond between the fiber and the matrix greatly improving adhesion. UHMWPE fibers are als o used in high strength sutures where poor adhesion is also an issue T he weak tissue/suture interface results in the suture pulling through the tissue Coating sutures with proteins such as collagen and gelatin has been shown to greatly increase cell a dhesion. Surface grafts can be used to bind protein and increase the amount adhered to the surface. This increase in protein at the surface will increase adhesion of cells Unlike the composites and orthopedic sutures, adhesion is undesirable in the ma rine environment. Since the ban of TBT in 2008, there has been a strong incentive to create surfaces which are resistant to fouling by marine organisms Surface grafts are particularly promising as a means to reduce fouling Zwitterionic grafts bind wat er to the surface and exclude organisms from adhering. Also, metal ions used in anti fouling paints such as zinc can be incorporated into surface grafts. Surface grafts which bind Zinc should greatly reduce fouling.

PAGE 43

43 CHAPTER 3 INCREASED PULL OUT ST R ENG TH OF ULTRA HIGH MOLECULAR WEIGHT POLYETHYLENE FIBERS IN EPOXY MATRIX USIN G POLY (ACRYLIC ACID CO GLYCIDYL METHACRYLAT E) SURFACE GRAFTS Introduction Gel spun, ultra high molecular weight polyethylene (UHMWPE) fibers have exceptional properties including sp ecific strength and tenacity due to their molar mass and processing [105] The crystallinity exceeds 98% with minimal evidence of chain folding and a high degree of orientation [21] They are drawn during a gel spin process which induce s this high crystallinity and orientation. This method is su perior to others where the UHMWPE is drawn from a melt. The presence of the solvent allows for easier chain motion. Disentanglement enables further drawing after spinning and therefore higher crystallinity and orientation resulting in much higher modulus than other forms of UHMWPE. Figure 3 1: Illustration of crystalline structures in polyethylene with folded chains ( L eft) and chain extended configurations ( R ight). UHMWPE fibers have a dramatically different crystal structure (Right side of figur e 3 1) than typical polyethylene (left side of figure 3 1) A typical crystalline structure for polyethylene consists of lamellar crystalline regions Here crystallites are formed by

PAGE 44

44 chains folding back on each other, with amorphous regions in between th e crystals. The chain extended crystalline structure of UHMWPE fibers does not have folding Chains are aligned with the fiber direction to form long crystallites called microfibrils. The fibers themselves are polycrystalline [106] A single fiber consists of many macrofibrils, which are 0.5 2.0 microns in diameter and are visible in scanning electron microscopy ( SEM ) (Figure 3 2). These macrofibrils are made of many microfibrils A morphous regions and taut tie molecules exist at the interfaces between these microfibrils These regions allow for easier chain motio n and reduce the modulus and creep resistance of the fibers. Additional processing techniques, including radiation cross linking, can limit these effects and improve fiber properties. Figure 3 2: Sketch depicting how a single fiber is made of many m acrofilaments (Left) and a n SEM image of a single fiber where macrofibrils are clearly visible (Right). I t has long been a goal to incorporate UHMWPE fibers into strong fiber reinforced composites for uses such as ballistic protection. The main challenge to creating such a composite is the poor wettability of the fibers. Polyethylene is non polar and hydrophobic. This is a strong contrast to the materials used in epoxy and urethane matrices which contain many polar groups such as carbonyls and amides. The low

PAGE 45

45 surface energy of polyethylene leads to poor adhesion between the fibers and the surrounding matrix. Many researchers have pursued better adhesion by altering either the surface of the fiber [2, 9, 15, 21 ] or the composition of the matrix [107 109] Raising the surface energy of the fiber allows for better wetting of a more polar matrix like epoxies, resulting in a stronger bond B y developing a hydrophobic mat rix material, such as polydicyclopentadiene bonding can be improved. The lower surface energy of the matrix makes for better wetting. It is difficult, however, to create a hydrophobic matrix with similar mechanical properties to epoxies and polyurethane s; therefore it is preferred to alter the surface energy of the fibers. A surface graft can be designed for a specific interaction with the matrix. A poly(acrylic acid co glycidyl methacrylate) ( PAAGMA) surface graft was chosen for improved bonding to an EPON 828 epoxy resin cured with Jeffamine T 403. A crylic acid ( AA ) and glycidyl methacrylate (GMA) the chemical structures of which are given in figure 3 3 were chosen for their chemical and reactive properties AA is very reactive in the photopolym erization, which will help to maximize the polymerization rate The GMA units have a pendant epoxy group which can react with the curing agent of the epoxy to form a covalent bond. Th is bond between the matrix and the fiber should greatly improve adhesio n to the epoxy It should be noted that the high crystallinity and orientation that give the UHMWPE fibers their excellent mechanical strength also make grafting onto them difficult [110] It has been suggested that this is because amorphous regions are easier to graft onto than crystalline regions. UHMWPE fibers, such as Dyneema have very low

PAGE 46

46 amorphous content making grafting onto them significantly more difficult than other forms of polyethylene. Figure 3 3: Ch emical structures of AA (left) and GMA (right). A dhesion between fibers and a matrix is evaluated using a single filament pullout test. Here a single filament is set inside the matrix and allowed to cure. The filament is then pulled out of the matrix u sing a tensile testing frame. There are two failure modes for pullout tests: fiber pullout and fiber fracture [111] Pullout failure occurs when the filament slides out of the matrix with the fiber intact. The peak load of pullout failure is a measure of strength of the bond between the filament and the matrix. Filament fracture occurs when the fi lament itself fails, and the bond between the fiber matrix remains intact Filament fr acture is a sign of strong adhesion between the fi lament and the matrix. It indicates that the fi lament /matrix bond is stronger than the fiber itself. Surface treatments have been shown to damage polyethylene fibers, reducing mechanical properties [ 2 ] UV radiation can also damage polyethylene [112] It is important to quantify damage to the fiber yarn during the grafting process. For this reason, single filament tensile tests are also necessary. Any damage to the fiber would be clearly indicated by a reduction in modulus and/or strain at break. T his work will show the relative strength of the fiber after grafting by comparing back to the mechanical pro perties of the untreated fiber

PAGE 47

47 Materials UHMWPE Fiber Dyneema SK78 UHMWPE fiber was used as a 780 filament yarn. The yarn was pulled through the various treatments using an electric motor. A servo motor at the payout regulated the tension in the yar n at 70 grams. Control of line tension was necessary to minimize mechanical damage to the individual filaments. A level wind at the take up distributed the yarn Monomer and Initiator GMA was passed thr ough an inhibitor removal column 3 times to remove the hydroquinone monomethyl ether inhibitor present for shipping. Removal of the inhibitor allowed for more effective grafting. Benzophenone (BP) was recrystallized in 100% ethanol to improve purity. Th e two monomers and initiator were dissolved in acetone at the molar concentrations given in table 3 1. Concentrations used were based on those from literature of the UV grafting process. T he monomer solution was purged with ul tra high purity nitrogen for 10 minutes. Table 3 1: Concentration of AA and GMA monomers and BP photoinitiator in acetone for grafting solution Component Concentration (M) AA 3.2 GMA 0.8 BP 0.4 Methods Fiber Cleaning The fiber yarn were cleaned in acetone prior to grafting to remove any surface contaminates. This was done by drawing the yarn or fiber tow, through a bath of 99.5% purity acetone at 32 ft/min. Immediately after leaving each bath, the fiber tow was

PAGE 48

48 drawn across a felt wiper to remove excess acetone as shown in figure 3 4 T he fiber tow was then dried in air overnight and sealed in a polyethylene bag. Figure 3 4 : Schematic of the fiber tow during the washing process. Fiber Grafting The grafting setup was adapted from the washing system as shown in fi gure 3 5 In between the 2 baths a Lesco CureMax UV chamber was added. T he monomer enters the UV chamber where it was irradiated after being soaked in the monomer solution A single section of fiber was irradiated for 5 seconds based on a similar proce ss in the literature [97] The first bath and the UV chamber were sealed in a polyethylene lining purged with nitrogen. The fiber tow was drawn through a bath of acetone to remove any ungrafted monomer or homopolymer after grafting After leaving the bath the fiber tow was dried under a stream of nitrogen prior to the take u p. Each sample was grafted three times to increase the amount of graft These samples are

PAGE 49

49 referred to as 3x PAAGMA grafts for this reason. T he fiber was dried under vacuum and stored in a polyethylene bag after the final graft Figure 3 5 : Path of the Dyneema SK78 fiber during the UV Grafting process. Graft Characterization Sections of yarn were wrapped around a 2 inch by 0.5 inch polyethylene mounting card to provide a stable sample for attenuated total reflectance Fourier transformed infrare d spectroscopy (ATR FTIR) The card wa s double wrapped to insure that the whole of the card was covered with fiber. T he wrapped yarn was pressed against a clean aluminum sheet in a two step process to create a flat surface The first stage was 6000 lbs. for 5 minutes. The second was 7000 lbs. for 5 minutes. Three groups of samples were examined: Untreated, washed and 3x PAAGMA grafted Dyneema Untreated samples were examined to determine the presence and composition of surface contaminants. Washed samples were given the acetone washing step only with no graft. ATR FTIR spectra were obtained with a Jobin Yvon

PAGE 50

50 Horiba LabRAM system equipped with an IR module A single reflection is analyzed using a ZnSe crystal. Five spot s were analyzed using 50 scan s which are averaged to create the spectr um Spectra were analyzed using KnowItAll software (to create overlays) and Microsoft Excel (to calculate peak areas). The area of the carbonyl peak at 1720 cm 1 ( ) was divided by the area of an inter nal standard, in this case the methyl bending peak at 1470 cm 1 ( ) to determine the carbonyl ratio ( ), as shown in equation 3 1. The internal standard is a common method for ATR FTIR quantit ative characterization and has been described in literature [113, 114] and standard characterization t exts [115, 116] An analysis of variance (ANOVA) was used with a Tensile Tests Tensile modulus, elongation at break and peak load were determined by single filament tensile tests. Single filaments are approximately18 microns in diameter, making them difficult to ha ndle without some sort of backing. The single filaments were first backed to construction paper using a cyanoacrylate adhesive The ends of the paper were then folded back in to lock the fiber in place as shown in figure 3 6. The paper allowed for the sa mple to be gripped in the testing frame as shown in figure 3 7 Prior to each test, the backing was removed so that only the filament is being stressed Single filaments with a 5 inch ga u ge length were tested at 2.5 mm/min using an Instron 1122 frame wi th an MTS ReNew software package. A 500 gf load cell was used to measure the load on the filament. Strain was calculated from the displacement of the (3 1)

PAGE 51

51 crosshead of the testing frame. Elastic modulus was calculated assuming a circular cross section with a diameter of 18.5 microns based on using optical microscopy Figure 3 6 : A series of images illustrating how single filaments are prepared for tensile tests. Modified from images made by Jim Seliga Pull out tests Single filaments were also used for p ull out tests. Poly (dimethyl siloxane) elastomer (PDMSe) molds are prepared to create a 1 x 0.5 x 0.5 cm epoxy block. Single filaments

PAGE 52

52 are placed in the PDMSe while the other end of the filament is adhered to a paper backing as in the tensile tests. Ne xt EPON 826 (an epoxy base) and Jeffamine T 403 (a diamine curing agent) are mixed at a 100:43 ratio (w/w), degassed and placed in the mold. Here it is cured for 36 hours at room temperature (~20 C). The mold is carefully removed from the cured epoxy w ith 0.5 cm of fiber embedded in it. Figure 3 7: A single filament used for mechanical testing. Construction paper in the gauge length is removed prior to testing. Image by Cliff Wilson. Samples were tested using an Instron 1122 frame with a MTS Re New Software package using a 500 gf load cell. The paper tab was gripped in the upper grip and epoxy was gripped in the lower grip. All samples were tested at 2 mm/min. Untreated and 3x PAAGMA grafted Dyneema were examined. The peak load of samples w hich

PAGE 53

53 exhibited pullout failure was record ed with a t test ( =0.05) used to establish statistical significance. Results ATR FTIR spectra from untreated washed and 3x PAAGMA grafted Dyneema fibers are shown in figure 3 8. The spectra were similar with th e exception of the carbonyl peak at 1720 cm 1 This peak indicated graft content as the carbonyl group is present in both graft monomers and not in the fibers themselves. There was not sufficient graft to determine the relative content of GMA with AA as the carbonyl peak was the only visible graft peak The 3x PAAGMA grafted Dyneema had statistically higher as shown in figure 3 9. This indicates that while the amount of graft was not substantial, there was a significant change in the surface chemistry of the fiber. Figure 3 8: ATR FTIR spectra of acetone washed Dyneema (Blue) untreated Dyneema( Red) and 3x PAAGMA grafted Dyneema Fiber (Black) Washed Dyneema Untreated Dyneema 3x PAAGMA grafted Dyneema

PAGE 54

54 Figure 3 9: Ratio of peak areas for untreated Dyneema washed Dyneema and 3x PAAGMA grafted Dyneema Black line indicates which statistically similar groups. Tensile Properties Grafted samples show ed an increase in modulus while keeping statistically similar values of peak stress and strain at break as shown in table 3 2 This is also shown in representative stress strain plots, given in figure 3 10. The results suggest ed that the grafting proces s improved modulus without sacrificing ductility. This may have been an effect of drawing occurring as the fibers were pulled through the process. Table 3.2: T ensile modulus, peak load and strain at break for untreated and 3x PAAGMA grafted Dyneema SK78 filaments. Standard deviation (STD) is in parentheses. Cross head displacement : 2 .5 mm/min Treatment # of Samples Tensile Modulus GPa (STD) Peak Stress GPa (STD) Strain at Break % (STD) Untreated Dyneema 21 75 (5) 2.7 (0.1 ) 4.5 (0.3) 3x PAAGMA Graft ed Dyneema 20 85 (6) 2.7 ( 0.2 ) 4.3 (0.3) 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20 Untreated Dyneema Washed Dyneema 3x PAAGMA Grafted Dyneema R c

PAGE 55

55 Figure 3 10: A stress strain plot for tensile tests of untreated (blue) and 3x PAAGMA grafted (red). Samples are representative of the average modulus for each group. Difference in modulus was statisti cally different, but differences in peak stress and strain at break were not statistically different. Representative examples of pullout and filament fracture are given in figure 3 11. All of the untreated samples pulled out of the epoxy without fracturin g the fiber, as shown in table 3 3. Filament fracture was much more common for the grafted samples. This indicated a stronger bond at the filament/epoxy interface due to the surface graft. The peak load of the samples separated by failure mode is given in figure 3 12. Grafting increased pull out load by 15%, which was statistically significant ( =0.05). 0 500 1000 1500 2000 2500 3000 3500 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 Stress (MPa) Strain (%) Untreated 3x PAAGMA Grafted

PAGE 56

56 Pull out Test Data Figure 3 1 1 : Examples of filament pull out (A) and fracture (B) during single filament pullout tests. Modified from an image made by Cliff Wilson. Table 3 3: Comparison of percent pullout for untr eated and 3x PAAGMA grafted Dyneema SK78 Treatment Number of Tests Percent Pullout Untreated Dyneema 17 100% 3x PAAGMA Graft ed Dyneema 12 56% ( A ) ( B )

PAGE 57

57 Figure 3 1 2 : Peak load for untreated Dyneema and 3x PAAGMA grafted Dyneema wh ich exhibited pullout failure Error bars represent one STD Asterisks indicate statistically significant groups Discussion The presence of the carbonyl peak in ATR FTIR spectra with a statistically significant height confirmed that grafting changed the surface chemistry. Graft content, however, appears to be very limited. Carbonyl ratios seen in the lite rature are typically 5x larger. This could have been because of the limited UV exposure time. Perhaps polymerization was incomplete with only 5 seconds of radiation. Also, with the high degree of orientation and crystallinity of UHMWPE it becomes more di fficult to graft to the surface of polyethylene. Dyneema SK78 is among the most highly oriented, highly crystalline fibers in the world, which could be why the grafting was so limited. It is interesting that the fiber modulus increased as a result of the grafting process. There are two possible explanations for the higher modulus, i.e., influence of surface grafts or induced higher orientation by drawing. Thi s could have been caused by grafted chains effecting deformation mechanisms of the fibers limit ing the effect of the surface defects. G raft chains could act like large pendant groups, limiting chain

PAGE 58

58 slippage and increasing the energy needed to deform the fibers. This seems improbable, since surface defects only represent a small portion of the ove rall defect structure. The vast majority of defects would be unaffected and a 10 GPa increase in modulus would be unreasonable. It is also possible that the process of passing the fibers through the grafting process served to draw them even further, incr easing orientation and crystallinity. This seems unlikely as 70 grams spread across 780 filaments (~0.09 g per fiber, 0.12% of peak tensile load) should not have been enough force to draw the fibers Adhesion to the epoxy matrix was improved despite th e relatively limited graft. Grafting reduced pull out failure by 40% and increased the peak load at failure by 15% compared to untre ated fiber. PAAGMA graft can greatly improve adhesion to UHWMPE fibers even if the graft content is nearly the detection limit of ATR FTIR spectroscopy It would be interesting to see how graft content increases with further repetitions. Perhaps 5x or 10x would provide even better results. Other changes to the process, such as grafting time or initiator concentration coul d increase adhesion even further. Summary Dyneema SK78 fibers were washed and grafted with PAAGMA using a continuous grafting process. Graft was characterized using ATR FTIR, where spectra show a significan t peak at 1720 cm 1 : Th e peak of the carbonyl gr oup of both the AA and GMA monomers. Mechanical testing showed the grafting process has actually increased the fiber modulus. This was unexpected, as normally surface treatments damage the fiber while changing its surface. This could be because of the g rafting process drew the filament, even though the load was very small.

