The Effect of Engineered Topographies on Marine Adhesive

MISSING IMAGE

Material Information

Title:
The Effect of Engineered Topographies on Marine Adhesive
Physical Description:
1 online resource (119 p.)
Language:
english
Creator:
Castro Cara, Francisco
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Materials Science and Engineering
Committee Chair:
BRENNAN,ANTHONY B
Committee Co-Chair:
BATICH,CHRISTOPHER D
Committee Members:
ALLEN,JOSEPHINE

Subjects

Subjects / Keywords:
antifouling -- attachment -- biofouling -- micro-topography -- mussel
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre:
Materials Science and Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
Mussels have the ability to attach to a wide variety of materials by the production of a structure known as byssus. This byssus is formed by a number of adhesive plaques each connected to the inside of the animal through a collagen thread. The byssal plaque is a highly cross-linked protein adhesive that is segregated by the foot, a specialized appendix organ capable to identify chemical and physical features on the surface in order to identify a suitable attachment spot. This experiment is intended to investigate the effect of the micro-engineered pattern size in mussel attachment process.
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 Francisco Castro Cara.
Thesis:
Thesis (M.S.)--University of Florida, 2014.
Local:
Adviser: BRENNAN,ANTHONY B.
Local:
Co-adviser: BATICH,CHRISTOPHER D.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2014
System ID:
UFE0046437:00001


This item is only available as the following downloads:


Full Text

PAGE 1

THE EFFECT OF ENGINEERED TOPOGRAPHIES ON MARINE ADHESIVE By FRANCISCO CASTRO CARA A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2014

PAGE 2

2014 F rancisco Castro Cara

PAGE 3

To my beloved parents and my amazing wife

PAGE 4

4 ACKNOWLEDGMENTS I can say with certainty that I could not have achieved half of what I have done without the encouragement and help of a select group of individuals. First and foremost, I must thank my advisor and committee ch airman Dr. Anthony Brennan for his encouragement and support during my studies at University of Florida. I also want to mention the other two members of my committee, Dr. Josephine Allen and Dr. Christopher Batich, who have offered her knowledge and expert ise when I have needed. I want to make a special mention to all my fellow group members, who provided me with a priceless experience, personal and professional, and who shortly after I joined the group became friends. This thesis would have never been wri tten if I did not count with all their help and patience with me. My deepest gratitude and appreciation go to the group members who have seen me through this entire process: Dr. Canan Kizilkaya, Cary Kuliasha, Clayton Argenbright, Danessa Jerome, Joe Decke r Laura Villada and Shanna Smith. It is very important for me the assistance provided in my investigation by Dr. Shane Stafslien (University of North Dakota), who helped me to improve my procedure and Dr. Daniel Rittschof (Duke University) who took part of his time to collect and deliver the mussels used in this research project. My biggest appreciation to my family and friends, who have offered emotional support and love during this entire process. Especially my parents, Francisco and Encarnacion who d id not doubt to trust completely in my skills to provide me with this opportunity, and my wife, Andrea, the one that has always been strong at my side especially in those moments I was more vulnerable.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ........................... 11 ABSTRACT CHAPTER 1 INTRODUCTION ................................ ................................ ................................ ....... 13 2 BACKGROUND ................................ ................................ ................................ ......... 16 Marine biofouling and marine adhesives ................................ ................................ 16 Mussel adhesion ................................ ................................ ................................ ..... 18 Byssus: water adhesive interaction and mechanism of attachment ................. 19 Byssus: chemistry ................................ ................................ ............................ 21 Polydimethylsiloxane elastomer (PDMSe) ................................ .............................. 23 Mechanical and chemical properties ................................ ................................ 24 Surface energy ................................ ................................ ................................ 25 Wettability, surface energy and topography ................................ ............................ 26 Contact angle: Young equation ................................ ................................ ........ 27 Wettability: Wenzel and Cassie Baxter models ................................ ................ 28 Engineered Roughness Index (ERI) ................................ ................................ 29 3 MATERIALS ................................ ................................ ................................ .............. 36 Setting up the tank: cultivating mussels ................................ ................................ .. 36 Testing kits ................................ ................................ ................................ .............. 36 ................................ ............................... 37 Silicone elastomer: SILASTIC T 2 ................................ ................................ .......... 37 Allyltrimethoxysilane (ATS): silane coupling agent ................................ ................. 38 Hexamethyldisiloxane (HMDS) ................................ ................................ ............... 38 4 METHODS ................................ ................................ ................................ ................ 41 Cultivating mussels ................................ ................................ ................................ 41 Setting up the aquarium. Safety and previous considerations .......................... 41 Conditions in the tank ................................ ................................ ....................... 43 Mussel aquarium maintenance protocol ................................ ........................... 43 PDMSe patterned samples preparation ................................ ................................ .. 45

PAGE 6

6 PDMSe patterned samples characterization ................................ ........................... 47 Mussel plaque production ................................ ................................ ....................... 48 Extraction procedure ................................ ................................ ............................... 50 Measuring number of plaques ................................ ................................ ................ 50 Measuring surface area and Nuclear Form Factor (NFF) ................................ ....... 51 Degree of penetration (Scanning Electron Microscope) ................................ .......... 52 Tensile test (texture analyzer) ................................ ................................ ................. 52 5 RESULTS AND DISCUSSION ................................ ................................ .................. 63 Number of plaques ................................ ................................ ................................ 63 Plaque sur face area ................................ ................................ ................................ 64 Nuclear Form Factor (NFF): Morphology of the plaques ................................ ......... 66 Degree of penetration ................................ ................................ ............................. 67 Textur e analyzer ................................ ................................ ................................ ..... 68 Tensile force ................................ ................................ ................................ ..... 72 Tensile force versus plaque surface area ................................ ......................... 73 Tensile stress ................................ ................................ ................................ ... 74 6 CONCLUSIONS AND FUTURE WORK ................................ ................................ .. 107 Conclusions ................................ ................................ ................................ .......... 107 Future work ................................ ................................ ................................ ........... 109 Comparison with commercial adhesives ................................ ........................ 109 Test other engineered topographies ................................ ............................... 109 Measure the penetration degree ................................ ................................ .... 110 Measure the exact volume of adhesive secreted by the mussel .................... 110 Analyze the surface where adhesive failure happens. ................................ ... 110 LIST OF REFERENCES ................................ ................................ ............................. 113 BIOGRAPH ICAL SKETCH ................................ ................................ .......................... 118

PAGE 7

7 LIST OF TABL ES Table page 2 1 DOPA containing proteins o f the byssus of Mytilus species .............................. 32 3 1 Typical properties of Silastic T2 (Dow Corning), from Dow Corning product information sheets ................................ ................................ .............................. 39 4 1 Water tank maintenance monthly schedule ................................ ........................ 55 5 1 Different sizes of the features in the different patterns ................................ ....... 77 5 2 Table with the more significant statistical values of number of plaques for the different samples (data collected from the texture analyzer tests) ...................... 79 5 3 Plaque surface area statistically most important parameters ............................. 81 5 4 ANOVA variance study for the plaque surface area values on the different topographies ................................ ................................ ................................ ....... 82 5 5 Tukey post hoc analysis of plaque surface area values (being 1=smooth, 2=5x5, 3=20x20, 4=50x50, 5=100x100, 6=200x200) ................................ ......... 83 5 6 Nuclear Form Factor statistically most important parameters ............................. 85 5 7 ANOVA variance study for the Nuclear Form Factor (NFF) values on each topography ................................ ................................ ................................ .......... 86 5 8 Tukey post hoc analysis of nuclear form factor values (being 1=smooth, 2=5x5, 3=20x20, 4=50x50, 5=100x100, 6=200x200) ................................ ......... 87 5 9 Values of tensile force (mN) and Tensile stress (mN/mm 2 ) including basic statistical analysis for every sample tested in the texture analyzer ................... 101 5 1 0 ANOVA variance study for tensile force values for each topography ............... 102 5 1 1 Tukey post hoc analysis of tensile force values (being 1=smooth, 2=5x5, 3=20x20, 4=50x50, 5=100x100, 6=200x200) ................................ ................... 103 5 1 2 Comparison with ANOVA statistical analysis (Minitab) between plaque surface area values of the 48 h and 72 h experiments ................................ ..... 105

PAGE 8

8 LIST OF FIGURES Figure page 2 1 Progress of movements during mussel plaque deposition. ................................ 31 2 2 Scheme of the distribution of the different mussel foot protein along the byssus structure. ................................ ................................ ................................ 32 2 3 Chemical structure of L 3,4 Dihydroxyphenilalanine ................................ .......... 33 2 4 Adhesive and cohesive interactions of mussel protein through DOPA ............... 33 2 5 Basic structure of a PDMS chain segment ................................ ......................... 34 2 6 Diagram of forces on molecules in a liquid.. ................................ ....................... 34 2 7 Schematic diagram of liquid droplet on solid surface. ................................ ......... 35 2 8 Schematics representation of Wenz el model (left) and Cassie Baxter (right).. ... 35 3 1 Curing reaction between the base resin and the curing agent of SILASTIC silicone ................................ ................................ ................................ ............... 39 3 2 ATS coupling agent reaction with silica glass ................................ ..................... 40 3 3 HMDS reaction onto silica glass slide. ................................ ................................ 40 4 1 Mold setup for preparation of patterned PDMSe on glass slides (sectional view) ................................ ................................ ................................ ................... 56 4 2 SEM captures the different topographi es used in this study (3.67 kV) ............... 57 4 3 General plaque production experiment set and the direction corresponding to the position of the mussel ................................ ................................ ................... 58 4 4 SEM micrograph of the edge of a plaque onto 10 CH 20x20 ............................ 59 4 5 Optical microscope picture of mussel plaques onto 20 CH20x20 at x10 magnification ................................ ................................ ................................ ...... 60 4 6 Optical microscope image of two mussel plaques onto 10 CH 20x20. On the left without image processing. ................................ ................................ ............ 60 4 7 Schematic of the measurements taken by ImageJ to calculate NFF ............... 61 4 8 Tensile analyzer (XT plus Texture analyzer from Texture Technologies Corp.) used in tensile force measurements ................................ ........................ 61

PAGE 9

9 4 9 Texture analyzer set up. Mussel is attached to the sample by different byssus structures. ................................ ................................ ............................... 62 4 10 Example of force versus distance plot produced from data collected by texture analyzer for an individual mussel in a plaque strength test. .................... 62 5 1 Pictures that illustrate the mussel plaque production extends beyond the patterned area. ................................ ................................ ................................ ... 78 5 2 Number of plaques in different patterns (texture analyzer test) compared with control sample (smooth PDMSe).. ................................ ................................ ...... 80 5 3 Plaque surface area values (mm2) of the different patterns tested. ................... 82 5 4 Plaque surface area values (mm 2 ) vs. size of the features ( m 2 ).. ..................... 84 5 5 NFF values of the different patterns tested. ................................ ........................ 86 5 6 NFF values vs. size of the features. ................................ ................................ ... 88 5 7 SEM pictures of detached mussel plaques on 20x20 microns channels feature ................................ ................................ ................................ ................ 89 5 8 SEM pictures of attached mu ssel plaques on 20x20 microns channels feature (A) and 50x50 microns feature (B) ................................ ......................... 90 5 9 SEM pictures of attached mussel plaques on 50x50 microns channels feature (A) and 20x20 microns feature (B) ................................ ......................... 91 5 10 SEM pictures of attached mussel plaques on 50x50 mic rons feature ................ 92 5 11 Tensile test performed in texture analyzer. A number of threads can be observed going from the shell of the mussel to the substrate.. ........................... 93 5 12 Results for the texture analyzer tensile test on smooth PDMSe.. ....................... 93 5 13 Results for the texture analyzer tensile test on channels 5x5 microns features.. ................................ ................................ ................................ ............ 94 5 14 Results for the texture analyzer tensile test on channels 20x20 microns features.. ................................ ................................ ................................ ............ 94 5 15 Results for the texture analyzer tensile test on channels 20x20 microns features.. ................................ ................................ ................................ ............ 95 5 16 Results for the texture analyzer tensile test on channels 100x100 microns features.. ................................ ................................ ................................ ............ 95

PAGE 10

10 5 17 Results for the texture analyzer tensile test on channels 200x200 microns features.. ................................ ................................ ................................ ............ 96 5 18 Histogram for the tensile force (mN) values of the different plaques for those tests on smooth PDMSe ................................ ................................ ..................... 96 5 19 Histogram for the tensile force (mN) values of the different plaques for those tests on 5x5 microns channels samples ................................ ............................. 97 5 20 Histogram for the tensile force (mN) values of the different plaques for those tests on 20x20 microns channels samples ................................ ......................... 97 5 21 Histogram for the tensile force (mN) values of the different plaques for those tests on 50x50 microns channels samples ................................ ......................... 98 5 22 Histogram for the tensile force (mN) values of the different plaques for those tests on 100x100 microns channels samples ................................ ..................... 98 5 23 Histogram for the tensile force (mN) values of the different plaques for those tests on 200x200 microns channels samples ................................ ..................... 99 5 24 Adhesive failures of mussel plaques in the different topographies. .................. 100 5 25 Tensile force (mN) for every sample (5x5, 20x20, 50x50, 100x100, 200x200, width x spacing in microns) and blank.. ................................ ............................ 1 02 5 26 Comparison between data collected for tensile force (left axis) and plaque surface area (right axis).. ................................ ................................ .................. 104 5 27 Tensile stress (x10 3 Pa) for every sample (5x5, 20x20, 50x50, 100x100, 200x200, width x spacing in microns) and blank applying roughness index .. ... 106 5 28 Comparison between data collected for tensile force (right axis) and tensile stress (left axis).. ................................ ................................ .............................. 106 6 1 Scheme of the effect of the topography size on mussel plaque adhesion (adhesive attachment) ................................ ................................ ...................... 112

PAGE 11

11 LIST OF ABBREVIATIONS 5x5 Channel pattern sample 5 micron by 5 micron (width by spacing) CH Channel pattern Mfps Mussel foot protein(s) NFF Nuclear Form Factor PDMSe Polydi methylsiloxane elastomer is a rubber (viscoelastic behavior), with a highly cross linked structure based on polydimethylsiloxane chains. SEM Scanning Electron Microscope SK Sharklet pattern

PAGE 12

12 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requir ements for the Master of Science THE EFFECT OF ENGINEERED TOPOGRAPHIES ON MARINE ADHESIVE By Francisco Castro Cara May 2014 Chair: Anthony Brennan Major: Materials Science and Engineering (MTL) Mussels have the ability to attach to a wide variety of materials by the production of a structure known as byssus. This byssus is formed by a number of adhesive plaques each connected to the inside of the animal through a collagen thread. The byssal plaqu e is a highly cross linked protein adhesive that is segregated by the foot, a specialized appendix organ capable to identify che mical and physical features on the surface in order to identify a suitable attachment spot. This experiment is intended to inve s tigate the effect of the micro engineered pattern size in mussel attachment process.

