University of Massachusetts Amherst ScholarWorks@UMass Amherst
Doctoral Dissertations 1896 - February 2014
1-1-2001
Ultrahydrophobic surfaces : effects of topography on wettability.
Didem, Oner University of Massachusetts Amherst
Follow this and additional works at: https://scholarworks.umass.edu/dissertations_1
Recommended Citation Oner, Didem,, "Ultrahydrophobic surfaces : effects of topography on wettability." (2001). Doctoral Dissertations 1896 - February 2014. 1022. https://scholarworks.umass.edu/dissertations_1/1022
This Open Access Dissertation is brought to you for free and open access by ScholarWorks@UMass Amherst. It has been accepted for inclusion in Doctoral Dissertations 1896 - February 2014 by an authorized administrator of ScholarWorks@UMass Amherst. For more information, please contact [email protected].
ULTRAHYDROPHOBIC SURFACES- EFFECTS OF TOPOGRAPHY ON WETTABILITY
A Dissertation Presented
by
DIDEM ONER
Submitted to the Graduate School of the University of Massachusetts Amherst in partial fulfillment of the requirements for the degree of
DOCTOR OF PHILOSOPHY
September 2001
Polymer Science and Engineering © Copyright by Didem Oner 2001
All Rights Reserved ULTRAHYDROPHOBIC SURFACES- EFFECTS OF TOPOGRAPHY ON WETTABILITY
A Dissertation Presented
by
DIDEM ONER
Approved as to style and content by:
Thomas J. McCarthy, Chair
Mark T. Tuominen, Member
Thomas J. McCarthy, Department Head Polymer Science and Engineering To Nesli§ah, Liitfi and Kerem Oner, and Sinan Deliormanli with all my heart ACKNOWLEDGMENTS
I would like to acknowledge advisor, my Dr. Tom McCarthy, for all his
enthusiasm, advice and understanding during the last 4 years while we worked on this
exciting Ph.D. thesis. Thank you Tom, for making this so special. Also, I want to thank
Dr. Tuominen and Dr. Strey for their time and input.
It would not have been possible if the McCarthy group were not there.
Especially, I want to thank: Jack for helping me out with many things in the lab; Chris for
his help, input and his friendship; Jeff for our interesting conversations (from nutrition to
ultrahydrophobicity); Xinqiao (a great office mate); Margarita for the wonderful
memories we shared in such a short period of time; Misha for his sincere help; Chuntao;
Terry for our discussions in teaching me how to become an 'engineer'; Kevin; Ebru; like
and Wei (was always so supportive). Special thanks to the PS&E faculty and staff for all
the things they have done- from teaching in class to the kind support they provided to
make me feel like home. I sincerely thank Ho-Cheol, Robin, Firat, Elif, Sabri and Serkan
for making this journey so enjoyable. Also, I will never forget the support from the very
special people in my life, Mutlu, Kutay, Iskender and Emel Yilgor.
I want to thank my brother, Kerem, for our unforgettable memories and for his
patience during our 'Amherst years' together. Also, I am always grateful to Sinan's
family for their sincere support and love.
And to Sinan Deliormanli, my other half. It was him who always believed in me,
filled me with his endless love and care. Thank you, hayatku§um.
Most of all, I feel so blessed to have the everlasting love of the most wonderful
parents in the universe, Nesli§ah and Liitfi Oner. Thanking them will never be enough.
V ABSTRACT
ULTRAHYDROPHOBIC SURFACES- EFFECTS OF TOPOGRAPHY ON WETTABILITY
SEPTEMBER 2001
DIDEM ONER, B.S., KOC UNIVERSITY
M.S., UNIVERSITY OF MASSACHUSETTS AMHERST
Ph.D., UNIVERSITY OF MASSACHUSETTS AMHERST
Directed by: Professor Thomas J. McCarthy
The overall objective of this Ph.D. thesis is to control the wetting behavior of surfaces by exploring the effects of topography on wettability, and ultimately make ultrahydrophobic surfaces. Tliree different approaches were taken in preparing rough surfaces with controlled wettability. The first approach involved the use of photolithography that resulted in a series of silicon surfaces with different post size, shape and separation (Chapter 2). The second approach was the surface modification of low density polyethylene (Chapter 3). The last one was to adsorb polystyrene colloids
with different diameters onto polyelectrolyte multilayers (Chapter 4).
The wettability of the patterned silicon surfaces prepared by photohthography and hydrophobized using reactive silane chemistry was explored. Surfaces containing square posts with X-Y dimensions of 2 )iim-32 )im exhibited ultrahydrophobic behavior with high advancing and receding contact angles. The contact angles were independent of the post height and surface chemistry. Surfaces with larger posts were not ultrahydrophobic-
vi water droplets pinned on these surfaces, hicreasmg the separation between the posts caused increases in receding contact angles up to the point that water intruded between the posts. Changing the shape of the posts also increased the receding contact angles due to the more contorted contact lines.
The oxidative etching of low density polyethylene films followed by uniaxial or biaxial tension resulted in the formation of micron size fragments. 5 minutes oxidized films had smaller islands than the 1 5 minutes oxidized ones. The fragments became smaller and more distant from each other with increase in strain that affected the wettability of the surfaces. At 400%, the films exhibited ultrahydrophobic behavior. At a higher strain, the islands were very small and apart from each other, the receding contact angle dropped significantly.
Submicron and micron scale rough surfaces were prepared by adsorbing polystyrene colloids onto polyelectrolyte muUilayers. The negatively charged colloids were efficiently adsorbed onto the outermost cationic polyelectrolyte surface, showing no aggregation. The advancing water contact angle increased and the receding contact angle decreased as the surface coverage increased, resulting in a remarkable hysteresis of
-122°. Thus, hydrophobic surfaces could not be achieved by making rough surfaces by colloidal adsorption.
vii TABLE OF CONTENTS
Page ACKNOWLEDGMENTS V ABSTRACT VI LIST OF TABLES XI
LIST OF FIGURES ••• Xlll CHAPTER
1 . INTRODUCTION 1
Overview I Wetting Behavior of Surfaces 3
Contact Angle Hysteresis 8 Topography of Roughness 10 Three-Phase Contact Line...: 12
Surface Characterization Techniques 14
Dynamic Contact Angle 14 Atomic Force Microscopy 15 X-Ray Photoelectron Spectroscopy 17
Notes and References 20
2. MODEL LITHOGRAPHIC SURFACES TO EXPLORE ULTRAHYDROPHOBICITY 22
Introduction 22 Photolithography 23 Dry Etching 25 Reactive Silane Chemistry 29 Experimental 31
Materials 31 Preparation of Silicon Substrates 32 Reaction of Silicon Substrates with Organosilanes 33
Results and Discussions 34
viii Reaction of Silicon Substrates with Organosilanes 34 Critical Size Scale of Roughness for Hydrophobicity 35 Structure of 3-Phase Contact Line '""^ 4j Laplace Pressure Critical Tilt Angle
Conclusions Notes and References
SURFACE MODIFICATION OF LOW DENSITY POLYETHYLENE 55
Motivation Surface Treatments of Low Density Polyethylene 56 Deformation of Thin Film Coatings Upon Uniaxial and Biaxial Tension 59 Experimental
Materials 64 General Methods 64 Preparation of Polyethylene Substrates 65 Oxidation of Polyethylene Substrates 65 Uniaxial Streching of LDPE 65 Biaxial Streching of LDPE 66 Self-Assembled Monolayers of 1-dodecanethiol 66
Results and Discussions 67
Oxidation of Polyethylene Substrates (o-LDPE) 67 Uniaxial Deformation 69
Change in Surface Roughness of o-LDPE with Strain 72
Biaxial Deformation of o-LDPE 76 Deformation of Gold Coating Bonded to LDPE 78
Conclusions 80 Notes and References 82
4. ADSORPTION OF POLYSTYRENE COLLOIDS ONTO POLYELECTROLYTE MULTILAYER ASSEMBLIES AND WETTING BEHAVIOR OF THESE ROUGH SURFACES 86
Introduction 86 Polyelectrolyte Multilayer Assemblies 87 Adsorption of Colloids onto Polyelectrolytes 89 Self-Assembled Monolayers of Alkanethiols on Gold 91 Experimental ^5
ix Preparation of Silicon Substrates c)y Pretreatment of Silicon Substrates 9-7
Reaction of Silicon Substrates with (4-aminobutyl)dimethylmethoxysilane 97 Layer-by-Layer Deposition of Polyelectrolytes '95
Adsorption of Polystyrene Colloids onto Polyelectrolyte Multilayers 98
Kinetic Study of the Adsorption Behavior of Colloids 99
Solution Casting Polystyrene Colloids onto Polyelectrolyte Multilayers 99 Gold Coating and Self-Assembled Monolayers (SAMs) of 1-dodecanethiol jqq
Results and Discussions 1 qq
Pretreatment of Silicon Substrates 1 00 Adsorption of Polystyrene Colloids onto Polyelectrolyte Multilayers.... 102
Kinetic Study of the Adsorption Behavior of Colloids 104
Summary of the Rough Surfaces Prepared 114 Solution Casting Polystyrene Colloids onto Polyelectrolyte Multilayers 118
Conclusions 120 Notes and References 122
BIBLIOGRAPHY 126 1
LIST OF TABLES
Table Page
2. Atomic concentration (XPS) 1 of silicon/silicon oxide surfaces reacted with DMDCS, ODMCS and FDDCS at 15° 75° and takeoff angles 35
2.2 Water contact angle data for silane-modified hexagonally arrayed 40 - high square post surfaces described in figures 2.7 and 2.8 38
2.3 Water contact angle data for ODMCS-modified surfaces containing 1 6 x 16 ^im, x 32 32 |im, 64 |im x 64 |im, 128 x 128 ^im square posts of different height 4q
2.4 Water contact angle data for silane-modified surfaces containing different shaped posts and 8 |im x 8 |nm square posts spaced at various distances (described in figures 2.10 and 2.12) 45
2.5 Laplace pressure calculated using equation 4.1 48
2.6 Tilt angle and hysteresis exhibited by the DMDCS-modified silicon surfaces 50
3.1 Tensile failure modes for thin brittle films^^ 60
5° 3.2 XPS atomic concentration data (at 1 and 75° takeoff angles) for LDPE and
oxidized LDPE (o-LDPE) with chromic acid for 1, 10, 30 and 50 minutes at 75 °C 68
3.3 Dynamic water contact angles for oxidized LDPE (o-LDPE) substrates that were then gold-coated and hydrophobized by SAMs of 1-DCT (o-LDPE/Au/l-DCT) 69
4.1 XPS atomic concentration data for a clean silicon substrate, an ABDMS- modified silicon substrate and after building 10 polyelectrolyte layers composed of PSS and PAH, PAH as the outermost layer 102
4.2 XPS atomic concentration data for an ABDMS-modified silicon substrate, and after building 10 polyelectrolyte layers composed of PSS and PAH, PAH as the outermost layer which were then hydrophobized using
self-assembled monolayers of 1-DCT 1 1
4.3 Dynamic contact angles for surfaces with 1 fxm colloids, before and after hydrophobizing with self-assembled monolayers of 1-DCT, using water and hexadecane (HD) as probe fluids, (time ^ 0 is for polyelectrolyte
multilayers with PAH as the outermost layer) 1 13
xi 9
4.4 Dynamic contact angles for surfaces with 0.35 |im, before and after hydrophobizing with self-assembled monolayers of 1-DCT, using water ' and hexadecane (HD) as probe fluids j j 3
4.5 Dynamic contact angles for surfaces prepared by adsorbing polystyrene colloids with the indicated diameters, before and after hydrophobizing with self-assembled monolayers of dodecanethiol (Au/l-DCT), using water and hexadecane (HD) as probe fluids, r is the roughness ratio calculated from the AFM height image with 40 ^im x 40 ^m scanned area 115
4.6 Dynamic contact angles for surfaces prepared by solution casting polystyrene colloids and hydrophobizing with self-assembled monolayers of
1 -dodecanethiol (Au/l-DCT), using water and hexadecane (HD) as probe fluids, r is the roughness ratio calculated from the AFM height image with 40 [im X 40 |um scanned area 1 1
xii 7
LIST OF FIGURES
Figure Page
1 .1 Schematic representation of Young's equation 4
1 .2 Cassic's analysis at high roughness ratios, showing that the water droplet is mostly in contact with air rather than the solid surface'^' 5
1 .3 (a) Wenzel's, Cassie's, and Johnson and Dettre's analyses on contact angle for different roughness ratios. The smooth surface had Bvoung = 120"; (b) Johnson and Dettre's analysis was done using sinusoidally rough surfaces with different roughness ratios'^ 6
1 .4 Water droplets behave differently on tilted surfaces due to the contact angle hysteresis 9
1 .5 Possible three-phase contact line structures on a real surface (— ) and an ideal one(-) 12
1 .6 2-Dimensional (X-Y) representations of two surfaces with the same f| and f2 values, but very different contact line structures. The dark lines are meant to represent possible contact lines 13
1 .7 Schematic representation of dynamic contact angle measurement 15
1 .8 An illustration of an AFM 1
1 .9 Illustration of the photoelectric effect"*^' 18
2. 1 Main lithographic steps in preparing the patterned silica substrates 25
2.2 Cross sections of photoresist etched by plasma or liquid etchants 26
2.3 The reactive ion etcher 28
2.4 The reaction of octyldimethylchlorosilane (and other monofunctional organosi lanes) with silicon dioxide surfaces 30
2.5 The possible reactions of dimethylchlorosilane (and other di functional organosi lanes) with silicon dioxide surfaces 31
2.6 XPS survey spectrum of silicon/silicon oxide surfaces reacted with FDDCS at 75° takeolT angle 34
xni 2.7 2-D,mensional (X-Y) representation of a series of silicon surtiices containing hcxagonally arrayed square posts, X 2, 8, 16, 32, 64 and 128 am Surfaces with post heights (Z) of 20, 40, 60, 80, 100 and 140 jim were prepared 35
2.8 SEM images of the surfaces described in figure 2.7. The water contact angles shown are for DMDCS-modificd surfaces 3(3
2.9 2-dimensional (X-Y) representations of surfaces containing different geometry posts
2. 10 SEM images of the surfaces described in figure 2.9. The water contact angles shown are for DMDCS-modified surfaces 42
2.11 2-Dimensional (X-Y) representation of a series of silicon surfaces containing hexagonally arrayed square posts, X = 8 ^m and heights (Z) = 40 ^im. Post separations were varied as 32 \im x 32 ^m, 32 |im x 80 |im, 80 X fxm 80 |im 43
2.12 SEM images of surfaces containing 8 )im x 8 |Lim square posts with different spacings as described in figure 2.11. The water contact angles shown are for DMDCS-modified surfaces 44
2.13 SEM images of surfaces containing (a) 32 ^m x 32 |im square posts; (b) 32 ).im X 32 |im hollow squares. The water contact angles shown are for DMDCS- modified surfaces 46
2. 1 4 Snapshot of a water droplet on the substrate with star-shaped posts, hydrophobized with DMDCS 47
2. 1 5 Optical micrographs of the Woods metal - silicon - air 3-phase contact line on
the (a) silicon substrate; surfaces named (b) 21 pmSP^" and (c) ,2 ^niSP^" in Table 2.4. The contact line is clearly contorted by the square posts on the surface 47
2.16 The apparatus used to measure the critical tilt angle at which the water droplet starts to roll off the surface 49