PAGE 59

59 PAAGMA greatly improved adhesion between Dyneema and an epoxy matrix. Pullout tests showed a reduction in fiber pull out due to the improved bonding between the fiber and the epoxy matrix. Filament fracture was 40% more common for the grafted fibers than for the untreated fibers Peak load for pull out samples increased by 15% compared to the untreated fibers. Future work will focus on increasing the amount of graft on the fiber sur face. This can be done either with more repetitions of the current procedure or by adjusting grafting time

PAGE 60

60 CHAPTER 4 INCREASED ADHESION OF GELATIN AND HUMAN OSTEOBLAST S ON TO POLYETHYLENE USING A GLYCID YL METHACRYLATE SURFACE GRAFT Introduction Rotator c uff tears are a very common injury among the active and the elderly and t he number of injuries is expected to increase as the American population ages [31, 41] The current methods for these repairs which use polyethylene sutures, yield unacceptably poor recurre nce rates [30] [ 31 ] There has been a great deal of research to develop new materials and technique s for treatment for this reason [33, 34, 36 39] Studies show that the most common failure is suture pull through [39] therefore improvi ng adhesion between the polymer sutures and the body may be able to improve patient outcomes. The goal of this work is to show that by grafting gelatin to the surface of polyethylene, adhesion between cells and the surface can be greatly increased. Coat ing of synthetic sutures has been shown improve healing. Mazzocca has shown improved bonding and adhesion of fibroblasts on UHMWPE sutures by solution coating sutures with gelatin and collagen [44] However, such a treatment is limited b y the adhesion of the proteins to the surface. Polyethylene has very low surface energy, therefore fe w protein chains would adhere and those that do would be weakly bound However, by using surface grafting to bind these proteins to the surface, one woul d expect to greatly increase the amount of protein irreversibly bound to the surface. I t is possible to irreversibly bind protein to the surface using glycidyl methacrylate (GMA), the structure of which is shown in Figure 4 1 A surface graft of poly (g lycidyl methacrylate) (PGMA) can covalently bind proteins such as collagen, gelatin and fibronectin via the epoxy ring The bound protein creates a surface that is favorable to cell adhesion

PAGE 61

61 Figure 4 1: Chemical structure of glycidyl methacrylate monomer. Recently, there has also been a great deal of research in the role of mesenchymal stem cells (hMSC s) in healing at the rotator cuff [117 119] hMSC s are pluripotent adult stem ce lls which can differentiate into a variety of cell types, including osteoblasts, tenocytes and chondrocytes Methods for injecting hMSCs into the wound ed rotator cuff have been found to promote healing [118, 119] Surgeons are currently developing methods to extract the cells directly from the humerus during the procedure [117] This procedure or wound site. A suture material that these cells could adhere to could facilitate better healing, greatly improv ing patient outcomes. Primary hum an osteoblasts (hOBs) are a standard cell type for assessing adhesion to orthopedic materials [44, 120] and will also be examined here. The adhesion of these cell correlates with strong adhesion between the materi al and the bone interface. While the material tendon interface is of more importance for high strength sutures, hOBs provide a valuable assessment of the biocompatibility of grafted polyethylene. Gelatin coatings will be evaluated by attachment assays o human osteoblasts. Gelatin is a very useful material in improving biocompatibility [121, 122] It is chosen here because the denatured protein will have higher conformational entropy, which should promote its adhesion to the surface. hMSC s were evaluated as an

PAGE 62

62 indication of how the coating promotes healing and may be increasingly relevant as the injection processes develop Human osteoblasts were used as a means of assessing the attachment to the bone, and allow a comparison between this treatment and others [44] Methods Film and Monomer Preparation Films were prepared by melt pressing Exxon Mobil LD135 MN ( Lot # D10400) low density polyethylene (LDPE) in between two poly (ethylene terephthalate) sheets attached to aluminum plates Sample films were created by taking 10 g of LDPE polymer pellets and pressing them at 1000 lbs force. T he films were quenched to room temperature using a water bath after 15 minutes Resulting films were ~0.5mm thick and had approximately 55% crystallinity (as measured using d ifferential s canning c alorimetry with ASTM F26 25 07 ). Samples measuring 2 in by 0.5 in were section from this film The samples were then washed in a Soxhlet extractor for 24 hours in acetone Dried samples were sealed in centrifuge tubes prior to grafting. Monomer solution used in grafting was p repared using the photoinitiator benzophenone (BP) and GMA in a 0.001/1 ratio without any solvent BP had been purified by recrystallization in 100% ethanol and GMA monomer had been purified by 3 passes through an inhibitor removal column prior to use T he solution was bubbled with nitrogen for 10 minutes to purge out oxygen. P GMA Grafting and covered with 0.5 mL of monomer solution and covered with a quartz plate (Figure 4 2). The quartz plate spread the monomer solution uniformly over the entire film, while still allowing the ultraviolet (UV) light to pass through. A Lesco CureMax UV

PAGE 63

63 chamber was purged with nitrogen gas for 10 minutes prior to grafting, as well as for the duration of the irradiation. Figure 4 2: Grafting setup for grafting P GMA to LDPE film. Samples were irradi ated for 45, 90 and 180 seconds to establish a preferred grafting time The sample s were then cleaned with acetone in a Soxhlet extractor for 24 hours. This w as demonstrated in literature as a way to remove any un reacted monomer or homopolymer [101] T he PGMA grafted samples (LDPE PGMA) were then dried under vacuum and taken for analysis. Gelatin Grafting Samples used for gelatin grafting were grafted with P GMA for 180 seconds and cleaned in ace tone in a Soxhlet extractor for 24 hours. T he grafted samples were immersed in 30 mL solution of 35 mg/mL solution of gelatin type B from bovine skin (Sigma) in carbonate/bicarbonate buffer (pH=9.4 ) and held at 38 C in a water bath for 48 hours. The sam ples referred to as LDPE PGMA Gel were rinsed after grafting to remove any excess gelatin.

PAGE 64

64 Control Groups and Standards There were 2 control groups used: Untreated polyethylene and gelatin treated polyethylene. The second control referred to as LDPE Gel w as treated with the gelatin grafting without the benefit of the P GMA graft. T he handling of the sample and duration of the treatment were identical to the experimental sample. There were two standard groups prepared: Pure P GMA and pure gelatin. A pure P GMA standard was created by polymerizing the same monomer solution on a glass slide for 6 minutes. A pure gelatin standard was prepared covering a glass slide in the gelatin solution and allowing it to dry in open air overnight at room temperatur e Both standards resulted in a layer of polymer several microns thick which could be used for characterization. Gelatin Desorption S amples were subjected to a series of aggressive solvents in an effort to remove any unbounded gelatin A brief summary of the treatments which included deionized water, bicarbonate buffer and dimethyl sulfoxide (DMSO), is given in table 4 1 below. Stronger solvents and detergents were not used because they could damage the polyethylene substrate or interact with the P GMA g raft. After each treatment the samples were vacuum dried and characterized. Table 4 1 : A summary of gelatin desorption treatments. Order Treatment Duration 1 Soak in Deionized Water 48 hours 2 Soak in 9.4 pH Bicarbonate Bu ffer 48 hours 3 Soak in 9.4 pH Bicarbonate Buffer 168 hours 4 Soak in DMSO 18 hours 5 Soak in Deionized water with Sonication 6 hours

PAGE 65

65 Surface Characterization P GMA grafted samples were characterized by sessile drop contact angle with deionized water Five 2 L deionized water drops were placed on the surface using a syringe. The contact angle of the triple point is measured using a goniometer. Five measurements were taken per sample, each from different regions of the sample. The contact angles of the water drops were measured on both sides and then averaged. An analysis of variance ( ANOVA ) ( =0.05) was used to determine statistically significant groups. Attenuated total reflectance Fourier transformed infrared spectroscopy (ATR FTIR) was also used to characterize the surface of the films. Spectra were taken using a Perkin Elmer Spectrum One spectrometer with an ATR module A ZnSe ATR crystal with 12 reflection points was used with an incident angle of 60 for a penetration depth of ~1 micron Reported spectra are the average of 20 scans. Each sample is measured 3 times, sampling a differen t region each time. All quantitative calculations are made using the average of the three measures for each sample. Quantitative measures of P GMA content for the different grafting times were made by the comparing peak area for the carbonyl peak ( ) to the peak area of an internal standard, in this case the CH 2 bending peak ( ) This method is common for ATR FTIR quantitative characterization and has been described in literature [113, 114] and standard characterization texts [115] The area for each peak was measured using Omnic spectra processing software. The carbonyl peak area was taken from1760 cm 1 to 1700 cm 1 and the CH 2 peak was taken from 1490 cm 1 to 1425cm 1 The carbonyl ratio ( ) was then measured using equation 4 1. An analysis of variance (ANOVA)

PAGE 66

66 A nother peak ratio is used to quantify the amount of gelatin on the surface. This is done using the height ratio of the amide peak ( ) and the height of the CH bending peak ( ). The height ratio is determined using equation 4 2. Height is used instead of area in this calculation because of some slight overlap of the carbonyl and amide peaks. This overlap would skew the area of the LDPE PGMA Gel samples, whereas the height shoul d not be affected. An analysis of variance (ANOVA) Cell Adhesion Assay s Cell adhesion assays were performed on 6 mm sections s amples after desorption treatments Samples were sterilized using 70:30 ethanol: water and rinsed with media prior to seeding. Media used for rinsing was corresponded to the media used for cell culture. Two cell types were used: primary hMSCs and primar y hOBs hMSCs were grown in basal MSC growth media (Linza Cat alog # PT 3001 ) and hOBs were grown in basal osteoblasts growth medium (Linza Cat alog # CC 3207 ). Prior to seeding, cells were grown in media and passaged three times with a 0.25% Trypsin/Ethylen ediaminetetraacetic acid ( EDTA ) solution. The cells detached after 2 minutes and centrifuged for 4 minutes. The pellet was then re suspended in new media to the desired cell density. (4 1) (4 2)

PAGE 67

67 LDPE, LDPE Gel and LDPE PGMA Gel samples were seeded with 17k hMSCs (L onza Cata log # PT 2501 ) in 30 L of basal mesenchymal stem cell media (Lonza) Cells were then incubated for 10 minutes to allow for attachment. Media was then aspirated off and the sample was rinsed three times in phosphate buffered saline (PBS) to remov e weakly adhered cells. Surface was then trypsinized to detach the strongly adhered cells, which were then counted using a hemocytometer. LDPE, LDPE Gel and LDPE PGMA Gel samples were seeded 24k human osteoblasts (Lonza Catalog # #CC2538 ) in 30 L of basa l mesenchymal stem cell media (Lonza) These cells were incubated for 10 minutes and rinsed in PBS as in the solution in water. The solution was stored at 20 C prior to PicoGreen d eoxyribonucleic acid ( D NA ) analysis. DNA content of the lysed solutions was determined using a Quant PicoGreen kit (Invitrogen), which uses a florescent dye to stain double stranded DNA. The PicoGreen dye is not influenced by proteins such as gelatin, as has been shown in the literature [121] Lyzed cell solutions were mixed with the dye and loaded into a 96 well plate. Standard DNA solutions between 2.0 g/ml and 0.125 g/ml were used to establish a standard curve. The PicoGreen dye was excited by a wavel ength of 485 nm and the emitted light was measured at wavelength of 528 nm using a BioTek Synergy Mx microplate reader Absorbance was correlated to DNA content using the standard curve. An analysis of variance (ANOVA) was used across all four treatments treatment.

PAGE 68

68 Results Contact Angle T he contact angle decreases linearly (R 2 =0.9974) with increasing grafting time, approaching the sessile drop contact angle of the PGMA stan dard as shown in figure 4 3 This suggests that grafting increases linearly with grafting time up to 180 seconds A t 180 seconds, there is no statistical difference between the grafted sample and the pure PGMA standard This suggests that at 180 seconds the polyethylene is completely covered. The contact angle of the completely grafted sample is in line with values reported by other researchers [101] Figure 4 3: Contact angle of LDPE PGMA samples at different irradiation times. Error bars represent one standard deviation (STD). Black l ines indicate statistically different groups. Blue line represents the value of PGMA standard. y = 0.17x + 90.14 R = 0.9974 0 10 20 30 40 50 60 70 80 90 100 0 20 40 60 80 100 120 140 160 180 200 Contact angle () Time (s) PGMA Standard

PAGE 69

69 ATR FTIR Grafting of PGMA onto LDPE Figure 4 4: Representative s pectra of LDPE PGMA grafted samples at 180 s econds eaks for the epoxide ring. Figure 4 5: Representative s pectra of LDPE PGMA at 180 seconds of grafting time (A), LDPE (B) and PGMA standard (C). Clearly, A is a combination of B and C.

PAGE 70

70 The ATR FTIR spectra for the LDPE PGMA show peaks consistent wit h the LDPE base (2900, 2800, 1450 and 710 cm 1 ), but also peaks for the PGMA graft (1700, 1250, 910 and 850 cm 1 ), as seen in figure 4 4. The 1250, 910 and 850 cm 1 peaks (highlighted in blue in figure 4 4) are indicative of the epoxy ring. When the LDPE PGMA is compared to the controls for clean LDPE and the pure PGMA standards, as in Figure 4 5, it is clear that the graft shows characteristic peaks of both. Figure 4 6: Carbonyl Ratio of LDPE PGMA as a function of grafting time. Error bars represe nt one STD The carbonyl ratio increases quadratically (R 2 =0.9981) with grafting time between 0 and 180 seconds, as shown in figure 4 6. This deviation from linear behavior can be y = 4E 05x 2 + 0.0006x 0.0472 R = 0.9981 0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 0 20 40 60 80 100 120 140 160 180 200 R c Time (s)

PAGE 71

71 explained by two phenomena : an induction period and autoacceleration. Between 0 and 45 seconds the reaction is still being inhibited by oxygen and other contaminates. There is no statistical difference between 0 seconds and 45 seconds for this reason. Autoacceleration is seen a later time points. Chain polymerization is a n exothermic process; therefore the heat produced by the reaction increases the reaction rate with time. The substantial jump in PGMA content between 90 and 180 seconds is the product of this. There is significantly more PGMA ( =.0.05) at 180 seconds tha n at any other time point, hence it was the time point chosen for gelatin grafting. Binding of Gelatin to LDPE The gelatin control shows the amide peaks (1637 and 1543 cm 1 ) consistent with previous literature and the pure gelatin standard as shown in Fig ure 4 7 These same peaks are present in the gelatin grafted samples of bo th grafted and ungrafted LDPE. There was more gelatin on the LDPE PGMA Gel sample than on the LDPE Gel control, as shown in figure 4 8 The removal steps significantly reduced the g elatin content of both the LDPE PGMA Gel and LDPE Gel samples as shown in figures 4 9 and 4 10 respectively. Both LDPE PGMA Gel and LDPE Gel samples showed irreversible adhesion of gelatin. The LDPE PGMA Gel samples have more gelatin on the surface than the LDPE Gel samples after removal, as shown in figure 4 11. Quantitative analysis, summarized in figure 4 12 shows there was statistically more gelatin on the LDPE PGMA Gel surface both before and after the desorption process. LDPE PGMA Gel samples had more than twice the gelatin content of the LDPE Gel samples after desorption. The percent reduction in gelatin content due to the desorption process was lower for the LDPE PGMA Gel samples, as shown in figure 4 13.