PAGE 13

13 CHAPTER 1 INTRODUCTION Biofouling is the formation of an undesired film generally on man made structures, produced by organic products and organisms present in the environment. The film may vary i n thickness and it exists a wide range of organisms (Schumacher et al. 2007) that take part in the fouling process from algal zoospores (Ulva linza ) to macroscopic species (adult cyprids), that differ from each other in size, type and mechanism of action Biofouling has a major economic and e nvi ronmental effect. It increases the fue l consumption rate of ships about 10% plus all the expenses of maintenance (paint, hull cleaning) and other extra cost associated with these treatments (Schultz et al. 2010). Power plants and other industry facilities where water intake from natural sources are used have remarkable losses too. Macrofouling affects the plant operation and safety as it may produce shutdowns (Wells et al. 2009). At the same time, biofouling has a critical impact in the environment. Adult a nd juvenile mussels specifically, attached to the hull of ships or inside the ballast water of boats, may manage to travel from far areas and invade different wetlands and coast waters such as zebra mussel in the Great Lakes region. Mussels attach successfully to a variety of substrates in an aqueous environment by using a strong protein based adhesive, including PVC, PMMA, Aluminum, Stainless Steel, wood and concrete (Ackerman et al. 1996; Ackerman et al. 1992). Mussels use a specialized organ kno select an appropriate site to produce the byssus, an anchor like structure that fix the organism to the surface and make it possible for the animal to survive in the intertidal

PAGE 14

14 zone where they live (Silverman and Roberto 2007). The byssus consist of an organ inside the shell, called stem, a collagen thread t hat may vary on length and a plaque, that is directly in contact with the surface. Mussels present two different stages in their life cycle in which they settle on the surface, known as primary settlement and secondary settlement (Bayne 1964 ). The first settlement process take place in an early stage in the development of the mussel, also called as pediveliger (170 300 m), and the second set tlement occurs when the mussel is transforming into the adult form also called as plantigrades/ juvenile (Ackerman et al. 1994). The animal preferences about attachment sites in each stage vary with complexity being a key factor. Mussels prefer more comple x structures in early stages of development for protection and less complex later to promote growth and more availability of food source (Carl et al. 2012c). Different coatings to prevent from the ad hesion of mussels have been tested Considerations about toxicity, resistance and long t erm use have come to be critical. Legal l evels of copper and zinc leaking into the environment are lower and lower each year, and toxicity becomes a very important issue. Foul ing release materials may not be appropriate for a high flow condition s like those in intake pipes for big facilities because they may be peeled off the treated surface, and secondary toxic effects have been reported. Most of the coatings require substitution after an average of 5 7 years (Wells et al. 2 009 ). Surface patterning of polymer systems has shown a very good performance in antifouling. Polymer surface properties such as surface energy can be altered by modifying the size, shape and type of material used (Carman et al. 2006 ).

PAGE 15

15 Microchannels and Sh arklet, have already been tested and proved valid for different organisms such as shellfish (Scardino et al. 2008) and algae ( Hoipkemeier Wilson et al. 2004), mussel pediveligers (Carl et al. 2012a) and barnacle cyprids (Aldred et al. 2009). It has also been shown that the performance of a specific topography is highly specie specific, due to the different properties, sizes and mechanism of adhesion of the wide range o f organisms that are part of biofouling (Schumacher et al. 2007 ). Some resea rch has been done about the effect of surface roughness and engineered topographies on the settlement and attachment of mussels and other macrofoulers such as barnacles, but all of them have used organisms in an early stage of development, like mussel pedi veliger (Carl et al. 2012a) and barnacle cyprids (Aldred et al. 2010). More studies have been done in adult mussels attachment mechanism and strength, but not taking into account topography, for example enhancing the efficacy of fouling release coatings us ing nanofillers (Carl et al. 2012b) or measuring the attachment strength onto different surfaces (Burkett et al. 2009). The purpose of this study was to test the effect of different engineered topographies on the process of attachment of adult mussels (sp ecifically Geukensia demissa). Surface topography size was introduced as a key factor for mechanical adhesion strength of the byssus, spreading behavior of the plaque adhesive onto the surface, and size and shape of the plaque. The substrate selected was s ilicone elastomer for its well known foul release properties and the good results in previous experimentations.

PAGE 16

16 CHAPTER 2 BACKGROUND Marine biofouling and marine adhesives Controlling biofouling in a marine environment represents a very difficult cha llenge. Water provides an ideal living medium for a wide range of species, some of them needing of a substrate to attach in order to grow and develop (i.e. algae, barnacles or tubeworms). Each of these organisms possess different mechanisms of attachment, as well as they use different adhesives, what makes bi ofouling phenomena very complex A good review of current commercially available products used nowadays can be found in Wells and Sytsma (2009). Chemical and physical properties of water have played a v ery important role on compounds that can bind to a surface. In adhesive bonding industry, water presence leads to a bad performance of the adhesive through some mechanism such as moisture induced plasticization, swelling, erosion, and hydrolysis of polymers and interfacial wicking and crazing (Wake and Kinloch 1977). Water decrease largely the electrostatic interaction between opposite charges as a consequence of the hig h diele ctric constant, and it was also further decreased as an effect for the strong solvation of ions (i.e. Mg 2+ or Li + ). When talking of surfaces interaction, the term adhesion energy and adhesion work must be taken into account, since the dispersion and polar interactions (as surface energy terms) in water produce the value of the adhesion work to be a lot less than it is expected in non wet conditions. This means that the strength of the interaction between the surface and the adhesive is much lower in water i f it exists (Lee et al. 2006).

PAGE 17

17 An adhesive in underwater conditions, in order to be effective needs to work quickly. First the adhesive needs to be fluid to spread onto the surface and second, at the same time it needs to have a surface tension to preven t from being dissolved by the water. These two conditions seem to be a bit of a contradiction, so a balance between spreading properties and surface tension is an important requirement. Adhesive must be able to displace water molecules that are already bon ded (electrostatically) to the surface, so the attachment process competes directly with water molecules. Finally, the adhesive has to undergo a crosslinking process, in order to acquire the stiffness and cohesion enough to endure (Smith and Callow 2006). The zoospore of the Ulva algae, has been extensively used as a model fouling agent in a lot of research projects. An important stage in the reproduction and spreading of this alga is the production of motile zoospores that need to attach to a surface to ge t to the next stage of development. The zoospore has the ability to actively identify a suitable site on the surface and the attachment is performed by the secretion of an adhesive which anchors the spore to the surface. This adhesive is stored inside vesi cles within the membrane and the composition has not been completely characterized yet (Callow and Callow 2006). Diatoms are a group of unicellular microalgae whose common feature is the production of a silicified cell wall after settlement. Diatoms do not actively search to attach to surfaces but they are moved by the water currents. The adhesive used by this organism has been described as mucilaginous material, as well known as extracellular polymeric substances, which are completely non soluble compounds (Chiovitti et al 2006).

PAGE 18

18 Barnacle s are a well known fouling agent. Barnacles are a type of macrofouler that can be described as a shellfish that attach to surfaces for protection, increase the access to resources and improve reproduction rates, since they usually are found in colonies. During barnacle reproduction process they produce larva, as well known as cyprids. These cyprids possess antennules that are used to get information from the substrate and identify a convenient site for attachment. The b indin g process happens by the secretion of a cement, which composition varies from earlier stages in the life of the barnacle to the adult and last stage, but can be described as a multi protein complex (over 90% of content in protein). The chemical analysis of the barnacle cement in any case has revealed DOPA, as a big difference with mussel adhesive. (Kamino 2006). A big effort in marine bio fouling prevention has been done in the direction of mimicking natural surfaces that are not fouled in natural conditio ns, so some biomimetic models and bioinspired surfaces have shown encouraging results (Scardino and de Nys 2011). Some of them worth to be mention are gorgonian corals ( Vrolijk et al. 1990 ), pilot whale skin ( Baum et al. 2001 ), shark skin ( Brennan et al. 2010) and molluscs ( Scardino et al. 2009 ) Mussel adhesion Mussels life cycle go through different stages, 1) pelagic larval forms, 2) primary settlement and metamorphosis, 3) post metamorphic settlement (secondary settlement), where the animal differ not only in size but in preferences for attachment (Ackerman et al 1994). During the primary settlement, the larvae actively search f or those sites that provide increased surface area and protection, being normally site s with a higher complexity. However the p references change when the animal develops into a more mature form, when space availability and a more abundant food source become more

PAGE 19

19 important. For this reason, during maturation the second settlement takes place on different sites, producing the mussel to move to a more open area (Carl et al. 2012c; Bayne 1964). Settlement always takes place through the same process at all stages of development. This process starts with the mussel foot, a retractile tongue like organ that surface previous to the attachment. The mussel foot is able to detect differences in chemistry and morphology onto the surface, what permits the animal to detect the more favora ble sites. Different studies have shown attachment preferences over different m aterial, being those materials with a higher surface energy more likely to be used by the animal (Ackerman et al. 1996; Ackerman et al. 1992; Aldred et al. 2006). It has been proved that surface topography play an important role in the process of site sele ction by the mussel, being preferred those topographies that protect the larvae in early stages (Carl et al. 2012c). Once the site is located, the mussel foot proceed s to secrete the byssus structure, a protein complex that anchors the mussel to the surfac e. This structure is connected by the stem inside the shell, and is bonded through the plaque on the surface. The process is shown with pictures of the most relevant moments during plaque production in Figur e 2 1 Plaque and stem are connected by a high collagen content thread. This structure provid es the mussel with the ability to survive the wave breaking and intertidal zone where they inhabit. Byssus: water adhesive interaction and mechanism of attachment Water and adhesives are in conflict This is a major challenge when designing a good adhesive to perform in underwater conditions or in moist areas. The water effects adhesion in 4 different ways: 1) f orming a strongly/weakly bonded layer at interfaces, 2)

PAGE 20

20 wicking or crazing into interfaces, 3) hydrolysis and erosion, 4) swelling or plasticization of adhesive by water absorption, or a combination of two or more of these effects (Wake and Kinloch 1977). Mussels need to overcome all the issues addressed above, 1) removing weak boundary layer s 2) efficient spreading of the adhesive onto the surface, 3) developing an appropriate chemistry to produce high energy and durable contacts and 4) crosslinking the p rotein structure to increase the tenacity of the byssus (Waite 1987). Water is squeezed out of the foot surface interface by physical pushing. Once the water layer has been removed, the foot is able to produce a depression of vacuum by cavitation, where th e adhesive precursor is injected in the absence of water, for spreading and later, crosslink (Waite 1987). There is a continuous and direct contact at the plaque substrate interface that presents a uniform, electron dense layer (around 10 20 nm thick), whi ch is completely created by the mussel (no biofilm deposited previously). This layer has been identified with the adhesive component of the plaque (Farsad and Sone 2012). It is controversial the correlation between plaque surface area and surface energy. W aite 1987 stated that in non polar surfaces (low surface energy), larger (surface area) plaques are produced with lower attachment strength. Smaller surfaces and stronger attachment ha ve been observed in polar (high surface energy) substrates. On non polar surfaces water interacts more weakly. Since the polyphenolic proteins are much larger, they offer a more dispersive interaction to the surface such that the water molecules are displaced and a high degree of spreading occurs. On polar substrates water mo lecules are more strongly attached to the surface so the spreading does not

PAGE 21

21 happen spontaneously, causing the mussel adhesive to create a front where high energy interactions take place (Waite 1987). On the other hand, a more recent study (Aldred et al. 20 05) with Self Assembled Monolayers (SAM) has been proved the opposite effect, that is, mussels produce larger plaques in high surface energy substrates. Other different study establish no relation at all between these two parameters (Burcket et al. 2009). When the process of attachment is finished, a byssus structure anchors the mussel onto the surface. Experimen ts with Mytilus Edulis demonstrate that an average of 10 20 plaques wa s produced after 72 hours (Burkett et al. 2009). T he values of the maximum lo ad were highly dependent o n the substrate (Ackerman et al. 1996 and Burkett et al. 2009). Generally when the byssus structure undergoes loading the failure appears at the interface adhesive substrate, rather than in the collagen thread or in the plaque th read connection (Burkett et al. 2009). A different approach to measure adhesion of the mussel foot proteins ( Mfps ) of the mussel has been developed with a surface forc e apparatus (SFA) (Lu et al. 2013). Byssus: chemistry The byssus structure has been characterized as a multi protein complex, the thread being mainly formed by collagen and the plaque formed by 6 different proteins, known generally as mussel foot proteins (Mfp). The composition of the byssus may vary between species of mussels while cert ain species have a byssus chemistry characterization that is a more challenging issue than others as is the case of the Dreissena Polymorpha. This is because immediately after (even during) the mussel secretes the adhesive it undergoes a curing process, i n other words, the proteins crosslink with each other to create a crosslinking network. This fact makes it very

PAGE 22

22 difficult to isolate the different proteins and its characterization (Gantayet et al. 2013; Gilbert and Sone 2010). Nevertheless, the whole sequ ence of amino acids has been obtained for the proteins that form the byssus structure in the mussel genre Mytilus. In Table 2 1 the different composition and sequences for each protein are shown (Lin et al. 2007). Something that makes the mussel adhesive so special is the content in the protein code of a rare component called L 3, 4 dihydroxyphenylalanine which is well known as DOPA ( Figure 2 3. ) DOPA is not only s uggested to mediate adhesion to the surface, but is a cross linking agent contributing to the cohesive streng th of the adhesive plaque (Wiegemann 2005). This component is found in different molar percentages in each of the Mfps, being an important characterizing factor. DOPA plays an important role in surface attachment. Table 2 1 shows how the highest content of DOPA is present in Mfp 3 and Mfp 5, so these two proteins need to be in a higher degree forming the plaque substratum interface ( Figure 2 2 ) The DOPA content of each protein and th e degree of crosslinking can vary according to several factors including the chemistry of the substrate. The DOPA can be modified to readjust different substrate chemistries. The time of the year is also an important factor (Matos Perez et al. 2012). Important factors when defining the adhesion strength of the plaque are the pH and the redox condition. It has been shown that lower values of pH improve adhesion, mainly because the oxidation of DOPA is less likely to happen (Nicklisch and Waite 2012). It is also worth mentioning the importance of the Mfp 6 in modulating the redox process. Mfp 6 present s low DOPA content and high cysteine content. Most of the thiol

PAGE 23

23 groups in cysteine are not coupled leaving a great number of them reduced as cysteine decomp oses Experiments have proved that modification of the pH is not enough to prevent oxidation of DOPA, so the antioxidant effect of Mfp 6 is an important factor (thiol groups produce the reduction of dopaquinone form from Mfp 3 and Mfp 5 in a stoichiometric fashion) (Yu et al. 2011). The spatial distribution of the proteins along the byssus structure is not well known due to the extensive crosslinking that prevents from an exact characterization. Threads and plaques present very similar amino acid content s b ut the composition is different The composition of the bulk plaque as well as the plaque substrate interface also differs in composition. Variation between fresh and aged mussel interfacial composition prove that the maturation and crosslinking process of the byssus does not stop directly following the initial curing process, but some of these proteins keep curing through a secondary period of time (Gilbert and Sone, 2010). The polyphenolic content of the mussel proteins interact with the substrate through 4 different energetic interactions, 1) h ydrogen bonds, 2) metal ligand complexes, 3) c ova lent cros s links which are initiated by a free radical mechanism between two quinones and 4) Michael type addition compounds derived from a o quinones (Wiegemann, 200 5). Hydrogen bonds and metal ligands are shown in Figure 2 4 It is also notable that a very unique iron induced mechanism may play a major role during crosslinking in the curing process (Wilker, 2010) Polydi methylsiloxane elastomer (PDMSe) Over the past decade silicon has become a very popular material in develop ing devices for biological adhesion testing, microfluidic devices and a variety of experiment

PAGE 24

24 set ups related to organic and biological systems interaction (Gibson 2002, Holm et al. 2006). Po lydimethylsiloxane (PDMS) is a semicrystalline polymer, with a glass transition temperature of 120C and a melting point of 40C. The basic structure is formed by silica, oxygen and hydrogen atoms. Silica atoms are bond through oxygen within the backbone The low electronegativity of silica (1.8) means a very polarized Si O bond forming a polymer chain with strong bonding energy (Si O 443 kJ/mol). This polarization however does not lead to strong intermolecular forces between the chains, since the interac tions between methyl groups are very weak and shield the interaction between polar groups within the molecules. PDMS also exhibits a low bond rotation energy (3.3 kJ/mol, nearly free rotation) being the nominal bond angle 145 degrees (tetrahedral), but var iable from 100 to 170 degrees without an important energy penalty. This effect makes the PDMS polymer chains very flexible and able to adapt many different configurations (Ltters et al 1997). The basic structure of a PDMS chain segment is represented in Figure 2 5 . PDMS elastomer (PDMSe) is a rubber (viscoelastic behavior), with a highly cross linked structure based on polydimethylsiloxane chains and approximately 10% fumed silica filler It is commerc ially available in different forms ( oil grease rubber resin or caulk). Mechanical and chemical properties The strong bond energy within the chain (Si O and Si C) makes silicones chemically stable. Silicones also present a good thermal stability and electrical insulation properties. Degradation starts only at around 400C and properties such as heat capacity are constant in a wide range of temperature (from 120 C to 240 C).

