<%BANNER%>

Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2012-12-31.

Permanent Link: http://ufdc.ufl.edu/UFE0042509/00001

Material Information

Title: Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2012-12-31.
Physical Description: Book
Language: english
Creator: Hsu, Shu-Hau
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

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

Notes

Statement of Responsibility: by Shu-Hau Hsu.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Sigmund, Wolfgang M.
Electronic Access: INACCESSIBLE UNTIL 2012-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042509:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042509/00001

Material Information

Title: Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2012-12-31.
Physical Description: Book
Language: english
Creator: Hsu, Shu-Hau
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

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

Notes

Statement of Responsibility: by Shu-Hau Hsu.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Sigmund, Wolfgang M.
Electronic Access: INACCESSIBLE UNTIL 2012-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042509:00001


This item has the following downloads:


Full Text

PAGE 1

1 BIOLOGICALLY INSPIRED HAIRY SURFACES FOR LIQUID REPELLENCY By SHU HAU HSU A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR O F PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Shu Hau Hsu

PAGE 3

3 To my lovely Mom and Dad

PAGE 4

4 ACKNOWLEDGMENTS I would like to first and foremost thank Dr. Wolfgang Sigmund who is not only just an adviser but a n incredibly energetic mentor and sc ientist. His vision, compassion, support, understanding and guidance help ed me through the entire work I would also like to thank my committee members, Drs. Moudgil, Baney, El Shall and C Y Wu for their constructive comments. Also I would like to thank Dr. Tonia Hsieh at Temple University for the extensive discussions on biological system. I would like to recognize the help of the staff of MAIC (Materials Analytical Instrument Center) and PERC (Particle Engineering Research Center) regarding the characte rization of the surface properties There are also a lot of students and friends who without their help I would not have finished this work. I would like to acknowledge all the past and current members in the group for assisting me in many ways d uring my work. I particularly thank former group member, Yi Chung Wang, for the preparation of the fluorocarbon coated samples. Special thanks would need to go to the people who had lived in Arbor (Howard, Ian, Kenneth, Sophia, Pei Ching and Ray) for shari ng the life in Gainesville. L ast but not least I am extremely grateful to my parents for their love and unselfish support th r ough out my study in United States. Without them this dissertation would have never been accomplished.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 14 ABSTRACT ................................ ................................ ................................ ................... 16 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 18 Water Repellent Interfaces: A Green Technology ................................ ................... 18 Superhydrophobic Surfaces in Nature ................................ ................................ .... 19 Research Objectives ................................ ................................ ............................... 21 2 PRINCIPLES OF LIQUID REPELLENT SURFACES ................................ ............. 24 Issues of Wetting on Solid Surfaces ................................ ................................ ....... 24 Surface Tension ................................ ................................ ............................... 24 Wetting on Ideal Smooth Surfaces ................................ ................................ ... 25 Wetting on Roughened Surfaces I: Wenzel Model ................................ ........... 26 Wetting on Roughened Surfaces II: Cassie Baxter Model ............................... 27 Transition Between Cassie Baxter and Wenzel State ................................ ...... 27 Contact Angle Hysteresis and Sliding Angle ................................ .................... 28 Water Repellent Surfaces from Nature ................................ ................................ ... 30 Plant Kingdom ................................ ................................ ................................ .. 31 Animal Kingdom ................................ ................................ ............................... 33 3 FABRICATION OF ARTIFICIAL HAIRY SURFACE S ................................ ............. 41 Experimental Work of Making Hairy Surfaces ................................ ......................... 41 Self Assembly of Hair Like Structure ................................ ................................ 42 Materials Selection ................................ ................................ ........................... 42 Experimental procedure ................................ ................................ ............. 42 Characterization ................................ ................................ ......................... 43 Making Hairy Surfaces via Moulding Techniques ................................ ............. 43 Casting with lithographed moulds ................................ .............................. 44 Casting with natural leaves ................................ ................................ ........ 45 Casting with commercial porous membrane ................................ .............. 45 Characterization of surface morphology ................................ .................... 47

PAGE 6

6 Characterization of thermal propert ies ................................ ....................... 47 Characterization of mechanical properties ................................ ................. 48 Experimental Results ................................ ................................ .............................. 48 Making Hairy Surface via Self Alignment of Carbon Nanotub es ...................... 48 Colloidal carbon nanotubes ................................ ................................ ........ 48 Results and discussion ................................ ................................ .............. 49 Making Hairy Surfaces via Mould ing Technique ................................ ............... 51 Casting with lithographed moulds ................................ .............................. 51 Casting with natural leaves ................................ ................................ ........ 52 Making Hairy Surfaces via Moulding with Commercial Membranes ................. 53 Membrane casting on elastomer ................................ ................................ 53 Membrane casting on thermoplastic polymers ................................ ........... 54 Issues of the current membrane casting ................................ .................... 59 General Discussion ................................ ................................ ................................ 62 The Artificial Hairy Surfaces ................................ ................................ ............. 62 The Scope of the Casting Process ................................ ................................ ... 63 Durability of the Cast Structure ................................ ................................ ......... 65 Summary ................................ ................................ ................................ ................ 66 4 WETTABILITY OF HAIRY SURFACES ................................ ................................ .. 92 Contact Angle Measureme nt ................................ ................................ .................. 92 Concerns of Contact Angle Measurement ................................ ........................ 93 Fitting Model of Contact Angle Measurement ................................ ................... 94 Experimental Work of Wettability Evaluation ................................ .......................... 95 Static Co ntact Angle Measurement ................................ ................................ .. 95 Contact Angle Hysteresis Measurement ................................ .......................... 96 Video Assessment of the Surface Hydrophobicity ................................ ............ 97 Surface Tension Determination ................................ ................................ ........ 97 Contact Area and Theoretical Contact Angle Interpretation ............................. 98 Wetting Property of Hairy Plants ................................ ................................ ............. 98 Observations on Hairy Leaves ................................ ................................ ......... 98 Droplets on Hairy Plants ................................ ................................ ................... 99 Wetting Property of Artificial Hairy Surfaces ................................ ......................... 101 Contact Angles of Cast PDMS Surface ................................ .......................... 101 Contact Angles of Cast Thermoplastic Surf ace ................................ .............. 102 Polypropylene substrate ................................ ................................ .......... 103 LDPE substrate ................................ ................................ ........................ 106 PVDF substrate ................................ ................................ ........................ 106 Perfectly Hydrophobic Response ................................ ................................ ... 106 Self Cleaning Ability ................................ ................................ ....................... 109 General Discussion ................................ ................................ ............................... 110 Summary ................................ ................................ ................................ .............. 114

PAGE 7

7 5 TOWARDS SUPEROLEOPHOBIC SURFACES ................................ .................. 134 Experimental Work of Plasma Treatment ................................ ............................. 134 Introduction ................................ ................................ ................................ ..... 134 Surface Treatment with Water Plasma ................................ ........................... 135 Surface Treatment with Fluorocarbon Plasma Deposition ............................. 136 Characterization of Plasma Treated Surface ................................ .................. 137 Surface morphology ................................ ................................ ................. 137 Surface chemistry ................................ ................................ .................... 137 Contact angle measurement ................................ ................................ .... 138 Experimental Results ................................ ................................ ............................ 138 Surface with Water Plasma Treatment ................................ ........................... 138 Surface with Fluorocarbon Coating ................................ ................................ 139 General Discussion ................................ ................................ ............................... 141 Summary ................................ ................................ ................................ .............. 143 6 CONCLUSIONS AND SUGGESTIONS FOR FUTURE WORK ............................ 153 Conclusions ................................ ................................ ................................ .......... 153 Suggestions for Future Work ................................ ................................ ................ 155 The Role of Elasticity on Water Repellency ................................ .................... 155 Quantitative Study of Self Cleaning Effect ................................ ...................... 155 LIST OF REFERENCES ................................ ................................ ............................. 157 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 165

PAGE 8

8 LIST OF TABLES Table page 3 1 List of materials for casting hairy surfaces in this study ................................ ...... 69 3 2 Glass transition temperature ( T g ) and melting point ( T m ) o f the thermoplastics used in this work.. ................................ ................................ ............................... 88 3 3 Mechanical properties of the thermoplastics used in mould casting. .................. 89 4 1 Measured s urface tensions of the liquids used in contact angle measurement. ................................ ................................ ................................ ... 118 4 2 Contact angles of all the cast thermoplastic surfaces and the theoretical contact angles calculated from Cassie Baxte r theory. ................................ ...... 122 5 1 Liquid drops on plasma treated 0.6 m peeled PP samples. ............................ 147 5 2 Contact angles of low surface tension liquids on cast PP surfaces before and after CF coatings. ................................ ................................ ............................. 151

PAGE 9

9 LIST OF FIGURES Figure page 1 1 SEM images of lotus leaf surface. ................................ ................................ ...... 23 1 2 SEM image of stem of Dicliptera Suberecta and stem of tom ato. ................................ ................................ ................................ ............... 23 2 1 Schematic diagram of l iquid molecules at the surface and t he surface tension .. 36 2 2 Wetting behavior of a liquid dropl et on solid surface and their mathematical models. ................................ ................................ ................................ ............... 37 2 3 Contact angle hysteresis is the angle difference between adv and rec ............. 38 2 4 Relationship between sliding angles and contact angle hysteresis .................... 38 2 5 Few examples of microscopic mophology of water repellent leaf surfaces. ....... 39 2 6 SEM micrograph s ................................ ........................... 39 2 7 ................................ ... 40 2 8 Hairs protrude from the leg of fish spider, and Plastron of fish spider, visible as a silver envelope around the body and legs. ................................ ................. 40 3 1 Schematic procedure of self alignment of functio nalized carbon nanotubes with polyelectrolytes. ................................ ................................ .......................... 67 3 2 The procedure of high aspect ratio surface structure by photolithography and e beam lithography. ................................ ................................ ............................ 68 3 3 Creating negative PDMS elastomer mould by using leaf Dicliptera Suberecta as a positive mould ................................ ................................ ............................. 69 3 4 Two commercial membranes AAO and PC, used for casting ............................ 70 3 5 The procedure of casting PDMS elastomer with polycarbonate and alumina membranes. ................................ ................................ ................................ ........ 71 3 6 The procedure of casting thermoplastics su bstrate with polycarbonate membrane. The membrane is removed by either direct peeling or dissolving. ... 72 3 7 T g is determined by the midterm point B from curve obtained by DSC analysis. ................................ ................................ ................................ ............. 73 3 8 Standard dog bone shaped samples for ultimate tensile test. ............................ 73

PAGE 10

10 3 9 A typical tensile stress strain curve and the represented proper ties for polymeric materials. ................................ ................................ ............................ 73 3 10 Zeta potential of functionalized and non functionalized carbon nanotubes showing the shift of the IEP (isoelectrical point). ................................ ................ 74 3 11 SEM picture of Si substrate after immersed into colloidal CNTs suspension ..... 74 3 12 The surface structure of cast PDMS elastomer from mould s developed by photolithography and by electron beam lithography ................................ ........... 75 3 13 SEM pictures showed the hairs Mantle. ................................ ................................ ................................ ................ 75 3 14 The surface morphology of cast PDMS elastomer by using Dicliptera Suberecta as master moulds. ................................ ................................ ............. 76 3 15 ult to be removed from the PDMS elastomer. ................................ ............................. 76 3 16 ................... 77 3 17 The i nterface of PDMS substrate cast by PC membrane ................................ ... 77 3 18 PDMS substrate cast with different pore size PC membrane. ............................ 78 3 19 The micro structure of PP substrate cast with AAO membrane ( =0.2 m) ......... 79 3 20 PP substrate cast with different pore size PC membrane. The membrane was dissolved after casting. ................................ ................................ ................ 80 3 21 Top viewed post structure of cast PP substrate after dissolving the membrane. ................................ ................................ ................................ ......... 81 3 22 SEM images of PP substrate cast with different pore size PC memb ranes. ....... 82 3 23 SEM images of LDPE substrate cast with different pore size PC membranes. .. 83 3 24 SEM images of PVDF substrate cast with different pore size PC membranes. .. 84 3 25 SEM images of PS and PMMA substrates cast with PC membrane. ................. 85 3 26 DSC analysis cu rve of PP substrate ................................ ................................ ... 86 3 27 DSC analysis curve of LDPE substrate ................................ .............................. 86 3 28 DSC analysis curve of PVDF substrate ................................ .............................. 87 3 29 DSC analysis curve of PS substrate ................................ ................................ ... 87

PAGE 11

11 3 30 DSC analysis curve of PMMA substrate ................................ ............................. 88 3 32 Comparsion between t he surface microstructure on PP substrate after peeling off the PC membrane and t abdomen. ................................ ................................ ................................ ........... 90 3 33 Images of Immersed regular PP substrate and PP substrate with artificial hairy structure. ................................ ................................ ................................ .... 90 3 34 SEM images of cast PP and PVDF substrate s before and after being rubbed by fingers. ................................ ................................ ................................ ........... 91 4 1 Contact angle measurement on a superhydrophobic surface with different diameters of syringe ................................ ................................ ......................... 116 4 2 Images of the same water droplet on a superhydrophobic surface under different fitting modes of the static contact angle ................................ .............. 116 4 3 The equipment set up of goiniometer. ................................ .............................. 117 4 4 The interface of Dr op Snake program. ................................ .............................. 117 4 5 Selected sequential images during contact/compression/release test on a superhydrophobic surface. ................................ ................................ ............... 118 4 6 Schematic picture of Wilhelmy plate method of measuring liquid surface tension. ................................ ................................ ................................ ............. 118 4 7 The fraction of contact area is estimated by adjusting the threshold of top view SEM pictures. ................................ ................................ ........................... 119 4 8 Water d roplet s on the leaf of ................................ ..................... 120 4 9 S tereo microscop ic Images of a droplet on the leaf of ............... 120 4 10 Variety of hair density of the leaves of .......................... 121 4 11 Morphology of different cast PDMS surfaces and their contact angles. ............ 121 4 12 The surface morphology of the membrane dissolved PP surfaces and their sessile drop images. ................................ ................................ ......................... 12 3 4 13 The surface morphology of the membrane peeled PP surfaces and their sessile drop images. ................................ ................................ ......................... 124 4 14 The surface morphology of the membrane peeled LDPE surfaces and their sessile drop images. ................................ ................................ ......................... 125

PAGE 12

12 4 15 The surface morphology of the membrane peeled PVDF surfaces and their sessile drop images. ................................ ................................ ......................... 126 4 16 The images of contact angle measurem ent acquiring from goniometer. (a) a steel ball and (b) a droplet on cast PP hairy surface. ................................ ....... 127 4 17 Selected images during contact/compression/release test on a perfectly hydrophobic hairy PP substrate. ................................ ................................ ....... 127 4 18 Selected images during contact/compression/release test on a non perfectly superhydrophobic PP substrate. ................................ ................................ ....... 128 4 19 Selected images during motion test on 0.6 m peeled hairy PP substrate. ....... 129 4 20 Schematic diagram of self cleaning effect (Lotus effect) ................................ .. 130 4 21 Water droplets on an uncast PP sheet contaminated by carbon powder. ........ 131 4 22 Water droplets on an cast hairy PP surface contaminated by dirt particles. ..... 132 4 23 Two configurations of models for water repellent hair pile. ............................... 133 4 24 The interaction between water droplet and artificial hairy surface ................... 133 5 1 Schematic diagram of the water plasma system used in this work. .................. 145 5 2 Photo of t he water plasma chamber used in this work ................................ .... 145 5 3 Schematic diagram of the STS deep reactive ion etch system. ........................ 146 5 4 Schematic diagram of the typical process of deep reacti ve ion etching ............ 146 5 5 T he morphology of hairy surface before and after Ar plasma ........................... 148 5 6 The surface morphology of CF coated PP s amples after casting process. ...... 148 5 7 The hairy structure before and after plasma deposition of CF layer in DRIE chamber. ................................ ................................ ................................ .......... 149 5 8 Contact angles of methanol water mixture on hairy surfaces coated with different plasma power. ................................ ................................ .................... 149 5 9 XPS results of CF coated surfaces under different plasma power. ................... 150 5 10 Low contact angle of nonpolar liquid (dodecane) on CF coated hairy PP surfaces. ................................ ................................ ................................ ........... 150 5 11 Low surface tension liquid (dodecane) on PVA and (b) pHEMA substrates ..... 152

PAGE 13

13 5 12 Schematic diagrams of interfacial force when a liquid with contact angle c on the non reentrant and reentrant structure ................................ ........................ 152

PAGE 14

14 LIST OF ABBRE VIATIONS AAO anodic aluminum oxide ASTM American Society for Testing and Materials CAD computer aided design CB Cassie Baxter CCD charge coupled device CF fluorocarbon DRIE deep reactive ion etching DSC d ifferential s canning c alorimetry HMDS h examethyl dis ilazane ICP induced coupled plasma IEP iso electrical point LDPE low density polyethylene MeOH methanol MWCNT multi walled carbon nanotube PC polycarbonate PDAC poly(diallyldimethylammonium chloride) PDMS poly dimethyl siloxane PMMA p olymethyl methacrylate PP polycarbonate PS polystyrene PVDF p olyvinylidene fluoride RF radio frequency RFGD radio frequency glow discharge SEM scanning electron microscope

PAGE 15

15 THF t etrahydrofuran XPS x ray photoelectron spectroscopy A LV contact area of liquid/vapor interface A SL con tact area of solid/liquid interface A SV contact area of solid/vapor interface f s area fraction of liquid/solid interface f v area fraction of liquid/vapor interface R f roughness factor T g glass transition temperature T m melting temperature surface tension c critical surface tension LV surface tension of liquid/vapor interface SL surface tension of solid/liquid interface SV surface tension of solid /vapor interface p linear fraction of contact line contact angle hysteresis c contact angle of liquid to solid w contact angle of Wenzel state CB contact angle of Cassie Baxter state adv advancing contact angle of liquid rec receding contact angle of liquid trans transition contact angle slide sliding angle of drops

PAGE 16

16 Abstract of Diss ertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy BIOLOGICALLY INSPIRED HAIRY SURFACES FOR LIQUID REPELLENCY By S hu H au H su December 2 010 Chair: Wolfgang M. Sigmund Major: Materials Science and Engineering Owing to remarkable features, such as self cleaning, anti biofouling and drag reduction, interest on rendering surfaces water repellent has significantly grown within this decade. At tempts on making surfaces superhydrophobic contact angle ( c >150 ) accompanied with only few degrees of roll off angle, have been extensively demonstrated through the mimicking of the surface chemistry and morphology of lotus leaves T his appealing phenomenon also exists on another structure from nature: surfaces comprising soft hairs. Although t he role of this piliferous integument has long been recognized for providing life, arthropods in particular, water repellency, the synthetic su perhydrophobic surfaces based on this structure are still very limited. In this study the goal was to develop a novel liquid repellent surface by mimicking the hair y exterior of species. The artificial hairy surfaces were prepared by means of pressurized membrane casting, in which thermoplastic sheets were forced to flow into porous membranes at elevated temperature. The shape d pillars on the membrane cast polypropylene substrate are particularly similar to the conformation of natural hairs.

PAGE 17

17 The princip le of this fabrication technique is relatively accessible and is expected to be compatible with large area fabrication of superhydrophobic interfaces. The artificial hairy surface features perfectly hydrophobic response where no contact angle hysteresis w as observed from video assessment Thus the artificial hairy surface of the current work appears to be the first report to have such extreme hydrophobicity with only structural modification from the original substrate. This ultralow adhesion to water dropl et is believed to be attributed to the hydrophobic methyl groups and the mechanical response of the artificial hairs. Liquid repellency of the hairy surfaces was further enhanced by coating with fluorocarbon ( CF ) layers via deep reactive ion etching (DRIE) The contact angle of water methanol mixture ( < 35.2 mN/m) was raised from 60 to around 140 The surface energy of coated samples, however, was still not low enough to repel non polar liquids. Moreover, the hairy structure is not favorable for maintaining the low surface tension liquid in Cassie B axter state.

PAGE 18

18 CHAPTER 1 INTRODUCTION Water Repellent Interfaces: A Green Technology Research on how liquids act on solid surfaces dates back to the early 19th century when P. S. Laplace and T. Young studied capillary action. 1 The study, dealing with the cohesion of water molecules and adhesion of those molecules to solid materials, uncovered some fundamental aspects of wetting on solid materials. In the modern study of this field, significant interest has been directed towards the control of the degree of surface wetting, especially in rendering surface liquid repellent. 2 6 Liquid droplets show low adhesion to these particular engineered surfaces and are easily to be shed. Some applications based on this similar response can be seen in our daily lives, suc h as non stick frying pans, rain repellent windshields, or waterproof fabrics, where less effort is needed to clean surfaces or keep them dry. In the last decade, extreme water repellency existing in some natural species has ttention. Their exteriors show a very high water contact angle ( c >150 ) with only a few degrees of contact angle hysteresis, which enables water droplets to form spherical beads and to be shaken away from the surfaces quickly. This superior water repellen t behavior is now generally recognized as Superhydrophobicity (where hydro and phobos stand for water and fear respectively in Attic Greek ) Some crucial functions for these species are provided through this property, such as the reduction of surface contamination from microorganisms, and the performance of locomotion on water or to sustain the impact from raindrops. Interests have not only increased concerning the science behind thi s unique phenomenon, but

PAGE 19

19 also concerning the design and fabrication of the superb anti wetting surfaces. A few remarkable applications of the superhydrophobic surface may include: Self cleaning surfaces : One of the most astonishing phenomena of the super hydrophobic surface is its self cleaning ability, on which water droplets pick up the dirt and remove it while rolling off from the surface. With less consumption of water as well as small required force for removing the dirt, the superhydrophobic surfaces effortlessly remain smudge free Anti biofouling surfaces: Surfaces showing superhydrophobicity may have potential to reduce the undesired adherence of the microorganisms, plants and algae to the surfaces. This function merely results from the architectur e of the surface, r ather than chemistry responses. 7 This is considered to be more environmentally friendly, since toxic coatings, such as tributyltin ( TBT ) and triphenyltin (TPT) f o r marine hulls may be avoided. 8 Drag force reduction: As a thin air layer is trapped on the supehydrophobic interface, the direct interaction between solid and fluid is diminished. The reduction of the drag force can be as high as 5 0 %. 9 A more energy effic ient watercraft could be designed by covering it with a water repellent surface. All these features are leading us to applications concerning the conservation of the natural environment and the reduction of resource consumption, which are two of the bigge st concerns this century. Along with other developments in surface science, superhydrophobic surfaces in turn provide a significant contribution to green technology. Superhydrophobic Surfaces in Nature Water repellent and hydrophobic properties of many pl ants and living creatures have been acknowledged for a long time. The most well known example is observed on the leaves of lotus ( Nelumbo Nucifera ), which beautifully demonstrates the self clean ability owing to its superior water repellency. 10,11 Despite emerging from muddy marshlands, the lotus flower has stood as a symbol of purity in Asian culture for ce nturies. Barthlott and Neinhuis 10 examined the surface morphology of water repellent leaves by us ing electron microscopy. Figure 1 1 shows these water r epellent leaves were found to have micrometer scale protrusions (~ 1 m) covered with nanometer sized

PAGE 20

20 epicuticular wax crystals (~20 nm). A water droplet on the lotus leaf typically shows a contact angle larger than 150 accompanying with only few degrees of the contact angle hysteresis Since then, many attempts have been made to imitate the now so called Lotus Effect combination of hydrophobic coatings and the surface roughness. The hydrophobic nature of the wax crystals as well as the multiple scaled roughn ess they generate on lotus leaf plays key roles in having strong water repellency. The mechanism, which heavily relies on chemistry, however turns out to have some restrictions. For instance, these so called superhydrophobic surfaces show remarkably decrea sed repellency to hot water (50 80 C ). 12 This may arise from the dissolving of the wax crystals while leaves getting in contact with hot water droplets. Another type of leaf showing this amazing water repellent property is found to be very different from vulgaris), Tomato or Dicliptera Suberecta comprising a plurality of flexible hairs Lotus effect droplets on their fuzzy leaves can be suspended by the trichomes, which effectively prevent the droplets from wetting the leaf. (Figure 1 2b, c, d) Although the leaf itself shows an excellent hydrophobicity, those trichomes were found to be hydrophilic. 13 This is somewhat surprising since based on a classical model a rough ened hydrophilic surface will tend to be even more hydrophilic 14 Which indicates making superhydro phobic surfaces out of hydrophilic materials is possible with an appropriate geometry of surface structur e. (b)

PAGE 21

21 In addition to the plants, this piliferous exterior plays a more crucial role for numerous arthropods by not only effectively protecting their bodies from getting wet, but by providing various functions for their living activities. These hairs protrude several micrometers from their cuticles, typically inclined at certain angles, with diameters in the micrometer to submicrometer range. These structures can resist the impact of raindrops, 15 allow locomotion on the surface of water, 16 or even trap a layer of air for respiration when submerged 17,18 L egs of w ater striders composed of numerous hair like setae with diameters in micro meter, have water contact angle above 160 which in turn provide a great supporting force and en able them to walk on water. 19 Research Objectives While the water repellent property of those hairy structures has been recognized for over 100 years, 20 unlike lotus effect, only a few studies have been reported on the creation of the superhydrophobic hairy surfaces. The main reason is making this high aspect ratio structure is relatively difficult by using top down technique. 21 Moreover, obtaining the unique arrangement of these natural hairs is also a challenge. Two studies were published demonstrating attempts to duplicate the hairy structure via a two step casting technique by using species as the template for the mo u ld. 22,23 Alt hough one report of a cast surface showed superhydrophobicity, only samples with small area could effectively be prepared. 23 Therefore, our major efforts foc us on creating artificial hairy surfaces and to study their wetting behavior in depth. The hypotheses of this study are set as: (1) The surface covered with hair like structure s can be made on some specific substrates at moderate size (2) S urface s with artificia l hair will feature superhydrophobicity.