2. 1 7 Critical tilt angle as a function of hysteresis exhibited by DMDCS-treated
silicon surfaces. As hysteresis decreased, the critical tilt angle decreased as expressed in Furmidge's equation 50
3.1 Chromic acid oxidation of LDPE 57
3.2 SEM micrographs of (a) LDPE; and LDPE oxidized with chromic acid solution for (b) 60 seconds; (c) 6 minutcs^"^ 58
xiv 3.3 Possible tensile failure mechanisms for thin coatings bonded to a substrate- la) through-thickness cracking; (b) through-thickness cracking followed by substrate failure; (c) through-thickness cracking followed by interfacial delamination 60
3.4 Schematic representation of distribution of stress throughout the metal coating bonded to a polymer support. Stress is maximum at the center of the coating. This grows with stretching and ultimately there is a crack forming at the center of the coating^^ ^
3.5 SEM micrograph of a deformed 3.8 nm platinum coating on PET support. The strain was not reported. The strain rate was 1 mm/min. The bar is 5 \im^^ ....62
3.6 Crack density and crack onset strain as a function of coating thickness." The thicker the coating was, the lower the onset strain and smaller crack density at saturation 53
3.7 SEM micrographs of LDPE substrates that were oxidized with chromic acid for 5, 15, 30 and 50 minutes at 75 °C 68
3.8 SEM micrographs taken in the necking region of unoxidized LDPE substrates that were drawn uniaxially at strains (e) of 260% (extension = 80 mm), 314% (extension = 120 mm) and 570% (extension = 200 mm). The surface microstructure did not change with strain 70
3.9 Strain as a function of extension for LDPE films that were oxidized for (a) 5 minutes and (b) 15 minutes. Strain was calculated by measuring the distance between two marked points on the LDPE film before and after
necking.Taking the ratio of the difference between the final and the initial
distance, over the initial one gives the strain. Strain = s = (final distance - initial distance)/initial distance 71
3.10 SEM micrographs of LDPE substrates that were oxidized with chromic acid for
5 minutes and then drawn uniaxially at strains (s) of 140% (extension = 60 mm), 314%) (extension = 120 mm), 320% (extension = 130 mm), 400% (extension = 150 mm). The surface microstructure changed with strain in the necking region of the polymer. The water contact angles shown were after gold coating and reacfing with 1-DCT (Au/l-DCT) 73
3.11 SEM micrographs of LDPE substrates that were oxidized with chromic acid for 15 minutes and then drawn uniaxially at strains (e) of 30% (extension = 30 = mm), 1 80% (extension = 60 mm), 260% (extension 80 mm), 250% = (extension = 120 mm), 400% (extension = 150 mm) and 570% (extension 200 mm). The surface microstructure changed with strain in the necking region of the polymer. The water contact angles shown were after gold coating and reacting with 1-DCT (Au/l-DCT) 75
XV 3.12 SEM micrographs of LDPE substrates that were oxidized with chromic acid for 5 minutes and then biaxially streched. The surface microstructure changed with strain. Part 1 had the lowest and Part 4 had the highest strain The strain increased in the order: Part 1 < Part < 2 Part 3 < Part 4 > Part 5 The water contact angles shown were after gold coating and reacting with 1-DCT (Au/l-DCT), except for Part 4 whose contact angles could not be measured due to the wrinkled necked region yy
3.13 SEM micrographs of virgin LDPE substrates that were coated with gold and then necked at strains (e) -140% (extension = 60 mm), 310% (extension = 120 mm), 320% (extension = 130 mm), 400% (extension = 150 mm). The distance between the gold stripes increased with strain. The water contact angles shown were after gold coating and reacting with 1-DCT (Au/l-DCT) 79
4. Dependence 1 of the surface coverage on adsorption time at a nanosphere concentration of 3.8 x lO'" mL"' at 10 °C onto (PAH-PSS)3-PAH surface that was prepared in the presence of 2 M NaCl: (a) 780 nm; (b) 548 nm- (c) 84 nm'' \ 90
4.2 Time courses for polystyrene adsorption (diameter: 548 nm) at 10 "C onto (PAH-PSS)3 - PAH surfaces prepared in the presence of NaCl: (a) 2 M; (b) 0.2 M; (c) 0.02 M and (d) 0 M^'
4.3 Schematic illustration of SAMs of thiols on gold substrate 92
4.4 STM image of an SAM of dodecanethiol on Au (111). Intermolecular spacings are 0.50 nm which is V3 times larger than the atomic spacing of gold atoms
(0.288 nm) in Au (1 1 1) planes^^ 93
4.5 Kinetics of adsorption of octadecanethiol (solid symbols) and decanethiol
(open symbols) from ethanol onto gold, (a) EUipsometric thickness, (b) Advancing contact angles of water and hexadecane (HD)^^ 95
4.6 XPS survey of silicon substrate at 15*" takeoff angle (a) reacted with ABDMS for 3 days in toluene; (b) with 10 polyelectrolyte layers, PAH as the outermost layer 101
4.7 AFM section analysis was performed using AFM height images to calculate the roughness ratios of the substrates. The software provided both the surface distance (peak to valley distance) and the horizontal distance. The roughness ratio was calculated using the following formula: r = (surface
distance)^ / (horizontal distance)^ 1 03
XVI 4.8 The change in roughness ratio, calculated from the AFM height images with adsorption time of polystyrene colloids (1 ^im) onto the polyelectrolyte multilayers
4.9 AFM height (left) and phase (right) images of the surfaces obtained when the time of adsorption of polystyrene colloids (1 ^m diameter) onto polyelectrolyte multilayers was for (a) 10, (b) 30, (c) 60 minutes; (d) 2, (e) 4, 6, 8, (h) 10 and (i) .' (0 (g) 24 hours, (continued next page).'. 105
4. 1 The change in 0 roughness ratio, calculated from AFM height images, with adsorption time of polystyrene colloids (0.35 ^m) onto the polyelectrolyte multilayers jQg
4. height (left) 11 AFM and phase (right) images of the surfaces obtained when the time of adsorption of polystyrene colloids (0.35 \im diameter) onto
polyelectrolyte multilayers was for (a) 1/2, (b) 2, (c) 4, (d) 6 and (e) 1 0 hours, (continued next page) 1 09
4. 1 2 The advancing () and the receding () water contact angle with time of adsorption of (a) 1 |im, (b) 0.35 |im colloids after the substrates were hydrophobized with self-assembled monolayers of 1-DCT 112
4. 1 3 AFM height (left) and phase (right) images of the surfaces obtained using
(a) 0.1, (b) 0.35, (c) 0.5, (d) 1, (e) 2, (f) 3.3 and (g) 4.5 |iim diameter polystyrene colloids that were adsorbed onto the polyelectrolyte multilayers for 6 hours, (continued next page) 116
4. 14 AFM height (left) and phase (right) images of the surfaces obtained by using
0.35, 0.5, 1, 2 and 3.3 |um diameter polystyrene colloids that were solution- casted onto polyelectrolyte multilayers, and dried for 3 days 118
xvii CHAPTER 1
INTRODUCTION
Overview
The wetting behavior of surfaces is an important property that has been studied
from both fundamental and practical perspectives. Controlling the wettability of surfaces has been the target in practical applications such as printing plates, anticorrosion paints, cookware, textiles and biomedical devices. There has been renewed interest in this phenomenon and numerous recent reports have been published on preparing water- repellent surfaces.'"' ' However, there are critical factors that determine hydrophobicity that have not been taken into consideration in these papers such as the topological nature of the surface roughness, the 3-phase (solid-liquid-air) contact line structure (shape,
length, continuity of contact, amount of contact), and the contact angle hysteresis
exhibited by the surface. The primary goal of this thesis work is to examine the conditions necessary to observe non-wetting behavior of surfaces. This will ultimately enable the prediction and tuning of wettability. Three different approaches have been taken in designing and preparing model hydrophobic surfaces with different roughness topographies.
The first chapter begins with an introduction to the theoretical and experimental
studies reported on wettability of surfaces. Here, hydrophobicity is considered from a
dynamic point of view and dynamic hydrophobicity is redefined as the minimum force
1 needed to move a droplet across a surface. The critical parameters mentioned above in determining hydrophobicity are discussed in detail. Also, surface characterization techniques such as x-ray photoelectron spectroscopy, dynamic contact angle analysis and atomic force microscopy are reviewed.
The second chapter involves the design and preparation of a series of silicon surfaces with various topographies by photolithography and reactive ion etching techniques. The surfaces contain posts of different size, shape and separation and are hydrophobized by using reactive silane chemistry to achieve alkyl-, fluoroalkyl- and silicone-like surfaces. The wetting behavior of these surfaces is then explored. The primary characterization techniques are scanning electron microscopy and dynamic contact angle analysis.
The motivation behind the approach explored in chapter 3 is to prepare ultrahydrophobic surfaces via a practical route. Surface modification of low density
polyethylene (LDPE) is carried out to achieve ultrahydrophobic surfaces. LDPE film is first treated with chromic acid followed by uniaxial or biaxial stretching. The strained and necked polymer film turns out to have a special surface microstructure with feature size and shape that can be controlled by the extent of strain. The wetting behavior of
these surfaces is analyzed after gold coating and treating with alkanethiols.
In the last chapter, the wetting behavior of a series of rough surfaces prepared by adsorbing polystyrene colloids onto polyelectrolyte multilayers is investigated. The
2 flexibility of altering the surface roughness is studied by varying both the diameter of the colloids and the time of adsorption. In-depth understanding of the adsorption behavior of colloids onto polyelectrolyte multilayers is crucial and this is probed by using atomic force microscopy in TappingMode'^ The wetting behavior of these surfaces is assessed by dynamic contact angle measurements after gold coating and hydrophobizing the surfaces using self-assembled monolayers (SAMs) of alkanethiols.
Wetting Behavior of Surfaces
Wettability of surfaces has been investigated both theoretically and experimentally 19"' since the century. There is an enormous amount of literature from the past 60 years directed at studying and controlling the interaction between fluids and solids.
Thomas Young was the first scientist to investigate the wetting behavior by measuring the static contact angle. ' According to his argument, wettabihty of an ideal-
homogeneous, smooth, rigid and insoluble solid can be expressed as in figure 1.1 and
equation 1.1. Gvoung is the equilibrium (or static) contact angle and it is the balance
between Ysv, Ysi and yiv , the interfacial tensions of solid-vapor, solid-liquid and liquid-
vapor interfaces, respectively. The interfacial tension is defined as the force per unit length on a surface or an interface with the units of mN/m or dyn/cm. Young's equation gives a single value for the contact angle exhibited by a given solid-liquid-vapor system.
However, as will be pointed out later, real surfaces cannot be characterized by only one
3 contact angle.
Figure 1.1. Schematic representation of Young's equation.
Wenzel introduced an equation to explain the effect of surface roughness on
'"'^ wettability.''* He derived the relation between surface roughness and contact angle for uniformly rough surfaces:
cos = r (ysv- ysi )/yiv = r cos 9w Bvoung ( 1 .2) where 9w (Wenzel's angle) and Bvoung are the equilibrium contact angles for the rough
and the smooth surface, respectively. He introduced the term called the roughness factor,
r, which is the ratio of the actual surface area to the geometric projected (or apparent) surface area. He proposed that for a smooth non-wetting (Gvoung > 90°) surface, the rough
surface is more water-repellent than the smooth one. Inversely, for a smooth hydrophilic
surface (Gvoung < 90°), the rough surface is more hydrophilic than the smooth one. Thus,
since the roughness ratio is always greater than one, depending on the equilibrium contact angle of the smooth surface, the surface roughness can increase, or decrease the wettability of the surface. Wenzel's analysis starts to fail at high rougliness ratios when water droplets rest mainly on air and the surface is so called a 'composite' one.
4 Cassic and linxtcr ' cxlciulcd Wcn/crs analysis lo porous siii laccs hy iho following equation:
= cos OCussie I'l COS 0 - h llus equation relates the eontaet angle (Oew) ofa liquid on a composite (air-solid nuxture) surlaee to the conlacl angle on a smooth surface of the same solid (0), and the fraction of wnlcr-solid interface area (fi) to water-air interface = area ((2), (fi + h 1).
According to this analysis, a Teflon surface with a sialic 10" conlacl angle of 1 would exhibit a 73" conlacl angle of 1 if it was made into a 99% porous Tcllon surface.
They proposed thai at high levels of roughness, the probe liquid (lor example, water)
would jirimarily sit on air contained in the jiores of the rough matei iaf as shown in figure
1.2.
Liquid /-Air
( Figure 1 .2. 'assie's analysis at high roughness ratios, showing that the water ilroplel is mostly in contact with air rather than the solid surface.
These analyses and many more put emjihasis on the static conlacl angles when describing Ihe welting behavior of surfaces. Our argument is that the static contact angle has no practical im|)orlance and one should focus on examining Ihe dynamic contact
angle behavior when assessing the surface's true wettability. I lence, the advancing and
5 receding contact angles, measured while the probe lluid is advanced or receded across the surface should be important. The dilTerence between the advancing and the receding contact angles is called the contact angle hysteresis.
Johnson and Dettre studied the wettability ol" sinusoidal surfaces with dilTerent degrees of roughness and tried to explain dynamic contact angle behavior.'^ For a hydrophobic surface having GYoung^ 120°, the hysteresis increases with the roughness ratio up to a critical roughness ratio. At this point, the receding angle (Or) increases to high values. The drops can no longer intrude between the pores and no hysteresis is observed.
Figure 1 .3. (a) Wenzel's, Cassie's, and Johnson and Detlrc's analyses on contact angle for different roughness ratios. The smooth surface had OYmmg = 120"; (b) Johnson and
Dettre \s analysis was done using sinusoidally rough surfaces with different roughness ratios.