PAGE 25

25 Optical clarit y is another highly appreciated property of silicones, mostly when forming films in biological system testing because it permits for easy observati on under the optical microscope. High chain flexibility produce s the presence of a high free volume giving t he polymer good gas permeability to oxygen, nitrogen and even water vapor, being an important factor when usi ng PDMSe for biological assays. (Ltters et al. 1997). PDMSe has a very low toxicity. An initial abiotic reaction may start depolymerization follow ed by biological degradation (fungi or bacteria) which can produce di methylsilanediol as a byproduct ( Stevens 1998) Nevertheless this degradation does not state a hazard to the environment. Some experiments have taken place in the lab to prove possible leaching of PDMSe into salt water but no appreciative quantities were observed (data not shown). Surface energy Silicones present a very low surface tension. The weak interaction between methyl group and the high flexibility of the chain allow the polymer to acquire a configuration that exposes the maximum number of methyl groups to the outside forming a very hydrophobic surface good for applications such as biomaterials and foul releasing technologies. Critical surface tension of wetting for silicones i s higher than their own surface tension, so silicones can wet themselves making it easier to produce films and form molded pieces. The number of low molecular weight siloxane species increases as the silicone age by the effect of prolonged radiation in field exposure tests, which produce a detriment in the mechanical behavior of the rubber that becomes more brittle and less elastic (Homma et al. 2000). W hen silicones are exposed to plasma treatment or partial

PAGE 26

26 electrical discharge the hydrophobic surface of the rubber can also be turned into hydrophilic by the production of radicals. Different mechanisms have been studied to explain hydrophobic recovery of PDMSe. When a PDMS surface is plasma treated, a mechanism associated with burial of polar groups in the bulk and the condensation of silanol groups occurs This was an early attempt to explain the process of hydrophobic recovery (Morra et al. 1990). A different study theorized that polar groups are created on the surface after plasma treatment, and those new polar groups created reorient themselves attempting to decrease the surface energy of the system (Tth et al. 1994). However the role of these polar groups in the hydrophobic recovery is much less important than the diffusion of LMW from the bulk to the surface. Later in time, a different study proposed that a wettable silica like brittle layer is produced on the surface of PDMSe as an effect of exposure to partial electrical discharge. Analysis of the surface composition with XPS at different depths of this layer shows that cracks are produced in the process and a diffusion of PDMS chains must occur through those cracks (Smith et al. 1992). In another experiment where thermally induced or mechanically deform ed PDMS films were studied confirmed that a cracked surface present s a faster rate of hydrophobic recovery than those untouched samples, so the silica like layer inhibit s diffusion of LMW chains (Hillborg and Gedde 1998). Wettability, surface energy and topography Some concepts need to be explain ed to understand the subject of this thesis. Wettability is the ability of a liquid to interact through different intermolecular forces with a solid surface when brought together. It is said that a liquid will wet well a surface when enough strong interact ions between the molecules in the liquid and those in the surface exist, so the liquid spreads. What happens during this process is that a new interface

PAGE 27

27 (liquid surface) is created and a new balance between cohesive and adhesive forces appears. Cohesive fo rces are attractive forces (intermolecular forces) that exist between molecules of the same substance which cause a tendency in liquids to resist separation. Adhesive forces are attractive forces between two different substances (in this case liquid and solid surface). In order to achieve a good wetting ability in a specific system formed by a liquid and a solid surface of different composition, the formation of the interface must be favorable, in other words, the system must lower its overall energy by spreading and forming new intermolecular interactions within the interface. Surface energy is a measure of the energy associated with the disruption of the chemical bonds that occurs when a new surface is created. Molecules in the surface of a material have a higher energy state than those same molecules in the bulk; since the forces acting on a molecule at the surface are pulling the molecule inward ( Figure 2 6.) The difference in energy between the surface molecules and the bulk molecules is the surface energy. Contact angle: Young equation Consider a droplet in equilibrium with its own vapor and a flat, solid surface at constant temperature The balance of cohesive and adhesive forces occurring in the system is going to determine the shape of the droplet, which can be geometrically defined by measuring the contact angle. L and S are defined as the liquid vapor and the solid vapor interfacial surface energies, and LS as the liquid solid interfacial surface energy ( Figure 2 7 ) The a ngle is called the contact angle, and gives an idea of whether the interaction between the liquid and the solid (wettability) is good. Contact angle over 90 represents a

PAGE 28

28 hydrophobic surface I f we are considering that the liquid is water and less than 9 0 would mean good interaction (hydrophilic). The Young equation is an application of Laplace derivation to the case of a single droplet on top of a flat surface. In this case, a simple force balance equation can be written as follow: (2 1) Wettability: Wenzel and Cassie Baxter models The wettability of a liquid over a solid surface is mostly determined by two different factors: the surface chemistry and the topography. It is important to recall that good chemical i molecules of the liquid and those on the surface of the solid will improve the contact and as a result the wettability increases. Wenzel model describes a homogenous wetting behavior ( Figure 2 8 ) On the other hand, Cassie and Baxter model represents better the wetting behavior in heterogeneous surfaces ( Figure 2 8 ). It is common to associate non wettability with the latter model. Contact angle (CA) is frequently used as a technique to characterize the wettability of a surface. CA greater than 150 are usually related to superhydrophobic surfaces. The larger the angle the better the hydrophobic character of the material. Roughness of the sample also has a great effect on wettability since it can positively contribute to the destabilization of the liquid on top of the material enhancing the hydrophobic character. Research has been done about the metastable state of liquid drops on top of a surface that has proved how the roughness geometry needs to follow

PAGE 29

29 a mathematical distribution in order to enable stable or unstable Cassie Baxter state (Marmur 2013). Engineered Roughness Index (ERI) Different methods of surface modification have been researched in order to prevent marine biofouling through a non chemical approach. Hierarchically wrinkled topographies has been produced on PDMS through u ltraviolet /ozone treatment, with reproducible results that go from tens of nanometers to a fraction of a millimeter in the same sample, decreasing barnacle settlement with no necessity of chemical modificatio n (Efimenko et al. 2009). In order to evaluate the effect of feature size, shape and roughness on the settlement of different microorganisms an empirical factor known as Engineered Roughness Index (ERI) was proved to be valid through several tests. A n ind irect relation ship exists between ERI and organisms settlement. A first attempt was made by D ) and freedom for movement (df) (Schumacher et al. 2007b). (2 2) takes into account the side area of the ridges as well as the top of the features and depressions, to the projected surface area. This factor is always more than 1 whe n roughness of any kind exists. Depressed surface fraction (f D ) is the ratio between the surface area between protruded features and the projected surface area. Freedom of movement (df) refers to the complexity of the features and how the organism is allow ed to move between them. The n ew ERI offers the possibility to analyze more accurately

PAGE 30

30 the effect of surface roughness with respect to the Wenzel model and antifouling behavior (Schumacher et al. 2007b). Other related experiments showed how the aspect rati o is an important factor when designing an effective engineered topography for antifouling (Schumacher et al. 2007a). In the same line of study, Long et al. (2010) refine d ERI by the substitution of the rent features on the surface. The term 1 S is equivalent to depressed surface fraction (f D ). (See equation 2 3). (2 3 ) Experiment regression is done by a natural logarithmic transformation that yields a linear correlati on. The equation states as follows: (2 4) S represents the number of organisms settled in the modified surface, S SM the number of organisms attached to the smooth surface (control), m is the value of the slope in the regression and b the Y axis interception when x=0. Typically l inear correlations values over 85% were reached.

PAGE 31

31 Figur e 2 1 Progress of mo ve ments during mussel plaque deposition. F resh deposited byssus can be observed (arrows point the location). The process takes around 3 minutes (Data not shown) (Photo taken by author).

PAGE 32

32 Table 2 1 DOPA containing proteins of the byssus of Mytilus species Reused from Lin, Q., Gourdon, D., Sun, C., Holten Andersen, N., Anderson, T. H., Waite, J. H., et al. (2007). Adhesion mechanisms of the mussel foot proteins mfp 1 and mfp 3. Proceedings of the National Academy of Sciences, 104(10), 3782 3786. Copyright 2007 by PNAS. Reused with permission. Figure 2 2 Scheme of the distribution of the different mussel foot protein along the byssus structure Reused from. Nicklisch, S. C., & Waite, J. H. (2012). Mini review: The role of redox in Dopa mediated marine adhesion. Biofouling, 28(8), 865 877 Copyright 2012 by Taylor & Francis Reused with permission.

PAGE 33

33 Figure 2 3 Chemical structure of L 3 ,4 Dihydroxyphenilalanine A) Basic structure of (S) 2 amino 3 (3,4 dihydroxyphenyl) propanoic acid, as well known as L DOPA B) Chain of DOPA molecules C) Reaction of two molecules of L DOPA Figure 2 4 Adhesive and cohesive interactions of mussel protein through DOPA, A) H ydrogen bond, B) Metal complex formation, C) Di DOPA formation and D) Michael addition.

PAGE 34

34 Figure 2 5 Basic structure of a PDMS chain segment Figure 2 6. Diagram of forces on molecules in a liquid. Above a molecule in the surface, being pulled inwards. Below a molecule in the bulk, being pulled in all directions. Blue dots represent water molecules and arrows are forces.

PAGE 35

35 Figure 2 7 S chematic diagram of liquid droplet on solid surface. Figure 2 8 Schematics representation of Wenzel model (left) and Cassie Baxter (right) Reused from Marmur, A. (2013). Superhydrophobic and superhygrophobic surfaces: From understanding non wettability to design considerations. Soft Matter. Copyright 2013 by Royal Society of Che mistry. Reused with permission.

PAGE 36

36 CHAPTER 3 MATERIALS Setting up the tank: cultivating mussels Mussels are cultivated in a tank i n the lab, where simulated conditions similar to the natural environment of the animals are kept at all times, in order to minimize negative effects in the experiment. The response of the mussels in the tests must be the closest to the response expected in natural conditions. For this experiment the next list of supplies has been required: Two glass tanks, 10 and 40 gallons respectively. Clear airline tubing (Top Fin Airline tubing Bubbling air stones (Top Fin Airstones with nozzles) Air pump (Top Fin Air pump AIR 3000) Gravel (Top Fin Premium quality aquarium gravel 50lb) Salt (Instant Ocean Sea Salt) Hydr ometer (Instant Ocean hydrometer) Dechlorinator (Kordon NovAqua Plus water conditioner and Dechlorinator) Water filter (Marineland Penguin Bio wheel aquarium power filter mod. 100B and 200B) Thermometer (Top Fin floating glass aquarium thermometer) V alves (Top Fin Metal T valve) Testing kits As recommended in the literature (Burkett et al 2009), to ensure appropriate conditions in the tank when raising mussels in the lab, different properties must be analyzed and controlled at all times. The water was tested for pH (8.5) salinity (specific gravity=1.025) temperature (22 25 C) ammonia (0.1 ppm max) nitrite (0.2 ppm max)

PAGE 37

37 nitrate (10 ppm max) carbonate (7 11 dKH) and phosphate (0.03 ppm) once a week using test kits. For salinity simple measures with a hydrometer were used A mmonia, nitrite, nitrate, carbonate and phosphate were tested with API (American Pharmaceuticals) test kits. Feeding animals when being raising in a lab is a cr itical element of the process. In this case, following advice from literature (Burkett et al Blend Phytoplankton was utilized Silicone elastomer: SILASTIC T 2 A silicone based elastomer polymer was used in order to modify the to pography. The product was Silastic T 2, shipped from Dow Corning. This product is a silicone moldmaking rubber, translucent and with high strength. It is based on two component material s consisting of SILASTIC T IC T 2 Figure 3 1 . SILASTIC T the material has shown to be suitable due to its transparency, low shrinkag e and its ability to reproduce topograph y (even those in the range of a few micrometers) in a high quality standard. The typical properties of this material after cured according to specifications and provided by the manufacturer ( Table 3 1 ) The ratio of base to curing agent in weight used was 10:1, even though it has been shown experimentally in previous work in the lab (data not shown) that ratios 9:1 and 11:1 did not modify significantly some im portant properties of the elastomer such as

PAGE 38

38 crystallization temperature, melting temperature or glass temperature, by mechanical properties measured with Dynamic Mechanical Thermal Analysis (DMTA). Allyltrimethoxysilane (ATS): silane coupling agent Si lane coupling agents such as ATS are silicon based components containing both organic and inorganic reactivity in the same molecule. These chemicals work as a bond between an inorganic substrate and an organic c omponent By using ATS a connecting interface can be created between PDMSe and the glass even though the dissimilarities. ATS is used in this study in order to achieve a higher degree of interaction between the silica glass slide and the PDMSe that was cast onto its surface, without which the PDMSe film would just peel off the glass surface. Methoxy groups from ATS react with the hydroxyl radical produced onto the surface from the silica structure through a condensation reaction prod ucing methanol as a byproduct ( Figure 3 2 ) ATS is applied onto the slide as a solution with ethanol in a controlled pH environment (pH=4.5), so more hydroxyl groups are produced from the silica structure. Hexamethyldisiloxane (HMDS) HMDS was used in our experiments to prevent strong bonding between the wafer and the PDMSe film cast on top of it. On water free surfaces, HMDS chemically bonds its Si atom to the oxygen of hydroxyl groups from the silica glass releasing ammonia in this reaction. The methyl groups of the HMDS fragment thereby form ing a hydrophobic surface and thus resist ing wetting and adhesion ( Figure 3 3.)