PAGE 22

22 (3) S urface with artificial hair is able to repel low surface tension liquids without a further modification of their surface chemistry. The following specific objectives are set in this dissertation to verify the hypo theses. (1) Develop an effective method to fabricate surfaces covered with high aspect ratio flexible materials for scientific investigation. (2) Characterize these hairy structures, their hair surface density, shape and geometry as well as key material properties such as thermal, mechanical and surface chemical properties. (3) Examine the wetting properties, including contact angles and contact angle hysteresis for a variety of liquids ranging from water to hydrocarbons. (4) Modify the surface chemistry to explore the li mitation of the prepared hairy structures towards specific liquids and analyze data with current theories. (5) Provide hypotheses for explaining the liquid repellency of hairy surfaces. In order to provide a comprehensive overview of this research, a general r eview of the science concerning wetting behavior is presented in Chapter 2. The classic Wenzel and Cassie Baxter models together with other recent theoretical works are described. The main work of this research is given in the rest of the chapters. Each ch apter contains its own perspective, and it starts with the main experimental work, which is followed by a presentation and discussion of the results. Chapter 3 focuses on the making of the artificial hairy surfaces and the major protocol of fabrication of the liquid repellent surfaces is presented. The main characterization of the wetting properties of the hairy surfaces will be described in Chapter 4. In Chapter 5, making artificial hairy surfaces towards repelling organic liquids is described and discusse d. The culmination of these findings lead to a general conclusion in Chapter 6, in which the possibility of future work on this topic is suggested, since the research initiates another possibility to make surface s liquid repellent. The possible implication s are also presented in this chapter as well.

PAGE 23

23 Figure 1 1. SEM images of lotus leaf surface where it is covered with micron sized protrusions (a), and (b) submicron sized wax crystals. Figure 1 leaf densely covered with hairs mantle, (c) stem of Dicliptera Suberecta and (d) stem of tomato. ( a ) ( b ) ( a ) ( b ) ( c ) ( d )

PAGE 24

24 CHAPTER 2 PRINCIPLES OF LIQUID REPELLENT SURFACES The purpose of this chapter is to provide readers with essential background knowledge while moving towards the preparation of liquid repellent inter faces. Hence an overview of the theoretical and experimental work regarding liquid repellency is presented. The chapter begins with the fu ndamental aspects of wetting behavior, and advances with some mathematical models. Since this dissertation is inspired by some cuticle s of natural species, a n exploration of different water repellent surfaces discovered from nature is provided. The major s trateg ies and examples that make surfaces liquid repellent are briefly given at the end of this chapter. Issues of Wetting on Solid Surfaces Surface Tension Surface tension is a property of the surface of a liquid or solid. Surface tension is caused by th e imbalance of the attraction force at the surface where t he molecules are surrounded by a reduced number of similar atom s or molecules than those in the interior. (Figure 2 1a) This cause s liquid molecules at the surface to be in an energetically unfavorable state, and therefore the surface generates a force to minimize its free surface area. This force is usually characterized quantitatively as the surface tension ( ) in fluid system or free surface energy for solid materials, which is the energy required for creating a unit area of the surface at constant pressure and temperature. The units can be interpreted either as tension force per unit length of a line (e.g. N/m ) or as energy per unit surface area (e.g. J/m 2 ) at the surface (Figure 2 1b) The system energy of two solid and liquid phases will be lowered if they are in contact with each other. This is exp ressed by the Young Dupr equation : 24

PAGE 25

25 W SL = S V + L V SL (2 1) where W SL is the work of cohesion per unit area between two surfaces, S V and SL are the surface tension s of the solid against vapor and liquid, and LA is the surface tension of liquid against vapor Wetting on Ideal Smooth Surfaces Wetting o f a solid surface is usually characterized by contact angle ( c ), which is the angle where the three different phases meet one another (Figure 2 2 a) Th e value of the angle is determined by th ree surface tensions where the chemical potential i n the three p hases should be equal Considering a liquid drop on a solid surface, t he total energy of the system, E is expressed as E = L V (A L V + A S L SL A SL (2 2) where A L V and A SL are the contact areas of the liquid with the solid and vapor respectively. At the equ ilibrium state, dE = 0; therefore, L A (dA L A + dA SL ) = W SL dA SL (2 3) It is generally assumed that the gravitational potential energy can be neglected and the volume and pressure are constant The angle between the three phases can be given by geometrical considerations, which is dA L V /dA SL = cos c (2 4) Therefore the contact angle c can be obtained from the combination of Eqs. (2 1), (2 3) and (2 4). cos c = ( S V SL ) / L V (2.5) of contact angle s 1

PAGE 26

26 The wettability of a surface is commonly evaluated by the apparent liquid contact angle. If the value of the static contact angle is 0 c the liquid tends to spread on the surface. It is usually referred to as hydrophilic or oleophilic surface in terms of aqueous or oily liquid, respectively. I f the value of the contact angle is 90 < c the wet area tends to shrink It is then referred to as hydrophobic or oleophobic surface in terms of aqueous or oily liquid, respectively. Surfaces with the water contact angle between 150 and 180 are usually called superhydrophobic. Wetting on Roughened Surfaces I: Wenzel Model The equation is an oversimplified expression, and is only valid for ideally flat surf aces that are atomically smooth a nd chemically homogeneous C onversely, very few solids are atomically flat. Wetting on rough surfaces was first considered by Wenzel. 14 In the Wenzel state where the roughness grooves are completely filled with liquid (Figure 2 2b) the contact angle ( w ) can be described by cos w = R f [ ( S V SL ) / L V ] (2 5) where R f is the surface roughness factor, which is defined as R f = (Actual Area) / (Projection Area) (2 6) Combining eq. (2 cos w = R f cos c (2 7) Since the roug hness factor is always larger than unity in a practical situation, therefore it is obvious that the apparent angle on a roughened surface will become smaller if its intrinsic contact angle on a smooth surface is less than 90. The apparent contact angle wi ll be larger, if its intrinsic contact angle is larger than 90.

PAGE 27

27 Wetting on Roughened Surfaces II: Cassie Baxter Model The Wenzel regime is usually recognized as homogeneous wetting since the liquid completely penetrates into the grooves. While under som e circumstances, especially the increase of the surface roughness, vapor pockets may be trapped underneath the liquid yielding a composite interface. (Fig ure 2 2 c) This heterogeneous wetting is usually described by Cassie Baxter (CB) theory 2 5 from which t he apparent contact angle ( CB ) is given by equation (2 8) cos CB = f s cos c + f v cos v (2 8) where c is the intrinsic contact angle on the original smooth surface, and f s and f v are the area fractions of the solid and vapor on the surface, respectively. Since f s + f v =1, and v = 180 (This implies that a suspended liquid droplet in air is a perfect sphere.), equation (2 8) can be rewritten as follows: cos CB = f s (cos c +1) 1 (2 9) From equations 2 9, it can be found that droplet s will have a higher apparent contact angle if less area is being in contact with the solid substrate. The CB equation simply indicates the contact angle can be increased even when the intrinsic contact angle of a liquid on the original smooth surface is less than 90 Transition B etween Cassie Baxter and Wenzel State T he apparent contact angle for a rough surface can be given by the equation (2 7) and (2 9), depending on the solid liquid interface It has also been reported that the solid liquid contact mode may change from the CB state to the Wenzel s tate by applying a small pressure, evaporating some liquid, or adjusting the volume of the droplet. 26 2 8 A transition between these two wetting states can occur and there is a critical value of f s

PAGE 28

28 below which the CB regime exists and above which the Wenz el regime is thermo dynamically more stable 29 3 2 During the transition, the droplet will penetrate into the groove of structure, and the equation should be still valid before and after the transition. The threshold value trans between the two modes can t herefore be obtained by combing eq. (2 7) and (2 9) 33 cos trans = ( f s 1)/(R f f s ) (2 10) Hence i f the contact angle is lower than the threshold angle trans the trapped air pocket s underneath will be thermodynamicall y unfavorable and the Wenzel mode will be easily obtained. To have a droplet to agree with the CB mode l the solid substrate must be hydrophobic enough or trans must be as small as possible, because the underneath trapped air pockets will be only stable w hen contact angle above trans Contact Angle Hysteresis and Sliding Angle The fundamental aspects of static contact angles are described in the previous sections. In addition to static behavior the contact angle hysteresis, and the sliding behavior of li quid droplets are also dominated factors for the evaluation of the liquid repellency of a surface. The contact angle hysteresis is an important characteristic of a solid liquid interface which is induced by the roughness and chemical inhomogeneity of a su rface. Superhydrophobicity means not only a high contact angle, but also a low hysteresis of the contact angle. Contact angle hysteresis, hys defined by equation (2 11), hys = = adv rec (2 11) Where adv and rec are advancing and receding angles, respectively. For a given solid substrate, a range of static contact angle may be observed. T he a dvancing contact

PAGE 29

29 angle is the maxim um value of contact angle before the liquid solid contact line starts to advance, and r eceding angle is the minimum value of contact angle before the liquid solid contact line starts to retreat (Figure 2 3) Therefore, the measured contact angle lies anywh ere within the range adv > c > rec Advancing and receding angles are usually determined by sessile drop measurement where liquid is progressively pumped into or withdrawn from the droplet to constantly vary the contact angle and record their values when the contact line mig rates L ow contact hysteresis of superhydrophobic surface s is responsible for the self cleaning properties, which means that a water droplet can easily roll off the surface and pick up dust from the surface. 11, 3 4 ,3 5 The degree of rolling off behavior of a liquid droplet is measur ed by sliding angle, slide at which a liquid droplet begins to roll or slide down an inclined plate. The sliding angle can be directly measured by titling the solid substrate to have the value. It can also be obtained from the contact angle hysteresis by using equation (2 11 ): 3 6 mg(sin slide ) / w= LA (cos rec cos adv ) (2 11) where rec and adv are the advancing and receding contact angles, respectively, g is the gravitational acceleration and m and w are the mass and width of the droplet, respectively The equation simp ly denotes that the lower contact angle hysteresis (smaller difference between the advancing and receding contact angles) the more easily the drop will slide. T he force F slide required to initiate the sliding of a drop over a solid substrate can also giv en by t he contact angle hysteresis : 40 F slide = LA (cos rec cos adv ) (2 12)

PAGE 30

30 If contact angle hysteresis is small, then only small e xternal forces, like wind, gravity, or mechanical shaking can easily shed off the drops to have the surface dry Similar to static contact angles, contact angle hysteres is is also greatly dominated by t he surface roughness. 33 When the droplet is in Wenzel state, t he liquid droplet is mainly pinned on the substrate even it is tilted to a significant angle while in CB state the pinning barrier is more easily to be overcom e even if the substrate is only slightly tilted Therefore, increasing the surface roughness can be used to lower the sliding angle; u nfortunately, to date there is no simple expression for the contact angle hysteresis as a function of roughness. 4,5,94 T he contact mode can be switched continuously from Wenzel to C B states with increasing surface roughness. 37,39 The sliding behavior of a liquid droplet is also governed by the movement of the three phase contact lin e toward its sliding direction. 3 8 ,3 9 With th e same fraction of the solid phase low contact angle or low sliding angle is favorable for a short continuous contact line It is also found that water slides off more easily when moving in the direction parallel to the pillars. 39 Therefore, not only the surface roughness but also the surface geometrical structure need to be taken into account in order to prepare a n effective superhydro phobic surface where a high water contact angle and low sliding angle can be achieved. Water Repellent Surfaces from Na ture M any plant s and arthropods are known for having their exterior water resistant With the advance of microscopy in the last few decades, a variety of microstructures is found on these surfaces which vigorously stimulate the research on making superhyd rophobic surfaces. Here some remarkable water repellent surfaces of plants and animals are presented from which the beauty of nature is amazed.

PAGE 31

31 Plant Kingdom Lotus leaf (Nelumbo nucifera) has become the epitome of natural superhydro phobic surfaces. This strong water repellency leads Lotus leaf to another impressive ability, self cleaning, where leaves remain unsmudged even being immersed into muddy water. Thus, in oriental cultures lotus is long considered as a sacred symbol of purity Nowadays the abilit y of self cleaning and water repellency is termed effect cleaning of leaves are suggested as a mechanism to resist harmful micro organism bounding to the leaf surface, since water is usually required for the germination. T he relationship between surface roughness and wettability or particle deposition of the leaves ha s long been known; 4 1 however, s ystematically detailed investigation was not conducted until Neinhuis and Barthlott 10,11 studied more than 200 water repellent plants via scanning electron microscopy where s everal types of epidermal relief, a single layered group of cells covering a plant, and epicuticular wax crystals are reported T he epidermal relief is ranged from 5 in multipapillate cells to 100 in lar ge epidermal cells. However, t his variation in scale had a lmost no effect on wettability. 11 Overall, t he SEM study reveals that the water repellent leaf surface s are all covered by small protrusions (Figure 2 5a and b) called papillae, which are covered by an additional layer of ep icuticular waxes. (Figure 2 5c and d) The wax is present in crystalline tubules with water contact angles of about 95 110 which is considered hydrophobic. 4 2 The superhydrophobicity of leaves is obviously dominated by their surf ace roughness. The papillae create the first degree of roughness on which the wax crystals generate an additional roughness in a smaller length of scale. Th is kind of hierarch ical

PAGE 32

32 roughness on superhydrophobic surfaces seems to play a crucial role, but the detailed mechanism is not yet completely clear. Some explanations ha ve been proposed to provide a reason why most natural surfaces are hierarchical. 43 45 This arrangement may be generated for having a general purpose : to repel both macroscopic and microsc opic droplets. 4 6 4 9 Surfaces with only one scale of roughness repel led macroscopic droplets fairly well, while the condensation may easily form microscopic droplets between the grooves of the surface structure. 50 While it is known that self cleaning is mai nly attributed to the strong water repellency to quantify the ability of self cleaning is still a challenge Therefore, a further study is needed to quantitatively address how these two properties are related to each other. It is usually believed that the dirt particles on a smooth surface are partially redistributed when water droplet s slide away while on a rough surface the dirt particles easily adhere to the droplet interface and are removed away w hen the droplet s roll off. 11 The second type of surfa ces is the leaves covered with soft dense hairs ratio above 1 00 have a diameter around 10 m and a height up to around 1 mm (Figure 2 6a). As presented in Chapter 1 water droplet rests on the trichomes as a perfect sphere (Figure 1 2b), whose apparent contact angle may be taken as 180, without contacting to the leaf cuticle. The droplets also run off these leaves very easily, and consequently the leaves keep unwetted. Despite the difference of the morphology, the composition of these trichomes has some interesting features. Electron microscopy covered with cuticular wax crystals (Figure 2 6b), similar to Lotus leaf. However, the contact angle of the trichomes is

PAGE 33

33 shown to be below 60 13 which indicates the trichomes are well hydrophilic. This is unexpected from a surface comprised of hydrophili c dense hairs, since liquid is supposed to be sucked into the brush as the capillary suction. A model was proposed by Otten and Herminghaus 13 in which the elasticity of the trichomes as well as the hydrophobicity from cuticular wax crystals are believed to be the main reasons for having strong water repellency. As trichomes are more contact angle is different from 90, the liquid surface will deform around the hai r order to form a bundle, which costs elastic energy For a sample of infinite size, the combination of both contributions leads to a minimization of the total energy of th e system if the hairs group into bundles of a particular size. This particular size depends on the height and separation of the hairs, their mechanical modulus and on the distance h between the leaf surface and the drop/air interface. As a result a water d rop resting on such a bundle of hairs cannot come in contact with the substrate. Animal Kingdom As presented in the previous section, the wettability of leaves is dominated by both their surface chemicals and the topography. Similar to plants, the princip les also apply to living creatures whose cuticles are water repellent; however, this water repellency or water proofing provides more crucial functions for their living activity, such as to resist the impact of raindrops, to perform locomotion on the surfa ce of water, or even to trap a layer of air for respiration when submerged. Two general criteria for having water repellent surfaces: the resistance to fluid penetration and the resistance to adhesion by droplets. Thorpe and Crisp referred these two conce pts as water and rain proofing

PAGE 34

34 respectively for water repellent insects 5 1 B oth water and rain proofing are enhanced by rigid roughened exterior but the preferred morphologies are different for these two features. Higher solid liquid contact is favored for w ater proofing while minimizing this contact will benefit rain proofing as water droplets is more easily to be shaken off Similar to the water repellent leaves, i nsect cuticle is covered with a layer of epicuticular wax that renders the exterior hydr ophobic to ease the penetration of water. The presence of this waxy layer has long been known. 5 2 ,5 3 Holdgate 15 reported that t he wax layer had two way benefit. It is not only important in water repellency but also in water retention: removing the wax laye r of insects lead to their rapid desiccation. 5 4 The cuticle typically has numerous irregular structures such as hairs, scales and regions of varying curvature, all of which make the contact angle highly variable. 5 5 Here the attention is on the hairy struct ure, since the piliferous exterior is found in almost in all water repellent arthropods and has been noted for over 100 years. 20 The characteristics of the hair layer of these water repellent arthropods however have just been established within this 30 y ears via using scanning electron microscopy. The arthropods cuticle and its interaction with the free surface can be now qualitative elucidated with macrophotography. 5 6 Figure 2 7 shows st towards the leg, which discourages their piercing the water surface. The hairs are typically 30 m long, tapering to a point from a 1 m diameter base and inclined at an angle around 30 to the underlying leg surface, forming a layer 10 m thick. These hairs, termed macrotrichia, point in the direction of the leg tip, and bend inwards at their

PAGE 35

35 tips s o as to lie roughly tangent to the leg and water surface. The hair density varies along the leg: the density is 4 6 10 3 hairs mm 2 on the front tarsus, 1.2 1.6 10 4 mm 2 on the middle tarsus and 8 10 10 3 mm 2 on the hind tarsus. 5 6 Their non wetting pilifer ous legs have shown to have contact angles above 150 and to provide a maximal supporting force of 152 dyn e for each leg, which effectively enables water striders to stand and walk on water. 19 Another example is the fish spider (Argyroneta aquatic) which is known to submerge and fish underwater with a n trapped air layer that serves as an air supply ( Figure 2 8). 5 7 When submerged the water spider as well as many insects appears to have shiny coatings owing to a total reflection from a thin layer of air trap ped by their hair coating. The hair structure is thus necessary not only to lower the contact area with the water surface, but to enable these creatures to breathe when submerged. This activity is termed Plastron respiration as the hairs acting like a shie ld. 17 The oxygen and carbon dioxide are supplied and removed through the diffusion across the bubble surface from the ambient water. The structure of plastron hairs was extensively studied by H inton 18 which illustrated the air layer is maintained by an arr ay of hairs that their tips lie roughly parallel to the body. The air layer is then retained before the hairs collapsed due to the increase of water pressure. 17,18

PAGE 36

36 Figure 2 1 (a) Liquid molecules at the surface have fewer neighbors than interior molec ules, and are in a higher state of energy. Therefore the system tends to minimize the total free surface area. (b) The surface tension ( ) is a force per unit length of a film to resist the expansion of the surface area, or the work required to create a ne w unit surface area.

PAGE 37

37 (a) (b) (c) Figure 2 2 Wetting behavior of a liquid droplet on solid surface and their mathematical models. (a) A liquid droplet on a n ideall y Liquid droplet on a rough surface, Wenzel model, (c) Vapor pockets are trapped between the grooves and the liquid droplet, Cassie Baxter model.

