<%BANNER%>

Nanocomposite Strategies For Limiting Medical And Marine Biofouling

Permanent Link: http://ufdc.ufl.edu/UFE0043701/00001

Material Information

Title: Nanocomposite Strategies For Limiting Medical And Marine Biofouling
Physical Description: 1 online resource (204 p.)
Language: english
Creator: Cooper, Scott P
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: algae -- antibiotic -- bacteria -- biofilm -- biofouling -- composite -- infection -- microsphere -- silicone -- sol-gel -- topography -- xerogel
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: Microorganisms affect many aspects of human life. When microorganisms colonize a surface, the resulting microbial community is called a biofilm. Biofilms can negatively affect human health and productivity. Osteomyelitis is caused by biofilms of bacteria attached to the bone. These biofilms pose a threat to human life and lead to the loss of healthy tissue. Biofilms attached to marine vessels decrease the fuel economy of ships, resulting in a significant economic cost. There is a need to develop new materials which eradicate and prevent biofouling. Nanocomposites and mixed-phase organic/inorganic materials are presented in various embodiments as a means to limit biofouling. Antibiotic-filled microspheres are created to improve the treatment of osteomyelitis. These microspheres consist of bioactive glass and poly(n-vinylpyrrolidone) (PVP) or gelatin. Bioactive glasses have historically been shown to promote the regeneration of bone. Sol-gel chemistry is used to make the bioactive glass component, in this case a calcium silicate. The low temperature of the reaction allows organic molecules such as drugs and polymers to be blended with the glass. The catalyst used during the sol-gel reaction affects the structure and composition of the microspheres. Base catalysis leads to microspheres that exhibit behavior indicative of a nanocomposite structure. Acid catalysis produces microspheres that appear to exist as more as a mixed phase between silica and PVP. These structures directly affect the stability of the microspheres in simulated body fluid (SBF): base-catalyzed microspheres degrade within the first day in SBF, while acid-catalyzed microspheres are stable for at least one week. The morphology of acid-catalyzed microspheres is directly affected by the following compositional parameters: molecular weight of PVP, concentration of PVP, and concentration of calcium. Solid, hollow, and core/shell morphologies are produced by adjusting these parameters. These morphologies are likely caused by various rates of silicate hydrolysis, condensation, and hydrogen bonding to PVP. Viscosity of the sol plays little role in determining the diameter of the dried microspheres. An antibiotic, vancomycin, is successfully incorporated into these hybrid microspheres. Vancomycin is released for 5-7 days as measured by UV absorption. An in vitro assay against cultures of Staphylococcus aureus demonstrates that the drug remains effective for 4 days. Marine biofouling is addressed by imparting topography onto silica-reinforced poly(dimethylsiloxane) elastomeric (PDMSe) films. Swimming zoospores from the green alga Ulva are used as a model fouling organism. A bio-inspired topography deterred attachment of the zoospores by 70-80% over a 4-hr assay. Image analysis of the zoospores suggests that the topography may inhibit biofilm formation by disrupting the early-stage aggregation of spores on the surface. The attachment kinetics fit, with high correlation, equations used to describe that adsorption of bacteria to surfaces. This suggests the same physical phenomena drives the attachment of bacteria and swimming algal zoospores to solid surfaces.
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 Scott P Cooper.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Brennan, Anthony B.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-06-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043701:00001

Permanent Link: http://ufdc.ufl.edu/UFE0043701/00001

Material Information

Title: Nanocomposite Strategies For Limiting Medical And Marine Biofouling
Physical Description: 1 online resource (204 p.)
Language: english
Creator: Cooper, Scott P
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: algae -- antibiotic -- bacteria -- biofilm -- biofouling -- composite -- infection -- microsphere -- silicone -- sol-gel -- topography -- xerogel
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: Microorganisms affect many aspects of human life. When microorganisms colonize a surface, the resulting microbial community is called a biofilm. Biofilms can negatively affect human health and productivity. Osteomyelitis is caused by biofilms of bacteria attached to the bone. These biofilms pose a threat to human life and lead to the loss of healthy tissue. Biofilms attached to marine vessels decrease the fuel economy of ships, resulting in a significant economic cost. There is a need to develop new materials which eradicate and prevent biofouling. Nanocomposites and mixed-phase organic/inorganic materials are presented in various embodiments as a means to limit biofouling. Antibiotic-filled microspheres are created to improve the treatment of osteomyelitis. These microspheres consist of bioactive glass and poly(n-vinylpyrrolidone) (PVP) or gelatin. Bioactive glasses have historically been shown to promote the regeneration of bone. Sol-gel chemistry is used to make the bioactive glass component, in this case a calcium silicate. The low temperature of the reaction allows organic molecules such as drugs and polymers to be blended with the glass. The catalyst used during the sol-gel reaction affects the structure and composition of the microspheres. Base catalysis leads to microspheres that exhibit behavior indicative of a nanocomposite structure. Acid catalysis produces microspheres that appear to exist as more as a mixed phase between silica and PVP. These structures directly affect the stability of the microspheres in simulated body fluid (SBF): base-catalyzed microspheres degrade within the first day in SBF, while acid-catalyzed microspheres are stable for at least one week. The morphology of acid-catalyzed microspheres is directly affected by the following compositional parameters: molecular weight of PVP, concentration of PVP, and concentration of calcium. Solid, hollow, and core/shell morphologies are produced by adjusting these parameters. These morphologies are likely caused by various rates of silicate hydrolysis, condensation, and hydrogen bonding to PVP. Viscosity of the sol plays little role in determining the diameter of the dried microspheres. An antibiotic, vancomycin, is successfully incorporated into these hybrid microspheres. Vancomycin is released for 5-7 days as measured by UV absorption. An in vitro assay against cultures of Staphylococcus aureus demonstrates that the drug remains effective for 4 days. Marine biofouling is addressed by imparting topography onto silica-reinforced poly(dimethylsiloxane) elastomeric (PDMSe) films. Swimming zoospores from the green alga Ulva are used as a model fouling organism. A bio-inspired topography deterred attachment of the zoospores by 70-80% over a 4-hr assay. Image analysis of the zoospores suggests that the topography may inhibit biofilm formation by disrupting the early-stage aggregation of spores on the surface. The attachment kinetics fit, with high correlation, equations used to describe that adsorption of bacteria to surfaces. This suggests the same physical phenomena drives the attachment of bacteria and swimming algal zoospores to solid surfaces.
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 Scott P Cooper.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Brennan, Anthony B.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-06-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043701:00001


This item has the following downloads:


Full Text

PAGE 1

1 NANOCOMPOSITE STRATEGIES FOR LIMITING MEDICAL AND MARINE BIOFOULING By SCOTT PATRICK COOPER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011

PAGE 2

2 2011 Scott Patrick Cooper

PAGE 3

3 For my loving wife, Jessica

PAGE 4

4 ACKNOWLEDGMENTS First I must thank the chair of my committee, Dr. Anthony Brennan. H e balances a demand for scientific rigor while also providing g enuine support and care for his students His philosophy of requiring students formulate their own ideas has taught me a tremendous amount about the challenges of research I would also like to thank the other members of my c ommittee for their insight on th is project : Dr. Chris Batich, Dr. also thank Dr. Jim Farese for the time he spent on my committee and the clinical perspective he brought to our discussions I would also like to thank Dr. David Greenspan, who provided technical and career guidance I must also express my sincere gratitude to Dr. Larry Hench, who sparked my interest in bioactive glasses in my senior year at the University of Arizona and who he lped guide me to the University of Florida. I thank my fellow students and postdocs for their assistance and companionship This includes members of the Brennan Research Group past and present : Dr. Leslie Hoipkemeyer Wilson, Dr. Cliff Wilson, Dr. Jim Schu macher, Dr. Michelle Carman, Ken Chung, Jim Seliga, Dr. Chris Long, Dr. Chelsea Magin, Dr. David Jackson, Dr. Julian Sheats, Jiun Dwomoh, Angel Ejiasi, Dr. Canan Kizilkaya, and Dr. Medhat Khedr. I also would like t o thank Dr. Douglas Rodriguez for his steadfast friendship. This work would not have been possible without the help of numerous undergraduate students. These students (and the chapters they contributed to) include: Nick Sexson (chapter 4), Miranda So (cha pter 5), Kristi Lim (chapter 5), Sean Royston (chapter 6), Melissa Goude (chapter 7), and Scott Brown (chapter 7). Working with

PAGE 5

5 these students has been a two way learning experience. I look forwar d to the bright futures that await them Much of this work was performed at the Major Analytical Instrumentation Center (MAIC), Nanoscale Research Facility (NRF), Interdisciplinary Center for Biotechnology Research (ICBR) and the Particle Engineering Research Center (PERC) I would like to ackn owledge the staff at these facilities for their training and technical support. I would also like to thank the labs of Dr Pat Antonelli, Dr. Josephine Allen and Dr. Jim Farese for assisting with cell studies. I must also express my sincere gratitude to our collaborators Jim and Maureen Callow, John Finlay, and Gemma Cone at the University of Birmingham (UK) They were critical for the work presented chapter 6. They are outstanding biologists and true academicians. It has been a pleasure collaborating with their group I would like to acknowledge the funding sources for this work, which include: the National Science Foundation Graduate Research Fellowship, the Office of Naval Research contract numbers N00014 10 1 0579 and N00014 02 1 0325, the UF Alumni Fellowship, and the Margaret A. Ross Professorship. I would also like to thank my mother and father for instilling the value of education in me My brother, Ryan, must also be thanked for his support and good humor. Finally, I would like to express my s incere gratitude to my wife Jessica, for her unwavering support and encouragement through out this process.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ .......... 10 LIST OF FIGURES ................................ ................................ ................................ ........ 11 LIST OF ABBREVIATIONS ................................ ................................ ........................... 16 ABSTRACT ................................ ................................ ................................ ................... 18 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 21 Biofouling on Medical and Marine Surfaces ................................ ............................ 21 Biofilm Mitigation and Prevention ................................ ................................ ............ 22 Scope of Research ................................ ................................ ................................ 23 Specific Aim 1: Identify the synthesis parameters that affect the size, morphology, and stability of microspheres containing bioactive glass and water soluble polymers. ................................ ................................ ................ 23 Specific Aim 2: Demonstrate that vancomycin maintains its effectiveness against Staphylococcus aureus and is released from glass microspheres for at least 1 week in vitro ................................ ................................ ............. 24 Specific Aim 3: Determine if the density of zoospores of Ulva is lower on micropatterned PDMSe versus smooth PDMSe over a 4 hr period. ............. 24 2 BACKGROUND ................................ ................................ ................................ ...... 26 Introductory Remarks ................................ ................................ .............................. 26 The Process of B iofouling ................................ ................................ ....................... 26 Properties of Biofilms ................................ ................................ .............................. 28 Instan ces of Biofouling ................................ ................................ ............................ 29 Marine biofouling ................................ ................................ .............................. 30 Medical biofouling ................................ ................................ ............................. 33 Approaches to Limit Biofouling ................................ ................................ ............... 35 Lethal Approaches ................................ ................................ ........................... 35 Non specif ic Compounds ................................ ................................ ................. 36 Semi Specific Compounds ................................ ................................ ............... 37 Non Lethal Approaches ................................ ................................ .................... 39 Chemical Cues ................................ ................................ ........................... 39 Physical Cues ................................ ................................ ............................ 42 Nanocomposite Materials ................................ ................................ ....................... 44 Sol gel Chemistry ................................ ................................ ................................ .... 46 Examples of Nanocomposites in Non Fouling Applications ................................ .... 49

PAGE 7

7 Antifouling/Fouling Release Nanocomposites ................................ .................. 49 Drug Eluting Nanocomposites ................................ ................................ .......... 52 Summary ................................ ................................ ................................ ................ 53 3 NANOCOMPOSITE BIOACTIVE GLASS MICROSPHERES CREATED BY A BASE CATALYZED WATER IN OIL EMULSION ................................ ................... 54 Introductory Remarks ................................ ................................ .............................. 54 Material s and Methods ................................ ................................ ............................ 59 Emulsion Formation ................................ ................................ ......................... 60 Microsphere Isolation ................................ ................................ ....................... 60 Scanning Electron Microscopy / Energy Dispersive Spectroscopy (SEM/EDS) ................................ ................................ ................................ .... 60 Thermal Analysis ................................ ................................ .............................. 61 Microsphere immersion ................................ ................................ .................... 61 X ray Diffraction (XRD) ................................ ................................ ..................... 62 Atomic Emission Spectroscopy ................................ ................................ ........ 62 Results ................................ ................................ ................................ .................... 62 Discussion ................................ ................................ ................................ .............. 74 Summary ................................ ................................ ................................ ................ 80 4 PARAMETERS AFFECTING THE SIZE AND MORPHOLOGY OF ACID CATALYZED BIOACTIVE GLASS MICROSPHERES ................................ ............ 81 Introductory Remarks ................................ ................................ .............................. 81 Acid Cat alyzed Microspheres ................................ ................................ ........... 81 Compositional Effects on Size and Morphology ................................ ............... 82 Materials and Methods ................................ ................................ ............................ 85 Aqueous Phase Preparation ................................ ................................ ............. 85 In itial acid catalysis study ................................ ................................ ........... 85 Size and Morphology study ................................ ................................ ........ 86 Microsphere Synthesis ................................ ................................ ..................... 86 Rheology ................................ ................................ ................................ .......... 87 Laser Light Scattering ................................ ................................ ....................... 87 SEM/ED S ................................ ................................ ................................ ......... 88 Thermal Analysis ................................ ................................ .............................. 89 Microsphere Immersion and XRD ................................ ................................ .... 89 Osteoblast Viability ................................ ................................ ........................... 89 Results ................................ ................................ ................................ .................... 90 Initial Acid Catalyzed Microsphere Study ................................ ......................... 90 Microsphere Size ................................ ................................ .............................. 94 Microsphere Morphology ................................ ................................ .................. 98 Cell Viability ................................ ................................ ................................ .... 101 Discussion ................................ ................................ ................................ ............ 102 Initial Acid Catalyzed Microsphere Study ................................ ....................... 102 Microsphere Size ................................ ................................ ............................ 104

PAGE 8

8 Microsphere Morphology ................................ ................................ ................ 106 Summ ary ................................ ................................ ................................ .............. 109 5 IN VITRO DRUG RELEASE FROM BIOACTIVE GLASS MICROSPHERES ....... 111 Introductory Remarks ................................ ................................ ............................ 111 Materials and Methods ................................ ................................ .......................... 115 Micros phere Synthesis ................................ ................................ ................... 115 UV Absorption ................................ ................................ ................................ 116 Thermal Analysis ................................ ................................ ............................ 116 SEM ................................ ................................ ................................ ................ 116 Drug Release Study ................................ ................................ ....................... 117 Bacterial Inhibition Study ................................ ................................ ................ 117 Results ................................ ................................ ................................ .................. 118 Microsphere Synthesis ................................ ................................ ................... 118 Vancomycin Release Study ................................ ................................ ............ 122 Ba cterial Inhibition Study ................................ ................................ ................ 126 Discussion ................................ ................................ ................................ ............ 129 Summary ................................ ................................ ................................ .............. 131 6 KINETIC ANALYSIS OF THE ATTACHMENT OF ZOOSPORES OF THE GREEN AGLA ULVA TO SILICONE ELASTMOERS ................................ ........... 133 Introductory Remarks ................................ ................................ ............................ 133 Materials and Methods ................................ ................................ .......................... 136 Sample Preparation ................................ ................................ ........................ 136 Attachment Assay for Zoospores ................................ ................................ ... 136 Zoospore Mapping and Grouping Distribution ................................ ................ 138 Statistical Analysis ................................ ................................ .......................... 139 Results ................................ ................................ ................................ .................. 139 Kinetic Attachment Assays ................................ ................................ ............. 139 Ma pping Spore Attachment ................................ ................................ ............ 140 Spore Grouping ................................ ................................ .............................. 143 Discussion ................................ ................................ ................................ ............ 146 7 CONCLUSIONS AND FUTURE WORK ................................ ............................... 153 Introductory Remarks ................................ ................................ ............................ 153 Compos ite Bioactive Glass Microspheres ................................ ...................... 153 Antifouling Topographies on Elastomeric Films ................................ .............. 155 Amphiphilic, Nanocomposite Hydrogels as Fouling Release Coatings .......... 156 APPENDIX A CALCULATION OF CRITICAL MOLECULAR WEIGHT OF ENTANGLEMENT FOR PVP IN SOLUTION ................................ ................................ ...................... 163

PAGE 9

9 B REAGENTS USED IN SIZE AND MORPHOLOGY STUDY OF ACID CATALYZED MICROSPHERES ................................ ................................ ........... 166 C VISCOSITIES AND WEBER NUMBERS OF ACID CATALYZED EMULSIONS .. 168 D IMAGES OF EXTERNAL AND INTERNAL MO RPHOLOGY OF BIOACTIVE GLASS MICROSPHERES ................................ ................................ .................... 169 E PROCESS FOR CREATING NANOCOMPOSITE HYDROGELS ........................ 184 LIST OF REFERENCES ................................ ................................ ............................. 186 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 203

PAGE 10

10 LIST OF TABLES Table page 3 1 SEM micrographs with corresponding EDS spectra of microspheres as a function of PVP concentration. ................................ ................................ ........... 64 3 2 SEM micrographs and EDS spectra of microspheres with various c oncentrations of PEG. ................................ ................................ ....................... 65 3 3 SEM micrographs and EDS spectra of microspheres with various concentrations of gelatin. ................................ ................................ .................... 66 3 4 Composition of base catalyzed microspheres as determined by EDS and TG/DTA. ................................ ................................ ................................ ............. 67 4 1 Sol gel glass compositions tested in this study ................................ ................... 8 6 5 1 Molar ratio of components (relative to TEOS) used to make vancomycin loaded microspheres. ................................ ................................ ....................... 115 5 2 Physical properties of vancomycin loaded microspheres. ................................ 120 5 3 Drug release properties of vancomycin loaded microspheres. ......................... 124 6 1 Parameters A e 1). ...... 150 7 1 Swelling ratio of nanocomposite hydro gels after 48 hrs in DI water. ................ 158 B 1 Moles of reagents used in synthesizing acid catalyzed microspheres for size/morphology study in chapter 4. ................................ ................................ 166 B 2 Relative moles of reagents (to TEOS) used in synthesizing acid catalyzed microspheres for size/morphology study in chapter 4. ................................ ...... 167 C 1 Viscosity, diameter, and Weber number of acid catalyzed microspheres. ........ 168 E 1 Concentrations of reagents (M) used in making multi functional silica nanoparticles. ................................ ................................ ................................ ... 184

PAGE 11

11 LIST OF FIGURES Figure page 2 1 Schematic of the biofouling process. ................................ ................................ .. 27 2 2 Example of macroscopic fouling on marine surfaces ................................ ......... 32 2 3 Common fouling organisms in the mari ne and biomedical environment. ........... 34 2 4 Chemical structures of lethal compounds used to inhibit biofouling. ................... 36 2 5 org anisms and critical surface energy ................................ ................................ 40 2 6 Scanning electron micrographs of the skin of a spinne r shark and Sharklet topography ................................ ................................ ................................ ......... 43 2 7 Chemical reactions that form silica from TEOS ................................ .................. 46 2 8 Polymerization of silica under various solution conditions ................................ .. 48 3 1 Chemical structures of comm on analgesics and antibiotics. ............................... 56 3 2 Mass loss versus temperature for PVP/bioactive glass microspheres. ............... 68 3 3 Mass loss versus temperature for PEG/bioactive glass microspheres. .............. 69 3 4 Mass versus temperature for gelatin/bioactive glass microspheres. ................... 70 3 5 First derivative of sample mass versus time (DTG) and thermal differential (DTA) signals for 6 wt% PVP nanocomposite microsphere s. ............................. 70 3 6 SEM micrograph of 10wt% PVP bioactive glass microspheres after heating to 800 C. ................................ ................................ ................................ ............... 71 3 7 Dissolution of 10 wt% PVP microspheres. ................................ .......................... 71 3 8 Electron micrographs bioactive glass particles prior to and after immersion in SBF for 7 days ................................ ................................ ................................ .... 72 3 9 XRD spectra of base catalyzed bioactive glass microspheres after immersion in SBF for 1 week. ................................ ................................ .............................. 73 3 10 Elemental composition of SBF containing various compositions of bioactive glasses as measured by ICP AES ................................ ................................ ...... 74 3 11 Phase behavior in the calcium nitrate tetrahydrate (CNT) PEG TEOS system. ................................ ................................ ................................ ............... 76

PAGE 12

12 3 12 Proposed st ructure of PVP and gelatin nanocomposite microspheres produced by base catalysis. ................................ ................................ ............... 79 4 1 r oplet breakup in an emulsion. .............. 84 4 2 SEM images of acid catalyzed bioactive glass microspheres made with 6 wt% PVP in the aqueous phase ................................ ................................ ......... 91 4 3 SEM images of acid catalyzed bioactive glass microspheres made with 6 wt% gelatin in the aqueous phase. ................................ ................................ ..... 92 4 4 SEM images of acid catalyzed bioactive glass microspheres made with without any polymer in the aqueous phase ................................ ........................ 93 4 5 Thermogravimetic analysis of acid catalyzed bioactive glass microspheres. ..... 94 4 6 XRD spectra of acid catalyzed bioactive glass microspheres after immersion in SBF for 1 week. ................................ ................................ .............................. 95 4 7 6 wt% gelatin microspheres remain intact after one week in SBF ...................... 96 4 8 Vicsosity versus shear rate for the oil phase and 77S; 1,000 kg/mol; 10 wt% aqueous phase ................................ ................................ ................................ ... 97 4 9 Relationship between Weber number and viscosity ratio ................................ ... 97 4 10 Ternary phase diagrams summarizing the observed morphologies of PVP nanocomposite microspheres prepared by acid catalysis. ................................ .. 98 4 11 Microsphere made with 2 wt% of 1,000 kg/mol PVP in the aqueous phase at a nominal glass composition of 77S ................................ ................................ .. 99 4 12 Microsphere made with 10 wt% of 1,000 kg/mol PVP in the aqueo us phase at a nominal glass composition of 77S. ................................ ............................ 100 4 13 Microspheres made with 10 wt% of 10 kg/mol PVP in the aqueous phase at a nominal glass composition of 77S. ................................ ................................ ... 101 4 14 Results of MTT assay of human osteoblasts exposed to 77S; 1,000 kg/mol; 10 wt% PVP microspheres ................................ ................................ ............... 102 4 15 Critical Weber number versus viscosity ratio for droplet breakup in different types of laminar flow ................................ ................................ ......................... 105 4 16 Proposed mechanism of microsphere formation by acid catalysis. .................. 107 4 17 SEM images of microspheres with wrinkled surfaces ................................ ....... 109 5 1 SEM images of vancomycin loaded microspheres prior to soaking in SBF ...... 119

PAGE 13

13 5 2 Thermogravimetric analysis of vancomycin loaded microspheres .................... 121 5 3 Absorbance of SBF solutions containing various compounds. ......................... 122 5 4 Calibration curve for vancomycin in Tris buffered SBF. ................................ .... 123 5 5 Vancomycin release from various microsphere compositions measured by absorbance at 280 nm. ................................ ................................ ..................... 124 5 6 Cumulative vancomycin release from composite bioactive glass m icrospheres.. ................................ ................................ ................................ .. 125 5 7 SEM images of vancomycin loaded mic rospheres after 7 days in SBF. ........... 126 5 8 Zone of inhibition for S. aureus measured around paper disks containing the SBF drawn from a solution of vancomycin encapsulated microspheres. .......... 127 5 9 Vancomycin release in the bacterial inhibition study. ................................ ....... 128 6 1 Schematic of sample arrangements on 2 .54 cm x 7.62 cm glass slides. ......... 135 6 2 Kinetic response of zoospore of Ulva to smooth and Sharklet PDMSe surfaces. ................................ ................................ ................................ ........... 140 6 3 Reduction in attachment of zoospores on Sh arklet microtopography versus smooth PDMSe (6.5 cm 2 areas). ................................ ................................ ...... 141 6 4 Maps of all zoospores attached to the Sharklet topography duri ng the kinetic study. ................................ ................................ ................................ ................ 142 6 5 Settlement maps of individual spor es on the Sharklet topography. .................. 143 6 6 Settlement maps of spores in groups of 2 on the Sharklet topography. ........... 144 6 7 Settlement maps of spores in groups of 3 on the Sharklet topography. ........... 145 6 8 Settlement maps of spores in groups of more than 3 on the Sharklet topography.. ................................ ................................ ................................ ...... 146 6 9 Distribution of zoospores in groups of 1, 2, 3, or more than 3. ......................... 147 7 1 Dynamic swelling of PEGMA gels containing various concentrations of functionalized SiO 2 nanoparticle s. ................................ ................................ .... 159 7 2 Contact angle on nanocomposite gels.. ................................ ............................ 160 7 3 SEM/EDS analysis of gels containing 25 wt% NHF/MAP SiO 2 particles. ......... 161 A 1 Spherical pervaded volume of a polymer chain (V sp ). ................................ ....... 163

PAGE 14

14 A 2 The condition of entan glement, in which two chains are present in the pervaded volume of a single chain (N=2). ................................ ........................ 164 D 1 Example of ternary phase diagram that is used as a guide throughout this appendix ................................ ................................ ................................ ........... 169 D 2 10 kg/mol, 77S, 2 wt% bioactive glass mi crospheres external morphology a nd in cross section ................................ ................................ ......................... 170 D 3 10 kg/mol, 77S, 6 wt% bioactive glass microspheres external morphology and in cross section ................................ ................................ .......................... 170 D 4 10 kg/mol, 77S, 10 wt% bioactive glass microsphe res external morphology and in cross section ................................ ................................ ......................... 171 D 5 10 kg/mol, 86S, 2 wt% bioactive glass microspheres externa l morphology and in cross section. ................................ ................................ ......................... 171 D 6 10 kg/mol, 86S, 6 wt% bioactive glass microspheres external morphology and in cross section. ................................ ................................ ......................... 172 D 7 10 kg/mol, 86S, 10 wt% bioactive glass microspheres external morphology and in cross section. ................................ ................................ ......................... 172 D 8 10 kg/mol, 100S, 2 wt% bioactive glass microspheres external morphology and in cross section ................................ ................................ ......................... 173 D 9 10 kg/mol, 100S, 6 wt% bioactive glass microspheres external morphology and in cross section ................................ ................................ ......................... 173 D 10 10 kg/mol, 100S, 10 wt% bioactive glass mi crospheres external morphology and in cross section. ................................ ................................ ......................... 174 D 11 40 kg/mol, 77S, 2 wt% bioactive glass microspheres external morphology and in cross section ................................ ................................ ......................... 174 D 12 40 kg/mol, 77S, 6 wt% bioactive glass mic rospheres external morphology and in cross section ................................ ................................ .......................... 175 D 13 40 kg/mol, 77S, 10 wt% bioactive glass mi crospheres external morphology and in cross section ................................ ................................ ......................... 175 D 14 40 kg/mol, 86S, 2 wt% bioactive glass mic rospheres external morphology and in cross section ................................ ................................ ......................... 176 D 15 40 kg/mol, 86S, 6 wt% bioactive glass mi crospheres external morphology and in cross section. ................................ ................................ ......................... 176

PAGE 15

15 D 16 40 kg/mol, 86S, 10 wt% bioactive glass microspheres external morphology and in cross section ................................ ................................ ......................... 177 D 17 40 kg/mol, 100S, 2 wt% bioactive glass mi crospheres external morphology and in cross section ................................ ................................ ......................... 177 D 18 40 kg/mol, 100S, 6 wt% bioactive glass mi crospheres external morphology and in cross section ................................ ................................ ......................... 178 D 19 40 kg/mol, 100S, 10 wt% bioactive glass mic rospheres external morph ology and in cross section. ................................ ................................ ......................... 178 D 20 1,000 kg/mol, 77S, 2 wt% bioactive glass mi crospheres external morphology and in cross sectio n. ................................ ................................ ......................... 179 D 21 1,000 kg/mol, 77S, 6 wt% bioactive glass mi crospheres external morphology and in cross section ................................ ................................ ......................... 179 D 22 1,000 kg/mol, 77S, 10 wt% bioactive glass mic rospheres external morphology and in cross section ................................ ................................ ..... 180 D 23 1,000 kg/mol, 86S, 2 wt% bioactive glass mi crospheres external morphology and in cross section ................................ ................................ ......................... 180 D 24 1,000 kg/mol, 86S, 6 wt% bioactive glass mi crospheres external morphology and in cross section. ................................ ................................ ......................... 181 D 25 1,000 kg/mol, 86S, 10 wt% bioactive glass m icrospheres external morphology and in cross section. ................................ ................................ ..... 181 D 26 1,000 kg/mol, 100S, 2 wt% bioactive glass microspheres external morphology and in cross section. ................................ ................................ ..... 182 D 27 1,000 kg/mol, 100S, 6 wt% bioactive glass microspheres external morphology and in cross section. ................................ ................................ ..... 1 82 D 28 1,000 kg/mol, 100S, 10 wt% bioactive glass microspheres external morphology and in cross section ................................ ................................ ..... 183 E 1 Mold assembly for polymerizing nanocomposite hydrogels. ............................. 185

PAGE 16

16 LIST OF ABBREVIATION S 45S5 Melt derived glass: 46 SiO 2 24 Na 2 O 27 CaO 3 P 2 0 5 (mole %) 100S Sol gel derived glass : 96 SiO 2 0 CaO 4 P 2 O 5 (mole %) 86S Sol gel derived glass: 90 SiO 2 6 CaO 4 P 2 O 5 (mole %) 77S Sol gel derived glass: 80 SiO 2 16 CaO 4 P 2 O 5 (mole %) AF Antifouling CNT Calcium nitrate tetrahydrate CPC Calcium phosphate cement EDS Energy dispersive spectroscopy EPS Exopolysaccharide ERI Engineered roughness index FR Fouling release HA Hydroxyapatite ICP AES Inductively coupled plasma atomic emission spectroscopy MAP 3 met hacryloxypropyltriethoxysilane MIC Minimum inhibitory concentration MRSA Methicillin resistant Staphylococcus aureus NHF Nona hexyl fluorotriethoxysilane PDMSe Poly(dimethylsiloxane) elastomer PEG Poly(ethylene glycol) PEGMA Poly(ethylene glycol) methacrylate PVP Poly(n vinylpyrrolidone) QAC Quaternary ammonium compound R Molar ratio of water:TEOS SBF Simulated body fluid

PAGE 17

17 TBT Tributyltin TEOS Tetraethylorthosilicate TEP Triethylphosphate Tris Tris(hydroxymethyl)aminomethane W/O Water in oil XRD X ray diffraction c Critical surface energy int Interfacial energy between two liquids c Viscosity of the continuous phase of an emulsion d Viscosity of the dispersed phase of an emulsion

PAGE 18

18 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 NANOCOMPOSITE STRATEGIES FOR LIMITING MEDICAL AND MARINE BIOFOULING By Scott Patrick Cooper December 2011 Chair: Anthony Bren nan Major: Materials Science and Engineering Microorganisms affect many aspects of human life. When microorganisms colonize a surface, the resulting microbial community is called a biofilm. Biofilms can negatively affect human health and productivity. O steomyelitis is caused by biofilms of bacteria attached to the bone. These biofilms pose a threat to human life and lead to the loss of healthy tissue. Biofilms attached to marine vessels decrease the fu el economy of ship s, resulting in a significant ec onomic cost. T here is a need to develop new materials which eradicate and prevent biofouling N anocomposite s and mixed phase organic/inorganic materials are presented in various embodiments as a means to limit biofouling. Antibiotic filled m icrospheres a re created to improve the treatment of osteomyelitis These microspheres consist of bioactive glass and poly(n vinylpyrrolidone) (PVP) or gelatin Bioactive glass es have historically been shown to promote the regeneration of bone S ol gel chemistry is used to make the bioactive glass component in this case a calcium silicate The low temperature of the reaction allows organic molecules such as drugs and polymers to be blended with the glass.