PAGE 39

39 Figure 3 1 Curing reaction between the base resin and the curing agent of SILASTIC silicone Table 3 1 Typical properties of Silastic T2 (Dow Corning), from Dow Corning product information sheets Characterization test Value Color a p preciation Tra n slucent/transparent Viscosity (Base) 50,000 cp Viscosity (Curing agent) 550 cp Mixed (10:1 weight) specific gr a vity 1.12 After 24 hours of curing Tensile strength 0.55 Mpa Elongation 290% Tear strength 120 ppi (Die B) Linear shrinkage <0.1%

PAGE 40

40 Figure 3 2 A TS coupling agent reaction with silica glass Figure 3 3 HMDS reaction onto silica glass slide.

PAGE 41

41 CHAPTER 4 METHODS Cultivating mussels Adult mussels ( Geukensia d emissa ) between 30 and 45 mm are obtained from Duke University (North Carolina), Dr. Rittschof research group. The mussels are raised in the lab in two different tanks running in parallel under the same conditions and parameters values. One of the tank s has a capacity of ten gallons and the other holds twenty gallons. Conditions for an appropriate environment are provided through adjustment of temperature, salinity and feeding source. The parameters in the tank are tested periodically. Setting up the aquarium. Safety and previous considerations The location of the aquarium has to be strong enough to resist the weight. It is also required to be o n a flat surface with easy access for researchers to manipulate the conditions as needed One must be sure that the standing researcher can reach all corners of the tank with his hands before start filling the tank. Electrical cords and tubing that run in and out of the tank must be arranged in a U shape or in a loop. This prevent s any leakage from the tank from run ning down the cord and onto the loop. Outlets around the tank are necessary, but they must be protected from any leakage, splash from the tank or experiment handling. An appropriate lid must be chosen and its placement ensure d onto the tank to prevent from splash ing Do not beg in adding components or filling with water until be sure the location is appropriate. All safety issues must be discussed with the research advisor before setting up the aquarium.

PAGE 42

42 For the first use of the tank, clean with a diluted (1/5) solution of bleach accompanied by a very intense water rinse. A lid must be placed onto the surface to avoid excessive evaporation and contamination. This also prevents from splash ing The tank should not be exposed to direct sunlight due to subsequent promotion of algae g rowth. For a first time use tank was also necessary to leave the whole system running (air pump and filter) for at least 5 days. The water in the aquarium was DI water (conductivity=18.0 ) obtained with Barnstead EASYpure II (ThermoScientific) and I ns tant Ocean Sea Salt until the water has a value of specific gravity (measured by hydrometer) between 1.023 and 1.025. For a better mix between the water in the tank and the salt, high concentrated dissolutions of salt are previously made in a beaker and t hen added to the tank. The air must be pumping at all time s A second smaller tank, around 10 gallon s, was located next to the big tank in which the tests were made The conditions on this tank are equal to that of the larger tank. In order to get accurate salinity values, approximately 5 0 grams of Sea Salt per liter must be added to the tank, but the measures must not be taken until at least 48 hours later during which aeration must be active all the time. Water in the tan ks was continuously and vigorously aerated by using a pump ( Top Fin Air Pump AIR 3000 ), with 4 bubbling air stones in the big tank and only 2 for the small ta nk. Clear line plastic tubing was used to connect the air system. A w ater filter with a biologicall y su pported system was used, Marineland Penguin Bio Wheel Aq uarium Power Filters (mod. 100B for the small tank and mod.200B for the big one). The f ilter must be working 24/7, except for a period of 8 hours after each feed.

PAGE 43

43 Conditions in the tank Conditions r eproduced in the tanks are those (averaged through the last 5 years) for Gulf Jackson (Levy County, FL) during spring, considering these values as representi ng the west coast of Florida. Data is provided by the University of Florida/Institute of Food and A gricultural Sciences (IFAS) Shellfish Aquaculture Extension Program (SAEP), through the website http://shellfish.ifas.ufl.edu/ Those conditions are 74F, salinity of 32ppt (around 1.024 specific gravity accor ding to NOAA conversion tables) and 7 7.5 pH. Temperature was measured with a floating thermometer. The value was 74F (24C) with no chilling or heating system attach to it. Salinity was measured weekly with a hydrometer and it ha d to be in the range of 1 .020 1.025 all the time. The water was tested also for pH, ammonia, nitrite and nitrate once a week, and calcium, carbonate and phosphate once every two weeks, using API (American Pharmaceuticals) test kits. Mussel aquarium maintenance protocol A new proto col to raise mussels in the laboratory setting was developed to reduce mortality following t he directions Dr. Stafslien (University of North Dakota) provided I have created a schedule to facilitate the maintenance of the tank conditions. It has been uploa ded to the Brennan group resear ch drive Protocols Water tank maintenance protocol). Table 4 1 shows the monthly schedule and some specific directions are explained in details in the following paragraphs: Water replacement: 10% of the tank volume was replaced by fresh artificial sea wa ter, produced by mixing DI water and Instant Ocean Sea salt. It was highly recommended to measure the specific gravity of water before adding the fresh water, so

PAGE 44

44 the salinity parameter could be adjusted daily by adding a more or less concentrated portion of artificial sea water (value of specific gravity must be around 1.025 at all time s ). Another consideration was that the water replacement must take place before feeding the mussel s and test ings water conditions Keep in mind, at least one hour must pass after the addition of fresh sea water before conditions can be tested. Mussel feeding: mium Reef Blend Phytoplankton was fed to the mussel twice a week (Mo nday and Thursday). A pipette was used to measure the volume added. Mussel feedin g was prob ably the most critical step in keeping healthy water conditions, since it can rapidly increase the ammonia content in the water and poison the animals The ratio of feeding volume was of 0.3 mL per mussel per feeding. The w ater filter must be disconnected at least f or 8 hours when algae culture was added to the tank. Water testing: The water parameters to be tested are: pH, salinity, temperature, ammonia, nitrite, nitrate, phosphate, hardness and calcium. The test s were done using API (American Pharmac eutic als) test kits. Salinity was tested with a hydrometer and the temperature with a floating thermometer. Complete water replacement: To prevent the build up of algae, a co mplete water replacement was done every 4 weeks in every tank. Since the muss els were going to be transferred fr om one tank to the other both tanks were operated continuously so the conditions were constant Complete water change could not occur at the same time in both tanks. After the water change the tank could not be used for the next 4 d ays, until proper parameters were reached agai n ( Table 4 1 ).

PAGE 45

45 PDMSe patterned samples preparation All the sam ples were prepared to be tested using PDMSe ( Silastic T2) from Dow Corning as a basis material, and the procedure followed instructions from the protocol s previously reported by D Protocol for Making Patterned Slides Directly from Wafer and Unpatterned PDMSe slide preparation All the samples were cast onto precleaned microscope glass slides (Fisher Scientific 75 x 50 x 1 mm). Before use, each glass slide bonding surface was pyrolyzed (both sides) using blue flame from a bunsen burner to remove the pretreated layer they come with from the factory The slides were cool ed in the air on top of clean kimwipe. T he glass slides were then moved to the hood and covered carefully with a solution of allyltrimethoxys ilane (ATS). The ATS solution was prepared in a polypropylene cup by mixing 30mL of ethanol (190 proof) with 0.17mL of ATS. Before the addition of the ATS, the pH had to be adjusted to around 4.5, generally by adding two drops of glacial acetic acid. The reaction was left for at least 5 minutes before use After the glass slides react ed with the ATS solution for 3 minutes, the glass slides were rinsed with ethanol and place onto a dry kimwipe. Finally after 10 minutes in the oven at 120C, the glass slides are ready to be used. The wafers used to cast the textures o n the silicones samples were made by photolithography as described in Alavi et al. 1992. The w afers were cleaned with ethanol before use, dried with compressed nitrogen gas and placed in a vacuum chamber for 3 minutes to assure all water and contaminants were removed from the surface. The chamber was then disconnected from the pump, but the vacuum was not released, since it was used for vapor depositio n of hexamethyldisiloxane. HMDS was coated onto the wafers by connecting a syringe line to the vacuum cham ber and

PAGE 46

46 allow ing the transfer process to occur for 5 minutes. Two glass plates were used as a mold. The plates were covered with polyethylene terephthalate (PET) films with double sided tape. The unmodified silicone elastomer was prepared by mixing in a 1 0:1 ratio in weight (10 parts of resin to 1 part of curing agent) using approximately 25 gr ams per wafer. The mix was put in a tri corner polypr opylene (PP) beaker, where it was stirred for 5 minutes with a PP spatula, making sure to scrape the sides and b ottom of the beaker to produce a better mix ture After this time, the beaker with the mixture was placed into a vacuum chamber for degassing over 15 minutes, to rid it of any trapped air bubbles. The process of patterned PDMSe on top of a glass slide has two steps. First, a PDMSe film with the texture was cast. Glass slides are used as spacers (1,000 microns thick) by taping them to eac h corner. After this, wafers were also taped with double sided tape to the PET film, and pre cured PDMSe was poured on top of it After pouring, bubbles still trapped on top of the mixture, were released by using a syringe needle Finally the other glass plate was placed on top of the PDMSe (with the PET facing down) and pressed until the top glass plate rest s on top of all spacers ( Figure 4 1 ) A 5 lb weight was placed on top of everything and allowed to cure for 24 ho urs. Once the 24 hours have passed, the PDMSe patterned sample needs to be carefully peeled from the wafers and the P ET film, making sure the film was pulled when peeling in the same direction of the texture, to prevent from producing damage to the feature s. Then the PDMSe film was placed, with texture facing down, onto a glass plate (without the PET sheet). Spacers had to be put (1,600 microns) to ensure the thickness of the sample.

PAGE 47

47 The other glass plate was covered with PET film and ATS treated glass slid es are taped using double sided tape, with the treated side facing away from the glass. Again, mixture 10:1 of base to resin was poured on the back part of the textured PDMSe sample obtained the day before, and the two plates are pressed together, allowing the fresh not cured PDMSe acting as an adhesive. It must be pressed until the top plate rest ed on top of the spacers. The w eight was placed on top of everything and the system was al lowed to cure for 24 hours undisturbed Finally, the glass slide was peeled and the excess PDMSe was cut along the edges ( Figure 4 1 ) The surface s of the samples were pretreated before starting the test. This pretreatment consists o f wetting the surface completely with water (from the aquarium). Since the samples present a rough topography, air pockets trapped between the spacing or air bubbles on the surface need to be prevented. To ensure that the surface was completely wet, the samples were submerged in the aquarium 24 hours previous to the mussel plaque production test. PDMSe patterned samples characterization Every sample was characterized in order to ensure the features were replicated from the wafer. In addition, the features were inspected for any deformation, missing ridges and heterogeneity of pattern Also the size of the features was measured Every sample produced was taken to the optical microscope and observed at x10 magnification in different spots on to the surface. If the sample does not present any problems it can then be used for testing. The direction of the channels must be examined as well in order to place the mussel always in the same direction, during the tests.

PAGE 48

48 To verify that the size of the channels, height of the ridges and width of the spacing was correct, a sample silicone from each different texture was observed with Scanning Electron Microscope ( FEI XL 40 FEG ). Flat metallic sta ges were used for sample preparation to measure the width of the ridges and the spacing, and vertical stages for the depth of the features. All of the samples were gold coated by plasma sputtering (Denton Vacuum, LLC, Desk II cold sputter/ECTH Unit and Car bon evaporation accessory). The process was carried out at a pressure of 50 mtorr and an intensity of 45 miliamps for 40 seconds The voltage used in the SEM was 4 kV, which was bright enough to obtain good quality and to also prevent damage to the sample. The results were obtained by comparison with the automatic measure bar display ed in the pictures taken ( Figure 4 2.) Mussel plaque production When a mussel was in contact with a surface, plaque production occurs naturally on top of any substrate as a mechanism of survival. Nevertheless, the mussel foot detects the suitability of different sites around the animal and it may produce the migration of the mussel out of the sample, onto a more favored substrate. To prevent Universal) was placed around the mussel and the sa mple so the plaque production was forced to happen onto the topography. The band must be strong enough to prevent the animals from moving but at the same time must permit t he open ing and clos ing of the shell enough to let the foot extend out of the sample to produce the plaques. Mussel selection was an other im portant issue. Only healthy mussel s were used. To ensure this good condition, byssus threads need ed to have been deposited previously onto the substrate of the aquarium, and the animal had to prove motion

PAGE 49

49 reaction when contact. No gaping motio nless organi sms were used. It was also necessary that the mussels have been living at least for a week in the aquarium before they are used for experimentation. Selection based upon the size of the mussel was also mandatory Since values of surface area of the plaques were highly dependent on the size of the mussels, they have to be selected with in the narrow range of size (between 35 45 mm). PDMSe topographies cast onto precleaned microscope glass slides (Fisher Scientific 75 x 50 x 1 mm) are be placed into the aquari um, and only ONE mussel for every sample (around 20 mm length) is attached with the rubber band to produce the interaction. Ten different mussels are tested for each sample. Topographies of a 1:1 ratio of 5, 20, 50, 100 and 200 microns plus the smooth PDM Se used as a control are tested, adding up to a total of 60 mussels per experiment. The width of the ridges and the spacing was the same. The experiment run for 72 hours, the necessary amount of time according to literature (Burkett et al, 2009, Aldred et al, 2009) and previous observations done in the lab (data not shown) to produce adequate interaction and plaque adhesion Mussels were observed daily during the experiment, ensuring proper placement throughout the trial The mussel was able to stick the fo ot out and move it around with a broad range of freedom since the organ presents a high degree of flexibility and the capacity of getting to a long extension Neverthele ss, it has been prove n previously in the laboratory (data not shown) that certain dire ctions in relation with the location of the mussel onto the sample were preferred. To ensure reproducibility and homogeneity of the obtained data the position of the mussel on the sample follow ed the same

PAGE 50

50 orientation in all tests, so the animals were al ways on the same position with respect to the pattern ( Figure 4 3.) This improved the statistical significance of the Nuclear Form Factor (NFF) values. Extraction procedure The method of extraction was very simple. First, the glass slide with the attached mussel was removed from the tank. The threads of each of mussel were cut carefully using small scissors or a razor. Detachment of the plaque at th e plaq ue surface interface was to be avoided by following appropriate procedure. Samples to be tested for strength were not cut from the surface Samples were cleaned using three different salt water solution s followed by rinsing with three different ethanol wa ter solutions Salt solutions with values of specific gravity o f 1.015, 1.005 and DI water, were prepared. S ample is submerged in a Petri dish with the first solution (1.015), later with the second solution (1.005) and finally in the DI water, with a spaci ng of 6 minutes between each The sample is then washed out with DI water, and cleaned with three ethanol water solution s. The solutions have a composition of ethanol to water of 1:4, 2:4 and 3:4 There is also a gap of 6 minutes between them. Finally, the sample is put into a vacuum for 5 minutes to ensure a dry surface. This treatment prevented the plaque and the surface of the sample from cracking with drastic changes in salinity ( Figure 4 4.) while also providing a clean surface. Measuring n umber of plaques Once the samples had been washed and dried, optical microscopy characterization was used to count the number of plaques pro duced on each sample. The whole patterned area of the sample was swept with the microscope (Axioplan 2,