PAGE 38

38 Figure 2 3 ( a) Advancing angle ( adv ) is the maximum contact angle of a droplet before its contact line (liquid solid interface) starts advancing. ( b) Receding angle ( rec ) is the minimum contact angle of a droplet before its contact line starts receding. Contact angle hysteresis is the angle difference between adv and rec Figure 2 4 Relationship between sliding angles and contact angle hysteresis (advancing and receding angles)

PAGE 39

39 Figure 2 5 Few examples of micromorphological mophology of wat er repellent leaf surfaces. Water repellent leaf surfaces of Nelumbo nucifera ( a ) and Lupinuspolyphyllos (b) Bars = 50 m Every epidermal cell forms a papilla and is superimposed by a dense layer of epicuticular waxes. An i ncreased roughness due to a diffe rentiation within the wax layer : on the leaf surfaces of Tropaeolum majus (c) and Melaleuca hypericifolia (d) Bars = 20 m. 11 (r ep ro duced by permission of Oxford University Press ) Figure 2 6 (a) SEM mic a spect ratio (a) (b) (a) (b) (c) (d)

PAGE 40

40 Figure 2 relative to the body surface. (b) The h discourages their piercing the water surface. Figure 2 8 (a) Hairs protrude from the leg of fish spider, which allow water spider to perform locomotion on water. (b) Plastron o f fish spider, visible as a silver envelope around the body and legs. 5 6 (r ep ro duced by permission of Elsevier Ltd ) (a) (b) (a) (b)

PAGE 41

41 CHAPTER 3 FABRICATION OF ARTIFICIAL HAIRY SURFACES This chapter focuses on the preparation of artificial hairy surfaces. Certain techni que s, including self assembly and moulding, that have been employed are presented here. The final objective is to obtain a practical approach that effectively fabricates artificial hairy surfaces which mimic the hairs from biological system and allow us to st udy their wetting properties. The chapter begins with a brief consideration on the strategy of making artificial hairy surfaces. All the experimental work is presented in the latter sections, although some of them are not used as the main process to make h airy surfaces. The challenge and the results, as well as the created morphology, are discussed in the second part of the chapter. Experimental Work of Making Hairy Surfaces The first major task in the study is to create an interface covered with hair like structure. The diameter of hairs on water repellent creatures or leaves is generally located at micrometer range while their length can be up to a millimeter, which means the target aspect ratio of the p illars could be as high as 100. 13 Making structure a t such high aspect ratio is always a great challenge in microfabrication. Moreover, direct machining on soft materials is another concern, so making hair like structure by means of top down route is not considered as an option. Hence two different approach es used here to develop the hairy structure are self assembly and moulding technique. Self assembly is a process in which the pre existing disordered substances spontaneously evolve into an ordered structure without applying an external guiding force. The moulding technique, which embosses the surfaces by forcing the materials into the

PAGE 42

42 moulds and commonly used for constructing surface features, is chosen as another major method in this study. Self Assembly of Hair Like Structure Materials S election Carbon nanotubes were selected to create the hair like structure in self assembly process. Arc discharge produced multi wall carbon nanotubes (MWCNTs) in the form of soot were ordered from Alfa Asear, which contains 30 40% of carbon nanotubes, carbon nanoparticle s, and fullerenes All the following materials were obtained from Aldrich and used as received: poly(diallyldimethylammonium chloride) (PDAC, 20 wt% in water, molecular weight 400,000 500,000), sulfuric acid (H 2 SO 4 98%), and nitric acid (HNO 3 70%). Diced Si wafer was used as the substrates for the self alignment process. Experimental procedure The carbon soot contains not only just carbon nanotubes but many other forms of carbon such as, carbon fibers, fullerenes and amorphous carbon. Since most of the impurities are carbon in nature, they can be removed by oxidization with strong acid. Acid treatment also functionalizes the carbon nanotubes in which the carboxylic groups are introduced onto the surface of carbon nanotubes which is essential for the foll owing self alignment coating process. The arc discharged carbon soot was initially ground with alumina molder and pestle to fine powder and then purified by oxidization to remove undesired impurities. The oxidization of ground carbon powder was performed b y sonification in nitric acid (HNO 3 70%) for 8 hours and then refluxed in an oil bath at 140 C for 24 hours The precipitate was rinsed with the deionized water three times and functionalized by sonication in a mixture of sulfuric acid (H 2 SO 4 98%) and ni tric acid (HNO 3 70%)

PAGE 43

43 (volume ratio 3:1) for 8 hours. The functionalized carbon nanotubes were then separated by centrifuging and washed with deionized water three times. After being dried at 100 C for 10 hours functionalized MWCNTs were dispersed by soni cated in deionized water to form a stable solution. PDAC was dissolved in deionized water at a concentration of 1 mg/mL with 0.01 M NaCl for layer by layer assembly. A Si wafer was diced into pieces and cleaned by sonification in acetone for 30 minutes and then rinsed with methanol and deionized water. A silicon substrate was immersed in PDAC solution for 10 min utes to form a polycationic surface layer. Functionalized MWCNTs were coated on the silicon substrate by immersing the substrates into the MWCNT so lution for 5 to 8 hours. These steps were repeated till desired layers were obtained, and then the substrate was rinsed with deionized water and dried in oven at 120 C The abstract of experimental procedure is given in Figure 3 1. Characterization The co lloid property of carbon nanotubes wa s characterized by Zeta potential. Dilute MWCNTs suspensions (0.01 wt%) at different pH were prepared by adding various amount of HCl and NaOH to the DI water, and their zeta potential was measured by ZetaPlus analyzer (Brookhaven Instruments Co., USA). The surface morphology of MWCNTs coated Si substrate was characterized by SEM (SEM JSM 6335, JEOL co.) with a thin layer of Pt Au conductive coating. Making Hairy Surfaces via Moulding Techniques A variety of template sou rce was considered and tested here, including those making from lithography, natural leaves and commercial porous membranes.

PAGE 44

44 Casting with lithographed moulds Two different irradiation source s, UV photons and electron beam, for lithography were tried here Figure 3 2 shows a general procedure of making high aspect ratio surface structure by means of lithography. Photolithgraphy Silicon wafers were used as the main substrate material here. They were cleaned by sonicaing in acetone and i sopropyl a lcohol for 5 10 minutes, and then dried by a nitrogen gas flow. Thick film photoresist (AZ 9260, MicroChemical GmbH) was coated onto the substrate by spinner coater and baked at 200 C for 30 minutes. The thickness, depending on the rotation speed, of the photoresist is around 10 m The following exposure was carried out at wavelength of 365 nm with intensity at 40 mW/cm 2 by Karl Suss MA 6 c ontact m ask a ligner After exposure, the surface feature was developed by immersing the substrate into the developer ( AZ 400K diluted with dei onized water in a 1:3 ratio, MicroChemical GmbH) for 5 minutes. A silicon elastomer PDMS (poly dimethyl siloxane Sylgard 184, Dow C orning Corp .) mixed with curing agent at 10 to 1 ratio was first degassed and then poured on the surface of the developed sil icon mould. After curing at 80 C for 2 hours, the silicon mould was separated from the elastomer leaving the transferred pattern on it. Electron beam l ithgraphy In order to create the patterns at submicron range, electron beam (e beam) lithography was als o use The preparation of substrates is similar to photolithgraphy, but the surface feature is directly written by an energetic electron beam on electron sensitive resist ( 950 PMMA Microchem Nano Inc.). The PMMA resist was spin coated onto a substrate at 3000 rpm for 30 seconds using a spin coater ( M odel p6700, Specialty Coating Systems). The thickness of the PMMA was

PAGE 45

45 ~500 nm. The substrate was then hot baked on a hot plate for 10 minutes at 180 C for curing the PMMA resist. The pattern writing was carried out at 30 KeV by a scanning electron microscope (SEM, Philips XL 40 FEG) equipped with NPGS system (Nano Pattern Generating System, JC Nabity Lithography Systems). The typical area dose for 2 at working distance 10 cm. After ex posure, the written pattern was developed by immersing the substrate in a 1 to 3 solution of methyl isobutylketone in isopropanol for about 75 seconds, and then rinsed by isopropanol before air dried. The deep reactive ion etching (DRIE) was presumed to be applied on the developed substrates to construct high aspect ratio wells as moulds for casting polymers. Casting with natural leaves Leaves from two different plants, Dicliptera Suberecta used here as the main positive molds. The elastomer PDMS (poly dimethyl siloxane Sylgard 184, Dow C orning corp. ) mixed with curing agent (10:1 ratio) was poured onto the leaves. In order to avoid a quick dehydration of the leaves, the cast elastomer was room temperature cured for at least 24 hours The leaves were gently peeled off from elastomer leaving it as negative mould for a subsequent casting. ( Figure 3 3) Before casting the second layer of elastomer, the moulds were treated with HMDS ( Hexamethyl disilazane, [(CH 3 ) 3 Si] 2 NH, Dow C orning ) vapor as a demoulding agent in the desiccators for 10 minutes. The second layer of elastomer was then poured on the negative moulds and cured at 80 C for 2 hours. Casting with commercial porous membrane Membranes Two type s of membranes ( F igure 3 2 ) were used a s here : (1) Anodic alumina membrane (Anopore, Whatman), pore size: 0.2 m

PAGE 46

46 (2) Track pore size= 0.6, 1.2 and 3.0 m Materials of substrates Thermosofting plastics were used as major substrate materials for membrane casting. T able 3 1 summaries the specification of the substr ates used in this study. The PS ( p olystyrene ) and PMMA (Polymethyl methacrylate) films were prepared by drying polymer solutions in which PS and PMMA granules were dissolved (at 15 wt%) in toluene and t etrahydrofuran (THF), respectively. Procedure of m emb rane c asting The liquid silicone elastomer and the curing agent were mixed at 10:1 ratio and degassed to prevent the formation of air bubbles. T he degassed silicone elastomer was then spun on a silicon substrate to form a uniform film where a PC or alumina membrane was placed over the coated film afterwards. After curing the film at 80 C for 2 hours, the alumina membrane was then removed by dissolving it in 45% KOH solution for 10 minutes while t he PC membrane was dissolved in d ichloromethane (CH 2 Cl 2 ) for 5 minutes. For other thermoplastic polymers, the plastic sheets w ere cut into 1.5 cm square and then sonicated in acetone and DI water for 5 min utes After dried in air, the substrate and a membrane were placed between two glass slides, where binder clips were applied to hold the pieces together. The casting process was then held by putting the sample set in a vacuum oven (vacuum pressure<1 kPa, VO914A, Lindberg/Blue M co.) at a certain temperature for 10 minutes The membranes were dissolved in proper solv ents as previously described. The PC membrane can be simply peel ed off by finger nails from the substrate. Figure s 3 5 and 3 6 show the general procedure of membrane casting for elastomer and thermoplastics.

PAGE 47

47 Characterization of surface morphology The surfa ce morphology was mainly characterized by scanning electron microscope with the electron beam from conventional tungsten filament (SEM JSM 6400, JEOL co. ) or cold field emission (JEOL 6335F JEOL co.) All of the cast samples were coated with a thin layer of Au Pd (thickness ~30 nm) to improve the conduc tivity prior to the examination, w hile for samples of natural species (leaves and striders), no conductive coating was deposited in order to reduce the dehydration during the coating process. The acceleratin g voltage for observing cast samples was at 10 15 KV, and for natural species the value was lowered to 5 10 KV to ease the surface discharging issue. Characterization of thermal propert ies The thermal properties of the thermoplastic substrates were measure d by d ifferential s canning c alorimetry (DSC Q1000, TA Instruments Inc.) in the Chemistry Department at University of Florida. Materials at around 10 mg were placed in the he rmetic alumina pans with lids. The heating profile was carried out as follows: the sample was first cooled down from room temperature to 50 C and steadily heated to 200 C at a rate of 10 C per minute. After holding temperature for 5 minutes, the sample was immediately quenched to 50 C and again heated to 200 C at the same ramp. T he gla ss transition temperature ( T g ) of samples w as determined as the middle point between the two intersection points by the three tangent lines on temperature transition slope. (Figure 3 7) The curve was analyzed by the software controlling DSC, from which the point of T g was also given. The melting point, T m is determined by the temperature at the maximum in the ( dH/dT ) plot near the transition. The onset of melting is taken as the initial rise of the curve above baseline. The maximum melting temperature at t his

PAGE 48

48 heating rate is taken as the final point which deviates from the baseline. All the analyses were conducted at a heating rate of 10C/min. Characterization of mechanical propert ies The mechanical strength of the substrates was measured by I nstron tensil e tester (Instron 1122, MTS System Co ) The test was performed at room temperature with the smallest 22 lb load cell. Specimens in dog bone shape (Figure 3 8) were first cut from plastic sheets by usin g a standard cutting die and then clamped onto the load cell. The tensile test protocol follow ed the standards ASTM D638. 59 The tensile test was performed at the speed at 1 (Figure 3 9) are determined from the controlling program Experimental Results Making Hairy Surface via Self Alignment of Carbon Nanotubes Colloidal carbon nanotubes MWCNTs were functionalized by chemical oxidation in which carboxylic groups were generated on the surface of carbon nanotubes. Figure 3 10 shows the zeta potential of functionalized and non functionalized MWCNTs at various pH values. Functionalized MWCNTs ha ve negative surface charge in different pH values with IEP (isoelectrical point) near pH=1 wher e IEP of non functionalized MWCNTs was around pH=5.5 which makes non functionalized MWCNTs unable to form a stable solution in water. Functionalized MWCNTs are well dispersed in water and ethanol without the assistance of surfactants. Carboxylic groups had a strong adhesion force with amine groups in deionized water (pH 6.5 ). 60 Some studies showed that carbon nanotubes can be aligned on substrates by electrical adsorption 6 1 ,6 2 This was the main reaction to form

PAGE 49

49 self aligned MWCNTs on the substrate where th e carboxylate anion groups of functionalized MWCNTs absorbed on the oppositely charged poly cationic polymer, PDAC, by Coulo mbic attractions. (Figure 3 1) After the immersion of PDAC coated Si substrates in colloidal MWCNT suspensions for 18 24 hours, the substrates showed no visible change The SEM image of 18hr immersed Si substrate (Figure 3 11) shows a bare, flat surface where only few pillar like structures were seen. Similar results were also discovered for those whose immersion time prolonged to 24 h ours. The dimensions of these pillars (rods) were relatively large, with the diameter around several hundred nanometers and the length up to few micrometers, indicating these were CNT bungles, formed due to van der Waals attraction between carbon nanotubes Although previous study conducted under the same procedure showed carbon nanotubes tended to be vertically absorbed onto substrate with the height of the layers ranges from 10 to 80 nm, 6 3 the current work shows that scanty CNTs were effectively absorbed onto the substrate coated with polycationic polymer. Results and discussion Using vertically aligned carbon nanotube arrays to create superhydrophobic surfaces is not rare. 12, 6 4 66 The carbon nanotube arrays, grown from chemical vapor deposition (CVD), c reate a rough surface where air can be trapped under these heterogeneous structures resulting in the Cassie Baxter (CB) regime. The areal density of CNTs on the CVD grown sample s can be as high as 10 m 2 Despite this density difference, the idea of using carbon nanotubes may conflict with the main objective at certain points. First of all, the purpose of this study is to have a surface covered with

PAGE 50

50 hair like structures that allows us to study the wet ting properties. The goal is to mimic the hairy integument of arthropods. The geometry of their hairs is all similar; they taper to a point from the cuticle with most tips curl ed towards their body, and the length is usually among several to tens of microm eters, sufficient to hold an air layer underneath the liquid. 5 6 The height of adsorption CNTs is tens of nanometers, which is much shorter than those hairs and may not be able to maintain the liquid interface in CB state. Moreover the flexibility of hairs on arthropods may play an important role while contacting liquid, although this is not fully understood. 17,5 6 The hair tends to buckle under Laplace pressure, which, in turn, eases the fluid impregnation. Nevertheless, the CNTs are considered the stiffest up to 1 TPa, which makes it impossible for them to have deflection while interacting with a liquid interface. 67 Last, but not least the CVD grown CNT arrays are not intrinsically superhydrophobic; they u sually require a coating with a low surface tension chemical compound (i.e. polytetrafluoroethylene PTFE ). 12,6 4 6 6 This indicates that the surface chemistry here dominates more on liquid repellency. This also shows that the functionalized CNTs of the curr ent study will need an extra modification since the surface carboxylic groups makes CNTs more hydrophilic, even if they were to be attached on the substrate. Overall, this technique originally intended to align CNTs vertically on the surface through a dire ct, simple self assembly process, but due to low adsorption of CNTs and other inherent issues, the experiment could not use this technique as the main fabrication method.

PAGE 51

51 Making Hairy Surfaces via Moulding Technique Casting with lithographe d moulds L ithogr aphy is by far the most successful technology in micro fabrication area, and has been used in semiconductor industry since the late of 50s 68 Typically, an irradiation source and a photosensitive polymer material are used to perform the pattern transfer. I t is referred as a top down approach where a pattern is transferred onto a substrate and followed by an etching process to construct the surface features Figure 3 12a shows an SEM image of the surface of a cast elastomer. The height of the surface structu re was around 10 m with the edge length of 10 m showing the geometry was able to be precisely duplicated from the moulds. One of the great advantage s of this technique is that the lateral structure can be well defined via pre designed pattern s from a photomask, while thi s may also restricts the freedom of having different dimensions. More sophisticated equipment as well as the operation skill is usually required here. Electron beam (E beam) lithography is considered as a maskless lithography to create a very small structu re, submicron to nanometer range, on an electron sensitive resist coated surface. The pattern can be designed by CAD software, and then transferred it to the pattern writing system. This makes e beam lithography a great tool in nanofabrication for research and development purpose. Figure 3 12 b shows a dot array of 1 m in diameter and 1 m in spacing created by e beam lithography. The deep reactive ion etching (DRIE) was planned to be applied to create high aspect ratio wells as a mould for casting polymers The major limitation of e beam lithography is throughput: an extremely long exposure time is required for making a larger area sample. Taking current experimental parameters, t he minimum time (t) required to make

PAGE 52

52 an array shown in Figure 3 6 on a 1 1 squ are inch substrate can be estimated by the following formula: Dose Exposed area = Beam current Exposed time 600 (mC/cm 2 ) 1. 61 (cm 2 ) = 0.4 (pA) t t = 1.9 10 6 sec 22 days It will require 22 days to finish writing this dot array, and this does not include time for the stage to move and possible time for beam corrections or adjustments. This is apparently not a practicable method for making a mould for casting hairy structure surfaces. Casting with natural leaves The surface features from natural species provided an alternate source of moulds. Using species, such as water striders, as master moulds for duplicating the hairy structure has been reported. O nly samples however, with small area (few square millimeters) could effectively be prepared so far, which makes the subsequent characteri zation very challenging. 22 Therefore the current work started with hairy leaves as moulds so that it may provide an acceptable sam ple size, and, more importantly, to avoid the involvement of living creatures. Figure 3 13a show s the surface structure of a Dicliptera Suberecta hairs (trichomes) are around 50 m wide with the height up to several hundred micrometers. Ru gged trichomes along with lower packing density made Dicliptera Suberecta a proper mould. The cast elastomer is shown in F igure 3 14a and b where the hairy structure was able to be duplicated from Dicliptera Suberecta On the other hand, L M a ntle was severely dehydrated after curing at oven for two hours

PAGE 53

53 (Figure 3 15), which made the leaf vulnerably trichomes with a diameter much smaller (~1 m ) and at a much higher areal density leading remained inside the elastomer while being peeled off from the leaves (Figure 3 1 6 a). Figure 3 1 6 b show s the pore density was significant ly lower (1.6 10 3 cm 2 vs 6.5 10 3 cm 2 ) than the negative mould. Smaller pore size also inhibits the elastomer from filling up the entire pores with only capillary action. All these make direct leaf casting an impractical way of making hairy structure. The wetting property is mainly studied on the leaf itself, rather than cast surface, and is presented in Chapter 4 Making Hairy Surfaces via Moulding with Commercial Membranes Membrane casting on elastomer Two commercial membranes, a nodic alumina (AAO) and track etched polycarbonate (PC), were applied as moulds for casting. Both membranes provide uniform, disconnected pores which are appropriate for creating the surface patterns by filling the pores. Membranes were first placed onto the silicone elastomer co ated substrates, and then dissolved by proper solvents. Figure 3 17 show s the surface of silicone elastomer after immersing into d ichloromethane The elastomer was not able to fill the pores ( = 0.6 m), and the surface was roughened after dissolving PC membrane. Similar results were found for using AAO membrane ( = 0.2 m), while the substrate remained smooth after etching process. Wetting difference between the elastomer and the membrane is the important factor that hinders the elastomer to flow into the pores The silicone elastomer is considered as hydrophobic ( c ~ 97 ), while PC and AAO membrane s are more hydrophilic ( c ~ 80 and ~ 40 for PC and AAO,

PAGE 54

54 respectively ) A n external assistan ce, rather than just by capillary action, is required to force the viscous elastomer to wet the membranes and fill the pores Alternatively, the elastomer was poured onto the PC membranes ( = 0.6, 1.2, and 3.0 m), and then cured in a vacuum oven. The mem branes were peeled off after the curing process Figure 3 18a to c show the elastomer surface cast with different pore size membranes. The surface feature becomes appreciable with the increase of the pore size, protruding from tiny spots ( = 0.6 m) to po sts ( =3.0 m). The height of the protruding posts is around 10 m, and they were more randomly oriented on the surface rather than vertically standing. The use of silicone elastomer on the fabrication of surface micro structure has become widespread in th e past decade. It is usually attributed to its chemical stability, non toxicity and a wide range of working temperature. Moreover the surface is easy to functionaliz e by silane based chemicals rendering surface hydrophobic or hydrophilic. The silicone ela stomer modulus below 1MPa) 69, 70 making the stability of the patterned structures as o ne of the main concerns especially the one with high aspect ratio where buckling and collapse of the struct ures may occur. Depending on the geometrical shape, diameter and the elasticity, the critical aspect ratio is usually under 10. 69 This makes silicone elastomer not to be a suitable material for the preparation of high aspect ratio hair like structure. Mem brane casting on thermoplastic polymers Thermoplastic is a polymer that can be melted to a liquid when heated and frozen to a solid, glass like state when cooled. The process is reversible, which makes thermoplastic recyclable. This also allows their surfa ces to be easily embossed with a

PAGE 55

55 particular pattern, since they can be remelted and remoulded to create surface features. The ultimate goal is always aiming to develop a method that not only makes hairy structures, but is also cost effective and time effic ient, so more effort can be devoted to the wetting study. The previous result on silicone elastomer denoted a promising way of creating surface pillars via membrane casting. Therefore, I applied a similar procedure to common thermoplastics so it could be e asily assessed on a daily basis. Polypropylene (PP) Polypropylene (PP) is a commercially available polymer and is used widely in a variety of applications due to its fair mechanical strength and superior resistance to many solvents. 7 1 The PP substrate in t he experiment was a general file jacket (No. 85781, SMEAD co.) purchased directly from Office Depot. The DSC analysis showed it had a melting temperature of 165 C ( Figure 3 2 6). The casting process was performed in a vacuum oven where PP substrates and mem branes were pressurized together between two glass slides, as described in the first section. The PP substrate was cast with an AAO membrane ( = 0.2 m ) at 190 C for 10 minutes; the AAO was then dissolved in KOH. Figure 3 19 shows the surface morphology after dissolving the membrane. The protruded structure formed a grass like surface where hundreds of submicron sized vertical fibers clumped together and curled randomly. The diameter of the fibers was i n good agreement with the pore size, but due to the high pore density (10 8 10 9 cm 2 ) the mean distance between pores was only ~50 nm, resulting in clumps of fibers rather than individual strands (Figure 3 19c). The substrate was also cast with polycarbonate (PC) membranes ( = 0.6, 1.2 and 3.0 m ). After the membrane was dissolved in d ichloromethane the microstructure

PAGE 56

56 of the cast surface was shown in Figure 3 20 For 3.0 and 1.2 m cast surfaces (Figures 3 19a and b), the protruded structure formed cylindrical posts with an average height of around 20 m The smooth base also indicated that the PP substrate did not react with the solvent. The pore of the PC membrane was made by the irradiation of high energy particles, so, unlike the AAO membrane, the pore distribution on th e PC membrane was disordered and not all perpendicular to the surface (Figure 3 4a). This made the cast structure randomly distributed with some of them inclined towards the surface (Figure 3 20a). The surface cast with the 0.6 m membrane (Figure 3 20c) s howed height variation at the posts, and some of the taller ones clumped together on the upper region. Close examination of top viewed pictures (Figure 3 21) revealed that the structure formed hollow cylinders as the diameter became smaller. The posts almo st formed tube structures for the 0.6 m cast surface (Figure 3 21c), although it was difficult to estimate the depth of the tube (or the height of the filled portion). More casting time is needed to fill up the pores with a higher aspect ratio. This is not a rare occurrence in the moulding pr ocess, as people use this casting technique to make nanotubes out of various materials. 72 ,7 3 The PC membrane could also be mechanically removed by simply peeling the membrane off from the substrate. The morphology of the peeled off structure heavily depen ded on the pore size shown in Figure 3 22 The posts on the 3.0 m cast surfaces were somewhat stretched, with the tips angled towards the direction in which they were peeled. The structures on the 1.2 m cast surfaces were much more disordered, where some of the posts were significantly elongated and randomly curled up to over 50 m As for the 0.6 m cast surface the surface structure was completely

PAGE 57

57 different from the membrane dissolved surface. After the peeling, the density of the remaining structure wa s about 7 10 5 cm 2 almost two order s of magnitude lower than the membrane (~4 10 7 cm 2 ) ; this indicate d that most of the posts were ripped off during the peeling process. The structure of the 0.6 m peeled surface featured a very unique geometry; it had a low areal packing density and curled structures with various leng ths and seemed to be randomly oriented. Their configuration appeared to be a de facto hair structure, similar to those in the natural biological system. More detail is presented in the section of general discussion. Low Density Polyethylene (LDPE) T he cas ting peeling process was applied on other thermoplastics to create the hair like surface structure LDPE is one of the most widely used thermoplastics for making containers, wrapping films, and, the most common, plastic bags. Compared to PP, the LDPE subst rate was softer, and had a lower melting temperature at 109 C (Figure 3 27). The LDPE substrate was cast with a PC membrane at 140 C for about 6 to 8 minutes, and the membrane was removed afterward by the peeling process only. The surfaces of the cast LDPE turned from translucent into opaque. The force needed to peel the membrane off from the LDPE substrate was less than that from the PP substrate. The SEM pictures in Figure 3 23 show that after the peeling process, the pore filled structures were greatly s tretched to several hundred micrometers all over the surface, much longer than the PP cast. The prolonged fibers were randomly oriented and entangled with one another, making the surface morphology more like a fiber mat. With the decrease in diameter, the fiber length was elongated more, but the overall surf ace structure remained similar.