PAGE 19

19 The catalyst used during the sol gel reaction affects the structure and composition of the microspheres. Base catalysis leads to microspheres that exhibit behavior indicative of a nanocomposite structure Acid catalysis produces microspheres that appear to exist as more as a mixed phase between silica and PVP These structures directly affect the s tability of the micro sphere s in simulated body fluid (SBF): b ase catalyzed microspheres degrade within the first day in SBF, while acid catalyzed microspheres are stable for at least one week. T he mo rphology of acid catalyzed microsphere s is directly affected by the following compositional parameters: molecular weight of PVP, concentration of PVP, and concentration of calcium. Solid, hollow, and core/shell morp hologies are produced by adjusting these parameters. These morphologies are likely caused by various rates of silicate hydrolysis, condensation, and hydrogen bonding to PVP V iscosity of the sol plays little role in determining the diameter of the dried microspheres. An antibiotic, vancomycin, is successfully incorporated into these hybrid microspheres. Vancomycin is released for 5 7 days as measured by UV absorption. A n in vitro assay against cultures of Staphylococcus aureus de monstrates that the drug remains effective for 4 days. Marine biofouling is addressed by imparting topography onto s ilica reinforced poly(dimethylsiloxane) elastomeric (PDMSe) f ilms Swimming zoospores from the green alga Ulva are used as a model fouling organism. A bio inspired topography deter red attachment of the zoospores by 70 80% over a 4 hr assay. Image analysis of the zoospores suggest s that the topography may inhibit biofilm formation by disrupting the early stage aggregation of spores on the su rface. The attachment kinetics fit, with

PAGE 20

20 high correlation, equations used to describe that adsorption of bacteria to surface s This suggest s the same physical phenomena drives the attachment of bacteria and swimming algal zoospores to solid surfa ce s

PAGE 21

21 C HAPTER 1 INTRODUCTION Biofouling on Medical and Marine Surfaces The earliest form of life to appear on earth was cyanobacteria also known as blue green algae These organisms thrived approximately 3.5 billion years ago and produced the oxygen that later allowed animal and human life to flourish. The only record of these early microorganisms is found in the oldest known fossils, stromatolites. These fossils have a multi layered structure which is caused by calcified mats or 1 It is likely that the cyanobacteria aggregated into biofilms as a defense mechanism from the harsh environment of the early Earth. Approximately 3.5 billion years later, biofilms are still one of the most predominant form s of life on e arth o These organisms are bound together with a layer of extracellular polysaccharide ( EPS ) to for m a dynamic community called a biofilm. The EPS protects the organisms and al lows life to exist in extreme environments. For example, the b acteria found in superheated hydrothermal vents or freezing polar ice are nearly always exist as b iofilm s 2 B iofilms also offer other advantages beyond protection from the environment Individual cells within the biofilm can perform specific tasks, allowing the microbial community to behave like a higher, complex organism. This allows biofilms to exist ubiquitous ly in diverse environments that would otherwise be uninhabitable for individual cells. C oexisting with micro organisms often comes at a cost to human health and productivity. Biomedical infections are o ne example Human tissue would be an ideal surface for biofouling to occur immune system that constantly

PAGE 22

22 eliminates foreign organisms W hen the immune system is compromise d, however, biofilms can proliferate i n human tissue resulting in an infection If left unchecked, the infection poses a serious risk for sepsis (systemic infection) and death. P atients are often hospitalized at the first sign of an infection and given extensive antibiotic therapy This can be costly and dangerous especially if numerous surgical procedures are required to clear the infection. Biofouling negative ly impact s productivity in other sectors as well The marine environment is full of organisms such as algae, barnacles, and tubeworms which form biofilms as part of their life cycle 3 Marine vessels that become coated with these organisms suffer a significant loss in efficiency 4 F ouling on ship hulls places an eco nomic burden on international commerce and can compromise naval defense capability 5 Marine fouling is also harmful to the environment because it serves a mechanism by which invasive species spread into new ecosystems 6 Biofilm Mi tigation and Prevention There are three general strategies that are used to control biofilms today The first is to use lethal compounds that kill the fouling organisms Some compounds such as antibiotics are lethal to a narrow range of organisms, while compounds such as copper are toxic to many organisms Another approach is to create a material that forms a very weak bond with the early biofilm. The idea is that only a small mechanical force is required to slough off weakly attached organisms. These (FR) coatings are gaining popularity in the marine industry but have limited use in biomedical applications. (AF) technologies prevent biofouling by avoiding the initial attachment altogether. One approach is to use the topography, or roughness, of a surface to deter the initial attachment of marine and biomedical organisms 3

PAGE 23

23 Scope of Research This research uses organic/ inorganic ma terials to eradicate or limit biofouling utilizing the various strategies li sted above The materials presented herein all contain silica as an inorganic phase. Silica is one of the most ubiquitous minerals on earth and has found numerous uses in composite materials. This work targets two applications in which biofouling poses a significant problem The first application is osteomyelitis, which is an infection of bacterial biofilms on bone tissue There is currently a need for a material that delivers drugs degrades in the body, and regenerates natural tissue Bioactive glass has historically been shown to slowly degrade and regenerate bone 7 and is therefore used in this work to improve osteomyelitis treatment L ow temperature processing is used to form the bioactive glass which allow s organic mole cules such as antibiotics to be incorporated The second application is marine biofouling. The Brennan Research Group has found that a topography inspired by shark skin is particularly effective at inhibiting the attachment of marine organisms. This topo graphy called Sharklet, is formed by patterning the surface of silica filled poly(dimethylsiloxane) elastomer (PDMSe). These nanocomposite materials are attractive for marine applications because of their toughness and low surface energy. T he ability of PDMSe topographies to inhibit fouling of algae spores is evaluated in an extended laboratory assay. Specific Aim 1: Identify the synthesis parameters that affect the size, morphology and stability of microspheres containing bioactive glass and water soluble polymers A process for forming organic/inorganic microspheres is presented. Microspheres are formed by sol gel synthesis in a water in oil (W/O) emulsion. Polymers are dissolved in the aqueous droplets to act as a binder for the calcium s ilicate glass. This

PAGE 24

24 results in a hybrid organic/inorganic microsphere Poly(n vinylpyrrolidone) ( PVP ) poly(ethylene glycol) ( PEG ), and gelatin are tested for their ability to form microspheres. The effects of pH are examined with respect to the organic /inorganic ratio of the final microspheres PVP molecular weight, PVP concentration, and calcium concentration are varied to identify the parameters that influence the diameter and morphology of the resulting microspheres. Since the microspheres are deri ved from droplets in an emulsion, the hypothesis is that the diameter of the microspheres will vary logarithmically with the v iscosity of the aqueous droplets according to a relationship described in the literature. Specific Aim 2: Demonstrate that vancomy cin maintains its effectiveness against Staphylococcus aureus and is released from glass microspheres for at least 1 week in vitro Vancomycin is loaded into the microspheres and released into si mulated body fluid (SBF). D rug release kinetics are measured by UV absorption. The efficacy of the released vancomycin is tested against cultures of S aureus The hypothesis is that microspheres will release a statistically significant concentration of vancomycin for at least 1 week as determined by 1 tailed t t est ( = 0.05). Specific Aim 3: Determine if the density of zoospores of Ulva is lower on micropatterned PDMSe versus smooth PDMSe over a 4 hr period. Zoospores of the green algae Ulva are used as a model marine fouling organism. The kinetics of fouling are measured on smooth and micropatterned PDMSe. The percent inhibition is compared between the two surface s for up to 4 hr s. A 4 hr assay is chosen because it is believed that zoospores become less selective at periods longer than 1 hr. Previous result s show that the Sharklet topography inhibits attachment by approximately 80% over smooth surfaces The hypothesis is that the fouling on PDMSe

PAGE 25

25 with the Sharklet microtopography w ill remain consistently lower (~ 80 % ) lower than smooth PDMSe for up to 4 hr s as determined by an ANOVA test ( = 0.05)

PAGE 26

26 CHAPTER 2 BACKGROUND Introductory Remarks becomes coated with biological substances. These substances include living organisms that can come from various kingdoms of life such as Bacteria Animalia Plantae and Fungi Bi omolecules produced by these organisms (including proteins, polysaccharides, lipids, and nucleic acids) are critical components because they mediate adhesion to the surface. The resulting encrustat ion creates a coating called a biofilm. Biofilms are dyna mic communities that can respon d to environmental stimuli and interaction with other organisms. Biofilms can negatively affect human health the environment, and the performance of engineered devices. Therefore, eradication and prevention of biofilms is d esirable in many applications Eliminating a biofilm that is already established typically requires careful selection of chemical compounds that will kill the problematic organisms Preventing biofilms altogether poses a si gnificant engineering challenge due to the diversity of fouling species. This typically involves intervention at the early stages of the fouling process. The Process of Biofouling Biofouling is a dynamic process that spans many length scales ( Figure 2 1 ). When a synthetic material is placed in an aqueous environment, ions from the liquid adsorb to the surface with in fractions of a second. An electric double layer is established in this manner, the properties of which are dictated by the surface charge and ion concentration in solution Larger molecules such as polypeptides interact with

PAGE 27

27 Figure 2 1 Schematic of the biofouling process. Note that the time and length scales increase from (A) to (E). Adapted from Ma gin, et al. 3 with permission (A) Ions adsorb from solution to form an electric double layer on a surface, in this case a linear polymer. (B) This electric layer mediates the interaction of larger molecules, such as peptide sequences (in this case Type II subunits of fibronectin). (C) The protein assume s a low energy conformation, which may expose certain parts of the protein to the microorganisms (in this case, Small organisms such as bacteria and s ecrete extracellular polysaccharide (EPS) which provides protection from the environment and increase adhesion to the substrate. (E) After a period of several days, a bacterial biofilm community is established which is covered by the EPS. Cells within t he biofilm will differentiate to perform different functions, such as swarmer cells that seek out new surfaces (shown in red). Larger cells (such as white blood cells) may interact with this initial biofilm. the surface through the electric double layer. These large molecules can be attracted to the surface by van der Waals forces, electrostatic and hydrophobic interactions. This occur s within a range of seconds to min ute s. If the polypeptide is part of a protein, the

PAGE 28

28 lowest energy conformation of the protein may expose certain domains to binding by microorganisms Small organisms such as bacteria will come in close contact with the surface via van der Waals forces Strong bonds are then established via specific interactions between the organism and t he proteins. The s mall organisms then begin establish a biofilm by secreting the EPS which increases adhesion to the substrate and offers protection from the environment. As a result, bacteria in biofilms are more resistant to antiseptic 8 and antibiotic 9 t reatments than free floating (p lanktonic ) cells. Bacteria undergo phenotyp ic changes as the biofilm grows. C ells deep in the biofilm become less metabolically active while cells near the surface of the biofilm can differentiate into motile cells which ca n leave the biofilm and colonize new surfaces. Large cells or other organisms may subsequently interact with this initial biofilm. In the body these may be white blood cells which attempt to eliminate the biofilm. In the marine environment, it may be un icellular organisms such as zoospores of algae or multicellular species such as tubeworms or barnacles. The result is a dynamic microbial community that can contain a multitude of species The species within this community are continually interacting and competing with one another to survive on the surface. Properties of B iofilms Biofilms possess unique properties which set them apart from planktonic cells. Bacteria in biofilms tend to be more resistant to attack from chemicals and biological organisms 10 This is attributed to the EPS, which is a complex polysaccharide network that contains lipids, surfactants and DNA. Biofilms of Bacillus subtilis a bacteria found in soil, have recently been shown to be resistant to s olutions containing 70% alcohol

PAGE 29

29 (which is commonly used to sterilize surfaces ) 8 This demonstrates how difficult i t may be to kill biofilms by chemical methods alone. The variety species that are often found within a biofilm also makes chemical treatments challenging D ental plaque for example, is known to consist of many species that co aggregate to form complex b iofilm structures 11 Since bacteria have different susceptibilities to chemical treatments, identifying a single compound that can eradicate a diverse community can prove especially challenging. Instances of Biofouling Biofouling has nega tive consequences in numerous applications. One example is tooth decay. Tooth decay is caused by biofilms of acid producing bacteria such as Streptococcus mutans and Lactobacillus which lower the local pH until the tooth dissolves 11 It has been shown that 92% of adults in the US will have a tooth decay by age 64 12 an d that the necessary treatment ( dental res torations ) costs $46 billion per year in the US 13 Biofouling has a dramatic impac t on other industries as well. Biofilm growth on heat exchangers causes a loss in efficiency of up to 10% 9 The critical nature of heat exchangers in power plants an d industrial processes makes fouling on th ese devices extremely costly. A 1985 estimate of the total cost of fouling on heat exchangers was in the range of $3 10 billion dollars per year a number that has certainly grown along with the demand for energy 14 Two applications in which biofouling has a significant impact are marine biofouling and persistent infections These two applications are the focus of this research and are explained in greater detail below.

PAGE 30

30 Marine biofouling Fouling o n ship hulls has been recognized as a significant problem since ancient times The first century Greek historian Plutarch wrote : 15 W hen weeds, ooze, and filth stick upon its sides, the stroke of the ship is more obtuse and weak; and the wat er, coming upon this clammy matter, doth not so easily part from it; and this is the reason why they usually scrape the sides of their ships Ancient shipbuilders used tar, pitch, sulfur, and sheets of lead to prevent fouling on their hulls. Copper was la ter discovered as a compound that is highly effective at preventing fouling By the mid 1700s, ship bottoms were covered with copper plates o r made entirely out of copper. The drawback of copper plates was that they corroded when placed in contact with ir on or steel. This was a significant problem because the nails used to attach the copper sheets to wooden ships were made of iron While performing galvanic protection experiments on copper surfaces in the early 1800s, Sir Humphry Davy noted slightly dissolve in seawater. He was the first to establish a correlation between copper leaching from a surface and its ability to prevent biofilms. T he use of copper pl ates became more challenging in the 19 th century a s more ships were built from iron 16 Copper continues to be used in marine paints even today, although in the form of an oxide. Despite the effectiveness of cop per based coatings, marine biofouling still occurs and comes at a significant cost. There are many factors that contribute to the economic impact of marine biofouling. A detailed analysis by Schultz, et al. 5 estimated that biofouling on the mid size US Navy destroyer class of ships ( Arleigh Burke class, DDG 51) costs $56 million

PAGE 31

31 per year and up to $1 billion over 15 years. This analysis showed that increased fuel consumption due to drag is the primary expense with dry docking, repainting, and hull cleaning also contributing to the cost T he cost of biofouling to all marine activity (naval as well as commercial) is much higher. The $56 million/year estimate accounts for only 1/5 th ( approximately 60 ships) of the US Navy fleet 17 T he number of cargo ships currently active in the world exceeds 50,000 18 Although cargo ships hav e different service lives than N avy vessels, a cons ervative extension of puts the total cost of marine fouling on the order of tens of billions of dollars per year. Beyond the economic penalties of marine biofouling, there are also troubling environmental consequences. Hull fouling act s as a transport mechanism for the spread of invasive species. A study by Hewitt, et al. 6 revealed that hull fouling was the primary mechanism for the delivery of invasive species at four ports around the world. Hull fouling is responsible for the spread of nonindigenous species over thousands of miles via intercontinental cargo ships Species can also be spread between nearby freshwater lakes by overland transport of small, recreational boats. 19 S mall boats have been attributed to the spread of the zebra mussel in North America for example 20 Evaluating the cost of damage s caused by invasive species is difficult due to the complex interactions of with the new ecosystem. Colautti, et al. 21 have estimated that six invasive aquatic species in Canada are responsible for damages of $343 million/ year Algae are one of the most common types of marine fouling organisms Algae are Figure 2 2 B). This is in contrast to Figure

PAGE 32

32 2 2 A). Hard foulers create calcified structures that can be cent imeters in length, generating a very rough surface that dramatically increases drag 4 A lgae however, are more ubiquitously observed on ship hulls than hard foulers partially because a lgae are more resistant to copper based coatings. 22 Soft foulers have been found to be the predominant type found on US Navy vesse ls 5 It is therefore necessary to have a model organism to study soft fouling. Figure 2 2 Example of macroscopic fouling on marine surfaces (A) Hard fouling of organisms such as barnacles and muscles encrust the bow of a ship. Image courtesy of Jacksonville Shipyards. (B) The soft fouling algae Ulva covers the surface of a submarine Reprinted with permission from Nature Publishing Group ( http://www.nature.com/ncomms/index.html ): Nat Commun 2:244, copyright 2011. Ulva (formerly called Enteromorpha ) is one of the most common types of algae found on ships. Algae from this genus are widespread, tolerant to various types of salinity, and are commonly found on copper based coatings 22 Ulva plants release zoospores as part of their reproductive cycle. These zoospores are 5 23 Zoospores of Ulva will probe a surface to find a suitable site for attach ment and can respond to light, chemical species,

PAGE 33

33 and texture. The zoospores spin on the surface to secrete their EPS and subsequently start the process of growing into a plant ( Figure 2 3 A) 24 25 Medical biofouling The first recorded observations of medically relevant biofilms w ere made by the inventor of the microscope, Antonie van Leeuwenhoek. In 1684, van Leeuwenhoek scraped the plaque of his teeth and observed a number of swimming organisms which 26 Yet the link between human healthcare and bacterial grown in a n attached (sessile) state was not recognized until m uch later. A major breakthrough came in the mid 1800s when the German bacteriologist Robert Koch postulated that diseases were caused by single types of bacteria that could be isolated grown in pure culture, and would cause disease when inoculated into a healthy organism. Koch made significan t contributions to establish ing the method s of bacterial culture and to correlat ing infection with disease 27 However, his postulates could not explai n disease s particularly those that involved a community of organisms or bacteria that cannot be grown in a culture plate. The concept of bacteria flourishing on a surface was not introduced until the early 1900s when Henrici 28 and ZoBell 29 noted that aquatic bacteria preferred to grow on surface, rather than free floating in sus pension. Biofilms were not recognized as a predominant mode of existence for disease causing bacteria until the 1970s 30 It is now recognized th at biofilms are responsible for afflictions such as tooth decay periodontal disease, cystic fibrosis, pneumonia, urinary tract infections, and chronic wounds 10 31 The Center for Disease Control estimates that approximately 65% of all human infections involve biofilms 32 Patients often acquire such infections while in the hospital being treated for a separate condition. There were

PAGE 34

34 1.7 million hospital acquired infections in 2002, which resulted in nearly 100,000 deaths 33 The financial costs associated with hospital acquired infections include longer hospital stays, more laboratory diagnoses, and extensive antibiotic therapy These costs amount to an annual total of ~ $4.5 billion (1995 estimate) 34 Figure 2 3 Common fouling organisms in the marine and biomedical environment. (A) Transmitted light micrograph of zoospores of Ulva attached to smooth PDMSe after 240 min s of exposure. Image taken by John F inlay, University of Birmingham. (B) Scanning electron micrograph of a biofilm of S. aureus grown in culture. Image taken from Biofilms: The Hypertextbook Montana State University with permission. 26 Although the biofilms appear similar, note that the images are taken at different magnifications The zoospores (~ 5 diameter ) Chronic infections are examples of biofilms that persist on human tissue 35 These wounds typically originate in soft tissues, but cause significant complications when th ey proliferate into hard tissue such as bone One example is periodontal disease tha t originates from the plaque in the mouth but spreads to infect the cementum of the tooth or the alveolar bone. Another example is diabetic ulcers A lack of blood supply in the ulcerated region allows bacteria to proliferate until they spread to bone. The condition of infected bone is c alled osteomyelitis Osteomyelitis is caused by Staphylococcus aureus in more than half of all cases 36 37 S. aureus is a bacterial species that is

PAGE 35

35 commonly found as a biofilm ( Figure 2 3 B). B acterial b iofilms produce acid as a byproduct of their metabolism lowering the pH of the local environment 38 H ydroxyapatite, the crystalline component of bone, dissolves below a pH of approximately 5 .8 11 Therefore, b one is dissolved in cases of chronic osteomyelitis, leaving the patient with a serious defect region that requires restoration Approaches to Limit Biofouling There are many ways to limit biofouling. The traditional method has been to eradicate the biofilm by using a lethal compound. The toxicity of the compounds can operate by a ny number of mechanisms including: lysis of cell membrane, inhibition of cell wall synthesis, or induction of apoptosis. Historically, leth al methods have been widely used in mar ine an d biomedical industries and are effective at removing biofilms that have already established C urrent trends however, have been to move away from lethal approaches due to long range environmental and health co ncerns surrounding these compounds. Many non lethal approaches focus on weakening the protein interaction with the surface, allowing the organisms and biomolecules to easily slough off. Other approaches use the physical structure of a surface to deter att achment. The long term efficacy of these non lethal approaches are still being evaluated. Lethal Approaches One of the most widely used methods to eliminate or prevent bi ofouling is the use of compounds that kill fouling organisms. Many marine and biome dical approaches still rely on lethal compounds to prevent or remove biofilms from surfaces. The specificity of these compounds varies widely, which makes them effective in different scenarios.

PAGE 36

36 Figure 2 4 Chemical structures of lethal compounds used t o inhibit biofouling. (A) Cuprous oxide, typically dispersed as a powder in acrylate paints. Image provided by North Carolina State University under Creative Commons license. (B) Tributyltin (TBT) copolymerized with methyl methacrylate form self polishi ng coatings that have been used in the marine industry 39 (C) Tetradecyldimethylammoniumchloride is one example of a quaternary ammonium compound (QAC) that lyse s cells on contact. The specificity of these compounds is often related to the alkyl chain length. (D) Vancomycin, a glycopeptide antibiotic, inhibits cell wall synthesis of gram positive bacteria. Non specific Compounds C opper is one is one of the oldest broad spectrum lethal compounds to be used in antifouling (AF) coating s Modern AF paints typically contain copper in the form of cuprous oxide (Cu 2 O), rather than the metallic form used by ancient shipbuilders. Such paints are often red in color and c an be commonly seen on commercial shipping

PAGE 37

37 vessels as the paint used below the waterline Cuprous oxide is blended with film fo rming vinyl or acrylic polymers with r osin added to slightly increase the solubility of the film in sea water. Once dissolved i n the sea water, the copper oxidizes to form the cupric (Cu 2+ ) ion, which is toxic to nearly all forms of marine life 40 41 S ome species of algae, including Ulva ha ve demo nstrated resistance to copper Therefore, o ther broad spectrum toxic compounds are typically used to increase the performance against algae. The incorporation of tributyltin (TBT) into marine paints was based on this principle. The development of coatings containing TBT was a landmark in the development of AF paints because it significantly extended the lifetime of the coatings. TBT is typically polymerized into the backbone of an acrylate polymer via carboxylate ester groups 39 The dissolution of these coating s is based on the hydrolysis of the TBT molecule and occurs at only the outermost molecular layer. These coatings are called self polishing becau se the top layer slowly erodes, keeping the surface hydrodynamically smooth. This is environmentally problematic because the carboxylic acid of tributyltin is soluble in seawater and highly toxic to marine life. TBT has been indicated in the impotence of bivalves and has the ability to bioaccumulate in organisms higher in the food chain T he International Maritime Organization banned use of TBT in paints in 2008 due to environmental concerns 40 41 However, the US Navy still uses TBT for critical underwater components such as sonar domes Semi Specific Compounds The environmental and health concerns over metals like copper and tin have led to an increased interest in biocides with a narrower range of toxicity such as q uaternary ammonium compounds (QACs). QACs are typically formulated as salts containing a

PAGE 38

38 positively charged quaternary ammonium with one or two long alkyl chains. Electrostatic and hydrophobic interactions cause the molecule to insert itself into the membrane. This increases permeability of the membrane, causing cell lysis and death 42 The sensiti vity of bacteria to various formulations of QACs differs from one organism to another For example, QACs have limited effectiveness against gram negative bacteria such as Pseudomonas aeruginosa which have an additional lipopolysaccharide layer with limit ed permeability 43 On the other hand QACs dispersed in solution are typically effective against gram positive bacteria such as Staphylococcus and are commonly used in disinfectants such as hand soa p. A relatively new strategy to limit biofouling is to covalently bind QACs into polymers or o nto surfaces to kill the cells on contact 44 This strategy has been shown effective against both marine and biomedical organisms 44 45 Other lethal compounds with semi specific toxicity are antibiotics. Antibiotics such as penicillin and vancomycin kill bacteria by binding to the peptidoglycans on the surface of the cell wall, arresting assembly of the cell wall and causing autolysis 46 Since eukaryotic cells do not have cell walls, these types of antibiotics can be safely used to kill bacteria infecting a larger organism. One serious disadvantage to using antibiotics is the ability of bacteria to develop resistance to these compounds. Approximately half of hospital isolates of Staphylococcus aureus are methicillin resistant S. aureus lactam class of antibiotics (such as penicillin). Recent reports have also indicated cases of S. aureus resistant to

PAGE 39

39 vancomycin, commonly viewed as the last line of defense against antibiotic resistant bacteria 47 Antibiotics present a dilemma in that they are currently the only defense against serious infections such as osteomy elitis y et their use leads to the creation of drug resistant strains of bacteria There are currently no other alternatives to removing established biof ilms from underlying tissue except for amputation The best practice for using antibiotics is to char acterize the infecting species and to deliver a high antibiotic dose ov er an extended period of time. Doing so gives the best chance that no bacteria will survive to develop drug resistance. Non Lethal Approaches Concerns over the environmental effects of broad spectrum toxins (such as copper and organotin) as well as antibiotic resistance have led to an increased interest in non lethal strategies Weakening t he initial adhesion of the biofilm to the surface is typically the approach that these strategies utilize. These concepts hold promise for both marine and biomedical applications Chemical Cues One of the earliest attempts to correlate surface chemistry with the degree of biofouling was made by Baier in the 1960s. This empirical relationship identified the regions of high and low fouling retention ( Figure 2 5 ) 48 based on the critical surface c ) of the substrate as determined by the method described by Zisman 49 The c = 2 0 25 mN/m. This critical surface tension corresponds to the dispersive component of the surface tension of water This energy corresponds to wet a surface after removing proteins or cells 3 48

PAGE 40

40 Figure 2 5 organisms and critical surface energy. Image taken from Baier 48 with permission. Coatings in this range of surface energy are and are typically made of silicones such as PDMSe Organisms may attach to FR coatings but only require a small s hear force to remove the m from the surface Hard fouling organisms such as barnacles are removed more easily from surface s because their bodies are larger and more rigid than soft foulers As a result, hard foulers experience a greater shear force over the length of their body than soft foulers 40 The Baier curve presented in Figure 2 5 also recognizes a low level of biofouling c > 60 mN/m. One of the most ubiquitous hydrated materials used for AF applications is poly(ethylene glycol) (PEG) PEG m olecules have repeatedly demonstrated a resistance to non specific protein adsorption, regardless of

PAGE 41

41 whether the polymer is chemically bound or physically adsorbed onto a surface 50 Jeon and de Gennes proposed that the antifouling nature of PEG is due to t he high affinity of water for the PEG molecule 51 Energy is required to displace water from the poly mer chain, resulting in an enthalpic energy barrier The polymer chain must also undergo a conformation change upon dehydration, which is entropically unfavorable The result is a polymer that binds water so strongly that biological molecules often canno t overcome the high energy barrier to adsorb onto the surface. Jeon and de Gennes predicted that the greatest resistance to protein adsorption would be caused by high molecular weight PEG. Szleif er modified this analysis to show that low molecular weight PEG at a high density could achieve the same effect 52 PEG can be implemented as an AF surface in many ways Self assembled mono layers (SAMs) with only a few ethylene glycol units have been shown to preve nt protein adsorption 53 54 as well as attachment of larger organisms such as zoospores of Ulva 55 56 Another approach is to create cross linked hydrogels from PEG. These gels swell in water and present a surface that is nearly 100% water, making protein adsorption very difficult. Such gels can be made by methacrylate 57 or thiol ene 58 chemistry and have proven effective at lowering the a ttachment of marine bacteria and algae zoospores. s in which both high and low surface energy groups are dispersed on the nanoscale. Several amphiphilic coatings are based on c c 59 chains of consisting of PEG and fluoropolymer blocks 60 These PEG fluoropolymer

PAGE 42

42 coatings have higher removal of Ulva spores and diatoms than PDMSe. The authors proposed a dynamic nature of these surfaces, such that either PEG or fluoropolymer can be prese nted at the surface depending on the environment. Si milar work by shown that hyperbranched fluoropolymers copolymerized with PEG display nanoscale phase separation 61 These cross linked materials have demonstrated re sistance to protein adsorption as well as low settlement and high re moval of Ulva zoospores and sporelings 62 Similar chemistries are currently being used in the Intersleek series of paints (International Paint), which is one of the most commercially successful FR coatings 40 Physical Cues New approaches to creat e non lethal AF coat ings focus on using the physical, rather than chemical, attributes of a surface to prevent adhesion. Nature serves as an inspiration for some of these physical attributes, since many marine organisms such as sharks do not foul The reason why living orga nisms do not foul while synthetic materials quickly become encrusted with biofilms is still an unanswered question. R ecent studies indicate that both physical and chemical mechanisms are responsible 63 The topography, or r oughness, of a surface is a crucial feature that can dictate how an organism attach es to the surface. A wide range of marine organisms including mussels, crabs, oysters, sea stars, porpoises, and sharks all have defined topographies on their surfaces, typ ically on the micron scale Some of these organisms are also reported to have surface energies in the range associated with minimum fouling (20 30 mN/m) as well as unique topograph ies 64

PAGE 43

43 Figure 2 6 Scanning el ectron micrographs of the skin of a spinner shark and S harklet topography The skin is covered with denticles which are tooth like scales with a regular series of ridges (A) Top down (B) and side on view of denticles from a spinner shark A PDMSe film with the bio inspired topography Sharklet is shown fro m top down (C) and side on (D). The Sharklet topography has a repeating diamond pattern of ridges, although they are smaller than those of the shark and are not offset at different heights. Previous work has shown that a topograph ic pattern inspired by sh ark skin called Sharklet ( Figure 2 6 ), is effective at deterring attac hment of a variety of organisms. Reductions in attachment density were 86% for zoospores of Ulva 65 97% for cyprids of the barnacle Balanus amphitrite 66 and up to 99% for cells of the marine bacteria Cobetia marina 57 T he Sharklet topography has also been effective at slowing the

PAGE 44

44 biofilm growth of biomedically relevant bacteria. Chung, et al. 67 showed that the Sharklet topography reduced the growth of biofilms of S. aureus for up to 3 weeks. The mechanism behind AF t opographies such as Sharklet deter settlement has not been clearly identified but some trends have been observed. Topographies that have spacing narrower than the size of the fouling or ganism reduce attachment, while spaces wider than the organism may in crease the attachment ( compared to smooth surfaces ) 25 68 Scardino, et al. 69 proposed a model correlat ing the attachment strength of diatoms to the number of attachment points the organism makes with the surface. The number of a ttachment points in this model serve as a proxy to the attachment area that the organism makes with the surface. Schumacher et a l. used areal terms which are typically area fraction, (f d )] to correlate surface topography with fouling behavior 68 This model correlates to set tlement reduction. Th e ERI model has since been extended to include the Reynolds number of the particular organism moving through the fluid above the surface. The resulting ERI model describes, with good correlation (r 2 = 0.77) the fouling behavior of a variety of organisms on textured surfaces 57 Nanocomposite Materials A composite is a blen d of two different materials which exist in separate domains (or phases). The reason for creating a composite is typically to achieve a desired property that could not be achieve d be either material individually Composites often take the form of one mat erial dispersed within a matrix as a second material. This can be in the form of particles, whiskers, or fibers. Fiberglass composites are one example. These composites have high strength and toughness that allows them to be used in

PAGE 45

45 applications which w ould otherwise not be suitable for either the glass or polymer materials alone. The term e case when the dispersed phase is on the order of hundreds of nanometers or less in size. The high interfacial area between components in a nanocomposite can offer desirable properties but may pose challenges such as effective dispersion on the discontinuous phase Bone is one example of a nanocomposite commonly found in nature that has outstanding properties. Bone is composed of approximately 65 % hydroxyapatite [Ca 10 (PO 4 ) 6 (OH) 2 ], 30 % proteins, and 5 % water by weight 70 The majority of the protein in bone is Type I collagen, a protein that forms a rod shaped triple helix with a molecular weight of approximately 300,000 g/mol 71 These triple helices self assemble into fibrils with narrow gaps that allow for the intercalation of hydroxyapatite plates that are approximately 25 nm wide and 2 3 nm thick. The dispersion of hydroxyapatite within the collagen fibril and the layering of these nanocomposite fibrils within the bone imparts considerable stre ngth, toughness, and elongation to failure 72 Nanocomposites have unique structural, optical, a nd biological properties and are thus desirable to create synthetically. There are various methods to create organic/inorganic nanocomposites 73 Hybrid materials can be created by polymerizing organic molecules in the presence o f inorganic particles I norganic particles can also be formed in situ within a polymer 74 The result in either case is often a network with elastomeric properties. The organic and inorganic phases c an be polymerized simultaneously to cre ate an interpenetrating network with unique mechanical and optical properties 75 Sol gel chemistry is the method typical ly used to create the inorganic phase and therefore necessitates further discussion.