6 According to Johnson and Dcllrc's analysis on geometrically rough surfaces, contact angle hysteresis can be qualitatively explained by assuming that the advancing and the receding angles are determined by the balance between the macroscopic vibrational energy of the drop and the heights of energy barriers. For non-composite surfaces (roughness ratio < 2) the energy barriers are high between the metastable slates leading to surfaces with high hysteresis. That is, the drop wants to pin at those metastable states resulting in a high advancing but a low receding contact angle. As surface roughness increases and passes beyond the critical value, the receding angle jumps up resulting in low hysteresis surfaces. The vibrational energy of the drop is now high compared to the height of the barrier. Hence the drop does not remain pinned at this position of metastable equilibrium, rather it moves spontaneously. The key point
Johnson and Dettre missed in their investigation is the parameters that have crucial control on the wettability of surfaces, such as the effect of different topographies of
roughness and the structure of the three phase contact line.
In addition to these theoretical studies, there are numerous experimental reports published on surfaces exhibiting high water contact angles, as discussed in the following sections. Each paper reported a single water contact angle, not the advancing or the
receding one. Even though they called these surfaces hydrophobic, in reality it was impossible to assess the surfaces' true hydrophobicity with only one contact angle value.
7 Contact AiiKle Hysteresis
We believe that dynamic hydrophobicity should be defined as the minimum force needed to move a water droplet across the surface.'' A hydrophobic surface is one on which the water droplet camiot pin when the surface is tilted at an angle. For the water droplet to start rolling off the tilted surface, the downhill side ofthe drop must be at its
advancing angle while the uphill side of it must be at its receding angle. According to
this argument, 1 figure .4 demonstrates the scenarios where water droplets behave differently tilted on surfaces due to the differences in the contact angle hysteresis.
The surfaces in the cases (a) and (b) exhibit water contact angles of 0a/ Or = 120°
80° the / and surface in case (c) exhibits Ba / Or = 70° / 70°. In case (a), a 5 |iL droplet was increased in volume to 10 ^L, thus it intersects the surface at its advancing angle of
120". hi case (b), a 15 |iL droplet was decreased in volume to 10 f.iL, thus it intersects the surface at its receding angle of 80". In case (c), a 10 |iL droplet is in equilibrium with 0a/
Or = 70° / 70°.
When surface (a) is tilted, the downhill side of the droplet can advance, but the
uphill side stays pinned until the receding angle is reached. On surface (b), the uphill side of the droplet can recede, but the downhill side stays pinned until the advancing
angle is reached. On surface (c), the droplet can advance and recede simultaneously.
8 - I ^ I I (a) ^ (b) (c)
Figure 1 .4. Water droplets behave differently on tilted surfaces due to the contact angle hysteresis.
In 1962, Furmidge derived an equation relating the minimum tilt angle at which the droplet starts rolling/sliding off the surface to the contact angle hysteresis:'^
= - (mg) sin e / w ylv (cos Gr cos Ga) (1 .4)
It is clear from these arguments that the difference between the advancing and the
receding angles (hysteresis) and not the absolute values of the contact angles is important for hydrophobicity.
There are various parameters that affect the contact angle hysteresis. One of them
is the heterogeneity of surfaces that are composed of domains of different molecules.
Also, if a surface component is soluble or if there is an interaction between the surface
9 and the liquid such as the adsorption of the liquid onto the surface, these will change the
free energy of the surface and hence result in hysteresis. The parameter that we are concerned about here is the effect of surface roughness on contact angle hysteresis.
Topography of Roughness
Nature has provided us with remarkable examples of the relation between surface
roughness and wettability. Barthlott and his coworkers, as well as others found that wetting of a leaf''^'^° or an insect wing^' was related primarily to the special
microstructures of their surfaces. 'Self-cleaning' or the lotus leaf effect is defined as a phenomenon that occurs if the surface is non-wettable and there is removal of
contaminating species by for instance, rain.
As mentioned in the previous section, the effects of roughness on wettability have
been studied theoretically and experimentally since Wenzel's and Cassie's initial publications. Cassie and Baxter prepared rough surfaces by coating wire gratings with
paraffin wax and reported a low hysteresis surface of 0a / 6r = 152° / 148°. Bartell and
Shepard studied the effects of roughness on contact angle and hysteresis by preparing
reproducible rough surfaces via machining pyramid-shaped asperities with various
heights and angles of inclination into paraffin surfaces.^'' They reported water contact
25° = 10° 99°). angles as high as Oa / Or = 1 58V 1 (the smooth surfaces had Ba / Or 1 /
They also predicted that if the angle of inclination was 90°, for example surfaces of
80°. feathers, furs, leaves and certain textiles, the advancing contact angle should be 1
10 Another important work was by Johnson and Dettre who reported water contact angles of
1 57° S?'* Oa / Br - 59V 1 and 0A / Gr = 1 58V 1 for rough fluorocarbon surfaces, and 0a / Gr
= 158V 153° for rough paraffin wax surfaces which were prepared by spraying solutions
of waxes onto glass slides.^^
In the last decade, there has been a renewed interest in achieving non-wettable
surfaces. For instance, Onda and coworkers prepared 'super water-repellent' fractal
surfaces that exhibited water contact angles as large as 174°.''^ 'Ultra water-repellent'
rough films were made by using microwave plasma-enhanced chemical vapor deposition
of organosilane compounds. The maximum contact angle was -160°.^ Porous alumina
thin films with a roughness of 20-50 nm, hydrophobized by fluoroalkyltrimethoxysilane
yielded 'super-water-repellent alumina' coating films with G = 165°.'*'^ The ion-plated
poly(tetrafluoroethylene) coatings that Drelich et al. prepared also had roughness
dimensions of a few nanometers and these had water contact angles of 150-160°.^ PTFE
powders with diameters of 7.6 \im and 0.83 |nm were used as coatings to make water-
repellent materials with static contact angles of -150°. The key point to all these
examples is that almost all the studies reported a single water contact angle, not the
advancing and the receding ones. Thus, it is impossible to assess the surfaces' true
hydrophobicity from the published contact angle values.
There is another important issue in the literature that needs to be resolved: the
issue of the size scale of roughness needed to impart ultrahydrophobicity. The recent
submicrometer-, reports of surfaces with high contact angles indicated that micrometer-,
11 and nanometer-scale roughness ' imparted this property." Older reports Irom the 194()'s,
1950's, and 196()'s, however, suggested much larger features (tens to hundreds of micrometers) could also function in this manner.^^'^^ By designing and preparing surfaces with a range of si/e scale of roughness, this issue is mvestigated in chapter 2.
Thrcc-Phase Contact Line
The intersection of the three interfaces, namely solid-liquid-vapor interfaces, results in the formation of the three-phase contact line. The three-phase contact line is ; perfect circle if the surface is smooth, homogeneous, planar and rigid. As the droplet advances (recedes) across the surface, the whole contact line advances (recedes) and it starts to deviate from its circular shape and becomes contorted ibr real surfaces that are
rough or heterogeneous, as demonstrated in figure 1 .5.
Figure 1 .5. Possible three-phase contact line structures on a real surface (— ) and an ideal one (-).
12 We argue that the structure cl lhe 3-plu.se contact line is important to dynamic
wettability (Oa, Ok and hysteresis) and C:assie's analysis does not take the 3-phase contact
line structure into account. Surfaces with many dilTcrent topographies can have the same
li and I values, - but can have very dilTerent contact line structures. I'igure 1.6 shows
representations of two surCaccs with the same f, and values 0 (eq 1 .2). Cassie's equation predicts the .same equilibrium contact angles, but the advancing and receding contact angles will be very different on these surfaces becau.se in (a), a nearly continuous contact line can form (pinning the drop), but in (b) the contact line is discontinuous and unstable.
(ii) (b)
I'igure 1.6. 2-Dimensional (X-Y) representations of two surfaces with the same fi and I2 values, but very different contact line structures. The dark lines are meant to represent po.ssible contact lines.
We are aware of very few experimental works in the literature where scientists
have paid attention to the importance of the three-pha.se contact line, its shape and its
26-28 input on the wetting behavior ol'surlaces. Drelich el al. prepared model heterogeneous surfaces with 2..S [un liydrophilic and 2.5 jun hydrophobic stripes that
were alternating and parallel to each other. 1 he gold surface was patternetl with self-
13 assembled monolayers (hydrophobic and a hydrophilic one) via microcontacl prnU.ng with an elastomer stamp developed by Whitesides and his group."^"^' Their argument was that the contortion of the three-phase contact line affected the contact angle. When measuring the contact angle of the surface with strips parallel to the three-phase contact line, the contact line was not contorted, hence the contact angles (both the advancing and the receding) were 2-10" higher. However, the contact line was observed to be contorted when it was perpendicular to the stripes. This led to a significant effect on the advancing and the receding contact angles. Actually, a new theory was developed describing the contact angle behavior for such systems."'^^'^^ j^-^ equation was the modification of the
Cassie's equation, incorporating the line/pseudo-line tension term:
cos e = f, cos 9, + f2 cos 02 - + 5^ n ' y X^ Xi
where = F, Lj) is ysLvi (5 / 5 r, v. Ay, ni the line/pseudo-line tension, F is the free energy of the system; is the yi.v surface tension of the liquid and ri is the radius of the three-phase contact line (in this case, the half width of the stripes) at the i-component of the surface.
Surface Characterization Techniques
Dynamic Contact Angle
Dynamic contact angle measurement is the most effective technique in this thesis
work to characterize the surface wettability. It is a surface-sensitive technique, measuring the wetting behavior of the outermost few angstroms of the surface. In the
McCarthy group, the measurements are done with a Rame-Hart telescopic goniometer
14 and a Gilmont syringe with a flat-tipped 24-gauge needle. A water droplet (or another
probe fluid) is introduced to the surface tliiough a microsyringe. As the volume of the
droplet is increased and the three-phase contact line of the water droplet starts to advance
across the surface, the contact angle is simultaneously measured which gives the
advancing contact angle (Ba). Likewise, the receding angle (6,0 is measured as the
volume of the droplet is reduced and the three-phase contact line starts to recede. The
contact angles are usually measured for 8-12 areas across the surface and the average
contact angles are reported. The accuracy of the measured values is ± 2". The difference
between the advancing and the receding contact angle is the hysteresis exhibited by the
surface. Milli-Q water and hexadecane are the probe fluids for our experiments.
syrmge
Figure 1 .7. Schematic representation of dynamic contact angle measurement,
Atomic Force Microscopy
Scanning probe microscopy (SPM) is a relatively new surface analysis technique, characterizing the probe-sample interactions, based on a sharp probe (usually a silicon
tip) scanning across the surface. SPM consists of two different microscopy techniques,
15 one of them is scanning tunneling microscopy and the other one is atomic force
microscopy (AFM), also called scanning force microscopy (SFM). AFM is a powerful
technique for probing the surface topography of a substrate. This technique was
developed in 1986 by Binning, Gerber, Rohrer, and Weibel at IBM in Zurich,
Switzerland."-^'' There arc three modes of operating an AFM: contact mode, non-contact
mode and TappingMode ".
TappingMode'" AFM is used in this work to characterize the surface
microstructure of the substrates. There is a silicon tip that is attached to a cantilever oscillating at or near its resonance frequency and its amplitude changes between 20-100 nm. The probe is brought closer to the surface so that it can lightly 'tap' on the surface.
As the tip contacts the surface at the bottom of its swing, the oscillation amplitude is reduced. This amplitude change is detected by the split photodiode via the reflection of the laser off the tip. An x-y-z piezo moves the cantilever in the z-direction in order to maintain a constant amplitude and this vertical position at each (x, y) data point is stored by the computer resulting in the topographical image of the surface. These images are height and phase images of the surface.
The AFM software lets the user do several different analyses of the image. In this
thesis work, section analysis software is used. Section analysis provides information
about the surface distance and the horizontal distance of a rough substrate when a line is drawn across the scanned image. Thus, based on 8 different locations over the scanned area, the ratio of (surface distance) to the (horizontal distance) can be calculated and one
16 can oblain the Wcn/cl's roughness ratio. Th.s calculation is schematically shown in
detail in chapter 4.
i eciihiick I ,oop Mamlains Constant Oscillution Amplitude NanoScopc Controller
Detector lilcclronics Scanner
( antilevLM and Tip
Spill Pholodiode Deleclor
Siibslrulc
Figure 1.8. An illustration oTan AVM
X-Ray IMiotoelectron Spectroscopy
X-ray pliotoclectron spectroscopy (XPS), also known as cicclron spectroscopy lor
chemical analysis (luSCA) is a powerful surface analysis technique hased on
photoelectric erfect. It provides a quantitative elemental analysis oflhe surface's
outermost 10- 100 A^\^reven - 200 A^^' depending on the experimental conditions and
the equi|)ment used. 1 he photoelectric effect is the phenemonon where an orhila
17 electron is ejected from an atom upon its interaction with a photon. The electron has a
kinetic energy, Ek, given by the equation:
Eb = hi) - El (1.6)
where Eb is the binding energy of the atomic orbital electron, Ito is the photon energy,
2p 2s photon photoelectron Is
Figure 1 .9. Illustration of the photoelectric effect.^*^
the When X-ray source with monochromatic X-ray photons (in our lab, either Al
Ka or Mg Ka) hits the targeted sample under a high-vacuum environment, photoelectrons are produced. These photoelectrons are detected by the analyzer that counts the number of electrons with different kinetic energies. Every element of the surface has a unique
binding energy and hence kinetic energy. Thus, it is possible to identify the elements according to the peaks in the XPS spectrum that plots photoelectron intensity versus binding energy. Also, the atomic compositions can be calculated by integrating the area under the peaks while taking into consideration the atomic sensitivity factors which depend on the photoelectric cross sections of the atoms.
The sampling depth of the XPS depends on the angle between the sample and the
detector, called the takeoff angle. As the takeoff angle increases, it samples atoms deeper
18 )
in the surface and as the angle decreases, it samples atoms closer to the surface. The
number of electrons detected (N) is proportional to the number of electrons ejected (N.)
at sampling depth (t) by the following equation:
N = N„exp(—!— XsinO (1-7)
where 9 is the takeoff angle and X is the electron mean free path. According to this
equation, 95% of the photoelectrons detected are from the outermost S^sinB. For
instance, Cis electrons have a mean free path"^^ of 14 A, so at 15" takeoff angle, the
ejected photoelectrons that are from the outemiost -10 A can be detected, whereas at 75" takeoff angle, this detection depth increases to the outennost -40 A. Hence, XPS is also useful in quantitatively obtaining the depth profile of a sample by varying the takeoff angles.
19 Notes and References
(1) Onda, T.; Shibuichi, S.; Satoh, N.; Tsujii, K. Langmuir 1996, 12, 2125.
(2) Shibuichi, S.; Onda, T.; Satoh, N.;Tsujii,K. J. Chem. 1996,700, 19512.
Hozumi, (3) A.; Takai, O. Thin Solid Films 1997, 303, 122.
(4) Tadanaga, K.; Katata, N.; Minami, T. J. Am. Ceram. Soc. 1997, 80, 1040.