PAGE 58

58 Polyvinylidene Fluoride (PVDF) PVDF is from the fluoropolymer family and has a relatively lower melting point. The one used in the experiment had a melting point of around 168 C (Figure 3 28). PVDF is commonly used in insulation, tubing, or high valued paint due to its flexibility, light weight, and resistance to chemicals. 74 Following the same casting process, the PVDF substrate was cast at 190 C with a PC membrane and the n peeled off. More force was generally needed for peeling the membrane off from the PVDF substrate. The microstructure of the cast PVDF surface was shown in Figure 3 24. For the 3.0 m cast surface, samples with adequate area were not able to be obtained, as the membrane was strongly stuck to the substrate. Only a very small portion of the membrane, mostly around the sample edge, could be peeled. Its peeled off structure (Figures 3 24 a and b), however, was very different from the LDPE or PP substrates. The diameters of the posts were not tapered towards the tip as those in the PP substrate. Instead, the heads became slightly larger than the body of the posts, making their shape similar to tulip This may be attributed to the neck forming while the posts were being pulled. Although the contact angle measurement was not able to be performed on the small sample area, this particular structure may have a n appealing property on adhesion. The ir shape appear ed to be like the s e t a es of a gecko foot which ha d more contact area and V an der Waal interaction with the surface than the fibers should have. 75 The structure of the 1.2 m cast surface was similar but more curled with a thinner fiber diame ter (Figure 3 2 4d ). For the 0.6 m cast surfaces, the posts were generally stretched into fibers but with a much shorter length than the LDPE. The fiber curled

PAGE 59

59 randomly and a large portion of them were ripped off during the peeling process (Figure 3 24f ). O ther Thermoplastics The casting was also tried on a few other common thermoplastics to exam the possibility of making hairy features. Polyethylene terephthalate (PET) was cast at 230 C (due to the limit of the vacuum oven and PC membranes), and the film sh owed no sign of casting (the surface remained transparent and glossy after removing the membrane). Polystryene (PS) and p olymethyl methacrylate (PMMA) films were prepared by dissolving the granules in proper solvents and then air drying them. The PS film w as later cast at 200 C while the PMMA was cast at a lower temperature, 160 C; membranes were peeled off for both substrates. As shown in Figures 3 25a and c, when cast with a 0.6 m membrane, both substrates had small bulges, rather than posts, stuck on the surface. For the 3.0 m cast surface (Figures 3 25b and d), the structure became more obvious, although the height was still much shorter when compared with the other thermoplastics. Issues of the current memb rane casting Substrate Materials During mouldin g, the polymeric substrates were partially melted or softened and then pressurized into the pores. This filling process wa s usually slow since t he polymer s need ed to flow long distances to fill up the cavities and, more importantly, because the polymers ha d a high viscosity. This wa s a consequence of the filling mechanism driven by the wetting properties of the polymer/ mould system 7 3 In this work, high aspect ratio surface structures were successfully made on several thermoplastic materials, including

PAGE 60

60 PP, LDPE and PVDF, although not all of them showed a vertical, pillar form structure. These materials had distinct melting points (Figures 3 26 to 28), and the processing temperatures were set higher than their melting points (Table 3 2) For semi crystalline polymers, the viscosity greatly dropped if the temperature was above the melting point. 76 The casting could therefore be effectively achieved within a few minutes at this elevated temperature, and the vacuum could also facilitate the filling process. On t he other hand, amorphous substrates, like PS and PMMA, hardly exhibited a sign of filling the pores even though the temperature was raised above the glass transition temperature to almost 80 C. The viscosity was not low enough for the materials to flow int o the pores. More importantly, the hair structure was greatly affected by plastic deformation while peeling off the membrane at ambient temperature. Table 3 3 summaries the mechanical properties of the substrate materials used for making hairy structure. T his plastic deformation, however, was not available for PS or PMMA as the ambient temperature was below 80% of the ir glass temperature s ( T g ) making t hem brittle and likely to fracture. 7 6 The pore filled structure would be easily broken during the peeling p rocess. Tensile tests on the PS and PMMA substrates could not be performed since the substrates were too brittle to be cut into dog bone shaped specimens with the standard cutting die. Restrictions The two types of commercial membranes used as moulds for casting, poly carbonate and anodic alumina, greatly reduced the time and complexity for the sample preparation. However, they also limited the control of the experiments, such as pore density, distribution, and thickness, especially for the PC membrane who se pores were

PAGE 61

61 not only randomly distributed, but o ccasional ly intersected with others. The range of the casting temperature was also restricted by the membrane. PC membranes, for example, were suggested by the manufacturer to be used below 200 C. The peeli ng process of the membrane prevented the ceramic AAO membrane from even being considered. Therefore it was quite challenging to test polymers with temperatures above 200 C, which included most of the thermoplastics in the fluorocarbon family, 77 whose low s urface energy may have been more preferable for creating a super hydrophobic or oleophobic surface. The majority of the membranes were removed by a direct peeling process without tools, but with fingernails. This was a very quick, effective, and zero cost process of removing the moulds. Some PC moulds could even be reused for another casting, although all the PC membranes curled severely into rolls after being peeled. As mentioned above, peeling at ambient temperatures hindered the use of thermoplastics wh ose T g were above room temperature. Additionally, it was very difficult to keep the process under a consistent condition, such as the speed and force of peeling. This may have had a great impact on the thermoplastic substrates, as the mechanical properties of polymers are strongly related to the strain rate. T he main goal of this study wa s not to reveal the detailed parameters of the process but to seek a feasible way of making the hairy structure. Therefore, the whole process as performed as consistently as possible throughout the entire fabrication. The later characterization was performed and discussed based on the surfaces created.

PAGE 62

62 General Discussion The Artificial Hairy Surfaces The surface structure varied with the thermoplastics after peeling the me mbrane. Pillars on the PP substrate easily formed hollow structures (tubes), especially with narrow pore membranes, due to its low surface tension and the ability to wet well 7 1 These tubes were not able to mechanically withstand the peeling force, and the refore the majority of them were ripped off with the membrane. The LDPE substrate had the lowest yield strength and could be stretched to several times the original under a limited pulling force; this made the LDPE surface turn into a fiber mat covered str ucture. PVDF, however, had the highest mechanical strength, and the pillars were not significantly stretched, with most of them still remaining on the surface after being peeled off the membrane. Among these surfaces, the membrane peeled ( =0.6 m ) structu re on the PP substrate specifically proved to be of importance. As shown previously in Figure 3 22f, the density of the remaining structure on this particular surface wa s almost two order s of magnitude lower than the membrane These remaining pillars were curled at various lengths and were to be oriented in a certain direction. A close up image in F igure 3 32 a reveal ed that the peeled structure of an individual post wa s similar to natural hairs. They taper ed to a point from a base with a pore size diameter while the length varie d from less than a micron to about 10 microns. Long er hairs we re tilted around 30 to 45 with their tips curled S ome hairs were also curled towards or parallel to the substrate. Interestingly, t his kind of arrangement happens to be easily found in many arthropods that have the capability of water walking or underwater breathing. 17,18,56 Figure 3 28c shows the microstructure of a water

PAGE 63

63 integument, where two different sets of hairs are easily seen The longer and shorter hai rs on their abdomen are around 30 m and 5 m long respectively. Longer hairs are inclined at a 30 angle with the tips roughly towards or parallel to the body surface Shorter hairs are usually thinner ( diameter~ 500 nm) and more randomly oriented. The d ensi ties of longer and shorter hairs are ~5 10 5 cm 2 and ~8 10 7 cm 2 respectively. While the density varies along the body and species, the overall geometry is similar to the PP cast surface. This particular structure enables the certain species to mainta in and control the air water interface adjacent to their skin, which is vital to them for daily activity For example, Figure 2 8b shows a thin layer of air trapped by a submerged fisher spider, enabling the spider to constantly perform respiration under w ater. Similar air trapping behaviors are also found on the artificial hairy surface as shown in Figure 3 33b. Moreover, the artificial hairy surfaces showed superior water repellency; water drops dripped on the surface immediately bounced back and then rol led off the surface effortlessly. Their hydrophobicity, along with other cast structures, is studied in the following chapter. The Scope of the Casting Process One of the main objectives in this work was to obtain an effective method to make a surface cove red with hair like structures. A great challenge was to make a surface feature with such a high aspect ratio (>10) with the diameter just a micrometer, not to mention that the ratio could also be in the hundreds for some natural hairs. Although high aspect ratio vertical posts or fibers, such as carbon nanotube forests, 6 4 ZnO, 78 or BN 83 nano fibers, have been well prepared via vapor phase condensation, it was not adopted here due to their complexity. Direct machining or the top down route was also

PAGE 64

64 not consid ered as a main approach. Despite the aspect ratio, as mentioned before, the shape and the conformation of the hairs needed to be arranged in a certain way. Therefore, even though there are plentiful works on mimicking natural superhydrophobic surfaces, rep orts specifically aimed at creating the structure of hairs are still very limited. All of the previous studies duplicated the hair structure from living species by a two step moulding process, similar to leaf casting. The first report came out in 2005, 22 i n which a 23 The hair structure was successfully duplicated from the master moulds. However, both of these works had the same defect, inad equate sample size, which resulted in difficulty for the characterization of wetting property The main manufacturing technique used in this research, moulding with commercial membranes, had been previously utilized for making arrays of pillars, 80 fibers 81 and nanotubes, 72 was not a novel approach. The technique that specifically involved polypropylene and the PC membrane was largely reported in research mimicking the gecko foot, 8 2 8 4 in which an array of millions of high aspect ratio microfibers was crea ted to imitate the gecko adhesive. Their strong adhesive and shear force are attributed to millions of micro spatulaes, which result in enormous Van der between this w ork and previous research is the way of removing the membrane. The membranes are normally dissolved in a proper solvent, which is CH 2 Cl 2 for PC membrane, to preserve the high areal density of fibers. In my experiment, I simply peeled the membrane off. Beca use of the force of pulling the membrane away, most of the material extruded into the pores was removed. The remaining columns were

PAGE 65

65 elongated and oriented with the peeling process, leaving the surface structure exactly as the microhairs discovered in many arthropods. T his work turned out to be the first to create the hair structure on polymeric substrates with a feasible sample area. Many methods have been proposed to render surfaces superhydrophobic, although very few of them have shown a promising way for large scale production. This current process is relatively simple and effective. It may be able to expand to the fabrication of large area superhydrophobic surfaces, since the surface morphology is only physically modified; it requires neither chemical re actions nor multiple steps. With the proper facilities, superhydrophobic sheets scaled up to several meters should be able to be prepared in a continuous process. A similar idea had been explored by Guo et al. 85 where a so was employed to imprint nanopillars on polycarbonate substrates with a tubular porous anodic alumina template. The generated pattern significantly increased the contact angle from 85 to 141 yet the contact angle hysteresis and sample size were reported Durability of the Cast Structure The casted surfaces are able to withstand the impact from fluid somewhat well, i.e. compressed air blowing or droplets striking. However, since this high aspect ratio structure was made by soft polymers, the mechanical s trength of the structure is not robust. The hairy structure may be damaged if the surface is abraded by solid matter and consequently loses their superhydrophobicity. Figure 3 34 shows the hair structure on PP and PVDF substrates w ere demolished after bei ng rubbed by fingers. While the durability is a general challenge among all superhydrophobic surfaces, publications

PAGE 66

66 which address this issue are rare. 86 ,87 Moreover, no such study has ever been carried out on superhydrophobic surfaces consisting of high as pect ratio structures. Summary To fulfill the first task, seeking a practicable method of creating surfaces covered with hair like structures, various possible methods were explored and presented. Two major techniques were employed: self assembling and mou ld casting. Self assembling was carried out by immersing polycationic polymer coated substrates into colloidal carbon nanotubes. Neither the experimental results nor the fundamental aspects supported the possibility of forming a vertical, self aligned surf ace structure. The effort then shifted to the moulding technique where different moulds and substrates were tested. Making moulds out of lithography was complicate d and extremely time consuming. Using of natural species as master moulds was also very chal lenging with the increase of aspect ratio. Therefore commercial membranes were chosen as the main moulds. This however limited the control of the experimental parameters. High aspect ratio vertical structures were prepared by casting thermoplastics with commercial membranes. A hairy surface was successfully achieved on polypropylene substrates ; after peeling off the PC membrane the surface was covered with shape d structure s analogous to the natural hairs on many arthropods. The experimental parameters were restricted by the membrane, oven and demoulding techniques here Additionally, th e durability of the cast structure wa s another concern. Nevertheless, the current process of making hairy structure s is straight forward and relatively accessible, and it has the possibility to be scaled up for the production of large area superhydrophobic surfaces. b )

PAGE 67

67 Figure 3 1. Sc hematic procedure of self alignment of functionalized carbon nanotubes with polyelectrolytes. Functionalized MWCNTs H 2 SO 4 +HNO 3 Sonication + Heat Si substrate Dip Coating PDAC Multiwall carbon nanotubes (MWCNTs) Poly(diallyldimethyl ammonium chloride) (PDAC) Self alignment of carbon nanotubes PDAC poly(diallyldimethyl ammonium chloride)

PAGE 68

68 (1) Spin coating of the photoresist (2) Exposure ( 3 ) Development (4 ) Etching (5 ) Casting of elastomer (6 ) Stripping of elastomer Figure 3 2 The procedure of high aspect ratio surface structure by photolithography and e beam lithography.

PAGE 69

69 Figure 3 3 (a) Creating negative PDMS elastomer mould by using leaf Dicliptera Suberecta as a positive mould, (b) Cured negative elastomer moulds after leaves were peeled off. Table 3 1. List of materials for casting hairy surfaces in this study Polymers Approximate Thickness ( m ) Surface Tension (20 C) 58 Sources LV mN/m) a c mN/m) b Silicone Elastomer 21.3 21.7 Sylgard 184 Dow Corning co rp. PP (Polyproplyene) 190 29.4 28.6 File jacket No. 85781, SMEAD co. PVDF (Polyvinylidene fluoride) 760 23.2 Kynar sheet West lake Chemical Inc. LDPE (Low density polyethylene ) 760 34.3 32.0 HIS 070335 G 01 Small Parts Inc. PET (Polyethylene terephthalate) 150 46 PES 19900 F 01 Small Parts Inc. PS (Polystyrene) 200 40.7 41.4 Lab prepared PMMA (Polymethyl methacrylate) 200 41.1 35.9 Lab prepared a Liquid surface tensions LV of solid polymers are from extrapolation from higher temperature studies of polymer melts. b. Zisman c ritical surface tension c is obtained from contact angle measurement with series of liquids of surface tension. (a) ( b)

PAGE 70

70 Figure 3 4 Two commercial membranes used for casting here (a) Track etched polycarbonate membranes. (b) Anodic alumina membrane, the bar is 1 m (b) (a )

PAGE 71

71 (1) Deposition of elastomer (2) Placing of the membranes (3) Dissolving of the membranes Figure 3 5 The procedu re of casting PDMS elastomer with polycarbonate and alumina membrane s

PAGE 72

72 Figure 3 6 The procedure of casting thermoplastics substrate with polycarbonate membrane. The membrane is removed by either direct peeling or disso lving. Thermoplastic substrate Polycarbonate membrane Heat and Pressure Peeling Dissolving

PAGE 73

73 Figure 3 7 T g is determined by the midterm point B from curve obtained by DSC analysis. Figure 3 8 Standard dog bone shaped samples for ultimate tensile test. Figure 3 9 A typical tensile stress strain curve and the represented properties for polymeric materials

PAGE 74

74 Figure 3 10 Zeta potential of functionalized and non functionalized carbon nanotubes showing the shift of the IEP (isoelectrical point). Figure 3 11 SEM picture of Si substrate after immersed into colloidal CNTs suspension for 18 hours. Two possible CNT bundles can be seen on the surface.

PAGE 75

75 Figure 3 12 (a) The surface structure of cast PDMS elastomer from moulds developed by photolithography, (b) dot array c reated by electron beam lithography on PMMA electron sensitive resist. Figure 3 13 SEM pictures showed the structures of hairs (trichomes) of two different leaves. (a) Dicliptera Suberecta (b) (a ) (b) (a )

PAGE 76

76 Figure 3 1 4 (a) The surface morphology of cast PDMS elastomer by using Dicliptera Suberecta as master moulds, (b) a close up view. Figure 3 15 to be removed from the PDMS elastomer. (b) (a )

PAGE 77

77 Figure 3 16 where the packing density of pores were significant lower than the original hair density, (b) leaf hairs were ripped off and remained inside the PDMS moulds. Figure 3 17 ( a) The interface of PDMS substrate cast by PC membrane (pore size= 0.6 m), where the membrane was partially dissolved in CH 2 Cl 2 (b) PDMS surface was roughened after dissolving PC membrane. a ) b ) (a) (a) (b) PC membrane PDMS substrate (b) (a )

PAGE 78

78 Figure 3 18 PDMS substrate cast with different pore size PC membrane. The membrane was removed by peeling. (a) 0.6 m (b) 1.2 m (c) 3.0 m and (d) a close view of the 3.0 m posts. ( a) (b) (d ) ( c )

PAGE 79

79 Figure 3 19 The microstructure of PP substrate cast with AAO membrane ( =0.2 m). (a) and (b) High areal density causing hundreds of submicron sized fibers clumped together and curle d randomly. (c) Tube like structure can be seen. (b) (c) ( a )

PAGE 80

80 Figure 3 20 PP substrate cast with different pore size PC membrane. The membrane was dissolved after casting. (a) and (b) =3.0 m (c) and (d) =1.2 m (e) and (f) =0.6 m ( a) ( f ) (c) (b ) ( d ) ( e )

PAGE 81

81 Figur e 3 21. Top viewed post structure of cast PP substrate after dissolving the membrane. (a) =3.0 m (b) =1.2 m and (c) =0.6 m. The tube structure is formed with the decrease of pore size. (b ) (c) ( a )

PAGE 82

82 Figure 3 22 PP substrate cast with different pore size PC membranes. The membrane was peeled off after casting. (a) and (b) =3.0 m (c) and (d) =1.2 m (e) and (f) =0.6 m ( a) (b) ( e ) (d ) ( f ) ( c )

PAGE 83

83 Figure 3 23. LDPE substrate cast with different pore size PC membranes. The membrane was dissolved after casti ng. (a) and (b) =3.0 m (c) and (d) =1.2 m (e) and (f) =0.6 m ( e ) (d ) (c ) (f ) ( a) (b) ( c ) (d ) ( e ) (f )

PAGE 84

84 Figure 3 2 4 PVDF substrate cast with different pore size PC membranes. The membrane was dissolved after casting. (a) and (b) =3.0 m (c) and (d) =1.2 m (e) and (f) =0 .6 m ( a) (b) ( c ) (d ) ( e ) (f )

PAGE 85

85 Figure 3 2 5 (a) and (b) are PS substrates cast with =0.6 m and 1.2 m PC membrane, respectly. (c) and (d) are PMMA substrates cast with =0.6 m and 1.2 m PC membrane, respectly. ( a) (b) ( c ) (d )

PAGE 86

86 Figure 3 2 6 DSC analysis curve of PP substrate shows a melting temperature at around 168 C. Figure 3 2 7 DSC analysis curve of LDPE substrate shows a melting temperature at around 109 C. Polypropylene Polyethylene

PAGE 87

87 Figure 3 2 8 DSC analysis curve of PVDF substrate shows a melting temperature at around 165 C. Figure 3 2 9 DS C analysis curve of PS substrate shows a glass transitional range between 103 C and 107 C. PVDF Polystyrene