PAGE 46

46 Sol gel Chemistry Sol gel chemistry 76 is a solution based method of making oxide ceramics. The most common precursors for thes e ceramics are metal alkoxides with the general formula M n+ (OR) n in which R is an alkyl group. Alkoxysilanes, Si(OR) 4 are used to create silica, which is one of the most ubiquitous ceramic phases used in nanocomposites. Figure 2 7. Chemical reactions that form silica from TEOS. (A) Hydrolysis of TEOS occurs by nucleophilic attack (S N mechanism) of water at the silicon atom. (B) Condensation can occur between two silanol groups, producing water or (C) between a silanol and alkoxy group, producing ethanol. Reaction a rrows are s hown as bidirectional, however all reactions (A C) heavily favor the product side. T etraethylorthosilicate (TEOS) which has R = CH 2 CH 3 is one of the most common alkoxysilanes used in sol gel chemistry. TEOS will hydrolyze in the presence of water, creati ng a silanol group ( Figure 2 7 A). Siloxane (Si O Si) bonds are formed by

PAGE 47

47 condensation of a silanol group with either another silanol ( Figure 2 7 B) or an alkoxy group ( Figure 2 7 C). A sol is created as the hydrolysis and condensation reactions produce a co lloidal suspension of silica. The molecular weight of silica increases as the colloidal particles begin to condense together T he silica polymer has infinite molecular weight at the point of gelation The resulting gel is porous, with water or alcohol o ccupying the pores. area. There are numerous factors that influence the structure of gels prepared by sol gel chemistry. One of the most critical parameters that a ffect gel morphology is pH. Under acidic conditions, the hydrolysis reaction occurs very quickly, while condensation occurs more slowly. The result is that gels polymerized under acidic conditions tend to form branched chains, much like organic polymers These chains cross link to form a network structure. U nder basic conditions, hydrolysis is the rate limiting step, while condensation occurs readily. Therefore s ols created at high pH tend to be more particulate in nature and gelation occurs as these particles aggregate At high pH, sols tend to be very stable and the colloidal particles can remain in solution for months. This is because the surface of silica has a significant negative charge at high pH and the electrostatic repulsion between particle s allows the solution to be stable. S a lts will cause the sol to gel as well as lowering the pH of the solution. At are m etastable This pH corresponds to the isoelectric point of silica, where particle growth is very slow due to low so lubility of silica

PAGE 48

48 Another crucial parameter is the molar ratio water to alkoxysilane ( R ratio). Technically, R = 2 is sufficient to fully hydrolyze an alkoxysilane. However, hydrolysis is rarely complete, leaving a mixture of silanol and alkoxy groups on the silica surface Generally, higher R ratios result in a greater degree of hydrolysis. Low R ratios (R < 4 ) create polymeric sols that can be spun into fibers while higher R ratios ( R >> 4) encourage more of a particulate morphology (similar to the track B shown in Figure 2 8 ). Figure 2 8 Polymerization of silica under various solution conditions. (A) Acid catalysis produces polymer like network structures. (B) Base catalysis encourages particle formation. Image from Iler 77 with perm ission from John Wiley and Sons Silane coupling agents can be used to bring or ganic functionality into the inorganic network. These compounds have the general formula R Si (OR ) 3 in which R may be a

PAGE 49

49 polymer, oligomer, or functional group capable of bonding with organic species. Covalently binding the organic and inorganic phases typically enhances the mechanical properties of the nanocomposite Examples of Nanocomposites in Non Fouling Applications Sol gel chemistry can be utilized to create hybrid materials with unique properties such as low surface energy and controlled drug rel ease Often, these materials are phase separated in a nanocomposite structure. Below are examples of mixed phase hybrid materials and nanocomposites aimed at preventing fouling in the marine and medical environments. Antifouling/Fouling Release Nanocompo sites PDMSe has historically been one of the most widely used nanocomposite materials for AF/FR applications. The surface energy of PDMSe is typically ~22 mN/m, C ross linked PDMS is soft and gum like. S ili ca fillers are commonly added to increase the modulus and strength of the material. The strength can increase by a factor of 40 with a dramatic increase for particles less than 100 nm in diameter 78 Silica f illed PDMSe is an attractive material for marine coating applications because it has a low surface energy and high toughness. However, the retention of fouling organisms changes with modulus. O ne study has shown that psuedobarnacle adhesion increases with the concentration of silica filler 79 Brady used a fracture analysis model to predict that higher modulus material s would retain more fouling organisms. 80 This was later experimentally demonstrated with silicones by Chaudhury, et al. 81 Schumacher, et al. 82 attempted to overcome fouling on silica filled PDMSe by patterning the surface with a heterogeneous microtopography. T he idea behind these

PAGE 50

50 surfaces was that deflection of adjacent, dissimilar features imparts a force gradient of ~100s nN on the organism. These gradients represent the mismatch in the force req uired to uniformly deform two adjacent features with different lengths. Indeed, the force gradients appear ed to loosely correlate to the lowest settlement of zoospores of Ulva More recent work has been focused on blending other chemistries with silica. Specifically, hybrid xerogels have been examined as AF/FR coatings Xerogels containing alkyl, fluoroalkyl, and aminoalkyl groups performed better than glass at releasing of zoospores of Ulva (up to 65%) and preventing settlement barnacle cyprids (as low as 10%). C oatings with c ) of approximately 20 mN/m had the highest release of Ulva sporelings and lowest settlement of barnacle cyprids 83 A later study s upported this result by demonstrating a distinct Baier curve behavior for the adhesion of protein coated AFM tips to xerogel surfaces. This work showed that c although the release of diatoms showed the opposite behavior 84 Most of these hybrid materials appear to be a single, mixed phase. However, a later study identified xerogels with phase separation on the nanoscale. These materials contained a blend of n octane and n octadecane groups with a critical surface tension of c = 21 23 mN/m. These phase segregated materials had higher release of zoospores of Ulva (up to 90 %) and lower settlement of barnacle cyprids (<10%) than previous hybrid xerogels 85 Because nanocomposites allow the incorporation of many different chemical species it is possible to vary the surface energy over a wi de range. This has allowed detailed studies

PAGE 51

51 or FR behavior. Liu, et al 86 plated 316 L stainless steel surfaces with various composi tions of Ni P TiO 2 polytetrafluoroethylene (PTFE) from a solution containing colloidal TiO 2 an emulsion of PTFE, and salts of Ni and P. The coatings ranged in surface energy from 20 45 mN/m. T he adhesion of Pseudomonas fluorescens Cobetia marina and V ibrio alginolyticus bacteria was direc tly proportional (r 2 = 0.82 0.92) to the ratio 2 LW 2 2 LW 2 are components of surface energy represent ing nonpolar and electron do nor interactions, respectively This analysis supports the results by Bennett, et al. 87 which established a strong (r 2 = 0.94) relationship between the attachment density of zoospores of Ulva and the percent of the surface energy contributed by polar inte ractions of xerogel surfaces While these two studies us ed different theories to calculate surface energies they both arrived at the same conclusion: lower biological attachment is achieved by increasing the polar contribution to the surface energy. Sol gel chemistry has the advantage of being a low temperature process This allows sensitive organic molecules such as enzymes to be encapsulated in the material Proteolytic enzymes, which digest the proteins that organisms use to adhere to surfaces, a re promising non toxic compounds that are gaining interests as components of AF coatings 88 Luckarift, et al. 89 used silica and titania xerogels to encapsulate lysozyme. The lysozyme containing gels maintained 70 100% of their ability to lyse Micrococcus lysodeikticus and had better resistance to he at treatment that the free molecule. Kim, et al. 90 chymotrypsin into silicate gels and found that it could digest human serum albumin. An extensive review on the subject states that polymers and oligom ers can be incorporated in to xerogels to protect the enzyme from

PAGE 52

52 being denatured. These studies show that while some enzyme activity may be lost in the process of encapsulating enzymes in a nanocomposite the activity is generally more resistant to aging and environmental degradation 91 Drug Eluting Nanocomposites Sol gel chemistry is also conducive to encapsulating drugs. Aughenbaugh, et al. 92 demonstrated that vancomycin can be incorporated into silica xerogels. Vancomycin release rate was directly proportional to the R ratio (moles H 2 O:moles alkoxysilane ) in the range of R = 4 10. The extent of vancomycin released from the material was also proportional to R ratio, with R = 10 gels releasing 90% of the initial amount of vancomycin. Radin, et al. 93 continued this work and showed that higher vancomycin loading produced a longer extent of first order release kinetics. Vancomycin released from these xerogels was effective against S. aureus for up to 21 days. These silica xerogels were made into microspheres (100 in a later study by Radin, et al. 94 by emulsifying the gel containin g vancomycin in vegetable oil But m icrospheres prepared by this method had slower release and less cumulative release of vancomycin than ground xerogel. It should be noted that these studies investigated vancomycin in silica xerogels without the addition of any polymer or salts Blending polymers with xerogels has the ability to change the release rate of encapsulated drugs. Costache, et al. 95 demonstrated that zero order release rates for anesthetic drugs could be obtained by b lending silica xerogels with poly(ether carbonate) PEG block copolymers. Similar results were obtained by Ahola, et al. 96 who demonstrated that adding PEG to a silica xerogel slowed the rele ase of an anti cancer drug. The authors cite dissolution of the PEG/silica matrix as the reason for the observed first order release kinetics. However, this group also found that the same

PAGE 53

53 caprolactone co DL lact ide) when it was encapsulated into a silica xerogel 97 This indicates that drug release from nanocomposites is dependent on complex interactions between the components (drug, polymer, inorganic phase) as well as the method by which the material is made. Summary Biofouling is a complex process that involves a wide variety of biomolecules and living organisms. Once a biofilm establishe s on a surface, it is often difficult to remove due to the defense mechanism inherent in the biofilm phenotype. Therefore, there is significant interest in creating AF coatings that prevent the initial adhesion event from taking place. This has been effe ctive in some instances; however due to the long working life of these coatings and the diversity of fouling species, attachment does still occur. FR s urfaces that require a low amount of energy to release any attached organisms are also of great interest for this reason Recent efforts have focused on developing materials that have limited toxicity. However, in some cases such as a life threatening infection, the use of toxic antibiotics is necessary. Nanocomposites are a class of materials that combin e the benefits of polymers and ceramics. There are a wide range of chemistries that can be created by using nanocomposites due to the flexible nature of sol gel processing. Nanocomposites created by sol gel processing have shown promising results as both AF and FR materials in the marine environment. Drugs have also been successfully encapsulated in nanocomposites to treat biofilms present in wounds or infected tissue.

PAGE 54

54 CHAPTER 3 NANOCOMPOSITE BIOACTIVE GLASS MICROSPHERES C REATED BY A BASE CATALYZED WA TER IN OIL EMULSION Introductory Remarks Osteomyelitis is the infection of bone caused by biofilms of bacteria. These biofilms lower the local pH and destroy the underlying tissue. The current treatment is to administer antibiotics locally over several w eeks to achieve a high, sustained dosage 37 98 Poly( methyl methacrylate) (PMMA) beads have been used in as the drug delivery material for osteomyelitis treatment for decades 99 100 However, there are several drawbacks to using PMMA First, the material tends to exhibit rapid, burst rel ease kinetics 101 Second, PMMA is not bioresor bable and a second surgery must be performed to prevent fibrous encapsulation Finally, PMMA does not stimulate regeneration of bone tissue that has been lost due to the infection. B ioceramics have been proposed as drug carriers because they can remain at the site and would encourage bone g rowth 102 Bioceramics such as the bioactive glasses in the SiO 2 Na 2 O CaO P 2 O 5 system are promising candidates for this application because they are well tolerated in the body 7 One reason why these glasses are tolerated so well is that they slowly dissolve into ionic byproducts such as Ca 2+ PO 4 3 and Si(OH) 4 These small molecules are easily metabolized These ions also allow the glass to physically bond the to the bone ti ssue. This occurs through a series of surface mediated reactions in which Ca 2+ and PO 4 3 precipitate as an amorphous solid on the glass 103 The amorphous calcium phosphate subsequently crystallizes into an apatite by incorporating OH CO 3 2 or F ions fr om solution. Bioactive glasses and their dissolution products have been shown to increase the proliferation 104 and collagen production 105 of the bone precursor cells This

PAGE 55

55 produces a mineralized interface that forms a strong bond between the glass and the surrounding tissue. M elt derived bioactive glasses have been proposed as drug delivery materials 106 However, drugs cannot be encapsulated in to the glass matrix due to the high temperature processing Instead, drugs must be adsorbed onto the surface of the material. Silanol groups on the surface of silicate materials offer sites for hydrogen bonding to occur with the drug. However, this leads to the adsorption of a monolayer of drug at the surface The low surface area and of melt derived glasses does not allow for an appreciable amount of drugs to be adsorbed. However, m esoporous silicates have very high surface areas and have attracte d attention as drug release materials. Mesoporous silicates are created by sol gel processing. Micelles of a surfactant are used as templates to create the pores in the final silicate. An alkoxysilane is polymerized around the micelles, which typically h ave spherical or tube like morphology. The micelles are removed by either heating to ~500 C or by acid treatment. The result is a xerogel with pores ~5 15 nm in diameter with extremely high surface area (~1,000 m 2 /g). Drugs are typically physically ads orbed from solution into the pores of the gel 107 The drug release rate d epends on numerous factors including : the size of the drug relative to the pores interaction between the drug and silica surface and the tortuosity of the pores The first drug release study with mesoporous silica showed that ibuprofen (~1 nm molecule ) was released over a period of 3 days when the average pore size was ~4 nm 108 M esoporous silicate s with larger pores ( ~12 nm mean size) exhibited a complete release of ibuprofen within 1 day 109 Arcos et al. 110 demonstrated that

PAGE 56

56 calcium can be incorporated mesoporous silicates and that these could be spray dried into 0.2 triclosan, an antibiotic over a period of ~12 hr s. Most studies show that untreated mesoporous silica releases drugs throughout a period of hours to a few days. Figure 3 1 Chemical structures of common analgesics and antibiotics. Approximate size of the drug molecules were estimated by measuring the molecule along tware Jmol. (A) Ibuprofen, ~ 0.8 nm. (B) Triclosan, ~0.7 nm. (C) Gentamycin, ~1. 2 nm. (D) Vancomycin, ~ 1.7 nm. Another approach to encapsulate a drug within a xerogel is to polymerize the silicate in a solution containing the drug Unsintered xerogels have porosity on the range of 1 20 nm, depending on the catalyst and processing co nditions 76 Duchey group has demonstrated the effectiveness of this technique in numerous publications One study found that vancomycin could be trapped within an acid catalyzed, 100% silica xerogel that p roduce transparent monoliths when dried 93 The release rate of vancomycin directly correlated to the molar ratio H 2 O:alkoxysilane ratio and the concentration of vancomycin in the gel 92

PAGE 57

57 A later study showed that microspheres (100 encapsulat ing vancomycin could be produced by emulsifying these sols in vegetable oil However, emulsifying the sol diminished the release rate and cumulative release of vancomycin This may have been due to vancomycin dissolving into the oil phase, because the average pore size (~2.5 nm) was the same xerogels regardless of whether it was emulsified. The release kinetics fit a model for diffusion limited release from spherical bodie s This model indicates that diffusion, rather than matrix dissolution, was responsible for driving the release of vancomycin (~1.7 nm molecule) from the porous xerogel 94 Polymers have been blended with ceramics t o change the nature of the drug release. Xia, et al. 111 d emonstrated that two different drugs could be released from mesoporous silica depending on the pH of the surrounding media. This was accomplished by coating the mesoporous silica with pH responsive benzyl L glutamate) block poly(ethylen e glycol). Ahola, et al. 96 showed that release of an anti cancer drug from xerogels was slower when poly(ethylene glycol) (PEG) was incorporated into the acid catalyzed, silica sol. Most stu dies to date on drug releasing ceramics have been performed on xerogels that did not contain polymers. Although some studies (mentioned above) have shown that polymers have the ability to slow the release kinetics. Polymer bioactive glass blends have bee n created in various morphologies. Hatcher, et al. 112 created poly(n vinyl pyrrolidone) (PVP) bioactive glass nanocomposites that could be sprayed or spun into fibers. Costa, et al. 113 produced macroporous foams of a bioactive glass poly(vinyl alcohol) hybrid material Incorporating CaO and P 2 O 5 generally increases the dissolution rate of xerogel materials 114 However, in the presence of polymers, the

PAGE 58

58 interaction between ions and the polymerizing silicate becomes more complex. Arcos, et al. 110 reported that Ca 2+ decreases th e ordering of pores within mesoporous silica formed in the presence of PEO PPO PEO block copolymer. The goal of this study is to characterize the morphology of hybrid microspheres made by a water in oil (W/O) emulsion process This work is an extension of a publication by Park, et al. 115 116 who created PVP /silica and PEG/silica microspheres by W/O emulsi on. The droplets of the emulsion serve as precursors to the microspheres in this method. Tetraethylorthosilicate (TEOS) is added to the emulsion and hydrolyzes at the oil/water interface to become water soluble. Polymers dissolved in the aqueous droplet s to slow the diffusion of hydrolyzed silica This study showed that t he microspheres have a hollow structure after heating to 600 C 115 This work differs from that by Park, et al. in that microspheres will be made with the 77S composition 114 of bioactive glass ( 80 SiO 2 16 CaO 4 P 2 O 5 mole % ). Three polymers are used in this study for the organic phase of the microsphere : PVP, PEG, and gelatin. PVP is commonly used as a binder in pharmaceutical applications and is generally well tolerated by the body 117 PEG is a water soluble polymer that is widely used as an excipient in drug formulations. Gelatin is a naturally occurring polyamide found in skin and bone. It is commonly used in drug delivery applications and is generally inexpensive 118 Furthermore, gelatin has been shown to significantly increase the early stage bonding strength of apatite wollastanite bioactive glass to bone 119 For these reasons, gelatin is a suitable polymer filler for drug d elivery using bioactive glass.

PAGE 59

59 Materials and Methods A process was adapted from the method shown by Park, et al 115 to create microspheres by a W/ O emulsion The oil phase consisted of 1 octanol with 3 % (wt/wt) Span 80 (an oi l soluble emulsifier) and 1.4 % (wt/wt) hydroxypropylcellulose (a rheological modifier). This was prepared by heating the 1 octanol to 80 5 C before slowly adding the hydrox ypropylcellulose over a 5 min period while stirring The solution was stirred for an additional 4 hr s at this temperature and subsequently cooled to 40 5 C before adding the Span 80. To prepare the aqueous phase, a 2.7 M solution of Ca(NO 3 ) 2 was prepar ed with DI strong ammonium hydroxide solution which titration revealed [NH 3 ] = 10.6 M. The ammonium hydroxide was added to the 2.7 M Ca(NO 3 ) 2 solution until pH = 10.5 as measured by a Thermo Orio n 2 Star Benchtop pH meter. Polymers were added to the aqueous solutions at concentrations of 2 % 6 % and 10 % (wt/wt) of the pH adjusted 2.7 M Ca(NO 3 ) 2 solution. The polymers tested in this st udy were: PVP (molar mass ~ 1,000 kg/mol, Polyscie nces #06067), PEG (molar mass ~ 20 kg/mol, Alfa Aesar #A17925), a nd gelatin type B (molar mass ~ 50 kg/mol, Sigma #G 939 ) The appropriate amount s of polymer were added to each sample and stirred until a transparent, homogenous mixture was achieved. Because gelatin is more soluble at higher temperatures, the aqueous solution was heated to 70 3C for 40 min s to allow the gelatin to dissolve The gelatin solutions then were cooled to ambient; i.e., below 30 C, prior to adding it to the oil phase.

PAGE 60

60 Emulsion Formation The oil phase was added by weight (46 2g) to a 250 ml glass beaker. A Caframo mechanical mixer was used to mix the emulsion at 500 rpm with a propeller type blade 6.1 ml of the aqueous solution was ad ded dropwise with a pipette to the continuously stirring oil phase. This corresponds to a n emulsion with a volume ratio of approximately 9:1 (oil:water). The emulsion was stirred for 30 min s. 18.8 ml of TEOS was added directly to the emulsion immediatel y followed by 1.4 ml of triethylphosphate (TEP) These reagent concentrations correspond to a 77S composition of bioactive glass ( 80 SiO 2 16 CaO 4 P 2 O 5 mole %) with a H 2 O:TEOS molar ratio of 4. The molar ratio of sol gel reactants H 2 O:TEOS:Ca(NO 3 ) 2 :TEP:NH 3 was 4:1:0.18:0.01:0.02. The emulsion was stirred for 24 1 hr s at 21 3 C. Microsphere Isolation The microspheres were washed three times by centrifugation at 2,000 rpm and were rinsed with ~20 ml of 200 proof ethanol after each centrifugation A vortex mixer was used to re disperse the microspheres after each rinsing. The microspheres were dried by placing them in a vacuum oven at 45 10 C and placing them under vacuum. The microspheres were dried until ethanol was no longer visible, typi cally 2 4 hr s Scanning Electron Microscopy / Energy Dispersive Spectroscopy (SEM/EDS) All samples were mounted on aluminum stubs with double sided tape f or SEM/EDS analysis. A layer of gold palladium was sputter coated on a Denton Desk II at 38 mA for 60 seconds. SEM/EDS analysis was performed on a JEOL 6400. All i mages were taken in secondary electron mode with 15 mm working distance and 15 kV accelerating voltage. Elemental content was determined by EDS at the following x ray

PAGE 61

61 1 = 0.2675 1 1 = 1 = 2.0075 keV. EDS spectra were taken on spots are indicated on the micrographs. Thermal Analysis Thermogravimetric and differential thermal analysis (TG/DTA) was performed on a Seiko TG/DTA 320. Each sample ha d a mass of 9 11 mg. The samples were purged with air at a flow rate of 100 ml /min. Heating was performed between 30 800 C at a rate of 2 C/min. For DTA analysis, the reference material was 14 mg of Al 2 O 3 powder DTA curves have been baseline corrected against an empty sample pan. Microsphere immersion Microspheres were immersed in simulated body fluid (SBF) to determine if hydroxyapatite w ould precipitate at the surface. The SBF was prepared as descri bed by Kokubo, et al. 120 and contained the following ions (in mM) : 142.7 Na + 5.0 K + 1.5 Mg 2+ 2.6 Ca 2 + 187.8 Cl 4.2 HCO 3 1.0 HPO 4 2 0.5 SO 4 2 The solution and was buffered to pH = 7.40 using tris(hydroxymethyl)aminomethane 120 Samples were immersed in a s haker bath at 37 C for up to 7 days. The concentration of microspheres to SBF was fixed at 0.001 g/ ml Melt m elt derived 45S5 bioactive glass 7 granules were includ ed in these experiments ( 46 SiO 2 24 Na 2 O 27 CaO 3 P 2 O 5 mole % ). The granules were prepared by Mo Sci Corp ( donated by Dr. Larry Hench) and had a particle size of 200 cellulose paper. The particles were rinsed three times with ~10 ml DI water. The samples were then dried at room temperature in a desiccator. Because SBF containing microspheres filtered very slowly, these samples were collected by centrifugation at 150 0 rpm. The

PAGE 62

62 pellet was rinsed with 50 mL of DI water. This process was repeated three times bef ore drying as described above. The SBF was analyzed by ICP AES to determine composition changes induced by dissolution of the particles. SBF was collected by f iltration cellulose filter paper a nd collected in centrifuge tubes. These solutions analyzed wi thout any further modification. X ray Diffraction ( XRD ) X ray diffraction (XRD) was used to determine whether crystalline hydroxyapatite forms at the surface of microspheres after immersion in SBF. The analysis was avg = 0.154 nm. Scans were performed with a step size of 0.02 and time per step of 0.5 seconds. Powdered samples were mounted on to a piece of glass slides with double sided tape. Atomic Emission Spectroscopy Inductively coupled plasma atomic emission spectroscopy (ICP AES) was used to monitor the presence of Ca, Si, and P in SBF as a function of time ICP AES analysis was performed on a Perkin Elmer Optima 3200. A 100 ppm standard of each element was serially diluted in DI water (18.2 M cm) to give standards of 10 and 1 ppm A calibration curve was established using for each elemen t using a non linear fit to an r 2 value of 0.9999 or greater. Each sample was measured 5 times and correlated to the calibration curve using a peak area algorithm. The wavelengths used for reporting concentrations we Ca Si P = 214.914 nm. Results Aqueous droplets which contained PVP ( Table 3 1 ) at concentrations of 6 wt% and 10 wt% The microspheres made with 6 wt% PVP ha d diameter s of approximately

PAGE 63

63 5 made with 10 wt% PVP were slightly larger with diameter s of approximately 10 The 2 wt% PVP composition produced few microsphere s. The material instead appeared largely as a film.

PAGE 64

64 Table 3 1 SEM micrographs with corresponding EDS spectra of microspheres as a function of PVP concentration. Low m agnification High m agnification EDS s pectra 2 wt% PVP 6 wt% PVP 10 wt% PVP 10 10 10

PAGE 65

65 Table 3 2 SEM micrographs and EDS spectra of microspheres with various concentrations of PEG. Low m agnification High m agnification EDS s pectra 2 wt% PEG 6 wt% PEG 10 wt% PEG 1 1 1 5 5 5

PAGE 66

66 Table 3 3 SEM micrographs and EDS spectra of microspheres with various concentrations of gelatin. Low m agnification High m agnification EDS s pectra 2 wt% Gelatin 6 wt% Gelatin 10 wt% Gelatin 5 5 5 2 2 2

PAGE 67

67 Microspheres made with PEG were significantly smaller, i.e., 0.2 1 at all concentrations tested. Microspheres made with gelatin appeared to be 1 diameter by visual inspection of the micrographs ( Table 3 3 ). T he concentration of gelatin appears to affect the morphology of the microspheres. Some microspheres prepared with 2 6 wt% gelatin appeared hollow, despite never havi ng been fired. Many of these microspheres also appeared collapsed. However, microspheres prepared with 10 wt% gelatin concentration appear more Table 3 4 Composition of base catalyzed microspheres as determined by EDS and TG/DTA Ca:Si ratios with error values indicate the average and standa rd deviation of 2 measurements. Numbers without error are a single measurement Polymer Polymer c oncentration ( %, wt/wt ) Ca:Si ratio C:Si ratio Mass loss between 150 800 C (%) PVP 2 0.8 0.9 1.8 2.3 68 6 0.8 0.4 3.2 0.7 84 10 1.1 0.1 2.7 0.4 85 PEG 2 0.1 0.0 0.1 0.1 28 6 0.1 0.0 0.3 0.3 26 10 0.1 0.0 0.3 0.2 29 Gelatin 2 0.3 0.0 0.4 0.0 65 6 0.9 0.6 2.0 1.0 75 10 0.9 0.4 81 Calcium concentration in the final microspheres was dependent upon t he type of polymer in the aqueous phase Silicon was used as an internal reference for comparing EDS signal s between samples. The ratio of Ca:Si was approximately 10 times lower for samples cont aining PEG than samples containing PVP or gelatin ( Table 3 4 ). The amount of carbon relative to silicon was also lower in samples cont aining PEG than other polymers.

PAGE 68

68 Figure 3 2 Mass loss versus temperature for PVP/bioactive glass microspheres TG/DTA analysis was used to determine the approximate ratio of organic and inorganic components in the microspheres. All samples lost approximately 10% of their initial mass upon heating to 150 C. This mass loss is largely associated with residual water or eth anol being driven out of the microspheres However, completion of the sol gel reaction by condensation of residual silanol or alkoxide groups will also produce a mass loss in this region Mass loss between 150 800 C is be attributed to the removal of th e organic polymer. Therefore, the percent mass loss in this range ( Table 3 4 ) is assumed to be the weight percent polymer in the microspheres Compositions prepared with 2 wt% PVP ( Figure 3 2 ) lost approximately 68 % of its original mass between 150 800 C. Microspheres prepared with 6 and 10 wt% PVP lost approximately 85% of their mass. This indicates that microspheres prepared with 6 and 10 wt% PVP had essentially the same ratio of organic to inorganic components. All nanocomposites made with PEG los t between 26 29 % of their original weight. There

PAGE 69

69 was no correlation between PEG concentration and mass loss ( Figure 3 3 ). Microspheres prepared with gelatin lost between 65 81 % of its mass between 150 80 0 C. The mass loss directly correlated to the con centration of gelatin ( Figure 3 4 ). Figure 3 3 Mass loss versus temperature for PEG/bioactive glass microspheres The mass loss curves for the PVP microspheres showed unusual backwards loop s at ~450 C. This is caused by a rapid mass loss followed by a significant exothermal event ( Figure 3 5 ). The slight delay (approximately 1 min ) between mass loss and temperature increase creates an plotted versus temperature This delay may have be en caused by the heat c apacity of the sample holders. The artifact was more prono unced at faster heating rates. Samples containing gelatin also displayed this behavior w hen heated at 10 C/min The PEG microspheres did not display this behavior at heating rates between 2 10 C/min. This sharp exothermic event may correspond to microspheres that ruptured during heating. SEM analysis of the remnants of the TG/DTA heat treatment reveals hollow

PAGE 70

70 spheres many of which had ruptured. These hollow spheres appear similar to that reported by Park, et al. 115 Figure 3 4 Mass versus temperature for gelatin/bioactive glass microspheres Figure 3 5 First derivative of sample mass versus time (DTG) and thermal differential (DTA) signals for 6 wt% PVP nanocomposite microspheres.

PAGE 71

71 Figure 3 6 SEM micrograph of 10wt% PVP bioactive glass microspheres after heating to 800 C. Microspheres containing PVP and gelatin dissolved within 1 day of immersion in SBF. There were no s observed by SEM at this time ( Figure 3 7 ). The material that remained predominantly consisted of calcium and phosphorous at a 1.4:1 ratio as measured by EDS ( Figure 3 8 ) Silicon was detected in the remnant s of the gelatin microspheres. The remnants of the PVP microspheres had a weak silicon signal. Figure 3 7 Dissolution of 10 wt% PVP microspheres. Microspheres (A) prior to immersion and (B) after 1 day in SBF.