PAGE 51

51 Zeiss) at a magnification of x5, and a picture of each plaque encountered is taken. In some cases, more than one plaque appear s on the screenshot. A sec ond swe ep at higher magnification (x10) was made for each sample. Pictures of individual plaques on the surface are also taken at this magnification ( Figure 4 5.) Since this method proved itself to be un reliable the number of plaques the mussel produced was measured directly by counting the number of steps (individual plaque detachment) from the data collected with the texture analyzer (See Tensile test: t exture analyzer). Pictures must be taken from the same orientation with respect to the topography pattern in order to certify that the measurements of the Nuclear Factor Form are all equal as to obtain a common conclusion. Measuring surface area and Nuclear Form Factor (NFF) Since one of the aim s of this study was to evaluate the effect of the topography on the mussel plaque adhesion and the spreading of the mussel adhesive onto a surface, measuring the surface area of the plaque s onto different topog raphies was a critical issue of this study. Images from the optical microscope with a magnification of x10 were processed using ImageJ. Eight of the pictures taken for every topography were selected randomly Images where the edges of the plaque were d iffused, where two or more plaques overlapped or the collagen threads or any other artifact decreased the quality of the image were automatically discarded. A threshold was established, by ImageJ, for each image to enhance the shape of the plaque relati ve to the pattern ( Figure 4 6 ) Once the image was enhanced and the edges easily identified, the shape was traced manually, since the accuracy is higher than if automatically traced by the software. The length and the width, values necessary to de fine the Nuclear Form Factor (NFF) were also obtained manually. NFF is well

PAGE 52

52 defined in Figure 4 7 . NFF was used previously by Carman et al. in 2006 for research on alignment of porcine c ardiovascular endothelial cells, and was previously defined by Dunn and Heath in 1976, in their research on heart fibroblast cells contact guidance study. Even though cells were not tested, the morphology of the adhesive (plaque) onto different samples was registered as a way of evaluating the capacity of the adhesive to adapt to different shapes according to the surface roughness and design, as to minimize the energy b arriers that the spreading front of the adhesive needs to undergo. Degree of penetration (Scanning Electron Microscope) The procedure and equipment used in this part were the same as those PDMSe patterned samples characterization section of this thesis. Other techniques and tools were tried unsuccessfully such as microtome and liquid nitrogen freezing split, but the size of the plaque and the properties (viscoelasti c behavior) of the silicone made these processes not viable. Tensi le test (texture analyzer) Tensile force of the mussel was measur ed in this project using a texture analyzer (XT plus Texture analyzer from Texture Technologies Corp.) ( Figure 4 8.) The mussel plaque production test was performed as described in the part production was ended after 48 hours instead of 72. Every mussel taped to the sample was extracted carefully from the water and the rubber band cut. In the case the plaque production was not satisfactory (less than 5 attachments by simple visual observation) the experiment was repeated on top of the same sample with a different mussel. The glass slide was mounted on the stage of the texture analyzer by taping just the side,

PAGE 53

53 and the mussel was gripped with a clamp, properly mounted on the movable arm of the device ( Figure 4 9.) The test r a n under steady elongation, which means that the mussel was pulled off at a continuous speed (0.5 mm/sec) while the force was measured at every instant. Settings for the test are 4 points per secon d, tension mode, 5 gr. trigger force and 70 mm maximum elongation. During the test the individual threads coming from the mussel and connected to the sample can be seen stretched and one by one they undergo failure (rupture). The experiment is finished whe n no threads can be observed connected to the sample. No more than 5 minutes passed from the mussel extraction to the end of the test, because Mfps change structure as they dry. According to literature plaques as well as threads removed from sea water expe rience changes in mechanical properties within 10 minutes (Burkett et al. 2009) (Data not published). The number of points varies according to the duration of the test. Number of plaques, strength of the adhesion and length of the byssus structure affect t he data acquisition time of the experiment. When the mussel is being pulled off, the byssus stretch es and create s a force that oppose s the movement. This force was registered by the texture analyzer. First the force increased until a maximum load was reach ed and then it began decreasing zero load (some noise was always registered). The bell shaped evolution of the for ce versus elongation ( Figure 4 10 ) was explained by the detachment of the individual thread can be ob served along the graph, cor responding each one with a different detachment event By data processing with Excel the pea ks were identified and the difference

PAGE 54

54 between a consecutive maxim um and minimum attributed to the ind ividual force component of each plaque. Data was collected for 5 different samples (channels 5x5, 20x20, 50x50, 100x100 and 200x200) plus the control (smo oth PDMSe), and each tensile test was run eight times in each different pattern. The d ata has been processed and the results are

PAGE 55

55 Table 4 1 Water tank maintenance monthly schedule MONDAY TUESDAY WEDNESDAY 1st week Water replacement (1) Mussel feeding (2) Water replacement Water replacement WATER TEST (3) 2nd week Water replacement Mussel feeding Water replacement Water replacement WATER TEST 3rd week Water replacement Mussel feeding Water replacement Water replacement WATER TEST 4th week Water replacement Mussel feeding Water replacement Water replacement WATER TEST THURSDAY FRIDAY 1st week Water replacement Mussel feeding Water replacement 2nd week Water replacement Mussel feeding Water replacement 3rd week Water replacement Mussel feeding Water replacement 4th week Water replacement Mussel feeding Complete water replacement (4)

PAGE 56

56 Figure 4 1 M old setup for preparation of patterned PDMSe on glass slides (sectional view)

PAGE 57

57 A B C Figure 4 2 SEM captures the different topographies used in this study (3.67 kV) (A) 5 CH5x5, (B) 20 CH20x20, (C) 50 CH50x50, (D) 100 CH100x100, (E) 200 CH200x200 (Photo taken by author).

PAGE 58

58 D E Figure 4 2 Continued Fig ure 4 3 General plaque production experiment set and the direction corresponding to the position of the mussel

PAGE 59

59 A B Figure 4 4 SEM micrograph of the edge of a plaque onto 10 CH 20x20. The plaque adhesive filling the spacing between the ridges can be ob se rved Cracks in both the PDMSe sample and the plaque are marked with yellow circles (Photo taken by author).

PAGE 60

60 Figure 4 5 Optical microscope picture of mussel plaques onto 20 CH20x20 at x10 magnification (Photo taken by author). Figure 4 6 Optical microscope image of two mussel plaques onto 10 CH 20x20. On the left without image processing. On the right after threshold enhanced processing with ImageJ (Photo taken by author).

PAGE 61

61 Figure 4 7 Schematic of the measurements taken by ImageJ to calculate NFF where L is the length of the plaque parallel to the ridges and W is the width of the plaque orthogonal to the ridges. Figure 4 8 Tensile analyzer (XT plus Te xture analyzer from Texture Technologies Corp.) used in tensile force measurements (Photo taken by author).

PAGE 62

62 Figure 4 9 Texture analyzer set up. Mussel is attached to the sample by different byssus structures (Photo taken by author). Figure 4 10 Ex ample of force versus distance plot produced from data collected by texture analyzer for an individu al mussel in a plaque strength test.

PAGE 63

63 CHAPTER 5 RESULTS AND DISCUSSION This study consisted of the measurement of geometrical factors of mussel plaques after plaque production experiment ation, as well as tensile force experiments using 5 different patterns described in Table 5 1 . PDMSe smooth surfaces were also tested and used for comparison as a control Number of plaques The number of plaques has prove n to be a challenging factor to measure in our laboratory because of the size of the samples. For technical reasons the size of the samples cannot be bigger than 50 by 50 mm and the size of the mussel permits the animal to reach out of the patterned area more often than desired with its extendable foot causing experiments to be repeated Actually it has been observed that most of the mussels tend to attach in those non patterned areas, in other words, in the border of the sample (Data not shown). It seems like they purposefully avoid ed the pattern ed area (See Figure 5 1 ) According to the experiments performed in this study, mussels produced an average of 16 and 23 plaques which is in agreement w ith literature: between 10 and 40 (Burkett et al. 2009) within a very similar period of time (48 hours versus 72 hours in this case ). It is important to mention that the specie of mussel used in this project is different than those in other papers. Table 5 2. show s that the low est number of attachments occurred in those samples with dimensions of 50x50 microns feature size ( =16.38, SE=1.903, n=8 ) and the highest to 20x20 microns size ( =22.75, SE=2.206, n=8 ). In Figure 5 2 a graph representing the same set of data can be observ ed. If the values are compared with the control value ( = 17.00 SE= 1.62, n=8 ), the average

PAGE 64

64 number of plaques for 50x50, 100x100 and 200x200 do not show statistical difference between themselves and the control. On the other hand average values for 5x5 and 20x20 are very similar with respect to each other but present a significant increas e relevant to the smooth PDMSe. An explanation for this may be that features corresponding to 5x5 and 20x20 slightly promote the plaque production, while the other feature s follow the control fashion. It could also be that the individual plaque adhesion strength is weaker in the 5x5 and 20x20 so the mussel produces more byssus structures in order to provide itself with a stronger final adhesion value. It is important to me ntion that the values of the standard deviation for all samples are considerably high, signifying the average values cannot be considered as highly accurate but still represent a valid approximation. For this reason the number of plaques was not considered for further discussion Plaque surface area The plaque surface area average values were calculated with 8 different values obtained by 8 different tests The area of 64 different plaques w as measured by testing 8 different mussels for each different t opography. The highest value was that for 5x5 ( = 0.541575 mm 2 SE= 0.019021, n=8 ) and the lowest value was on 200x200 samples ( = 0.349550 mm 2 SE= 0.049025, n=8 ) ( Table 5 3.) Plaque surface area values for 5x 5, 20x20 and control (smooth) did not show any significant difference ( Table 5 3.) The value for 50x50 pattern goes slightly below the value range ( = 0.510939 mm 2 SE= 0.031342 n=8 ). In the plaque surface area value for the 100x100 microns topography, a step appeared ( = 0.368991 mm 2 SE= 0.051340 n=8 ), showing a significant difference with respect to the preceding size scale. In the following size (200x200 microns) the value for the plaque surface area did not present an important

PAGE 65

65 change with the previous size, even though it decreased Variance test ANOVA was applied to the different values of plaque surface area on each topography ( Table 5 4 ). The homogeneity of these parame ters could not be rejected at 90 % confidence level for each sample or between samples of a same group Additional Tukey statistical analysis ( Table 5 5.) proved that 50x50 topography did not exhibit a statistically significant difference with any other sample except for the 200x200 topography and at the same time the 200x200 tensile force value present ed a significant difference with all topographies except that of the 100x100 It is noteworthy that the size of the plaque showed a steady decreasing while increasing the size of the features. A n important step exists in the graph ( Figur e 5 3.) show ing a transition in the range of the size (reducing the plaque surface area value to approximately 37%) that was exclusively related to the size of the feature. It could be that as the height of the ridges increases, the size becomes too large for the mussel adhesive to overcome. Thus, spreading is restricted such that the plaque size is reduced The impossibility of a higher degree of spreading can not be achieved, this may be considered as an energy barrier related to the surface tension of the adhesive. spread onto the substrate but also, needs to have a high enough value to prevent from being dissipated in the water since all the process takes place under water. To prove a mathematical correlation between the two parameters at hand ( plaque surface area and size of the channels ) a new graph was created ( Figure 5 4.) and a linear correlation calculated. Even if the graph presents a good correlation (R 2 =0.8424) it was cle ar that a pronounced step existed when increasing the size of the

PAGE 66

66 features from 50x50 m icron to 1 00x100 micron. This was indicative of the existence of a critical point in the features size between 50 and 100 marking an important transition of the spreading behavior of the adhesive. In this range of values the channels topography had a significant e ffect o n the plaque surface area. It was assumed that the amount of adhesive the mussel secreted every time it produced a byssus was the same no matter what substrate, so due to the larger features present ing larger wells between ridges it seemed that a bigg er portion of the adhesive filled the space between the ridges. Nuclear Form Factor (NFF): Morphology of the plaques Nuclear Form Factor (NFF) is a method based on the measurement of the length and the width of a two dimension feature, in thi s case the plaques. It indicated the degree to which the plaque adhesive adapted to various topographies by acquiring a n elongated shape by following the direction of the channels. It was expected that during spreading the adhesive front moved first into t he spacing of the features since it is energetically favored due to two reasons The first of these reasons is the space availability and pressure effect Secondly is the difference in surface tension between the adhesive and the water being less than the difference between the adhesive and the substrate. This means that the expected shape of the plaque was to be roughly elongated in the direction of the channels. For more information, please refer to Table 5 6.) Data for the control are not shown because smooth samples did not present any topography. Statistical analysis was conducted by using ANOVA variance study ( Table 5 7. ) T he means were statistically different from each other (p=0.001) A Tukey st atistical analysi s ( Table 5 8.) showed that only the NFF value for 5x5 topography was

PAGE 67

67 the only topography statistically different from the rest of NFF mean values Also the 200x200 and 20x20 were shown to be statisticall y different. The lowest NFF value was for 5x5 sample s ( = 0.88 SE= 0.076 n=8 ). The largest NFF value was for 200x200 feature size samples ( = 1.91 SE= 0.092, n=8 ). This means that the plaques present ed a more elongated shape in the directions of the channel in the bigger size features. A continuous increas e appeared in NFF with increasing size of the features which can be observed ( Figure 5 5.). NFF values were graphed versus size of the features ( Figure 5 6.) The graph showed a good corre lation (R 2 =0.8471) even though it is important to point out that the chang e initially from 5x5 to 50x50 is larger than the change from 50x50 to 200x200, although the difference in size between the latter set of values is much larger. That can be explained considering that after a certain value around 50 microns of the features the degree of orientation of the plaque was affected more slowly. It was expected that larger features permit a better penetration of the adhesive between th e ridges, since the adhesive could flow better. Degree of penetration The observation of the plaques onto the sample by Scanning Electron Microscope was very useful for characterization. It can be observed that the adhesive produced by the mussel fill ed the spacing between th e ridges of the sample, as shown in the SEMs ( Figure 5 7. to Figure 5 10.) S amples after the plaque production test, illustrate that the adhesive filled spaces changed with feature size In the first two figures the plaques were detached from the surface probably during the pr ocess of gold coating or by the effect of the electron beam in the SEM. It is easy to observe in these first two figures that the outline of the shape is not completely round and smooth but is

PAGE 68

68 spacing. S ome of the ridges have been bent and displaced out of the normal shape ( Figure 5 10. D ) It seems that the adhesive actively pulled the top of the ridges and caused them to shift together to the cen ter point where the thread grew from. It could be that after spreading, the adhesive protein cures that produc es the cross linked structure W hen curing the structure shrink s around a central point in the middle of the plaque, causing the shape of the pattern to rearrange in order to adapt to the new stress. It could also be related to the dehyd ration process. The degree of penetration of the adhesive between the ridges could have some implications in the attachment strength of the mussel to the substrate. If the adhesive has the ability of creating an effective adhesion to the wall of the ridges this would mean that topography could contribute critically to the overall adhesion strength of the animal. In other words, those sizes in which the penetration in to the topography is reduced the overall attachment strength would be increased by the increased area of the plaque. This subject is addressed later in this thesis when measurements o f the actual adhesion are presented. Texture analyzer Eight different mussels for each of the 6 samples were used (48 different mussel s in total). Each mussel was attached to the texture anal yzer and then a tension force was applied to the animal in the vertical direction causing the plaques to det ach from the substrate. Force was measured as well as distance and time. A s eries of da ta o f force versus distance were displayed. I t can be observed one of the moment s during