PAGE 88

88 Figure 3 30 DSC analysis curve of P MMA substrate shows a glass transitional range between 1 08 C and 1 1 7 C. Table 3 2. Casting temperature, g lass transition temperature ( T g ) and melting point ( T m ) of the thermo plastic substrates used in this work. The data was measured and given by DSC. Thermoplastics H air Like Structure Casting Temperature ( C) Glass Transition Temperature ( T g C) Melting Temperature ( T m C) PP (Polyproplyene) yes 140 168.1 LDPE (Low density polyethylene ) yes 190 108.9 PVDF (Polyvinylidene fluoride) yes 190 164.2 PS (Polystyrene) N/A 200 106.6 PMMA (Polymethyl methacrylate) N/A 160 112.7 PMMA

PAGE 89

89 Figur e 3 31 Stress strain curve of the thermoplastics used in mould casting. Table 3 3. Mechanical properties of the thermoplastics used in mould casting. Thermoplastics modulus (GPa) Yield strength (MPa) Tensile strength (Mpa) Elongaton at break (% ) PP (Polyproplyene) 0.62 34.0 15.8 630 LDPE (Low density polyethylene) 0.14 6.1 51.7 485 PVDF (Polyvinylidene fluoride) 1.23 51.7 46.1 180

PAGE 90

90 Figure 3 32 (a) The surface microstructure on PP substrate after peeling off the PC membrane. (b) The m Note that there are two different sets of hairs on water strider. Longer hair s are bent inwards towards the body which hinder the piercing the water surface. Figure 3 33 (a) Immersed PP sheet without any p articular surface structure. (b) Immersed PP sheet with artificial hairy structure on it. The silver look comes from the total reflection from thin air layer trapped on the substrate surface. ( a) ( a ) (b ) (b)

PAGE 91

91 Figure 3 34 (a) and (c) are microstructure of cast PP ( =0.6 m) and PVDF ( =1.2 m) substrate. (b) and (d) are the microstructure after being rubbed by fingers ( a) (b) ( c ) (d )

PAGE 92

92 CHAPTER 4 WETTABILITY OF HAIRY SURFACES The central property wettability or hydrophobicity of different surfaces including natural and artif icial hairy structure, is presented in this chapter. The wettability was evaluated by contact angle measurement via image based technique. Due to superior water repellency on some of the artificial surfaces an alternative video assessment was also used he re. T he concerns and experimental work regarding the determination of contact angles are given in the first section The results of the surfaces created in this work are described in the following section, where natural hairy surfaces are also included. Th e final part is to associate the wetting properties with the surface structure, and to draw a conclusion towards making optimal superhydrophobic interfaces with hairy structure. Contact Angle Measurement The contact angle is the angle at which three diffe rent phases meet one another and is not limited to liquid/vapor on solid surfaces. The case people are specifically interested in is how the liquid droplets act on solid surfaces. Hydrophobicity is often characterized by the static contact angle measuremen t of water. The general principles are to assume: 88 (1) The drop is symmetric about a central vertical axis, which indicates that the drop shape is irrelevant from which direction it is viewed. (2) The drop shape is only determined by the balance of interfacial fo rces and the gravity. Their viscosity or inertia of the liquid does not shape the drop. Typically, contact angle measurements are carried out by acquiring images of resting drops on a surface through a specific apparatus, a goniometer. The image is

PAGE 93

93 then an alyzed by fitting a mathematical expression to the shape of the drop where the slope of the tangent to the drop at the liquid solid vapor interface line will be calculated. Concerns of Contact Angle Measurement The concept of interpreting contact angles h as been developed over 150 years; 1 however, a recognized standard protocol of measuring contact angles is not yet established, although contact angle is widely measured and used in academia and industry nowadays. Therefore, even within the same system, the value of angles may vary if the measurements are performed differently. The purpose of this section is to present a few general concerns in contact angle measurement from the practical perceptive, especially for samples showing extreme hydrophobicity, and in addition, to specify some details of the measurements taken. (1) Drop size: If the drop is too large, the shape will be distorted because of the gravity. Too small of a drop, however, will be difficult to place on the surface, especially for the surfaces showing superior hydrophobicity. Small drops may also cause a state transition from the CB state to Wenzel state, 29 leading to an entirely false angle. There is no definitive volume of drops for contact angle measurement. The value is generally below 10 l for each drop, and most reports have their volume around 5 l Using the same volume of drops throughout the entire measurement is essential. (2) Drop placement: All superhydrophobicity is located in the meta stable Cassie Baxter regime, which means it may tu rn to a more stable Wenzel state if an external force applied. Therefore, the drops should be placed on the surface with as small a force as possible. In other words, rather than just falling upon the surface, the drops should touch the surface by either r aising the sample stage or lowering the syringe until the drop touches the sample surface. This will minimize the kinetic energy as well as the drop spreading before taking the images. (3) Syringe size: Ideally, the diameter of the syringe tip should be infini tesimal when compared with the drop size. For a surface with moderate hydrophobicity, this may not be a major issue, since the syringe is not in contact with the drop during the measurement. However, when dealing with an extremely hydrophobic surface, due to the low adhesion between the surface and drops, the drops will stay on the tip of the syringe resulting in a possible distortion of the drop shape (Figure 4 1a). Thus, for the experiment, the plastic tip of the syringe equipped on the goniometer

PAGE 94

94 was rep laced with a 30 gauge needle (Figure 4 1b, OD: 0.260 mm, ID: 0.159 mm) to reduce the influence. (4) Image quality: Acquiring a clear image is probably the most crucial part, since the modern contact angle determination merely relies on mathematical analysis of the enough to have a clear, distinguishable profile, especially for the bottom portion where the drop is in contact with the substrate. This is usually a challenge when dealing w ith the superhydrophobic surfaces ( c >150 ) since the drop has a very limited contact area with the substrate. An auxiliary light source is sometimes needed to obtain a clear image. Constantly adjusting the stage, focus, aperture, and lighting from sample to sample is always necessary for optimizing image quality. (5) Fitting model: As stated above, the contact angle was analyzed by fitting it to a mathematical model, and consequently, the value significantly varies with the model applied. Unfortunately, not ev ery previous work specified the fitting model used when reporting their results. Zhang et al. 4 reported that even for the same image, the angle value could vary from 150 to almost 180 if different fitting models were applied. (Figure 4 2) This variation is related to the mathematical expression and also to the deformation of the water droplet caused by gravity. Therefore, in dealing with superhydrophobic surfaces, the value of contact angles above 150 may not exactly reflect the real situation of surface wetting, if the fitting mode is not clearly mentioned. All of the above are important issues that should be taken into consideration in contact angle measurement. The key is to perform the measurement under the same condition, and report it clearly. Sinc e no standard protocol to date has been built for the measurement, a direct comparison between different values may not be sound without specifying the detailed setting. Fitting Model of Contact Angle Measurement Various methods of drop shape fitting exist but the commercial software designated for contact angle analysis is still very limited. T hey are often accompanied with particular facilities and may not be reliable if used under other conditions. Because of this, the d irect measurement of a drop by us ing a goniometer with a telescope d protractor is still used nowadays, even with a major drawback coming from the

PAGE 95

95 subjectivity. The fitting method used in this experiment was based on a so called B spline snakes that was developed by Stalder et al. 89 for me asuring high accuracy contact angles. Snakes were originally defined as a spline energy minimization under internal and external forces 9 0 and are now widely used in computer assisted tools for segmentation such as medical image analysis or feature tracki ng in video sequences. The whole model ha s been programmed as a Java plug in named Drop Snake for the ImageJ image processing software 91 The program uses not only the whole drop contour to provide global informati on but the local polynomial fit ting as w ell in order to extract the value of contact angle s. Presenting its detailed mathematics is far beyond the scope of this dissertation. This approach generally improved some drawbacks from only using polynomial fitting models and ADSA (a xisymmetric drop sha pe analysis ), in which the assumption of axisymmetric al drops is not actually fulf illed in many cases. This is extremely important for my system since the surfaces observed are mostly with non uniformly distributed micro features that usually result in a n on axial symmetrical in for the freeware ImageJ which is very cost effective and allows for the analysis of images independently without sticking to a particular goniometer. The following results of c ontact angles were all measured by this program. More detailed information and the program itself can be found from their website. 9 2 Experimental Work of Wettability Evaluation Static Contact Angle Measurement All the contact angle measurements were carrie d out at ambient temperature (~23C) with a goniometer, Ram Hart Model 100, equipped with an automated dispensing system and a 30 gauge flat tipped needle. Figure 4 3 show s the equipment

PAGE 96

96 of goniometer, which consists of a CCD camera, light source and an a utomated syringe pump. The images were recorded by its operation software. Water purified through Milli Q system ( 18 M cm Millipore Inc.) was used as the main probe fluid, while water methanol mixtures and other organic solvent s were also used as a lo w surface probe liquid. (Table 4 1) For sessile drop test, images were photographed by putting 4 l of water droplet onto at least 5 different areas of the sample. The needle was lower ed towards the sample to carefully place drops on the surface, and then the needle was removed away from the drop. For surfaces with extreme hydrophobicity, the drops may not be able to be placed on the surface, and the needle was kept within the drop during the measure ment. The captured images were analyzed by DropSnake pro gram a plug in for the ImageJ software. (Figure 4 4) The mean value of the water contact angle and its standard deviation were reported as contact angle ( c ). Contact Angle Hysteresis Measurement Advancing angle and receding angle were also measured by goniometer. Images for advancing angle were recorded as follows: fluid was gradually added into the droplet, typically 0.05 l at a time, and images were cap tured after the fluid introduced each time. The advancing angle was determined specifically from the image frame before the advance of the liquid interface was observed. Receding angle determination was performed by the same process but withdrawing the flu id from the droplet instead. The receding angle was determined specifically from the image frame before the retreat of the liquid interface was observed. The measurement was carried out on three

PAGE 97

97 different places for each sample. The value of angles was als o analyzed by Drop Snake analysis program. Video Assessment of the Surface Hydrophobicity For extremely hydrophobic surfaces, the degree difference between advancing angle and receding angle may be too small to be measured. I employed a video based assessm ent, a so called c ontact/ c ompress/ r elease t est to examine the degree of hydrophobicity or water adhesion of the surface. Without using sophisticated equipment, softwar e while performing the experiment. The test did not provide quantitative numbers but a qualitative evaluation. The test was conducted by using a supported water droplet on an upper side, typically supported by a plastic syringe or a substrate. The supporte d droplet was slightly compressed on the surface with the advancement of the syringe, and then the syringe was slowly withdrawn from the surface. The key frame was the one right before the droplet detached from the surface. The interface distortion during the detachment (Figure 4 5) represented the degree of the surface hydrophobicity. During the droplet compression, the sample stage could also be moved to examine the surface uniformity. Surface Tension Determination The surface tension of the liquid used f or contact angle measurement was measured by the Wilhelmy plate method. The method consists of a thin platinum plate hanging under a microbalance via a thin metal wire The platinum surfac e was roughened, along with its high surface energy, to ensure complete wetting ( ~ 0 ). Before the measurement, the platinum plate was fired by a gas burner, and then rinsed with acetone and DI water to thoroughly remove any substances on the surface. Th e

PAGE 98

98 plate was immersed into the liquid and then the container was gradually lowered away from the plate. The force on the plate due to wetting was showed on the microbalance. T he surface tension ( ) of the liquid can be calculated by using the Wilhelmy equat ion : (4 1) where L is the wetted perimeter (2 w + 2 l ) of the Wilhelmy plate and is the contact angle between the liquid phase and the plate (Figure 4 6 ) In practice the contact angle is rarely measured, complete wetting ( cos = 1 ) is usually assumed for the calculation. Contact Area and Theoretical Contact Angle Interpretation Theoretical contact angles, CB were calculated by using the Cassie Baxter equation (2 9): cos CB = f s (cos c +1) 1 (2 9) where c is the contact angle measu red on the uncast, original substrate, and f s is the area fra ction of the droplet in contact with cast structure. The cast structure is generally not regular, so it is a great challenge to estimate the actual fraction of contact area. This number was estim ated through analyzing the top viewed SEM pictures by ImageJ (Figure 4 7 ) Since the emission of secondary electrons is higher at the tips, it is assumed that t he brightest regions showing on the SEM picture are those in contact with the droplet. Therefore by adjusting the threshold, a range of the contact area fraction can be determined by the software. Wetting Property of Hairy Plants Observations on Hairy Leaves Observing nature, as a fundamental step in the scientific method, is always stimulating. Thi s research is inspired by the hairy exterior of natural species that show

PAGE 99

99 water repellency. As mentioned in Chapter 2 a number of examples have been found in both the animal and plant kingdom. Since the hairy integument from the animal kingdom is usually just few millimeters, observing hairy plants is a more feasible option. However, even though water repellency on hairy plants has been broadly acknowledged, they have not all been extensively studied. For example, the Lotus Effect has been extensively stud Observations for some hairy plants, along with the latest experimental work from other studies, have been presented to provide a better understanding on this phenomenon. The hairy plants, Tomato ( Sol anum lycopersicum ), Hummingbird Plant (D icliptera S uberecta behavior. The emblem of superhydrophobicity, the Lotus leaf, was also included. Most of the analysis was performed on blocked the view from the base of the droplets, contact angle measurement was unable to be performed precisely on these hairy plants. Due to a lack of a CCD camera mounted microscope, the interaction bet ween the droplets and the hairs was mainly investigated under a stereo optical microscope (Olympus SZ 40), and the images were directly captured simply by placing a commercial digital camera (Sony DSC T 33) on the lens. The images were restricted at certai n angles, but they allowed us to perform a qualitative study. Other observations were also filmed by digital camera. The morphology and the areal density of the hairs were studied by SEM without a conductive coating on the leaves. Droplets on Hairy Plants Water deposited on these hairy plants all exhibited a similar response; they formed spherical droplets and slipped quickly off the leaves or to a concave spot. From the

PAGE 100

100 goniometer, as shown in Figure 4 8 a, the droplet, supported by the hairs (also known a s trichomes ), exhibited the apparent contact angle of exactly 180 The surface of the leaves remained in a dry state, as these hairs acted like a shield. These hairs had an aspect ratio above 1 00, a diameter around 10 m and a height up to around 1 mm cuticular wax crystals, similar to the Lotus leaf. Figure 4 9 a is a typical image taken under the optical microscope showing a droplet ( penetrated into the droplet, yet still maintained their original configuration. In order to reveal if the hairs buckled when confronting the interface, a droplet hung on a syringe was slid on the leaf under a microscope. The results showed that the hairs in contact with the droplet penetrated directly into the interface and leaned towards the direction the droplet moved, as shown in Figure 4 9 b (due to the limitation of the equipme nt, the picture quality was not optimum). The hairs neither seriously changed their arrangement, nor did they bend before and after they contacted the droplet. The force exerted from the droplet, including gravity and capillary attraction, did not seem lar ge enough to cause a severe deflection. Interestingly enough, during the composing this dissertation, a very similar result was reported by Bernardino et al. in their latest publication. 93 With a more modern imaging facility, they confirmed the hairs were fairly hydrophilic and caused a relatively large hysteresis, around 30 The hairs were hardly bent when they were partially immersed into water. They also observ ed that the droplets were usually strongly pinned by the hairs and only rolled when the leaves tilted largely. Additionally, a simplified

PAGE 101

101 model to estimate the total energy of the system was proposed by them. The model was based on three terms: the wetting energy, the capillary interaction between the hairs, and the elastic energy of bending the ha irs They found the existence of a local energy minimum c ould hardly be fulfilled. Both the hair length and the distance between them would need to be unusually large to have bending and clustering. Together with my observations, it is clear that the hairs are too stiff to be bent by the capillary interactions and the hydrophobic behavior of the leaf is simply coming from the strong pinning of contact lines. Due to a large contact angle hysteresis, the hairs penetrating into the droplet hinder the motion o f the interface and prevent the droplet from coming down to fully wet the hairs and substrate. As a result, droplets on the leaves with a lower hair density moved downwards more easily and got in contact with the substrate (Figure 4 10 b). Wetting Property of Artificial Hairy Surfaces The wetting properties of the various cast surfaces are presented in this section. The main characterization was based on contact angle measurement. A video assessment was also used to provide a qualitative evaluation of the hy drophobicity. Contact Angles of Cast PDMS Surface The PDMS silicone elastomer was cast with different moulds, including PC membranes and natural leaves. Their morphology and the resulting contact angles are shown in Figure 4 1 1 The cured PDMS elastomer wa s intrinsically hydrophobic with the contact angle at 97 The surface structure duplicated from natural leaves was able to slightly increase the contact angle to 102 After the PDMS elastomer surface was roughened by dissolving the PC membrane, the conta ct angle on the roughened

PAGE 102

102 surface increased to 137 For the membrane peeled surface (p ore size =3.0 m ), due to its smaller fraction of contact area, the contact angle had been significantly increased to the superhydrophobic region of 153 The PDMS has low surface energy (21.3 mN/m) and exhibits a hydrophobic property. The contact angle increases with i ncreasing surface roughness. The leaf structure was successfully duplicated because of its low areal density of the hairs, and it was not able to hold the droplet in a Cassie state. As a result, the contact angle was increased slightly, ~120 but was stil l lower than the leaf itself. This may be caused by a lack of the cuticular wax crystals, which commonly exist all over the leaves and enhance the hydropobicity. For the membrane dissolved surface, the surface was left with small cavities that enabled air to be trapped underneath and resulted in a high contact angle. From the image analysis, the fraction of contact area was estimated from 45% to 50%, and the consequent theoretical contact angle based on Cassie Baxter equation was 127 to 124 The smaller c ontact area from membrane peeled surface causes a contact angle larger than 150 Contact Angle s of Cast Thermoplastic Surface The pictures of the sessile drops from contact angle measurement and their associated microstructures have been brought together to illustrate the wetting behavior. In order to elucidate the results, the values of contact angles, advancing angles, and receding angles, along with their estimated contact areas and theoretical contact angles, of all the surfaces are listed in Table 4 2

PAGE 103

103 Polypropylene substrate The polypropylene sheets were cast with two types of membrane, polycarbonate (PC) and anodic alumina (AAO). Due to the small sample size of AAO cast surfaces, the following contact angle measurements were conducted on the PC cast surfaces only. Since most of the membrane peeled samples were somewhat curved, the samples were taped onto glass slides to have a larger flat region for contact angle measurement. The original PP sheet was relatively hydrophobic and had a contact angle sli ghtly larger than 90 (Figure 4 1 2 a). The images of sessile drop measurement on membrane cast PP surfaces are shown in Figures 4 1 2 and 4 1 3 a e The contact angles on these surfaces we re around 150 or above significantly increased over the original PP sheet (9 4 0.7 ). The high contact angle denoted that the droplet was in a Cassie Baxter state, where a fair amount of air pockets were trapped underneath the water interface. Regarding the contact angles, the values were increased with the decrease of po re size, and they could be estimated by the Cassie Baxter equation: cos CB = f s (cos c + 1) 1 where c wa s the intrinsic contact angle on the polypropylene sheet, and f s was the fraction of solid area in contact with the droplet. This equation simply indicates that the less solid the droplet contacts with (or the more air th e droplet contacts with), the higher the contact angle will be. It was the main reason why membrane peeled surfaces had higher contact angles than membrane dissolved surfaces, as a considerable number of pillars were torn down ( = 0.6 and 1.2 m ) lowering the contact area. Smaller contact angle hysteresis (difference between advancing and receding angles) for membrane peeled surfaces was also attributed to this factor, where fewer pinning sites existed to

PAGE 104

104 impede the movement of contact lines. Thus, due to the high areal density of pillars for membrane dissolved surfaces, the contact angle hysteresis was fairly large on these surfaces. This sticky effect had been previously reported on surfaces with high density nanotubes, where a strong adhesion force was g enerated by Van der Waals interaction 66 This phenomenon has raised an interesting question on how to determine the superhydrophobic state. To address this, the role of contact angle hyst eresis should be taken into account as the main factor. Unfortunately, the behavior of dynamic contact angle is a lot more complicated, and, unlike the Cassie Baxter equation, a general mathematical expression has yet to be established to estimate the valu e. 94 Despite the angle difference, the macroscopic wetting behavior of these surfaces were similar overall: water drops (~50 l ) falling from several centimeters easily rolled off the tilted surface (>10 ) leaving the surface unwetted. The force from falli ng was large enough to overcome the pinning force from the pillar structure. Generally, the cast PP surfaces are considered having reasonable water repellency. For surfaces with smaller angle hysteresis (< 20 ), drops bounced back after impacting the surfa ces. The bouncing height increased with the increase of the hydrophobicity as less kinetic energy was transferred to the wetting surface area. 9 5 As shown in the Chapter 3, the casting structure on 0.6 m peeled PP surface had a configuration very close to the hairs appearing on arthropods. This cast surface also featured exceptional hydrophobicity. Owing to the ultralow adhesion between the surface and water, the droplets were not able to be deposited o n the surface during the measurement (Figure 4 1 3 c). Furthermore, while changing the volume of the drop, no

PAGE 105

105 visible difference between advancing and receding angles was found, making the determination of the hysteresis extremely difficult. As a result, ano ther technique through captured videos was used to assess the hydrophobicity. The results and related issues are presented in the section regarding perfectly hydrophobic response. If the cast temperature was not high enough, the lower viscosity may have hi ndered the filling process, making the height of the protrusions not tall enough to maintain the droplets in the Cassie Baxter state. As shown in Figure 4 1 3 d, a PP substrate was subjected the same casting process but at a lower temperature, 170 C. This me mbrane was easily peeled off, but left small bumps on the surface afterwards. Droplet s on the surface were basically stuck, showing no sign of repelling water. By c asting at a higher temperature, the materials flow ed into the pores easily resulting in bet ter mechanical strength to withstand the peeling force. Some of t he pilla rs were able to be stretched to over 50 m long (Figure 4 1 3 e) and increase d the contact area, which in turn, lower ed the contact angle and its hysteresis. To compare with other PP s tructures, I followed a renowned work published in Science 9 6 where superhydrophobic surfaces were made by casting dissolved PP onto the substrate. Isotactic PP granules were first dissolved in p xylene at 130 C; the solution was then dip coated onto silic on substrates. The resolidified PP solution formed a porous network structure (Figure 4 1 3 f) raising the contact angle to around 150 with contact angle hysteresis at 27 This comparison simply showed that my current work with the high aspect ratio struct ure clearly was not only easier to prepare but also had better hydrophobicity.