PAGE 72

72 Figure 3 8 Electron micrographs bioa ctive glass particles prior to and after immersion in SBF for 7 days (A C) Prior to immersion in SBF (D F) Material remaining after 7 days immersion in SBF with corresponding EDS spectra. The surface of the 45S5 particles drastically changed after 7 d ays in SBF ( Figure 3 8 D). The surface appeared coated with a material that had a coral like morphology. This morphology is commonly associated with hydroxyapatite nucleate d on bioactive glass surfaces hydroxyapatite 7 110 120 122 This coating was also rich in calcium in phosphorous, with a Ca:P ratio of 1.4. XRD analysis was performed to examine the crysta llinity of the calcium phosphate compounds observed after immersion in SBF ( Figure 3 9) All samples showed a weak peak around 2 = 31.8 which is the (211) reflection and strongest peak for

PAGE 73

73 hydroxyapatite. The peaks were broad, which indicates a small crystal size. No other characteristic peaks of hydroxyapatite were observed. Figure 3 9 XRD spectra of base catalyzed bioactive glass microspheres after immersion in SBF for 1 week Nucleation of hydroxyapatite occurs by reacting with ions in SBF, which was monitored by ICP AES. Hydroxyapatite formation is indicated by a decrease in phosphorous concentration. This occurred for all samples although more rapidly for SBF containing microspheres ( Figure 3 1 0 B and C) tha n SBF containing 4 5S5 particles ( Figure 3 10 A). M icrospheres produced a higher cal cium concentration after 1 day which subsequently decreased at days 3 and 7 The calcium concentration decreased more slowly in SBF containing for 45S5 particles All of t he samples released silicon (likely in the form of Si(OH) 4 ). T he concentration of Si was highest for the 6 wt% gelatin composition, which reached a value of ~30 ppm after the first day of immersion.

PAGE 74

74 Figure 3 1 0 Elemental composition of SBF containing various compositions of bioactive glasses as measured by ICP AES. The elemental composition at day 0 is the composition of SBF for all samples. Error bars indicate 95% confidence intervals. A) SBF containing 45S5 glass particles. B) SBF containing 6 wt % gelatin nanocomposite microspheres. C) SBF containing 6 wt% PVP nanocomposite microspheres. Discussion This study demonstrates as a proof of concept, that microspheres consist ing o f calcium silicate and water soluble polymers can be made by a water in oil process Solutions containing PVP (6 and 10 wt%) and gelatin (2, 6 and 10 wt%) produced This particle size corresponds to the size of the droplets in the initial emulsion as shown by Park, et al. 115 Calcium was detected by

PAGE 75

75 EDS in these microspheres at a ratio of ~1:1 with silicon. The inorganic content was 15 32 % and was consistent for microspheres over this composition range. However, the substit ution of PVP or gelatin with PEG cause d a significant change in morphology and composition. associated with the initial droplets of the emulsion. EDS revealed that these particles contain less calcium tha n microspheres made with PVP or gelatin. The PEG particles contain ed much more i norganic material (~63%) than those made with PVP or gelatin Therefore, it is likely that a different phenomenon is responsible for the morphology of the PEG silicate materi al. The incorporation of organic polymers into inorganic sol gel systems can be described in terms of classical polymer polymer mixing. The silicate phase is assumed to be a polymeric species which has a molecular weight that is increasing with time Phase separation can occur depending on the compatibility of the two polymeric components This behavior can be described by the Flory Huggins equation for a two component polymer mixture: 123 124 in which i is the volume fraction of component i, P i is the degree of polymerization of component i 12 is the interaction parameter between the two components. PEG silica gels have been shown to undergo phase separation to produce structures ranging from macroporous, bi continuous structures to aggregated particles. Nakanishi, et al. 125 reported that PEG SiO 2 phase separated gels consist of 22 32 wt%

PAGE 76

76 organic material. This correlates well to the polymeric con tent of the PEG silicate particl es measured in this study ( 26 29 wt%). Figure 3 1 1 Phase behavior in the calcium nitrate tetrahydrate (CNT) PEG TEOS system. (A B) Particulate structures observed by Kim, et al. 126 are indicated by circle and star shapes. (C E) Compositions produced in this study, which also produced submicron particulate m orphology. Phase diagram and images reproduced from Kim, et al. with permission. Kim, et al. 126 de monstrated that the Ca(NO 3 ) 2 PEG TEOS system is capable of producing a range of morphologies by phase separation ( Figure 3 11 ) Aggregated microparticles ( Figure 3 11 A B) were observed when PEG was present in the sol. T he primary particle size was inversely correlated to the concentration of PEG. These aggregated particles were observed in a similar region of the phase diagram to the

PAGE 77

77 compositions tested in this study ( Figure 3 11 C E) Although t he PEG silicate particles in this chapter appear smaller than would be predicted by the phase diagram established by Kim, et al This is likely due to the catalyst that was used in this study (basic) compared that used by Kim, et al. (acidic) Base catalyzed systems encourage silicates to form a particulate morphology. It is worth noting that despite using different catalysts similar structures were observed in both studies Kim et al. 126 also showed that the Ca:Si ratio in PEG silicate gel s (as measured by EDS ) is approximately 1/3 the molar ratio Ca:Si of the initial sol A similar observation was made in this study ; t he PEG silicate particles had a Ca:Si ratio of ~ that of the initial sol Both studies demonstrate that calcium is not completely incorporated into the silicate The slightly hig her Ca:Si ratio in this study can be attributed to pH T he water surrounding c alcium ion is more hydrolyzed at high pH. 76 This produces h ydroxyl groups which can condense with the silicate or undergo hydrogen bonding with PEG Elemental analysis of SBF showed an increase in the concentration of silicon and calcium within the first day for all compositions Calcium and silicon indicate that the glass is dissolving. T he concentration of phosphorous decreased more rapidly in SBF containing microspheres than 45S5 particles These results are consistent with those of Saravanapavan et al. 127 which showed that sol gel derived glasses react more rapidly with the ions in SBF than melt derived, 45S5 glass. The rapid loss of phosphorous indicates that the microspheres precipitate hydroxyapatite more quickly than the 45S5 particles. The rapid deg radation of the microspheres likely caused a burst release of calcium, which saturated the solution and drove calcium phosphate precipitation. The

PAGE 78

78 burst release of calcium was observed by ICP AES on the first day of immersion ( Figure 3 18 ) The PVP and ge latin microspheres did not retain their initial shape upon immersion in SBF. Rather, they appear dissolve or leaving behind hydroxyapatite and a small amount of submicron silica particles. Heating the microspheres to 800 C led to ruptured, hollow silica shells These results point to a nanocomposite morphology ( Figure 3 12 ) in which the organic polymer is the matrix that binds silicate nanoparticles at the edge of the microsphere. The submicron silica particles are generated by the base c atalyzed condensation of silicate species at the edge of the droplet. The interior of the particle remains largely organic. The interaction between the polymer and silica surface dictates the stability of the nanocomposite Microspheres made with PVP or gelatin are stable, indicating a strong preference of these polymers for the silica surface. Hydrogen bonding of the carbonyl group of PVP to silanol groups on ammonia catalyze d silica has been demonstrated in the literature by NMR spectroscopy 128 et al. observed hydrogen bond ing between PVP and fumed silica by FTIR 129 It is likely that gelatin forms a si milar hydrogen bond to the silanol groups due to the structural similariti es between gelatin and the amide like bond in the PVP ring. The free energy of the PVP silanol hydrogen bond has been calculated by density functional theory as 25 kJ/mol in the presence of water 129 This is greater than the free energy of binding polyethylene oxide to silica surface, which has been estimated to be 5.9 kJ/m ol in the presence of water 130

PAGE 79

79 Figure 3 12 Proposed structure of PVP and gelatin nanocomposite microspheres pro duced by base catalysis. (A) Silica nanoparticles form at the surface of the droplet and are bound together by the polymer. (B) H ydrogen bonding between the silanol groups and the carbonyl of PVP binds the organic and inorganic phases together. Image ad apted from Xu, et al. 128 with permission. (C) High magnification SEM image of 10 wt% PVP microsphere, which exhibi ts a surface texture indicative of the structure shown in (A). The nanocomposite structure proposed in Figure 3 12 can be disrupted by water Dissolution of the organic matrix leave s behind submicron silica particles Silicate material was observed in the remnants of the gelatin microspheres ( Figure 3 8 ) Future experiments should use small angle X ray scattering (SAXS) and transmission electron microscopy (TEM) to confirm the nanocomposite structure of the microspheres.

PAGE 80

80 Summary Three water soluble polymers (PEG, PVP, and gelatin) were tested for their ability to form microspheres with sol gel derived bioactive glasses. PVP and gelatin produced micr ospheres 1 These microspheres contain ed a significant amount of calcium and dissolve within 1 day in SBF. C ompositions made with PEG produced 0.2 silicate particles with low concentration of calcium and polymer concentration Phase separation appear ed to be driving th e formation of these PEG silicate particles. Microspheres made with PVP or gelatin appear to have a nanocomposite structure in which the polymer acts as a matrix and binder for silicate nanoparticles Future work attempting to encapsulate drugs in bioact ive glass microspheres should use PVP or gelatin b ecause they produce more stable particles T he dissolution rate of the microspheres should be slower if they are to be used as extended drug release materials.

PAGE 81

81 CHAPTER 4 PARAMETERS AFFECTING THE SIZE AND MORPHOLOGY OF ACID CATALYZED BIOACTIVE GLASS MICR OSPHERES Introductory Remarks Microspheres comprising calcium silicate glass and various water soluble polymers were made by a W/O emulsion in the previous chapter. The organic polymer was the largest comp onent of these microspheres by weight. Th e polymer component seemed to act as the matrix in a nanocomposite structure which caused the microspheres to rapidly degrade in SBF The ideal material for treating osteomyelitis should fight the disease and re generate tissue lost from the infection The microspheres described in chapter 3 degraded within one day in SBF. Microspheres should hold their shape over a period of at least one week to effectively stabilize a wound Increasing the glass content in th e nanocomposite may slow the degradation of the particles. It may be possible to do this by making microspheres at a lower pH. Acid Catalyzed Microspheres O ne of the most important factors dictating the structure of sol gel derived materials is the pH of the sol TEOS hydrolyzes more readily in acidic solutions than in basic solutions. Hydrolysis is restricted to the oil/water interface in this procedure This produces silicate species which become more soluble in the aqueous phase as the degree of hydro lysis increases Once drawn into the droplet, the silicate species can subsequently interact with PVP and/ or undergo condensation. Because hydrolysis is more rapid at low pH, droplets containing an acid catalyst should generate more hydrolyzed silica, re sulting in microspheres with more inorganic material This may also change the structure of the microsphere from discrete, silica particles dispersed in an organic coating to a more cross linked and interconnected network. The first goal of

PAGE 82

82 this chapter is to identify the differences between microspheres presented in chapter 3 (base catalyst) with those made with an acid catalyst An initial study was performed to evaluate the effects of acid catalysis on microsphere formation and composition Three compositions of microsph eres were made for this study: compositions containing PVP (molar mass ~ 1,000 kg/mol ), gelatin (molar mass ~ 50 kg/mol ), and one composition without any polymer present in the aqueous phase SEM/EDS and thermogravimetric an alysis was performed to examine the general structure and composition of these acid catalyzed microspheres. Compositional Effects on Size and Morphology The second goal of this chapter is to identify how the initial sol composition affects the final micros phere size and morphology Size and morphology are important factors that can affect the release rate of drugs and attachment of cells. Therefore, it is important to understand how the composition affects the size and shape of the final microsphere. The relationship between sol composition and microsphere size is examined in the context of droplet formation. The emulsion is crucial in this procedure because the droplets act as templates for the final microspheres. An emulsion is a mixture of two immisci ble liquids consisting of a discontinuous phase dispersed in a continuous phase Stirring the emulsion leads to the formation of droplets of the discontinuous phase. The work required for forming droplets is given by: w int is the interfacial tension. Smaller droplets can be created by adding more work into the system (such as stirring the emulsion faster). Lowering the interfacial tension also

PAGE 83

83 produces smaller droplet s for a fixed amount of work. One of the most common methods for lowering interfacial tension is through the use of a surfactant. Surfactants are molecules that contain both hydrophilic and hydrophobic groups. These molecules selectively partition at the interface between two immiscible liquids. liquids, resulting in a lower interfacial tension. The surfactant molecule may have more of a hydrophilic or hydrophobic characte r, depending on the chemistry. The HLB number (hydrophilic/lipophilic balance) is a semi quantitative measure of this cha racter. Surfactants with HLB < 10 tend to be oil soluble, while HLB > 10 are water soluble. Surfactants are most successfully employed when they are soluble in the continuous phase of the emulsion. Span 80, the non ionic surfactant used in these experiments, has HLB ~4, which makes it suitable for use in the oil phase, 1 octanol. Emulsions are thermodynamically unstable syste ms. In the absence of stirring there is a constant driving for droplets to coalesce and reduce the interfacial area between the two liquids. Polymers dissolved in the continuous phase slow the process of coalesce by limiting the diffusion of droplets. Thickeners, such as hydroxypropylcellulose used in this system, are commonly added to stabilize the emulsion. However, increasing the viscosity of the continuous phase means that more work must be added to stir the system. This balance between viscous forces and interfacial forces is characterized by a dimensionless number called the Weber number (We). The Weber number for an emulsion is given as:

PAGE 84

84 i n which d is the diameter of t he droplet, is the shear rate, and is the viscosity of the continuous phase T he droplet will rupture w hen the Weber number exceeds a critical value (We cr ) A relationship between We cr and the ratio of dispersed and continuous phase viscosities w curve 131 The Grace curve predicts that under simple shear flow, the droplet size can var y over several orders of magnitude by changing the viscosity of the dispersed phase ( Figure 4 1 ). Polymers dissolved in the aqueous phase can alter the viscosity of the d ispersed phase in this system. This establishes a connection between solution composition and size of the final microspheres, which is one of the goals of this chapter. Figure 4 1 elationship describing droplet breakup in a n emulsion Droplets undergoing breakup in simple shear are follow the relationship of the upper curve. Droplets undergoing breakup under elongational flow are shown by the bottom curve. Note that the ordinate (critical capillary number, Ca c ) is equivalent the crit ical Webe r number (We c r ) Image from Stegeman 132 with permission from John Wiley and Sons This chapter examines whether the size of the microspheres follow the behavior described by Grace 131 PVP is chosen to perform this study. PVP made distinct microspheres in chapter 3 and is soluble over at a wide range of concentrations and

PAGE 85

85 molecular weights The variables tested in this study are: PVP molecular weight (10, 40, and 1,000 kg/ mol), concentration of PVP (2, 6, and 10 wt% of the aqueous phase), and calcium concentration ( Table 4 1 ). The PVP molecular weights tested in this study are predicted to be above the critical molecular weight for entanglement for PVP in water, which is c alculated to be approximately 8 kg/mol (Appendix A ). The morphology of the microspheres should also be affected by the composition This was observed in chapter 3, most notably for the gelatin microspheres Ternary phase diagrams are used in this chapter to map out the observed morphologies of the various PVP microspheres This is done with the aim of correlating composition to the final microsp here structure. Finally, if the microspheres presented in this chapter are to be used for treating osteomyeliti s, it is important to test whether they are toxic to bone cells. Residual ethanol or alkoxy groups can impart a toxic effect to sol gel derived ceramics. An MTT assay is performed to evaluate the viability of osteoblasts exposed to the microspheres made in the size and morphology study. This assay relies on enzymes in the mitochondria of viable cells to reduce m eth ylthiazolyldiphenyl tetrazolium bromide (MTT, a yellow compound) to a purple formazan. The formazan precipitates as crystals, which are then d issolved and measured by absorbance. Materials and Methods Aqueous Phase Preparation Initial acid catalysis s tudy Microspheres were made with PVP (molar mass ~ 1,000 kg/mol, Polysciences #06067), gelatin (molar mass ~ 50 kg/mol, Sigma #G 939), and one witho ut any polymer. Aqueous solutions were made by dissolving 6 wt% of polymer in a 2.8 M

PAGE 86

86 solution of Ca(NO 3 ) 2 This corresponds to a 77S composition with molar H 2 O:TEOS ratio of 4, similar to the conditions in chapter 3. Nitric acid (1 M) was added to reach pH = 1 as measured by a digital pH meter. The volume of nitric acid added to each solution was l ess than 10 % (vol/vol) for all solutions Size and Morphology s tudy S olutions of Ca(NO 3 ) 2 and nitric acid were prepared in 100 ml volumes for each glass composition ( Table 4 1 ). PV P was dissolved in 10 ml of the appropriate salt solution at concentrations of 2, 6, or 10 wt%. PVP powders were dried at 125 C for 1 hr prior to adding to the solution. The aqueous solutions were covered and stirred overnight at room temperature with a magnetic stir bar to allow for complete dissolution. PVP was tested at three molecular weights (weight averaged) : 10 kg/mol (Polysciences #033 15); 40 kg/mol (Polysciences #01051); and 1,000 kg/mol (Polysciences #06067). The molar ratio of reactants for each composition is listed in Appendix B. Table 4 1. Sol gel glass compositions tested in this study. The calcium concentration in the glass w as controlled by the amount of CaNO 3 dissolved in the aqueous phase of the emulsion. Abbreviation 114 Initial glass c omposition ( %, mole ) Ca(NO 3 ) 2 (M) HNO 3 (M) 100S 100 SiO 2 0 CaO 0 P 2 O 5 0 0.02 86S 90 SiO 2 6 CaO 4 P 2 O 5 0.9 0.02 77S 80 SiO 2 16 CaO 4 P 2 O 5 2.8 0.02 Microsphere Synthesis The same oil phase was used in all studies. The oil phase was prepared as described in chapter 3 and consisted of 1 octanol with 1.4 % (wt/wt) hydroxypropylcellulose and 3 % (wt/wt) Span 80. The emulsion was formed by the same method described in chapter 3. Briefly, 45 g of the oil phase was added to a 250 ml glass beaker. The oil phase was stirred at 500

PAGE 87

87 rpm with a Caframo overhead laboratory mixer equipped with a pitched blade impellor As the oil phase was being stirred, 6.1 ml of the aqueous phase was added dropwise with a pipette. The emulsion was stirred for 30 min s before adding 18.7 ml of tetraethylorthosilicate (TEOS) and triethylphosphate (TEP) (0, 1.3, and 1.4 ml for glass com positions 100S 86S, and 77S, respectively). This mixture was stirred constantly for 24 hr s at room temperature. The resulting slurries were poured into 50 ml c entrifuge tubes and centrifuged at 1,500 rpm for 15 min s to isolate the microspheres. The supernatant was poured out and 30 ml of 200 proof ethano l was added to each centrifuge t ube. The samples were vortexed and centrifuged again at 1,000 rpm This pro cess was repeated twice. The samples were then dried under vacuum until the ethanol had evaporated by visual inspection. Samples were stored in a bell jar with desiccant. Rheology Rheological measurements were made on the oil and aqueous phases using a B rookfield cone and plate viscometer. A circulating water bath was used to maintain a constant temperature of 25 C during all measurements. The calibration of the viscometer was checked against PDMS standards prior to taking measurements. Laser Light Sca ttering Particle size distributions were measured using light scattering with a Coulter LS13320. Samples were analyzed using the Universal Liquid Module by suspen ding the particles in 200 proof ethanol. The optical model used for calculating particle siz e assumed completely spherical particles of silica suspended in ethanol. This optical model assumes: fluid refractive index = 1.359, real refractive index of the sample = 1.46, and imaginary refractive index of the sample = 0.01.

PAGE 88

88 SEM/EDS Upon drying, the microspheres typically produced a single, agglomerated pellet. Agglomerates were broken up by placing them between two pieces of wax paper and lightly grinding them on a flat glass surface with a ceramic pestle. The resulting powder was spread onto doubl e sided tape and blown with nitrogen gas at 10 psi to remove any weakly attached particles. The double sided tape was then placed on aluminum SEM stubs and sputter coated as described in chapter 3. Cross sectional views of the microspheres were made by em bedding the particles in epoxy and fracturing the resulting composite Microspheres were first treated with a silane coupling agent to improve the bonding with the epoxy matrix. Approximately 0.2 ml of microsphere powder was added to a 2 % (vol/vol) solu tion of 3 aminopropyltrimethoxysilane in 190 proof ethanol. This solution was vortexed for 2 min s and then centrifuged for 5 min s at 1 500 rpm Microspheres were washed twice with 5 ml of 190 proof ethanol The powder was dried on a watch glass at 100 C for 15 min s in a convection oven. The two part epoxy consisted of Epon 826 (Hexion) and Jeffamine D230 (Huntsman) in a 1:0.33 weight ratio. The epoxy components were first mixed before adding the silane t reated microspheres. The mixture was cast into a silicone mold and degassed for 30 min s under vacuum. Samples were cured at room temperature for 24 hr s, then at 60 C for 2 hr s. The epoxy was fractured by hand and mounted onto aluminum SEM stubs with do uble sided tape. Samples were then sputter coated with gold palladium as described in chapter 3.

PAGE 89

89 Thermal Analysis Thermogravimetric analysis was performed on a Seiko 320. Samples were heated at 2 C/ min under air at a flow rate of 100 ml /min. The initial sample masses were 7 12 mg. Microsphere Immersion and XRD SBF was prepared as described in chapter 3. Microspheres were immersed in SBF at a fixed ratio of 0.001 g/ ml. 45S5 bioactive glass (US Biomaterials) powder with a particle size of rticles was included as a standard material. Microspheres were removed from SBF by filtration at through a 0.4 5 stored overnight in a desiccator. XRD analysis was performed as described in chapter 3. Osteoblast Viabilit y The assay was performed using human osteoblasts (NHOst, Lonza) Cells were grown in a T75 flask to 75% confluence in growth in medi a containing 10% fetal bovine serum, 0.1% ascorbic acid, and 1000 units of g entamicin /amphotericin A. Cells were trypsini zed with a solution of 0.25% trypsin and 1 mM ethylenediaminetetraacetic acid and counted using a hemocytometer. Cells were plated into 96 well plates at a density of 5,000 cells/cm 2 and incubated for 2 hours at 37 C and 5% CO 2 prior to adding microspheres Microspheres (77S; 1,000 kg/mol PVP; 10 wt% composition ) were prepared for cell culture by washing three times with 200 proof ethanol. This was performed by vortexing 15 mg of microspheres with 10 mL of ethanol followed by c entrifugation for 5 minutes. The supernatant was removed by pipette The microspheres were then dried by placing them in an oven at 150 C for 16 hrs The microspheres were pre treated by

PAGE 90

90 soaking in 3.8 ml of media for 15 mins This solution was centri fuged and the media replaced prior to seeding the microspheres Cells were allowed to attach to the wells for 2 hrs before adding microspheres. The microspheres were suspended by repeated pipetting before adding them to the plate. Microspheres were seed ed at a density of 1 mg/cm 2 whi ch corresponded appeared to cover half of the well by microscopic evaluation The MTT assay was performed in accordance to the instructions provided in the kit (Caymen Chemical Cat No. 10009365). After incubating the cells f or 24 hrs with the incubated again for 4 hrs crystal dissolving solution was added to each well. The plate was gentl y mixed by han d for 5 mins prior to reading a bsorbance on a plate reader at 570 nm. Six replicates were tested for each condition. Cells grown in the polystyrene well served as the positive control. Cells grown in media containing 0.05% Triton X 100 serv ed as n egative control. Wells containing particles and cells were used as an optical blank to correct for any absorbance that may have been caused by the particles Results Initial Acid Catalyzed Microsphere Study One of the goals of this chapter was to u nderstand the differences between acid and base catalysis on the resulting microsphere composition. Microspheres were made with either 6 wt% PVP, 6 wt% gelatin, or no polymer in the aqueous solution at pH = 1. All of the resulting microspheres were appro ximately 5 Figure s 4 2 through Figure 4 4 ), similar to the PVP and gelatin microspheres made by base catalysis in chapter 3.

PAGE 91

91 Figure 4 2 SEM images of acid catalyzed bioactive glass microspheres made with 6 wt% PVP in the aqueous phase. EDS spectrum was taken on the spot indicated in the image. Secondary electron image, 15 keV accelerating voltage, 16 mm working distance. Microspheres made with PVP were not completely smooth and appeared to have dimples on the ir surface. A few microspheres were fractured and appeared to have an interior structure that contained many voids ( Figure 4 2 ). Microspheres made with PVP appeared larger by examination of the micrographs with gelatin (up Gelatin microspheres did not appear to have dimples or voids such as those made with PVP ( Figure 4 3 ). It was possible to form microspheres without any polymer dissolved in the aqueous droplets. These microspheres were also dimpled although not to the extent as compositions made with PVP ( Figure 4 4 ). This composition had a much higher proportion of fractured particles than compositions containing PVP or gelatin. The

PAGE 92

92 contents of the fractured particles appeared to be a fine particulate material This indicates these microspheres may have be en mechanically weaker than those containing PVP or gelatin. Figur e 4 3 SEM images of acid catalyzed bioactive glass microspheres made with 6 wt% gelatin in the aqueous phase. EDS spectrum was ta ken on the spot indicated in the image. Secondary electron image, 15 keV accelerating voltage, 16 mm working distance. All compositions had EDS peaks indicative of silicon, oxygen, and carbon ( Figure 4 2 through Figure 4 4 ). Microspheres made with PVP an d gelatin showed peaks for calcium, although they were very weak. The Ca:Si peak ratio was 3 x 10 2 and 5 x10 2 for PVP and gelatin samples, respectively. This is much lower than the Ca:Si ratio of approximately 1 for samples made by base catalysis ( cha pter 3 ) Microspheres made without any polymer in the droplets did not have a detectable calcium peak.

PAGE 93

93 Thermogravimetric analyis revealed that acid catalyzed microspheres contained mostly inorganic by weight ( Figure 4 5 ). The mass los s between 150 800 C was 48%, 38%, and 21% for microspheres made with PVP, gelatin, and no polymer, respectively. The mass loss for equivalent compositions made by base catalysis were much higher: 84% for PVP and 75% for gelatin. Furthermore, the mass los s curves for acid catalyzed microspheres did not show the that were observed in chapter 3. Figure 4 4 SEM images of acid catalyzed bioactive glass microspheres made with without any polymer in the aqueous phase. EDS spectrum was taken on the spot indicated in the image. Secondary electron image, 15 keV accelerating voltage, 10 mm working distance. XRD showed peaks characteristic of hydroxyapatite after soaking the microspheres in SBF for one week ( Figure 4 6 ) The (211) reflection at 31.8, which is the strongest peak for hydroxyap atite, was observed on all samples, although it was

PAGE 94

94 weak for 6 wt% gelatin and no polymer microspheres. The (002) reflection at 25.9 was detected on the 45S5 particles and possibly the 6 wt% PVP sample. Figure 4 5 Thermogravimetric analysis of acid catalyzed bioactive glass microspheres. Microscopic evaluation of the immersed samples revealed globules growing off of the microspheres ( Figure 4 7 A). These globules are rich in calcium and phosphorous ( Fig ure 4 7 B) and have a coral like morphology ( Figure 4 7 C) that has been associated with hydroxyapatite 7 110 120 121 This material occasionally appeared as a coating in which the underlying microspheres had dissolved or delaminated from the calcium phosphate ( Figure 4 7 D). Yet t he micr ospheres made in this study appeared to maintain their shape after soaking in SBF. They did not dissolve into submicron, particles suc h as the microspheres in chapter 3. Microsphere S ize E mulsion droplets act as t emplates for the microspheres made by thi s process. P article size s were measured in an effort to correlate size with aqueous phase viscosity

PAGE 95

95 A Weber number was calculated for each microsphere composition. The diameter was assumed to be the median (volume weighted) particle size measured by la ser light scattering. This metric of particle size appeared to correlate best with SEM observations. The interfacial tension in all instances was assumed to be 4 mN/m, which was the value Xu, et al. 133 measured for deionized water and 2 wt% span 80 in n octanol. The shear rate was estimated as 83 s 1 This value is 1/6 of the stirring speed ( rpm ) and was the recommended estimate of the shear rate given by the manufacturer of the stirrers (Caframo). The oil phase of the emulsion showed strong shear thinning behavior ( Figure 4 8 ). The viscosity of the oil phase was estimated at a shear rate of 83 s 1 by fitting the viscosity versus shear rate to a power function. This resulted in a viscosity 557 cP. The aqueous phases exhibited no shear thinning behavior (Figure 4 8) Figure 4 6 XRD spectra of acid catalyzed bioactive glass microspheres after immersion in SBF for 1 week.

PAGE 96

96 The values for dispersed phase viscosity to continuous phase viscosity d / c ) span ned nearly three orders of magnitude. The median particle sizes measured by ligh t scattering ranged between 6 (Appendix C) T hese values were used to calculate the Weber number which shows no slope on a log log plot versus viscosity ratio ( Figure 4 9 ). This means that changing the viscosity of the dispersed phase alone has little effect on the final particle size. The calculated Weber numbers are also lower than the predicted values for droplet breakup in simple shear. Figure 4 7 6 wt % gelatin microspheres remain intact after one week in SBF. (A) The surface of these spheres appears to have nodules that are rich in calcium and phosphorous (B). Higher magnification of these nodules (C) shows that they appear to have a coral like morph ology, which is indicative of hydroxyapatite. (D) A calcium phosphate rich structure that appears to have delaminated from the surface of micro spheres with made from a 6 wt% PVP solution.

PAGE 97

97 Figure 4 8 Viscosity versus shear rate for the oil phase and 77S; 1,000 kg/mol; 10 wt% aqueous phase The power function fitted to the oil phase was used to calculat e the viscosity during emulsification ( 557 cP at a shear rate of 83 s 1 ). Figure 4 9 Relationship between Weber number and viscosity ratio. Calcul ated values for microspheres are shown in blue. The theoretical value for simple shear and was calculated by the equation given by de Bruijn 134

PAGE 98

98 Microsphere Morphology Several different morphologies were observed in this study. Top down and cross sectional images of all the compositions prepared in this study can be foun d in Appendix D Mapping these morphologies on ternary phase diagrams reveals trends that hold between the three different molecular weights tested ( Figure 4 10 ). The 100S surface. This was appeared to diminish with increasing concentration of PVP Figure 4 10 Ternary phase diagrams summarizing the observed morphologies of PVP nanocomposite microspheres prepared by acid catalysis

PAGE 99

99 Many of the microspheres with the 86S composition did not fracture, particularly those made with 10 and 40 kg/mol PVP. A second attempt was made to embed the microspheres in epoxy using a slightly higher concentration of silane coupling agent. This yielded the same results. However, 86S microspheres prepared with 1,000 kg/mol PVP did fracture when embedded in epoxy. Compositions made with 2 and 6 wt% PVP appeared topologically uniform in cross section. However the contrast of the backscatte red electron signal appeared to be higher near the edge of the microsphere and darker towards the inside of the microsphere. When the PVP concentration was increased to 10 wt%, voids were observed on the inside of the particles. Figure 4 1 1 M icrospher e made with 2 wt% of 1,000 kg/mol PVP in the aqueous phase at a nominal glass composition of 77S. Micrographs on the left were taken in secondary electron (SE) and backscattered electron (BSE) modes. These voids were also observed in microspheres prepared with the highest concentration of calcium, 77S. However, voids were only observed when the molecular

PAGE 100

100 weight of PVP was 40 or 1,000 kg/mol. Voids were typically surrounded by a region that appeared smooth and had low backscattered electron signal ( Figure 4 11 ). This s mooth region was surrounded by stronger backscatter electron signal. EDS mapping revealed that the outer, rough layer contains silicon, oxygen, and calcium. The inner, smooth region does not con tain these elements. The relative size of the inner, smooth region appeared to be directly correlated to the concentration of PVP Figure 4 1 2 M icrosphere made with 10 wt% of 1,000 kg/mol PVP in the aqueous phase at a nominal glass composition of 77S. Micrographs on the left were taken in secondary electron (SE) and backscattered electron (BSE) modes. 77S m icrospheres made with the highest molecular weight (1,000 kg/mol) and concentration (10 wt%) of PVP consisted almost entirely of voids ( Figure 4 1 2 ). Many microspheres with this composition had a dense bulge that appeared to be growing off the primary particle. EDS mapping shows that these regions are rich in silicon.