PAGE 69

69 the texture analyzer test when the byssus is being stretched as an effect of the applied load ( Figure 5 11.) The graphs are approximately bell shaped. In general the force that the mussel oppose s stretch ing rises first gets to a maximum peak and then goes back to zero or a value close to zero. A number of steps are observed along the graph Counting the steps can be correlated to counting the number of plaques on top of the surface. This means that the number of steps equals the number of plaques onto the sample. There are a number of parameters mentioned in the anal ysis of the graphs that need to be addressed before go ing further in to the study. The se are described in the following lines: Number of plaques and tensile load : Each of the members on the list of Tensile d to the difference in force values between the upper limit and the lower limit of each step, each of them corresponding to an individual plaque. M aximum load: Is the maximum value reached along the test. This is a me asure of the tensile strength as a comb ination of a group of plaques work together A verage tensile strength: This tensile strength is the average of all the values for individual plaques for each test It is an approximation of the adhesion strength value for the specific sample Cohesive fail ure: Two different failures are contemplated in this project A dhesive failure is that taking place at the plaque material interface C ohesive are those produce d at some point along the collagen thread. It is important to be aware of the number of cohesive failures since those values do not give information of interaction between plaque interface. This can be done by simple observation under the optical microscope once the test is finished. Average tensile stress: This tensile stress is obtained by dividing the different values of tensile strength by the value of the plaque surface area (fixed in every feature size, see results in plaque surface area). Tensile force versus elongation was graphed for every different sample ( Figure 5 12. to Figure 5 17.) In these graphs the eight different tests per sample can be

PAGE 70

70 observed in the same plot for each topography and the blank i.e. (smooth PDMSe). A large variety of results can be detected within the same sample such as largely different values of maximum tensile force and the shape s of the curves This is because every mussel produce s a different number of plaques and the distribution of the plaques were not the same each time Many plaques can be produced around a very small s pot, generating a high load by working collectively against the force applied. On the other hand the plaques were spread out all over the sample so the resulting force does not count as a whole but the tension created by individual plaques in different st ages of the test. When the plaques were produced in a small region of the sample, none or very few steps are observed along the graph before the peak T he peak shape tends to be sharp as for the example in Test 1 ( Figure 5 12.). When the sample exhibited two different regions of plaque attachment then the graph displays a primary maximum peak and a secondary peak as in Test 5 ( Figure 5 14.). The other ext reme is that the mussel produced the byssus on the substrate so the graph spread s out and does not include a differentiated maximum peak as in the case of Test 2 ( Figure 5 12.). Those steps that can be observed before the maximum are related to those threads that were weaker or located further from the shell T he f a rther they are from the shell the faster they reach their maximum elongation and fail One may notice that the graph always starts at 50 mN, because it was set as the trigger force in the texture analyzer system. It is important to mention that in some of the graphs the end of the test shows neither a final value that is neither close to zero nor the starting value. This happened for two different reasons. First of all the noise present during the testing

PAGE 71

71 process, which was assumed to be around 100 mN above and below the starting value. A second contribution to this deviation was the weight of the mussel that some tests seem to not correctly take into account These deviations did not n egatively affect to the results. They cause the values to go off the range of the graph in a certain degree but the individual drop in tensile stress that configure the steps along the graph can be correlated with individual plaque attachments without conc ern. For this study it was assumed that the tensile strength values were produced only through the breaking process of the plaques when detachment happens by the application of the external tensile load with the texture analyzer. The fact that a correlat ion between the tensile force values and the individual plaque detachment process exists is supported when taking a look at the histograms. Histograms for all the registered tensile force values acquired with the tensile analyzer test are shown for each fe ature size ( Figure 5 18. to Figur e 5 23.) In all cases a regular distribution is observed with a narrow range of values. The b lank or the control, presents the highest frequency in the range 10 15 mN, being the lowest tensile average force 5x5 and 100x100 features present t he highest frequency in the 15 20 mN range 50x50 and 200x200 micron feature samples present the highest frequency in the 20 25 mN range 20x20 features show the highest frequency in the 25 30 mN range, being the highest tensile average force. These distributions of values are significant because they back up the theory that the steps in the tensile fo rce versus elongation represent single detachments.

PAGE 72

72 It is important to mention th at two different type s of failures were considered in this study. Adhesive failure is that produced at the interface between the adhesive and the substrate while cohesive is failure produced any point along the collagen thread or on top of the plaque. The differences were easily detected because the plaques are large enough to report the two different failures by optical microscopy. Those areas where adhesive failure happened were not chemically characterized to prove if any adhesive remained, so it was ass umed the failure was complete and no adhesive stayed on the area. Values from 19% cohesive failure in 5x5 microns channels (minimum) to 35% cohesive failure in 20x20 microns features (maximum) were reported as averages. This involved that the rest of the d ata analysis in this project include an error since all failures are considered adhesive, nevertheless since cohesive failure percentage range is similar the effect in the study is minimized Examples of adhesive failures in the different topographies are listed in Tensile force Table 5 9 lists the average values for tensile force obtained as an average of all individual plaque values gathered for each sample. The highest value in tensile force corresponds to 20x20 microns feature ( = 45.161 mN SE= 2.067, n=182 ), followed closely by that value on 5x5 ( = 42.368 mN SE= 2.331, n=177 ). The rest of the average values for tensile force in 50x50 ( = 36.425 mN, SE= 2.301, n=131) 100x100 ( =31.862 mN, SE= 1.597 n=142) and 200x200 ( = 31 .732 mN, SE= 1.677 n=137 ) are along the same line of the value corresponding to the blank ( = 32.706 mN, SE= 2.107, n=136 ) which does not represent a significant difference (p=1E 7) ANOVA statistical analysis show the differences in the tensile force value s that were statistically different ( Table 5 10.).

PAGE 73

73 Additional ly a Tukey statistical analysis ( Table 5 11. ) showed no stat istical differences between 50x50, 100x100 and 200x200 tensile force mean values with respect to control. However 5x5 and 20x20 tensile force mean values were significant differen t from the blank A lso the 20x20 mean value wa s statistically different from the values of all topographies except that of the 5x5. In Figure 5 2 5 the graph shows all values for each sample analyzed. Looking at the graph it is easier to see how values for the first two features in size display an important difference with respect to t he smooth surface. Decreasing the feature size of the topography improve d the adhesion of the mussel adhesive on the substrate. Samples with 50x50 microns feature show a significant decreas e. 100x100 and 200x200 features have values similar to the blank Th is indicates, that according to tensile force measurements a larger size do es not have a significant effect on the adhesion strength of the mussel adhesive. In general terms the tensile force scales are inversely with respect to the feature size. Tensile force versus plaque surface area Considering that the adhesive secreted for the mussel has always the same composition and that volume and chemistry of the substrate is the same in every test, it should be expected that a higher tensile force corresponds with a larger plaque surface area This is because a large r adhesive surface area would create a larger number of molecular interactions with the substrate. Observing the tensile force is reduced with the decreasing size of the mussel plaque ( Figure 5 26.) In order to use plaque surface average values obtained from 72 plaque production experiments and relate them through tensile force values obtained from 48 hours experiments it was necessary to establish that no statistically difference is detected. For that purpose 10 different micrographs from the 48 hours experiment were evaluated in terms of plaque surface

PAGE 74

74 values over 0.05) between plaque surface area values from both 48 h and 72 h experiments ( Table 5 1 2 ). When comparing values with the blank a number of important points should be addressed. First of all, in 5 CH5x5 the value for the tensile force was around 1/3 larger if compared with the blank, but the size of the plaqu e was not significantly different. This means that the 5x5 topography must promote adhesion. The same effect is observed when looking at the values for 20 CH20x20. This enhancement of adhesion was attributed to the topography and the fact that it provides more surface area for the mussel adhesive to attach. For 50x50, 100x100 and 200x200 features the values of tensile force are in the same range as that of the smooth samples N evertheless the plaque surface a rea values decrease with respect to the value for the blank. This suggests that even if the area was smaller in the topographies a certain degree of penetration was achieved on these sizes too. Tensile stress Table 5 9 also shows the average values for tensile stress (mN/mm2), obtained from the values of tensile force and the plaque surface area values calculated previously in this project. The maximum valu e corresponds to 200x200 samples ( = 90.779 mN/mm 2 SE= 4.796 n=137 ) and the lowest value to 50x50 samples ( = 71.291 mN/mm 2 SE= 4.504, n=131 ). Nevertheless all the values for the different features are significantly higher than th e value for tensile stress in the blank ( = 60.977 mN/mm 2 SE= 3.928, n=136 ) which proves the topography has an effect on the attachment stress. The values obtained in this study for tensile stress were

PAGE 75

75 compared to those obtained for smooth PDMSe in literature (93.25 mN/mm 2 in Burcket et al. 2009). Recall that the specie used in the cited literature (Mytilus Edulis) is different than the one used in this study (Geukensia demissa). Mytilus Edulis is bigger in size however the content o f DOPA and the number of proteins in the byssus are essentially the same between different species and therefore allows this comparison The bar chart of tensile stress values for the different topographies tested as well as the blank clearly illustrates t he values increase with the size of the feature ( Fig ure 5 27.) In this figure tensile stress values are calculated by applying roughness i ndex (1 for the blank, 2 for 5x5, 20x20, 50x50 and 100x100, and 1. 5 for 200x200). T he tensile stress of the 50x50 was an exception which is interesting. A comparison of the tensile stress (roughness index applied) and tensile force as a function of the fea ture dimension, provides some insight into the effect of the 50 micron features ( Figure 5 28. ). Both stress and force increase in response to the features until the +50SK50x50, at which point both decrease rapidly until the +100CH100x100 features are evaluated. The stress and force diverge at the +100CH200x200 topography The overall tensile stress increases to the maximum valu e while the tensile force decrease d to the same value of the smooth PDMSe. We have made the assumption that the composition of the adhesive is constant as w ell as the adhesion mechanism. This is reasonable since the animals were also a constant as well as the experimental conditions So, i n order to explain this it must be mentioned that plaque surface area was measured as a planar projection of the pl aque onto the substrate. At the same time s ome degree of penetration into the spacing has also been previously observed (see

PAGE 76

76 Results, Degree of penetration). This data show s that mussels achieve some degree of effective adhesion to the side of the ridges that is a function of the specific size of the features This shows t he effective area is not that measure d of planar project ion but instead, some other larger value.

PAGE 77

77 Table 5 1 Different sizes of the features in the different patterns Sample number P attern Design Dimensions Width x Spacing ( m x m) Height ( m) 1 Channels 5 x 5 5 2 Channels 20 x 20 20 3 Channels 50 x 50 50 4 Channels 100 x 100 100 5 Channels 200 x 200 100

PAGE 78

78 A B Figure 5 1 Pictures that illustrate the mussel plaque production extends beyond the patterned area. The border between the pattern and unpatterned area is marked with a red line (Photo taken by author).

PAGE 79

79 Table 5 2. Table with the more significant statistical values of number of plaques for the different samples (data collected from the texture analyzer tests) Number of plaques SMOOTH Average 17 N 8 st. dev. 4.58 90%CI 2.66 SE 1.62 5X5 Average 22 n 8 st. dev. 9.97 90%CI 5.80 SE 3.52 20X20 Average 23 n 8 st. dev. 6.24 90%CI 3.63 SE 2.21 50X50 Average 16 n 8 st. dev. 5.38 90%CI 3.13 SE 1.90 100X100 Average 18 n 8 st. dev. 8.09 90%CI 4.70 SE 2.86 200X200 Average 17 n 8 st. dev. 4.40 90%CI 2.56 SE 1.56

PAGE 80

80 Figure 5 2 Nu mber of plaques in different patterns (texture analyzer test) compared with control sample (smooth PDMSe). Values are average s and the error bars correspond to standard error (SE). 17 22 23 16 18 17 0 5 10 15 20 25 30 # of plaques Number of plaques

PAGE 81

81 Table 5 3. Plaque surface area statistically most important parameters Plaque surface area (mm 2 ) SMOOTH Average 5.36E 01 N 8 st. dev. 5.38E 02 90%CI 3.13E 02 SE 1.90E 02 5X5 Average 5.42E 01 n 8 st. dev. 8.84E 02 90%CI 5.14E 02 SE 3.12E 02 20X20 Average 5.40E 01 n 8 st. dev. 1.06E 01 90%CI 6.19E 02 SE 3.76E 02 50X50 Average 5.11E 01 n 8 st. dev. 8.86E 02 90%CI 5.16E 02 SE 3.13E 02 100X100 Average 3.69E 01 n 8 st. dev. 1.45E 01 90%CI 8.44E 02 SE 5.13E 02 200X200 Average 3.50E 01 n 8 st. dev. 1.39E 01 90%CI 8.06E 02 SE 4.90E 02

PAGE 82

82 Figure 5 3. Plaque surface area values (mm2) of the different patterns tested (5x5, 20x20, 50x50, 100x100, 200x200 microns, width x spacing) compare to control (smooth PDMSe) Values are average s and the error bars correspond to standard error (SE). Table 5 4 ANOVA variance study for the plaque surface area values on the different topographies ANOVA Source of Variation SS df MS F P value F crit Between Groups 0.3257 5 0.0651 4.8707 0.0013 2.43769264 Within Groups 0.5617 42 0.0134 Total 0.8874 47 5.36E 01 5.42E 01 5.40E 01 5.11E 01 3.69E 01 3.50E 01 3.00E-01 3.50E-01 4.00E-01 4.50E-01 5.00E-01 5.50E-01 6.00E-01 Plaque surface area (mm 2 ) Plaque surface area

PAGE 83

83 Table 5 5. Tukey post hoc analysis of plaque surface area values (being 1=smooth, 2=5x5, 3=20x20, 4=50x50, 5=100x100, 6=200x200)

PAGE 84

84 Figure 5 4. Plaque surface area values (mm 2 ) vs. size of the features ( m 2 ) (width/spacing, ratio 1:1) to show linear correlation between the two parameters. Values are average and the error bars correspond to standard error (SE). 5.36E 01 5.42E 01 5.40E 01 5.11E 01 3.69E 01 y = 9E 4x + 0.566 R = 0.8843 3.00E-01 3.50E-01 4.00E-01 4.50E-01 5.00E-01 5.50E-01 6.00E-01 0 50 100 150 200 Area (mm 2 ) Size of the feature ( m) Plaque surface area vs size of the features

PAGE 85

85 Table 5 6. Nuclear Form Factor statistically most important parameters NFF SMOOTH Average 1.66E+00 N 8 st. dev. 1.77E 01 90%CI 1.03E 01 SE 6.27E 02 5X5 Average 8.85E 01 n 8 st. dev. 2.15E 01 90%CI 1.25E 01 SE 7.59E 02 20X20 Average 1.26E+00 n 8 st. dev. 4.47E 01 90%CI 2.60E 01 SE 1.58E 01 50X50 Average 1.46E+00 n 8 st. dev. 6.84E 01 90%CI 3.98E 01 SE 2.42E 01 100X100 Average 1.49E+00 n 8 st. dev. 3.70E 01 90%CI 2.15E 01 SE 1.31E 01 200X200 Average 1.91E+00 n 8 st. dev. 2.60E 01 90%CI 1.51E 01 SE 9.20E 02

PAGE 86

86 Figure 5 5. NFF values of the different patterns tested (5x5, 20x20, 50x50, 100x100, 200x200 microns, width x spacing) including Smooth (control). Values are average s and the error bars correspond to standard error (SE). Table 5 7. ANOVA variance study for the Nuclear Form Factor (NFF) values on each topography ANOVA Source of Variation SS df MS F P value F crit Between Groups 4.904494 5 0.981 5.4189 0.0006 2.43769264 Within Groups 7.602537 42 0.181 Total 12.50703 47 1.66 0.88 1.26 1.46 1.49 1.91 0.50 0.70 0.90 1.10 1.30 1.50 1.70 1.90 Nuclear Form Factor (NFF) Nuclear Form Factor (NFF)

PAGE 87

87 Table 5 8. Tukey post hoc analysis of nuclear form factor values (being 1=smooth, 2=5x5, 3=20x20, 4=50x50, 5=100x100, 6=200x200)

PAGE 88

88 Figure 5 6. NFF values vs. size of the features ( m 2 ) (width/spacing, ratio 1:1) to show linear correlation between the two parameters considered. Values are average and the error bars correspond to standard error (SE). y = 4.4E 4x + 1.0757 R = 0.8471 0.00 0.50 1.00 1.50 2.00 2.50 0 50 100 150 200 250 NFF Size of the features (um) Nuclear Form Factor (NFF)

PAGE 89

89 A B Figure 5 7. SEM pictures of detached mussel plaques on 20x20 microns channels feature (Photo taken by author).

PAGE 90

90 A B Figure 5 8. SEM pictures of attached mussel plaques on 2 0x20 microns channels feature (A) and 50x50 microns feature (B ) (Photo taken by author).