PAGE 106

106 LDPE substrat e Similar to the PP substrate, the LDPE sheet had a contact angle around 90 (Figure 4 1 4 a), but rather than forming high aspect ratio pillars, due to low yield st rength and high percentage in elongation, the peeling process turned the LDPE surfaces into soft, thick layers of fiber mat. Measuring contact angles of this structure also required extra care as the drops sank slightly into the mat, blur ring the view of t he bottom. In contrast to PP or PVDF, high density protruding structures on the 0.6 m peeled surface were not ripped off, but stretched to highly entangled fibers; this greatly increased the fraction of contact area and resulted in the lower contact angle of 134 and the higher hysteresis (~23 ). The 1.2 m and 3.0 m peeled surfaces showed higher contact angles (>140 ) and more importantly, had a low contact angle hysteresis (~5 ) making the surfaces extremely water repellent. PVDF substr ate The contact angl e on the PVDF sheet was below 90 (~81 Figure 4 1 5 a), and was considered to be hydrophilic. After casting, the contact angles were increased to around 150 converting the surface from hydrophilic to superhydrophobic. The contact angles were almost the s ame, while the contact angle hysteresis of 1.2 m peeled surface was 15 degrees higher than that of the 0.6 m peeled The value was also higher than the PP and LDPE of 1.2 m peeled surfaces, which was attributed to its lower hydrophobicity nature and higher contact area of the pillar structure. Perfec tly Hydrophobic Response The contact angle was determined based on the two dimensional image generated from a three dimensional droplet. With the increase of hydrophobicity, the

PAGE 107

107 contact area between the droplet and the substrate became smaller, making the contact angle measurement on the ultrahigh hydrophobic surfaces very challenging. Among all the cast samples, exceptional water repellency was found on the 0.6 m peeled PP surfaces. The droplet could not be placed onto th is hairy substrate during the measurement due to the low adhesion to water. After the droplet volume was varied and the contact angle hysteresis analyzed, the captured images revealed that the a dvancing and receding angles were indistinguishable in these images. Figures 4 1 6 a and b are the images of a steel ball and a drop on the hairy surface acquired from the goniometer. The comparison showed that the contact angle of the drop was not smaller t han a real sphere. This also indicated that a contact angle above 170 may not represent the actual wetting property. Gao et al. 97,98 have even pointed out that contact angles may not be measured accurately in the range of extreme hydrophobicity ( c >176 ) if the conventional image captured analysis is used. Alternatively, they videotaped the response of water droplets being compressed onto and decompressed from the substrate. A similar strategy was adopted here, but instead of using videotaping t he video was captured directly from the computer screen during the operation of the goniometer. It was a cost effective method to avoid using an extra telescoped video capturing system. Figure 4 1 7 shows the sequential images from the c ontact/ c ompress/ r ele ase t est of the 0.6 m peeled hairy PP surface. During the test, a hung 2 l droplet was slightly compressed on to the surface with the advancement of the syringe, and then the syringe w as slowly withdrawn from the surface. The droplet leaves the surface with no visible distort ion (Figures 4 1 7 g and h), which represent s a very limited adhesion to this surface. Gao et al. 97 concluded that their organosilane coated

PAGE 108

108 surface, named Lichao s surface comprised of a network of 40 nm fiber adv = rec = 180 ) based on the very similar response show n here. This process has been reproduced multiple times to establish statistical significance of the interaction between the suspended droplet and the surface. Figure 4 1 8 is the set of sequential images from the sam e test but performed on the cast PP surface having a higher contact hysteresis ( ~10 ). The last two frame s showed that the surface adhesion made by the droplet leaving the substrate caused a visible distortion and was clearly not a perfectly hydrophobic surface. Another test is shown in Figure 4 1 9 where a water droplet was being p u mped out through a metal needle. The droplet slid aside after touching the surface. The stage supporting the hairy PP film was then gently moved horizontally. Although not shown in the figure, the shape and contact interface of the droplet remained and n o change was discovered throughout the entire movement. This is ascribed to the ultralow adhesion of the hairy surface. The frames confirmed that the adhesion between the surface and the droplet was of very limited range. All these pointed out that the hai ry surface created here was by far one of the best superhydrophobic surfaces within the non chemically treated surfaces. Thus it is believe d that the current work appears to be the first report to have such extreme hydrophobicity with only structural modi fication s from the original substrate. The original videos can be accessed elsewhere on the Internet. 99 Contact angle measurement via image analysis is generally considered to have an error range around 3 89 and it can also be very subjective depending o n the operator. To precisely evaluate the differences of hydrophobicity is therefore difficult when dealing with surfaces having high contact angles ( c >170 ). The video assessment provides a

PAGE 109

109 relatively simple solution through the investigation of interfac e responses. Although this technique only has a qualitative result, rather than quantitative numbers, the idea can be expanded to measure the actual adhesive force between the droplet and surface. This concept has been demonstrated to measure the force req uired for taking a droplet away from the surface of polystyrene nanotubes. 10 0 A droplet was suspended by a metal ring hung in a high sensitivity microelectromechanical balance, and the force around the micron Newton range was measured. Compared with contac t angles, the amount of force measured should more accurately represent the degree of hydrophobicity. Self Cleaning Ability The most astonishing effect of the superhydrophobic surfaces is their self cleaning ability, in which water droplets pick up the dir t and remove it while rolling off from the surface. This is also usually termed the Lotus Effect as the superhydrophobic nature of their leaves allows surface dirt to be washed off by rain or morning dew easily. Figure 4 20 schematically depicts the motion of a drop on an inclined ordinary surface and superhydrophobic surface. It is believed that dirt particles have less adhesion to topologically structured superhydrophobic surfaces and tend to adhere to the liquid interface as droplet s roll off the surface (Figure 4 20 b). 11 Figure 4 2 1 is the set of sequential images showing a dirt contaminated (carbon powder) uncast PP substrate where a few drops slid down from the surface with a very limited portion of particles being carried away. Moreover, the droplets slide very slow ly and usually requires external assistance, either a largely tilted surface or a mechanical shaking. In contrast, dirt particles on superhydrophobic hairy PP surfaces were easier to be picked up and washed off by rolling beads of water (Fig ure 4 2 2 ). Droplets dripping

PAGE 110

110 onto the contaminated surface immediately turned into black beads (Figures 4 2 2 b and e), as dirt particles adhered to the liquid interface. Even though the dirt particles formed a thick layer on the hairy surface (Figure 4 2 1 a vs. 4 2 2 a), the whole cleaning process only took a few seconds with very little mechanical force involved. Although this self cleaning has been a subject of considerable scientific interest for over a decade, compared to the efforts in making surfaces supe rhydrophobic, insightful studies on understanding this effect are still very rare. Important factors or relationships, such as particle size, particle hydrophobicity, and contact angle hysteresis vs. cleaning efficiency, have yet to be extensively addresse d. The cast hairy surface may provide good media for the study as samples with different morphology and hydrophobicity are easy to be prepared. The samples also have adequate surface area to perform the test. General Discussion This section aims to ration alize the hydrophobic response of the cast artificial hairy surfaces. Given a limited knowledge in fluid mechanics, I focused on delivering a qualitative explanation with a semi quantitative expression for the liquid repellency of hairy surfaces based on m y results and the findings of others. The cuticle of water repellent species consists of a waxy or oily substance that makes the surface contact angle above 90 and the additional surface roughness enhances the hydrophobicity further to superhydrophobic it y where liquid on the cuticle remains in a Cassie Baxter rather than Wenzel state From the animal kingdom, most water repellent arthropods present a rough integument comprising of a thick, dense layer of microhairs. Surface roughness arises from the arran gement of hairs, while the

PAGE 111

111 form and location of hair varies from species to species Therefore many arthropods possessing hairy integument are reported to have contact angles reaching over 160 19 The high contact angles on the 0.6 m peeled PP hairy surfac es arise mainly from the small contact area with the droplet and can be easily interpreted by the Cassie Baxter T heory ( Eq. 2 9 and Table 4 2 ). Strong water repellency as well as the self cleaning feature, however, depends not so much on the high contact angle, but more on the small contact angle hysteresis. While a low interfacing area is the main criteria in water repellency, the arrangement of this artificial hairy structure seems to be another importa nt factor for the occurrence of th e ultralow adhesio n observed here. The hairs created we repellent a rthropods having a hairy exterior. Although the size and density of the hairs vary among different species, this geometry may benef it the water repellency in some ways. First, all the natural hairs of water repellent arthropods are inclined at a certain angle to their cuticle (~30 45 of our cast surface). Assuming a droplet on a uniform array of pillars m aintains a CB state, then the contact angle hystere sis can be simply expressed as: 3 8 p ( 4 2 ) where p is the linear fraction of the contact line and can be treated as a size to spacing ratio (= d/L Figure 4 2 3 a); is the inclined angle to the substrate of the individual pillars, and is the intrinsic contact angle hyst eresis of the pillars materials. According to the equation, the contact angle hysteresis is decreased by decreasing the size to spacing ratio of the pillars but increased with the increase of the tilted angle. This may increase

PAGE 112

112 the retention force of the contact line and generate more thrust for water walking species. 16 It may, however, increase the difficulty of shedding droplets as well. Note that the model assumes the contact line migrates on the top, flat area of the pillars, which is far from the cas e for hair tips. On the other hand, when the layer of hair comes in contact with a liquid, the liquid interface is suspended by the hairs with a pressure difference. The pressure required (impregnation pressure) to force liquid through v arious arrangement s of pillars has been calculated. 51 65 The calculation revealed that the required pressure is increased by decreasing the spacing of the array ( L ). In addition, the results showed that the array of tilted pillars increased the impregnation pressures since the free space between the pillars was reduced, making tilted pillars favorable to resisting water. A horizontal array of cylindrical pillars was also considered (Figure 4 2 3 b), and it was concluded that such an arrangement may provide greater water resist ance. Liquid that penetrated between the horizontal pillars needed to wet the maximum surface area and required maximum surface energy. Overall, c ompared to simple vertical or inclined pillar arrays, it is energetically unfavorable for a liquid to intrude between the pillars parallel to the interface. Most of the hairs of natural species and our cast surfaces are not only inclined but curled with their tips parallel to the bodies, which in turn provides a greater resistance to wetting Another crucial role of the hairs is the mechanical response to the liquid interface. This response was previously considered in plastron respiration, in which the submerged insect s breathe the oxygen from a layer of air maintained by their hair y integument. 17,18,5 1 The curvat ure pressure and the mechanical strength of hairs sustain

PAGE 113

113 the interface, while higher ambient hydraulic pressure may buckle the hairs causing the collapse of the plastron air. Earlier observations found that the collapse for most insects occurred in the pr essure between 1 to 5 atm. With newer and better imaging equipment, a calculat ion, however, estimated that the hairs should be stable at a pressure up to 40 atm 18 indicating the bald patches of the cuticle may be responsible for the failure. In terms of t he dynamic behavior, this has not been extensively studied. A recent study showed the hairs on water walking insects can be bent during the interaction with the liquid interface. This flexibility associated with the second scale of roughness on the hairs r educe d the contact force while performing locomotion 57 Another newly published work revealed the hierarchical design of the anti wetting termite wings, in which the longer hairs (~30 m long and 2 m wide) and the shorter star shaped structure (~5 m high and 5 m wide) serve to repel macro and micro droplets, respectively. Macro droplets are mainly supported by longer hairs causing the hairs to deflect and transfer the kinetic energy to elastic energy. The mechanism may also contribute to the extremely hydrophobic response of the artificial hairy surfaces Considering how a droplet rolls on the hairy structure, the solid liquid interface would advance and retreat upon the hairs if the e nergy barriers of interface pinning we re overcome (Figures 4 2 4 a and b). The hairs in contact with the liquid may also be deflected Rather than wetting more surface area, the kinetic energy is transferred into potential energy (Figures 4 2 4 c and d) This interaction with the flexible hairs would prevent the further intrusion of liquid between hairs and allow the contact area fraction to be minimized during the motion of droplets The adhesion force also remains at the lowest level The hypothetical mechani sm proposed requires further examination, such

PAGE 114

114 as with computational modeling or microscopy techniques, in order to reveal the interaction on the interface. Although studies on the dynamic role of the flexibility of hairs and hydrophobicity are still very limited, the principle may yield a new route for the f abrication of superhydrophobic surface s. In addition to the geometrical factor, the surface chemistry should be taken into account. The PP substrate contain ed a methyl side group which is considered hi ghly hydrophobic. The nanofiber structure has been shown to enhance hydrophobicity because the hydrophobic portion of the molecule chain tends to be exposed to air, enhancing the surface hydrophobicity. Summary The wetting property of natural and cast surf aces was characterized by measuring contact angles. Although contact angle measurement with goniometer has been widely used, few important issues should be taken into careful consideration to lower the showed contact angle at 180 while the contact angle hysteresis was relatively large. The hydrophobicity imply com es from the strong pinning of contact lines rather than elastic response of the hairs. Contact angles of all the cast surfaces were significa ntly increased over their original flat sheet. The high contact angle denotes the droplet is in Cassie Baxter state, where a fair amount of air pockets were trapped underneath water interface. The value of contact angles was generally increased with the de crease of membrane pore size, and they can be estimated by Cassie Baxter equation. Remarkable water repellency was found on 0.6 m peeled PP hairy surfaces showing no contact angle hysteresis. Video assessment confirmed that the PP hairy

PAGE 115

115 surfaces is perfectly hydrophobic ( adv = rec = 180 ) and is by far one of the best super hydrophobic surfaces within those non chemically treated s urfaces. The confirmation of natural and artificial hairs (inclined and curled with their tips parallel to the bodies) provides a greater resistance to wetting A possible mechanical response to the liquid interface where hairs are deflected to avoid the f urther wetting is another important factor lowering the contact angle hysteresis. Related studies addressing the relationship between elastic materials and hydrophobicity are still rare, and the mechanism needs further examination with either computational simulation or imaging technique.

PAGE 116

116 Figure 4 1. Contact angle measurement on a superhydrophobic surface with different diameters of syringe. (a) a larger plastic syringe, and (b) a narrower, 30 gauge, metal needle. Figure 4 2. Images of the same water droplet on a superhydrophobic surface under different fitting modes of the static contact angle: (a) ellipse fitting; (b) circle fitting; (c) tangent searching; (d) and Laplace Young fitting. The figures include the simulation lines of the shape of t he water droplets and the horizontal baselines. ( r eproduced by permission of The Royal Society of Chemistry .) (a) (b)

PAGE 117

117 Figure 4 3. The equipment set up of goiniometer, where the droplet was loaded by a syringe pump and placed on the sample surface. The image wa s later captured by the CCD camera connected to the controlling program. Fig ure 4 4. The interface of Drop Snake program. The value of contact angle was determined by fitting the contour of the whole drop and polynomial fit ting of the interfacing area. CCD Camera Main Light ing Auxiliary Light ing Syringe Pump Sample Stage CCD Camera Main Light ing Auxiliary Light ing Syringe Pump Sample Stage

PAGE 118

118 Table 4 1 Measured s urface tensions of the liquids used in contact angle measurement. Liquid Surface Tension (mN/m) Liquid Surface Tension (mN/m) Water 72.8 Hexdecane 27.6 Oliver oil 34.6 Octane 21.8 Water Methanol (MeOH vol%) 0.00 72.8 29.41 44.1 2 .24 67.8 35.53 41.4 4.71 62.9 45.45 37.8 6.66 62.5 55.75 34.5 7.39 59.7 65.41 31.8 12.67 55.0 74.83 28.9 17.84 51.3 80.71 25.2 23.81 47.7 100.00 22.3 Figure 4 5 Selected sequential images during contact/compression/release test on a superh ydrophobic surface. The last frame shows the droplet left the substrate with a visible distortion. Figure 4 6 Schematic picture of Wilhelmy plate method of measuring liquid surface tension.

PAGE 119

119 Figure 4 7 The fraction of con tact area is estimated by adjusting the threshold of top view SEM pictures.

PAGE 120

120 Figure 4 mantle impede the droplet being pulled, (c) droplet on the Lotus leave, and (d) t he microstructure of Lotus leaves. Figure 4 (b) hairs in contact with the droplet are rarely deflected by the interface. ( a) (b) (c ) (d ) ( a) (b)

PAGE 121

121 Figure 4 eaves. (a) Higher contact angles on dense and ordered hairs. (b) Lower contact angles on random oriented hairs. Figure 4 11. Morphology of different cast PDMS surfaces and their contact angles. (a) intrinsic PDMS surface, (b) leaf cast PDMS surface, (c) membrane dissolved PDMS surface (pore size=0.6 m), and (d) membrane peeled PDMS surface (pore size=3.0 m ). The contact angle is increased with the surface roughness ( a) (b) (d ) ( c ) ( a) (b)

PAGE 122

122 Table 4 2. Contact angles of all the cast thermoplastic surfaces and the theo retical contact angles calculated from Cassie Baxter theory ( Eq. 2 9 ) PP (Polypropylene) Pore Size Contact Angle ( c ) Advancing Angle ( adv ) Receding Angle ( rec ) Contact Angle Hysteresis a ( ) Estimated Contact Area Fraction ( f s ) Cassie Baxter Contact Angles ( cb ) Original sheet 94 0.7 98 2.1 75 2.6 23 Membrane dissolved 0.6 m 155 3.6 161 3.7 147 3.1 24 0.11 0.03 15 4 1 3.7 1.2 m 153 3.6 159 4.9 111 3.1 48 0.17 0.02 14 7 8 1.51 3.0 m 147 1.9 169 1.6 113 6.3 56 0.23 0.01 141.8 0.9 Membrane peeled 0.6 m >170 >170 >170 <1 0.02 0.01 169.3 2.9 1.2 m 157 3.8 160 4.3 152 3.1 8 0.15 0.04 15 0.1 4.3 3.0 m 152 4.0 162 2.5 145 4.4 17 0.21 0.03 143.6 2.7 0.6 m membrane peeled 170 C cast 134 5.5 138 6.3 91 6.4 47 200 C cast 163 4.2 164 1.8 154 4.8 10 Dissolved resolidfied 151 4.2 162 4.7 135 4.0 27 LDPE (Low density Polyethylene) Pore Size Contact Angle ( c ) Advancing Angle ( adv ) Receding Angle ( rec ) Contact Angle Hysteresis a ( ) Estimated Contact Area Fraction ( f s ) Cassie Baxter Contact Angles ( cb ) Original sheet 91 2.3 93 1.3 70 4.6 23 0.6 m 134 4.0 150 4.2 127 2.5 23 0.36 0.04 130.4 3.0 1.2 m 154 5.7 159 4.5 152 2.6 7 0.31 0.06 134.3 4.8 3.0 m 141 2.3 151 3.2 146 3.5 5 0.24 0.04 140.0 3.6 PVDF (Polyvinylidene Fluoride) Pore Size Contact Angle ( c ) Advancing Angle ( adv ) Receding A ngle ( rec ) Contact Angle Hysteresis a ( ) Estimated Contact Area Fraction ( f s ) Cassie Baxter Contact Angles ( cb ) Original sheet 81 1.6 82 0.5 42 5.1 38 0.6 m 152 2.7 159 2.9 133 5.8 26 0.13 0.02 148.9 3.2 1.2 m 147 1. 8 153 2.0 112 5.3 41 0.23 0.02 137.8 2.5 a. This angle was the difference between the mean value of advancing angle and receding angl e

PAGE 123

123 Figure 4 1 2 The surface morphology of the membrane dissolved PP surfaces and their sess ile drop images. (a) original PP sheet. The membrane pore size is (b) 3.0 m (c) 1.2 m and (d) 0.6 m ( a) (b) (d ) ( c )

PAGE 124

124 Figure 4 1 3 The surface morphology of the membrane peeled PP surfaces and their sessile drop images. The membrane pore size is (b) 3.0 m (c) 1.2 m and (d) 0.6 m (d) and (e) are 0.6 m peeled surfac e cast at 170 and 200 respectively. (f) Si substrate dip coated with i PP solution. ( a) (b) ( d ) ( c ) ( e ) (f )

PAGE 125

125 Figure 4 1 4. The surface morphology of the membrane peeled LDPE surfaces and their sessile drop images. (a) original LDPE sheet. The membrane pore size is (b) 3.0 m (c) 1.2 m and (d) 0.6 m ( a) (b) (d ) ( c )

PAGE 126

126 Figure 4 1 5 The surface morphology of the membrane peeled PVDF surfaces and their sessile drop images. (a) original PVDF sheet. The membrane pore size is (b) 1.2 m and (c) 0.6 m (b) (c) ( a )

PAGE 127

127 Figure 4 1 6 The imag es of contact angle measurement acquiring from goniometer. (a) a steel ball and (b) a droplet on cast PP hairy surface. Figure 4 1 7 Selected images during contact/compression/release test on 0.6 m peeled hairy PP substrate. The last th ree frame s show the droplet left the substrate without a visible distortion, and the response represents the surface is nearly perfect hydrophobic. (a) (b) (c) (d) (e) (f) (g) (h) (i) ( a ) ( b )

PAGE 128

128 Figure 4 1 8 Selected images during contact/compression/release test on 0.6 m peeled PP substrate (cast at 200 C) The last two frame s show the droplet left the substrate with a visible distortion, and is apparently not a perfectly hydrophobic surface. (a) (b) (c) (d) (e) (f) (g) (h) (i)

PAGE 129

129 Figure 4 1 9. Selected images during motion test on 0.6 m peeled hairy PP substrate. A water droplet slid aside after touching the surface while water was being pu mped out through a metal needle T he shape and contact interface of the droplet remained and no change was discovered. (a) (b) (c) (d) (e) (f) (g) (h) (i)

PAGE 130

130 Figure 4 2 0 (a) Drop slides on an ordinary hydrophobic surface hardly carrying away dirt particles. (b) A drop rolls on a superhydrophobic surface picking up most dirt particles. (a) (b )

PAGE 131

131 Figure 4 2 1 Water droplets on an uncast PP sheet contaminated by carbon powder. From (a) to (f), the droplets slide down slowly from the surface if the tilted angle is large enough. A very limited portion of dirt particles was washed out with the droplets. (a) ( b ) ( c ) ( d ) ( e ) ( f ) ( g ) ( h ) ( i )

PAGE 132

132 Figure 4 2 2. Water droplets on a cast h airy PP surface contaminated by dirt particles. From (a) to (i), the droplets were formed beads and rolled off from the surface effortlessly. The surface demonstrates the self cleaning ability where the dirt particles were effectively washed out with the d roplets rolling. (a) ( b ) ( c ) ( d ) ( e ) ( f ) ( g ) ( h ) ( i )

PAGE 133

133 Figure 4 23 Two configurations of models for water repellent hair pile (a) A uniform array of pillars titled at an angle. (b) An array of horizontally aligned cylindrical hai rs which suspend water interface. Figure 4 2 4 (a) and (b) the motion of a droplet on the hairy surface causing the deflection of hairs. (c) hair bending un der hydraulic pressure resulting in very little or no change in wet area, (d) stiffer hairs under hydraulic pressure resulting in the increase of wet area. (a) ( b ) (a) ( b ) ( c ) ( d )