(5) Tadanaga, K.; Katata, N.; Minami, T. J. Am. Ceram. Soc. 1997, 80, 3213.
Veeramasuneni, (6) S.; Drelich, J.; Miller, J.D.; Yamauchi, G. Prog Ors Coat 1997 57,265.
(7) Yamauchi, G.; Miller, J.D.; Saito, H.; Takai, K.; Ueda, T.; Takazawa, H.; Yamamoto, H.; Nislhi, S. Coll. and Surf.. A: Physicochemical and Engineering. Aspects 1996, 77^, 125.
Ogawa, K.; (8) Soga, M.; Takada, Y.; Nakayama, I. Jpn. J. Appl. Phys Pt 2 - Lett 1993, 32, L614.
(9) Kunugi, Y.; Nonaku, T.; Chong, Y.B.; Watanabe, N. J. Electroanal. Chem 1993 353, 209.
(10) Schakenraad, J.M.; Stokroos, I.; Bartels, H.; Busscher, H.J. Cells and Materials 1992, 2, 193.
(11) Miller, J.D.; Veeramasuneni, S.; Drelich, J.; Yalamanchili, M.R.; Yamauchi, Y. Polym. Eng. Sci. 1996, 36, 1849.
(12) Oner, D.; McCarthy, T. J. Langmuir 2000, 16, 1111.
(13) Young, T. Philos. Trans. R. Soc. London 1805, 95, 65.
(14) Wenzel, R.N., Ind. Eng. Chem. 1936, 27, 105.
(15) Wenzel, R. N., J. Phys. Colloid Chem. 1949, 55, 1466.
(16) Cassie, A. B. D. Discuss. Faraday Soc. 1948, 5, 1 1.
(17) .lohnson, R.E.; Dettre, R.H. Surface Coll. Sci. 1 969, 2, 85.
(18) Funnidge, C. G. L. J. Colloid Sci. 1 962, 7 7, 309.
20 Walanabc, T.; (19) Yamaguchi, \. Jounuil of Pesticide Science 1991, /6,(,51.
(20) Barlhlolt, W.; Nciiihuis, C. PUmta 1997, 202, 1.
(21) Wagner, T.; Ncinhuis, C; BarlhloU, W. Acta Zool. 1996, 77, 213.
(22) Cassie, A.B.D.; Baxter, S. Trans. Faraday Soc. 1944, 40, 546.
(23) Barlell, V. C; Shepard, .1. W. J. P/iys. Clwm. 1953, 57, 211.
(24) .lohnson, R. .Ir.; Dcltrc, R. II. Adv. Clwm. Ser. 1963, 43, 136.
(25) Dettrc, R.ll.; Johnson, R.E. in Wetting; S.C.I. Monograph No. 25, Sociely of Chemical Industry, London, 1967, p 144.
Drelich, .1.; Wilbur, .1. (26) L.; Miller, J. D.; Whitcsides, G. M Lanf^nuiir 1996 12 ' 1913. '
(27) Drelich, .1. Ph.D. Dissertation, University of Utah, 1993.
(28) Drelich, J.; Miller, J. D. Langmuir 1993, 9, 619.
(29) Kumar, A.; Whitesides, G. M. Appl. Phys. Lett. 1993, 63, 2002.
(30) Kumar, A.; Bicbuyck, II. A.; Whitesides, G. M. Langmuir 1994, 10, 1498.
(31) Drelich, J.; Miller, .1. D.; Kumar, A.; Whitesides, G. M. Colloids Surf., A:
Physicochem. Eng. Aspects 1994, 93, 1.
(32) Drelich, .1.; Miller, .1. D. Part. Sci. Tcchnol. 1992, 5, 1 1.
(33) Digital Instruments, Scanning Prohc Microscopy Training Handbook, 1998, Santa Barbara, CA
(34) Binning, G.; Quate, C. F.; Gcrber, C. H. Phys. Rev. Lett. 1986, 56, 930.
(35) Clark, D. T.; Thomas, II. R. ./. ofPolym. Sci. 1911, 15, 2843.
(36) Andradc, J. D. X-ray Photoelectron Spectroscopy; Andrade, .1. D., Ed.; Plenum
Press: New York, 1985, Vol. I, p 105.
(37) Clark, D. T.; Thomas, 1 1. R. J. ofPolym. Sci. 1977, 15, 2843.
21 CHAPTER 2
MODEL LITHOGRAPHIC SURFACES TO EXPLORE ULTRAHYDROPHOBICF
Introduction
In this phase of the dissertation work, model silica surfaces were designed and prepared in order to examine the critical parameters, such as the structure and discontinuity of the three-phase contact line, that control ultrahydrophobicity. The surfaces containing hydrophobic posts should form discontinuous and unstable 3-phase contact lines with water droplets. The posts must be close enough together and hydrophobic enough that water does not intrude between them. Surfliccs such as these are referred to as composite surfaces as the intersection of the water droplet with the surface consists of a composite mixture of water-solid interface area and water-air interface area. The wettability of such surfaces was first addressed by Cassie and
Baxter,' as introduced in the first chapter of the thesis. We also stressed that this equation and many others were derived for a drop at equilibrium on a surface, thus the contact angles predicted were "equilibrium contact angles". We argue that the structure
of the 3-phase contact line is important to dynamic wettability (Oa, Or and hysteresis). A nearly continuous contact line causes the drop to pin on the surface, whereas a discontinuous contact line leads to an unstable droplet that rolls off the surface with a
small till. By varying the size, shape and the separation between the posts we were able to make surfaces where the water droplet pinned or rolled off easily.
22 It is well known that photolithography and dry etchnig are the most common
techniques used in the fabrication of integrated circuits and solid-state devices where the
substrate is usually silicon wafer, hi the McCarthy group, the chemistry of reactive
organosilanc reagents on silica surfaces has been investigated for several years now.
Hence, the techniques developed in microelectronics manufacturing and the reactive
silane chemistry were combined in this work to study the effect of surface microstructurc on wettability.
Photolithography
Photolithography is the most widely used technique in the printing industry as
well as in semiconductor processing.' The ultimate goal of this process is to transfer three dimensional patterns onto the semiconductor substrate (most commonly silicon wafer) accurately and precisely. The process first involves designing the targeted features using computer software (CAD) and preparing the photomask consisting of these patterns. This is followed by transfering the pattern on the photomask into the photoresist. This pattern can then be replicated on the underlying semiconductor
substrate by a subsequent etching step. Photolithography uses UV radiation of A, = 360-
410 nm when transfering the patterns from the photomask to the photoresist.
In this work, two different kinds of photolithography technique - contact lithography and projection lithography - were used. Contact lithography is the oldest technique developed in this field.
23 Figure 2.1 shows the Hthographic steps involved in the fabrication of a patterned
silicon wafer. The process starts with cleaning the silicon wafer and spni-coating the primer (if needed) and the photoresist (in our case, a positive photoresist). If contact
lithography^ is used, the mask is then placed close to the wafer and the mask pattern is aligned with respect to the crystallme planes of the wafer. Next, the mask is brought into contact with the wafer and is simultaneously exposed to UV light. After that, the sections of the photoresist that have been exposed to UV light dissolve if the wafer is immersed into an appropriate solvent called the developer. This results in a pattern of the mask on the resist. The time spent in the developer will affect the shape and the size of the resulting patterns on the photoresist. It is very important to make sure that the pattern on the photoresist has the targeted dimensions of the original mask, since this pattern is going to be used as the in 'mask' the further step of reactive ion etching. Likewise, if the photoresist is not developed long enough, then the 'mask' will not be accurate. Thus, it needs to be examined by optical microscopy before any further steps such as reactive ion etching.
The contact lithography has some drawbacks such as loss of resolution and sharpness of the features as the pattern size decreases. Hence, if small and sharp features
(< 8 |j,m) are designed, then one needs to consider a different technique such as projection lithography (a stepper). In this technique, the mask patterns are projected onto the wafer
and hence, the wafer is no longer in direct contact with the mask. This results in more
defect-free and accurate patterns. It also has the advantage that the mask doesn't wear out easily.
24 UV light
Mask Photoresist Silicon wafer
Develop
Silicon wafer
RIE
Plasma I
Remove photoresist
Figure 2. 1 lithographic . Main steps in preparing the patterned silica substrates,
Dry Etching
After the photolithography step, there are two ways of replicating the pattern onto the substrate, dry or wet chemical etching. Especially for the last couple of decades, the dry etching, or plasma-enhanced etching has been replacing the conventional wet
chemical etching due to its unique process.'* It resuhs in anisotropic features at low
25 temperatures and in dry environment. Additionally, the technological need to etch features with < 0.5-1 ^m accurately and precisely requires the use of dry etching.
The conventional wet chemical etching is a well-established technique. However, the resolution of the features is limited by undercutting. As the solvent etches downward, it also etches laterally at the same rate resulting in an isotropic etch, as shown in figure 2.2. This causes changes in pattern size and shape. Also, the use of corrosive acids and toxic solvents causes the wet chemical etching process to be less attractive.
One of the biggest advantages of the plasma etching technique is that it etches
faster much in the vertical direction than in the horizontal direction resulting in an anisotropic profile as shown in figure 2.2. It can be performed in full automation and does not require any harmful chemicals.
Mask > Mask Photoresist to be etched
Substrate
Dry Etch Wet Chemical Etch
> w
Anisotropic Profile Isotropic Profile
Figure 2.2. Cross sections of photoresist etched by plasma or liquid etchants,
26 There are several different dry etching technologies, such as plasma etchuig, reactive ion etching and ion beam milling. Among these, reactive ion etching (RIE) is the best-controlled technique offering high selectivity and anisotropy and it was used m this work.
Usually, radio-frequency (rf) or microwave electric fields are used to excite the low-pressure glow discharges.' The etching gas (or combination of gases) decomposes and ionizes into electrons, free radicals and ions due to this electric field. The reactive species have longer mean free paths due to the lower 10"^ operating pressures of 1 x to 1
X 10"' Torr in RIE. This lets the ions impinge vertically at the substrate's surface leading to an anisotropic etch.^
The cycle of plasma etching starts with the collision of gas molecules and the energetic electrons resulting in highly reactive species that are responsible for the etching
reactions even at low temperatures. These are some of the possible reactions:"*'^
1) excitation (rotational, vibrational, electronic)
e' + A2 ^ A2* + e"
2) dissociative attachment
e" + A2 ^ A" + A"" + e"
e" + A2 A" + A^
3) dissociation
e" + A9 -> 2A + e"
27 4) ionization
e" + A2 -> Ai^ + 2e'
5) dissociative ionization
e" + A2 + A + 2e"
Next, these chemical species (for instance free radicals) diffuse and adsorb onto the surface. It has been stated that the free radicals chemisorb and react with the surface/
Other primary processes include: 1) ion-surface interactions - leading to neutralization and secondary electron emission, sputtering and ion-induced chemistry; 2) electron- surface interactions - leading to secondary electron emission and electron-induced chemistry; and 3) radical-surface interactions - resulting in surface etching as well as film deposition.
The critical step is the desorption step. The product of the reaction between the free radicals and the surface should have an appropriate vapor pressure in order to desorb and diffuse into the gas phase.
Gas Inlet
Wafer Plasma
To Vacuum Electrode Pump
Figure 2.3. The reactive ion etcher.
28 Reactive Silane Chemistry
Reactive organosilicon compounds have been '^ used to modify metal oxidc*^ and
^'^ polymer surfaces.' Due to the covalent attachment of silane monolayers onto silica
surfaces, they are stable and ideal for many applications.'' From enhancmg the biocompatibility of a surface,"'^ modifying the surface for sensor applications"" to
immobilizing chemically active species,^^ the silane reagents have been used in a variety of areas.
Recently, in the McCarthy group, there has been an extensive study done on exploring silica surfaces modified with monofunctional (R.^SiX), difunctional (R2SiX2) and trifunctional (RSiX.^) organosilanes (X = CI, OR, NMe2).^''" Unlike other groups' work, in these reports the reaction conditions such as the kinetics and the amount of water in the system were well optimized in order to have well-characterized surfaces. The dynamic contact angles changed if the reactions were performed in a solvent (usually toluene) or in the vapor phase at -75 "C. The vapor phase reaction resulted in surfaces with higher contact angles than the liquid phase reaction. Si-O-Si covalent bonds are
formed when the monofunctional silanes are reacted with the silica surfaces, as shown in figure 2.4. The n-alkyldimethylsilane groups have cross-sectional area of 32-38 A^, and they form densely packed monolayers.*^"' The water does not penetrate the monolayers, resulting in contact angles of the surfaces that are independent of the chain length of the alkyl group. They stated that the vapor phase reaction was a convenient and an easy way of preparing surfaces with high yields.
29 Figure 2.4. The reaction of octyldimethylchlorosilane (and other monofunctional organosilanes) with siHcon dioxide surfaces.
In contrast to the monofunctional silanes, the difunctional silanes can react in two different ways with hydrated sihca. As shown in figure 2.5, they can either covalently attach to the surface or they can polymerize to form grafted polysiloxanes depending on the reaction conditions. It was shown that the vapor phase reaction of short chain length alkylsilanes results in oligomeric structure and in surface-grafted alkylsiloxanes in the presence of water. The wettability of these surfaces was very impressive. The water
(as well as hexadecane and methylene iodide) droplets slide off the surfaces easily due to the negligible hysteresis behavior they exhibited when prepared in the vapor phase. For
instance, for dimethyldichlorosilane the dynamic water contact angles were 9a / 6r = 105°
I 104"*. We call such surfaces without any hysteresis 'ultralyophobic'.
30 8
H3C H,C^ /CH ' \ .CH °» o o ^ OH OH OH CH H3C CI Si Covalent
CI Attachment
CH HO^ 3 / CH Vertical si-
Polymerization H,C I CH, \ Six. Si \ / ^OH / CH O OH O
Figure 2.5. The possible reactions of dimethylchlorosilane (and other difunctional organosilanes) with silicon dioxide surfaces.
Experimental
Materials
Ethanol, toluene, sulfuric acid and hydrogen peroxide (30%) were used as received from Fisher. Organosilanes were obtained from Gelest and used as received.
House purified water (reverse osmosis) was further purified using a Millipore Milli-Q®
1 system that involves reverse osmosis, ion exchange and filtration steps (10 ohm/cm).
Contact angle measurements were made using a Rame-Hart telescopic goniometer with a
24-gauge flat-tipped needle; dynamic advancing and receding contact angles were
31 recorded as the probe fluid, water, purified as described above, was added to and withdrawn from the drop, respectively. Scanning electron micrographs (SEM) were obtained using an Amray 1 803TC instrument with an accelerating voltage of 25 kV. X- ray photoelectron spectra (XPS) were obtained using Perkin-Elmer-Physical Electronics
5 1 00 spectrometer with Mg excitation at takeoff 5" angles of 1 and 75". Atomic concentrations of the substrates were calculated using the sensitivity factors, Si is, 0.270;
0.250; Cis, Ois, 0.660; Fis, 1.00, obtained from the samples of known composition.