PAGE 72

72 Figure 4 7 : Comparison of LDPE P GMA Gel (Red) LDPE Gel (Blue) and pure gelatin (Black) Figure 4 8: Representative ATR FTIR spectra of LDPE PGMA Gel and LDPE Gel samples before desorption treatment. LDPE P GMA Gel LDPE Gel Pure Gelati n LDPE PGMA Gel Initial LDPE Gel Initial

PAGE 73

73 Figure 4 9 : Representative ATR FTIR spectra of LDPE PGMA Gel samples before and after desorption treatment. Figure 4 10 : Representative ATR FTIR spectra of LDPE Gel samples before and after desorption treatment. LDPE PGMA Gel Initial LDPE PGMA Gel Final LDPE Gel Initial LDPE Gel Final

PAGE 74

74 Figure 4 1 1 : Representative ATR FTIR spectra of LDPE PGMA Gel and LDPE Gel samples after desorption treatment. Fi gure 4 1 2 : Gelatin content for LDPE PGMA Gel and LDPE Gel before and after gelatin removal. Error bars represent one STD. Single, double and triple asterisks indicate statistically different groups. 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 Initial After desorption R Gel LDPE PGMA Gel LDPE Gel ** ** *** LDPE PGMA Gel Final LDPE Gel Final

PAGE 75

75 Figure 4 1 3 : Percent of gelatin content retained af ter desorption Error bars represent one STD Asterisk indicates statistically different groups. Cell Adhesion There is a substantial increase between the untreated control and the samples with gelatin as shown in figure 4 14 This indicates that the p olyethylene surface was a very poor surface for initial cell attachment and the presence of gelatin increase d adhesion considerably. There is also a large difference (>50%) between the LDPE Gel and the LDPE PGMA Gel sample. The increased gelatin content seen in the ATR FTIR studies seems to correlate with increased hMSC adhesion. This increase in adhesion is also seen in the osteoblasts study, as shown in figure 4 14. Significantly more DNA, and therefore cells, was on the LDPE PGMA Gel samples than on t he LDPE Gel cells. Also, LDPE Gel also had a statistically significant increase in adhesion compared to the untreated LDPE sample. Clearly, the cells 0% 10% 20% 30% 40% 50% 60% 70% 80% LDPE GMA Gel LDPE Gel Coat % of Initial Gelatin Retained

PAGE 76

76 respond to the protein coated surfaces than the untreated polymer and this preference increases with inc reasing protein content. Figure 4 1 4 : Results of 10 minute adhesion study using human mesenchymal stem cells. Figure 4 1 5 : Results of 10 minute adhesion study using human osteoblasts. Error bars represent one STD Single and double asterisks indi cate statistically significant groups. 0 1000 2000 3000 4000 5000 6000 7000 8000 LDPE LDPE Gel LDPE PGMA Gel # of cells adhered 0 50 100 150 200 250 300 350 400 LDPE LDPE Gel LDPE PGMA Gel DNA from Surface (ng/ml) **

PAGE 77

77 Discussion Data from ATR FTIR and contact angle strongly indicates that grafting was effective. Water c ontact angle dropped from 90 to 59, approaching the contact angle of the PGMA standard. This indicates that t here is complete coverage of the surface with PGMA at this reaction time. ATR FTIR of PGMA grafted samples showed peaks consistent with the chemistry of PGMA including the carbonyl and epoxy rings. The 180 s econd graft was considered preferable for use i n gelatin binding studies for this reason The gelatin desorption data strongly indicates that the PGMA graft irreversibly binds more gelatin at the surface than the ungrafted surface. The initial concentration is higher for the LDPE PGMA Gel samples than the ungrafted controls. More gelatin desorbed from the LDPE Gel samples than the LDPE PGMA Gel samples. Furthermore, more gelatin remains after removal for L DPE PGMA Gel than LDPE Gel. The cell adhesion assay shows the effectiveness of surface treatm ents. Surface gelatin greatly increased the number of cells adhered to the surface. Considerably more cells adhered to the LDPE PGMA Gel surface due to the increased gelatin content The relative increase here is much greater than reported by Mazzocca with a more than 10 fold increase just by the addition of gelatin [44] The most likely explanation for this is that even the LDPE Gel samples reported here have more protein on the surface. The increase in due to gelatin was greater than some values reported for collagen in several ca ses [43, 46] It would be interesting to see how the hMSCs attach, prolifer ate and differentiate on the different samples. Increased cell proliferation could greatly improve healing,

PAGE 78

78 especially if hMSCs could be differentiated into osteoblasts and tenocytes to create new bone and tendon, respectively. Conclusions An effective me thod for grafting PGMA onto LDPE was shown. PGMA g raft is characterized using sessile drop water contact angle and ATR FTIR Water contact angle was reduced to 59 and ATR FTIR showed characteristic peaks for the carbonyl and epoxy groups in the PGMA G rafting time was optimized at 180 s econds for a 0.001/1 molar ratio of initiator to monomer. An effective method for binding gelatin to the LDPE surface was shown. A 48 hour treatment in bicarbonate buffer solution bound gelatin to the LDPE surface. T his was determined by ATR FTIR analysis which showed peaks distinct to the amide groups in the protein. The presence of the P GMA graft bound more gelatin to the surface. It also held more gelatin on the surface after the removal treatments This additi onal gelatin resulted in more hMSCs and hOB adhering to the surface. Future work will focus on the proliferation and differentiation of hMSCs on these surfaces.

PAGE 79

79 CHAPTER 5 REDUCTION OF ULVA SPORE ATTACHMENT ON TO POLYETHYLENE USIN G POLY (SULFOBETAINE M ET HACRYLATE) SURFACE G RAFTS Introduction Biofouling is a major problem for materials in the marine environment. The fouling of ships, pipes, buoys and other objects greatly increases operating costs and reduces product lifetimes and effectiveness [5] The weight and drag of bacteria, algae and barnacles greatly increase fuel costs and the r emoval of the organisms is also very expensive [5, 54] The accumulation of these organisms is also known to cause corrosion and damage the underlying substrate. Fouling can occlude cooling p ipelines, thereby limiting the effectiveness. It is, therefore, very important to develop materials which resist fouling. Materials with surface grafts present an opportunity to create low fouling surfaces [123] Surfaces grafted with hydrophilic chains such as poly ( ethylene glycol) (PEG) form a hydration layer in water due to hydrogen bonding H ydrogen bonding between the water and the graft in this layer can be so s trong that the freezing point of the water is lowered. Remarkably, in some cases water at the surface will not crystallize at as low 70 C [124] The presence of such a layer correlates with limited adsorption of protein and attachment of cells. It is believed that it is this strong hydration layer that causes the low fouling of these hydrophilic surfaces PEG and its derivatives are the standard for hydrophilic surface gra fts, but there are drawbacks. Degradation of PEG limits its effectiveness in practical applications [123] More stable groups such as zwitterionic grafts are emerging as an anti fouling graft option. These grafts have ionic character but are net charge neutral. The ionic character of these groups leads to strong hydration with water much stronger than the

PAGE 80

80 hydrogen bonding seen in PEG. Unlike polyanionic or polycatio nic grafts, these grafts do not attract organisms of the opposite charge. Zwitterionic grafts are thought to be an ideal surface graft for this reason. Surface grafts of poly ( [2 (Methacryloyloxy) ethyl] dimethyl (3 sulfopropyl) ammonium hydroxide ) ( poly ( sulfobetaine methacrylate ) P SBMA) have been shown to be very effective at reducing fouling. The chemical structure of sulfobetaine methacrylate ( SBMA ) monomer is given in figure 5 1. P SBMA grafted to glass slides showed a 97% reduction of fibrinogen pr otein compared to untreated glass [52] An extensive study has shown P SBMA grafts lower adsorption of organic molecules than any other surface grafts [70] It has also been effective at reducing attachment of marine organisms including a 92% reduction of Ulva zoospores and an 85% reduction of Navicula bacteria diatoms [74] Figure 5 1: Chemical structure of an SBMA monomer unit. One major limitation for practical applications of P SBMA grafted surfaces is the current grafting atom transfer radical polymerization (ATR P) process [52, 72 74, 125] ATRP gives excellent control of molar mass and branching, but it is a very slow (4 hr), batch process. Furthermore surface ATRP can only be used on pretreated, brominated surfaces. The long reaction time and limited substrates both must be improved for P SBMA grafts to be viable in practical applications Ultraviolet ( UV ) photopolymerization can remedy both of these problems

PAGE 81

81 UV photopolymerization using BP as a photoinitiator can be used on any polyolefin, polyester or poly ( methyl silane ) [98, 114, 126] There are also a wide variety of vinyl monomers that can be used Reaction times take on the order of minutes rather than hours. The drawback to using UV polymerization is that the reaction is less controlled. Competing reactions and termination mechanisms lead to wide disparities in molar mass, branching and cross linking. The goal of this work is to establish a method for creating P SBMA grafted polyethylene (LDPE P SBMA) and evaluating its resistance to fouling using Ulva algae zoospores. Methods Sample Preparation Samples were prepared using a thermal pressing technique. A 2 step process was used because p ressing the sample in a s ingle step lead s to surface roughness and surface defects caused by the original pellets. By first pressing the LDPE pellets into a film, these defects were avoided. First, a sheet of polyethylene was formed by melting 10 g of Exxon Mobil LD135 MN ( Lot # D10400) against polyethylene terephthalate ( PET ) sheet s backed by stainless steel plates PET wa s used to cover because it allows for eas ier removal of the final LDPE film. The plates we re heated to 180 C using Carver Model M hydraulic press. A 1000 lb load wa s applied after the polymer had melted Glass slides we re used as spacers to insure the final thickness of the film wa s ~ 1 mm. The molten polymer wa s held at temperature and pressure for 15 minutes and quenched in room temperature water. The re sulting film wa s then cut into 3 in x1 in sections that were subsequently bonded to a glass slide as shown in figure 5 2 Briefly, t we re stacked with a trimmed glass slide between the second and third piece. The glass slide wa s

PAGE 82

82 trimmed to 2 in x in so that it will be completely encased in the polyethylene The glass sinks into the polymer, since it is more dense than the LDPE. The enchased glass slide makes the sample negatively buoyant, which is necessary for the zoospore assay The s tack of polyethylene and glass was placed on a PET sheet backed to a silicon wafer The PET sheet allowed for easy removal, while the silicon wafer insure d the final product ha d a smooth surface. A steel plate wa s placed on the bottom to support the sili con wafer. A n additional steel plate with a PET sheet attached to it wa s placed on top of the stack. Spacers 2.3 mm tall we re used to control sample thickness. The sample wa s then heated to 180 C and 10 kg weight wa s applied once the mold is at tempera ture A dead weight wa s used to maintain constant load for the entire process Also, a lower load was desired to ensure the glass slide and silicon wafer d id not fracture Figure 5 2: An illustration of the second step of the sample making process. Here the films made in the first step are pressed with an embedded glass slide.

PAGE 83

83 The sample was held at temperature and pressure for 15 minutes then quenched Samples wi th any visible surface abnormalities or roughness were discarded. Prior to grafting, samples were cleaned by soaking in acetone at room temperature for 24 hours. The samples were then dried under nitrogen and sealed in 50 mL centrifuge tubes for grafting Preparing Monomer A solution of SBMA and BP (0.5 M and 0.005 respectively) was prepared in methanol. A slightly higher ratio of monomer and initiator, compared to other graft chemistries, wa s used here to increase the rate of polymerization due to the l arge size of SBMA The increased concentration of initiator will increase the reaction rate at the cost of kinetic chain length. Monomer solution wa s purged with argon for 10 minutes prior to grafting to drive out any residual oxygen Grafting Procedure A Lesco CureMax UV chamber was used after being purged with nitrogen for 10 minutes. Samples we 1.0 mL of monomer solution wa quartz plate wa s placed on top of the sample to spread the monomer solution evenly across the sample. The sample wa s then placed in the center of the chamber and irradiated for 4 minutes. T he grafted sample wa s cleaned with deionized water ( DI 18 M resistivity) and dried with nitrogen New monomer wa s then applied to the surface and the 4 minute graft ing process wa s repeated. S amples we re soaked in deionized water for 24 hours to remove any residual monomer or homopolymer after the second grafting After

PAGE 84

84 the soak samples we re rinsed with deionized water again and dried under nitrogen. Samples we re t hen store d in a centrifuge tube and taken for characterization Of the four LDPE PSBMA samples made, t hree would be analyzed by Ulva assays with fourth being retained for destructive characterization. This was required for vital characterization methods such as attenuated total reflectance Fourier transformed infrared spectroscopy ( ATR FTIR ) and X ray photoelectron spectroscopy ( XPS ) Pressing against the ATR FTIR crystal would leave an impression that would affect Ulva Spore attachment The retained sa mple w ould need to be cut into smaller sections for XPS analysis. Therefore, only the retained sample could be examined using ATR FTIR and XPS. All four samples were characterized by sessile drop contact angle to ensure the samples were reproducible and that the retained sample was representative of the 3 shipped samples. Experimental Controls There we re 2 control groups: PDMSe and LDPE. PDMSe sample wa s used as a standard because of the extensive use of this material in similar studies [63, 65, 66, 68] Ungrafted LDPE wa s used as a baseline. It controlled for the fouling characteristics of the LDPE substrate. Material Characterization Sessile Drop Water Contact Angle Sessile drop contact angle was done on al l samples prior to the attachment assay Five drops of DI water 2 l in volume were placed on the surface of the sample. The contact angle of the drops was measured using a goniometer. Statistical significance between the LDPE PSBMA samples and the untre ated control was established using a t test ( =0.05, n=3).