PAGE 91

91 A B Figure 5 9. SEM pictures of attached mussel plaques on 5 0x50 microns channels feature (A ) and 20x20 microns feature (B ) (Photo taken by author).

PAGE 92

92 A B Figure 5 10. SEM pictures of attached mussel plaques on 50x50 microns feature (Photo taken by author)

PAGE 93

93 Figure 5 11. Tensile test performed in texture analyzer. A number of threads can be observed going from the shell of the mussel to the substrate. The grip appears in the figure producing the load to the byssus system (Photo taken by author). Figure 5 12. Results for the texture analyzer tensile test on smooth PDMSe. The graph shows tensile force (mN) versus elongation (mm) for 8 different tests. -100 0 100 200 300 400 500 600 700 800 0 5 10 15 20 25 30 35 Tensile force (mN) Elogation (mm) SMOOTH Test 1 Test 2 Test 4 Test 5 Test 6 Test 7 Test 8 Test 3

PAGE 94

94 Figure 5 13. Results for the texture analyzer tensile test on channels 5x5 microns features. The graph shows tensile force (mN) versus elongation (mm) for 8 different tests. Figure 5 14. Results for the texture analyzer tensile test on channels 20x20 microns features. The graph shows tensile force (mN) versus elongation (mm) for 8 different tests. -400 -200 0 200 400 600 800 1000 1200 0 10 20 30 40 50 Tensile force (mN) Elongation (mm) Channels 5x5 Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 -200 0 200 400 600 800 1000 1200 1400 1600 0 5 10 15 20 25 30 35 40 Tensile force (mN) Elongation (mm) Channels 20X20 Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8

PAGE 95

95 Figure 5 15. Results for the texture analyzer tensile test on channels 20x20 microns features. The graph shows tensile force (mN) versus elongation (mm) for 8 different tests. Figure 5 16. Results for the texture analyzer tensile test on channels 100x100 microns features. The graph shows tensile force (mN) versus elongation (mm) for 8 different tests. -400 -200 0 200 400 600 800 1000 1200 1400 1600 0 5 10 15 20 25 30 35 40 Tensile force (mN) Elongation (mm) Channels 50x50 Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 -400 -200 0 200 400 600 800 1000 0 10 20 30 40 50 60 70 Tensile force (mN) Elongation (mm) Channels 100x100 Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8

PAGE 96

96 Figure 5 17. Results for the texture analyzer tensile tes t on channels 200x200 microns features. The graph shows tensile force (mN) versus elongation (mm) for 8 different tests. Figure 5 18. Histogram for the tensile force (mN) values of the different plaques for those tests on smooth PDMSe -400 -200 0 200 400 600 800 1000 0 5 10 15 20 25 30 35 Tensile force (mN) Elongation (mm) Channels 200x200 Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 0 5 10 15 20 25 30 0-5 10-15 20-25 30-35 40-45 50-55 60-65 70-75 80-85 90-95 100-105 110-115 120-125 130-135 140-145 150-155 160-165 170-175 180-185 190-195 Frequency Tensile force (mN) Histogram SMOOTH (Blank)

PAGE 97

97 Figure 5 19. Histogram for the tensile force (mN) values of the different plaques for those tests on 5x5 microns channels samples Figure 5 20. Histogram for the tensile force (mN) values of the different plaques for those tests on 20x20 microns channels samples 0 5 10 15 20 0-5 10-15 20-25 30-35 40-45 50-55 60-65 70-75 80-85 90-95 100-105 110-115 120-125 130-135 140-145 150-155 160-165 170-175 180-185 190-195 Frequency Tensile force (mN) Histogram Channels 5x5 0 5 10 15 20 25 0-5 10-15 20-25 30-35 40-45 50-55 60-65 70-75 80-85 90-95 100-105 110-115 120-125 130-135 140-145 150-155 160-165 170-175 180-185 190-195 Frequency Tensile force (mN) Histrogram Channels 20x20

PAGE 98

98 Figure 5 21. Histogram for the tensile force (mN) values of the different plaques for those tests on 50x50 microns channels samples Figure 5 22. Histogram for the tensile force (mN) values of the different plaques for those tests on 10 0x100 microns channels samples 0 5 10 15 20 25 0-5 10-15 20-25 30-35 40-45 50-55 60-65 70-75 80-85 90-95 100-105 110-115 120-125 130-135 140-145 150-155 160-165 170-175 180-185 190-195 Frequency Tensile force (mN) Histogram Channels 50x50 0 5 10 15 20 25 0-5 10-15 20-25 30-35 40-45 50-55 60-65 70-75 80-85 90-95 100-105 110-115 120-125 130-135 140-145 150-155 160-165 170-175 180-185 190-195 Frequency Tensile force (mN) Histogram Channels 100x100

PAGE 99

99 Figure 5 23. Histogram for the tensile force (mN) values of the different plaques for those tests on 200x200 microns channels samples 0 5 10 15 20 25 0-5 10-15 20-25 30-35 40-45 50-55 60-65 70-75 80-85 90-95 100-105 110-115 120-125 130-135 140-145 150-155 160-165 170-175 180-185 190-195 Frequency Force (mN) Histogram Channels 200x200

PAGE 100

100 Figure 5 24. Adhesive failures of mussel plaques in the different topographies. Those failures on 100 CH100x100 and 100 CH200x200 are more difficult to observe in the picture due to the scale of the topography (Photo taken by author)

PAGE 101

101 Table 5 9 Values of tensile force (mN) and Tensile stress ( Pa ) including basic statistical analysis for every sample tested in the texture analyzer Tensile force (mN) Tensile stress x10 3 (Pa ) SMOOTH Average 32.706 60.977 N 136 136 st. dev. 24.567 45.803 90%CI 3.465 6.460 SE 2.107 3.928 5X5 Average 42.368 78.231 n 177 177 st. dev. 31.014 57.266 90%CI 3.834 7.080 SE 2.331 4.304 20X20 Average 45.161 83.604 n 182 182 st. dev. 27.892 51.634 90%CI 3.401 6.295 SE 2.067 3.827 50X50 Average 36.425 71.291 n 131 131 st. dev. 26.339 51.551 90%CI 3.785 7.408 SE 2.301 4.504 100X100 Average 31.862 86.350 n 142 142 st. dev. 19.031 51.576 90%CI 2.627 7.119 SE 1.597 4.328 200X200 Average 31.732 90.779 n 137 137 st. dev. 19.623 56.138 90%CI 2.758 7.889 SE 1.677 4.796

PAGE 102

102 Figure 5 2 5 Tensile force (mN) for every sample (5x5, 20x20, 50x50, 100x100, 200x200, width x spacing in microns) and blank. Values are average s and the error bars correspond to standard error (SE). Table 5 1 0 ANOVA variance study for te nsile force values for each topography ANOVA Source of Variation SS df MS F P value F crit Between Groups 27207 5 5441.379652 8.305585654 1E 07 2.2241 Within Groups 588977 899 655.1470153 Total 616184 904 32.71 42.37 45.16 36.43 31.86 31.73 25 30 35 40 45 50 Tensile force (mN) Tensile force

PAGE 103

103 Table 5 1 1 Tukey post hoc analysis of tensile force values (being 1=smooth, 2=5x5, 3=20x20, 4=50x50, 5=100x100, 6=200x200)

PAGE 104

104 Figure 5 2 6 Comparison between data collected for tensile force (left axis) and plaque surface area (right axis). Values are average s and the error bars correspond to standard error (SE). 0.3 0.35 0.4 0.45 0.5 0.55 0.6 30 32 34 36 38 40 42 44 46 48 50 Plaque surface area (mm 2 ) Tensile force (mN) Tensile force vs Plaque surface area

PAGE 105

105 Table 5 1 2 Comparison with ANOVA statistical analysis (Minitab) between plaque surface area values of the 48 h and 72 h experiments One way ANOVA: 5CH5X5 (48 h), 5 CH5X5 (72 h) Source DF SS MS F P Factor 1 0.00166 0.00166 0.39 0.539 Error 16 0.0674 1 0.00421 Total 17 0.06907 S = 0.06491 R Sq = 2.41% R Sq(adj) = 0.00% One way ANOVA: 20CH20X20 (48 h), 20CH20X20 (72 h) Source DF SS MS F P Factor 1 0.00295 0.00295 0.31 0.586 Error 16 0.15266 0.00954 Total 17 0.15561 S = 0.09768 R Sq = 1.89% R Sq(adj) = 0.00% One way ANOVA: 50CH50X50 (48 h), 50CH50X50 (72 h) Source DF SS MS F P Factor 1 0.00001 0.00001 0.00 0.969 Error 16 0.07976 0.00498 Total 17 0.07977 S = 0.07060 R Sq = 0.01% R Sq(adj) = 0.00% One way ANOVA: 100CH100X100 (48 h), 100CH100X100 (72 h) Source DF SS MS F P Factor 1 0.0213 0.0213 1.56 0.230 Error 16 0.2192 0.0137 Total 17 0.2405 S = 0.1 171 R Sq = 8.86% R Sq(adj) = 3.17% One way ANOVA: 100CH200X200 (48 h), 100CH200X200 (72 h) Source DF SS MS F P Factor 1 0.0012 0.0012 0.10 0.756 Error 16 0.1924 0.0120 Total 17 0.1936 S = 0.1097 R Sq = 0.62% R Sq(adj) = 0.00%

PAGE 106

106 Figure 5 2 7 Tensile stress (x10 3 Pa) for every sample (5x5, 20x20, 50x50, 100x100, 200x200, width x spacing in microns) and blank applying roughness index Values are average s and the error bars correspond to standard error (SE). Figure 5 2 8 Comparison between data collected for tensile force (right axis) and tensile stress (left axis). Values are average s and the error bars correspond to standard error (SE) 60.98 156.46 167.21 142.58 172.70 136.17 55 75 95 115 135 155 175 195 Tensile stress x10 3 (Pa) Tensile stress (roughness index applied) 30 35 40 45 50 50.000 100.000 150.000 200.000 Tensile force (mN) Tensile stress x10 3 (Pa) Tensile stress vs Tensile force (Roughness index) Tensile stress x10-3 Pa Tensile force (mN)

PAGE 107

107 CHAPTER 6 CONCLUSIONS AND FUTURE W ORK Conclusions As was previously mentioned in this thesis the number of plaques does not represent a confident measurement for technical reasons However it was worth mention ing that the values, although rough, are considered valid estimations that can be compared with previous studies. Plaque surface area has been shown in this study to decrease with increasing size of the channels on the topography, since the adhesive spreads faster when the the adhesive was the shape of the plaques. The degree of alignment of the plaques with the topography is proportional to the size of th e features. The wider the spacing the easier it was for the adhesive to adva nce between ridges producing an elongation in the direction of the channels. Tensile force scal ed inversely proportional to the feature size. Since the only interaction measured w as the interaction between the adhesive and the PDMSe, the only effect that possibly produces those changes is the size of the topography. In this way, the topography proves to have a significant effect on mussel attachment, reducing the force of every in dividual plaque attachment by increasing the size of the features. Nonetheless the difference between the biggest topography and the smooth PDMSe is not significant. The tensile stress measured in this project scal ed directly proportional to the feature s ize. If we consider all the assumptions made, the value of the tensile stress for

PAGE 108

108 each topography should be constant To explain th e observed unexpected effect, some considerations need to be made. The plaque surface value measured in this experiment was that obtained by the planar projection of every plaque on the topography. This means that the only effective area considered to calculate stress was the planar projection when in reality an extra effective adhesion area must exist that was not consider ed but plays an important role in adhesion. In other words, topography creates a larger effective area (side of the ridges) where attachment happen s So the degree of penetration of the adhesive is big enough to produce an important effect. A good method to estimate the effect of the topography is the Engineering Roughness Index (ERI) but in this case the parameters of the equation are the same for most of the samples used in this study. Two concepts play an equally important role in this study: plaque surface area and degree of penetration, both of them affected by the size of the different topographies. First of all, the bigger the size of the topography the smaller the plaque surface area because the features act as obstacles in the mussel adhesive s preading and therefore decrease adhesion strength. The degree of penetration promotes adhesion strength by filling the space between the features. It could be that the adhesive undergoes a different degree of effective adhesion depending on the different t opographies Th e adhesive may create an effective adhesion to most of the small features by adapting to them but not achieve an extensive adhesion in t hose larger features ( Figure 6 1. ) Nevertheless, no degree of penetration measurements have been done in this study and this is only a hypothesis. Further study in this subject could help explain the observed behavior

PAGE 109

109 In general this study provid ed enough in formation to state that channel topographies (ratio 1:1) within the range of 5x5 microns and 50x50 microns enhance the adhesion providing extra surface to the adhesive where a ttachment happens, and that bigger topographies present values very similar to t hose found on smooth PDMSe. For this reason, based on this study it is recommended the use of patterns around 100 200 microns, in order to reduce attachment strength of the mussel to a surface. Future work There are many options to obtain a deeper understa nding of mussel adhesion and mussel adhesive spreading behavior in engineered topographies. Most of which will lead to further experimentation. Comparison with commercial adhesives If the same technique and method described in this study is repeated using different commercially available adhesives this could help to understand what makes the mussel attachment procedure so unique. This could be done easily by depositing a known volume of adhesive on top of different topographies and allowing the spreading t o occur naturally underwater. Another modification of this test could be to force the spreading of the adhesive by applying a specific load also in underwater conditions. Values of plaque surface area, NFF and tensile force could be compared with those ob tained in this study. Test other engineered topographies Further experimentation with different topograph y design would be beneficial to understand the response of the mussel with respect to the surface in the early stage. It could help us to prove if the mussel foot is sensitive to changes and if the mussel has

PAGE 110

110 the ability to spot and discriminate different sites on the topography. This study helps to understand which size enhance s or prevents the adhesion of the mussel. Measure the penetration degree In this study the penetration degree of the mussel adhesive has been reported qualitatively but no measurement technique has been applied. Microtomy was tried unsuccessfully in the early stage of the study Due to the nature of the PDMSe and the proteins in the plaque this method was abandoned Not a single sectional slice was obtained by this method because the process completely ruins the plaque attachment due to shear forces on the polymer. One of the techniques recommended is cryogenic microtomy. Temper atures below the Tg will ensure the polymer has a sufficient modulus to achieve clean section can be obtain ed Another possible way to observe and to measure the degree of penetration between the ridges is direct observation with a high resolution camera. It may be possible to obtain an image clear enough to quantify distance. Measure the exact volume of adhesive secreted by the mussel In order to measure the volume of the plaque all dimensions need to be taken into account and not only the planar projectio n. To get this information some techniques such as confocal microscopy or even atomic force microscopy (AFM) could be helpful. This would provide information about the height of the plaque B ut there is still some of the adhesive that also fill s the spac e between the channel walls that has to be measured. Analyze the surface where adhesive failure happens. This study assumed that all the mussel adhesive was removed when the plaque undergoes adhesive failure, In order to make this assumption more confidentl y a

PAGE 111

111 chemical analysis of the specific spot on the topography should take place to ensure the assumption is correct. The plaque detachment leaves an easily detected spot, even visually, so it is big enough to reduce the characterization surface to that spe cific area. Some of the techniques could be infrared analysis (FTIR) or X Ray Photoelectron Spectrometry (XPS).