PAGE 134

134 CHAPTER 5 TOWARDS SUPEROLEOPHOBIC SURFACES This chapter aims to extend the cast surfaces to oleo phobic (oil repellent) region. Since most organic liquids or oil have much lower surface tensions (20 30 mN/m), therefore making a surface oil repellent is a great challenge in engineering level. An additional surface treatment of the surface is essential to render superhydrophobic surfaces into oleophobic or super oleophobic. Plasma surface treatment was used as the main tool for surface modification. The chapter starts with experimental work where two different plasma systems were introduced, a home const ructed and a commercial system. The resulting structure and their contact angles are presented at the second section. While the results showed the oil contact angle of the surfaces treated with fluorocarbon compound can be significantly increased, strong o il repellency or super oleophobicity was not able to be obtained on these hairy surfaces. Experimental Work of Plasma Treatment Introduction Surface treatment by means of p lasma is an effective technique for many bulk materials 10 1 One of the special inter est s is growing in manipulating the degree of surface hydrophobicity. P lasma is usually considered the fourth state of matter, comprised of highly excited atom s molecu les, ions, and radical species; these are typically generated by exciting gas molecules into energetic states with different sources, such as radio frequency ( RF ) electric fields microwave s or electrons from a hot filament discharge. The electromagnetic fields accelerate the electrons to impact the molecules, which in turn result s in the ex citation of atoms and ions, dissociation of gas molecules and production of more electrons. Therefore, p lasma is a highly unbalanced

PAGE 135

135 and reactive chemical environment in which the high density of ionized and excited species can promote chemical reactions and change the surface properties of normally inert materials. The materials used for casting hairy structures, polypropylene, polyethylene, and polyvinyl lidene fluoride are considered to have relatively inert surfaces where the absence of reactive sites ( e g. OH or NH 2 ) makes it difficult for them to couple with other chemical groups. Plasma treatment is practicable and can introduce radicals or reactive functional groups to the polymer surface through ion bombardment and UV radiation; this enhances t he corporation with other new species. T wo opposite routes were applied to test the possibility of enhancing oil repellency: (1) to increase the surface hydrophilicity and (2) to increase the surface hydrophobicity. To increase the hydrophilicity, the sur face was treated with water plasma to induce more hydroxyl groups on the surface. To increase the surface hydrophobicity, the surface was coated with a fluorocarbon (CF x ) layer using commercial equipment. Using the Inductive Coupled Plasma (ICP) Deep React ive Ion Etching (DRIE), a fluorocarbon compound was deposited as a passivation layer during an etching process. The fluorocarbon family is known for having low surface energy and hydrophobic behavior, such as Teflon. DRIE, originally designed for creating high aspect ratio surface structures, provides a stable environment and effective way for performing the pla sma deposition of CF compounds. 10 2 ,103 Surface Treatment with Water Plasma Water plasma treatment was carried out by a home assembled plasma system at Dr. Eugene Goldberg layout of the system was schematically showed in Figure 5 1. Plasma was generated through Radio Frequency Glow Discharge (RFGD) at 13.56 MHz in a modified bell jar

PAGE 136

136 wit h a copper coil to carry the current (RF Plasma Products, Inc). The chamber vacuum was maintained by a mechanical pump. S ample s w ere placed in the bell jar till t he pressure was below 100 mTorr for 10 minutes. Argon was then purged into the chamber at 1000 sccm for 3 times and then turn on the RF power to generate the plasma at 50 Watts to pre clean the chamber for 5 minutes. The source of water vapor was simply provided by connecting a bottle of DI water into the gas tube. Water vapor was then introduced into the chamber and then purged for 3 times After purging, pressure was kept at 100 mTorr and plasma was generated by adjusting RF power with incident power at around 50 W and reflected power at 25 W During treatment water vapor was continuously flowed into the chamber and adjusted to have a stable plume Cast hairy samples were placed ~3.5 inches below RF coil ( F ig ure 5 2 ). The duration of plasma treatment was from 1 minute to 5 minutes After plasma treatment, the samples were immediately tested with distilled water, and other organic liquid to have preliminary results. Surface Treatment with Fluorocarbon Plasma Deposition Fluorocarbon coating was deposited by using deep reactive ion etching (DRIE) in Nanoscience Institute for Medical and Engineering Technology (NIMET) at University of Florida. The equipment has two independent RF power source (at 13.56 MHz), the coil around the etching chamber and the platen coil connected to the sample stage (Figure 5 3). The chamber coil is used to dissociate the sp ecies and generate radicals while platen coil is to control the RF bias potential of the sample for having directional etch. Figure 5 4 shows a typical ion etching process where the plasma generated from SF 6 /O 2 mix gas flow is used to directionally etch t he silicon, and C 4 F 8 is then introduced to deposit a passivation layer on the sidewalls to protect etched structures. The steps

PAGE 137

137 are repeated to create deep structures. In our hydrophobic layer deposition, the plasma was generated with only C 4 F 8 gas introdu ced to the system. Cast thermoplastic samples were taped on a 6 inch wafer, and the wafer was fixed by a set of alumina clamps to the electrode. The whole set was then loaded into the DRIE system. The vacuum was first lower to below 1 mTorr and then mainta ined at 50 mTorr during the deposition with C 4 F 8 flow rate at 90 sccm The fluorocarbon plasma deposition was carried out for 30 seconds at coil power 600 W without applying platen bias. The thickness of 30 second deposition layer was around 60 nm measured by ellipsometer Different sets of coil power were tested to obtain the optimum condition. After plasma treatment, the samples were immediately tested with distilled water, and other organic liquid to have preliminary results. Characterization of Plasma Treated Surface Surface morphology The surface morphology before and after plasma treatment was characterized by scanning electron microscope (SEM JSM 6400, JEOL co. ). All the samples were coated with a thin layer of Au Pd (thickness ~30 nm) to improve the conductivity prior to the examination. The voltage for observing the plasma treated samples was at 10 15 KV. Surface chemistry Surface chemistry was analyzed by X Ray photoelectron spectrometry (XPS, Perkin Elmer XPS/ESCA PHI 5100 ESCA system ) with Mg K exciting radiation source (1253.6 eV) The typical analysis depth is around 1 10 nm from the bulk surface, since the kinetic energy of the photoelectrons is not large enough to escape from deeper in the sample. The photo electrons were collected with tak e off angle at 45 and quantitatized by h emispherical analyzer with pass energy of 22.36 eV. Carbon 1s peak

PAGE 138

13 8 was used to determine the relative areas of CF, CF 2 CF 3 and C CF x in fluorocarbon deposited samples. The commercial software that came with the inst rument was used to smooth the data, and the peaks were fitted with Gaussian Lorenztian at ratio 60/40. Contact angle measurement Contact angle measurement was mainly carried out by sessile drop test as described in C hapter 4. The Images were photographed b y putting 4 l of liquid droplet onto at least 5 different areas of the sample. The syringe was lower towards to the sample to carefully place drops on the surface, and then the needle was removed away from the drop. The captured images were analyzed by Dro p Snake program a plug in for the ImageJ software. The mean value of the water contact angle and its standard deviation were reported as contact angle ( c ). Experimental Results Surface with Water Plasma Treatment T he water plasma treatment was mainly appl ied to the highest hydrophobic surfaces, 0.6 m peeded PP samples. The samples were loaded into the chamber and treated with water plasma for 1 minute (including the prior chamber clean process in Ar plasma for 5 minutes). After the treatment, drops of diff erent liquids (~10 l ) were put onto the surface to quickly exam the wettability. Neither hydrophobicity nor oelophobicity was enhanced after plasma treatment. The water contact angle was greatly lowered to around 97 annihilating its superior water repell ency, and other organic liquids were spread on the treated surfaces showing no difference from the untreated samples.

PAGE 139

139 Some of the samples were treated with Ar plasma only; however, as pictures shown in Table 5 1, wetting property of all the samples wa s si milar. The superhydro phobicity was totally vanished even w ith Ar plasma treatment for only 1 minute. Figure 5 5 shows the morphology of the plasma treated surfaces. The hairy structure was severely damaged after 1 minute Ar plasma treatment (Figure 5 5a a nd b). The plasma also flattened the original PP substrate slightly lowering the contact angle to around 88 (Figure 5 5c and d). This implies the surfaces had been sustained a certain level of ion bombardment which commonly exists in the plasma. The plasm a contains highly reactive and charged species, and easily to be attracted to impacting an oppositely charged surface. The energy of reactive species is transferred into the material surface during the collision resulting in etching, degradation or chain scissions. 10 1 Degradation yield and rate depend on plasma and polymer natures. Gaseous radicals such as H OH, or O 2 generated in the water plasma effectively enhance the degradation process. Even with the Ar plasma, since the chamber is simply seale d with o ring and equipped with a mechanical pump, the residual oxygen coming from the poor vacuum may also have the same effect. AC power was also lowered to reduce the energy of plasma, but it was difficult to obtain a stable plump. Since the surface str ucture was majorly damaged from the process, water plasma was not further used for making oleophobic hairy surfaces Surface with Fluorocarbon Coating Fluorocarbon coatings were deposited on the samples by plasma polymerization of C 4 F 8 in the DRIE chamber. To avoid a possible damage from the plasma, fluoro carbon coating was deposited onto the PP sheet prior to casting process As shown in

PAGE 140

140 Figure 5 6, t he coated CF layer was like a thermal barrier that completely hinder ed the formation of surface structures during casting. Although the CF coatings jeopardize the casting process, it could actually be applied as a control mechanism if the surface structure is undesired on some particular area. The plasma deposition of CF therefore has to be carried out onto th e cast hairy surfaces. Figure 5 7 shows the hairy structure remained the same after plasma deposition. The CF layer on the original PP sheets increased the hydrophobicity raising water contact angle from 94 to around 100 Figure 5 8 shows a methanol wate r mixture ( ~ 25.2 mN/m ) on hairy PP surfaces coated with CF layers under different plasma power. The contact angles of methanol water mixture were increased with the increase of plasma power. In contrast to many fluorocarbon precursors, the film deposite d from C 4 F 8 is more sensitive to the applied RF power. 10 4 Compared to other smaller fluoro carbons ( e.g. CF 4 C 2 F 6 etc.), the cyclic structure of C 4 F 8 (Figure 5 4b) may take significant RF power t o dissociate the molecules. 10 5 The concentration of C F 2 fro m XPS spectra (Figure 5 9) also show s that the polymerization of CF molecules was increased with the increase of the RF power Table 5 2 lists some of the images of the sessile drops on uncoated and coated cast PP surfaces. Test with water methanol mixture s showed the contact angles were significantly increased from around 60 to above 130 The contact angle of olive oil was also raised to 100 The low surface tension drops should be placed onto the surface with extra careful, since they tend to reach the more stable Wenzel state. However, the coated surfaces still had contact angles around 30 with nonpolar liquids.

PAGE 141

141 General Discussion Making surfaces become oil repellent is extremely challenging on an engineering level, since the surface tensions of oily liquids are usually in the range of 20 30 mN/m. The first strategy used was to increase the surface hydrophilicity and thus, the capability of repelling hydrophobic liquids. The hydrophilicity was enhanced by using water plasma to generate more hydroxyl gr oups on the hairy surface. In order to have something to compare with the PP polymer, two common polymers containing the hydroxyl group, p olyvinyl alcohol (PVA) and poly(2 hydroxyethyl methacrylate) (pHEMA), were synthesized. The contact angles of oily liq uids on the PP surface were around 30 ; however, the liquids on the PVA and pHEMA surfaces were completely spread out, making contact angles unable to be measured (Figure 5 11). Besides the surface structure was severely damaged, the goal and fundamental concepts seemed to be in conflict. The proposed strategy needed to be reconsidered at this point in terms of the most basic and essential criterion for having liquid repellency: maintaining liquid drops in Cassie Baxter state. The energy of the CB state is considered meta stable and has the possibility to transfer into the Wenzel state. The energy state of the meta stable CB state can be quite complicated, depending on the surface structure. 45 47 The threshold value of the equilibrium contact angle ( trans ) however, can be calculated simply by combining the Wenzel and CB equations, as described in Chapter 2 eq. (2 10): cos trans = ( f s 1)/(R f f s ) (2 10) where R f and f s are the surface roughness factor and fraction of liquid/solid contact area. Since R f > 1 > f s makes cos trans < 0 the value of this transition contact angle is larger

PAGE 142

142 than 90 This implies a liquid with a surface contact angle smaller than 90 cannot be in 5), the contact angle ( c ) is determined by: cos c = ( S V SL ) / L V (2 5) where S V SL and L V are surface tensions solid/vapor, solid/liquid and liquid/vapor, respectively. When the contact angle is 90 the SL and S V should be equal The term SL can be approximated as : 106 (5 1) From eq. (2 5) and eq. (5 1), when the contact angle is at 90 we can have SV = LV /4 Therefore, for low surface tension liquids, such as o ctane (21.6 mN/m) to have a contact angle lar ger than 90 the surface tension of the substrate material must be at the range of just several mN/m. On the other hand, to induce hydrophilic groups on the surface actually increases the polar contribution to the surface energy and reduces the contact an gle of liquid, pulling it further away from oleophobicity. Hence, the strategy should turn to lowering the surface energy of the solid. Among all the solid surfaces, fluorocarbon compounds are renowned for their low surface energy, since the high electronegativity of fluorine atoms makes F C bonds difficult to be polarized; this, in turn, lowers the hydrogen bonding and dispersion interaction with liquids. The degree of th is nonattractive feature increases with the number of fluorine atoms, i.e., CF 3 > CF 2 > CF, and also with the chain length of perfluoroalkyl compounds. It is the only material to date that is capable of rendering the surface oleophobic. Note that the subst rate was either fluorinated compounds or coated with CF materials having a longer perfluoroalkyl chain ( CF 3 (CF 2 ) n n>7). 10 5 1 07

PAGE 143

143 In addition to lowering the surface energy of substrates, the surface structure also plays a crucial role in superoleophobicity The geometry of the so called re entrant structures, such as mushroom heads, micro hoodoos, or horizontally aligned cylindrical rods, is proven to have good repellency against low surface tension liquids. 1 08 11 0 The re entrant structure implies that a li ne drawn vertically up from a point on the projection of the solid surface may meet the actual solid interface more than once. Two examples are shown in Figure 5 12, where (b) represents the re entrant structure. From the diagrams, when c < the interfa cial force is downwards, dragging the liquid towards to the substrate. If c > the interfacial force is upwards, preventing the liquid from moving downwards and creating a possibility for having a composite CB state. The CF coated hairy surface showed a high contact angle with polar liquids, but the surface energy was not low enough to repel non polar liquids. The yield of CF 2 from the plasma polymerization of C 4 F 8 is lower than using compounds with a longer perfluoroalkyl chain 106 In addition, the hai ry structure cannot be categorized as a re entrant structure, as most of the hairs are inclined at a certain angle with tips of smaller diameters making the profile of liquid interface as it shows in Figure 5 12 Summary In this chapter, the efforts aimed to extend the remarkable water repellency into oil repellency (superoleophobicity). It is a great challenge in engineering because the surface tensions of oily liquids are only a half to one third that of water. Two opposite strategies were adopted to test the possibility of rendering hairy surface oleophobic: (1) to increase the surface hydrophilicity and (2) to increase the surface hydrophobicity. Water plasma was first used to increase the hydrophilicity by inducing a hydroxyl group

PAGE 144

144 onto the surface. How ever, the surface structure was destroyed, resulting in the complete loss of superhydrophobicity and oleophobicity. Moreover, the idea of increasing the surface hydrophilicity intrinsically increases the surface energy, making the surfaces more difficult t o repel low surface tension liquids. Therefore the water plasma was not used for the purpose of increasing oleophobicity. To increase the surface hydrophobicity, commercial equipment, DRIE, was used to plasma polymerize a CF layer onto the hairy surfaces. The contact angle of low surface tension polar liquids ( < 35.2 mN/m) was raised from 60 to around 140 The surface energy of the coated samples was still not low enough to create a high contact angle with non polar liquids. To achieve superoleophobici ty, the first criterion is to lower the surface energy of the substrate to about one quarter of the oily liquid. Fluorocarbon compounds are the only substances that are able to lower the surface energy enough to the oleophobic range. The so called re entra nt structure also enhances the surface to maintain the liquid in a composite CB state

PAGE 145

145 Figure 5 1. Schematic diagram of the plasma system used in this work. The plasma was induced by radio frequency coil. The vacuum was gen erated simply by a mechanical pump. Figure 5 2 The plasma chamber is a modified bell jar with a copper cable coiled in the upper part. The sample was placed under the plume of water plasma during the treatment. s ample

PAGE 146

146 Figure 5 3 Schematic diagram of the STS deep reactive ion etch system Figure 5 4. (a) Schematic diagram of the typical process of deep reactive ion etching (1) directional etching, (2) deposition of passivation layer, (3) next e tch step. (b) O ctafluorocyclobutane (C 4 F 8 ) molecule for passivation layer. (a) (b) C F

PAGE 147

147 Table 5 1. Liquid drops on plasma treated 0.6 m peeled PP samples. Treat ment Liquid Water ( ~ 72.8 mN/m) Oliver oil ( ~ 34.6 mN/m) Hexane ( ~ 18.4 mN/m) Ar 5 min + H 2 O 1 min Ar 5 min Ar 1 min

PAGE 148

148 Figure 5 5. (a) and (b) are the morphology of hairy surface before and after Ar plasma. The contact angle was lowered from 170 to 95 (c) and (d) are the morphology of original PP sheet before and after Ar plasma. The contact angle was slightly lowered from 94 to 88 Figure 5 6. The surface morphology of CF coated PP samples after casting process. (a) The CF layer hinders the formation of cast structure. (b) and (c) a re closer views of the different regions, and the bars are 50 m and 20 m respectively. Ar plasma 1 min (c ) (d ) (a ) (b ) Ar plasma 1 min ( a) ( b ) ( c ) Coated region Uncoated region

PAGE 149

149 Figure 5 7. The hairy structure before and after plasma deposition of CF layer in DRIE chamber. Figure 5 8. Contact angles of methanol water mix ture ( ~ 25.2 mN/m ) on hairy surfaces coated with different plasma power. 30 sec 600 W c = 151 c = 74 c = 35 c = 29 10W 30sec 50 W 30 sec 150W 30sec 600W 30sec

PAGE 150

150 Figure 5 9. XPS results of CF coated surfaces under different plasma power. Figure 5 10. Low contact angle of nonpolar liquid ( dodecane ) on CF coated 0.6 m peeled hairy surfaces. Peaks Area % 1. CF 3 15.6 % 2. CF 2 18.6 % 3. CF 22.4 % 4. C CF x 18.5 % 5. C C/C H 21.4 % Peaks Area % 1. CF 3 2 2.3 % 2. CF 2 25.1 % 3. CF 17.0 % 4. C CF x 16.9 % 5. C C/C H 14.7 % Dodecane ~25.3 mN/m ~ 32 Hexane ~ 18.4 mN/m ~ 3 1

PAGE 151

151 Table 5 2. Contact angles of low surface tension liquids on cast PP surfaces before and after CF coatings. Uncoated Coated H 2 O MeOH ( ~ 35.2 mN/m) H 2 O MeOH ( ~ 35.2 mN/m) H 2 O MeOH ( ~ 25.2 mN/m ) Olive oil ( ~ 34.6 m N/m) = 69 = 145 = 151 = 100 = 55 = 6 7 = 149 = 126 = 112 = 133 = 127 = 125 0.6 m peeled 0.6 m dissolved 3.0 m dissolved

PAGE 152

152 Figure 5 11. Low surface tension liquid (dodecane) completely spread out on the hydrophilic polymers, (a) PVA and (b) pHEMA. Figure 5 12 Schematic diagrams of interfacial force when a liquid with contact angle c on the same substrate material but having different geometry. (a) non reentrant structure, and (b) reentrant structure. ( a) (b ) Dod ecane ~25.3 mN/m Dodecane ~25.3 mN/m ( a) (b )

PAGE 153

153 CHAPTER 6 CONCLUSIONS AND SUGGESTIONS FOR FUTURE WORK Conclusions The main goal of this study, as delineated in the introduction is to develop a liquid repellent surface by mimicking the hairy exterior of arthropods. The synthetic superhydrophobic surfaces based on this type of structure are still very limited, though the role of the piliferous integument has long been recognized as rendering many arthropods water repellent and providing critical living functions. Therefore the main efforts of the study were devoted to creating the artificial hairy surfaces and to studying their wetting behaviors. The majority of the surfaces in the study were prepared by casting the thermoplastics with commercial porous membranes. The fabrication method was relatively simple and cost effective, but the membrane limited the control of the experimental parameters, such as the areal density and distrib ution of structures. The particular hairy surface was successfully created on cast PP substrate s after peeling off the PC membrane, where the surface was covered with s hape d pillars analogous to the natural hairs of the arthropods They taper ed to a point from a base with the pore size diameter while the length varie d from less than a micron to about 10 microns. The l ong er hairs we re tilted around 30 to 45 with thei r tips curled towards or parallel to the substrate. The process to create hair like surface structure has been demonstrated on other thermoplastics, including LDPE and PVDF. The casting temperature needs to be higher than the melting point of the substrat e, while the peeling has to be performed at a temperature below the glass transition range. The cast structure is not robust and easily

PAGE 154

154 damaged with an external force. However, the durability is a general concern on all superhydrophobic surfaces. The curre nt process has the possibility to be scaled up for the production of large area superhydrophobic surfaces. In general, the water contact angles on the cast structures were all significantly increased. The high contact angle denotes that a fair amount of ai r pockets were trapped underneath the droplet; the angle value can be rationalized by the CB equation. Remarkable water repellency was found on the PP surfaces possessing artificial hairy structures. Video assessment has confirmed that the PP hairy surface s is perfectly hydrophobic ( adv = rec = 180 ) and is by far one of the best superhydrophobic surfaces within the non chemically treated surfaces. The h ydrophobic methyl group of PP and a possible mechanical response to the liquid interface, where the deflect ed hairs resist further wetting, are considered the major factors in lowering the contact angle hysteresis. However, the hydrophobicity of hairy leaves is quite different from the synthetic surfaces. The 180 e comes simply from the strong pinning of contact lines rather than the elastic response of the hairs. To extend the repellency to oily liquids (oleophobicity) is a great challenge on an engineering level as the surface tensions of oily liquids are only ha lf to one third of water. To achieve superoleophobicity, the first criterion is to lower the surface energy of the substrate to about one quarter of the oily liquid. Fluoro carbon (CF) compounds are the only substances that are able to lower the surface en ergy enough to the oleophobic range. The CF layer was coated onto the hairy PP surface with commercial equipment, DRIE, to enhance the liquid repellency. The contact angle of the water methanol liquid