Preparation of Silicon Substrates
4-in. and 3-in. silicon wafers were obtained from International Wafer Service
(<100> orientation, P/B-doped, resistivity from 20 - 40 Q-cm, thickness from 450 - 575
|Lim). A contact lithographic mask (with hexagonally arrayed square posts of 16, 32, 64 and 128 length and width) was constructed by Photronics Inc. The preparation of the
silicon substrates, including the mask preparation and patterning of the wafers, were all performed at the Cornell Nano fabrication Facility. The other masks were designed using a CAD program and prepared with a GCA PG3600F optical pattern generator.
Photolithography was used to transfer the patterns of the masks onto the silicon wafers.
Subsequent to irradiation, the wafers were etched with a proprietary mixture of gasses using a Plasma Therm SLR-770 reactive ion etcher for different durations. When the etching process was complete, the wafers were cleaned oxidatively using a Branson IPC
P2000 Barrel Etcher. Wafers were then placed in a solution of ammonium hydroxide, hydrogen peroxide and water (4:1:1) for 15 minutes, rinsed with copious amounts of
32 water and spin dned. The wafers were then cut into 1.5 cm x 1.5 cm pieces, placed in a custom designed (slotted hollow glass cylinder) sample holder and were cleaned by submersion into a mixture of concentrated sulfuric acid and hydrogen peroxide (30%)
(7:3) overnight. The wafers were then rinsed with copious amounts of purified water and dried in a clean oven at 120 °C for 5 hours immediately prior to silanization reactions.
Reaction of Silicon Substrates with Organosilanes
The silicon substrates were placed in a custom-designed (slotted hollow glass cylinder) sample holder and placed in a flask containing 0.5 mL of organosilane reagent: dimethyldichlorosilane (DMDCS), n-octyldimethylchlorosilane (ODMCS) or
heptadecafluoro- , 1 1 ,2,2-tetrahydrodecyldimethylchlorosilane (FDDCS). The wafers were not in contact with the liquid silanes. The vapor phase reactions were carried out for days at 65-70 3 °C. The hydrophobized wafers were rinsed with toluene (2 aliquots),
ethanol aliquots), 1 : (3 1 ethanol/water (2 aliquots), water (two aliquots), ethanol (2 aliquots) and then water (3 aliquots) and were then dried in a clean oven at 120 °C for 30 minutes.
33 Results and Discussions
Reaction of Silicon Substrates with Qrganosi lanes;
The silicon substrates were hydrophobized with dimethyldichlorosilane
(DMDCS), n-octyldimethylchlorosilane (ODMCS) or heptadecafluoro-1,1,2,2-
tetrahydrodecyldimethylchlorosilane (FDDCS). The reactions were carried out in vapoi
phase for days to give 3 dense monolayers. Figure 2.6 shows a representative XPS
survey spectrum for a smooth silicon/silicon oxide surface reacted with FDDCS. Table
gives 2.1 the atomic concentration of the surfaces reacted with DMDCS, ODMCS and
FDDCS at 15° and 75° takeoff angles.
1 1 ! 1 1 1 I 1 1 1 1 ] i H- h"! i 1 1 10 — — f-
1000.0 900.0 800.0 700.0 600.0 500.0 -tOO.G 300. 0 200.0 lOO.O 0.8 ABIDING ElOGY. eV
Figure 2.6. XPS survey spectrum of silicon/silicon oxide surfaces reacted with FDDCS at 75° takeoff angle.
34 Table 2.1 Atomic concentration (XPS) . ofsilicon/silicon oxide surlaces leictcd will. DMDCS, ODMCS and I'DDCS at 15" and 75" takeolT angles.
DMDCS ODMCS 15° 75° 15° 75° 15° 75° %Si 14.44 39.48 26.82 48.69 14.22 41.92 %C 55.05 29.05 48. 11 14.32 26.91 11.16 VoO 30.51 31.47 25.07 37.00 20.73 32.60 %F 38.14 14.32
Critical Size Scale ol'Rouuhness for llvdiophobicitv
Figure is 2.7 a 2-dimensional schematic representation of one series of silicon oxide surfaces that was prepared using photolithographic techniques. These surfaces are hcxagonally arrayed square posts varying in size from 2 |im x 2 |am to 128 |im x 128 |luii
X X D 2x
2x
Figure 2.7. 2-l)imensional (X-Y) representation of a series of silicon surlaces containing hcxagonally arrayed square posts, X - 2, 8, 16, 32, 64 and 128 (.uii. Surfaces with post heights (Z) of 20, 40, 60, 80, 100 and 140 ^m were prepared.
35 2 |im Ba/Gr^ 176V 14P 8 |.im Oa / Ok - 173V 134"
16 |im eA/eR= IVr/ 144° 32}im eA/eR= 168°/ 142°
64 |am eA/eR= 139°/81° 128 urn Ga/Or^ 116°/80°
Figure 2.8. SEM images of the surfaces described in figure 2.7. The water contact angles shown are for DMDCS-modified surfaces.
36 Different post heights (20, 40, 60, 80, 100 and 140 ,m) were prepared by vaiying
the etching time for 1 6, 32, 64 and 1 28 size posts. The posts are spaced as indicated
in the figure. Figure 2.8 contains SEM images of the surfaces explained in figure 2.7. These surfaces were oxidatively cleaned (to remove any of the lithography mask and/or
etching chemicals) and chemically modified by reaction with silanizafion reagents in the vapor phase.
Table 2.2 shows water contact angle data for surfaces that were etched to give 40
^m high posts and reacted with DMDCS, ODMCS or FDDCS. Data for smooth silicon/silicon oxide surfaces are shown for comparison. The surfaces are named, ^SP^', with the superscript X indicating the post length and width and the superscript Y indicating the post height. The surfaces with post sizes of 32 and less exhibited both high advancing and receding contact angles. We ascribe the slight differences in contact angles between the differently modified surfaces to experimental error in the modification reactions or in the measurement of contact angles and not to the size of the posts or the identity of the silanization reagents. All of these surfaces are extremely hydrophobic. Water droplets were unstable on these surfaces. Droplets moved spontaneously on slightly tilted surfaces. For the surfaces with post sizes 64 |im and greater, low receding contact angles (lower than those on the smooth surfaces) were observed. Advancing angles were higher than those on the smooth surfaces, but lower than those on the surfaces with smaller features. The water drops pinned on these surfaces and did not rotate nor move spontaneously when the surfaces were slightly tilted.
37 It is apparent that water intruded between the greater spaced posts and that the contact line was pinned by the greater solid-liquid contact. We note that all of these surfaces contained 25% solid-liquid interfacial area (f, in Cassie's equation in chapter 1 ); this equation predicted equilibrium contact angles (using the measured advancing contact angles on smooth surfaces) of 145°, 143° and 150° for the DMDCS, ODMCS and
FDDCS surfaces, respectively. These were in line (between the measured advancing and receding contact angles) with the surfaces containing features of 32 |im and less.
Table 2.2. Water contact angle data for silane-modified hexagonally arrayed 40 [im - high square post surfaces described in figures 2.7 and 2.8.
X Y = SP ; X post width and length Y = post height
DMDCS-modified ODMCS-modified FDDCS-modified Silicon Surface Oa Or Oa Or Oa Br
smooth 107° 102° 102° 94° 119° 110°
2 nmgp40 |im 176° 141° 174° 141° 170° 146°
8 |imgp40 |jin 173° 134° 173° 139° 170° 140°
16 |imgp40 [im 171° 144° 174° 134° 168° 145°
32 ^|ngp40 \im 168° 142° 170° 132° 170° 146°
64 ).imgp40 |im 139° 81° 114° 65° 149° 100°
128 |iingp40 \im 116° 80° 95° 58° 131° 93°
38 Surfaces containing 16, 32, 64 and 128 posts with post heights varynig from
to 20 Mm 140 were prepared and derivatized using ODMCS. The dynamic contact
angles were measured and are reported in Table 2.3. The contact angles were
independent ofthe height of the posts for 16 and 32 m nm posts, ind.catmg that water
did not intrude between them (the probe fluid could not detect the depth ofthe region that
it did not penetrate).
The surfaces described thusfar (square posts) with features 32 ^m and smaller exhibited very high advancing and receding contact angles, but significant (30-40°) hysteresis. Water droplets came to rest on these surfaces when the surfaces were horizontal (but rotated continuously). This behavior differs from the randomly rough surfaces that have we prepared^^'^^' on which droplets moved spontaneously on horizontal surfaces. suspect We that the regular hexagonal array of features with regions that can support a straight contact hne over significant lengths is the source of this hysteresis.
39 m1
Tabic 2.x Walcrconlacl angle data for ODMCS-nrndillal surfaces conlaining K, ,„„ x 16 ^m, ,m x ,n,, 32 32 M x (,4 12« x 128 sc,ua,c posts ol d.ricra'r
^SP^ ; X post width and length Y = post height
Silicon Post
Surface Height (|iim)
1 6 |ini^p2() ^im 20 173" 138°
1 ()\ / 111 Mic^1 1 1 r )4() II tii ^ IV'' Mill 40 174° 34" 1
I 6 linic^ iim 60 169° 138°
1 () finiigpXO fini
80 1 69° 36° 1
1 6 |nnc n 1 00 iim 100 173° 138° 16 |jmgp!4() fim 140 168° 136°
20 170° 137° 32 uiiic n4() mil 40 170° 132° 32 ^iniQ T)6f) urn 60 168° 39° 1 32 umc inn 80 167° 134°
32 (iiiicj I) 1 00 )i 100 173° 134°
32 fiiiigpl4() fiMi 140 166° 131°
lit ftiii^iji,\/ |nii 09° 20 1 80°
64 ti mo r^4() mid
40 1 14° 65°
(\A IMlinr\^k(l iiiti 1 1 1 1' " V 1 ^ 1 1 1 ^ ' ' 1 1 1 1 i 60 128° 64°
1 1 (^A iiir~i " III!) i/*T LM 1 1 ' n 1 1 ^ I J'' 80 149° 113°
iiiiifiT~\l()(t iirn ' ' Mill ' IJ'^'*' f 100 145° 57"
64 pmepl4() |im 140 1 A°
1 2X nnn^p2() |mi 20 93° 74° I2H |iiin^p40 fuii 40 95° 58°
1 28 ^tnigp()0 |im 60 120° 69"
1 28 [ini^ pKO fit}} 80 116° 73°
128 ftnij^jjIOO ftiii 100 126° 65"
128 |iiii(^pl40 (ini 140 118° 15"
40 >
Structure of 3 -Phase Contact Line
We designed and prepared three different surfaces with the objective of contorting the 3-phase contact line. These are described in Figure 2.9 and SEM images are shown in
Figure 1 2. 0. These surfaces are named 'RP' for the staggered rhombus-sliaped posts, TP' for the indented square-shaped posts and 'StP' for the 4-point star-shaped posts. Tlie post height of these surfaces was maintained at 40 fim.
8 |Lim 8 |im
32 |im X 32 (.im
< X X > 16 |im 16 |im IP StP
8 |im <— 4 (^O ^ O-
32 |im
<> 0 ^-
32 |im RP
Figure 2.9. 2-dimensional (X-Y) representations of surfaces containing different geometry posts.
41 Figure 2.10. SEM images ofthc surfaces described in figure 2.9. The water contact angles shown arc for DMDCS-modillcd surlaces.
Wc also prepared a series of surfaces containing 8 |.mi x 8 |.uii liexagonaily
arrayed square posts (40 \xm high) that were spaced at greater distances than the surfaces
described in figures 2.7 and 2.8. fhese were prepared with the objective of making tlic }-
phase contact line less continuous.
42 X
4x 4x
4x lOx
40 |.im 23 (iinSP
X 6
lOx
Ox 40 |im 56 ^mSP
iMgurc 2. 1 1 . 2-Dimcnsional (X-Y) rcprcscntalion of a series of silicon surfaces containing hexagonally arrayed square posts, X = 8 ^im and heights (Z) = 40 \im. Post separations were varied as 32 fim x 32 |j.m, 32 |im x 80 |im, 80 |im x 80 |im.
43 Figure 2.12. SEM images of surfaces containing 8 |im x 8 |,im square posts with dilTcrcnt spacings as described in figure 2.11. The water contact angles shown are for DMDCS- modified surfaces.
Water contact angle data for the silane-modificd surfaces shown in figures 2.10
and 2. 1 2 are reported in Table 2.4. The subscripts in the structure names indicate the
minimum distance between square post centers. The entry labeled i6fimSP'*"^"" is the
^ same surface as that labeled ^'"\sp'*** (these squares are spaced by 1 6 |.im) in Table 2.3
and is displayed there for comparison. Changing the shape of the posts from square to
44 indented square, star or staggered rhombus did not affect the advancing contact angles, but caused significant increases (up to 22°) m receding contact angles and con-esponding decreases in hysteresis. The receding contact angles were higher for the staggered rhombus posts than those for the star-shaped posts which were higher than those for the indented square posts. We interpret these changes as due to more contorted and longer 3- phase contact lines. Increasmg the spacing between the posts also resulted in no changes in the advancing contact angles and increases in the receding contact angles (up to 2(r) for square posts spaced up to 32 ^im. We interpret these increases as due to decreases in the contact length of 3-phase contact lines. Contact angles decreased significantly for the substrate with posts spaced by 56 )nm. Water intruded between these posts, increasing the contact length of the 3-phase contact line.
Table 2.4. Water contact angle data for silane-modified surfaces containing different shaped posts and 8 |im x 8 |im square posts spaced at various distances (described in figures 2.10 and 2.12).
DMDCS ODMCS FDDCS Silicon Surfacee r OA Ba Gk Oa 176° 156" 174° 155° 168° 153°
^im jp40 175° 143° 173° 140° 169° 146°
^im |g|-p4() 175° 149° 174° 147° 170° 148°
Qp4() 173° 134° 173° 139° 16 ^m*^^ 170° 140°
Qn4() \im 175° 146° 172° 147° 167° 147° 23 \iuP^ 173° 154° 172° 154° 169° 155° 32 ixmP^ c p4() 121° 67° 112° 68° 134° 92° 56 ^ini'^*^
45 In addition to the surfaces discussed above, in order to observe the pinning effect
of the water droplet upon formation of a continuous contact line of water droplet on the
surface, we prepared two surfaces described in figure 2.13. Both of these surfaces were
treated with DMDCS and resulted in high advancing but low receding water contact
angles. Hence, the water droplets were pinning on these surfaces that exhibited high
hysteresis.