PAGE 112

112 Figure 6 1. Scheme of the effect of the topography size on mussel plaque adhesion (adhesive attachment)

PAGE 113

113 LIST OF REFERENCES Ackerman, J. D., Cottrell, C. M., Ethier, C. R., Allen, D. G., & Spelt, J. K. (1996). Attachment strength of zebra mussels on natural, polymeric, and metallic materials. Journal of Environmental Engineering, 122 (2), 141 148. Ackerman, J. D., Ethier, C. R., Allen, D. G., & Spelt, J. K. (1992). Investigation of zebra mussel adhesion strength using rotating disks. Journal of Environmental Engineering, 118 (5), 708 724. Ackerman, J. D., Ethier, C. R., Spelt, J. K., Allen, D. G., & Cottrell, C. M. (1995). A wall jet to measure the attachment strength of zebra mussels. Canadian Journal of Fisheries and Aquatic Sciences, 52 (1), 126 135. Ackerman, J. D., Sim, B., Nichols, S. J., & Claudi, R. (1994). A revie w of the early life history of zebra mussels (Dreissena polymorpha): comparisons with marine bivalves. Canadian Journal of Zoology, 72 (7), 1169 1179. Alavi, M., Bttgenbach, S., Schumacher, A., & Wagner, H. J. (1992). Fabrication of microchannels by laser machining and anisotropic etching of silicon. Sensors and Actuators A: Physical, 32 (1), 299 302. Aldred, N., Ista, L., Callow, M., Callow, J., Lopez, G., & Clare, A. (2006). Mussel (Mytilus edulis) byssus deposition in response to variations in surface wet tability. Journal of The Royal Society Interface, 3 (6), 37 43. Aldred, N., Scardino, A., Cavaco, A., de Nys, R., & Clare, A. S. (2010). Attachment strength is a key factor in the selection of surfaces by barnacle cyprids (Balanus amphitrite) during settlem ent. Biofouling, 26 (3), 287 299. Baum, C., Meyer, W., Roessner, D., Siebers, D., & Fleischer, L. G. (2001). A zymogel enhances the self cleaning abilities of the skin of the pilot whale (Globicephala melas). Comparative Biochemistry and Physiology Part A: Molecular \ & Integrative Physiology, 130 (4), 835 847. Bayne, B. (1964). Primary and secondary settlement in Mytilus edulis L.(Mollusca). The Journal of Animal Ecology 513 523. Brennan, A. B., Baney, R. H., Carman, M. L., Estes, T. G., Feinberg, A. W., Wil son, L. H., et al. (2010, January). Surface topographies for non toxic bioadhesion control. Surface topographies for non toxic bioadhesion control Google Patents. Burkett, J. R., Wojtas, J. L., Cloud, J. L., & Wilker, J. J. (2009). A method for measuring the adhesion strength of marine mussels. The Journal of Adhesion, 85 (9), 601 615.

PAGE 114

114 Callow, J. A., & Callow, M. E. (2006). The Ulva spore adhesive system. In Biological adhesives (pp. 63 78). Springer. Carl, C., Poole, A. J., Williams, M. R., & de Nys, R. (2 012). Where to Settle Settlement Preferences of Mytilus galloprovincialis and Choice of Habitat at a Micro Spatial Scale. PloS one, 7 (12), e52358. Carl, C., Poole, A., Sexton, B., Glenn, F., Vucko, M., Williams, M., et al. (n.d.). Enhancing the settlement and attachment strength of pediveligers of Mytilus galloporovincialis by changing surface wettability and microtopography. Journal Title, 2012 1 27. Carl, C., Poole, A., Vucko, M., Williams, M., Whalan, S., & de Nys, R. (2012). Enhancing the efficacy of f ouling release coatings against fouling by Mytilus galloprovincialis using nanofillers. Biofouling, 28 (10), 1077 1091. Carman, M. L., Estes, T. G., Feinberg, A. W., Schumacher, J. F., Wilkerson, W., Wilson, L. H., et al. (2006). Engineered antifouling micr otopographies -correlating wettability with cell attachment. Biofouling, 22 (1), 11 21. Chiovitti, A., Dugdale, T. M., & Wetherbee, R. (2006). Diatom adhesives: molecular and mechanical properties. In Biological adhesives (pp. 79 103). Springer. Dunn, G., & Heath, J. (1976). A new hypothesis of contact guidance in tissue cells. Experimental cell research, 101 (1), 1 14. Efimenko, K., Finlay, J., Callow, M. E., Callow, J. A., & Genzer, J. (2009). Development and testing of hierarchically wrinkled coatings for marine antifouling. ACS applied materials \ & interfaces, 1 (5), 1031 1040. Farsad, N., & Sone, E. D. (2012). Zebra mussel adhesion: Structure of the byssal adhesive apparatus in the freshwater mussel, Dreissena polymorpha. Journal of Structural Biology, 177 (3), 613 620. Gantayet, A., Ohana, L., & Sone, E. D. (2013). Byssal proteins of the freshwater zebra mussel, Dreissena polymorpha. Biofouling, 29 (1), 77 85. Gibson, A. L. (2002). C hemical modificaton of silicone elastomer for controlled biological interactions on microtextured surfaces Ph.D. dissertation, University of Florida. Gilbert, T. W., & Sone, E. D. (2010). The byssus of the zebra mussel (Dreissena polymorpha): spatial variations in protein composition. Biofouling, 26 (7), 829 836. Hillborg, H., & Gedde, U. (1998). Hydrophobicity recovery of polydimethylsiloxane after exposure to corona discharges. Polymer, 39 (10), 1991 1998.

PAGE 115

115 Hoipkemeier Wilson, L., Schumacher, J. F., Carman, M. L., Gibson, A. L., Feinberg, A. W., Callow, M. E., et al. (2004 ). Antifouling potential of lubricious, micro engineered, PDMS elastomers against zoospores of the green fouling alga Ulva (Enteromorpha). Biofouling, 20 (1), 53 63. Holm, E. R., Kavanagh, C. J., Meyer, A. E., Wiebe, D., Nedved, B. T., Wendt, D., et al. (20 06). Interspecific variation in patterns of adhesion of marine fouling to silicone surfaces. Biofouling, 22 (4), 233 243. Homma, H., Kuroyagi, T., Izumi, K., Mirley, C., Ronzello, J., & Boggs, S. (2000). Evaluation of surface degradation of silicone rubber using gas chromatography/mass spectroscopy. Power Delivery, IEEE Transactions on, 15 (2), 796 803. Kamino, K. (2006). Barnacle underwater attachment. In Biological adhesives (pp. 145 166). Springer. Lee, B. P., Messersmith, P. B., Israelachvili, J. N., & Wa ite, J. H. (2011). Mussel inspired adhesives and coatings. Annual review of materials research, 41 99. Lee, H., Scherer, N. F., & Messersmith, P. B. (2006). Single molecule mechanics of mussel adhesion. Proceedings of the National Academy of Sciences, 103 (35), 12999 13003. Lin, Q., Gourdon, D., Sun, C., Holten Andersen, N., Anderson, T. H., Waite, J. H., et al. (2007). Adhesion mechanisms of the mussel foot proteins mfp 1 and mfp 3. Proceedings of the National Academy of Sciences, 104 (10), 3782 3786. Long, C. J., Schumacher, J. F., Robinson, P. A., Finlay, J. A., Callow, M. E., Callow, J. A., et al. (2010). A model that predicts the attachment behavior of Ulva linza zoospores on surface topography. Biofouling, 26 (4), 411 419. Ltters, J., Olthuis, W., Velti nk, P., & Bergveld, P. (1997). The mechanical properties of the rubber elastic polymer polydimethylsiloxane for sensor applications. Journal of Micromechanics and Microengineering, 7 (3), 145. Lu, Q., Danner, E., Waite, J. H., Israelachvili, J. N., Zeng, H. & Hwang, D. S. (2013). Adhesion of mussel foot proteins to different substrate surfaces. Journal of The Royal Society Interface, 10 (79). Marmur, A. (2013). Superhydrophobic and superhygrophobic surfaces: From understanding non wettability to design consi derations. Soft Matter Matos Prez, C. R., White, J. D., & Wilker, J. J. (2012). Polymer Composition and Substrate Influences on the Adhesive Bonding of a Biomimetic, Cross Linking Polymer. Journal of the American Chemical Society, 134 (22), 9498 9505.

PAGE 116

116 Mor ra, M., Occhiello, E., Marola, R., Garbassi, F., Humphrey, P., & Johnson, D. (1990). On the aging of oxygen plasma treated polydimethylsiloxane surfaces. Journal of Colloid and Interface Science, 137 (1), 11 24. Nicklisch, S. C., & Waite, J. H. (2012). Mini review: The role of redox in Dopa mediated marine adhesion. Biofouling, 28 (8), 865 877. Scardino, A. J., & de Nys, R. (2011). Mini review: Biomimetic models and bioinspired surfaces for fouling control. Biofouling, 27 (1), 73 86. Scardino, A., Guenther, J., & De Nys, R. (2008). Attachment point theory revisited: the fouling response to a microtextured matrix. Biofouling, 24 (1), 45 53. Scardino, A., Hudleston, D., Peng, Z., Paul, N., & De Nys, R. (2009). Biomimetic characterisation of key surface parameters for the development of fouling resistant materials. Biofouling, 25 (1), 83 93. Schultz, M., Bendick, J., Holm, E., & Hertel, W. (2011). Economic impact of biofouling on a naval surface ship. Biofouling, 27 (1), 87 98. Schumacher, J. F., Aldred, N., Callow, M. E., Finlay, J. A., Callow, J. A., Clare, A. S., et al. (2007). Species specific engineered antifouling topographies: correlations between the settlement of algal zoospores and barnacle cyprids. Biofouling, 23 (5), 307 317. Schuma cher, J. F., Carman, M. L., Estes, T. G., Feinberg, A. W., Wilson, L. H., Callow, M. E., et al. (2007). Engineered antifouling microtopographies -effect of feature size, geometry, and roughness on settlement of zoospores of the green alga Ulva. Biofouling, 23 (1), 55 62. Silverman, H. G., & Roberto, F. F. (2007). Understanding marine mussel adhesion. Marine Biotechnology, 9 (6), 661 681. Smith, A. M., & Callow, J. A. (2006). Biological adhesives. Springer Berlin. Smith, P. J., Owen, M. J., Holm, P., & Toskey, G. (1992). Surface studies of corona treated silicone rubber high voltage insulation. Electrical Insulation and Dielectric Phenomena, 1992. Annual Report. Conference on (pp. 829 836). Stevens, C. (1998). Environmental degradation pathways for the breakdown of polydimethylsiloxanes. Journal of inorganic biochemistry, 69 (3), 203 207. Tth, A., Bertti, I., Blazs, M., Bnhegyi, G., Bognar, A., & Szaplonczay, P. (1994). Oxidative damage an d recovery of silicone rubber surfaces. I. X ray photoelectron spectroscopic study. Journal of applied polymer science, 52 (9), 1293 1307.

PAGE 117

117 Vrolijk, N., Targett, N., Baier, R., & Meyer, A. (1990). Surface characterisation of two gorgonian coral species: impl ications for a natural antifouling defence. Biofouling, 2 (1), 39 54. Waite, J. (1987). Nature's underwater adhesive specialist. International Journal of Adhesion and Adhesives, 7 (1), 9 14. Wake, W. C., & Kinloch, A. (1977). Developments in adhesives. I. Ap plied Science Publishers. Wake, W., & Kinloch, A. (1981). Developments in adhesives 2. Applied Science. Wells, S., & Sytsma, M. (2009). A Review of the Use of Coatings to Mitigate Biofouling in Freshwater. Portland State University Center for Lakes and Res ervoirs Wiegemann, M. (2005). Adhesion in blue mussels (Mytilus edulis) and barnacles (genus Balanus): mechanisms and technical applications. Aquatic Sciences Research Across Boundaries, 67 (2), 166 176. Wilker, J. J. (2010). Marine bioinorganic materials: mussels pumping iron. Current opinion in chemical biology, 14 (2), 276 283. Yu, J., Wei, W., Danner, E., Ashley, R. K., Israelachvili, J. N., & Waite, J. H. (2011). Mussel protein adhesion depends on interprotein thiol mediated redox modulation. Nature che mical biology, 7 (9), 588 590.

PAGE 118

118 BIOGRAPHICAL SKETCH Francisco Castro Cara was born in Zurich Switzerland in 1986 to Encarnacion Cara Lo pez and Francisco Castro Castro. In order to be close to the rest of the family, Francisco moved to Churriana de la Vega, Spain at only two years old with his mother and father where he was raised. Soon after arriving to Spain, Francisco became a big broth er to Esteban Castro Cara. Francisco graduated from his hometown high school as valedictorian in 2004, where he made some of the best memories and closest friends. During his high school studies, Francisco chose to focus on science courses, sparking his i nterest in the field of engineering. Directly after high school, he was accepted to the University of The next six years were spent learning, both in and out of the classroom. He worked part time as a scien ce and math tutor in which he developed excellent organization and communication skills. The experience of sharing his knowledge and passion for science was also very rewarding. He explored his creative side by joining a theater group for a year and by sub mitting creative writing stories to various contests. His summers were spent working at his The last year at the University of Granada was spent working on a design project continuing his education and acquiring international experience, which lead him to the University of Flor ida. Getting there was not an easy path. After a long application process, Francisco moved to the other side of the ocean on August 12 th 2012, to begin the Materials Science and Engineering program at UF. Shortly after starting his first

PAGE 119

119 semester, he joine project based on biofouling. This work permitted him to write the present thesis. It also allowed him to meet and work alongside a great group of people who became not only colleagues but frie nds. Along with the help from Dr. Brennan, the group collaboration as a whole was a vital part in the success of this project. Francisco made an important decision last November, when he married his wife Andrea Castro Cara, who has played the most imp ortant part in helping him achieve his goals. He hopes that his knowledge in the chemical engineering and materials science field, combined with his research experience and determination will make him an excellent engineer at his future place of employment