PAGE 85

85 Dynamic Water Contact Angle Figure 5 3: Advancing ( A ) and receding ( R ) contact angles of a water drop on untreated LDPE. The retained samples were analyzed using dynamic contact angle using a tilting method. Dynamic contact angle was done using a Ram Hart goniometer, which used an automated drop dispenser, tilting stage and live electronic imaging. A 30 l drop was placed on the surface. Smaller drops, such as those used in the sessile drop analysis, did not slip on the surface and therefore could not be used. The stage was tilted one degree at a time with an image captured after each degree. Tilting was stopped when the drop began to slip and the slip angle was recorded. The last image before slipping is used for analysis using ImageJ software to measure the advancing and receding angles at the leading and trailing sides of the drop respectively as shown in figure 5 3 Five of such measurements were made from each treatment. Analysis of variance

PAGE 86

86 (ANOVA) was combined with a Tuke groups (n=3, Measurement of Surface Energy The retained samples of both untreated and LDPE P SBMA grafted were given more detailed analysis for determining surface energy. These samples were analyzed using 10 drops (5 l in volume) of deionized water, glyce rol and diiodomethane. The relevant physical constants for these liquids are given in table 5 1. The contact angles of each of these liquids were used to measure surface energy via the Owens Wendt Kaelble (OWK) and Lifshitz van der Waals (LW) acid base m ethods. The OWK method uses polar and non polar liquids to determine the surface energy of a solid. It also determines the polar and dispersive components of the surface energy. This wa s important for analyzing the graft, since a large increase in the polar component is expected due to the ionic nature of P SBMA The LW method takes in to account electrostatic and acid/base components and requires three different liquids The LW method will provide an additional measurement of the surface. The use of multiple methods for measuring surface energy is common for evaluating a surface [127] The variance of the contact angle measurements was carried through the surface energy calculation so that statistical significanc e could be established ( =0.05). Table 5 1: Physical constants for deionized water, diiodomethane and glycerol. These are the liquids used to calculate the energy of the surface [7] Liquid L V L V LW L V AB L V + L V Deionized water 72.8 21.8 51.0 25.5 25.5 Diiodomethane 50.8 50.8 0.0 0.0 0.0 Glycerol 64.0 34.0 30.0 3.9 57.4

PAGE 87

87 ATR FTIR Infrared spectra were obtained with a Perkin Elmer Spectrum One FTIR spectrometer equipped with an ATR module The module consisted of a 2 inch by inch ZnSe crystal with an incident angle of 60. Reported spectra we re the average of 20 scans. Each sample was measured 3 times, moving the sample slightly between measurements to sample differen t regions each time. All quantitative calculations we re made using the average of the three measures for each sample. Quantitative measures of PSBMA content were made by the comparing peak height for the carbonyl peak (H C=O ) to the peak height of the CH 2 bending peak ( ). This method wa s common for ATR FTIR quantitative characterization and has been described in literature [113, 114] and standard characterization texts [115] The area for each peak wa s measured using Omnic spectra processing software. The carbon yl ratio ( ) wa s then measured using equation 5 1. Values of R c are then evaluated using a t test ( =0.05) to establish statistical significance. X ray P hoto electron S pectroscopy A section of the LDPE PSBMA retain was exam ined using a XPS/E SCA Perkin Elmer PHI 5100 ESCA s ystem with a magnesium anode. A section of the sample approximately 2 cm x 2 cm was cut using a razor blade and immediately taken to the XPS for analysis. A survey scan was taken from 1100 to 0 keV. The incident angle of the x rays was 45 Atomic percentages were calculated using AugerScan software which integrates peak area and uses literature values for sensitivity factors for its (5 1)

PAGE 88

88 calculation The following peaks were used for quantitative analysis: C1s, O1s, N1z and S2p. Ulva Attachment Ulva attachment was evaluated using an established protocol described elsewhere [68] S amples were immersed in DI water for 24 hours prior to testing immediately removed from the ASW and placed in an assay dish. The dish was then loaded with 10 ml of zoospore solution with a concentration of 1. 0 x10 6 spores/ml and samples we re incubated in darkness for 45 minutes at 20 C After incubation, t he samples are rinsed to remove any non adhered spores and then fixed with a 2 .5 % glutaraldehyde solutio n in ASW. The fixing solution was then removed with sequential rinses of ASW, 50 :50 ASW:DI and DI water. Samples were imaged using an Axiovision 4.8 and a fluorescence microscope where spores are identified by the autofluorescence of chlorophyll. Each slide was imaged 30 times, with a field of view of 0.15 mm 2 An analysis of vari test was used to establish statistically significant groups (n=90, =0.05). Results Sessile Drop Water Contact Angle P SBMA grafting reduces water contact angle from 98 to 64 with a statistically significant differen ce of 34 as shown in figure 5 4 This was in keeping with the hydrophilic nature of P SBMA. The untreated surface wa s more uniform, as seen by the lower variance. LDPE PSBMA ha d a more variant contact angle, suggesting it has regions of different graf t density. This suggests that graft coverage is significant, but not complete.

PAGE 89

89 Figure 5 4 : Sessile drop contact angle of LDPE and LDPE PSBMA Error bars represent one standard deviation (STD) LDPE PSBMA is statistically different from LDPE (t te st, =0.05, n=3) as indicated by the asterisk Dynamic Water Contact Angle LDPE PSBMA showed significantly ( =0.05) reduced advancing ( A ) and receding ( R ) contact angle compared to LDPE, as shown in table 5 2. The more hydrophilic nature of the grafts lowered the contact angles, but it did not have a significant effect on the hysteresis ( A R ). The graft did result in a statistically higher slip angle, which indicates stronger adhesion. The dynamic contact angle supports the conclusions of the stati c drop method. Table 5 2: Results from dynamic contact angle for untreated LDPE and LDPE PSBMA Asterisks indicate values where the grafted sample is statistically different from the untreated LDPE. A R Slip Angle A R Untreated LDPE 110 (3 ) 84 (1 ) 20 (2 ) 26 (3 ) PSMBA Grafted LDPE 75 (5 ) 51 (4 ) 25 (3 ) 24 (6 ) Surface Energy Measurements Contact angle data from the surface energy study showed increase wetting of all three liquids, as shown in table 5 3 This data wa s used to make two surface energy 0 20 40 60 80 100 120 LDPE LDPE PSBMA Contact Angle ()

PAGE 90

90 calculations: The OWK method using deionized water as the polar liquid and diiodomethane as the non polar liquid and another using the LW method, which requires all 3 liquids. The polar and dispersive components, along wi th the overall surface energy, are presented in table 5 4. There was a large increase in the polar component after grafting, likely due to the ionic nature of PSBMA. Both the OWK and LW methods result in similar values for the overall surface energy. A statistically significant increase ( =0.05) in surface energy is seen independent of calculation method. Table 5 3: Contact angle data for deionized water (DI) glycerol (GL) and diiodomethane (DM) on both LDPE and LDPE PSBMA Sessile Drop Contact Angle ( STD ) D W GL DM LDPE 98 (1.7 ) 84 (3.6 ) 52 (1.6 ) LDPE P SBMA 64 (6.4 ) 52 (6.7 ) 39 (5.4 ) Table 5 4: Surface energy of LDPE and LDPE P SBMA grafted samples as determined using the OWK method Surface Energy OWK: DW & DM (STD) Surface Energy LW (STD) S V P S V d S V S V LDPE 0.4 (0.1) 32.8 (0.6) 33.2 (0.8) 33.2 (1.2) LDPE P SBMA 12.3 (2.8) 34. 2 (1 .8 ) 46.5 (4.6) 47.4 (4.4) ATR FTIR LDPE PSBMA showed significantly altered surface chemistry, as shown in figure 5 5 Peaks unique to the graft include the carbonyl ( 1720 cm 1 ), amide (1650cm 1 ) and sulfoxide (1050 cm 1 ). The results of the quantitative analysis also show a significant increase in graft content, as seen in figure 5 6 This data suggests that there is a substantial amount of P SBMA graft on the surface

PAGE 91

91 Figure 5 5 : ATR FTIR spectra of SMBA grafted LDPE and untreated control. Figure 5 6: Quantitative ATR FTIR data showing the carbonyl ratio (R c ) of LDPE and LDPE PSBMA. Asterisk indicates statistically different groups ( =0.05). 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 LDPE LDPE PSBMA R c LDPE PSBMA LDPE

PAGE 92

92 XPS Figure 5 7 : XPS spectra of LDPE PSBMA LDPE. Sulfur and nitrogen peaks indicate the presence of graft. All values are reported in atomic percent. XPS confirmed in presence of PSBMA graft, as shown in figure 5 7 The substrate is largely po lyethylene, therefore carbon is the main constituent (84.1 atomic %). Oxygen wa s the next most prevalent element (12.6 atomic %) coming from the graft as well as any surface oxidation or contamination The distinct elements for the graft (sulfur and nit rogen) we re detected as well (1.8 atomic % and 1.7 atomic % respectively). The amount of sulfur and nitrogen wa s practically identical as expected considering graft chemistry The graft content based is much lower than those reported in studies where P SB MA was grafted using ATRP (1.7 atomic % compared to 3.8 atomic % [125] ).

PAGE 93

93 Ulva Attachment LDPE had slightly higher number of attached spores compared to PDMSe as seen in Figur e 5 8 LDPE PSBMA showed drastically lower attachment of zoospores with a 92% reduction compared to smooth LDPE (90% relative to the PDMSe). This seems to be in keeping with data presented in the literature for P SBMA grafted glass slides where a 97% red uction was seen [74] The differences in the substrate and reference material may be the cause of the slight difference. Figure 5 8 : Attachment of Ulva Zoospores to PDMSe, LDPE and LDPE P SBMA samples. Error bars represent one STD Single and double asterisks indicate statistically significant groups. Discussion Graft Characterization Sessile drop contact angle dynamic contact angle ATR FTIR and XPS confirm ed the presence of graft. Se ssile drop contact angle shows that the new surface energy of 46.0 mJ/m 2 (up from 33. 5 mJ/m 2 ). Dynamic contact angle showed a statistically significant decrease in advancing and receding contact angle and a significant increase 0 100 200 300 400 500 600 700 800 900 1000 PDMSe LDPE LDPE PSBMA Spore density (no/mm 2 ) Zoospore Attachment **

PAGE 94

94 in slip angle. ATR FTIR sh owed characteristic peaks for several chemical groups found in the graft. XPS indicate d that the surface was 1.7 at % sulfur due to grafting Each characterization method indicated a significant change in both surface energy and surface chemistry has occu rred. Each characterization method showed the presence of graft, but not all show it to the same degree. The ATR FTIR data suggests that there is substantial graft on the surface. The carbonyl ratio was high, whereas XPS seemed to indicate a more limit ed graft. With less than 2 atomic percent sulfur and nitrogen, XPS did not indicate there was much graft at the surface. There are several possible explan ations. T he material surface examined in XPS may have been different under vacuum than at ambient pressure. The more polar groups tend to bury themselves in vacuum, resulting in lower detection [15] Also, the two methods were sampling the surface differently. The ATR FTIR examined 12 spots with eac h scan, but XPS examined only a single spot. Water contact angle data seemed to indicate that some regions have more graft than others. XPS could be examining a region with relatively low graft content while ATR FTIR is averaging together several spots, including both the high and the low. This indicates that the graft is significant but not complete or uniform. Ulva Attachment UV grafted P SBMA was very effective at reducing the attachment of Ulva zoospores. The reduction in a ttachment was similar to values listed in the literature indicating that the graft created by UV polymerization was as effective as those made by ATRP despite having much less graft content Perhaps both values are above some critical point where the graft becomes dominant. Sin ce the UV polymerization takes

PAGE 95

95 much less time ( 8 minutes compared to 4 hours) it would be better for mass production It would be interesting to see if graft properties such as kinetic chain length could be further fine tuned to reduce attachment It may be worthwhile to seek the critical point where graft content begins to become dominant. The LDPE and PDMSe samples both appear to fit with the Baier curve which predicted LDPE would have more spores attach to it than the PDMSe since it lies outside the foul release zone The LDPE PSBMA samples are not in agreement with the Baier curve LDPE PSBMA ( surface energy = 47 mJ/m 2 ) ha d not reached the reduced fouling region O ne would have expect ed these samples to have higher attachment than the LDPE or PDM Se standards based on the Baier Curv e It is also possible that the ionic nature of the surface is not taken into account in the Baier curve. It would be interesting to compare the LDPE PSBMA to a non ionic material with a similar surface energy, such as a polyamide This would test surface energy independent of ionic character. Conclusions T he effectiveness of UV photografting on LDPE using SBMA as a monomer was demonstrated by sessile drop contact angle, ATR FTIR and XPS. The surface energy of the L DPE was increased from 33 mJ/m 2 to 47 mJ/m 2 ATR FTIR of LDPE PSBMA showed carbonyl, amide and sulfoxide peaks characteristic to the P SBMA graft chemistry XPS showed 1.7 atom ic % sulfur and nitrogen on the surface after grafting LDPE PSBMA showed re duced attachment of Ulva by 92% compared to untreated LDPE. The reduction in attachment is similar to reported values for ATRP grafts on glass slides despite have less sur face graft and a shorter grafting time. T he surface energy

PAGE 96

96 and attachment on the L DPE PSBMA sample does not appear to be in agreement with the Baier curve. The ionic nature of the surface grafts may be the cause. This could be examined by testing non ionic surfaces with similar surface energy.

PAGE 97

97 CHAPTER 6 REDUCTION OF U LVA ATTACHMENT O NTO POLYETHYLENE SUR FACES USING BOUND ZINC VIA AN AC RYLIC ACID SURFACE G RAFT Introduction Surface grafts to limit fouling of marine organisms have been largely focused on creating a surface that impedes cell attachment. Hydrophilic grafts with neutral cha rge are effective at this as was demonstrated in chapter 5 Another approach is to use the graft to bind a biocidal agent as commonly done in antifouling paints. One example of a b iocid e used in antifouling paint is tributyl tin. This organometallic c ompound was the antifoulant of choice before it was banned for its systemic effects [58] It has since been replaced by other me tals ions such as zinc and copper [128 130] These coatings work with a slow release of the metal ions. C alcium and sodium ions in sea water will displace the zinc ion, creat ing an artificially high zinc concentr ation at the surface. The zinc ion acts as a biocide, killing organisms that try to attach to the surface [128 130] One of the drawbacks to these paints is that they can flake off of surfaces [129] The flakes of paint then sink to the sea bot tom, maximizing the environmental impact without providing their intended benefit. This has been shown to increase zinc concentrations in seawater, which may cause systemic effects such as seen with tributyl tin. Binding zinc ions with a surface graft ra ther than paint could avoid this problem. Polymers with carboxylic acid functional groups are known to form complexes with metal ions lik of polyethylene and poly (methacrylic acid) which has been cross linked with zinc, is one example. It is commonly used in commercial products such as golf balls. The z inc ions f orm l inks between the carboxylic acid groups of the methacrylic acid mers as shown in figure 6 1. Zinc has also been used to cross link acrylic acid (AA) and maleic acid [131] Such

PAGE 98

98 cross link ing binds the zinc to the polymer with a similar strength to those used in antifouling paints Using a similar method with surface grafts would create a coating which would release zinc much like the antifouling paints. Figure 6 1: Zinc forming cross link s between carboxylic acid groups. It is hypothesized that a su rface graft of poly (acrylic acid) (PAA) can be used to bind zinc to the surface of poly ethylene. This zinc would be released in a seawater environment, as in the antifouling paints. The goal of this the research in this chapter is to create PAA grafted p olyethylene with bound zinc ions (LDPE PAAZn) and evaluate it as an antifouling coating. Methods Sample Preparation Samples were prepared using the same method described in chapter 5. A brief summary is given here. LDPE pellets (Exxon Mobil LD135 MN ( Lot # D10400 )) are melt which are then stacked and melted together with an encased glass slide The resulting sample is cut to size and cleaned in ethanol. Clean, dry samples are placed in 50 mL centrifuge tubes prior to grafting. samples ( poly ( ethylene co methacrylic acid zinc salt) Aldrich ) were prepared using the same procedure with identical temperature and wt % ) methacrylic acid (the manufacturer does not give the amou nt of zinc)

PAGE 99

99 Monomer Preparation AA and benzophenone (BP) are mixed without any additional solvent. The monomer solution has a 0.001:1ratio of initiator to monomer. This concentration is chosen to maximize kinetic chain length. AA is very reactive (hi gh k p ) in photoinitiated polymerizations [92 95] ; therefore a high rate of polymerization can be achieved even with low initiator concentration. T he monomer solution is purged with argon for 10 minutes to drive out any oxygen prior to grafting Grafting Procedure T he ultraviolet ( UV ) grafting wa s done in a similar method to the samples described in chapter 5 with t he most notable difference being the amount of irradiation time. A Lesco CureMax UV chamber wa s purge d with nitrogen for 10 minutes prior to grafting Samples we monomer solution. A three inch diameter circular quartz plate wa s placed on top of the sample. This evenly distribute d the monomer while allowing the UV light to pass through. The samples we re then placed in the center of the chamber and irradiated for 1 minute. After grafting, samples we re soaked in deionized water for 24 hours to remove residual monomer and homopolymer. S amples then were rinsed with deionized water and dried under nitrog en. Clean, dry samples were stored in centrifuge tubes for characterization. Eight samples of P AA grafted polyethylene (LDPE PAA) were grafted using this protocol Four would be taken for zinc treatment while t he other four would serve as controls.