9A/eR= 161°/ 110° eA/eR= 161° / 137°
^" ^1 llMi i^ffi ^IttHI llMi
(a) (b)
Figure 2.13. SEM images of surfaces containing (a) 32 x 32 |im square posts; (b) 32 |im X 32 |im hollow squares. The water contact angles shown are for DMDCS-modified surfaces.
We attempted to view the 3-phase contact line between water and the surfaces
reported here using optical microscopy - with no success, shown in figure 2. 14. The
droplets acted as lenses and we could not focus on the contact lines. To overcome this problem, we put drops of molten Woods metal on the surfaces and viewed the contact
lines after the metal solidified and was removed from the surfaces. Figure 2.15 shows an
optical micrograph of the Woods metal - silicon - air 3-phase contact line on the silicon
46 substrate and also on the surfaces ^"^ ^'"^ named 23 ..SP^"^ and 32 ,n,SP^" in Table 2.4. The contact line was clearly contorted by the square posts on the surface.
Figure 2.14. Snapshot of a water droplet on the substrate with star-shaped posts, hydrophobized with DMDCS.
25 |im 25 iim
(a) (b) (c)
Figure 2.15. Optical micrographs of the Woods metal - silicon - air 3-phase contact line ^"^ on the (a) silicon substrate; surfaces named (b) 23 nmSP'*^ and (c) 32 pmSP''° in Table
2.4. The contact line is clearly contorted by the square posts on the surface.
47 .
Laplace Pressure
Walcr starts penetrating between the hydrophobic posts iflhe hydrostatic
pressure, AP, is exceeded. The general equation lor this phenomenon is as follows:"
= y'" AP -(p cos ()a)/A ^4^^
where is the perimeter p ofthe unit cell, A is the area of air y'^' space in the unit cell, is the
surface tension of liquid (in this case, water), Oa is the advancing water contact angle
(greater than 90°). AP was calculated for the hydrophobic surfaces prepared and it is
listed in Table 2.5. 2 square posts had the highest pressure, followed by the 8 |iim square posts and staggered rhombus-shaped ones, fhc pressure trend was consistent with the hydrophobicity ofthe surfaces and their dynamic contact angles.
Table 1 2.5. .aplace pressure calculated using equation 4. 1
Silicon Surface AP (dync/^im^) X 10^
2 |ini^p40 jini 72.4
X )itn<^|j'1() |iin 18.0
1 () |iin^p4() fiiii 9.0
32 l^^^gp'^*^ 4.4
6-1 |iiii(^p'U) )ini 1.7
j)'1f) 1 2H |im^^ )iin 0.5
7.2 23 |mi^<
6.6 32 ^m^^'
1.9 56 (im'-'*
j^p40 fiin 9.6
48 Critical Tilt An^ ki
As the hysteresis decreases, the eritieal tilt angle at which the water droplet starts rolling oir the surlace should also decrease, as expressed in Furmidge's equation introduced in the lirst chapter. To observe this, we built an apparatus shown in figure
2. 1 6. A water droplet was put onto the DMnc^S-niodilied substrates and the whee was turned slowly until the droplet started rolling off the surface. The average of 5 tilt angles for each surface was calculated and is given in Table 2.6. When cos (hysteresis) was plotted against the critical tilt angle as shown in figure 2.17, an inverse relationship was observed as predicted in Furniidgc's equation.
Figure 2.16. The apparatus used to measure the critical tilt angle at which the water droplet starts to roll off the surface.
49 Tilt Angle (degrees)
Figure 2.17. Critical tilt angle as a function of hysteresis exhibited by DMDCS-treated silicon surfaces. As hysteresis decreased, the critical tilt angle decreased as expressed i Furmidge's equation.
Table 2.6. Tilt angle and hysteresis exhibited by the DMDCS-modified silicon surfaces.
Silicon Surface Tilt Angle Hysteresis
smooth 29° 5° 2 |imgp40 jim 20° 35°
8 |.imgp40 jim 17° 40°
16 fimgp40 19° 27°
32 ^mgp40 22° 26°
64 |.imgp40 jim 35° 58°
128 ^mgp40 |im 30° 36°
6° 21°
jp40 19° 32°
g^p40 11° 27°
cp40 m 14° 29° 23 nmor 9° 7° 1 32 |imor ^p40 urn 38° 53° 56 iimoi
50 Conclusions
The wettabihty of a number of patterned silicon surfaces that were prepared by
photoHthography and hydrophobized using silanization chemistry was determined.
Surfaces contaimng square posts with X-Y dimensions - of 2 ^m 32 nm and similar
distances between the posts exhibited ultrahydrophobic behavior with high advancing and
receding water contact angles. Water droplets moved very easily on these surfaces and
rolled off of slightly tilted surfaces. Contact angles were independent of the post height
from 20- 140 ^m and independent of surface chemistry (siloxane-, hydrocarbon- and
fluorocarbon-modified surfaces were prepared). Surfaces containing square posts with
X-Y dimensions of 64 and 128 ^m ^m with similar distances between them were not
ultrahydrophobic- water droplets pinned on these surfaces and water apparently intmded between the posts. Increasing the distance between posts caused increases in receding contact angles up to the point that water intruded between the posts. This is due to decreases in the contact length of the 3-phase contact line. Changing the shape of the posts from square to staggered rhombus, star or indented square also increased the receding angles due to the more contorted contact lines that form on these surfaces. The
maximum length scale of roughness that imparts ultrahydrophobicity is -32 |im.
Although the surfaces explored here are extremely hydrophobic, they are not as hydrophobic as many of the surfaces that we have prepared that contain random
roughness (perhaps at multiple length scales). ' We believe that this is due to two effects: First, on these surfaces that were prepared by photolithography and etching, the
51 3-phase contact line is tortuous only in the X-Y plane of the surface (all of the posts are the same height). Contact lines on randomly rough surfaces are tortuous m 3 dimensions, thus the contact lines can be longer and less stable. Second, these lithographed surfaces are very regular and the contact line can register with the posts in metastable geometries that will differ around the perimeter of the drop as a function of the angle between the contact line and the hexagonal spacing direction. We predict that surfaces with randomly arrayed posts of different heights will be even more hydrophobic than those reported here.
52 Notes and References
1) Cassic, A. B. D. Discuss. Faraday Soc. 1948, 3, 11.
2'"' 2) ' Microlilho^raphv, cel., !• k Thompson 1 , ^' "''^ ^'''^ Washmglon" DC (^aplcM
Bowdcn, M. .1. in 2"'' 3) Introduction to Microlithoi^rapliv, cd., • Thompson 1 F Willson, C. G.; Bowdcn, M. .!., luls; American dicmical Society Washinutcin^ DC, 1994, Chapter 2. ^
4) Mncha, .1. A.; Hess, D. W.; Aydil, E. S. in Introduction to Microlitlio^niphy 2"' ed., Thompson, L. F.; Willson, C. G.; Bowdcn, M. .1., Hds; American' Chemical Society, Washington DC, 1994, Chapter 5.
Powell, 5) R.A. in Materials Processing Theory and Practices, Vol. 4: Dry etching ' for Microelectronics, North-Holland Physics Publishing, Netherlands 1984 'In 119. '
6) Bell, A. T. Solid State Tcchnol. 1978, 21, 89.
7) Winters, H. F. in Plasma Chemistry III; Topics in Current Chemistry Series 94, Springer-Vcrlag, Heidelberg, Germany, 1980, p 69.
8) Leyden, D. E., Ed., Silancs. Surfaces, and Interfaces, Gordon and Breach: New York, 1986.
9) Lisichkin, CJ.V., Ed., Chemically Modiped Silicas in Adsorption, Chromatography and Catalysis, IChimiya, Moscow, 1986.
10) Plueddemann, E. P. in Silane Coupling Agents; 2"'' cd.; Plenum, New York, 1991
1 1 ) Mittai, K. L., Ed., Silane ami Other Coupling Agents, VSP, Zeist, 1 992.
12) Pesek, .1. J., Leigh, 1. E., Eds., in Chemically Modified Sinfaces; Royal Society of Chemistry, Cambridge, 1994.
13) Chaudhnry, M. K.; Whitesides, G. M. Langmuir 1991, 7, 1013.
14) Fadeev, A. Y.; McCarthy, T. .1. Langmuir 1998, 14, 5586.
15) IJlman, A. MRS Bull, .lune 1995, 46.
Biomedical Polymers, 1 6) Andrade, .I.D. in Surface and Interfacial aspects of
Andrade, J.D., Ed., Plenum Press,'New York, 1985; Vols. 1 and 2.
./. Soc 17) 1 hoden van Vclzen, E. U.; Engberscn, .1. F. J.; Reinhoudt, D.N. Am. Chem. 1994, 116, 3597.
53 (18) Weiss, T., Scheirbaum, K.D.; Thoden • van Velzen, • U Reinhoudti^einnouat, U.N.,D N Gopel,Tnn.i W. Sens. Actuators 1995, B26-27, 203.
Yang, (19) X; Shi, J.; Johnson, S.; Swanson, B. Langmuir 1998, 14, 1505. (20) Ingall M.D.K.; Honeyman, C.H.; Mercure, J. V.; • Bianconi P A Kunz / ' R R Am. Chem. Soc. 1999, 121, 3607. '
(21) Fadeev, A. Y.; McCarthy, T. J. Langmuir 2000, 16, 7268.
(22) Fadeev, A. Y.; McCarthy, T. J. Langmuir 1999, 1 5, 7238.
(23) Fadeev, A. Y.; McCarthy, T. J. Langmuir 1999, 15, 3759.
(24) Claudy, P.; Letoffe J. M , Gaget, C; Morell, D.; Serpinet, J. J. Journal of Chromatography 1987, 395, 73.
Chen, (25) W.; Fadeev, A. Y.; Hsieh, M.; Oner, D.; Youngblood, J.; McCarthy^^ciuiiy, Ti Jj. Langmuir 1999, 15,3395. .
(26) Youngblood, J. P.; McCarthy, T. J. Macromolecules 1999, 32, 6800.
(27) Dettre, R H.; Johnson, R. E. in Wetting; S. C. I. Monograph No. 25, Society of Chemical Industry, London, 1967, p 144.
54 CHAPTER 3
SURFACE MODIFICATION OF LOW DENSITY POLYETHYLENE
Motivation
The purpose of this phase of the thesis work was to extend the achievements of
observing ultrahydrophobic behavior discussed in chapter 2 to practical apphcations.
Conventional techniques such as lithography are capable of producing well-defined
structures with desired dimensions and shapes. However, these operations are expensive
and time-consuming and can be used to prepare only small samples. The motivation was
to explore a new practical way to make ultrahydrophobic surfaces. Low density
polyethylene (LDPE) was chosen since it is an inexpensive semicrystalline polymer and
there have been some studies in making its surface rough by various chemical reactions.
In order to impart roughness to LDPE, the following approach was taken.
Chromic acid etching of LDPE, known to form a patchy surface via etching the amorphous regions prefentially faster than the crystalline regions, was perfomied. Then, the oxidized polymer was deformed by uniaxial or biaxial tension at a constant rate and the surface morphology of the necking region was examined by scanning electron microscopy. The wettability of these rough surfaces was explored after gold coating and
hydrophobization using self-assembled monolayers of 1 -dodecanethiol.
55 Surface Treatments of Low Denc^i'ty Polyethvlene
Surface modification of polyolefms has been studied over tlie last 40 years in
order to improve adhesion, printabihty and wettabihty properties. The most common
techniques used are plasma,'"' ozone and corona '" '^ discharge treatments, chemical
etching,'^"2° photooxidation^'-^^and flame treatment.^^
Our interest was to utilize a surface modification technique to achieve rough
surfaces of LDPE. There are couple of different techniques in the literature where
researchers have observed rough surfaces upon such treatments. For instance, the
commercial pretreatment of corona discharge technique has been shown to cause some
surface roughening as a function of humidity and corona energy. Very recently, Feijen et
al. investigated the effects of CF4 gas plasma treatment on surface properties of polyethylene (both high density and low density polyethylene).^*^ The LDPE's surface
microstructure changed from a lamellar one to a nanoporous-like one upon plasma treatment (> 1 5 minutes) whereas this was not observed for HDPE. The etching rate of
CF4 gas was inversely proportional to the crystallinity of the polyethylene substrate used, indicating that the amorphous phase was etched faster than the crystalline phase. Surface fluorination occured with a XPS -determined atomic ratio of F:C = 1.4. The interesting point to make here is that these researchers called the plasma-treated LDPE surfaces
'superhydrophobic' because they exhibited an advancing water contact angle of -140°.
However, they did not comment on the low receding contact angle of -95°.
56 The other effective technique of oxidizing polyolefins is chemical etching using very strong oxidizing agents. Oxidants such as cliromic acid have been well documented in the literature.^«-^°-3'^ Whitesides and coworkers have extensively characterized these
'polyethylene carboxylic acid' surfaces by contact angle measurement, XPS and ATR-IR experiments.'' This technique yields species with hydroxyl groups and then with ketone, aldehyde and carboxylic acid groups via chain scission.
CrOj/HjSO^/Hp
75 "C
Figure 3.1. Chromic acid oxidation of LDPE.
Whitesides and coworkers were the first ones to publish the surface morphology of chromic acid-treated LDPE, as shown in figure 3.2." Recently, the surface properties
of the etched LDPE (up to 15 minutes reaction time) was also investigated with an in situ contact mode scanning force microscopy (SFM) using gold coated and thiol modified
probe tips and also with TappingMode™ SFM. LDPE had an initial morphology consisting of stacked lamellae/^^"'*^ These workers concluded that the amorphous phase was etched prefentially by showing that the stacks of lamellae were more exposed to the surface upon etching and that they started to corrode with etching time. The roughness value increased from an initial value of -20 nm to --50 nm during 10 minutes of oxidation.
57 (a) (b) (c)
Figure 3.2. SEM micrographs of (a) LDPE; and LDPE oxidized with chromic acid solution for (b) 60 seconds; (c) 6 minutes."
Due to environmental considerations, large scale use of chromic acid is
undesirable. Hence, Price and coworkers studied milder oxidizing agents such as
K2Cr207 / H2SO4 , H2O2 and K2S2O8. By using a high-intensity ultrasonicator, they attempted to promote the oxidation of the polyethylene surface via enhanced decomposition of the oxidizing agents and production of excited species such as radicals.
The first agent did not give any indication of an oxidized surface, however the other two resulted in oxidized LDPE surfaces. The paper did not report any surface roughness analysis using these reagents but only stated that there was no visible change in LDPE films.