PAGE 101

101 77S m icrospheres made with low molecular weight PVP (10 kg/mol) showed similar behavior to 40 kg/mol particles at 2 6 wt% PVP. However, at 10 wt% PVP, the microspheres had a very unusual morphology. The particles appeared to have a solid ( Figure 4 13 ) EDS mapping showed that both the the elements of the inorganic phase : silicon, morphology. Figure 4 13 M icrosphere s ma de with 10 wt% of 10 kg/mol PVP in the aqueous phase at a nominal glass composition of 77S. Micrographs on the left were taken in secondary electron (SE) and backscattered electron (BSE) modes. Cell Viability Osteoblasts grown in the presence of microsphe res showed a two fold increase in absorbance over the cells grown on the polystyrene dish ( positive control, Figure 4 14 ). This indicates that the microspheres are not only non toxic, but they also encourage the

PAGE 102

102 proliferation of osteoblasts. No absorbance was observed for cells treated with Triton X 100, as anticipated. Figure 4 14 Results of MTT assay of human osteoblasts exposed to 77S; 1,000 kg/mol; 10 wt% PVP microspheres Each condition was statically different according to an ANOVA ( = 0.05) with multiple comparisons test ( = 0.05). Error bars indicate 95% confidence intervals. Discussion Initial Acid Catalyzed Microsphere Study The goal of the first study in this chapter was to identify the differences between acid and base catalyzed microspheres. Acid catalyzed microspheres have a similar diameter to base catalyzed microspheres (~5 to the initial droplets of the emulsion. This indicates that the emulsion is stable over pH = 2 10 These microspheres contained were more than 50 wt% inorganic. This is attributed to the faster hydrolysis of TEOS under acidic conditions Efficient hydrolysis should allow more silicate species to be drawn into the aqueous phase. These hydrolyzed silicate species condense in the droplet to form the inorganic silica Higher

PAGE 103

103 silica content is one explanation for why acid catalyzed microspheres do not dissolve as readily in SBF as those made by base catalysis The structure of the silicate may have also increased the stability of the microspheres Acid catalysis tends to produce cross linked, polymeric gels of silica rather than particles of silica Cross sectional images did not indicate the nanocomposite structure suggested in chapter 3. Rather most of the micr ospheres appeared as a mixed phase of PVP and silica with some exhibiting domains that were rich in either the organic or inorganic components. C alcium was not incorporated efficiently into microspheres created by acid catalysis. This may be caused by a number of factors. First, c alcium ions are far less likely to be hydrolyzed at low pH This reduces the chance for direct condensation with silica. Second, the polymer s present in the aqueous phase are capable o f associating with calcium The polymers are more likely to be protonated at low pH, limiting the ability of the polymer to bind positively charged calcium ions. Although this is less likely since the pyrrolidone group is a ve ry weak base with a pKa arou nd 0.2 ( pKa of n methylpyrrolidone) 135 P hosphoro us was not detected by EDS in microspheres made by either catalyst While it may be surprising that the microspheres deficient in calcium and phosphorous form hydroxyapatite, these results are consistent with the literature. P 2 O 5 has been shown to have little effect on the precipitation of hydroxyapatite onto xerogels 127 Other studies have demonstrated that hydroxyapatite can form on silica gels that contain no calcium 136 The high surface area may be responsible for the high reactivity of xerogels with SBF It is commonly believed that the formation silica gel precedes the nucleation

PAGE 104

104 of calcium phosphate on melt derived glass 103 127 Glasses made by sol gel chemistry exhibit this porous, gel like surface, which is one explanation for why they have been observed to form hydroxyapatite more quickly than melt derived glasses 114 The hydroxyapatite peaks appeared more distinct for acid catalyzed microspheres than base catalyzed microspheres. Acid catalyzed microspheres did not dissolve as readily in SBF, which meant that more material could be recovered from the SBF. This meant that a greater quantity of powder was analyzed for these samples The 45S5 particles also showed more distinct hydroxyapatite peaks than those presented in chapter 3. This is likely due to the difference in size of the 45S5 particles tested in this chapter ( previous chapter (200 accounting for 50 times more surface area This higher surface area may cause more hydroxyapatite to precipitate from SBF. Microsphere Size The critical Weber number analysis ( Figure 4 9 ) demonstrated that particles do not follow the behavior at tributed to simple shear flow. However, the Grace curve for simple shear is not applicable to all types of mixing. A number of assumptions are made in this model. First, both the dispersed and continuous fluids are assumed to be Newtonian fluids. Second, shear is assumed to be applied slowly to the droplets in a stepwise equilibrium manner. Finally, the flow is assumed to be simple shear 137 The se conditions cause the droplet undergoes rotation The viscosity of the dispersed phase is responsible for translating the ext ernal shear stress into rotation. Rotation causes the droplet to e longate and fracture into two separate drops. These smaller drops undergo fracture until the cri tical Weber number is reached. Viscous coupling between the

PAGE 105

105 dispersed and continuous phase is critical to determining droplet size by this mechanism = 0, F igure 4 1 5 ). However, the conditions mentione d above are rarely applicable in practical mixing operations. Stegeman, et al. 132 have noted that in most industrial mixing operations, droplet breakup occurs under elongational flow. This type of flow arises from non Newtonian behavior, sudden exposure to high shear stress or from inhomogeneous flow. Elongational flow caus es d roplet s extend into long thread s The Rayleigh instability of the elongated droplet causes spontaneous break up into numerous small drops. The viscosity of the droplet plays little role in this case = 1 Figure 4 1 5 ) Figure 4 15 Critical Weber number versus viscosity ratio for droplet breakup in different types of laminar flow. The amount of elongation a droplet undergoes The Grace curve, which assumes s imple shear is described by 0 curve. Two dimensional, purely planar flow is described by the curve at 1. Image reproduced from Walstra with permission from Taylor and Francis, copyright 2002 138

PAGE 106

106 The critical Weber number analysis shows that m icrospheres formed from droplets created by elongational flow. The experimental conditions are appropriate for elongational flow and t he continuou s phase is clearly non Newtonian ( Figure 4 8 ). Sudden shear stress was imparted on the aqueous phase by adding to the oil phase under stirring. Finally, the mixing apparatus used in these experiments likely imparts a two dimensional, rotational shear, ra ther than simple shear. A more effective means to controlling droplet size in this system would be to change the shear stress (i.e. stirring speed) during emulsion formation. The critical Weber number analysis assumed that the microspheres did not shrink during drying The 100S microspheres did appear to undergo more shrinkage than other compositions (as discussed below). The effect of this shrinkage is a lower calculated Weber number. Since the 100S solutions typically had lower viscosity than other co mpositions, this may have contributed to the lack of slope in Figure 4 9 Microsphere Morphology S ystematic ally varying the composition of the particles allowed phase diagrams to be constructed which showed distinct regions of similar morphologies ( Figure 4 10 ). The phase behavior o f particles can be explained in terms of the sol gel reactions that form the microspheres Hydrolysis of the TEOS molecule occurs at the interface of the water droplet ( Figure 4 16 A) This occurs readily under the acidic condi tions. This allows the orthosilicate to become more water soluble. Silanol groups hydrogen bond with the carbonyl groups of PVP 128 139 as the molecule enters the water droplet. Unbound silanol or alkoxy groups can undergo condensation The condensation reaction is more rapid in the presence of calcium salts.

PAGE 107

107 Compositions with high calcium and PVP concentrations form a gel before the silicate has a chance to completely diffuse into the droplet. This creates a silica rich shell surrounding a core of PVP ( Figure 4 1 6 B). This phenomena explains wh y at 77S microspheres with high concentrations of PVP (40 and 1,000 kg/mol), appears to have Figure 4 10 evaporation of water from the hydrated PVP at the core of the particl e. EDS mapping has shown that the core of these particles contain very little inorganic material ( Figure 4 1 1 ). Figure 4 1 6 Proposed mechanism of microsphere formation by acid catalysis. All drawings are highly schematic do not show the surfactant, and are not drawn to scale. PVP networks are indicated by black lines. Silicate networks are indicated by purple lines.

PAGE 108

108 C ondensation does not occur as rapidly a t lower concentrations of calcium This allows the hydrolyzed silicate species to diffuse mo re thoroughly into the droplet, creating a particle that appears homogeneous ( Figure 4 10 black circles). The PVP may slow the diffusion of hydrated silica into the center of the droplet ( Figure 4 16 C). This accounts for why some of the solid particles appear ed to have a middle region with lower backscattered electron contrast than the outer region. This contrast is indicative of a difference in the average atomic number of the material caused by regions rich in PVP or silica (Appendix D ). Microspheres made without calcium (100S compositions) display ed a very unusual to phase separation by spinodal decomposition, this structure is not maintained throughout the cross section of the particle It is possible that very little condensation has occur red in these compositions prior to washing the particles A limited number of branched silicates would create a fluid droplet with relatively low density ( Figure 4 16 D). The droplet may still be fluid at the time that the reaction is stopped. Drying the microspheres at this stage would cause the microspheres to collapse. Zhao, et al. 140 observed this phenomenon for polystyrene microspheres that were cross linked in the presence of solvent. Drying the microspheres caused a wrinkled surface (Figure 4 1 7 ). L atex particles with a binary polymer composition have displayed similar morphologies to those observed in this chapter. This includes core shell structures particles with inclusions, and particles with lobes 141 142 Latex particles are also formed in an emulsion although t he polymerization mechanism is different Instead of acting as a template for the final particle, the droplets serve as

PAGE 109

109 reservoirs for the monomer. L atex particles are produced in micelles (~100s nm) outsi de of the monomer droplets. Despite the different mechanism, it is worth noting that the similar morphologies can be created in both processes. Figure 4 1 7 SEM images of microspheres with w rinkled surfaces (A) 100S; 1,000 kg/mol PVP; 2 wt% bioactive glass composite microsphere produced in this chapter. (B) Microspheres produced by Zhao, et al. 140 by drying c ross linked polystyrene microspheres that were previously swollen with toluene Reprinted w ith permission from the American Chemical Society. Cop yright 2011. Summary Acid catalysis produce d composite microspheres that contain more inorganic material and less calcium. These microspheres were less soluble in SBF and formed hydroxyapatite on their surface within 7 days The viscosity of the aqueous droplets had little effect on the final particle size due to elongational flow in the emulsion. However, viscosity and calcium concentration affect ed the morphology of the final microspheres. These morphologies were potentially caused by differences in t he rate of diffusion and gelation of silicat e species inside the droplets Microspheres approximately doubled the proliferation of osteoblasts, indicating that they are a suitable material to treat bone infections. Acid catalyzed microspheres should be u sed for

PAGE 110

110 encapsulat ing drugs because they are more stable than microspheres made by base catalysis.

PAGE 111

111 CHAPTER 5 IN VITRO DRUG RELEASE FROM BIOACTI VE GLASS MICROSPHERE S Introductory Remarks Osteomyelitis is a localized infection of bone that destroys tissue, causes severe pain, and can lead to loss of limb. Osteomyelitis affects approximately one in 5 ,000 people 143 and costs approximately $10,000 per patient to treat 36 The infection is typically caused by biofilms of Staphylococci, Streptococci, and Pseudomonas bacteria 36 37 98 143 The Gram positive bacterium Staphylococcus aureus ( S. aureus ) causes approximately 50% of all cases 37 One study reported that 53% of the S. aureus isolates from osteomyelitis and septic arthritis w ere methicillin resistant (MRSA) 36 Another study of osteomyelitis in pediatric patients showed that 44% of S. aureus isolates were MRSA 144 Resistant bacteria such as MRSA are becoming increasingly problematic for hospitals due to the limited number of antibiotics that are effective against these strains. Vancomycin, a glycopeptide a ntibiotic, is one of the few drugs effective against MRSA as well as other Gram positive bacteria. Bacteria can inoculate bone by various routes. One route is through open fractures, such as those sustained on the battlefield. Shrapnel and bone fragments present surfaces that can harbor the growth of biofilms. Osteomyelitis can also be caused by the spread of bacteria from a neighboring infection. Diabetic ulcers are a major cause of osteomyelitis for this reason. Vascularity and immune response are ty pically compromised in these cases, which renders systemic antibiotic therapy ineffective. Antibiotics must therefore be applied directly to the site of the infection to achieve a local concentration high enough to kill the biofilm.

PAGE 112

112 Buchholz 100 introduced the concept of mixing antibiotics into PMMA bone cements in 197 0. While these monolithic cements filled the void space in the bone, they were difficult to remove and did not allow for adequate drainage. Klemm 99 addressed these issues in 1976 by stringing drug loaded PMMA beads onto surgical wire which was packed into the infected site. This allowed the PMMA to be removed in a second surg ery. It has been estimated that approximately 90% of infections are cleared using this approach 99 The most common antibiotics delivered by PMMA beads are gentamicin, tobramycin, and vancomycin because they are resistant to heat given off from the polymerization of PMMA (up to 100 C). The relative success of PMMA beads i s attributed to the high concentration achieved at the site of the infection. Systemic buildup of antibiotics as measured in serum or urine does not occur with this local therapy 145 However, PMMA has disadvantages as a drug delivery material. PMMA beads must be removed within approximately 4 weeks to prevent fibrous encapsulation caused by a foreign body response 100 PMMA has also been reported to release only 25 50% of the loaded drug 146 147 This is a wasteful use of expensive antibiotics. Finally, PMMA does not encourage bone ingrowth into the dead area. A bone graft must be implanted in a second surgery to restore the bone tissue Therefore, drug loaded PMMA beads are not the ideal mate rial for treating osteomyelitis. Resorbable polymers have also been investigated for drug delivery to bone. Poly(lactic co glycolic acid) (PLGA) copolymers can exhibit drug release above the minimum inhibitory concentration (MIC) for up to 5 weeks 101 148 However, the PL GA family of copolymers lowers the local pH and can induce a late immune response 149

PAGE 113

113 Collagen sponges have the advantage of being a naturally derived material that is remodeled during the bone healing process. Yet, water soluble drugs tend to exhibit a burst release from these sponges 150 151 Chitosan is a naturally derived polysaccharide that has been used in many bi omedical and pharmaceutical applications. Chitosan microspheres loaded with vancomycin showed complete drug release within 2 weeks 152 Dissolutio n of these materials helps prevent f ibrous encapsulation However, no structural support is offered after the se material s dissolves because bone growth is not typically encouraged Calcium phosphate cements (CPCs) can offer more long term structural suppo rt. CPCs form either an apatite or brushite ceramic by a rapid acid base reaction of an amorphous calcium phosphate. These cements are attractive in space filling applications because they set to form a hard material within approximately 15 min s 153 Stallman, et al. 154 de monstrated that mixing gentamicin with the precursors of the CPC causes a slower release than adsorbing the drug onto the CPC once it has been set Hamanishi 155 showed that vancomycin could be released above the MIC for periods of 2 to 9 weeks in vivo depending on the amount of drug incorporated into the cement. Some CPC compositions exhibited f ibrous encapsulation 3 weeks. CPCs undergo limited dissolution and remodeling in vivo which causes the fibrous encapsulation observed by Hamanishi 155 Silicate glasses slowly dissolve in vivo particularly when the glass contains ions such as Na + Ca 2+ P 3 Mg 2+ and K + As discussed in chapter 3, mesoporosity can be introduced into glasses to incr ease the surface area available for the adsorption of

PAGE 114

114 drugs 107 110 Drugs such as vancomycin can also be directly encapsulated in sol gel matrices 93 Aughenbaugh, et al. 92 demonstrated that vancomycin could be encapsulated in silica xerogels in concentrations up to 30 mg/g. These xerogels were produced at H 2 O:alkoxysila ne (R) ratios of 6 and 10. Radin, et al. 94 extended this wor k by making microspheres (100 70 loaded silica. The microspheres were made by emulsifying the silica/vancomycin sols in vegetable oil. The amount of water in the sol was crucial to be able to dissolve the va ncomycin. Vancomycin was only soluble up to 6 mg/ ml in sols with H 2 O:alkoxysilane molar ratio of 4. This is far lower than the solubility of vancomycin in water (200 mg/ ml ) 156 which indicates poor solubility of vancomycin in the sol. The poor solubility is likely due to ethanol, which is a byproduct of the TEOS hydrolysis. Ethanol is a non solvent for vancomycin. This explains why more water was needed to dissolve vancomycin in the sol. The microspheres described in previous chapters were made by interfacial hydr olysis and condensation of TEOS. The initial aqueous droplets contain calcium nitrate and PVP. Alcohol is generated when TEOS hydrolyzed at the oil/water interface although most is likely carried away into the oil phase. Forming microspheres around aqu eous droplets that contain a high concentration of vancomycin may produce microspheres with greater drug loa ding than described previously. The goal s of this chapter are to demonstrate that vancomycin encapsulated in composite microspheres can be released over a period of one week and that the drug remains effective its effectiveness against Staphylococcus aureus Release was measured by UV absorption over a two week period from various microsphere

PAGE 115

115 compositions. A b acterial inhibition assay was used to en sure that vancomycin remained effective af ter the encapsulation process. Materials and Methods Microsphere Synthesis Vancomycin was loaded into bioactive glass microspheres that were prepared by acid catalysis, similar to the method described in chapter 4. Hydrochloric acid was used as the catalyst rather than nitric acid, because vancomycin is provided as a hydrochloride salt. The final hydrochloric acid concentration in the aqueous phase was 0.02 M. Microspheres were made in emulsions half the volume described in chapter 4. This was done in an effort to conserve vancomycin. All other re agents were scaled accordingly. The molar ratio of components used to create the microspheres ( Table 5 1 ) was the same as in chapter 4. Table 5 1. Molar ratio of com ponents (relative to TEOS) used to make vancomycin loaded microspheres. The concentration of PVP was 10 wt% in the initial aqueous solution for all samples. Glass c omposition PVP molecular w eight (kg/mol) TEOS CNT TEP H 2 O HCl Vancomycin 77S 1 0 1 0.20 0.10 4 1.4 x 10 3 2.4 x 10 3 77S 40 1 0.20 0.10 4 1.4 x 10 3 2.4 x 10 3 77S 1,000 1 0.20 0.10 4 1.4 x 10 3 2.4 x 10 3 86S 1,000 1 0.07 0.09 4 1.4 x 10 3 2.4 x 10 3 100S 1,000 1 0 0 4 1.4 x 10 3 2.4 x 10 3 Vancomycin was added to aqueous solutions containing of calcium nitrate and 10 wt % PVP to achieve a concentration of 36 mg/ ml These solutions were stirred overnight at room temperature to allow the vancomycin to dissolve Approximately 3 ml of the aqueous phase was added to 22 g of the oil phase while stirring at 500 rpm.

PAGE 116

116 TEOS and TEP were added and the emulsion and stirred for 24 hr s. The microspheres were washed three times with 200 proof ethanol. The samples were centrifuged at 1700 rpm and vortexed to resuspend the parti cles. The samples were dried as described in previous chapters. UV Absorption UV absorption spectra were measured on a Shimadzu UV 2410 PC UV Vis spectrophotometer. Solutions were measured in quartz cuvettes. Spectra were collected using the following p arameters: 1.0 slit width, fast scan speed, and 0.5 nm interval. Vancomycin concentration was measured by absorption of the SBF solution at 280 nm transparent 96 well plate (Corning 3635) on a M olecular Devices plate reader All SBF solutions were centrifuged prior to analysis. ml ) were prepared by serial dilution in volumetric flasks. Thermal Analysis Thermogravimetric analysis was performed on vanc omycin loaded microspheres using the following temperature profile: heating from 30 100 C ( at 10 C/min ), holding at 100 C for 2 hrs, then heating to 800 C ( at 10 C/min) The hold at 100 C was performed to determine how long the powder should be drie d at this temperature for future ex periments. The samples were purged with air at 100 ml/min during all stages of heating. SEM Samples were prepared for SEM as described in previous chapters. The microscope was operated in secondary electron mode at 15 kV accelerating voltage and 16 17 mm working distance for the images presented in this chapter.

PAGE 117

117 Drug Release Study Samples were immersed in tris(hydroxymethyl)aminomethane (Tris) buffered SBF which was prepared in the same manner as described in previous c hapters. 25 mg of vancomycin loaded microspheres were immersed in 5 ml of SBF (5 mg/ ml concentration). The samples were immersed in a shaker bath at 37 C for the duration of the study. SBF was replaced at daily intervals by centrifuging the samples at 1000 rpm, removing 4 ml of the supernatant by pipette, and replenishing with 4 ml of pre warmed SBF. The solutions were then vortexed for 5 seconds to resuspend the particles. Th e supernatant solutions were stored at 70 C and thawed prior to measuring absorbance. The calculations for cumulative release took into account the 1 ml of SBF that was not replaced each day. Release on the first day was calculated by multiplying the me ml ) by the volume drawn off (4 ml ). The amount of vancomycin in the 1 remaining ml was subtracted from the concentration measured on the following day Bacterial Inhibition Study The SBF used in the bacterial inhibition study was not buffered with Tris This is because s ome studies have indicated that bioactive glasses have a bactericidal effect which is claimed to be caused by an increase in pH caused by the dissolution of the glass 157 Tris was not used in the SBF to capture this potential effect The pH of the SBF was adjusted to 7.5 with sodium hydroxide The unbuffered SBF was aut oclaved to maintain sterility. Prior to immersing in SBF, 25 mg of microspheres (77S; 40 kg/mol PVP) were heated at 100 C for 2 hr s to remove any residual alcohol. 5 ml of SBF was added to

PAGE 118

118 the microspheres to achieve a final concentration of 5 mg/ ml The solutions were incubated at 37 C and 5% CO 2 during the experiment. 1 ml SBF was drawn off and replaced at 0.5, 1, 2, 4, and 24 hr s and at d aily intervals thereafter. The solutions drawn off of the microspheres were stored at room temperature until the bacterial assay could be performed. Inhibition of S. aureus (ATCC # 25923) was measured using a disk diffusion test. Bacteria were grown at 37 C in tryptic soy broth to an optical density of 0.2, corresponding to a cell density of ~10 6 CFU/ ml The bacteria were streaked onto 150 mm Petri dishes containing Muell er Hinton agar. Paper disks (6 mm diameter) were immersed in SBF that had been drawn off from the microspheres at certain time intervals. The SBF was centrifuged at 1000 rpm prior to immersing paper disks. Three disks were placed onto each agar plate in a randomized order. The plates were incubated overnight at 37 C. Bacterial inhibition was observed as transparent rings surrounding the paper disks The distance from the center of the disk to the edge of the transparent ring was measured using digita l calipers. This distance was measured in four orthogonal directions per disk for three replicate disks (n = 12) Results Microsphere Synthesis Microspheres containing vancomycin appeared morphologically similar to those presented in chapter 4 ( Figure 5 1 ). The particles appear ~2 were consistent with the microsph eres with the same composition prepared in chapter

PAGE 119

119 morphology that was observed in chapter 4. Figure 5 1 SEM images of vancomycin loaded microspheres prior to soaking in S BF. (A) 77S; 10 kg/mol PVP. (B) 77S; 40 kg/mol PVP. (C) 77S; 1,000 kg/mol PVP. (D) 86S; 1,000 kg/mol PVP. (E) 100S; 1,000 kg/mol PVP.

PAGE 120

12 0 Microsphere yield directly correlated to the calcium concentration in the aqueous solutions (Table 5 2). Yield was calculated by assuming no loss of PVP or vancomycin and complete conversion of TEOS, Ca(NO 3 ) 2 and TEP into SiO 2 CaO, and P 2 O 5 respectively. Compositions with the highest calcium co ncentration (77S) yielded 86 92 % of the theoretical maximum. This was significantly higher than the 16% yield of microspheres made without calcium (100S). Table 5 2 Physical properties of vancomycin loaded microspheres. Glass c omposition PVP m olecular w eight (kg/mol) Microsphere y ield (%) Mass l oss after 2 hr, 100 C h eat t reatment (%) Mass l oss between 150 800 C (%) 77S 1 0 86 8.1 45 77S 40 88 8.5 44 77S 1,000 92 12 56 86S 1,000 68 7.8 48 100S 1,000 16 5.8 87 Thermogravimetric analysis demonstrated that t he 100S; 1,000 kg/mol PVP composition had the highest organic content of all the compositions ( Figure 5 2 A). This composition lost ~ 87 % of its mass between 150 800 C. The other compositions lost 44 56 wt% of their mass in this temperature range ( Table 5 2 ). Micro spheres lost approximately 6 12 % of their initial mass after heating to 100 C for 2 hr s. The masses appeared to reach an equilibrium value at the end of this heat treatmen t ( Figure 5 2 B) This indicates heating at 100 C for 2 hours should be sufficient to dry the microspheres of any remaining water or alcohol. Microspheres were given this heat treatment prior to being exposed to bacteria

PAGE 121

121 Figu re 5 2 Thermogravimetric analysis of vancomycin loaded microspheres. (A) Heating to 800 C. The sudden drop in ma ss at 100 C is caused by the 2 hr hold at this temperature. (B) Change in mass over 2 hr s at 100 C for vancomycin loaded microspheres.

PAGE 122

122 Vancomycin Release Study UV absorption is a common ly used method to detect vancomycin in solution. Vancomycin is typ ically measured by absorption at 280 nm. However, it is important to ensure that any other compounds that may be released from the microspheres do not absorb at this wavelength. Vancomycin and PVP both show high absorbance in SBF at 220 nm ( Figure 5 3 ). The SBF drawn from microspheres without vancomy cin (77S; 40 kg/mol PVP; 10 wt% ) also showed high absorbance at 220 nm. This may have been caused by PVP released from the sample Figure 5 3. Absorbance of SBF solutions containing various compounds. microspheres (77S, 40 kg/mol PVP, 10 wt%) that did not contain vancomycin. SBF was used as the background prior to measurements being taken. Absorbance at 280 nm correlate d strongly (r 2 > 0.99) to a standard series of vancomycin solutions ( Figure 5 4 ) S BF containing PVP or microspheres had weak

PAGE 123

123 absorbance at this wavelength, corresponding to ~ ml of vancomycin. Therefore, the error for measuring vancomycin by UV abso rption ml Figure 5 4 Calibration curve for vancomycin in Tris buffered SBF. Absorbance was measured at 280 nm on a plate reader. Vancomycin release was monitored by absorbance for up to two weeks in Tris buffered SBF ( Figure 5 5 ). The highest concentration was detected on the first day for all samples. 100S; 1,000 kg/mol PVP microspheres exhibited a vancomycin concentration of ~ ml on the first day. This was the highest concentration for all the compositions. This com position released vancomycin at concentrations > ml during the first three days of the experiment. Other compositions released ~ ml o f vancomycin on the first day. The end of the release was determined by testing whether the vancomycin ml ). A single tailed t test ( = 0.05, df = 2) was used as the statistical test. The release was considered insignifica nt on the first day that the measured concentration failed this test. The release lasted 5 7 days for all samples using this criterion ( Table 5 3 ).

PAGE 124

124 Figure 5 5 Vancomycin release from various microsphere compositions measured by absorbance at 280 nm. Error bars indicate 95% confidence intervals. Table 5 3 Drug release properties of vancomycin loaded microspheres. Glass c omposition PVP m olecular w eight (kg/mol) Vancomycin c oncentration in m icrospheres (mg/g) Day s of v ancomycin c oncentration > Cumulative v ancomycin r elease Cumulative v ancomycin r elease (%) 77S 1 0 31 7 700 92 77S 40 33 5 630 77 77S 1,000 29 5 650 91 86S 1,000 43 6 790 74 100S 1,000 220 7 3600 65 The greatest quantity of vancomycin was released from the 100S; 1,000 kg/ mol from this composition over a period of 7 days ( Figure 5 6 A ) 86S and 77S compositions released approximately 5 times less ( Table 5 3 ) While the 100S composition released the greatest quantity of vancomycin, t his

PAGE 125

125 co rresponded to only 65 % of the initial drug loaded into the samples The release was more complete (74 92%) from 86S and 77S microspheres containing calcium ( Figure 5 6 B ) Figure 5 6 Cumulative vancomycin release from composite bioactive glass microspheres Release expressed as (A) mass and (B) percent vancomycin released.

PAGE 126

126 Microscopic imaging of the particles after soaking in SBF revealed that the 77S compositions made with 10 and 40 kg/mol PVP appeared hollow and broken ( Figure 5 7 ). These co mpositions did not appear to be hollow prior to soaking in SBF ( Figure 5 1 ). The 100S; 1,000 kg/mol PVP microspheres had completely dissolve d after 7 days in SBF. Figure 5 7 SEM images of vancomycin loaded microspheres after 7 days in SBF. (A) 77S; 1 0 kg/mol PVP. (B) 77S; 40 kg/mol PVP. (C) 77S; 1,000 kg/mol PVP. (D) 86S; 1,000 kg/mol PVP. The 100S; 1,000 kg/mol PVP composition is not shown because it completely dissolved after 7 days. Bacterial Inhibition Study The SBF drawn from the vancomycin loaded 77S; 40 kg/mol PVP microspheres was tested against S. aureus Bacteria were inhibited for up to 5 days ( Figure 5 8 ).

PAGE 127

127 SBF drawn from microspheres that did not contain vancomycin served as a negative control. No inhibition was observed f rom these microspheres at either 0.5 or 24 hr s. Figure 5 8 Zone of inhibition for S. aureus measured around paper disks containing the SBF drawn from a solution of vancomycin encapsulated microspheres. A 3 mm zone corresponds to the radius of the disk and hence zero inhibition. Error bars indicate 95% confidence intervals. Vancomycin concentration was calculated from the zone of inhibition data. This was done by establishing a calibration curve by for the zone of inhibition to standard vancomycin sol utions ( Figure 5 9 B). The concentration was also measured by absorbance at 280 nm for comparison ( Figure 5 9 A). Both techniques yielded similar average concentrations up to 48 hr s. Bacteria l inhibition diminished after this time while UV absorbance stil l indicat ed the presence of vancomycin.

PAGE 128

128 Figure 5 9 Vancomycin release in the bacterial inhibition study. (A) Zone of inhibition and UV absorbance were both used to measure the concentration of vancomycin. (B) Calibration curves used for calculating v ancomycin concentration by both zone of inhibition and UV absorbance.