PAGE 100

100 B inding Zinc to the PAA Surface Graft The zinc binding process was based on a cross linking protocol found in the literature [131] PAA g rafted samples we re treated in a basic solution to g ave the AA chains a negative charge to bind the zinc ions Samples were soaked in deionized water. A 1 N sodium hydroxide solution wa s added until the pH of the solution reaches 10. Since the pK a of acrylic acid wa s 4.2, nearly all of the AA groups we re negatively charged. Samples we re soaked in this solution for 24 hours, followed by a rinse with DI water. Zinc was then bound to the surface by soaking in a 0.25 M ZnCl 2 solution for 24 hours. The samples are then rinsed with deionized water to remove any unbound zinc Samples referred to as P AAZn we re then dried under nitrogen and stored in 50 mL centrifuge tubes. Experimental Controls There were 4 control groups: PDMSe, untreated LDPE PAA. PDMSe sample wa s used as a standard because of the extensive use of this material in similar studies [63, 65, 66, 68, 69] Ungrafted LDPE wa s used as a baseline. It controls for the f is analogous to the PAAZn grafted samples because it contains zinc ions bound by carboxylic acid groups. Finally, the LDPE PAA samples controlled for the fouling characteristics of the graft without th e biocidal ions. Four samples of each group were prepared. Three would be sent out for Ulva assays with t he fourth being retained for characterization. This was required for vital characterization methods such as attenuated total reflectance Fourier tra nsformed infrared spectroscopy (ATR FTIR) and X ray photoelectron spectroscopy (XPS). Pressing against the ATR FTIR crystal would leave an impression that would affect Ulva

PAGE 101

101 Spore attachment and the sample would have to be cut into small sections for XPS. O nly the retained sample could be examined using ATR FTIR and XPS for these reasons All four samples were characterized by sessile drop contact angle to ensure the samples were reproducible and that the retained sample was representative of the 3 shipp ed samples. Material Characterization Sessile Drop Water Contact Angle Sessile drop contact angle was done on all samples. Five drops of deionized water (18M resistivity) 2 l in volume were placed on the surface of the sample. The contact angle of the drops was measured using a goniometer. Data was collected from each sample group. Analysis of variance (ANOVA) establish statistical difference between groups (n=3, Dynamic Water Contact Angle The retained samples were analyzed using dynamic contact angle using a tilting method. Dynamic contact angle was done using a Ram Hart goniometer, which used an automated drop dispenser, tilting stage and live electronic imaging. A 30 l drop was placed on the surface. Smaller drops, such as those used in the sessile drop analysis, did not slip on the surface and therefore could not be used. The stage was tilted one deg ree at a time with an image captured after each degree. Tilting was stopped when the drop began to slip and the slip angle was recorded. The last image before slipping is used for analysis using ImageJ software to measure the advancing and receding angle s at the leading and trailing sides of the drop respectively. Five of such measurements were made from each treatment. Analysis of variance (ANOVA) was

PAGE 102

102 Meas urement of Surface Energy The retained samples were given more detailed analysis for determining surface energy. These samples were analyzed using 10 drops (5 l in volume) of deionized water, glycerol and diiodomethane. The relevant physical constants f or these liquids are given in table 5 1 in the previous chapter. C ontact angles of each of these liquids were used to measure surface energy via the Owens Wendt Kaelble (OWK) and Lifshitz van der Waals (LW) acid base methods. The OWK method uses polar a nd non polar liquids to determine the surface energy of a solid. It also determines the polar and dispersive components of the surface energy. This is valuable for analyzing the graft, since we would expect a large increase in the polar component due to the polar nature of acrylic acid The LW method also takes in to account electrostatic and acid/base components. The use of multiple methods for measuring surface energy is common for evaluating a surface [127] Th e variance of the contact angle measurements is carried through the surface energy calculation to es tablish statistical significance. An ANOVA combined with =0.05). ATR FTIR A Perkin Elmer Spectrum One FTIR spectrometer was used with an ATR module which consisted of 2 inch by inch ZnSe crystal with an incident angle of 60. Reported spectra were the average of 20 scans. Each sample was measured 3 times, s ampling different regions each time. All quantitative calculations were made using the average of the three measures for each sample.

PAGE 103

103 Quantitative measures of carboxylic acid content were made by the comparing peak height for the carbonyl peak (H C=O ) to the peak height of the CH 2 bending peak ( ). This method was common for ATR FTIR quantitative characterization and has been described in literature [113, 114] and standard characterization texts [115] The area for each peak was measured using Omnic spectra processing software. The carbonyl ratio ( ) was then measured us ing equation 6 1. An ANOVA combined with =0.05). XPS A section of LDPE PAAZn and were examined using a XPS/E SCA Perkin E lmer PHI 5100 ESCA s ystem with a magnesium anode. A section of the sample approximately 2 cm x 2 cm was cut using a razor blade and immediately taken to the XPS for analysis. A survey scan was taken from 1100 to 0 keV with an incident angle of 45 Atom ic percentages of carbon, oxygen and zinc were calculated using AugerScan software, which integrates peak area and uses literature values for sensitivity factors. T he following peaks were used for quantitative analysis: C1s, O1s and Zn2p3. Ulva Attachmen t Ulva attachment was evaluated using an established protocol described elsewhere [68] Samples were immersed in DI water for 24 hours pri or to testing, immediately removed from the ASW and placed in an assay dish. The dish was then loaded with 10 (6 1)

PAGE 104

104 ml of zoospore solution with a concentration of 1.0x10 6 spores/ml, and samples were incubated in darkness for 45 minutes at 20 C. After incubation, the samples are rinsed to remove any non adhered spores, and then fixed with a 2.5% glutaraldehyde solution in ASW. The fixing solution wa s then removed with sequential ri nses of ASW, 50:50 ASW:DI and DI water. Samples were imaged using an Axiovision 4.8 and a fluorescence microscope, where spores are identified by the autofluorescence of chlorophyll. Each slide was imaged 30 times, with a field of view of 0.15 mm 2 An A establish statistically significant groups (n=90, =0.05). Results Static Water Contact Angle Results from static water contact angle are given in figure 6 2. Contact angle of LDPE PAA and LDPE PAAZn samples we re not statistically different indicating PAA grafting was consistent in both groups Both grafted samples and the control we re statistically different from the untreated LDPE control due to the influence of the carboxylic acid groups. The con tact angle data indicates the grafting process has increased the wettability of the samples; however the reduction in contact angle is not nearly as large as has been reported in the literature PAA grafted polyethylene has been reported with a contact an gle of 31 [114] The addition of the zinc treatment did not appear to have an effect on the wettability of the samples. Dynamic Water Contact Angle Grafting greatly influenced wetting in dynamic contact angle, as shown in table 6 1 The untreated LDPE has a s ignificantly higher advancing ( A ) and receding ( R ) contact angle than any other sample due to its hydrophobic nature. LDPE PAA and

PAGE 105

105 LDPE PAAZn showed no significant difference in advancing or receding contact angle. The control had statistically lower advancing and receding c ontact angle than the other treatments. There was no statistical difference in contact angle hysteresis ( A R ) between the four treatments. and LDPE PAAZn samples each had a significantly higher slip angle than the other treatments. Figure 6 2: Static water contact angle for LDPE LDPE PAA and LDPE PAAZn samples. Error bars show one standard deviation (STD) Black line indicates statistically significant groups. Table 6 1: Results from dynamic contact angle PE PAA and LDPE PAAZn samples. Single and double asterisks indicate statistical significance between treatments ( =0.05). A R Slip Angle A R LDPE 110 (3 ) 84 (1 ) 20 (2 ) 26 (3 ) LDPE PAA 98 (4 ) 77 (7 ) 19 (5 ) 20 (9 ) LDP E PAAZn 98 (6 ) 74 (10 ) 29 (7 ) 24 (4 ) Surlyn 90 (1 ) ** 65 (2 ) 35 (5 ) 24 (2 ) 0 20 40 60 80 100 120 LDPE LDPE PAA LDPE PAAZn Contact Angle ()

PAGE 106

106 Surface grafting reduced the contact angles of all three liquids, as shown in table 6 2 Surface energy, as determined by OWK and LW methods, are reported in table 6 2 The OWK method calculated surface energy using deionized water as the polar liquid and diiodomethane as the non polar liquid. A graphical representation of the data is given in figure 6 3 indicating statistical significance. The LW method calculated surface energy using values for all three liquids the results of which are shown in figure 6 4 There was no statistical difference in surface energy between any samples Each method produces a similar value for the surface energy for each treatment. LDPE PAA and LDPE PAAZ n both had much lower surface energy than the 55 mJ/m 2 reported in literature [132] Table 6 2 : Contact angle data for deionized water (DI) glycerol (GL) and d iiodomethane (DM) for LDPE LDPE PAA LDPE PAAZn and Surlyn samples. Sessile Drop Contact Angle Used for Surface energy Measurements DI (STD) GL (STD) DM (STD) LDPE 98 (1.7 ) 84 (3.6 ) 52 (1.6 ) LDPE PAA 89 (3.6) 80 (3.9) 47 (3.9 ) LDPE PAAZn 84 (9.3 ) 77 (7.7 ) 45 (2.9 ) Surlyn 89 (3.5 ) 78 (1.4 ) 48 (2.4) Table 6 3 : Surface energy (in mJ/m 2 ) of LDPE LDPE PAA, LDPE PAAZn and Surlyn samples as determined using the OWK and LW method s Surface Energy OWK (STD) Surface Energy LW (STD) S V P S V d S V S V LDPE 0.4 (0. 2 ) 32.8 (0.7) 33. 2 (0.8) 33. 2 (1.2) LDPE PA A 1.9 (0.6) 34.3 (1.7) 36. 1 (2.3) 35. 3 (2.5) LDPE PAAZn 3.2 ( 2.9) 34.4 (0 .0 ) 37. 7 (2.9) 36.2 (2.5) Surlyn 2.1 (0.8) 33.3 (0.8) 35. 4 (1.6) 35.5 (1.0)

PAGE 107

107 Figure 6 3: Average surface energy for LDPE, LDPE PAA, LDPE PAAZn and samples as determined by the OWK method Error bars represent one STD Black lines indicate statistically significant groups. Figure 6 4: Average surface energy for LDPE, LDPE PAA, LDPE PAAZn and Surlyn samples as determined by the LW method. Error bars represent one STD Black lines indicate statistically significant groups. 0 5 10 15 20 25 30 35 40 45 LDPE LDPE PAA LDP PAAZn Surface Energy (mJ/m 2 ) Surface Energy Via OWK Method 0 5 10 15 20 25 30 35 40 45 LDPE LDPE PAA LDPE PAAZn Surface Energy (mJ/m 2 ) Surface Energy Via LW Method

PAGE 108

108 ATR FTIR Figure 6 5: Representative ATR FTIR spectrum of LDPE PAAZn LDPE PAA and Figure 6 6: Quantitative ATR FTIR data showing the carbonyl ratio (R c ) of LDPE P AA LDPE P AA Z n an d 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 LDPE PAA LDPE PAAZn Surlyn R c LDPE PAAZn LDPE PAA

PAGE 109

109 The LDPE PAA, LDPE carboxylic acid content, as shown in figure 6 5. The presence of graft is confirmed by the carbonyl peak seen at 1720 cm 1 Both LDPE PAA and LDPE PAAZn grafted samples as is shown in 6 There is no statistical difference between LDPE PAA and LDPE PAAZn samples, indicating that the PAA grafting is consistent. XPS Figure 6 7 : XPS spectrum of Surlyn Carbon, oxygen, and zinc are seen. PAAZn, as shown in figures 6 7 and 6 8 respectively. Quantitative atomic concentrat ion, as shown table 6 3,

PAGE 110

110 This was expected because the PAAZn sample has zinc concentrated at the surface, Figure 6 8 : XPS spectrum of LDPE PAAZn Carbon, oxygen, and zinc are seen. Table 6 4: A tomic concentration of carbon, oxygen and zinc from XPS analysis. Element (atom %) C O Zn Surlyn 87.2% 12.4% 0.4% LDPE PAAZn 86.3% 12.3% 1.5% Ulva Attachment LDPE had slightly higher number of attached spores compared to PDMSe as shown in Figure 6 of the other groups: a more than 100% increase compared to LDPE. LDPE PAA and LDPE PAAZn have greatly reduced spore atta chment. LDPE PAAZn reduced attachment by 87% relative to untreated LDPE (84% compared to PDMSe) and LDPE

PAGE 111

111 PAA reduced attachment of 91% compared to LDPE (89% compared to PDMSe). LDPE PAAZn and LDPE PAA groups were not statistically different. This indica ted that the zinc was not a contributing factor, and the reduction in attachment was likely caused by PAA. Spore density on PAA and PAAZn was similar to PAA homopolymer recently reported in the literature [133] Figure 6 9: Zoospore attachment to PDMSe, LDPE, LDPE PAAZn, LDPE PAA and lines indicate which groups are statistically different. Discussion Graft Characterization Sessile Drop Water Contact Angle The sessile drop contact angle showed that grafting was effective at increase wettability of the surface, with statistically lower contact angles for LD PE PAAZn and LDPE PAA grafted samples. The water contact angle of the grafted samples was grafted on LDPE in the literature Perhaps the bulk reaction of the acrylic acid allows for 0 200 400 600 800 1000 1200 1400 1600 1800 2000 PDMSe LDPE LDPE PAAZn LDPE PAA Surlyn Spore density (#/mm 2 )

PAGE 112

112 competing reactions such as chain transfer to monomer to reduce grafting efficiency by creating homopolymer. Perhaps using monomer dissolved in a solvent such as acetone or methanol would improve this. Dynamic Contact Angle Dynamic contact ang le does show some significant differences between the samples. The carboxylic acid content reduces the advancing contact angle for the PAA, contact angle and slip angle, which is likely caused by the higher carboxylic acid content compared to LDPE PA A and LDPE PAAZn There is no statistical di fference in hysteresis between any of the samples. Surface Energy There is not a statistical difference in the surface energy of the grafted samples. This is largely due to the high variance of the PAAZn samp le. Examining more samples per treatment might reduce variance and enable a statistical difference to be seen. The Baier curve would predict that these samples all will have similar fouling characteristics since the surface energies are all very similar ATR FTIR ATR FTIR LDPE PAA and LDPE PAAZn surfaces. The LDPE PAA and LDPE PAAZ n samples do not have statistically different graft content, indicating that any difference between them would be caused by the zinc This supports the contact angle data, which showed little change in surface energy between the untreated and grafted groups. The LDPE PAA and LDPE PAAZn grafted samples have relatively low carbonyl content, much lower than the Sur It is important to consider the sampling depth of ATR FTIR is

PAGE 113

113 much greater than that of sessile drop contact angle. This is why two analogous materials such as LDPE PAAZn and with similar surface energies can have very different car bonyl content XPS The and LDPE PAAZn samples have similar amounts of carbon and oxygen but differ greatly in zinc content, where the LDPE PAAZn sample has considerably more zinc It wa s expected that the high zinc content of LDPE PAAZn would g reatly reduce the adhesion of Ulva zoospores. Characterization Summary The combination of these three methods creates a detailed description of the surface. LDPE PAA and LDPE PAAZn samples were statistically different from LDPE in static and advancing contact angle, but samples are not as hydrophilic as some have reported in the literature. receding contact angle, but there is no statistical difference in contact angle hysteresis. The ATR FTIR spectra also show that while the grafting process has altered surface chemistry, the surface grafting is limited However, the XPS data shows that there is enough graft to bind zinc which was the objective of the grafting process It was expected that the presence of zinc would cause a reduction in a ttachment with the LDPE PAAZn sample having the lowest a ttachment The samples without zinc would be expected to have similar Ulva spore attachment since th ey have similar surface energy Ulva Attachment Ulva attachment results were very different from what was expected as the amount of attach ed zoospores varies widely despite generally similar surface energies.