58 Deformation of Thin Film Coatinf^s IJgcmUnuu^^
Deformation of a tiiin Him bonded to a thicker substrate has been the subject of
various pubhcations for the last two decades, mostly due to the concerns of the
electronics industry.^^ When an external force is applied to a substrate with a coating, a
stress is induced within the surface layer. The relaxation of the stresses developed can
lead to cracking^'*-''\ buckling^^'-^^ or delamination.^^'^*^ For instance, if the film is brittle
and the substrate it is bonded to is ductile, upon a tensile stress greater than the fracture
strength of the coating, cracks form.''^ These cracks then tunnel along the film.
'fhcrc are different crack patterns observed for biaxial and uniaxial stresses.
Biaxial stress results in cracks called 'mud cracks' because they are very similar to the cracks observed upon drying mud."^" On the other hand, upon uniaxial stress long and parallel cracks with approximately uniform spacings are observed.'''
Strawbridge et al. schematically showed the most common modes of coating failure upon tensile stress, as shown in figure 3.3."''^ The relative ductility (or britlleness)
of the substrate and the coating, as well as the interfacial bonding are important in identifying failure which of these tensile modes occur, as compared in Table 3. 1 . One of
the points they made was that the failure occurs in the favor of least resistance or the
greatest motivation. Hence, more irregular cracks would be observed if the stresses are not distributed evenly in the system.
59 (a)
(b)
(c)
Figure 3.3. Possible tensile failure mechanisms for thin coatings bonded to a substrate (a) through-thickness cracking; (b) through-thickness cracking followed by substrate failure; (c) through-thickness cracking followed by interfacial delamination.'^^
52 Table 3.1. Tensile failure modes for thin brittle films
Film Substrate Interface Bonding Decohesion Mechanism
Good Film cracking-no decohesion Brittle Ductile Poor Film cracking-interface decohesion
Good Film cracking-interface decohesion Ductile Brittle Poor Edge decohesion at interface
60 One of the pioneer studies in this field is by Volynskii and coworkers."-^^ They
work on the deformation of a polymer such as PET with thin metallic coatings. In one of
their recent papers, they studied the mechanism of formation of cracks upon tensile
drawing. They coated unoriented PET with a 4-20 mn thick platinum." The coating was
bonded to the sample well so that there was no slip upon stretching. When the polymer was drawn uniaxially, they believed that the stress was not uniform throughout the sample, as shown in figure 3.4. The maximum stress was at the center of the coating that initiated the nucleation of a crack. As the sample was stretched more, the stress grew bigger and ultimately, a crack formed. In another publication, they stressed on the fact that the interfacial adhesion was important in forming cracks.^^'* When the energy of deformation of the coating was greater than the energy of adhesion, the coating started to peel off.
Figure 3.4. Schematic representation of distribution of stress throughout the metal coating bonded to a polymer support. Stress is maximum at the center of the coating.
This grows with stretching and ultimately there is a crack forming at the center of the coating.""
61 They studied the fracture of gold, platinum and carbon coatings (1-15 nm) on a
PET support upon tensile deformation/^ They observed that the coating failed in the necking region resulting m ribbon-like stripes of platinum coating throughout the sample, as shown in figure 3.5. The unnecked region remained smooth without any feilure occurring.
Figure 3.5. SEM micrograph of a deformed 3.8 nm platinum coating on PET support. strain The was not reported. The strain rate was 1 mm/min. The bar is 5 ^im.^'
On platinum coated PET, they found out that at low strains (5-15%), cracks were
""^^ not distributed uniformly across the sample. At higher strains (-100%), the crack distribution was regular in the form of stripes aligned with each other. At further elongation, the distance between the cracks grew larger. Also, the faster the strain rate, the higher the number of cracks formed at a given strain. An important observation was that when PET was coated with aluminum which is more ductile than platinum, there were no cracks formed at low elongations. The thickness of the coating is another
62 parameter that affects the crack formation. The thinner coatn.gs had less cracks than the thicker ones. Tn other words, they argued that there was a minnnum thickness of a coating after which it would start fonning cracks.
Letterier and coworkers also studied the fracture failure of silica coatings on PET films.'' They showed that the coating thickness was an important parameter that affected the crack density and the cohesive strength of the coating. They observed a higher crack density at saturation for thinner coatings, as shown in figure 3.6. Also, a thinner coating started to crack at a higher strain than a thicker one due to its cohesive strength.
5(1 100 150 200 Coating Thickness (nm)
Figure 3.6. Crack density and crack onset strain as a function of coating thickness. ''^ The thicker the coating was, the lower the onset strain and smaller crack density at saturation.
63 Experimental
Materials
Low density polyethylene film (0.07 mil) was a gift from Flex-O-Film
Corporation. To ensure reproducibility, it was always made sure that the film surl^ice facing inside the role was reacted and analyzed. Chromium (VI) oxide (Aldrich, 99%), sulfuric acid (Fisher, Certified ACS Plus grade), methylene chloride (Fisher, Optima grade), anhydrous ethanol (Aldrich, 99.5%) and 1-dodecanethiol (1-DCT) (Aldrich, 98%) were used as received.
General Methods
House purified water (reverse osmosis) was further purified using a Milliporc
Milli-Q® system that involves reverse osmosis, ion exchange and filtration steps (1()"^ ohm/cm). Contact angle measurements were made using a Ramc-Hart telescopic goniometer with a 24-gauge flat-tipped needle; dynamic advancing and receding angles were recorded as the probe fluid, water, purified as described above, was added to and withdrawn from the drop, respectively. Scanning electron micrographs (SEM) were obtained using a JEOL-35CF scanning electron microscope with an accelerating voltage of 20 kV. X-ray photoelectron spectra (XPS) were obtained using Perkin-Elmer-Physical
Electronics 5100 spectrometer with Mg K„ excitation at takeoff angles of 15" and 75".
Atomic concentrations of the substrates were calculated using the sensitivity factors, Cis,
64 u.2iu; U,„ 0.660; Au,„ 4.950 and 0.540, oblaincd from the samples of known composition.
Preparation of Polyethylene Substrates
The polyethylene films were eut into 1 3 x .5 inch pieces, placed in a thimble and extracted in a Soxhlet extraction apparatus using methylene chloride (or 2-3 days. They were then dried in yacuum overnight.
Oxidation of Polyethylene Substrates
The polyethylene substrates were placed in a custom designed (slotted hollow glass cylinder) sample holder and were oxidi/ed by submersion into a stirred mixture of
concentrated sulfuric acid, chromium trioxide and water (29:29:42 by weight) at 75 "C
for 1,5, 1 0, 1 5, 30 and 50 minutes. The oxidi/ed substrates were then rinsed with copious amounts of purified water and dried in yacuum oycrnight prior to uniaxial and biaxial deformation.
Uniaxial Streching of LDPE
The 5 minutes and 15 minutes oxidized polyethylene films were drawn uniaxially at extensions of 60, 120, 130 and 150 mm, and 30, 60, 80, 120, 150 and 200 mm,
respectiyely at a rate of 25 mm/min, using a Instron 441 1 mechanical testing machine.
65 Tlic extension values were the programmed values at wluci, the maehme stopped streching the fdm. The strains were suhscc,ucntly calculated hy measurmg the distance between two marked points on the Him before and after necking. Taking the ratio of the difference between the Tmal and the initial distances, over the initial one gave the strain.
Biaxial Streching of LDIMi
3x3 inch polyethylene lilm oxidi/ed for 5 minutes was biaxially streched using ; home-built apparatus in Lcsser's laboratory. The dim was constrained at all sides and at
a controlled How rate, it was blown biaxially with flowing water from underneath. The film was then characterized by SRM.
Self-Assembled Monolayers of 1 -dodecanethiol
The LDPli substrates were coated with 130 A gold using a Polaron Instruments
SEM Coating Unit (E51()()) at 15 niAmps for 2 and a half minutes. They were then placed in a custom designed (slotted hollow glass cylinder) sample holder and
subsequently immersed into a freshly prepared 10 niM solution of 1 -dodecanethiol (1-
DCT) in anhydrous ethanol for 24 hours. After that, the substrates were rinsed with anhydrous ethanol twice, and immersed in anhydrous ethanol for 2 hours prior to drying
in vacuum.
66 Results aiul Discussions
Oxidation of Polyethylene Substrates (o-LDPLi)
LDI>1< substrates were oxidized in chromic acid solution and the oxidation kinetics was studied. Throughout the discussion, the LDPE fdms treated with chromic acid solution are abbreviated o-LDPE. The reaction was carried out for 1,5,1 0, 1 5, 3() and 50 minutes. The atomic concentrations were calculated from the XPS spectra as shown in Table 3.2. As observed from the SEM images in figure 3.7, the chromic acid solution etched the amorphous regions much faster than the crystalline regions. Thus, after 5 minutes of oxidation, the bundles of crystalline lamellae were more visible on the surface. These lamellae started to get corroded at longer oxidation times. I he advancing and receding water contact angles - 16" 84" were Oa / 0|< 1 / for a clean LDPE substrate,
and they decreased to Oa / 0|< 73" / 6" upon oxidizing for I minute. After 10 minutes of oxidation, the dynamic contact angles stayed almost constant with a high hysteresis. The oxidized LDPE films were Anther gold coated and hydrophobizcd with self-assembled
monolayers (SAMs) of 1 -DCT. The dynamic water contact angles were about the same
for all these hydrophobizcd surfaces, Oa / 0|< 165" / 90". fable 3.3 siiows the dynamic water contact angle analysis for the LDPE surfaces that were oxidized and then
hydrophobizcd using SAMs of 1 -DCT.
67 oxicli/cd 1.131 L (o-LI)Pl,) with chromic acid lor 1,10, 30 and 50 minutes at 75 T.
o-LUPE LDPE o-LDPE o-LDPE o-LDPE (1 min) (10 mill) (30 min) (50 min) 15° 75°
%C 98 .88 99.58 91.64 94.25 91.48 93.81 91.84 92.39 91.71 93.07
"-""Q ^-^2 0.42 8.36 5.75 8.52 6.19 8.16 7.61 8.29 6.29
1 5 minutes
Figure 3.7. SEM mierographs of LDPE substrates that were oxidized with ehromie aeid for 5, 15, 30 and 50 minutes at 75 '^C.
68 Table 3.3. Dynamic water contact angles for oxidized LDPE (o-LDPE) substr ites that were then gold-coated and hydrophobized by SAMs of 1-DCT (o-LDPE/Au/l-DCT).
Oxidation Time o-LDPE o-LDPE/Au/ 1-DCT (minutes) g A Oa Gr
1 73" 6" 153" 92"
5 91° 11" 152° 92"
10 90" 9" 161" 89"
15 92" 9° 167" 90"
30 94" 9° 160° 90"
50 88" 8" 167" 106"
Uniaxial Deformation
First of all, for a control experiment, unoxidized polyethylene films were drawn uniaxially at strains (c) of 260% (extension = 80 mm), 314% (extension = 120 mm) and
570%) (extension = 200 mm) with 25 mm/min strain rate. The surface morphology was then studied by SEM. Figure 3.8 shows that there was no significant surface roughness evolving upon necking of the unoxidized polymer. The surfaces were relatively smooth, with some buckling observed for the surfaces strained at 260%.
The polyethylene films, oxidized for 5 minutes, were uniaxially streched at extensions of 60 mm, 100 mm, 120 mm, 130 mm and 150 mm, and the ones oxidized for
15 minutes were streched at extensions of 30 mm, 60 mm, 100 mm, 120 mm, 150 mm
69 3KU mm m\ l eu jeol
2eKu Kiseee 0001 i.eu jeol
Figure 3.8. SEM micrographs taken in the necking region of unoxidized LDPE substrates that were drawn uniaxially at strains (e) of 260% (extension = 80 mm), 314% (extension = 120 mm) and 570% (extension - 200 mm). The surface microstructure did not change with strain.
and 200 mm, at a rate of 25 mm/min. The corresponding strains were calculated and
plotted against extension, as shown in figure 3.9. At the first stages of drawing the
polymer, the strain increased linearly up to -300%, then stayed almost constant until it reached 150 mm extension. After that, due to the strain hardening of the polymer, the strain started to increase again. We expected to see a surface morphology change
70 >
between the three distinct regions, that is at low strains, at the pi ateau region and at very
high strains.
(a)
X/1
80 100 120 140 160 Extension (mm)
600 (b)
500
400
300 • 1—1 cd a 4— 200
100
0 0 50 100 150 200 250 Extension (mm)
Figure 3.9. Strain as a function of extension for LDPE films that were oxidized for (a) 5
minutes and (b) 1 5 minutes. Strain was calculated by measuring the distance between two marked points on the LDPE film before and after necking. Taking the ratio of the difference between the final and the initial distance, over the initial one gives the strain:
Strain = e = (final distance - initial distance)/initial distance .
71 ^^^^^^^^^^^^^^^^ In the lUerature discussed above, it was shown that thinner brittle coatings resulted ni smaller fragments after deformation than the thicker ones. We believe that the oxidation reaction can be considered as forming a thm brittle coating on the polymer. The thickness of this coating can be varied by controlling the oxidation time. According to this argument, oxidizing
LDPE for 5 minutes resulted in a thinner brittle layer on the surface than oxidizing it for
15 minutes. The microstructure resulted from the deformation of 5 minutes oxidized substrate was compared with the 1 5 minutes oxidized one.
Figure 3.10 shows the SEM micrographs of LDPE substrates that were oxidized with chromic acid for 5 minutes and then necked at strains of 140, 310, 320 and 400%.
The surface microstructure changed in the necking region with different strain values which ultimately had an effect on wettability of these surfaces. The water contact angles shown in the figure were after gold coating and reacting with 1-DCT. The surfaces
consisted of submicron size fragments and the size of the fragments were not observed to change dramatically with strain. This was the reason why the water contact angles of the drawn films were very similar to each other in the necking region up to the strain of
320%. These surfaces were hydrophobic with high advancing and receding contact
angles, but still exhibited hysteresis -25". At a higher strain of 400%), the fragments became smaller and more distant from each other. The receding contact angle increased to -150° at the strain of 400%, leading to an ultrahydrophobic behavior.