PAGE 155

155 ( < 35.2 mN/m) was raised from 60 to around 140 However, the surface energy of the coated samples was still not low enough to have a high contact angle with non polar liquids. Suggestions for Future Work The R ole of E lasticity on W ater R epellency Althou gh the superior water repellency of the synthetic hairy surface is believed to be mainly based on the physical interaction of the interface, no direct evidence was available in this study. Direct observation on the interface between the artificial hairs an d the droplet will be the best investigation to solve th is mystery. However, it will be a challenge for optical microscope s to observe an event this size Environmental scanning electron microscopes (E SEM) may provide a better solution. T he imaging chambe r pressure of the E SEM can be as high as 10 2 atm, and it can be filled with water vapor, keep ing the sample hydrated over an extended period. It has already been operated in this fashion to examine the interaction of droplets and the lotus leaf The tech nique will be able to discover the direct interaction of droplets and the hair arrays. Quantitative Study of Self Cleaning Effect As demonstrated in Chapter 4, the synthetic hairy surface features a strong, self cleaning ability. Although this effect has b een a subject of considerable scientific interest for over a decade, compared to the efforts in making surfaces superhydrophobic, insightful studies on understanding this effect are still very rare. The synthetic hairy surfaces provide appropriate media fo r the study, as the adequate surface area of samples with different morphology and hydrophobicity are easily prepared. The research should rule out the factors and relationships, such as particle size, particle hydrophobicity, and contact angle hysteresis vs. cleaning efficiency. The

PAGE 156

156 challenge will be mainly on how to quantify the cleaning ability, since the mass of washed off dirt particles will be small Observation under a microscope, either through fluorescent particles or direct count, is a possible so lution

PAGE 157

157 LIST OF REFERENCES 1. Young, T. An Essay on the Cohesion of Fluids Phil. Trans. R. Soc. Lond. 1804 95 65 87 2. Feng, X. J. ; Jiang, L. Design and creation of superwetting/antiwetting surfaces Adv Mater 2006 18 3063 3078 3. Ma, M. L. ; H ill, R. M. Superhydrophobic surfaces Curr Opin Colloid In 2006 11 193 202 4. Zhang, X.; Shi, F.; Niu, J.; Jiang, Y. G. Wang, Z. Q. Superhydrophobic surfaces: from structural control to functional application J Mater Chem 2008 18 621 633 5. Nosonovsky, M. ; Bhushan, B. Superhydrophobic surfaces and emerging applications: Non adhesion, energy, green engineering Curr Opin Colloid In 2009 14 270 280 6. Roach, P.; Shirtcliffe, N. J. ; Newton, M. I. Progess in superhydrophobic surface develo pment Soft Matter 2008 4 224 240 7. Genzer, J. ; Efimenko, K. Recent developments in superhydrophobic surfaces and their relevance to marine fouling: a review Biofouling 2006 22 339 360 8. Zhang, H.; Lamb, R. Lewis, J. Engineering nanoscale roug hness on hydrophobic surface preliminary assessment of fouling behaviour Sci Technol Adv Mat 2005 6 236 239 9. Daniello, R. J.; Waterhouse, N. E. ; Rothstein, J. P. Drag reduction in turbulent flows over superhydrophobic surfaces Phys Fluids 20 09 21 085103. 10. Barthlott, W. ; Neinhuis, C. Purity of the sacred lotus, or escape from contamination in biological surfaces Planta 1997 202 1 8 11. N einhuis, C. ; Barthlott, W. Characterization and distribution of water repellent, self cleaning p lant surfaces Ann Bot London 1997 79 667 677 12. Li u, Y. Y.; Chen, X. Q. ; Xin, J. H. Can superhydro phobic surfaces repel hot water? J Mater Chem 2009 19 5602 5611 13. Otten, A. ; Herminghaus, S. How plants keep dry: A physicist's point of view Langmuir 2004 20 2405 2408 14. Wenzel, R. N. Resistance of solid surfaces to wetting by water Ind Eng Chem 1936 28 988 994

PAGE 158

158 15. Holdgate, M. W. The w etting of i nsect c uticles by w ater J Exp Biol 1955 32 591 617 16. Hu, D. L.; Chan, B. ; B ush, J. W. M. The hydrodynamics of water strider locomotion Nature 2003 424 663 666 17. Thorpe, W. H. ; Crisp, D. J. Studies on p lastron r espiration .1. The b iology of a phelocheirus [Hemiptera, Aphelocheiridae (Naucoridae)] and the m echanism of p lastro n r etention J Exp Biol 1947 24 227 269 18. Hinton, H. E. Plastron r espiration in b ugs and b eetles J Insect Physiol 1976 22 1529 1550. 19. Gao, X. F. ; Jiang, L. Water repellent legs of water striders Nature 2004 432 36 36 20. Comstock J H Note on the respiration of aquatic bugs. Am Nat 1887 21 577 578 21. del Campo, A. ; Arzt, E. Fabrication approaches for generating complex micro and nanopatterns on polymeric surfaces Chem Rev 2008 108 911 945 22. Mock, U.; Forster, R.; Men z, W. ; Ruhe, J. Towards ultrahydrophobic surfaces: a biomimetic approach J Phys Condens Mat 2005 17 S639 S648 23. Goodwyn, P. P.; De Souza, E.; Fujisaki, K. ; Gorb, S. Moulding technique demonstrates the contribution of surface geometry to the super hydrophobic properties of the surface of a water strider Acta Biomater 2008 4 766 770 24. Dupre, A. Theorie Mechanique de la Chaleur; Gauthier Villars: Paris, 1869 36W 2 5 Cassie, A. B. D. ; Baxter, S. Wettability of porous surfaces. T Faraday Soc 1944 40 0546 0550 26 Bico, J.; Marzolin, C. ; Quere, D. Pearl drops Europhys Lett 1999 47 220 226 27 Lafuma, A. ; Quere, D. Superhydrophobic states Nat Mater 2003 2 457 460 28 Quere, D. ; Non sticking drops Rep Prog Phys 2005 68 2495 2532 29 Dupuis, A. ; Yeomans, J. M. Modeling droplets on superhydrophobic surfaces: Equilibrium states and transitions Langmuir 2005 21 2624 2629 30 Patankar, N. A. Transition between superhydrophobic states on rough surfaces Langmuir 2004 20 7097 7102

PAGE 159

159 31 Patankar, N. A. Mimicking the lotus effect: Influence of double roughness structures and slender pillars Langmuir 2004 20 8209 8213 32 Marmur, A. The lotus effect: Superhydrophobicity and metastability Langmuir 2004 20 3517 35 19 33 Bico, J.; Thiele, U. ; Quere, D. Wetting of textured surfaces Colloid Surface A 2002 206 41 46 34 Miwa, M.; Nakajima, A.; Fujishima, A.; Hashimoto, K. ; Watanabe, T. Effects of the surface roughness on sliding angles of water droplets on supe rhydrophobic surfaces Langmuir 2000 16 5754 5760 35 McHale, G.; Shirtcliffe, N. J. ; Newton, M. I. Contact angle hysteresis on super hydrophobic surfaces Langmuir 2004 20 10146 10149 36 Furmidge, C. G. Studies at Phase Interfaces 1. Sliding of L iquid Drops on Solid Surfaces and a Theory for Spray Retention J. Coll. Sci. Imp. U. Tok 1962 17 309 324. 37 Dettre, R. H.; Johnson R. E. Jr. Contact Angle, Wettability and Adhesion; Adv. Chem. Ser. ; Am er. Chem. Soc., Washington, DC: 1964 ; 43 p 136 144. 38 Extrand, C. W. Model for contact angles and hysteresis on rough and ultraphobic surfaces Langmuir 2002 18 7991 7999 39 Yoshimitsu, Z.; Nakajima, A.; Watanabe, T. ; Hashimoto, K. Effects of surface structure on the hydrophobicity and slidin g behavior of water droplets Langmuir 2002 18 5818 5822 40 Wolfram, E. ; Faust, R. Wetting, Spreading, and Adhesion Academic, London : 1978 p 213. 41 Holloway, P. J. Surface factors affecting the wetting of leaves. Pestic. Sci. 1970 1 156 163. 4 2 Koch K .; Dommisse A .; Barthlott W Chemistry and crystal growth of plant wax tubules of Lotus (Nelumbo nucifera) and Nasturtium (Tropaeolum majus) leaves on technical substrates Crystl. Growth Des. 2006 6 2571 2578 43 Gao L .; McCarthy T J Th e l otus e ff ect e xplained: Two r easons w hy t wo l ength s cales of t opography are Important Langmuir 2006 22 2966 2967 44 Nosonovsky M, Multiscale roughness and stability of superhydrophobic biomimetic interfaces Langmuir 2007 23 3157 3161

PAGE 160

160 45. Noson ovsky M Model for s olid l iquid and s olid s olid f riction for r ough s urfaces with a dhesion h ysteresis J. Chem. Phys. 2007 126 224701 46 Nosonovsky M .; Bhushan B Hierarchical r oughness m akes s uperhydrophobic s urfaces s table Microelectronic Eng. 2 007 84 382 386 47 Nosonovsky M .; Bhushan B Hierarchical roughness optimization for b iomimetic superhydrophobic surfaces Ultramicroscopy 2007 107 969 979 48 Nosonovsky M .; Bhushan B Multiscale f riction m echanisms and h ierarchical s urfaces i n n ano and b io tribology. Mater. Sci. Eng. R 2007 58 162 193 49 Nosonovsky M .; Bhushan B Roughness induced superhydrophobicity: a way to design non adhesive surfaces J. Phys.: Condens. Matter 2008 20 225009 50 Furstner R .; Barthlott W .; Neinhuis C .; Walzel P Wetting and s elf c leaning p roperties of a rtificial s uperhydrophobic s urfaces Langmuir 2005 21 956 961 51 Thorpe, W. H. ; Crisp, D. J. Studies on plastron respiration III. The orientation responses of Apelocheirus [Hemiptera, Aphelocheiridae (Naucoridae)] in relation to plastron respiration; together with an account of specialized pressure receptors in aquatic insects. J. Exp. Biol. 1947 24 310 328. 5 2 Wigglesworth, V. B. The physiology of the cuticle and of ecdysis in Rhod nius prolixus (Triatomidae, Hemiptera); with special reference to the function of the oenocytes and of the dermal glands. Q. J. Microsc. Sci. 1933 76 269 318. 5 3 Wigglesworth, V. B. Transpiration through the cuticle of insects. J. Exp. Biol. 1945 21 97 114. 5 4 Pal, R. The wetting of insect cuticle. Bull. Ent. Res. 1951 51 121 139. 5 5 Stratton, G. E. ; Suter, R. B. ; Miller, P. R. Taxonomic variation among spiders in the ability to repel water: surface adhesion and hair density. J. Arachnol. 2004 32 11 21. 5 6 Bush, J. W. M.; Hu, D. L. Prakash, M. The integument of water walking arthropods: Form and function Adv Insect Physiol 2007 34 117 192 5 7 Lamoral, B. H. On the ecology and habitat adaptations of two intertidal spiders, Desis formidab (Kommetjie, Cape Peninsula), with notes on the occurrence of two other spiders. Ann. Natal. Mus. 1968 20 151 193. 58 Lange N .A. ; Dean J A Lange's Handbook of chemistry 11th ed. McGraw Hill N Y: 1973 p 1661 1665.

PAGE 161

161 59 ASTM D638, Standard t est m ethod for t ensile p roperties of p lastics ASTM International: USA, 2008 60 Zheng, H. P.; Lee, I. S.; Rubner, M. F.; Hammond, P. T. Two component particle arrays on patterned polye lectrolyte multilayer t emplates. Adv. Mater. 2002 14 569. 61 Liu, Z. F.; Shen, Z. Y.; Zhu, T.; Hou, S. F.; Ying, L. Z.; Shi, Z. J.; Gu, Z. N. Organizing s ingle w alled c arbon n anotubes on g old u sing a w et c hemical s elf a ssembling t echnique. Lagnmuir 2000 16 3569 3573. 62 Wu, B.; Zhang, J.; Wei, J.; Cai, S. M.; Liu, Z. F. Chemical a lignment of o xidatively s hortened s ingle w alled c arbon n anotubes on s ilver s urface J. Phys. Chem. B 2001 10 5, 5075 5078. 6 3 Kim B .; Sigmund W M. Self a lignment of s hortened m ultiwall c arbo n n anotubes on p olyelectrolyte l ayers Langmuir 2003 19 4848 4851 6 4 Lau, K. K. S.; Bico, J.; Teo, K. B. K.; Chhowalla, M.; Amaratunga, G. A. J.; Milne, W. I.; McKinley, G. H. Gleason, K. K. Superhydrophobic carbon nanotube forests Nano Lett 2003 3 1701 1705 6 5 Journet C. ; Moulinet S. ; Ybert, C. ; Purcell S. T. ; Bocquet L. Contact angle measurements on superhydrophobic carbon nanotube forests: Effect of fluid pressure Europhys. Lett. 2005 71 104 109 6 6 Joseph, P. ; Cottin Bizonne, C.; B enoi t, J. M.; Ybert, C. ; Journet, C. ; Tabeling, P. ; Bocquet L. Slippage of w ater p ast s uperhydrophobic c arbon n anotube f orests in m icrochannels Phys. Rev. Lett. 2006 97 156104 6 7 Demczyk, B. G.; Wang, Y. M.; Cumings, J.; Hetman, M.; Han, W.; Zettl A.; Ritchie, R. O. D irect mechanical measurement of the tensile strength and elastic modulus of multiwalled carbon nanotubes. Mat. Sci. Eng. A 2002 334, 173 178 68 Quake, S. R. ; Scherer, A. From micro to nanofabrication with soft materials Science 20 00 290 1536 1540 69 Roca Cusachs, P.; Rico, F.; Martinez, E.; Toset, J.; Farre, R. Navajas, D. Stability of microfabricated high aspect ratio structures in poly(dimethylsiloxane) Langmuir 2005 21 5542 5548. 70 Sharp, K. G.; Blackman, G. S.; Glassm aker, N. J.; Jagota, A. Hui, C. Y. Effect of stamp deformation on the quality of microcontact printing: Theory and experiment Langmuir 2004 20 6430 6438

PAGE 162

162 7 1 Moore E. P. Polypropylene Handbook. Polymerization, Characterization, Properties, Processing, Applications Hanser Publishers: NY, 1996 7 2 Kang, T. S.; Smith, A. P.; Taylor, B. E. Durstock, M. F. Fabrication of Highly Ordered TiO2 Nanotube Arrays and Their Use in Dye Sensitized Solar Cells Nano Lett 2009 9 601 606 7 3 Steinhart, M.; Wehrs pohn, R. B.; Gosele, U. Wendorff, J. H. Nanotubes by template wetting: A modular assembly system Ange Chem Interna t. Ed 2004 43 1334 1344 7 4 Zhang, Q. M. ; Bharti, V. ; Kavarnos, G. ; Schwartz, M. (Ed.), Poly (Vinylidene Fluoride) (PVDF) and its Copol ymers Encyclopedia of Smart Mat erials. John Wiley & Sons : NY 2002, 1 2 p 807 825. 7 5 Geim, A. K.; Dubonos, S. V.; Grigorieva, I. V.; Novoselov, K. S.; Zhukov, A. A. ; Shapoval, S. Y. Microfabricated adhesive mimicking gecko foot hair Nat Mater 2003 2 461 463 7 6 Baeurle, S. A.; Hotta, A. ; Gusev, A. A. On the glassy state of multiphase and pure polymer materials. Polymer 2006 47 6243 6253 77 Biswas, S. K. ; Vijayan, K. ; Friction and Wear of PTFE a Review Wear 1992 158 193 21 1. 78 Kwak, G. ; Seol, M.; Tak, Y. ; Yong, K. Superhydrophobic ZnO n anowire s urface: c hemical m odification and e ffects of UV i rradiation. J Phys Chem C. 2009 113 12085 1208 9. 79 Li, L. H. ; Chen, Y. Superhydrophobic Properties of Nonaligned Boron Nitride Nanotube Fi lms. Langmuir 2010 26 5135 5140 80 Yeh, K. Y.; Cho, K. H. Chen, L. J. Preparation of s uperhydrophobic s urfaces of h ierarchical s tructure of h ybrid from n anoparticles and r egular p illar l ike p attern Langmuir 2009 25 14187 14194 8 1 Ma, M.; Hill, R M. Rutledge, G. C. A Review of Recent Results on Superhydrophobic m aterials b ased on m icro and n anofibers J Adhes Sci Technol 2008 22 1799 1817 8 2 Majidi, C.; Groff, R. E.; Maeno, Y.; Schubert, B.; Baek, S.; Bush, B.; Maboudian, R.; Gravish, N .; Wilkinson, M.; Autumn, K. Fearing, R. S. High friction from a stiff polymer using microfiber arrays Phys Rev Lett 2006 97 076103. 8 3 Lee, J. ; Fearing, R. S. ; Contact self cleaning of synthetic gecko adhesive from polymer microfibers Langmuir 2 008 24 10587 10591

PAGE 163

163 8 4 Autumn, K.; Majidi, C.; Groff, R. E.; Dittmore, A. Fearing, R. Effective elastic modulus of isolated gecko setal arrays J Exp Biol 2006 209 3558 3568 8 5 Guo, C. W.; Feng, L.; Zhai, J.; Wang, G. J.; Song, Y. L.; Jiang, L. ; Zhu, D. B. Large area fabrication of a nanostructure induced hydrophobic surface from a hydrophilic polymer Chem Phys Chem. 2004 5 750 753 86. Wu, Y. Y.; Bekke, M.; Inoue, Y.; Sugimura, H.; Kitaguchi, H.; Liu, C. S.; Takai, O. Mechanical durability of ultra water repellent thin film by microwave plasma enhanced CVD Thin Solid Films 2004 457 122 127. 87. Cui, Z.; Wang, Q. J.; Xiao, Y.; Su, C. H.; Chen, Q. M. The stability of super hydrophobic surfaces tested by high speed current scouring Appl S urf Sci 2008 254 2911 2916. 88. Marmur, A. From hygrophilic to superhygrophobic: Theoretical conditions for making high contact angle surfaces from low contact angle materials Langmuir 2008 24 7573 7579 89. Stalder, A.F.; Kulik, G.; Sage, D.; Barb ieri, L.; Hoffmann, P. Colloids Surf., A: Physicochem. Eng. Aspects 2006 286 92 103 90. Kass M. ; A. Witkin, D. Terzopoulos, Snakes: Active contour models Int. J. Comput. Vis. 1988 2 321 331. 91. The program ImageJ can be downloaded from: http://rsbweb.nih.gov/ij/ 92. The program DropSnake can be downloaded from : http://bigwww.epfl.ch/demo/dropanalysis/ 93. Bernardino, N. R.; Blickle, V.; Dietrich, S. Wetting of s urfaces c overed by e lastic h airs Langmuir 2010 26 7233 7241 94. Quere, D. Wetting and roughness Annu Rev Mater Res 2008 38 71 99 9 5 Jung Y. C. ; Bhushan B. Dynamic e f f ects i nduced t ransition of d roplets on b iomimetic s uperhydrophobic s urfaces Langmuir 2009 25 9208 9218 96. Erbil, H. Y.; Demirel, A. L. ; Avci, Y. ; Mert, O. Transformation of a simple plastic into a superhydrophobic surface Science 2003 299 1377 1380 97. Gao, L. C. ; McCarthy, T. J. A perfectly hydrophobic surface J Am Chem Soc 2006 128 9052 9053 98. Gao, L. C. ; McCarthy, T. J. A commercially available perfectly hydrophobic material Langmuir 2007 23 9125 9127

PAGE 164

164 99. The videos can be accessed through ACS Langmuir website on: http://pubs.acs.org/doi/abs/10.1021/la9038 13g 100. J in, M. H.; Feng, X. J.; Feng, L.; Sun, T. L.; Zhai, J.; Li, T. J. Jiang, L. Superhydrophobic aligned polystyrene nanotube films with high adhesive force Adv Mater 2005 17 1977 101. Lejeune, M.; Lacroix, L. M.; Bretagnol, F.; Valsesia, A.; Colpo, P. Rossi, F. Plasma based processes for surface wettability modification Langmuir 2006 22 3057 3061 102. Andersson, H.; van der Wijngaart, W.; Griss, P.; Niklaus, F. Stemme, G. Hydrophobic valves of plasma deposited octafluorocyclobutane in DR IE channels Sensor Actuat B Chem 2001 75 136 141 103. Zhuang, Y. X. Menon, A. Wettability and thermal stability of fluorocarbon films deposited by deep reactive ion etching Journal of Vacuum Science & Technology A 2005 23 434 439 104. Labelle, C B. ; R Opila, ;. Kornblitc A Plasma deposition of fluorocarbon thin films from c C4F8 using pulsed and continuous rf excitation J. Vac. Sci. Technol. A 2005 23 190 196. 105. Gnanappa, A. K.; Slattery, O.; Peters, F.; O'Murchu, C.; O'Mathuna, C.; Fahey R.; Taylor, J. A. Krupenkin, T. N. Factors influencing adhesion of fluorocarbon (FC) thin film on silicon substrate Thin Solid Films 2008 516 5673 5680 106. Coulson, S. R.; Woodward, I. S.; Badyal, J. P. S.; Brewer, S. A. Willis, C. Ultralow surfac e energy plasma polymer films Chem Mater 2000 12 2031 2038 107. Coulson, S. R.; Woodward, I. S.; Badyal, J. P. S.; Brewer, S. A. Willis, C. Plasmachemical functionalization of solid surfaces with low surface energy perfluorocarbon chains Langmuir 2000 16 6287 6293 108. Marmur, A. From hygrophilic to superhygrophobic: Theoretical conditions for making high contact angle surfaces from low contact angle materials Langmuir 2008 24 7573 7579 109 Tuteja, A.; Choi, W.; Ma, M. L.; Mabry, J. M.; M azzella, S. A.; Rutledge, G. C.; McKinley, G. H. Cohen, R. E. Designing superoleophobic surfaces Science 2007 318 1618 1622 110. Tuteja, A.; Choi, W.; Mabry, J. M.; McKinley, G. H. Cohen, R. E. Robust omniphobic surfaces P Natl Acad Sci USA 2008 105 18200 18205

PAGE 165

165 BIOGRAPHICAL SKETCH Shu Hau Hsu was born in a rural township on the coast in northern Taiwan. He did his undergraduate and master studies both in M aterials S cience D epartment at National Cheng Kung University, Tainan, Taiwan. During his u ndergraduate, he was awarded Dr. Paul C.W. Chu Material Science Scholarship, a prestigious four year national scholar ship, which ultimately led him to explore superconductivity. Together with his team mates, they have built up several systems to demonstra te the super conducting magnetic levitation. They also have a video on YouTube showing how this phenomenal event works. The video has been watched over 750,000 times (as of Aug. 2010). What he ha d enjoyed most there was being an instructor for a science ca mp where he met hundreds of high school students each summer, and they brainstorm ed together to bring up any possible ideas of utilizing this phenomenon. After a two year military service and another two years as a research assistant in Academia Sinica, h e continued his education in pursuit of a Doctor of Philosophy degree in the Department of Ma terials Science and Engineering at University of Florida, Gainesville. He worked with Dr. Wolfgang Sigmund in UF since 2006 on various topics including colloidal pr ocessing, novel ceramic casting and his dissertation work another filed of super property Superhydrophobicity. They are the first one that successfully developed an artificial hairy surface exhibiting perfect hydrophobicity. This work also made him on the national and international media, including USA Today and Scientific American. He hopes that he can find a way of using his knowledge well and always have the enthusiasm for delivering the aspects of science to the young generation.