PAGE 114

114 had more than twice as many spores attached than any other group therefore the zinc present did not prevent fouling. There was no statistical difference between the LDPE PAA and the LDPE PAAZn samples, further indicating that the zinc was not influencing fouling. It may be that the zinc grafted to the LDPE PAAZn samples was removed in the preparation of the slides for the attachment studies. The samples were exposed to seawa ter prior to the attachment assay ; p erhaps a significant portion of the zinc was lost in that time frame. A different binding agent for zinc might be a path for future work The fouling reduction of LDPE PAA is particularly interesting. The reduction i n a ttachment for LDPE PAA alone was similar to the sulfobetaine grafted LDPE (LDPE SBMA) seen in chapter 5 suggesting that PAA might be a quicker, cheaper graft. The ionic nature of PAA may be the issue since it is largely deprotonated at neutral pH (pK a =4.2). The anionic nature of PAA, however, would attract positively charged organisms, such Cobetia marina to the surface. It would be interesting to see if the antifouling properties of PAA can be extended to other organisms. It may also be the case tha t the UV polymerization is influencing attachment. The UV grafted LDPE PAA and LDPE PAAZn samples each have similar a ttachment to the UV grafted PSBMA samples examined in chapter 5. These grafted chains create a nano scale topography on the surface [15, 101] Perhaps the organisms are responding to this t opography more than surface energy or surface chemistry. This could be evaluated by using a more hydrophobic surface graft such as Conclusions A method for grafting of PAA to LDPE was demonstrated and confirmed using sessile drop contact angle and ATR FT IR A method for binding zinc to the acrylic acid

PAGE 115

115 graft was established and confirmed using XPS Grafting did not, however, produce a statistically significant increase in surface energy. The zinc content of the PAAZn sample and the sample did not appear to reduce the attachment of zoospores as hoped. PAA graft on its own, however, greatly inhibited attachment of the zoospores compared to the untreated control Samples d id not appear to follow the Baier curve with highly va riant attachment on surfaces with similar surface energy was enormously different from the other samples, despite having similar surface energy. Future work will focus on understanding how such a wide range of attachment could be seen from s amples with similar surface free energy.

PAGE 116

116 CHAPTER 7 FUTURE WORK Adhesion to Polyethylene Fibers for Composites Optimizing Grafting Procedure The results of chapter 3 showed that even a relatively slight amount of graft could greatly increase adhesion of Ultra high molecular weight polyethylene (UHMWPE) fibers It seems likely that an increase in the amount of graft will further increase adhesion. There are several methods for increasing graft content First is by increasing the repetitions of graf ting. A 3x graft was effective, but one would expect 5x or 10x to be much more effective. Second, increasing grafting time should increase graft content. It appears 5 seconds is not enough time for all of the monomer to be consumed. Longer times, such as 1 minute, should produce considerably more graft. Finally, more concentrated solutions of monomer should produce graft more quickly. Increasing the concentration from 4 M to 6 or 8 M should dramatically increase the rate of polymerization and the amou nt of graft Cross linking GMA using Ethylene Diamine Chapter 3 showed how PAAGMA could improve bonding between UHMWPE fibers and epoxy. The GMA groups in the graft might serve another purpose as they could be covalently cross link ed using ethylene diami ne These covalent cross links would greatly reduce creep, which would be a substantial improvement There may, however, be a tradeoff between cross link ing and adhesion to the matrix. Since the epoxy rings would be consumed by cross link ing, fewer wou ld be available to bond to the matrix. It is possible that this could be optimized by varying the

PAGE 117

117 concentration of GMA in the monomer or by controlling the amount of surface cross link ing. It would be vital to understand the reaction kinetics of the cros s linking reaction optimize both adhesion and cross linking. Adhesion to Polyethylene in the Body Grafting onto Commercial UHMWPE Sutures The work presented in chapter 4 used smooth polyethylene to demonstrate the effect of gelatin on cell adhesion. S utur es themselves have curved surfaces made from woven fibers Th is additional surface topography will influence cell adhesion. The goal of this work was to evaluate adhesion of cells based on surface grafting, so chemistry was examined independent of topogr aphy. It would be interesting to see if grafting on actual sutures magnifies or minimizes the increased adhesion seen in chapter 4. Examining Other Proteins and Growth Factors Gelatin was an excellent starting example, but other proteins, such as fibron ectin or collagen could also be bound using PGMA. Proteins which have not been denatured will more closely mimic the extracellular matrix (ECM), but the binding process may be more complicated. Denaturing of proteins is possible during covalent bonding, which might be undesirable. It would be interesting to see if covalent bonding influences the functionality of fibronectin or collagen. Both fibronectin and collagen are much more expensive than gelatin, which brings into question whether any improvement in adhesion is worth the increase in cost. These bound proteins could be further developed by adding growth factors. Combining growth factors with surface proteins has been shown as a way to improve adhesion and influence cell behavior. Basic fibroblast growth factor and bone morphogenic proteins could also increase adhesion and improve wound healing.

PAGE 118

118 Stem Cell Differentiation The cell studies presented in chapter 4 show higher immediate cell attachments The y do not, however, explore how the cells grow develop and differentiate. The growth and differentiation of cells will greatly influence the viability of a potential product. It would be ideal if the suture surface would direct differentiation into chondrocytes and osteoblasts to form a new tendon/ bone interface. Adhesion to Polyethylene in the Marine Environment Optimizing Graft Kinetics for LDPE P SBMA The graft concentration in chapter 5 was very effective, but perhaps dif ferent graft concentrations can further reduce adhesion Tuning graft ki netics should make it possible to create grafted chains with different molar masses which should be an important factor in surface behavior [134] S hort chain grafts should reduce adhesion better than bulky, high molar mass chains. Kinetic chain length can be shortened by increasing initiator concentration or by addi ng a chain transfer agent, such as a quinone. The effect of changes in graft chain length can be visualized using atomic force microscopy [101] and it may be possible to relate adhesion to the radius of gyration of the grafted chains. Examining the Effectiveness of LDPE P SBMA for Limiting Att achment of Other Marine Organisms Chapter 4 also focuses on a single fouling organism. P SBMA grafts have been very effective at limiting the attachment of Ulva however these algae are not the only sources of fouling of concern for the polyethylene in the marine environment. B arnacles and mussels are a much bigger concern as their larger size makes them more problematic to the cooling system. There has not yet been a study of how P SMBA

PAGE 119

119 grafts effect the attachment of multiple organisms or a study in an actual marine environment If P SBMA grafts are an effective means of reducing fouling across multiple species then it may be a valuable replacement for toxic anti foulants. Examining Other Polymers for Binding Zinc Chapter 6 only examined a single method for binding zinc. Acrylic acid is only one of several monomers which could be used. Other acid groups, such as 2 a crylamido 2 methylpropane sulfonic acid (AMPS) or maleic acid also have a net negative charge and could potential ly bind zinc better than a crylic acid. Since it appeared that the zinc on the PAA surface was either not substantial enough or too weakly bonded, increasing the amount of zinc and the relative strength of its bond is of vital importance. AMPS is a stronger acid than AA and may b e able to bind zinc more strongly. Maleic acid would have two carboxylic acid groups per monomer unit. For a similar graft molecular weight, it could bond twice the zinc of acrylic acid. This may be beneficial because an increased amount of zinc might f urther reduce adhesion. Both of these monomers have been shown to bind metal ions and could provide a valuable comparison to the data presented in chapter 6. Combining Graft with Surface Topography E as discussed in chapter 2, has been extremely effective at limiting the fouling of marine organisms. It is possible to make these micron scale f eatures in LDPE, as shown in figure 7 1 and described in appendix A The combination of surface grafts and surf ace topography could greatly reduce fouling. Grafts combined with surface roughness have been shown to create

PAGE 120

120 Figure 7 superhydrophobicity in polyethylene [114] Early attempts were unable to be evaluated here due to over grafting; where the grafted layer was so large that it nullified the pattern below it, as shown in Figure 7 2 Fine tuning of the grafting process is needed to compensate for this. By changing grafting time and kinetics, it should be possible to combine surface grafting without damaging or obscuring surface topography. Figure 7 2: Surface graft ing of PSBMA covers up topography. Unaffected Sharklet is visible in the upper left corner, but graft overwhelms topography on the right edge. 20 m Working Distance= 8 mm; Accelerating Voltage = 15 keV Working Distance= 8 mm; Accelerating Voltage = 15 keV Unaffected Sharklet Sharklet covered up by Graft

PAGE 121

121 APPENDIX A CREATING PATTERNED POLYETHYLENE USING A MELT PRESSIN G PROCESS Introduction A ttachment of Ulva zo ospores to smooth samples with graft was explored extensively in chapters 5 and 6, but the effect of engineered topography is also an important consideration. M arine animals like sharks and crabs have topographical features on their exterior to limit foul ing. Engineered topography has been shown to greatly reduce the fouling of poly (dimethyl siloxane) elastomer (PDMSe ) [63, 66, 68, 69] Expanding to include other materials would make it possible to extend the mo del to include the effects of modulus and other material properties such as crystallinity or phase segregation. The goal of the work summarized here is to create and characterize patterned surfaces of polyethylene. It is possible to create micron scal e features in thermoplastic polymers using either heat, pressure or both. Despa et al. made 200 m pillars of high density polyethylene using an injection molding process with patterned silicon wafers [135] Feng et al. was able to mold 10 20 micron hairs using a silicone mold of a lotus leaf [136] thermal embossing process. The dimensions of the SK pattern are vital to its effectiveness The most successful data has been seen with features that are 2 microns wide and s eparated by 2 microns. Larger spacing allows the spores to fit in between features more easily. For this reason, surfaces had to be carefully examined and characterized before they could be assayed.

PAGE 122

122 A patterned surface can be characterized using sessil e drop water contact angle. The Young equation mentioned in chapter 2 assumes a perfectly flat surface however, a roughened surface can affect contact angle in 2 regimes : Wenzel wetting and Cassie Baxter wetting [67, 137, 138] In the Wenzel regime, the liquid still wets the entire surface. T he roughness ( ) of the material will affect the contact angle differently in the Wenzel regime depending on its hydrophilic or hydrophobic nature. If the actual contact angle ( ) is less than 90 additional roughness will reduce the apparent contact angle ( ) as shown in equation A 1 If the contact angle is greater than 90 additional roughness will increase the contact angle. The Cassie Baxter regime is when surface wetting is incomplete, i.e., there exist regions where the drop is contacting air rather than a solid as shown in figure A 1. ion given in equation A 2, [137, 138] applies will increase with decreasing which is the fraction of solid liquid interface. For engineered topographies, is the area of the feature tops. In this regime, contact angle increases with decreasing solid liquid interactions. For very low water contact angle can be above 150 at which point the surface is considered superhydrophobic. This superhydrophobic region is of great interest for marine fouling and other applications [114, 136, 139] beca use a superhydrophobic surface exhibits weak adhesion and self cleaning properties (A 1) (A 2)

PAGE 123

123 Figure A 1 : Incomplete (Cassie Baxter) and complete (Wenzel) wetting of a rough surface. Methods Sample Production There are two methods of s ample production: Films patterned off silicon wafers and films patterned of f PDMSe films. In each method, a smooth LDPE film wa s melted for patterning in a process identical to those described in chapter 5 and chapter 6. Both are treated with hexamethyl disilazan e (HMDS) with the intent of limiting adhesion between the LDPE and the mold. While the PDMSe molds were considered preferable, both methods produced interesting results. Figure A 2 : Creating patterned LDPE using a si licon wafer. Heat Source 10.0 kg Weight PET Film LDPE Patterned Sili con Wafer Aluminum Plate

PAGE 124

124 The method for creating samples from silicon wafers is depicted in figure A 2. The LDPE film is heated to 180 C and held under pressure for 15 minutes. Cooling is done slowly to prevent the wafer from shattering due to thermal shock. Cooling in air takes about 2 hours. The resulting film is carefully removed from the wafer and attached to a glass slide using double sided tape. These films were positively buoyant; therefore they could not be studied for Ulva zoospore attachment. The method for creating samples from PDMSe molds as shown in figure A 3, is similar to the smooth samples described in chapter 5. The main difference is that a patterned mold has replaced the PET sheet and the silicon wafer These samples were heated to 180 C and held under a 1 0 kg load for 15 minutes then quenched in water and removed from the mold. T he embedded glass slide makes them negatively buoyant just as with the smooth samples used in Chapter 5 and Chapter 6 Figure A 3 : Creating patterned LDPE using a PDMSe mold. Spacer Glass Slide PDMSe Mold Heat Source Heat Source LDPE films Al Plates Spacer PET Sheet

PAGE 125

125 Characterization Patterns were characterized by scanning electron microscopy (SEM), optical microscopy and sessile drop contact angle. SEM was used to determine feature dimensions. Patterned sections were sput ter coated with AuPd for 60 seconds at 45 mA prior to SEM analysis. A JEOL SEM 6400 was used at 15 keV accelerating voltage. Feature dimensions were measured using ImageJ software with the scale bar being used as a standard dimension. Optical microscopy was used to determine pattern fidelity (the % of features with no damage or defects). Contact angle was measured using 6 l drops of deionized water using a goniometer. Measurements were taken with t he features parallel to the field of view Results SE M Micrographs of samples produced from silicon wafers are presented in figure A 4 Results were inconsistent : Some regions had very high fidelity and properly shaped features while others were misshapen or completely ripped out. Due to the irregularly s haped features, no dimensional analysis was done There was also a problem with wafers breaking during the removal of the LDPE. The wafers were simply too fragile to withstand removal. S ilicon wafers were considered inferior to PDMSe molds for this reas on The ripped up features could prove to be valuable, however, as a potential superhydrophobic surface. Patterned LDPE made using PDMSe films were much more successful than those made against silicon wafers, as shown in figure A 5. Both positive and rec essed

PAGE 126

126 than samples made from silicon wafers. Since features were consistent, feature dimensions could be measured. Figure A 4 : SEM micrographs of LDPE patterned using a s ilicon wafer as a mold. Top down (A) and edge views (B D) indicate that feature damage is common and the shape of feature varies widely. A 1. w idth s pacing the empty distance between them. Feature widths are smaller than 2 microns, with the spacing making up most of the difference. This minor change in feature dimensions is a major problem, as the spores wi ll now be able to settle in between features more easily. It is necessary to correct for this with new PDMSe molds.

PAGE 127

127 Figure A 5 : Positive (Top down (A) and edge view (B)) and negative (Top down (C) and edge view (D)) SK patterns made with PDMSe molds. Table A 1: Feature dimensions of positive and recessed Sharklet. Positive Recessed Spacing Width Height Spacing Width Height ( m) ( m) ( m) ( m) ( m) ( m) Sample 2.5 1.3 2.8 2.2 1.7 3.9 STD 0.18 0.08 0.17 0.10 0.07 0.10 Pattern Fidelity Feature fidelity of LDPE patterned off of PDMSe was excellent, as shown in figure A 6. Fifteen of such images were taken at 400x magnification for each sample. Fidelity was measured by counting the number of defects in an image and dividing by the total number features in the image. The total number of features was determined by the

PAGE 128

128 product of the number of rows and columns. Representative values for samples of 2. Figure A 6 : Optical micrograph of r ecessed SK in LDPE. The recessed SK has higher fidelity because the features have more structural support. Positive SK features are isolated and can be deformed or damaged during mold release and subsequent handling. The recessed SK features are all conn ected, providing support and strength This added stability results in higher feature fidelity. Table A 2: Fidelity of features in positive and recessed SK patterned LDPE using PDMSe molds. Pos i tive Recessed %Good 98.3% 99.6% STD 1.68% 0.33%

PAGE 129

129 Cont act Angle Samples with topography have a statistically significant increase in contact angle compared to the smooth sample as seen in figure A 7 and table A 3 This is due to wetti ng in the Cassie Baxter regime, since Wenzel wetting could not account for the sharp increase in contact angle The positive SK has a lower value than the recessed SK, resulting in a higher water contact angle. The LDPE sample cast from the silicon mold appears to be in the superhydrophobic regime. This is becaus e of the hairs seen in the SEM ( Figure A 4) create a very low value. If t hese superhydrophobic samples could be reproduced consistently, they would be an excellent low fouling surface It may be desirable to create silicon molds to deliberately pull out the features, as this could make the superhydrop hobic surfaces more reprod ucible Figure A 7 : Representative water drops on surfaces of smooth LDPE (A), + SK ( B), SK (C) and +SK made from PDMSe (D). Table A 3: Contact angles for s mooth and patterned surfaces of LDPE. Smooth +SK SK +SK From Si Wafer Average 94.0 145 .7 130.2 157.5 STD 2.7 1.1 4.1 5.1

PAGE 130

130 Discussion The optical and electron micrographs show that it is possible to pattern LDPE with SK on the scale of a few microns. The PDMSe molds were much better than the silicon molds at creating consistent patterns wi th high fidelity. SEM micrographs show consistent, well formed features for the samples patterned against PDMSe. The samples patterned against silicon have misshapen features and hairs extending from the surface. It is possible that the LDPE adheres mo re strongly to the silicon wafers than to the PDMSe molds and this could be causing the features to deform and fracture during removal T hermal expansion mismatch during cooling may also put pressure on the features. This would be minimized in PDMSe mol ds due their lower modulus. The consequence of damaged is the superhydrophobic nature of the hairy LDPE surfaces which could be extremely low fouling if they could be reproduced. The samples molded off of PDMSe have excellent feature fidelity, but poor f eature dimensions. The features are narrower and the spacing is too wide. These would not be expected to reduce fouling, since the zoospores could more easily fit down inside the features. This must be corrected by creating PDMSe mold with slightly diff erent dimensions. Summary It is possible to create patterned LDPE samples using a thermal process. Both silicon and PDMSe were used as molds. PDMSe was shown to be much better in terms of pattern consistency fidelity and feature shape. The width and sp acing of the features was an issue. The narrow features and wider spacing resulted in spores being able to settle in between features more easily. New PDMSe molds need to be used to

PAGE 131

131 correct for this. Silicon molds created LDPE samples with missing featu res, pointed features and hairs This resulted in a superhydrophobic surface but the nature of the pattern is inconsistent. This could potentially be overcome by designing new wafers and molds.