PAGE 129

129 Discussion Adding vancomycin to the aqueous phase of the emulsion did not appear to significantly affect the formation of microspheres Vancomycin loaded microspheres appear similar to those produced in chapter 4. Some compositions appeared hollow and fractured after soaking in SBF ( Figure 5 7 ). This may have been caused by the vortexing that was performed each day after replacing SBF. Yet microspheres did not appear fractured after v ortexing three times during the washing procedure. It is possible that PVP leaches out of the microspher e; leaving it more brittle. Mass loss upon heating to 100 C is partially attributed to the removal of residual water and alcohol in the particle. The residual solvent in the microspheres directly correlated to both calcium concentration and molecular weight of PVP. The 77S; 1,000 kg/mol PVP composition lost the most weight (12%) after 2 hr s at 100 C. This was approximately 4% more than any other composition. This partially accounts for the lower final mass than the other 77S compositions. However, this comp osition had approximately 10 wt % more organic material than the other particles, a s indicated by the mass loss between 150 800 C ( Table 5 2 ). Microsphere yield was directly correlated with calcium concentration ( Table 5 2 ). This can be attributed to the faster condensation rate of silicate species in the presence of salts, as discusse d in chapter 4. The impact of yield was that it changed the effective loading rate of vancomycin in the microspheres. This effect was most dramatic for the 100S composition, which had the lowest yield (16%) and the highest loading rate ( 220 mg/g). This w as calculated under the assumption that no vancomycin was lost during in the process of forming the microspheres

PAGE 130

130 Cumulative percent release also directly correlated to calcium concentration ( Table 5 3 ). The 77S compositions released 93 95% of the ir initi al vancomycin The 100S composition released only 65%. The disparity in percent release does not correspond to the dissolution of the microspheres The 100S compositions completely dissolve after 7 days in SBF while the 77S particles were still intact a t this time. The calculation for percent release assumes that no vancomycin is lost during the process of forming the microspheres. Gupta, et al. 158 showed that vancomycin partially diffuses into the oil phase of a two part octanol/water mixture. The partition coefficient was measured as 0.2:1 (octanol:water). It is conceivable that surfactants may increase the solubility of vancomycin in octanol. Diffusion of vancomycin out of the aqueous droplets may account for the low cumulative release observed for the 100S samples. Calcium in the aqueous phase may affect this partitioning process. Calcium has been shown to change the partitioning of v ancomycin aqueous two phase system s 159 This is typically done with t wo incompatible polymers such as PEG an d dextran which form a two phase solution. Lee, et al. 160 found that 2 M NaCl greatly increase d the partitioning of vancomycin into the PEG phase The authors claim that the salt reduced the activity of water in solution resulting in increased hydrophobic interactions between vancomycin and PEG. A similar hydrophobic interaction between PVP and vancomycin may also occur in this study in the presence o f calcium Calcium also accelerates the condensation of silicate species, which may also slow the diffusion of vancomycin out of the droplet. Vancomycin can be dissolved at a higher concentration in an aqueous solution than a silica sol. This allowed the microspheres to attain a higher concentration of

PAGE 131

131 vancomycin (up to 220 mg/g) than reported for xerogels in previous studies (up to 30 mg/g) 92 161 Microspheres made by this study released a greater quantity of vancomycin (680 the same immersion conditions 94 The release was also more complete (65 95%) from these microspheres than those made in previous studies (6 36%). S. aureus was inhibited for up to 5 days b y the vancomycin released from the microspheres. This indicates that the process of encapsulating the drug does not compromise its viability. T he efficacy of the drug was diminished at days 4 and 5. This may be have been caused by degradation of the van comycin molecule. Vancomycin has been shown to degrade at 37 C in solution with a half life of 29 days 162 It is possible that the salts in the SBF accelerate the degradation of vancomycin. The degraded vancomyci n molecule may still absorb in the UV. This would account for the difference in concentration at times >3 days between bacterial inhibition and UV absorption ( Figure 5 9 ). Summary This study has demonstrated that vancomycin maintain s its bactericidal acti vity upon encapsulation in a calcium silicate composite Calcium played an important role in the encapsulation process. Calcium increased the yield of the microspheres, the inorganic content, and aided in the retention of vancomycin T he high yield of m icrospheres containing calcium led to less vancomycin per gram of microspheres. A high loading capacity (up to 220 mg/g) was achieved by calcium deficient microspheres The highest release occurred on the first day for all compositions and continued for 4 6 days under the fluid exchange conditions tested. Future work should focus on ways to

PAGE 132

132 extend the release kinetics so that the drug can be released over a period of several weeks.

PAGE 133

133 CHAPTER 6 1 KINETIC ANALYSIS OF THE ATTACHMENT OF ZO OSPORES OF THE GREEN AGLA ULVA TO SILICONE ELASTMOE RS Introductory Remarks Marine biofouling is a costly, complex and environmentally harmful phenomenon. Biofouling on ship hulls increases fuel consumption and neces sitates regular hull cleaning. A recent estimate by Schultz et al. 5 put the cost of biofouling at $1 billion over 15 years for the US Navy Arleigh Burke (DDG) class of destroyers. Hull fouling is also a primary vector for the introduction of invasive species 19 163 The majority of antifouling coatings contain biocides 40 164 but environmental issues have increased the interest in the development of non toxic antifouling coatings, which encompass a wide range of technologies 165 170 P olydimethylsiloxane elastomers (PDMSe) are non toxic materials that exhibit characteristically low adhesion to fouling organisms. Native, c ross linked PDMS is soft and not durable enough to function as a marine coating Silica fillers are commonly added to increase the strength and modulus of the material These mat eria ls have a nanocomposite structure, with silica particles forming aggregates which increase s the modulus of the rubber. Increasing the elastic modulus, however, has been shown to increase the retention of fouling organisms. Imparting a topography on the s urface of reinforced PDMSe is one route to preventing the attachment of fouling organisms. Many antifouling topographies are inspired by marine organisms that naturally exhibit low fouling such as crabs, mussels and sharks 3 64 171 Our group has developed an antifouling topography inspired by the skin of a shark, called Sharklet. This 1 Portions of this chapter have been published in Biofouling 27(8):881 892 Reprinted with permission from Taylor and Francis.

PAGE 134

134 topography has been shown to inhibit attachment of zoospores of Ulva 65 66 68 82 172 cyprids of Balanus amphitrite 66 ; and cells of the bacteria Cobetia marina 57 and Staphylococcus aureus 67 In a ki netic study, the Sharklet pattern inhibited biofilm formation of non motile S. aureus cells for up to 21 days 67 However, no kinetic analysis of attachment rates for the motile zoospores of Ulva which is a common fouling alga in the marine environment, has been p erformed on microtopographies The purpose of this study was to mea sure the initial attachment kinetics of the zoospores of Ulva on Sharklet patterned and smooth PDMSe This should provide insight into how topography functions as an antifouling surface. Observations of zoospores of Ulva reveal that the cells do not nece ssarily adhere to a surface on first contact. Rather, through temporary and therefore reversible contacts (previously 173 committing themselves to a permanent, irreversible adhesion involving the secretion of an adhesive pad 23 24 range of surface cues or features, potentially resulting in prefer ential attachment to specific locations rather than being randomly distributed over the surface. For example, spores may attach in close vicinity to previously attached spores, i.e. gregarious attachment 24 In the context of the present paper, attachment of single spores to specific topographical elements has been observed on Sharklet over a 45 min assay 174 Yet, it is unknown whether preferential attachment occurs only during the early stages of fouling or spores become less selective with time, leading to the topograp hy becoming a less effective fouling deterrent. In the present study, the distributions of spores, which

PAGE 135

135 are attached either as individuals or in groups, on the topography are mapped versus time in order to address this question. The experiment was also d settle on a smooth surface rather than the Sharklet pattern if given a choice. Test slides (19.4 cm 2 ) were prepared with various areal covera ge of the Sharklet pattern ( Figure 6 1 A C) to examine this i patterning were made by covering a 6.5 cm 2 area with the Sharklet pattern ( Figure 6 1 comparing the attachment kinetics among different sample geometries ( Figure 6 1 ). Figure 6 1 Schematic of sample arrangements on 2.54 cm x 7.62 cm glass slides. (A) 19.4 cm 2 area of smooth PDMSe coating on glass slide. (B) 6.5 cm 2 area of Sharklet micropatterned (+2.5SK2x2) PDMSe on glass slide adjacent to two 6.5 cm 2 areas of smooth PDMSe. (C) 19.4 cm 2 area Sharklet micropatterned (+2.5SK2x2) PDMSe on glass slide. (D) Scanning electron microscope image of +2.5SK2x2 topography.

PAGE 136

136 Materials and Methods Sample Preparation All test su rfaces were fabricated with polydimethylsiloxane elastomer (PDMSe, Dow Corning Silastic T2). The PDMSe surfaces were attached to glass slides 2.54 cm x 7.62 cm (total area 19.4 cm 2 ) for all experiments. All PDMSe surfaces were prepared by mixing T2 base and curing agent at a 10:1 weight ratio for 5 min s. The T2 was degassed for 30 min s under vacuum and cured fo r 24 hr s at ~ 22 C. To create smooth surfaces, T2 was cast against glass with a vapor deposited c oating of hexamethyldisilazane PDMSe surfaces with the Sharklet topography were made by casting against etched silicon wafers. The silicon wafer molds were prepared by photolithography and deep reactive ion etching as previously described 174 Silicon wafers were vapor deposited with HMDS to prevent adhesion of cured PDMSe. All samples were rinsed with 95% ethanol for approximately 10 seconds and blown with nitrogen pr ior to shipping to Birmingham, UK As noted previously 172 the topography on the Sharklet samples (e.g., +2.8SK2x2) is designated as: (feature height in m) SK (feature width in m) x (feature spacing in m) Attachment Assay for Zoospores Two kinetic attachment assays were performed on different dates. Assay 1 was performed in October, 2009 and tested sample arrangements A and B in Figure 6 1 Assay 2 was performed in April, 2010 and tested sample arrangements A, B, and C in Figure 6 1 Samples were shipped dry to the University of Birmingham. Each test surface was placed in an individual compartment (83 x 30 mm) of a Quadriperm dish (Greiner Bio one Ltd). Test surfaces were immersed in 0.22m filtered artificial seawater (ASW)

PAGE 137

137 (Tropic Marin, Aquarientechnik GmbH) for 24 h prior to the experiment. Any air visibly trapped in the features was dispersed using a stre am of seawater from a pipette. Zoospores were released from fertile tips of mature plants of Ulva linza into 0.22 described in by Schumacher, et al. 68 After filtering through 3 layers of nylon plankton net of decreasing pore size (100, 35 and 20 m ) into a glass beaker, the spore suspension was concentrated by plunging the beaker into crushed ice (spores rapidly swim towards the bottom of the beaker). The concentrated suspension of spores was pipetted into a clean beaker and the procedure was repea ted. The spore suspension was diluted with filtered ASW to give an absorbance of 0.15 at 660 nm, equivalent to 1 x 10 6 spores/ ml The suspension was kept on a magnetic stirrer to ensure the spores did not attach to the container. Ten ml of the spore sus pension were added to each compartment of the dishes containing the test slides. The dishes were incubated at room temperature (~ 20 o C) in darkness for periods up to 4 hr s. Three replicates (slides) were analyzed at each time point. At each time point th e slides were washed to remove unattached (i.e. still swimming spores by passing the slides 10 times through a beaker of ASW. Slides were subsequently fixed with 2.5% glutaraldehyde in ASW for 15 min before washing sequentially in ASW, 50% ASW and distill ed water, then dried. The density of zoospores attached to the surface was determined using a Zeiss Kontron 3000 image analysis system attached to a Zeiss Axioplan fluores cence microscope as described by Callow, et al. 25 Spores were visualized by autofluorescence of chlorophyll. Spore densities on slides A and C ( Figure 6 1 ) were counted in a line across the middle of the slide at 1 2 mm intervals, 30 counts per

PAGE 138

138 replicate slide (each 0.17 mm 2 ). Spores in the 6.5 cm 2 smooth areas adjacent to the topography (slide B, Figure 6 1 ) were counted in the same way with 15 fields of view on either side of the Sharklet pattern. Counts on the 6.5 cm 2 Sharklet pattern (slide B, Figure 6 1 ) were taken with 15 counts horizontal and 15 counts vertical to the approximate midpoint of the patter ned area. Zoospore Mapping and Grouping Distribution Attachment maps were created as described previousl y 174 175 to show the preferred attachment locations of zoospores on the topography. Thirty six transmitted light microg 2 ) per time point were used from the October 2009 assay for attachment maps and spore grouping analysis. These images were selected because they contained spores. The center of each spore was identif ied using Image J (US National Institute of Health) by overlaying the spore body with a black ellipse and subsequently using the threshold and particle measurement functions of the software. The coordinates for the center of the spore were transformed to correct for the orientation of the topography in the image. The resulting coordinates of each spore location were graphed on a unit cell of the topography and the spore location was individually checked against the original image for accuracy. The coordi nates for the spores were processed with a Kernel smoothing algorithm in Matlab (The MathWorks, Inc). This algorithm creates a smooth probability density function from a two dimensional histogram. Warm colors on the map indicate a high probability of att achment at the given time point. During the mapping process, each individual zoospore was identified as having attached in a group of 1, 2, 3, or >3 spores. The group size was determined by the number of spores visually touching in the micrographs. This data were used in the analysis of spore grouping. Attachment

PAGE 139

139 maps for groups of 2, 3, or >3 spores represent the location of each individual spore within the group. Statistical Analysis Attachment densities were counted as the average over 90 fields of view (3 replicates, 30 fields of view per replicate). 95% confidence intervals for reduction of attachment density were calculated by an arcsine transformation. SigmaStat 3.1 (Systat Soft ware, Inc.) was used to perform analysis of variance (ANOVA), Tukey, and chi squared tests. Fitting of the kinetic response was performed by a least squares fit algorithm in Matlab with the EzyFit package (Universit Paris Sud). Results Kinetic Attachment Assays Zoospores attached at a higher density on smooth surfaces than on Sharklet surfaces for every time point during the 4 hr experiment ( Figure 6 2 ). The attachment response was the same for the smooth surface regardless of whether it completely cover ed the slide or was adjacent to the Sharklet topography. Spores were inhibited by the Sharklet topography regardless of the areal coverage on the test slide (6.5 cm 2 or 19.4 cm 2 ). This kinetic attachment assay was conducted twice to e nsure reproducibility ( Figure 6 3 ). The zoospores responded to the surfaces in the same manner for both tests. The first assay, performed in October 2009, had lower zoospore attachment densities than the assay performed in April 2010. However, previous studies show that there is natural variability in the absolute numbers of attached spores from one experiment to another 24

PAGE 140

140 Figure 6 2 Kinetic response of z oospore of Ulva to smooth and Sharklet PDMSe surfaces. Error bars indicate 95% confidence intervals. Correlation c oefficients are for the fit of e quation ( 6 1) to the attachment data. Despite the variability in absolute settled spore density, a one way ANOVA (d f = 12, F = 14.065, p < 0.001) with a Tukey test (p < 0.05) showed that the reduction in zoospore density on the Sharklet surfaces is constant at 80% for up to 2 hr s ( Figure 6 3 ). The attachment reduction is lower at 70% for the 180 and 240 min time point s. These results were consistent between the two assays with the exception of assay 1 at 45 min s. The percent reduction in spore density is comparable to previous studies 65 66 68 82 172 in which the positive SK2x2 surface reduced attachment of spores by 63 86% over a 45 min assay. Mapping Spore Attachment Mapping the attachment of zoospores on the Sharklet topography qualitatively illustrates the kinetic fouling process. The maps for all spores attached to the Sharklet

PAGE 141

141 topography ( Figure 6 4 ) reveal that zoospores prefer to attach in the depressed areas between fe atures, rather than on top of them. The spores exhibit a higher selectivity for three particular sites on the Sharklet topography, as noted in a previous publication 174 Two of these sites are located at the ends of the features and a third site adjacent to midpoint of the longest feature. Fifty percent of the spores attach in these three sites, ppears to favor site 1 relative to site 3 for all times other than 15 min s. There also appears to be a decreasing selectivity for site 3 beyond 45 min s. Site 1 and site 2 appear to be equivalent in terms of spore selectivity Figure 6 3. Reduction in zoospore attachment density on Sharklet microtopography versus smooth PDMSe (6.5 cm 2 areas). Results are shown for two assays. Groups indicated by asterisks were shown to be similar (p < 0.05) by a Tukey test. Error bars indicate 95% confidence interva ls.

PAGE 142

142 Figure 6 4. Maps of all zoospores attached to the Sharklet topography during the kinetic study. The map at 45 min s indicates the sites of high preferential settlement that were originally described by Long et al. (2010a). Warm colors indicate high attachment density for that time point. Colors cannot be compared between time points. Breaking down the attachment maps for all spores ( Figure 6 4 ) into substituent groups ( Figures 6 5 through 6 8 ) provided additional insight. Most of the spores attach to the Sharklet topography as single organisms. As a result, the attachment map for single spores ( Figure 6 5 ) most closely represents the behavior of all spores attaching to the surface. The preference for attaching at the edge of the diamond pattern i s maintained as the size of the settlement group increases to two and three spores ( Figure 6 6 and 6 7 ). It also appears that as the size of the group increases, the attachment spreads out from the preferential sites observed for single spores. This

PAGE 143

143 indi cates that additional spores attach on the surface next to a spore that has already settled. Large groups (>3 spores) were not as selective, especially after 120 min s ( Figure 6 8 ). Large rafts of cells were occasionally seen on th e surface at these later times which reflects the lack of preference in the settlement maps at these time points. Figure 6 5 Settlement maps of individual spores on the Sharklet topography. Colors cannot be compared between time points. Spore Grouping The attachment of spores of Ulva on smooth surfaces shows the characteristics of appears to promote the attachment of additional spores in the same vicinity resultin g in groups attached to the surf ace. 24 The grouping of spores was compared between

PAGE 144

144 Sharklet and smooth surfaces to test whether the topography influ enced gregariousness ( Figure 6 9 ). The Sharklet topography inhibited the ability of spores to form groups of more than three spores. Chi squared tests revealed that the distributions of spore groups was statistically different between Sharklet and smooth surface at both early ( Figure 6 9 A 2 = 371, df = 3, p < 0.001) and late ( Figure 6 9 B 2 = 1143, df = 3, p < 0.001) time points. Seven percent of spores attached on smooth PDMSe as single spores after 240 min s compared to 46% of the on the Sharklet mic rotopography ( Figure 6 9 B). Figure 6 6. Settlement maps of spores in groups of 2 on the Sharklet topography. Sharp color contrast at early time points arises from low numbers of spores observed on the surface (less than approximately 50 spores). Color s cannot be compared between time points.

PAGE 145

145 Figure 6 7. Settlement maps of spores in groups of 3 on the Sharklet topography. No groups of 3 spores were observed after 15 min s. Sharp color contrast on some images arises from low numbers of spores observe d on the surface (less than approximately 50 spores). Colors cannot be compared between time points. To ensure that the results of this analysis were not caused by a lower density of spores on the Sharklet surface, the spore density after 30 min s on smooth (236 spores/ mm 2 ) and 120 min s on Sharklet (207 spores/ mm 2 ) were chosen to compare grouping at the same approximate spore density. The distribution of spore groupings is significantly different for the two surfaces ( Figure 6 9 C 2 = 446, df = 3 p < 0.001). Sixty nine percent of the spores attached as single cells on the Sharklet topography compared to only 14% on the smooth surface. Chi squared tests confirmed that the

PAGE 146

146 distributions of spore groupings are statistically different (p < 0.001) f or Sharklet and smooth ( Figure 6 9 A C). Figure 6 8. Settlement maps of spores in groups of more than 3 on the Sharklet topography. No groups of more than 3 spores were observed after 15 min s. Sharp color contrast on some images arises from low number s of spores observed on the surface (less than approximately 50 spores). Colors cannot be compared between time points. Discussion The kinetics of spore attachment on smooth surfaces over a 4 hr period reveal an approximately linear rate over the first 60 min s, but thereafter the attachment rate progressively decreases. Similar kinetics were reported by Callow, et al. 24 and were ascribed to reduced competence of the zoospore population to attach with time under

PAGE 147

147 Figure 6 9. Distribution of zoospores in groups of 1, 2, 3, or more than 3 (as determined microscopically by spores touching) at (A) 30 min s, (B) 240 min s, and (C) a spore dens ity of approximately 200 spores/ mm 2 A chi squared test for each condition (A C) showed that the Sharklet surface has a statistically different distribution from the smooth surface (p < 0.001). These samples were taken from the smooth and 6.5 cm 2 Sha rklet surface (Fig. 1A and B).

PAGE 148

148 propensity for settlement so that with time the population of spores in the assay becomes bia sed towards those with reduced propensity to settle). In the context of these experiments, two other causes of progressively reduced attachment rates may be operating beyond reduced competence. First, it is likely that with time the surface became progre ssively conditioned by organic materials either present in the water in which the spores were released, or which was secreted by the swimming spores. Such conditioning materials could make surfaces less attractive for attachment. Second, the number of sw imming spores in solution was reduced over time by attachment to both the test surfaces and especially those of the container in which the test slides are immersed. The spore attachment density on Sharklet topographies was nearly linear with time and was s ignificantly lower at each time point compared to smooth PDMSe (70 80% reduction). These results are consistent with previous assays 65 66 68 82 172 in which the microtopography reduced attachment densities f rom 63 to 86% over 45 min s. Mapping the spore attachment sites on the Sharklet microtopography showed that the zoospores preferentially attach in the recessed areas. This is consistent with previous results showing that spores attach preferentially in the recessed channels of PDMSe patterned surfaces 25 specifically at the long ends of the features 174 The maps generated in this study capture t he attachment at various time points and also categorize the spores as individuals or part of a group of spores. This was done to identify any cooperative behavior. It appears from the maps that attachment of single spores occurs at preferential sites al ong the edge of the diamond pattern, with attachment becoming more diffuse along the edge as the size of the group increases.

PAGE 149

149 This indicates that additional spores are settling adjacent to the first spore during the initial colonization of the surface. I t is worth noting that the results of mapping all spores ( Figure 6 4 ) represent a weighted average map for spore groups of 1, 2, 3 and >3 cells. The attachment maps for all cells most closely resemble the map for single spores because spores predo minantly attach as individuals The kinetic attachment curve can be fitted to the asymptotic exponential function in Equation ( 6 1) to provide a quantitative measure of the rate of adhesion. This function was used to describe attachment kinetics of bacteria to su rfaces : 176 180 ( 6 1) where A(t) is the number of attached organisms at time t, A e is the equilibrium density of reversible process: ( 6 2) in which k 1 and k 1 ar e the rates of attachment and detachment from the surface, respectively. In the case of Ulva contact with the surface warrant a model such as that in Equation ( 6 2) 24 Since these experiments were not performed in situ a direct measure of the adsorption (k 1 ) and desorption (k 1 ) cannot be obtained. Equation 6 1 is fit to the data as an estimation of the kinetic response. A least

PAGE 150

150 this time characteristic time constant, the initial flux of spores at the surface can be approximated by the quantity J o : ( 6 3) in which J o has equivalent units of flux (sporesmm 2 min 1 ). Table 6 1 Parameters A e 6 1). A e characteristic time constant and J o is the flux of spores at the surface. Assay Test a rea (cm 2 ) Surface A e 2 ) 3 ( min s 1 ) J o 2 min 1 ) r 2 October 2009 19.4 Smooth 748 9.2 6.9 0.924 13 Flat Area 639 11 7.0 0.960 6.5 +2.8SK2x2 371 3.4 1.3 0.979 19.4 Smooth 1042 11 11.5 0.975 April 2010 13 Flat Area 1135 10 11.4 0.963 6.5 +2.5SK2x2 639 2.7 1.7 0.978 19.4 +2.5SK2x2 1046 1.9 2.0 0.988 Fitting Equation 6 1 to the kinetic data show that the characteristic time constant 3.4 min s 1 ) than on smooth surfaces (9.2 11 min s 1 Table 6 1 ). The projected flux, J o offers a more physically intuitive sense of the early stage response of the spores to the surface. J o for Sharklet was consistent ly lower (1.3 2.0 sporesmm 2 min 1 ) than J o for smooth surfaces (6.9 11.5 sporesmm 2 min 1 ). The value of J o was approximately 80% lower on Sharklet than on smooth surface s within the same assay. This fitted value for the initial flux at the surface correlates well with the observed 80% reduction in spore density on Sharklet surfaces during the first 2 hr s of the assays. ll surface phenomena in bacteria including the synthesis of extracellular matrix and the exclusion of bacteria from settling next to neighbors 176 180 181 While Equation 6 1 has been effective in describing

PAGE 151

151 adhesion of non motile bacteria, it was unknown whether this Langmuir type adsorption process would effectively describe the attachment of motile spores. Indeed, the surface is far from saturated with spores. The surface area coverage is only 6.4% if two assumptions are made, i.e., area of an attached spore with its halo of adhesive on PDMSe is 64 m 2 182 and density of spores is 1000 spores/ mm 2 (maximum on the smooth PDMSe). The assay only captures the response of the zoospores which have strongly committed to adhesion a process that is fundamen tally different to the dynamic adhesion/detachment (or adsorption/desorption) observed for bacterial adhesion. Yet Equation ( 6 1 fit with high correlation (r 2 > 0.9) and J o accurately reflected the observed 80% reduction in attachment rate. The lower val Sharklet topography may reflect a reduced ability for attached spores to cooperatively recruit other spores to the surface. Additional evidence to this end is provided by comparing the grouping of spores on the smooth and Sharklet surface s. Spores attach as larger groups on smooth PDMSe ( Figure 6 9 ). After 45 min s, the percentage of spores attached as individuals on smooth PDMSe (contact angle 113) was 19%. This correlated well with a previously published value of ~ 22% spores attached a s singles on flat SAMs with a similar contact angle (110) after 1 hr 5 5 In contrast, a majority of spores (69%) on the Sharklet surface were attached as singles after 45 min s. It seems that zoospores were not able to form gregarious communities on the Sharklet pattern. This may be simply due to the fact that the early colonizing spores physically occupy those sites on the Sharklet surface that are topographically complex mechanisms may be at work whereby the Sharklet surface inhib its the ability of

PAGE 152

152 attached zoospores to recruit neighboring spores to attach. Future quantitative analysis of spore behavior on Sharklet topographies by digital holographic microscopy 23 183 may provide further insight into the mechanism(s) involved. Cooperative behavior has been observed for the attachment of Ulva to surfaces 24 A Sca tchard plot of zoospores binding to glass showed a positive slope at densities b elow approximately 1,000 spores/ mm 2 This means that at low coverage, the attached cells recruited other cells to the surface. This positively cooperative behavior has also b een shown for bacteria binding to surfaces 181 184 185 and aggregating in suspension 186 However, at high densities of attached spores, a different behavior has been observed. The Scatchard analysi s showed a negative slope, indicating that the attached spores deterred swimming spores from settling 24 The communication mechanisms underlying coop erative effects between zoospores of Ulva are essentially unknown but may include both diffusible (chemical) cues or physical, topographical cues presented by previously attached spores. It is possible that the inhibitory nature of the Sharklet topography is due to interference in the dynamics of such cooperative effects. For example, the Sharklet topography may simulate a high density of attached spores leading to negative cooperativity, a slower rate of attachment (k 1 ) and a smaller characteristic time co experiments do not identify the individual rate constants (k 1 and k 1 ), future experiments such as Scatchard analyses may be able to identify any cooperative be havior on the Sharklet surface.

PAGE 153

153 CHAPTER 7 CONCLUSIONS AND FUTU RE WORK Introductory Remarks Microbial biofilms were the first form of life that existed on earth Their long history has given rise to diverse, complex organisms from several kingdoms of life. Biofilms protect microorganisms from the environment, ensuring the survival of the species from one generation to the next. As a result, biofilms are difficult to eradicate when they threaten human health and productivity. Organic/inorganic nanocomposites were implemented in this work to address the issue of biofouling. These composites took two forms: polymer bioactive glass microspheres and antifouling topographies in silica reinforced PDMSe. Both approaches aimed to understand how surface structure correlates to the prevention of biofilms. Composite Bioactive Glass M icrospheres Organic/inorganic microspheres were created to eliminate biofilms and restore the function of bone tissue. M icrospheres encapsulated vancomycin, an antibiotic that has semi specific toxicity to Gram positive bacteria. The inorganic phase diss olved to produce calcium and silicate ions, which have been shown to stimulate bone growth 187 When the inorganic phase is formed by acid catalysis, the microspheres were s table over a period of 1 week The organic phase acted to bind the microspheres together and played a key role in determining the final morphology. The pH of the sol influenced the composition of the microspheres. 77S microspheres prepared at pH ~10.5 were 65 85 wt% organic, while 77S microspheres at pH ~2 were 38 48 wt% organic. This can be attributed to the higher rate of hydrolysis

PAGE 154

154 of TEOS under acidic conditions. The resulting microspheres were more stable in SBF when produced by acid catalysis This may be attributed t o the greater inorganic content or the structure of the gel. Calcium silicate glasses have been previously shown to stimulate osteogenic differentiation of osteoblasts. 104 187 188 However, this was not tested in this work. Alkaline phosphatase 188 osteocalcin 104 and collagen production 105 are osteogenic indicators that can be measured in vitro in future experiments. In vivo stu dies should be performed if these assays indicate that the microspheres encourage bone formation. One in vivo study has shown that vancomycin loaded xerogels cause a mild inflammatory response and fibrous encapsulation. 161 This may have been due to slow dissolution of the calcium deficient xerogel. The microspheres presented in this work may dissolve more quickly because they contain a water soluble organic component. Future work should also focus on extending the release of vancomycin. Linear drug release over 2 3 weeks is ideal for treating osteomyelitis. 101 Several approaches can be used to extend the release. Soaking the microspheres in an alkaline solution after forming them by acid catalysis should reduce the pore size by driving condensation of any rem aining silanol or alkoxy groups. Another approach is to covalently attach vancomycin to the polymer present in the aqueous phase. Carbodiimide chemistry could be used to perform this covalent attachment. 189 This could also crosslink the polymer, which may slow the drug release even further. Bone morphogenic proteins (BMPs) could also be delivered using these microspheres. BMPs have been approved by the FDA for in both spinal fusion and long bone defects. While BMPS are very effective at causing bone growth, there have been

PAGE 155

155 some issues rela ted to ectopic bone formation. 190 BMPs are typically delivered by a collagen sponge, which is soaked in a solution of the protein. This method of drug loading is dependent on handling by the surgeon which could be highly variable. Incorporating the BMPs into microspheres may provide a more accurate way to dose the drug. These microspheres should be incorporated into a paste or gel that can be easily applied by the surgeon. Ideally this paste or gel would harden to provide some mechanical stability. This may also slow the release of the encapsulated drug. The current method for creating microspheres can only be used to encapsulate water soluble drugs. This work could be extended by identifying a way to encapsulate oil soluble drugs into these microspheres. One approach is to create a double emulsion (oil in water in oil) in which the drug is dissolved in the innermost oil phase. A ntifouling Topographies on Elastomeric Films The surface topography of a silica filled PDMS elastomer was used to deter the settlement of marine algae. The organic phase, PDMS, has a surface energy of ~22 mN/m, which forms a weak adhesion to a variety of fouling organisms 48 The nanometer sized silica particles increase the toughnes s and durability to the coating, which is important for long term performance in marine applications. The long term performance was examined in an extended laboratory. The surface texture consistently inhibited marine algae by 70 80% over a 4 hr period. Image analysis showed that agal zoospores cannot easily form groups on the micro textured surface. Conversely, the organisms form clusters on smooth surfaces. This indicates that topography may prevent the early stage of biofouling by preventing the agg regation

PAGE 156

156 of individual cells. However, future work is needed to fully understand how microtopographies inhibit biofilm formation. A Scatchard analysis is one experiment that will provide additional insight into the behavior of zoospores attaching to topog raphies 181 This experiment will show whether the zoospores act cooperatively, i.e., whether the attachment of one zoospore facilities the attachment of a second. A previous study with zoospores of Ulva showed positive cooperativit y at low cell densities and negative cooperativity at high cell densities 24 If topography mimics cells already attached to the surface negative co operativity would be observed by a negative slope o n a Scatchard plot. The results in chapter 6 showed that equations derived from Langmuir type adsorption fit kinetic attachment of zoospores. This suggests that the attachment process is driven largely by physical phenomena. The Brennan Research Group has developed a model called the engineered roughness index (ERI) as a way to quantify the physical structure of the surface. The ERI model has the ability to predict the relative attachment for a variety o f species to textured surfaces 57 172 The ERI was derived from steady state (45 min ) adhesion assays. A logical extension of this work would be to combine the kinetic parameters identified in chapter 6 ( and A e ) with the ERI to create a model that can predict settlement over longer per iods of time. Amphiphilic, Nanocomposite Hydrogels as Fouling Release Coatings Amphiphilic surfaces have garnered recent interest as fouling release (FR) coatings. These materials contain both hydrophobic and hydrophilic chemical groups dispersed on the n anoscale. This chemical heterogeneity is believed to weaken the protein interaction with the surface, resulting in FR behavior.