PAGE 132

132 LIST OF REFERENCES [1] Piringer OG, Baner AL. Platic Packaging: Interactions with Food and Pharmaceuticals 2ed: Wiley VCH; 2008. [2] Brennan A. Surface Modification of Polyethylene Fibers for Enhanced Performance in Composites. Trends in Polymer Science. 1995;3:12 21. [3] Baier RE, Meyer A E. Surface Analysis of Fouling Resistant Marine Coatings. Biofouling. 1992;6:165 80. [4] de Gennes P G, Brochard Wyart F, Quere D. Capillarity and Wetting Phenomena: Drops, Bubbles, Pearls, Waves. New York, NY: Springer Verlag; 2004. [5] Townsin RL. The S hip Hull Fouling Penalty. Biofouling. 2003;19:9 15. [6] Young T. An Essay on the Cohesion of Fluids. Philosophical Transactions of the Royal Society of London. 1804;95:65 87. [7] Good RJ. Contact angle, wetting, and adhesion: a critical review. Journal of Adhesion Science and Technology. 1992;6:1269 302. [8] Owens DK, Wendt RC. Estimation of the surface free energy of polymers. Journal of Applied Polymer Science. 1969;13:1741 7. [9] Amornsakchai T, Doaddara O. Grafting of Acrylamide and Acrylic Acid on Poly ethylene Fiber for Improved Adhesion to Epoxy Resin. Journal of Reinforced Plastics and Composites. 2008;27:671 82. grafted ultra high molecular weight polyethylene for biomedical applicati ons. Chemical Papers. 2008;62:580 8. [11] Silva R, Muniz EC, Rubira AF. Multiple hydrophilic polymer ultra thin layers covalently anchored to polyethylene films. Polymer. 2008;19:4066 75. [12] Park S J, Jin J S. Effect of Corona Discharge Treatment on the Dyeability of Low Density Polyethylene Film. Journal of Colloid and Interface Science. 2001;236:155 60. [13] Allan JM, Dooley RL, Shalaby S, W. Surface Phosphonylation of Low Denisty Polyethylene. Journal of Applied Polymer Science. 2000;76:1870 5. [14] Ch anunpanich N, Ulman A, Malagon A, Strzhemechny YM, Schwarz SA, Janke A, et al. Surface Modification of Polyethylene Films via Bromination: Reaction of Brominated Polyethylene with Aromatic Compounds. Langmuir. 2000;16:3557 60. [15] Arnold JJ, Brennan AB. H ydroperoxide initiated grafting onto ultra high modulus polyethylene fibers. Polymeric Materials Science and Engineering. 1996;75:337 8.

PAGE 133

133 [16] Lanauze JA, Myers DL. Ink Adhesion on Corona Treated Polyethylene Studied by Chemical Derivatization of Surface Fu nctional Groups. Journal of Applied Polymer Science. 1990;40:595 611. [17] Courval GJ, Gray DG, Goring DAI. Chemical Modification of Polyethylene Surfaces in Nitrogen Corona. Journal of Polymer Science: Polymer Letters Edition. 1976;14:231 5. [18] Kato K, Uchida, E., Kang, E., Uyama, Y., Ikada, Y. Polymer surface with graft chains. Progress in Polymer Science. 2003;28:209 59. [19] Applications of Dyneema. 2008. [20] Moon SI, Jang J. The effect of the oxygen plasma treatment of UHMWPE fiber on the transverse properties of UHMWPE Fiber/vinylester composites. Composites Science and Technology. 1998;59:487 93. [21] Afshari M, Sikkema DJ, Lee K, Bogle M. High Performance Fibers Based on Rigid and Flexible Polymers. Polymer Reviews. 2008;48:230 74. [22] Muraoka Y, Rich MJ, Drzal LT. Sulfonation of UHMW PE Fibers for Adhesion Promotion in Epxoy Polymers. Journal of Adhesion Science and Technology. 2002;16:1669 85. [23] Vaisman L, Gonzalez MF, Marom G. Transcrystallinity in brominated UHMWPE fiber reinforced HDPE com posites: morphology and dielectric properties. Polymer. 2003;44:1229 35. [24] Mader E. Study of Fibre Surface Treatment For Control of Interphase Properties in Composites. Composites Science and Technology. 1997;57:1077 88. [25] Silverstein MS, Breuer O. A dhesive Properties and Failure of Etched UHMW PE Fibres. Journal of Materials Science. 1993;28:4718 24. [26] Tissington B, Pollard G, Ward IM. A Study of the Influence of Fibre/Resin Adhesion on the Mechanical Behavior of Ultra High Modulus Polyethylene Fr ibre Composites. Journal of Materials Science. 1991:82 92. [27] Anifantis NK, Kakavas PA, Papanicolaou GC. Thermal Stress Concentration Due to Imperfect Adhesion in Fiber Reinforced Composites. Composites Science and Technology. 1997;57:687 96. [28] Number of Physician Visits for Rotator Cuff Problems. National Ambulatory Medical Care Survey: National Center for Health Statistics; 1998. [29] Rotini R, Fini M, Giavaresi G, Marinelli A, Guerra E, Antonioli D, et al. New perspectives in rotator cuff tendon re generation: review of tissue engineered therapies. Chir Organi Mov. 2008;91:87 92.

PAGE 134

134 [30] Matsen FA. Rotator Cuff Failure. New England Journal of Medicine. 2008;358:2138 47. [31] Ejnisman B, Andreoli C, Soares B, Fallopa F, Peccin M, Abdalla R, et al. Interv entions for tears of the rotator cuff in adults (review). Cochorane Database of Systematic Reviews. 2004:1 26. [32] Aurora A, McCarron J, Iannotti JP, Derwin K. Commercially available extracellular matrix materials for rotator cuff repairs: State of the ar t and future trends. Journal of Shoulder and Elbow Surgery. 2007;16:S171 S8. [33] Castagna A, Garofalo R, Conti M, Borroni M, Snyder SJ. Arthroscopic Rotator Cuff Repair Using a Triple Loaded Suture Anchor and a Modified Mason Allen Technique (Alex Stitch) Arthroscopy: The Journal of Arthroscopic and Related Surgery. 2007:4. [34] Huijsmans PE, Pritchard MP, Berghs BM, Rooyen KSv, Wallace AL, Beer JFd. Arthroscopic Rotator Cuff Repair with Double Row Fixation. The Journal of Bone and Joint Surgery (American ). 2007;89:1248 57. [35] Deutsch A. Developments in Arthroscopic Rotator Cuff Repair. US Musculoskeletal Review. 2006. [36] Abbi G, Espinoza L, Odell T, Mahar A, Pedowitz R. Evaluation of 5 Knots and 2 Suture Materials for Arthroscopic Rotator Cuff Repair: Very Strong Sutures Can Still Slip. Arthroscopy: The Journal of Arthroscopic and Related Surgery. 2005;22:38 43. [37] Boehm TD, Werner A, Radtke S, Mueller T, Kirschner S, Gohlke F. The effect of suture materials and techniques on the outcome of repair of the rotator cuff: a Prospecive, Randomized study. Journal of Bone and Joint Surgery [Brittish]. 2005;87 B:819 23. [38] Sclamberg SG, Tibone JE, Itamura JM, Kasraeian S. Six month magnetic resonance imaging follow up of large and massive rotator cuff repai rs reinforced with porcine small intestinal submucosa. Journal of Shoulder and Elbow Surgery. 2004;13:538 41. [39] Cummins CA, Murrell GAC. Mode of failure for rotator cuff repair with suture anchors identified at revision surgery. Journal of Shoulder and Elbow Surgery. 2003;12:123 33. [40] Buess E, Steuber K U, Waibl B. Open Versus Arthroscopic Rotator Cuff Repair: A Comparative View of 96 Cases. Arthroscopy. 2002;21:597 604. [41] Ruotolo C, Nottage WM. Surgical and nonsurgical management of rotator cuff t ears. Arthroscopy. 2002;18:527 31.

PAGE 135

135 [42] Kato K, KIikumura Y, Yamamoto M, Tomita N, Yamada S, Ikada Y. Collagen immobilization onto the surface of artificial hair for improving the tissue adhesion. Journal of Adhesion Science and Technology. 2000;14:635 50. [43] Wang Y, Ke Y, Ren L, Wu G, Chen X, Zhao Q. Surface engineering of PHBV by covalent collagen immobilization to improve cell compatibility. Journal of Biomedical Materials Research Part A. 2009;88A:616 27. [44] Mazzocca AD, McCarthy MB, Arciero C, Jhav eri A, Obopilwe E, Rincon L, et al. Tendon and bone responses to a collagen coated suture material. Journal of Shoulder and Elbow Surgery. 2007;16:222S 30S. [45] Zheng MH, Hinterkeuser K, Solomon K, Kunert V, Pavlos NJ, Xu J. Collagen Derived Biomaterials in Bone and Cartilage Repair. Macromolecular Symposia. 2007;253:179 85. [46] Ma Z, Gao C, Gong Y, Jiacong S. Cartilage tissue engineering PLLA scaffold with surface immobilized collagen and basic fibroblast growth factor. Biomaterials. 2005;26:1253 9. [47] Rocha LB, Goissis G, Rossi MA. Biocompatibility of Anionic Collagen Matrix as Saffold for Bone Healing. Biomaterials. 2002;23:449 56. [48] Ratner BD, Hoffman AS, Schoen FJ, Lemons JE. Biomaterials Science: An introduction to Materials in Medicine. New Yor k, NY: Elsevier Academic Press; 2004. [49] Xu FJ, Wang ZH, Yang WT. Surface Functionalization of Polycaprolactone Films via surface initiated atomTransfer Radical Polymerization for Covalently Coupling Cell Adhesive Biomolecules. Biomaterials. 2010;31:3139 47. [50] Olenych SG, Moussallem MD, Salloum DS, Schlenoff JB, Keller TCS. Fibronectin and Cell Attachment to Cell and Protein Resistant Polyelectrolyte Surfaces. Biomacromolecules. 2005;6:3252 8. [51] Salloum DS, Olenych SG, Keller TCS, Schlenoff JB. Vasc ular Smooth Muscle Cells Directed Adhesion and Growth. Biomacromolecules. 2004;6:161 7. [52] Zhang Z, Chao T, Chen S, Jiang S. Superlow Fouling Sulfobetaine and Carboxybetaine Polymers on Glass Slides. Langmu ir. 2006;22:10072 7. [53] Champ MA. A review of organotin regulatory strategies, pending actions, related costs and benefits. The Science of the Total Environment. 2000;258:21 71. [54] Schultz MP, Bendick JA, Holm ER, Hertel WM. Economic impact of biofouli ng on a naval surface ship. Biofouling: The Journal of Bioadhesion and Biofilm Research. 2011;27:87 98.

PAGE 136

136 [55] Ricciardi A, MacIsaac HJ. Recent mass invasion of the North American Great Lakes by Ponto Caspian species Trends in Ecology & Evolution. 2000;15: 62 5. [56] Yebra DM, Kiil S, Dam Johansen K. Antifouling technology -past, present and future steps towards efficient and environmentally friendly antifouling coatings. Progress in Organic Coatings. 2004;50:75 104. [57] Bulteel J. A Letter Written to the P ublisher Concerning a New Way, by an English Manufacture, to Preserve the Hulls of Ships from the Worm, &c; Better for Sailing and More Cheap and Durable than Any Sheathing or Graving Hitherto Used. Philosophical Transactions. 1673;8:6192 4. [58] Evans SM. TBT or not TBT: that is the Question. Biofouling. 1999;14:117 29. [59] Ruiz JM, Bachelet G, Caumette P, Donard OFX. Three decades of tributyltin in the coastal environment with emphasis on Arcachon Bay, France. Environmental Pollution. 1996;93:195 203. [ 60] Baier RE, Shafrin EG, Zisman WA. Adhesion: Mechanisms That Assist or Impede It. Science. 1968;162:1360 8. [61] Magin CM, Cooper SP, Brennan AB. Non toxic antifouling strategies. Materials Today. 2010;13:36 44. [62] Baier R. Surface behaviour of biomat erials: The theta surface for biocompatibility. Journal of Materials Science: Materials in Medicine. 2006;17:1057 62 62. [63] Magin CM, Long CJ, Cooper SP, Ista LK, Lopez GP, Brennan AB. Engineered antifouling microtopographies: the role of Reynolds number in a model that predicts attachment of zoospores of Ulva and cells of Cobetia marina Biofouling: The Journal of Bioadhesion and Biofilm Research. 2010;26:719 27. [64] Bers AV, Wahl M. The Influence of Natural Surface Microtopographies on Fouling. Biofoul ing: The Journal of Bioadhesion and Biofilm Research. 2004;20:43 51. [65] Schumacher JF, Long CJ, Callow ME, Finlay JA, Callow JA, Brennan AB. Engineered Nanoforce Gradients for Inhibition of Settlement (Attachment) of Swimming Algal Spores. Langmuir. 20 08;24:4931 7. [66] Schumacher JF, Carman ML, Estes TG, Feinberg AW, Wilson LH, Callow ME, et al. Engineered antifouling microtopographies effect of feature size, geometry, and roughness on settlement of zoospores of the green alga Ulva. Biofouling. 2007; 23:55 62. [67] Wenzel RN. Resistance of Solid Surfaces to Wetting by Water. Industrial & Engineering Chemistry. 1936;28:988 94.