PAGE 157

157 Amphiphilic coatings in the literature typically use fluoropolymer and PEG as the hydrophobic and hydrophilic constituents, res pectively. These chemical groups have been implemented as pendant chains of a polymer 60 in cross linked diacrylate gels 188 189 or as hybranched networks 62 Wang, et al. 188 have created perfluoropolyether/PEG gels that that have high ( ~ 60) contact angle hysteresis, which is indicative of chemical or topographical heterogeneity. The contact angle hysteres is was inversely proportional to fibrinogen adsorption. Nanocomposites have not been extensively investigated as a way of creating gel chemistry to create xerogels which contain al kyl, fluoroalkyl, and aminoalkyl groups 83 85 Trialkoxy(silanes) were used to introduce these various organic groups. Xerogels with a wide range of surface energies have be en created using this method. Relationships have been established for the surface energy of these materials and the attachment of barnacles 84 and zoospore of Ulva 87 However, only a few xerogel compositions have sh own to have discontinuous or dispersed phases with a nanoscale domain size 85 Most of the xerogels appear to be a single, mixed phase. Preliminar y work has been done to create amphiphilic nanocomposites based on poly(ethylene glycol ) methacrylate (PEGMA) and fluorocarbon functionalized silica particles. The intent is to co functionalize silica nanoparticles with methacryloxypropyl (MAP) and nonahe xyl fluorocarbon (NHF) groups These particles then act as the crosslinking agents in PEGMA gels The MAP groups allow the particles to react with the PEGMA polymer. Fluorocarbon grou ps impart nanoscale hydrophobicity to the gel. The preliminary results are shown below.

PAGE 158

158 Silica nanoparticles were made by a Stber type process. These multi functional particles were mixed with PEGMA at various concentrations and photopolymerized with Irgacure 2959 (Appendix E ).Dynamic light scattering was performed on solu tions of both methacrylate functionalized silica particles (MAP SiO 2 ) and fluorocarbon/methacrylate functionalized silica particles ( NHF /MAP SiO 2 ). The particle diameters calculated from this technique were 13 nm and 7 nm for MAP SiO 2 and NHF /MAP SiO 2 respectively. This assumed that the refractive index of the particles was n real = 1.420 and n imaginary = 0.001 and that he solvent was 190 proof alcohol. The measured diameters were smaller than the anticipated size (>50 nm) based on previous reports of this process 19 0 191 When these particles were photopolymerized with PEGMA, gels were created that swelled 600 1,000 wt% in water ( Table 7 1 ). Gels containing 25 wt% NHF /MAP SiO 2 appeared more opaque than gels containing 25 wt% MAP SiO 2 Fluorocarbon functionalized nanoparticles did not influence the final water cont ent of the gels ( Table 7 1 ), but did appear to slow the initial swelling of the gel ( Figure 7 1 ). Table 7 1. Swelling ratio of nanocomposite hydrogels after 48 hrs in DI water. Error values represent 95% confidence intervals Particle Type Particle c oncentration in d ry g el ( wt%) Swelling r atio, Q Q = (mass wet mass dry )/mass dry MAP SiO 2 1 15 1.7 MAP SiO 2 10 14 1.1 MAP SiO 2 25 6.2 1.3 NHF /MAP SiO 2 25 6.2 0.4 The captive air bubble contact angle for gels with and without fluorocarbon groups was approximately 30 ( Figure 7 2 ). This indicates that PEG dominated the surface

PAGE 159

159 when immersed in water. The contact angle was only 2 higher for gels containing fluorocarbon groups (p = 0.039, n = 20 measurements). Figure 7 1. Dynamic swelling of PE GMA gels containing various concentrations of functionalized SiO 2 nanoparticles Error bars indicate 95% confidence intervals. FESEM indicated that the material that is in contact with the gel as it was ition. Surfaces polymerized 3 ). EDS showed that surfaces polymerized against PDMSe had a very small fluorine peak. The fluorine peak was not apparent on surfaces polymerized against quartz (Figure 7 3 ). This indicated that NHF/MAP SiO 2 nanoparticles may have selectively adsorbed

PAGE 160

160 on the hydrophobic PDMSe surface, increasing the concentration of particles near this surface. Figure 7 2. C ontact angle on nanocomposite gels. (A) The samples are inverted and a angle is shown as (B) Contact angle measurements for 25 wt% MAP SiO 2 PEGMA gels and 25 wt% NHF /MAP SiO 2 PEGMA gels. Contact angle was measured for surfaces polymerized versus quartz. The results suggest that gels produced by this method may not be covalently cross linked. Rather, the gels may be physically entangled. The particle diameters measured b y dynamic light scattering are likely too small to have the numerous methacrylate groups required to crosslink the system. This would explain why there no apparent difference in swelling ratio between gels made with 1 or 10 wt% MAP SiO 2 particles. The su rfaces appear to be largely dominated by PEGMA. Dynamic swelling and EDS measurements indicate that fluorocarbon is incorporated into the material, although the effect appears to be very weak.

PAGE 161

161 Figure 7 3. SEM/EDS analysis of gels containing 25 wt% NHF/MAP SiO 2 particles. The surface polymerized against PDMSe (top row) appears to have a Surfaces polymerized against PDMSe have a very small fluorine peak, which is not apparent when the surface is polymerized against quartz. Future work on these amphiphilic nanocomposites should use larger particles to increase the number of methacrylate groups on the particles. Increasing the crosslink density of the gels should restrict the ability of the network to rearrange and bury the hydrophobic particles. This could be accomplished by incorporating PEG dimethacrylate or increasing the number of particles in the nanocomposite.

PAGE 162

162 Fluorocarbon groups with higher molecular weights could also be u sed to increase the hydrophobicity of the particles

PAGE 163

163 APPENDIX A CALCULATION OF CRITI CAL MOLECULAR WEIGHT OF ENTANGLEMENT FOR PVP IN SOLUTION The molecular weight of entanglement ( M e ) can be calculated from the principles of chain dimensions according to the paper by Fetters 192 This calculation is based on the pervaded volume of the chain ( Figure A 1 ). This pervaded volume is approximated as the smallest sphere which can encompass the chain (V sp ), which is expressed as: in which A is a constant on the order of 1 and o is the unperturbed mean square radius of gyration. Figure A 1. Spherical pervaded volume of a polymer chain (V sp ). The radius of gyration can also be written as: in which C is the characteristic ratio, M is the molecular weight of the entire chain, l o is the length of each repeat unit, and m o is the molecular weight of each repeat unit. Substituting this expression into the equation for pervaded volume:

PAGE 164

164 Now consider the number of chains (N) which completely fill the volume V sp This number of chains is calculated as: A is number. Substituting the expression for V sp from above into this equation, N is given as: It can be assumed that the number of chains with which a given chain is entangled is N 1. The condition for entanglement can be considered as when N = 2. This condition is shown in Figure A 2 In this case, there are two chains present in the pervaded volume of a single chain This means that the primary chain (in blue) is entangled wit h another chain (shown in red). Figure A 2. The condition of entanglement, in which two chains are present in the pervaded volume of a single chain (N=2) The equation above can be expressed for the entanglement condition. In this case, N is replaced with 2 and M is replaced with M e or the critical molecular weight of entanglement.

PAGE 165

165 Solving this equation for M e yields: in which B is a constant which is: This equation can be used to calculate the critical molecular weight of entanglement for PVP. Shenoy 193 used this equation to estimate the M e for PVP in ethanol (16.8 kg/mol). In this calculation, polystyrene (PS) was used to approximate the values of B and l o for PVP in the equation above However, M e for PVP in water will be different than described by Shenoy. To calculate M e of PVP, a simple ratio is set up using PS as a model: Using the values for PVP and PS given by Shenoy : 193 M e (PS) = 16.6 x 10 3 g/mol C (PVP in H 2 O) = 14 m o (PVP) = 111 g/mol = 1.13 g/cm 3 C (PS) = 10.8 m o (PS) = 104 g/mol 3 The value of M e for PVP in water is 8.15 x 10 3 g/mol.

PAGE 166

166 APPENDIX B REAGEN TS USED IN SIZE AND MORPHOLOGY STUDY OF ACID CATALYZED MICROSPHERES The reactants used in the synthesis of microspheres for the size/ morphology study i n chapter 4 are listed as moles ( Table B 1 ) and moles relative to TEOS (Table B 2) Note that the same oil phase was used for all samples, which consisted of (in moles): 0.3 1 octanol, 3.1 x 10 3 Span 80, and 1.7 x 10 6 hydroxypropylcellulose. Table B 1. Moles of reagents used in synthesizing acid catalyzed microspheres for size/morpholo gy study in chapter 4. PVP m olecular w eight (kg/mol) PVP in a queous p hase (wt %) Glass TEOS (moles x 10 3 ) CNT (moles x 10 3 ) TEP (moles x 10 3 ) H 2 O (moles) HNO 3 (moles x 10 3 ) PVP (moles x 10 6 ) 77S 8 5 17 8 .4 0.3 0.1 0.2 2 86S 85 5.6 7.5 0.3 0.1 0.1 100S 85 0.0 0 .0 0.3 0.1 0.1 77S 85 17 8.4 0.3 0.1 0.6 1,000 6 86S 85 5.6 7.5 0.3 0.1 0.5 100S 85 0.0 0.0 0.3 0.1 0.4 77S 85 17 8.4 0.3 0.1 1.0 10 86S 85 5.6 7.5 0.3 0.1 0.8 100S 85 0.0 0.0 0.3 0.1 0.7 77S 85 17 8.4 0.3 0.1 4.5 2 86S 85 5.6 7.5 0.3 0.1 3.6 100S 85 0.0 0.0 0.3 0.1 3.1 77S 85 17 8.4 0.3 0.1 14 40 6 86S 85 5.6 7.5 0.3 0.1 11 100S 85 0.0 0.0 0.3 0.1 9.7 77S 85 17 8.4 0.3 0.1 25 10 86S 85 5.6 7.5 0.3 0.1 20 100S 85 0.0 0.0 0.3 0.1 17 77S 85 17 8.4 0.3 0.1 18 2 86S 85 5.6 7.5 0.3 0.1 14 100S 85 0.0 0.0 0.3 0.1 12 77S 85 17 8.4 0.3 0.1 57 10 6 86S 85 5.6 7.5 0.3 0.1 45 100S 85 0.0 0.0 0.3 0.1 39 77S 85 17 8.4 0.3 0.1 98 10 86S 85 5.6 7.5 0.3 0.1 78 100S 85 0.0 0.0 0.3 0.1 68

PAGE 167

167 Table B 2. Relative moles of reagents (to TEOS) used in synthesizing acid catalyzed microspheres for size/morphology study in chapter 4. PVP m olecular w eight (kg/mol) PVP in a queous p hase (wt %) Glass TEOS CNT TEP H 2 O HNO 3 (x 10 3 ) PVP 77S 1 0.20 0.10 4 1.44 2.1 x 10 6 2 86S 1 0.07 0.09 4 1.44 1.7 x 10 6 100 S 1 0.00 0.00 4 1.44 1.5 x 10 6 77S 1 0.20 0.10 4 1.44 6.7 x 10 6 1,000 6 86S 1 0.07 0.09 4 1.44 5.3 x 10 6 100 S 1 0.00 0.00 4 1.44 4.6 x 10 6 77S 1 0.20 0.10 4 1.44 1.2 x 10 5 10 86S 1 0.07 0.09 4 1.44 9.2 x 10 6 100 S 1 0.00 0.00 4 1.44 8.0 x 10 6 77S 1 0.37 0.11 4 1.44 6. 8 x 10 5 2 86S 1 0.20 0.10 4 1.44 5. 4 x 10 5 100 S 1 0.07 0.09 4 1.44 4.2 x 10 5 77S 1 0.00 0.00 4 1.44 3. 7 x 10 5 40 6 86S 1 0.37 0.11 4 1.44 2.1 x 10 4 100 S 1 0.20 0.10 4 1.44 1. 7 x 10 4 77S 1 0.07 0.09 4 1.44 1.3 x 10 4 10 86S 1 0.00 0.00 4 1.44 1.2 x 10 4 100 S 1 0.37 0.11 4 1.44 3.7 x 10 4 77S 1 0.20 0.10 4 1.44 2.9 x 10 4 2 86S 1 0.07 0.09 4 1.44 2.3 x 10 4 100 S 1 0.00 0.00 4 1.44 2. 0 x 10 4 77S 1 0.37 0.11 4 1.44 2.7 x 10 4 10 6 86S 1 0.20 0.10 4 1.44 2.1 x 10 4 100 S 1 0.07 0.09 4 1.44 1.7 x 10 4 77S 1 0.00 0.00 4 1.44 1.5 x 10 4 10 86S 1 0.37 0.11 4 1.44 8.5 x 10 4 100 S 1 0.20 0.10 4 1.44 6.7 x 10 4

PAGE 168

168 APPENDIX C VISCOSITIES AND WEBE R NUMBERS OF ACID CATALYZED EMULSIONS The Weber number (We) for each microsphere composition was calculated as : i n which d is the median, volume weighted particle diameter measured by light scattering (Table C 1), is the shear rate ( 83 s 1 ) is the viscosity of the continuous phase (557 cP), and i nt is the interfacial tension (4 mN/m as measured by Xu, et al 133 ). Table C 1. Viscosity, diameter, and Weber number of acid catalyzed emulsions PVP m olecular w eight (kg/mol) PVP in aqueous phase (wt %) Glass composition Viscosity, d (cP) Particle diameter, d Weber number, We x 10 2 77S 9.10 30.9 18 2 86S 6.11 12.6 7.3 100 S 4.68 12. 5 7.3 77S 67.8 5 23. 2 13 1,000 6 86S 43.5 12.1 7.0 100 S 27.6 7.94 4.6 77S 432 30.2 18 10 86S 178 8.6 7 5.0 100 S 93.8 6.3 9 3.7 77S 4.02 15.0 8.7 2 86S 1.99 19.0 11 100 S 1.5 9.5 7 5.6 77S 10.9 18. 9 11 40 6 86S 4.52 11.2 6.5 100 S 3.54 16. 8 9.7 77S 28.2 15.2 8.8 10 86S 9.72 9.72 5.6 100 S 6.81 16. 4 9.5 77S 3.17 11.1 6.5 2 86S 1.47 16.2 9.4 100 S 1.12 23. 4 14 77S 5.62 13. 2 7.6 10 6 86S 2.38 12.0 7.0 100 S 1.67 30.2 18 77S 10.3 11.2 6.5 10 86S 3.85 7.8 1 4.5 100 S 2.45 16. 8 9.8

PAGE 169

169 APPENDIX D IMAGES OF EXTERNAL A ND INTERNAL MORPHOLOGY OF BIOACT IVE GLASS MICROSPHERES Bioactive glass microspheres were created by an acid catalyzed process over a wide range of compositions. These microspheres were imaged externally and in cross section by SEM. This appendix is a complete listing of these images. Each composition is accompanied by a ternary phase diagr am for reference ( Figure D 1 ). Figure D 1 Example of ternary phase diagram that is used as a guide throughout this appendix. M olecular weight of PVP is indicated by the color of the diagram: green= 10 kg/mol, red = 40 kg/mol, and blue = 1,000 kg/mol.

PAGE 170

170 Figure D 2 10 kg/mol, 77S, 2 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image). Figure D 3. 10 kg/mol, 77S, 6 wt% bioactive glass microspheres external mo rphology (left, secondary electron image) and in cross section (right, backscattered electron image).

PAGE 171

171 Figure D 4 10 kg/mol, 77S, 10 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backs cattered electron image). Figure D 5 10 kg/mol, 86S, 2 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, secondary electron image).

PAGE 172

172 Fig ure D 6 10 kg/mol, 86S, 6 wt% bioactive glass mi crospheres external morphology (left, secondary electron image) and in cross section (right, secondary electron image). Figure D 7 10 kg/mol, 86S, 10 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross sec tion (right, secondary electron image).

PAGE 173

173 Figure D 8 10 kg/mol, 100S, 2 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image). Figure D 9 10 kg/mol, 100S, 6 wt % bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image).

PAGE 174

174 Figure D 10. 10 kg/mol, 100S, 10 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image). Figure D 1 1 40 kg/mol, 77S, 2 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image).

PAGE 175

175 Figure D 1 2 40 kg/mol, 77S, 6 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image). Figure D 1 3 40 kg/mol, 77S, 10 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image).

PAGE 176

176 Figure D 1 4 40 kg/mol, 86S, 2 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image). Figure D 1 5 40 kg/mol, 86S, 6 wt% bioactive glass microspheres external morpholo gy (left, secondary electron image) and in cross section (right, backscattered electron image).

PAGE 177

177 F igure D 1 6 40 kg/mol, 86S, 10 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image). Figure D 1 7 40 kg/mol, 100S, 2 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image).

PAGE 178

178 Figure D 1 8 40 kg/mol, 100S, 6 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image). Figur e D 1 9 40 kg/mol, 100S, 10 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image).

PAGE 179

179 Figure D 2 0 1,000 kg/mol, 77S, 2 wt% bioactive glass microspheres external mo rphology (left, secondary electron image) and in cross section (right, backscattered electron image). Figure D 21. 1,000 kg/mol, 77S, 6 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, bac kscattered electron image).

PAGE 180

180 Figure D 22. 1,000 kg/mol, 77S, 10 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image). Figure D 23. 1,000 kg/mol, 86S, 2 wt% bio active glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image).

PAGE 181

181 Figure D 2 4 1,000 kg/mol, 86S, 6 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image). Figure D 25. 1,000 kg/mol, 86S, 10 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image).

PAGE 182

182 Figure D 26. 1,000 kg/mol, 100S, 2 wt% bioactive glass microspheres external morphology (left, secondary elect ron image) and in cross section (right, backscattered electron image). Figure D 2 7 1,000 kg/mol, 100S, 6 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image).

PAGE 183

183 Fi gure D 2 8 1,000 kg/mol, 100S, 10 wt% bioactive glass microspheres external morphology (left, secondary electron image) and in cross section (right, backscattered electron image).

PAGE 184

184 APPENDIX E PROCESS FOR CREATING N ANOCOMPOSITE HYDROGE LS Silica nanoparticles were created using a process similar to that described by Stber, Fink, and Bohn 194 First, 775 ml of 200 proof ethanol and 62 ml of a 0.8 M aqueous ammonia solution were added to a 1000 ml round bottom flask. A second solution was prepared that contained 132 ml of 200 proof ethanol and 31 ml of TEOS. Both solutions were heated in separate water baths at 60 C for at least 1 hr The solution containing TEOS was then added to the round bottom flask and the mixture was stirred at 60 C for 24 hr s. After stirri ng for 24 hr s, the silane coupling agents 3 methacryloxypropyltriethoxysilane (MAP) and/or nonahexylfluorotriethoxysilane ( NHF ) were added to the solution to at the concentrations indicated in Table D 1 The nanoparticle solutions were stirred for another 24 hr s at 60 C. The solutions were then boiled at 80 C to increase the concentration 10 fold. The solutions were then cooled and stored at approximately 4 C. Table E 1. Concentrations of reagents (M) used in making multi functional silica nanoparticles. Particle d esignation H 2 O TEOS NH 3 MAP (x 10 3 ) NHF (x 10 3 ) SiO 2 3.47 0.14 0.05 0.0 0.0 MAP SiO 2 3.47 0.14 0.05 6.5 0.0 MAP/ NHF SiO 2 3.47 0.14 0.05 3.2 3.2 The nanoparticle solutions were diluted with 190 proof ethanol. PEGMA was added by to achieve a concentration of 0.41 M, or 25 wt% of the prepolymerization mixture. Irgacure 29959 was then added by mass to a concentration of 0.01 M. The solution was sti rred with a magnetic stir bar until visibly homogenous. The

PAGE 185

185 prepolymerization mixture was then placed in a glass test tube and degassed in a sonicator for 5 min s. The mold assembly ( Figure E 1 ) was cleaned with 190 proof ethanol between uses and purged n itrogen gas for 5 min s immediately before adding the prepolymerization mixture. The samples were set inside a box 3 inches below a Blak Ray UVL 56 lamp (366 nm) and exposed to UV light for 2 hr s. A slight nitrogen purge was kept flowing during the polymerization. Figure E 1. Mold assembly for polymerizing nanocomposite hydrogels. After polymerization, the samples were removed from the mold and swollen in approximately 25 ml of deionized water. The water was exchanged three times over a 24 hr per iod before any further analysis was done. The gels were cut into 6 mm discs and dried under vacuum to perform swelling experiments. Gels were frozen in liquid nitrogen and freeze dried prior to SEM/EDS analysis.

PAGE 186

186 LIST OF REFERENCES 1. Riding R. Microbial carbonates: the geological record of calcified bacterial algal mats and biofilms. Sedimentology 2000;47:179 214. 2. Nicolaus B, Kambourova M, Oner ET. Exopolysaccharides from extremophiles: from funda mentals to biotechnology. Environ Technol 2010;31(10):1145 1158. 3. Magin CM, Cooper SP, Brennan AB. Non toxic antifouling strategies. Mater Today 2010;13(4):36 44. 4. Schultz MP. Effects of coating roughness and biofouling on ship resistance and powering. Biofouling 2007;23(5):331 341. 5. Schultz MP, Bendick JA, Holm ER, Hertel WM. Economic impact of biofouling on a naval surface ship. Biofouling 2011;27(1):87 98. 6. Hewitt CL, Gollasch S, Minchin D. Chapter 6: The vessel as a vector biofouling, ballast water, and sediments. In: Rilov G, Crooks JA, editors. Biological Invasions in Marine Ecosystems. Berlin: Springer Verlag; 2009. p 117 131. 7. Hench LL. Bioceramics from concept to clinic. J Am Ceram Soc 1991;74(7):1487 1510. 8. Epstein AK, Pokroy B, Sem inara A, Aizenberg J. Bacterial biofilm shows persistent resistance to liquid wetting and gas penetration. Proc Natl Acad Sci U S A 2011;108(3):995 1000. 9. Costerton JW, Cheng KJ, Geesey GG, Ladd TI, Nickel JC, Dasgupta M, Marrie TJ. Bacterial Biofilms in Nature and Disease. Ann Rev Microbiol 1987;41(1):435 464. 10. Costerton JW, Stewart PS, Greenberg EP. Bacterial biofilms: a common cause of persistent infections. Science 1999;284(5418):1318 1322. 11. Lamont RJ, Burne RA, Lantz MS, LeBlanc DJ, editors. Or al Microbiology and Immunology. Washington, DC: American Society for Microbiology; 2006. 12. National Health and Nutrition Examination Survey, Dental Caries (Tooth Decay) in Adults (Age 20 to 64). Bethesda, MD: National Institute of Dental and Craniofacial Research, National Institute of Health; 1999 2004. 13. Beazoglou T, Eklund S, Heffley D, Meiers J, Brown LJ, Bailit H. Economic impact of regulating the use of amalgam restorations. Public Health Rep 2007;122(5):657 63.

PAGE 187

187 14. Garret Price B. Fouling of heat exchangers: characteristics, cost, prevention, control and removal. Park Ridge, NJ: Noyes Publications; 1985. 417 p. 15. Plutarch. Book II, Question 7. In: Goodwin WW, editor. Plutarch's Miscellanies and Essays. 6th edition ed. Volume 3. Boston: Little, B rown, and Company. 16. Marine Fouling and its Prevention. Woods Hole Oceanographic Institution. Menasha, WI: George Banta Publishing Co.; 1952. 17. US Navy. Status of the Navy. Volume 2011: http://www.navy.mil/navydata/navy_legacy_hr.asp?id=146; 2011. 18. International Chamber of Shipping. Shipping and World Trade. Volume 2011: http://www.marisec.org/shippingfacts/worldtrade/number of ships.php; 2010. 19. Drake JM, Lodge DM. Hull fouling is a risk factor for intercontinental species exchange in aquatic ecosystems. Aquatic Invasions 2007;2(2 ):121 131. 20. Minchin D, Floerl O, Savini D, Occhipinti Ambrogi A. Small craft and the spread of exotic species. In: Davenport J, Davenport JL, editors. The Ecology of Transportation: Managing Mobility for the Environment: Springer Netherlands; 2006. p 99 118. 21. Colautti R, Bailey S, Overdijk C, Amundsen K, MacIsaac H. Characterised and projected costs of nonindigenous species in canada. Biol Invasions 2006;8(1):45 59. 22. Callow ME. Chapter 1: A world wide survey of slime formation on anti fouling paint s. In: Evans LV, Hoagland KD, editors. Algal Biofouling. Amsterdam: Elsevier; 1986. p 1 20. 23. Heydt M, Divs P, Grunze M, Rosenhahn A. Analysis of holographic microscopy data to quantitatively investigate three dimensional settlement dynamics of algal zo ospores in the vicinity of surfaces. Eur Phys J E 2009;30(2):141 148. 24. Callow ME, Callow JA, Pickett Heaps JD, Wetherbee R. Primary adhesion of Enteromorpha (Chlorophyta, Ulvales) propagules: quantitative settlement studies and video microscopy. J Phyco l 1997;33(6):938 947. 25. Callow ME, Jennings AR, Brennan AB, Seegert CE, Gibson A, Wilson L, Feinberg A, Baney R, Callow JA. Microtopographic cues for settlement of zoospores of the green fouling alga Enteromorpha Biofouling 2002;18(3):237 245. 26. Biofi lms: The Hypertextbook. In: Cunningham AB, Lennox JE, Ross RJ, editors: Montana State University; 2011.

PAGE 188

188 27. Brock TD. Robert Koch: A Life in Medicine and Bacteriology. Washington, DC: ASM Press; 1988. 28. Henrici AT. Studies of freshwater bacteria: I. a di rect microscopic technique. J. Bacteriol. 1933;25(3):277 287. 29. Zobell CE. The effect of solid surfaces upon bacterial activity. J Bacteriol 1943;46(1):39 56. 30. Hall Stoodley L, Costerton JW, Stoodley P. Bacterial biofilms: from the natural environment to infectious diseases. Nat Rev Micro 2004;2(2):95 108. 31. Spratt DA, Ready D, Pratten J. Chapter 11: Biofilms in medicine. In: Drr S, Thomason JC, editors. Biofouling. Chichester, United Kingdom: Wiley Blackwell; 2010. p 154 164. 32. Potera C. Forging a link between biofilms and disease. Science 1999;283(5409):1837 1839. 33. Klevens RM, Edwards JR, Richards CL, Jr., Horan TC, Gaynes RP, Pollock DA, Cardo DM. Estimating health care associated infections and deaths in U.S. hospitals, 2002. Public Health R ep 2007;122(2):160 6. 34. Weinstein RA. Nosocomial infection update. Emerg Infect Dis 1998;4(3):416 20. 35. Edwards R, Harding KG. Bacteria and wound healing. Curr Opin Infect Dis 2004;17(2):91 6. 36. Lipsky BA, Weigelt JA, Gupta V, Killian A, Peng MM. Ski n, soft tissue, bone, and joint infections in hospitalized patients: epidemiology and microbiological, clinical, and economic outcomes. Infect Control Hosp Epidemiol 2007;28(11):1290 8. 37. Waldvogel F. Osteomeylitis. In: Gorbach S, Bartlett J, Blacklow N, editors. Infectious Diseases. Philadelphia: W.B. Saunders Company; 1998. p 1339 1343. 38. Simmen H P, Blaser J. Analysis of pH and pO2 in abscesses, peritoneal fluid, and drainage fluid in the presence or absence of bacterial infection during and after ab dominal surgery. Am J Surg 1993;166(1):24 27. 39. Yebra DM, Kiil S, Dam Johansen K. Antifouling technology -past, present and future steps towards efficient and environmentally friendly antifouling coatings. Prog Org Coat 2004;50(2):75 104. 40. Finnie AA, Williams DN. Paint and coatings technology for the control of marine fouling. In: Drr S, Thomason JC, editors. Biofouling. Oxford (UK): Wiley Blackwell; 2010. p 185 206.

PAGE 189

189 41. Howell D, Behrends B. Consequences of antifouling coatings -the chemist's persp ective. In: Drr S, Thomason JC, editors. Biofouling. Oxford (UK): Wiley Blackwell; 2010. p 227 242. 42. Walker EB. Quaternary ammonium compounds. In: Paulson DS, editor. Handbook of Topical Antimicrobials: Industrial Applications in Consumer Products and Pharamceuticals. New York: Marcel Dekker; 2003. p 99 116. 43. MacGregor DR, Elliker PR. A comparison of some properties of strains of Pseudomonas aeruginosa sensitive and resitant to quaternary ammonium compounds. Can J Microbiol 1958;4(5):499 503. 44. Ken awy E R, Worley SD, Broughton R. The chemistry and applications of of the art review. Biomacromolecules 2007;8(5):1359 1384. 45. Majumdar P, Lee E, Patel N, Ward K, Stafslien SJ, Daniels J, Chisholm BJ, Boudjouk P, Callow M E, Callow JA and others. Combinatorial materials research applied to the development of new surface coatings IX: An investigation of novel antifouling/fouling release coatings containing quaternary ammonium salt groups. Biofouling 2008;24(3):185 200. 46. A lberts B, Johnson A, Lewis J, Raff M, Roberts K, Walter P. Molecular Biology of the Cell New York: Garland Science; 2002. 47. Arias CA, Murray BE. Antibiotic resistant bugs in the 21st Century a clinical super challenge. N Engl J Med 2009;360(5):439 443. 48. Baier RE. Surface behaviour of biomaterials: the theta surface for biocompatibility. J Mater Sci Mater Med 2006;17(11):1057 62. 49. Zisman W. Relation of the equilibrium contact angle to liquid and solid constitution. Contact Angle, Wettability, and Adhesion. Washington, D.C.: American Chemical Society; 2009. p 1 51. 50. In: Harris J, Zalipsky S, editors. Poly(ethylene glycol): Chemistry and Biological Applications. Washington, DC: American Chemical Society; 1997. 51. Jeon SI, Lee JH, Andrade JD, De Gennes PG. Protein -surface interactions in the presence of polyethylene oxide: I. simplified theory. J Colloid Interface Sci 1991;142(1):149 158. 52. Szleifer I. Polymers and proteins: interactions at interfaces. Curr Opin Solid State Mater Sci 1997;2(3) :337 344.

PAGE 190

190 53. Prime KL, Whitesides GM. Adsorption of proteins onto surfaces containing end attached oligo(ethylene oxide): a model system using self assembled monolayers. J Am Chem Soc 1993;115(23):10714 10721. 54. Harder P, Grunze M, Dahint R, Whitesides GM, Laibinis PE. Molecular conformation in oligo(ethylene glycol) terminated self assembled monolayers on gold and silver surfaces determines their ability to resist protein adsorption. J Phys Chem B 1998;102(2):426 436. 55. Callow ME, Callow JA, Ista LK, Coleman SE, Nolasco AC, Lopez GP. Use of self assembled monolayers of different wettabilities to study surface selection and primary adhesion processes of green algal ( Enteromorpha ) zoospores. Appl Environ Microb 2000;66(8):3249 3254. 56. Schilp S, Rosenha hn A, Pettitt ME, Bowen J, Callow ME, Callow JA, Grunze M. Physicochemical properties of (ethylene glycol) containing self assembled monolayers relevant for protein and algal cell resistance. Langmuir 2009;25(17):10077 10082. 57. Magin CM, Long CJ, Cooper SP, Ist a 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 2010;26(6):719 727. 58. Lundberg P, Bruin A, Klijnstra A. Poly(ethylene glycol) based thiol aqueous stability, and potential marine antifouling applications. ACS Appl Mater Interfaces 2010;2(3):903 912. 59. Krishnan S, Weinman CJ, Ober CK. Advances in polymers for anti biofouling surfaces. J Mater Chem 2008;18(29):3405 3413. 60. Krishnan S, Ayothi R, Hexemer A, Finlay JA, Sohn KE, Perry R, Ober CK, Kramer EJ, Callow ME, Callow JA and others. Anti biofouling properties of comblike block copolymers with amphiphilic side chains. Langmuir 2006;22(11):5075 5086. 61. Gan D, Mueller A, Wooley KL. Amphiphilic and hydrophobic surface patterns generated from hyperbranched fluoropolymer/linear polymer networks: minimally adhesive co atings via the crosslinking of hyperbranched fluoropolymers. J Polym Sci, Part A: Polym Chem 2003;41(22):3531 3540.

PAGE 191

191 62. Gudipati CS, Finlay JA, Callow JA, Callow ME, Wooley KL. The antifouling and fouling release perfomance of hyperbranched fluoropolymer ( of biomacromolecules and the green fouling alga Ulva. Langmuir 2005;21(7):3044 3053. 63. Salta M, Wharton JA, Stoodley P, Dennington SP, Goodes LR, Werwinski S, Mart U, Wood RJ, S tokes KR. Designing biomimetic antifouling surfaces. Philos Transact A Math Phys Eng Sci 2010;368(1929):4729 54. 64. Scardino AJ, de Nys R. Mini review: Biomimetic models and bioinspired surfaces for fouling control. Biofouling 2011;27(1):73 86. 65. Carman M, Estes T, Feinberg A, Schumacher J, Wilkerson W, Wilson L, Callow M, Callow J, Brennan A. Engineered antifouling microtopographies correlating wettability with cell attachment. Biofouling 2006;22(1):11 21. 66. Schumacher JF, Aldred N, Callow ME, Finla y JA, Callow JA, Clare AG, Brennan AB. Species specific engineered antifouling topographies: correlations between the settlement of algal zoospores and barnacle cyprids. Biofouling 2007;23(5 6):307 317. 67. Chung KK, Schumacher JF, Sampson EM, Burne RA, An tonelli PJ, Brennan AB. Impact of engineered surface microtopography on biofilm formation of Staphylococcus aureus Biointerphases 2007;2(2):89 94. 68. Schumacher JF, Carman ML, Estes TG, Feinberg AW, Wilson LH, Callow ME, Callow JA, Finlay JA, Brennan AB. Engineered antifouling microtopographies effect of feature size, geometry, and roughness on settlement of zoospores of the green alga Ulva Biofouling 2007;23(1):55 62. 69. Scardino AJ, Harvey E, De Nys R. Testing attachment point theory: diatom attachm ent on microtextured polyimide biomimics. Biofouling 2006;22(1):55 60. 70. Thula TT, Rodriguez DE, Lee MH, Pendi L, Podschun J, Gower LB. In vitro mineralization of dense collagen substrates: A biomimetic approach toward the development of bone graft mater ials. Acta Biomaterialia 2011;7(8):3158 3169. 71. Friess W. Collagen biomaterial for drug delivery. Eur J Pharm Biopharm 1998;45(2):113 136. 72. Weiner S, Traub W. Bone structure: from angstroms to microns. FASEB J 1992;6(3):879 885. 73. Wen JY, Wilkes G L. Organic/inorganic hybrid network materials by the sol gel approach. Chem Mater 1996;8(8):1667 1681.

PAGE 192

192 74. Landry CJT, Coltrain BK, Wesson JA, Zumbulyadis N, Lippert JL. In situ polymerization of tetraethoxysilane in polymers: chemical nature of the intera ctions. Polymer 1992;33(7):1496 1506. 75. Brennan AB, Miller TM, Vinocur RB. Hybrid organic inorganic interpenetrating networks. In: Mark JE, Lee CY C, Bianconi PA, editors. Hybrid Organic Inorganic Composites. Washington, D C: American Chemical Society; 1 995. p 142 162. 76. Brinker C, Sherer G. Sol gel Science: The Physics and Chemistry of Sol Gel Processing. San Diego, CA: Academic Press, Inc; 1990. 908 p. 77. Iler R. The Chemistry of Silica. New York: John Wiley & Sons; 1979. 866 p. 78. Lewis F. The Scie nce and Technology of Silicone Rubber. Rubber Chem. Technol. 1962;35(5):1222. 79. Stein J, Truby K, Darkangelo Wood C, Takemori M, Vallance M, Swain G, Kavanagh C, Kovach B, Schultz M, Wiebe D and others. Structure Property Relationships of Silicone Biofou ling Release Coatings: Effect of Silicone Network Architecture on Pseudobarnacle Attachment Strengths. Biofouling 2003;19(2):87 94. 80. Brady R. A fracture mechanical analysis of fouling release from nontoxic antifouling coatings. Progress in Organic Coati ngs 2001;43:188 192. 81. Chaudhury M, Finlay JA, Chung JY, Callow ME, Callow JA. The influence of elastic modulus and thickness on the release of the soft fouling green alga Ulva linza (syn. Enteromorpha linza) from poly(dimethylsiloxane) (PDMS) model netw orks. Biofouling 2005;21:41 48. 82. 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 2008;24(9):4931 4937. 83. Tang Y, Finlay JA Kowalke GL, Meyer AE, Bright FV, Callow ME, Callow JA, Wendt DE, Detty MR. Hybrid xerogel films as novel coatings for antifouling and fouling release. Biofouling 2005;21(1):59 71. 84. Finlay JA, Bennett SM, Brewer LH, Sokolova A, Clay G, Gunari N, Meyer AE, Walker GC, Wendt DE, Callow ME and others. Barnacle settlement and the adhesion of protein and diatom microfouling to xerogel films with varying surface energy and water wettability. Biofouling 2010;26(6):657 66.

PAGE 193

193 85. Gunari N, Brewer LH, Bennett SM, So kolova A, Kraut ND, Finlay JA, Meyer AE, Walker GC, Wendt DE, Callow ME and others. The control of marine biofouling on xerogel surfaces with nanometer scale topograp hy. Biofouling 2011;27(2):137 149. 86. Liu C, Zhao Q. Influence of surface energy components of Ni P TiO2 PTFE nanocomposite coatings on bacterial adhesion. Langmuir 2011;27(15):9512 9519. 87. Bennett SM, Finlay JA, Gunari N, Wells DD, Meyer AE, Walker GC, Callow ME, Callow JA, Bright FV, Detty MR. The role of surface energy and water wettability in aminoalkyl/fluorocarbon/hydrocarbon modified xerogel surfaces in the control of marine biofouli ng. Biof ouling 2010;26(2):235 246. 88. Olsen SM, Pedersen LT, Laursen MH, Kiil S, Dam Johansen K. Enzyme based antifouling coatings: a review. B iofouling 2007;23(5):369 383. 89. Luckarift HR, Dickerson MB, Sandhage KH, Spain JC. Rapid, room temperature synthesis of antibacterial bionanocomposites of lysozyme with amorphous silica or titania. Small 2006;2(5):640 643. 90. Kim J, Delio R, Dordick JS. Protease containing silicates as active antifouling materials. Biotechnol Progr 2002;18(3):551 555. 91. Jin W, Brennan JD. Properties and applications of proteins encapsulated within sol gel derived materials. Anal Chim Acta 2002;461(1):1 36. 92. Aughenba ugh W, Radin S, Ducheyne P. Silica sol gel for the controlled release of antibiotics. II. The effect of synthesis parameters on the in vitro release kinetics of vancomycin. J Biomed Mater Res 2001;57(3):321 326. 93. Radin S, Ducheyne P, Kamplain T, Tan BH. Silica sol gel for the controlled release of antibiotics. I. Synthesis, characterization, and in vitro release. J Biomed Mater Res 2001;57(2):313 320. 94. Radin S, Chen T, Ducheyne P. The controlled release of drugs from emulsified, sol gel processed sili ca microspheres. Biomaterials 2009;30(5):850 858. 95. Costache MC, Qu H, Ducheyne P, Devore DI. Polymer xerogel composites for controlled release wound dressings. Biomaterials 2010;31:6336 6343. 96. Ahola M, Kortesuo P, Kangasniemi I, Kiesvaara J, Yli Urpo A. Silica xerogel carrier material for controlled release of toremifene citrate. Int J Pharm 2000;195(1 2):219 227.

PAGE 194

194 97. Rich J, Kortesuo P, Ahola M, Yli Urpo A, Kiesvaara J, Seppl J. Effect of the caprolactone co lactide) on t oremifene citrate release from copolymer/silica xerogel composites. Int J Pharm 2001;212(1):121 130. 98. Brooks G, Pons V. Osteomyelitis. In: Hoeprich P, Jordan M, Ronald A, editors. Infectious Diseases: A Treatise of Infectious Processes. Philadelphia: J. B. Lippincott Company; 1994. p 1390 1391. 99. Klemm K. The use of antibiotic containing bead chains in the treatment of chronic bone infections. Clinical Microbiol Infec 2000;7(1):28 31. 100. Henry SL, Galloway KP. Local antibacterial therapy for the manag ement of orthopaedic infections. Pharmacokinetic considerations. Clin Pharmacokinet 1995;29(1):36 45. 101. Kanellakopoulou K, Giamarellos Bourboulis EJ. Carrier systems for the local delivery of antibiotics in bone infections. Drugs 2000;59(6):1223 32. 102 Habraken WJEM, Wolke JGC, Jansen JA. Ceramic composites as matrices and scaffolds for drug delivery in tissue engineering. Adv Drug Delivery Rev 2007;59(4 5):234 248. 103. Filgueiras MR, Latorre G, Hench LL. Solution effects on the surface reactions of a bioactive glass. J Biomed Mater Res 1993;27(4):445 453. 104. Xynos ID, Hukkanen MVJ, Batten JJ, Buttery LD, Hench LL, Polak JM. Bioglass 45S5 stimulates osteoblast turnover and enhances bone formation in vitro : implications and applications for bone tiss ue engineering. Calcified Tissue Int 2000;67(4):321 329. 105. Bosetti M, Zanardi L, Hench L, Cannas M. Type I collagen production by osteoblast like cells cultured in contact with different bioactive glasses. J Biomed Mater Res A 2003;64A(1):189 195. 106. Kawanabe K, Okada Y, Matsusue Y, Iida H, Nakamura T. Treatment of osteomyelitis with antibiotic soaked porous glass ceramic. J Bone Joint Surg Br 1998;80 B(3):527 530. 107. Vallet Regi M, Balas F, Colilla M, Manzano M. Bone regenerative bioceramic implants with drug and protein controlled delivery capability. Prog Solid State Chem 2008;36(3):163 191. 108. Vallet Regi M, Ramila A, del Real RP, Perez Pariente J. A new property of MCM 41: drug delivery system. Chem Mater 2001;13(2):308 311.

PAGE 195

195 109. Cauda V, Fiori lli S, Onida B, Vern E, Vitale Brovarone C, Viterbo D, Croce G, Milanesio M, Garrone E. SBA 15 ordered mesoporous silica inside a bioactive glass ceramic scaffold for local drug delivery. J Mater Sci Mater Med 2008;19(10):3303 3310. 110. Arcos D, Lopez Noriega A, Ruiz Hernandez E, Terasaki O, Vallet Regi M. Ordered Mesoporous Microspheres for Bone Grafting and Drug Delivery. Chem Mater 2009;21(6):1000 1009. 111. Xia W, Chang J, Lin J, Zhu J. The pH controlled dual drug release from mesoporous bioactive g lass/polypeptide graft copolymer nanomicelle composites. Eur J Pharm Biopharm 2008;69(2):546 552. 112. Hatcher B, Seegert C, Brennan A. Polyvinylpyrrolidone modified bioactive glass fibers as tissue constructs: In vitro mesenchymal stem cell response. J Bi omed Mater Res A 2003;66A(4):840 849. 113. Costa HS, Rocha MF, Andrade GI, Barbosa Stancioli EF, Pereira MM, Orefice RL, Vasconcelos WL, Mansur HS. Sol gel derived composite from bioactive glass polyvinyl alcohol. J Mater Sci 2008;43(2):494 502. 114. Li R, Clark AE, Hench LL. An investigation of bioactive glass powders by sol gel processing. J Appl Biomater 1991;2(4):231 239. 115. Park J H, Oh C, Shin S I, Moon S K, Oh S G. Preparation of hollow silica microspheres in W/O emulsions with polymers. J Colloid Interface Sci 2003;266(1):107 114. 116. Park JH, Bae SY, Oh SG. Fabrication of hollow silica microspheres through the self assembly behavior of polymers in W/O emulsion. Chem Lett 2003;32(7):598 599. 117. Leuner C, Dressman J. Improving drug solubility for oral delivery using solid dispersions. Eur J Pharm Biopharm 2000;50(1):47 60. 118. Young S, Wong M, Tabata Y, Mikos AG. Gelatin as a delivery vehicle for the controlled release of bioactive molecules. J Control Release 2005;109(1 3):256 74. 119. Yan WQ, O ka M, Nakamura T. Bone bonding in bioactive glass ceramics combined with bone matrix gelatin. J Biomed Mater Res 1998;42(2):258 65. 120. Kokubo T, Takadama H. How useful is SBF in predicting in vivo bone bioactivity? Biomaterials 2006;27(15):2907 2915.

PAGE 196

196 121 Kokubo T, Kushitani H, Sakka S, Kitsugi T, Yamamuro T. Solutions able to reproduce in vivo surface structure changes in bioactive glass ceramic A W. J Biomed Mater Res 1990;24(6):721 734. 122. Hench L, Splinter R, Allen W, Greenlee T. Bonding Mechanisms at the Interface of Ceramic Prosthetic Materials. Journal of Biomedical Materials Research 1971;5(6):117 141. 123. Nakanishi K. Sol gel process of oxides accompanied by phase separation. Bull Chem Soc Jpn 2006;79(5):673 691. 124. Nakanishi K. Pore Structur e Control of Silica Gels Based on Phase Separation. Journal of Porous Materials 1997;4(2):67 112. 125. Nakanishi K, Komura H, Takahashi R, Soga N. Phase separation in silica sol gel system containing poly(ethylene oxide) 1. phase relation and gel morpholog y. Bull Chem Soc Jpn 1994;67(5):1327 1335. 126. Kim IY, Kawachi G, Kikuta K, Cho SB, Kamitakahara M, Ohtsuki C. Preparation of bioactive spherical particles in the CaO SiO2 system through sol gel processing under coexistence of poly(ethylene glycol). J Eur Ceram Soc 2008;28(8):1595 1602. 127. Saravanapavan P, Jones JR, Pryce RS, Hench LL. Bioactivity of gel glass powders in the CaO SiO2 system: A comparison with ternary (CaO P2O5 SiO2) and quaternary glasses (SiO2 CaO P2O5 Na2O). J Biomed Mater Res A 2003;66A(1):110 119. 128. Xu Y, Wu D, Sun Y, Chen W, Yuan H, Deng F, Wu Z. Effect of polyvinylpyrrolidone on the ammonia catalyzed sol gel process of TEOS: Study by in situ 29Si NMR, scattering, and rheology. Colloids Surf, A 2007;305(1 3):97 104. 129. Gun 'ko VM, Voronin EF, Nosach LV, Pakhlov EM, Voronina OE, Guzenko NV, Kazakova OA, Leboda R, Skubiszewska Zieba J. Nanocomposites with fumed silica/poly(vinyl pyrrolidone) prepared at a low content of solvents. Applied Surface Science 2006;253(5):2801 2811. 130. Pattanayek SK, Juvekar VA. Prediction of adsorption of nonionic polymers from aqueous solutions to solid surfaces. Macromolecules 2002;35(25):9574 9585. 131. Grace HP. Dispersion phenomena in high viscosity immiscible fluid systems and application of static mixers as dispersion devices in such systems. Chem Eng Commun 1982;14(3 6):225 277.

PAGE 197

197 132. Stegeman YW, Vosse FNVD, Meijer HEH. On the applicability of the grace curve in practical mixing operations. Can J Chem Eng 2002;80(4):1 6. 133. Xu JH, Li SW, L an WJ, Luo GS. Microfluidic approach for rapid interfacial tension measurement. Langmuir 2008;24(19):11287 11292. 134. Jansseune T, Mewis J, Moldenaers P, Minale M, Maffettone PL. Rheology and rheological morphology determination in immiscible two phase po lymer model blends. J Non Newtonian Fluid Mech 2000;93(1):153 165. 135. Adelman RL. Studies on the base strengths of n,n disubstituted amides. J Org Chem 1964;29(7):1837 1844. 136. Li PJ, Ohtsuki C, Kokubo T, Nakanishi K, Soga N, Nakamura T, Yamamuro T. Ap atite formation induced by silica gel in a simulated body fluid. J Am Ceram Soc 1992;75(8):2094 2097. 137. Walstra P, Smulders PEA. Emulsion Formation. In: Brinks BP, editor. Modern Aspects of Emulsion Science. Cambridge, UK: The Royal Society of Chemistry ; 1998. p 56 114. 138. Walstra P. Physical Chemistry of Foods. New York: Marcel Dekker, Inc.; 2003. 139. Chan C K, Chu IM, Ou C F, Lin Y W. Interfacial interactions and their influence to phase behavior in poly(vinyl pyrrolidone)/silica hybrid materials pr epared by sol gel process. Mater Lett 2004;58(17 18):2243 2247. 140. Zhao T, Qiu D. One Pot Synthesis of Highly Folded Microparticles by Suspension Polymerization. Langmuir 2011. 141. Sundberg DC, Durant YG. Latex particle morphology, fundamental aspects: A review. Polym React Eng 2003;11(3):379 432. 142. Stubbs JM, Sundberg DC. The dynamics of morphology development in multiphase latex particles. Prog Org Coat 2008;61(2 4):156 165. 143. Smeltzer M, Nelson C. Osteomyelitis and septic arthritis. In: Honeyman A, Friedman H, Bendinelli M, editors. Staphylococcus Aureus Infection and Disease. New York: Kluwer Academic Publishers; 2002. p 213 246. 144. Hawkshead JJ, 3rd, Patel NB, Steele RW, Heinrich SD. Comparative severity of pediatric osteomyelitis attributabl e to methicillin resistant versus methicillin sensitive Staphylococcus aureus. J Pediatr Orthop 2009;29(1):85 90.

PAGE 198

198 145. Wahlig H, Dingeldein E, Bergmann R, Reuss K. The release of gentamicin from polymethylmethacrylate beads. An experimental and pharmacokin etic study. J Bone Joint Surg Br 1978;60 B(2):270 275. 146. Rushton N. Applications of local antibiotic therapy. Eur J Surg Suppl 1997(578):27 30. 147. Kent ME, Rapp RP, Smith KM. Antibiotic beads and osteomyelitis: here today, what's coming tomorrow? Orth opedics 2006;29(7):599 603. 148. Liu SJ, Ueng SWN, Lin SS, Chan EC. In vivo release of vancomycin from biodegradable beads. J Biomed Mater Res 2002;63(6):807 813. 149. Athanasiou KA, Agrawal CM, Barber FA, Burkhart SS. Orthopaedic applications for PLA PGA biodegradable polymers. Arthroscopy 1998;14(7):726 737. 150. Stemberger A, Grimm H, Bader F, Rahn HD, Ascherl R. Local treatment of bone and soft tissue infections with the collagen gentamicin sponge. Eur J Surg Suppl 1997(578):17 26. 151. Drognitz O, Thor n D, Kruger T, Gatermann SG, Iven H, Bruch HP, Muhl E. Release of vancomycin and teicoplanin from a plasticized and resorbable gelatin sponge: in vitro investigation of a new antibiotic delivery system with glycopeptides. Infection 2006;34(1):29 34. 152. C evher E, Orhan Z, Mulazimoglu L, Sensoy D, Alper M, Yildiz A, Ozsoy Y. Characterization of biodegradable chitosan microspheres containing vancomycin and treatment of experimental osteomyelitis caused by methicillin resistant Staphylococcus aureus with prep ared microspheres. Int J Pharm 2006;317(2):127 135. 153. Bohner M, Gbureck U, Barralet JE. Technological issues for the development of more efficient calcium phosphate bone cements: a critical assessment. Biomaterials 2005;26(33):6423 6429. 154. Stallmann HP, Faber C, Bronckers AL, Nieuw Amerongen AV, Wuisman PI. In vitro gentamicin release from commercially available calcium phosphate bone substitutes influence of carrier type on duration of the release profile. BMC Musculoskelet Disord 2006;7:18. 155. Ham anishi C, Kitamoto K, Tanaka S, Otsuka M, Doi Y, Kitahashi T. A self setting TTCP DCPD apatite cement for release of vancomycin. J Biomed Mater Res 1996;33(3):139 143. 156. Robison RL; Eli Lilly and Company, assignee. Vancomycin HCL solutions and the lyoph ilization thereof. United States patent 4885275. 1989.

PAGE 199

199 157. Stoor P, Soderling E, Salonen JI. Antibacterial effects of a bioactive glass paste on oral microorganisms. Acta Odontologica Scandinavica 1998;56(3):161 165. 158. Gupta MM, Saini TR. Preformulatio n parameters charaterization to design, development and formulation of vancomycin hydrochloride tablets for psudomembranous colitis (sic). International Journal of Pharmaceutical Research and Development 2009;1(9). 159. Godbole PP, Tsai R S, Clark WM. Affi nity partitioning of vancomycin. Biotechnol Bioeng 1991;38(5):535 544. 160. Lee C K, Sandler SI. Vancomycin partitioning in aqueous two phase systems: Effects of pH, salts, and an affinity ligand. Biotechnol Bioeng 1990;35(4):408 416. 161. Radin S, El Bass youni G, Vresilovic EJ, Schepers E, Ducheyne P. In vivo tissue response to resorbable silica xerogels as controlled release materials. Biomaterials 2005;26(9):1043 1052. 162. Claudius JS, Neau SH. The solution stability of vancomycin in the presence and ab sence of sodium carboxymethyl starch. Int J Pharm 1998;168(1):41 48. 163. Piola RF, Dafforn KA, Johnston EL. The influence of antifouling practices on marine invasions. Biofouling 2009;25(7):633 644. 164. Thomas KV, Brooks S. The environmental fate and eff ects of antifouling paint biocides. Biofouling 2010;26(1):73 88. 165. Lindgren JF, Haeffner M, Ericsson CT, Jonsson PR. Oxygen depleted surfaces: a new antifouling technology. Biofouling 2009;25(5):455 461. 166. McMaster D, Bennett S, Tang Y, Finlay JA, Ko walke GL, Nedved B, Bright FV, Callow ME, Callow JA, Wendt DE and others. Antifouling character of 'active' hybrid xerogel coatings with sequestered catalysts for the activation of hydrogen peroxide. Biofouling 2009;25(1):21 33. 167. Tasso M, Pettitt ME, Cordeiro AL, Callow ME, Callow JA, Werner C. Antifouling potential of Subtilisin A immobilized onto maleic anhydride copolymer thin films. Biofouling 2009;25(6):505 516. 168. Bressy C, Hellio C, Marchal JP, Tanguy B, Margaillan A. Bioassays and field imme rsion tests: a comparison of the antifouling activity of copper free poly(methacrylic) based coatings containing tertiary amines and ammonium salt groups. Biofouling 2010;26(7):769 777.

PAGE 200

200 169. Kristensen JB, Olsen SM, Laursen BS, Kragh KM, Poulsen CH, Besenb acher F, Meyer RL. Enzymatic generation of hydrogen peroxide shows promising antifouling effect. Biofouling 2010;26(2):141 153. 170. Sommer S, Ekin A, Webster DC, Stafslien SJ, Daniels J, Van der Wal LJ, Thompson SEM, Callow ME, Callow JA. A preliminary st udy on the properties and fouling release performance of siloxane polyurethane coatings prepared from poly(dimethylsiloxane) (PDMS) macromers. Biofouling 2010;26(8):961 972. 171. Scardino AJ, Hudleston D, Peng Z, Paul NA, de Nys R. Biomimetic characterisat ion of key surface parameters for the development of fouling resistant materials. Biofouling 2009;25(1):83 93. 172. Long CJ, Schumacher JF, Robinson PAC, Finlay JA, Callow ME, Callow JA, Brennan AB. A model that predicts the attachment behavior of Ulva linza zoospores on surface topography. Biofouling 2010;26(4):411 419. 173. Callow ME, Callow JA. Marine biofouling: A sticky problem. Biologist 2002;49:10 14. 174. Long CJ, Finlay JA, Callow ME, Callow JA, Brennan AB. Engineered antifouling microtopographi es: mapping preferential and inhibitory microenvironments for zoospore attachment. Biofouling 2010;26(8):941 952. 175. Long CJ. Analytical methods for predicting biosettlement on patterned surfaces [Dissertation]. Gainesville (FL): University of Florida; 2 009. 134 p. 176. Dabros T, Van De Ven TGM. Kinetics of coating by colloidal particles. J Colloid Interf Sci 1982;89(1):232 244. 177. Cowan MM, Taylor KG, Doyle RJ. Kinetic analysis of Streptococcus sanguis adhesion to artificial pellicle. J Dent Res 1986;6 5(10):1278 1283. 178. Pratt Terpstra IH, Weerkamp AH, Busscher HJ. Microbial factors in a thermodynamic approach of oral streptococcal adhesion to solid substrata. Journal of Colloid Interf Sci 1988;129(2):568 574. 179. Sjollema J, Busscher HJ, Weerkamp AH Experimental approaches to studying adhesion of microorganisms to solid substrata: applications and mass transport. Journal of Microbiol Meth 1989;9:79 90. 180. Busscher HJ, Sjollema J, van der Mei HC. Relative importance of surface free energy as a meas ure of hydrophobicity in bacterial adhesion to solid surfaces. In: Doyle RJ, Rosenberg M, editors. Microbial cell surface hydrophobicity. Washington (DC): American Society for Microbiology; 1990. p 335 359.

PAGE 201

201 181. Van der Mei HC, Cox SD, Geertsema Doornbusch GI, Doyle RJ, Busscher HJ. A critical appraisal of positive cooperativity in oral streptococcal adhesion: scatchard analyses of adhesion data versus analyses of the spatial arrangement of adhering bacteria. J Gen Microbiol 1993;139(5):937 948. 182. Callow JA, Callow ME, Ista LK, Lopez G, Chaudhury MK. The influence of surface energy on the wetting behaviour of the spore adhesive of the marine alga Ulva linza (synonym Enteromorpha linza ). J R Soc Interface 2005;2(4):319 325. 183. Heydt M, Rosenhahn A, Grunz e M, Pettitt M, Callow JA, Callow ME. Digital in line holography as a three dimensional tool to study motile marine organisms during their exploration of surfaces. J Adhes 2007;83:417 430. 184. Krishnamurti K, Soman S. Studies in the adsorption of bacteria P Indian Acad Sci B 1951;34(2):81 91. 185. Sjollema J, van der Mei HC, Uyen HM, Busscher HJ. Direct observations of cooperative effects in oral streptococcal adhesion to glass by analysis of the spatial arrangement of adhering bacteria. FEMS Microbiol Le tt 1990;69:263 270. 186. Lin L, Rosenberg M, Taylor KG, Doyle RJ. Kinetic analysis of ammonium sulfate dependent aggregation of bacteria. Colloid Surface B 1995;5(3 4):127 134. 187. Xynos ID, Edgar AJ, Buttery LDK, Hench LL, Polak JM. Ionic products of bio active glass dissolution increase proliferation of human osteoblasts and induce insulin like growth factor II mRNA expression and protein synthesis. Biochem Biophys Res Commun 2000;276(2):461 465. 188. Wang Y, Finlay JA, Betts DE, Merkel TJ, Luft JC, Callo w ME, Callow JA, DeSimone JM. Amphiphilic co networks with moisture induced surface segregation for high performance nonfouling coatings. Langmuir 2011:10365 10369. 189. Wang Y, Betts DE, Finlay JA, Brewer L, Callow ME, Callow JA, Wendt DE, DeSimone JM. Ph otocurable amphiphilic perfluoropolyether/poly(ethylene glycol) networks for fouling release coatings. Macromolecules 2011;44(4):878 885. 190. Bogush GH, Tracy MA, Zukoski CF. Preparation of monodisperse silica particles control of size and mass fraction J Non Cryst Solids 1988;104(1):95 106. 191. Giesche H. Synthesis of monodispersed silica powders I. Particle properties and reaction kinetics. J Eur Ceram Soc 1994;14(3):189 204. 192. Fetters LJ, Lohse DJ, Richter D, Witten TA, Zirkel A. Connection betwe en polymer molecular weight, density, chain dimensions, and melt viscoelastic properties. Macromolecules 2002;27(17):4639 4647.

PAGE 202

202 193. Shenoy SL, Bates WD, Frisch HL, Wnek GE. Role of chain entanglements on fiber formation during electrospinning of polymer s olutions: good solvent, non specific polymer polymer interaction limit. Polymer 2005;46(10):3372 3384. 194. Stber W, Fink A, Bohn E. Controlled growth of monodisperse silica spheres in the micron size range. J Colloid Interface Sci 1968;26:62 69.

PAGE 203

203 BIOGRAPHICAL SKETCH Scott Patrick Cooper was born in Scottsdale, Arizona in 1985 to Pat and Larry Cooper. He grew up in Arizona, Arkansas, Tennessee, and California but calls Phoenix, Arizona home. As a child Scott enjoyed many activities including: outdoor adventures with his brother, playing golf with his dad, and going on family road trips to North Dakota in the summer His interest in materials science began when he took a ceramic art class in high school. He later worked for a glass artist prod ucing pieces of fused, dichroic glass. He graduated from Desert Vista High School in Phoenix in 2003 Scott attended the University of Arizona in Tucson, where he studied m aterials s cie nce and e ngineering and German s tudies. Scott traveled during summer breaks, studying at the University of Leipzig, Germany in 2004 and performing undergraduate research at North Dakota State University in 2005. Scott was selected as one of 25 students to take part in Transatlantic Program in 2 006 a German business immersion program A s a part of this program h e worked at the Schott glass company in Marienborn, Germany where he characterized the fluorescent properties of glasses and glass ceramics F or his senior design project Scott worked on a team with 4 other students to design and build a glassblowing furnace. Scott graduated summa cum laude w m aterials s cience and e ngineering and German s tudies. He was named an nities Scott received a National Science Foundation Graduate Research Fellowship to attend the University of Florida. He joined the Brennan Research Group in the summer of 2007. Scott married his high school sweetheart in June of 2008. In January 2010, Scott was selected to attend the US China Winter School on Glass at Zhejiang University in Hangzhou, China. The trip was a learning experience both academically

PAGE 204

204 and culturally. In July 2010, Scott presented his work in Newcastle, England at the Internat ional Congress on Marine Corrosion and Fouling, where he received an award for best poster. While at the University of Florida, Scott was actively involved in the Society for Biomaterials (SFB) He served as treasurer for the UF chapter from 2007 2008 and president from 2008 2009. He is currently serving a term as the president of the national student chapter of SFB. Scott has also written several summaries of research articles for the MRS Bulletin In his free time, Scott enjoys playing golf, homebrewi ng beer, and traveling with his wife.