PAGE 137

137 [68] Carman ML, Estes TG, Feinberg AW, Schumacher JF, Wilkerson W, Wilson LH, et al. Engineered antifouling microtopographies correlating wettability with cell attachment. Biofouling: The Journal of Bioadhesion and Biofilm Research. 2006;22:11 21. [69] Long CJ, Schumacher JF, Robinson PAC, Finlay JA, Callow ME, Callow JA, et al. A model that predicts the attachment behavior of Ulva linza zoospores on surface topography. Biofouling: The Journal of Bioadhesion and Biofilm Research. 2010;26:411 9. [70] Zhou M, Liu H, Kilduff JE, Langer R, Anderson DG, Belfort G. High Throughput Membrane Surface Modification to Control NOM Fouling Environmental Science and Technology. 2009;43:3865 71. [71] Zhang Z, Finlay JA, Wang L, Gao Y, Callow JA, Callow ME, et al. Polysulfobetaine Grafted Surfaces as Environmentally Benign Ultralow Fouling Marine Coatings. Langmui. 2009;In Press:In Press. [72 ] Li G, Cheng G, Xue H, Chen S, Zhang F, Jiang S. Ultra low fouling zwitterionic polymers with a biomimetic adhesive group. Biomaterials. 2008;29:4592 7. [73] Li G, Xue H, Cheng G, Chen S, Zhang F, Jiang S. Ultralow Fouling Zwitterionic Polymers Grafted fr om Surfaces Covered with an Initiator via an Adhesive Mussel Mimetic Linkage. Journal of Physical Chemistry Part B. 2008;112:15269 74. [74] Zhang Z, Finlay JA, Wang L, Gao Y, Callow JA, Callow ME, et al. Polysulfobetaine Grafted Surfaces as Environmentally Benign Ultralow Fouling Marine Coatings. Langmuir. 2009;25:13516 21. [75] Ytreberg E, Karlsson J, Eklund B. Comparison of Toxicity and Release Rates of Cu and Zn From Anti Fouling Paints Leached in Natural and Artificial Brackish Seawater. Science of the Total Envirnment. 2010;408:2459 66. [76] Yebra DM, Kiil S, Dam Johansen K, Weinell CE. Mathematic Modeling of Tin Free Chemically Active Antifouling Paint Behavior. AIChE Journal. 2006;52:1926 40. [77] Sebastian N, George B, Mathew M. Metal Complexes of Po ly(acrylic Acid): Synthesis, Characterization and Thermogravimetric Studies. Polymer Degradation and Stability. 1998;60:371 5. [78] Cho CK, Kim BK, Cho K, Park CE. The aging effects of repeated oxygen plasma treatment on the surface rearrangement and adhe sion of LDPE to aluminum. Journal of Adhesion Science and Technology. 2000;14:1071 83. [79] Moon SI, Jang J. Surface Characterization Method of Oxygen Plasma Treated UHMPE Fiber Using DRIFT Spectroscopy. Journal of Applied Polymer Science. 1997;68:117 1124

PAGE 138

138 [80] Li ZF, Netravali AN, Sachse W. Ammonia plasma treatment of ultra high strength polyethylene fibres for improved adhesion to epoxy resin. Journal of Materials Science. 1992;27:4625 32. [81] Holmes S, Schwartz P. Amination of ultra high strength polye thylene using ammonia plasma. Composites Science and Technology. 1990;38:1 21. [82] Li ZF, Netravali AN. Surface Modification of UHSPE Fibers Throught Allylamine Plasma Deposition. I. Infrared and ESCA Study of Allylamine Plasma Formed Polymers. Journal of Applied Polymer Science. 1992;44:319 31. [83] Li ZF, Netravali AN. Surface Modification of UHSPE Fibers Throught Allylamine Plasma Deposition. II. Effect on Fiber and Fiber/Epoxy Interface. Journal of Applied Polymer Science. 1992;44:333 46. [84] Jensen C, Zhang C, Qiu Y. The Aging of Atmospheric Plasma Treated Ultrahigh Modulus Polyethylene Fibers. Composite Interfaces. 2003;10:277 85. [85] Austin ME, Busfield WK, Pomery PJ. Radiation induced grafting of allyl alcohol onto oriented high density polyethy lene. European Polymer Journal. 1995;31:683 7. [86] Klein PG, Gonzalez Orozco JA, Ward IM. Structure and morphology of highly oriented radiation crosslinked polyethylene fibres. Polymer. 1991;32:1732 6. Radiation Induced Cross Linking of Polyethylene. Polymer Journal. 1973;4:79 86. [88] Woods DW, Busfield WK, Ward IM. Improved Mechanical Behaviour in Ultra High Modulus Polyethylenes by Controlled Crosslinking. Polymer Communications. 1984;25:289 302. [89] Wilski H. The rad iation induced degradation of polymers. International Journal of Radiation Applications and Instrumentation Part C Radiation Physics and Chemistry. 1987;29:1 14. [90] Pruitt LA. Deformation, yielding, fracture and fatigue behavior of conventional and highl y cross linked ultra high molecular weight polyethylene. Biomaterials. 2005;26:905 15. [91] George GA, Willis HA. Fiber matrix Adhesion of Hydroperoxidized and Corona Discharge Treate Ultra high Modulus Polyethylene. High Performance Polymers. 1989;1:335 5 1. [92] Yang W, Ranby B. Photoinitiation performance of some ketones in the LDPE acrylic acid surface photografting system. European Polymer Journal. 1999;35:1557 68. [93] Ranby B. Surface modification and lamination of polymers by photografting. Internati onal Joural of Adhesion and Adhesives. 1999;19:337 43.

PAGE 139

139 [94] Ranby B, Yang WT, Tretinnikov O. Surface photografting of polymer fibers, films and sheets. Nuclear Instruments and Methods in Physics Research B. 1999;151:301 5. [95] Yang W, Ranby B. Bulk Surfac e Photografting Process and Its Applications. I: Reactions and Kinetics. Journal of Applied Polymer Science. 1996;62:533 43. [96] Yang W, Rnby B. Radical Living Graft Polymerization on the Surface of Polymeric Materials. Macromoleules. 1996;29:3308 10. [9 7] Feng Z, Ranby B. Photoinitiated surface grafing of synthetic fibers: Photoinitiated surface grafting of ultra high molecular weight polyethylene fibers. Die Angerwandte Makromolekulare Chemie. 1992;195:17 33. [98] Kim D K, Choi YM, Jung C H, Kwon H J, C hoia J H, Nho Y C, et al. Patterned grafting of acrylic acid onto polymer substrates. Polymers for Advanced Technologies. 2008;20:173 7. [99] Yu H Y, Xu Z K, Yang Q, Hu M X, Wang S Y. Improvement of the antifouling characteristics for polypropylene micropo rous membranes by the sequential photoinduced graft polymerization of acrylic acid. Journal of Membrane Science. 2006;281:658 65. [100] Wang H, Brown HR. Ultraviolet Grafting of Methacrylic Acid and Acrylic Acid on High Density Polyethylene in Different So lvents and the Wettability of Grafted High Density Polyetheylene. Journal of Polymer Science Part A: Polymer Chemistry. 2004;42:253 62. [101] Wang H, Brown HR. Atomic force microscopy study of the photografting of gylcidyl methacrylate onto HDPE and the mi crostructure of grafted chains. Polymer. 2007;48:477 87. [102] Wang H, Brown HR. Lamination of High Density Polyethylene by Bulk Photografting and the Mechanism of Adhesion. Journal of Applied Polymer Science. 2005;97:1097 106. [103] Odian G. Principles in Polymerization. 4th ed. New York, NY: Wiley Interscience. [104] Koening JL. Spectroscopy of Polymers. Washington, DC: American Chemical Society; 1992. [105] Capaccio G, Ward IM. Preparation of Ultra high Modulus Linear Polyethylenes; Effect of Molecular Weight and Molecular Weight Distribution on Drawing Behavior and Mechanical Properties. Polymer. 1974;15:233 8. [106] Berger L, Kausch HH, Plummer CJG. Structure and deformation mechanisms in UHMWPE fibres. Polymer. 2003;44:5877 84.

PAGE 140

140 [107] Deng M, Shalaby S W. Properties of self reinforced ultra high molecular weight polyethylene composites. Biomaterials. 1997;18:645 55. [108] Devaux E, Caze C, Recher G, Bielarski D. Characterization of interfacial adhesion in a ultra high molecular weight polyethylene reinfo rced polydicyclopentadiene composite. Polymer Testing. 2002;21:457 62. [109] Lin SP, Han JL, Yeh JT, Chang FC, Hsieh KH. Surface modification and physical properties of various UHMWPE fiber reinforced modified epoxy composites. Journal of Applied Polymer S cience. 2007;104:655 65. [110] Amornsakchai T, Kubota H. Photoinitiated grafting of methyl methacrylate on highly oriented polyethylene: Effect of draw ratio on grafting. Journal of Applied Polymer Science. 1998;70:465 70. [111] Masse P, Cavrot JP, Francoi s P, Lefebvre JM, Escaig B. Adhesion Improvement of High Modulus Polyethylene Fibers b Surface Plasma Treatment: Evaluation by Pull Out Testing. Polymer Composites. 1994;15:247 51. [112] Huapeng Z, Shi M, Zhang J, Wang S. Effects of sunshine UV irradiation on the tensile properties and structure of ultrahigh molecular weight polyethylene fiber. Journal of Applied Polymer Science. 2003;89:2757 63. [113] Han J, Wang H. Photografting of Acrylic Acid and Methacrylic Acid onto Polyolefines Initiated by Formaldeh yde in Aqueous Solutions. Journal ofAppliedPolymer Science. 2009;113:2062 71. [114] Han J, Wang X, Wang H. Superhydrophobic surface fabricated by bulk photografting of acrylic acid onto high density polyethylene. Journal of Colloid and Interface Science. 2 008;326:360 5. [115] Willis HA, Maas JHVD, Miller RGJ. Laboratory Methods in Vibrational Spectroscopy. Third Edition ed. New York NY: John Wiley and Sons; 1987. [116] Miller RGJ, B.C.Stace. Laboratory Methods in Infrared Spectroscopy. 2nd ed. New York, NY: Heyden & Son LTD.; 1972. [117] Mazzocca AD, McCarthy MBR, Chowaniec DM, Cote MP, Arciero RA, Drissi H. Rapid Isolation of Human Stem Cells (Connective Tissue Progenitor Cells) From the Proximal Humerus During Arthroscopic Rotator Cuff Surgery. The America n Journal of Sports Medicine. 2010;38:1438 47. [118] Gulotta LV, Kovacevic D, Montgomery S, Ehteshami JR, Packer JD, Rodeo SA. Stem Cells Genetically Modified With the Developmental Gene MT1 MMP Improve Regeneration of the Supraspinatus Tendon to Bone Inse rtion Site. The American Journal of Sports Medicine. 2010;38:1429 37.

PAGE 141

141 [119] Gulotta LV, Kovacevic D, Ehteshami JR, Dagher E, Packer JD, Rodeo SA. Application of Bone Marrow Derived Mesenchymal Stem Cells in a Rotator Cuff Repair Model. The American Journal of Sports Medicine. 2009;37:2126 33. [120] Anselme K, Bigerelle M, Noel B, Dufresne E, Judas D, Iost A, et al. Qualitative and quantitative study of human osteoblast adhesion on materials with various surface roughnesses. Journal of Biomedical Materials R esearch. 2000;49:155 66. [121] Lu Q, Zhang X, Hu X, Kaplan DL. Green Process to Prepare Silk Fibroin/Gelatin Biomaterial Scaffolds. Macromolecular Bioscience. 2010;10:289 98. [122] Ai H, Mills DK, Jonathan AS, Jones SA. Gelatin Gultaraldehyde Cross linking on Silicone Rubber to Increase Enothelial Cell Adhesion and Growth. In Vitro Cellular & Developmental Biology Animal. 2002;38:487 92. [123] Chen S, Li L, Zhao C, Zheng J. Surface hydration: Principles and applications toward low fouling/nonfouling bioma terials. Polymer.In Press, Corrected Proof. [124] Tanaka M, Sackmann E. Polymer supported membranes as models of the cell surface. Nature. 2005;437:656 63. [125] Zhao Y H, Wee K H, Bai R. Highly hydrophilic and low protein fouling polypropylene membrane pr epared by surface modification with sulfobetaine based zwitterionic polymer through a combined surface polymerization method. Journal of Membrane Science. 2010;362:326 33. [126] Gupta B, Hilborn JG, Bisson I, Frey P. Plasma Induced Graft Polymerization of Acrylic Acid onto Poly(ethylene terephthalate) Films. Journal of Applied Polymer Science. 2000;81:2993 3001. [127] Barshilia HC, Mohan DK, Selvakumar N, Rajam KS. Effect of substrate roughness on the apparent surface free energy of sputter deposited superh ydrophobic polytetrafluoroethylene thin films. Applied Physics Letters. 2009;95:033116 3. [128] Bellotti N, Deya  C, del Amo B, Romagnoli R. Antifouling Paints with Zinc Tannate. Industrial & Engineering Chemistry Research. 2010;49:3386 90. [129] Bellas J Granmo A, Beiras R. Embryotoxicity of the antifouling biocide zinc pyrithione to sea urchin (Paracentrotus lividus) and mussel (Mytilus edulis). Marine Pollution Bulletin. 2005;50:1382 5. [130] Yonehara Y, Yamashita H, Kawamura C, Itoh K. A new antifouli ng paint based on a zinc acrylate copolymer. Progress in Organic Coatings. 2001;42:150 8. [131] Sebastian N, George B, Mathew B. Metal complexes of poly(acrylic acid): synthesis, characterization and thermogravimetric studies. Polymer Degradation and Stabi lity. 1998;60:371 5.

PAGE 142

142 [132] Lavielle L, Schultz J. Surface properties of graft polyethylene in contact with water: I. Orientation phenomena. Journal of Colloid and Interface Science. 1985;106:438 45. [133] Cao X, Pettitt ME, Wode F, Arpa Sancet MP, Fu J, Ji J, et al. Interaction of Zoospores of the Green Alga Ulva with Bioinspired Micro and Nanostructured Surfaces Prepared by Polyelectrolyte Layer by Layer Self Assembly. Advanced Functional Materials. 2010;20:1984 93. [134] Brennan AB, Kumar D, Rabinovich Y I, Adler JJ, Butler GB, Moudgil BM. Steric and depletion forces between silica surfaces with free and grafted PEO. Polymeric Materials Science and Engineering. 1999;81:394 5. [135] Despa MS, Kelly KW, Collier JR. Injection molding of polymeric LIGA HARMs. Microsystem Technologies. 1999;6:60 6. [136] Feng J, Huang M, Qian X. Fabrication of Polyethylene Superhydrophobic Surfaces by Stretching Controlled Micromolding. Macromole cular Materials and Engineering. 2009;294:295 300. [137] Cassie, D. AB. Contact Angles. Discussions of the Faraday Society. 1948;3:11 6. [138] Cassie, D. AB. Wettability of Porous Surfaces. Transactions of the Faraday Society. 1944;40:546 51. [139] Hsu S H Sigmund WM. Artificial Hairy Surfaces with a Nearly Perfect Hydrophobic Response. Langmuir. 2010;26:1504 6.

PAGE 143

14 3 BIOGRAPHICAL SKETCH David Carnaby Jackson, son of Jim and Cathy Jackson was born in Midland Michigan in 1983. He has one older brother name d Steve. While in Midland, he th grade. In 1996, the family moved to Harbor Beach, Michigan. There he attended Harbor Beach Community High School and excelled at football, soccer, track, band, ja zz band, Chinese, mock trial and quiz bowl. He graduated 7 th in his class in 2001. David went to Michigan State University for his undergraduate degree in Materials Science and Engineering He spent the summer of 2003 at the University of Florida Particl e Engineering Research Center as part of a National Science Foundation Research Experience for Undergraduates. His time there working under Dr. Kevin Powers and Marie Kissinger sparked his interest in research as a career. When he returned to Michigan S tate, he sought out a research position with Dr. Meliss a Baumann. He would work in he r lab until he graduated in December 2005. David had the highest GPA in his major when he graduated. In 2006, David enrolled in graduate school at the University of Flor ida in the Department of Materials Science and Engineering with an Alumni Fellowship. He worked under the advisement of Dr. Anthony B. Brennan, specializing in polymers and David has many hobbies and interests, most notably Chinese language and American football. In 2002, he spent 2 months in Tianjin, China as part of a Michigan State study abroad program. There he learned about Chinese language and culture and visited landmarks such as The Great Wall, Tiananmen Square, and the Forbidden City. In 2003, he became the director of Corner Blitz, a student group for M ichigan

PAGE 144

144 State football fans. As director from 2003 2005, he grew the organization rapidly, from 500 members in 2003 to 2000 members in 2005. He organized road trips from East Lansing, MI to South Bend, IN, Ann Arbor, MI, Bloomington, IN and Columbus, OH to follow the Michigan State football team. Upon coming to Florida, David adopted the Gators as a second team. He was an avid fan from the 2006 season to on including National Championships in 2006 and 2008.