72 Figure 3.1 0. SI Figure 3.1 1 shows the SFM micrographs of 15 minute oxidized LDIM' films that were necked at strains ranging from 30 to 570%. When the o-Ll)PF was drawn to a small strain of -30%), the formation of small cracks was observed. This type of surface topography would not have much effect on wettability as proven by the dynamic water - 56" " contact angles of Oa / Or 1 / 95 after hydrophobi/ation. At strains > 250%, the 73 island formation became very visible in the necking region. These fragments were considerably bigger in size than the ones formed upon necking the 5 minute oxidized LDPE. There were couple of differences between the surface topography of the films that were drawn at strains of -250% and -400%, shown the m figure. First of all, the distance between the islands started to increase. Also, the island size became smaller with the increase in stram. These were reflected in the wettability of the surfaces such that the LDPE film strained at 250% exhibited / ~ 34° 0a Or 1 59V 1 whereas the one strained at had higher 400% advancing and receding water contact angles, 0a / 0r -167" / 150°. The interesting phenomenon was that when the oxidized LDPE was extended more to a strain of -570%, the surface was no longer hydrophobic. The size of the islands were < 0.5 ^m, and both the advancing and the receding contact angles decreased to 0a / 0R-158V 12r. 74 e = 59" 260% Oa / Gr = 1 / 1 29" e = = 250% Oa / 0,, 1 63^ / \ 340 Figure 3.1 1. SEM micrographs of LDPE substrates that were oxidi/ed with chromic acid for 1 5 minutes and then drawn uniaxially at strains (e) of 30% (extension = 30 mm), 1 80% (extension ^ 60 mm), 260% (extension = 80 mm), 250% (extension = 120 mm), 400% (extension = 150 mm) and 570% (extension = 200 mm). The surface microstrueture changed with strain in the necking region of the polymer. The water contact angles shown were after gold coating and reacting with 1-1)C'T (Au/l-DC f). 75 Biaxial Deformation of o-LDPE Biaxial deformation was also performed usmg 5 minutes oxidized LDPE films. An o-LDPE film, constrained at all sides, was blown biaxially by flowing water from underneath at a controlled flow rate. This resulted in a hemisphere. The strain was maximum at the top of the hemisphere and decreased going down the hemisphere. Figure 3.12 shows the change in surface topography across the hemisphere. Part 1 was the bottom of the hemisphere and hence it corresponds to the lowest strain. Part 2 and 3 were the regions with higher strain and Part 4 was the necked region at the top of the hemisphere. The SEM micrographs in the figure indicated that the surface topography changed with strain. As the strain increased from Part 1 to Part 3, the island fomiation started to be more apparent. These are called 'mud cracks' in the literature.'^" The islands were very discrete at the necked region. Part 5 was the unnecked region very close to the necked region and its SEM micrograph shows that the islands were about to be fonned with cracks in between them. The wettability of the substrates was studied after gold coating with a thickness of -130 A and hydrophobizing using self-assembled monolayers of alkanethiol. Figure 3.12 shows the corresponding dynamic water contact angles for each surface. The unnecked regions with various strain values exhibited low advancing and receding water contact angles. Since the islands were not discrete, the wettability of the surfaces was not very interesting. The necked region was predicted to have a hydrophobic behavior, however because it was a very small wrinkled area, contact angle measurements could not be performed on that region. 76 Pari 1 eA/eR= 121V69° I>art2 Qa/Or = 119"/ 73" 0824 1.0U JE. Part 3 1 0A / Ok - 1 2 77" " / Part 4 Oa / ()„ = can not be determined 20KIJ xi?0e9 oe24 i.eu jeol Parts Oa/Gr^ 121V75" Figure 3.12. SEM micrographs of LDPE substrates that were oxidized with chromic acid lor 5 minutes and then biaxially streched. The surface microstructure changed with strain. I^art 1 had the lowest and Part 4 had the highest strain. The strain increased in the order: Part 1 < Part 2 < Part 3 < Part 4 > Part 5. The water contact angles shown were after gold coating and reacting with 1-DCT (Au/l-DCT), except for Part 4 whose contact angles could not be measured due to the wrinkled necked region. 77 1 Deformation of Gold Coating Bonded to LDPE 50 nm thick gold film was sputtered coated onto clean LDPE films that were further deformed uniaxially with strains of 140%, 310%, 320% and 400%. Figure 3.13 shows the SEM micrographs of the streched films at the necking region. All of the substrates except for the last one shown m the figure were stretched at a stram rate of 25 mm/min. It was observed that there were small cracks forming at 140% strain and the long gold stripes became more apparent at higher strains. The length and the distance of the stripes did not change between the strains of 300% and 400%. The strain rate also affected the surface morphology. At 400% strain, the stripes were more separated from each other if streched at a faster rate. The wettability of the substrates was studied after gold coating with a thickness of -100 A and hydrophobizing using self-assembled monolayers of 1-dodecanethiol. In the figure, the dynamic water contact angles were also shown for each surface. The film streched at a low strain of 140% exhibited Ga / Or = 150V 1 T. Both the advancing and receding contact angles increased to -168" / 134" at higher strains when the gold stripes were more distant from each other. The contact angles of the surfaces remained very similar to each other at 300%) and 400%) strain. The surfaces were ultrahydrophobic leading to the water droplets rolling off of them at low tilt angles. 78 8 = 400% Oa/ Or - 165V 138" rate = 15 mm/min Figure 3.13. SEM micrographs of virgin LDPE substrates that were coated with gold and then necked at strains (e) -140% (extension = 60 mm), 310% (extension = 120 mm), 320% (extension = 130 mm), 400% (extension = 150 mm). The distance between the gold stripes increased with strain. The water contact angles shown were after gold coating and reacting with 1-DCT (Au/l-DCT). 79 Conclusions In this work, our aim was to explore a new practical way of tuning the wettability of polyethylene surfaces, and ultimately make ultrahydrophobic polyethylene films by controlling the surface microstructure. We achieved this through an oxidative etching of LDPE followed by either uniaxial or biaxial deformation. Chromic acid etching of LDPE led to the formation of a thin brittle surface layer. It etched the amorphous regions prefentially at first. 5 minutes and 15 minutes oxidized surfaces had different surface morphologies. Uniaxial deformation of LDPE and oxidized LDPE was perfonned with strains ranging from to 30% 570%. The effect of oxidadon time and strain on the formation surface microstructure was studied. Upon uniaxial tension, the brittle surface of the polyethylene film started to defomi into islands in the necking region. These fragments became smaller and more distant from each other with the increase in strain. Also, the surface topography was different depending on the oxidation time. 5 minutes oxidized LDPE consisted of smaller fragments at the necking region compared to the 15 minutes oxidized one. When LDPE was oxidized for 15 minutes, at low strains, the size of the islands were in the order of a few microns, and the dynamic water contact angles were 6a/ 0r -169" / 122" after hydrophobization. These contact angles were not significantly different from the unstreched o-LDPE films with Ga / 6r -167" / 90" after hydrophobization. However, at higher strains of 400%, the island fragments were further apart from each other and the 67" 50" substrates started to exhibit ultrahydrophobic behavior with 9a / Or -1 / 1 after 80 hydrophobization. The strain limit for impartmg hydrophobic behavior was ^570o/o. At this strain, the islands were very small and apart from each other, hence the receding water contact angle dropped sigmficantly to 121". They were no longer ultrahydrophobic. Biaxial deformation resulted in a microstructure called 'mud cracks', but with no interesting wettability. The water droplets pinned on the samples. Deformation of a gold coating on LDPE resulted in the formation of stripes of gold perpendicular to the streching direction. The distance between the gold stripes increased with strain. The surface morphology as well as the wettability of the substrates were very similar at high strains, -168" with Ga / Qr / 139" after hydrophobizing with an alkanethiol. 81 Notes and References (3) Arefi, F.; Montazer-Rahmati, P.; Andre, V.; Amouroux, J. J Polym ^u.,Sci Appi.Annl Polym.Symp 1990, 46,3. (4) Aouroux, J.; Arefi, F. Polym. Mater. Sci. Eng. 1990, 62, 418. Foerch, R.; Mclntyre, (5) N.S.; Hunter, D. H. J. Polym. Sci., Polym Chem Ed 1990 , 25,803. " — (6) Inagaki, N.; Tasak, S.; Kawai, H.; Kimura, Y. J. Adhes. Sci. Technol. 1990, 4, 99. Mutel, B.; Dessaux, (7) O.; Goudmand, P.; Grimblot, J.; Crapentier, A.; Szarzynski S. Rey. Phys. Appl. 1988, 2J, 1253. Dcrco, J.; (8) Lodes, A.; Lapcik, L.; Blecha, J. Chem. Zyesti 1984, 38, 463. (9) Sharma, A.K.; Millich, F.; Hellmuth, E.W. J. Appl. Polym. Sci. 1981, 26, 2205. (10) Cooper, G.D.; Prober, M. J.Polym. Sci. 1960, 44, 379. (11) Stardal, M.; Goring, D. A. I. Polym. Engineering Sci. 1977, / 7, 38. (12) Van der Linden, R. Polym. Sci. Technol. 1980, 12B, 563. (13) Carley, J. F.; Kitze, P. T. Polym. Eng Sci. 1980, 7/, 1 13. (14) Briggs, D. J. Adhes. 1982, 13, 287. (15) Lanauze, J. A.; Myers, D. L. J. Appl. Polym. Sci. 1990, 40, 595. (16) Blais, P.; Carlsson, D. J.; Csullog, G. W. ; Wiles, D. M. J. Coll. Interface Sci. 1974,47, 636. (17) Haridoss, S.; Perlman, M. M. J. Appl. Phys. 1984, 55, 1332. (18) Briggs, D.; Brewis, D. M.; Konieczo, M. B. ./. Mater. Sci. 1976, //, 1270. 82 Briggs D, Brewis D. M.; Comyn, J, Dalim, R. H. Green, M. A, B. Konieczkoi^onieczKo, MM. Surf. Interface Anal. 1980, 2, 107. ^ ^^^-^'^"^ T. Clark and W. 1 Feast Eds., ' Wiley, New York, 1978, Chapter 15. Rasmussen, J. R.; Stedronsky, E. R.; Whitesides, G. M. J. Am. Chem. Soc. 1977, Peeling, J.; Clark, D. T. J. Polym. Set, Polym. Chem. Ed. 1983, 27, 2047. Winslow, F. H Matreyek, ; W.; Trozzolo, A. M. Polym. Prepr., Am. Chem. Soc Div. Polym. Chem. 1969, 70, 1271. Blais, P.; Carlsson, D. J.; Wiles, D. M. J. Polym. ScL, PartA-1 1973, W, 1077. Fumeaux, G. C; Ledbury, K. J.; Davis, A. Polym. Degrad. Stab. 1981, i, 431. Adams, J. H. J. Polym. Sci, PartA-1 1970, 8, 1279. Heacock, J. F.; Mallory, F. B.; Gay, F. P. J. Polym. Sci., PartA-1 1968, 6, 2921. Davidson, R. S.; Meek, R. R. Eur. Polym. J. 1975, 77, 85. Owens, K. K. J. Appl. Polym. Sci. 1975, 79, 265. Olde Riekerink, B.; M. Terlingen, J. G. A.; Engbers, G. H. M.; Feijen, J. Langmuir 1999, 75, 4847. Rasmussen, J.R.; Stedronsky, E.R.; Whitesides, G.M. /. Am. Chem. Soc 1977 99 4736. Ferguson, G.S.; Whitesides, G.M., in Modern Approaches to Wettability, M. E. Schrader and G. I. Loeb, Eds., Plenum, New York, 1992, p 143. Holmes-Farley, S. R.; Reamey, R. H.; McCarthy, T. J.; Deutch, J.; Whitesides, G. M. Langmuir 1985, 7, 725. S. Wu, in Polymer Interface and Adhesion, Marcel Dekker, New York, 1982, p 279. Brewis, D.M.; Birggs, D. Polymer 1981, 22, 7. Holloway, F.; Cohen, M.; Westheimer, F. H. J. Amer. Chem. Soc. 1973, 9151, 65. Wiberg, K. B.; Eisenthal, R. Tetrahedron 1964, 20, 1 151. 83 . (38) Schonherr, H.; Vancso, G. J. J. Polym. ScL, Polym. Phys. 1998, 36, 2483. (39) Smith, P. F.; Chun, I.; Liu, G.; Dimitrievich, D.; Rasbum, J.; Vancso G J.' ' Polym. Eng. Sci. 1996, 36, 2129. ' (40) Kato, K. J. App. Polym. Sci. 1 977, 27,2735. (41) Bassett, D. C, in Principles ofPolymer Morphology, Cambridge University Press, Cambridge, 1981. (42) Price, G. J.; Keen, F.; Clifton, A. A. Macromolecules 1996, 29, 5664. (43) Thouless, M. D. Annu. Rev. Mater. Sci. 1995, 25, 69. Gill, G. in Current (44) Topics in Materials Science, Kaldis, E., ed., Amsterdam/New York, North Holland, 1985, 12:420, p 72. (45) Thouless, M. D. J. Vac. Sci. Technol 1991, A9, 2510. (46) Nir, D. Thin Solid Films 1 984, / 72, 4 1 (47) Amaratunga, G. A. J.; Welland, M. E. J. Appl. Phys. 1990, 68, 5140. (48) Kendall, K. J. Phys. D. Appl. Phys. 1971, 4, 1 186. (49) Thouless, M. D. Acta Metall. 1988, Jd, 3131. (50) Garino, T. J.; Bowen, H. K. J. Am. Ceram. Soc. 1987, 70, C315. (51) Nagl, M. M.; Saunders, R. J.; Evans, W. T; Hall, D. J. Corros. Sci. 1993, 35, 965. (52) Strawbridge, A.; Evans, H. E. Engineering Failure Analysis 1995, 2, 85. (53) Volynskii, A. L.; Voronina, E. E.; Lebedeva, O. V.; Yaminskii, I. V.; Bazhenov, S. L.; Bakeev, N. ¥ .Vysokomolekulyarnye Soedineniya Seriya A 2000, 42, 262. (in Russian) (54) Volynskii, A. L.; Lebedeva, O. V.; Bazhenov, S. L.; Bakeev, N. F. Vysokomolekulyarnye Soedineniya Seriya A 2000, 42, 665. (in Russian) (55) Volynskii, A. L.; Voronina, E. E.; Lebedeva, O. V.; Yaminskii, I. V.; Bazhenov, S. L.; Bakeev, N. F. Vysokomolekulyarnye Soedineniya Seriya A 2000, 42, 658. (in Russian) (56) Volynskii, A. L.; Voronina, E. E.; Lebedeva, O. V.; Yaminskii, I. V.; Bazhenov, S. L.; Bakeev, N. F. Vysokomolekulyarnye Soedineniya Seriya A 1999, 41, \627. (in Russian) 84 Letterier (57) Y, Andersons, J.; Pitton, Y.; Manson, J.-A. E. J. Polym Sci Pari^ B Polym. Phys. 1997, 35, 1463. ^ 85 CHAPTER 4 ADSORPTION OF POLYSTYRENE COLLOIDS ONTO MULTILAYER POLYELECTROI YTF ASSEMBLIES AND WETTING BEHAVIOR OF THESE ROUGH SURFACES Introduction In the literature, one of the ways that the researchers have tried to impart roughness to the solid surfaces is hy using spherical particles such as PTFE particles,'"^ silica particles^"^ and glass beads.*^ Our attempt is to prepare a series of micron and submicron scale rough surfaces by adsorbing monodisperse polystyrene lattices and explore their wettability. The polystyrene colloids with carboxylic acid groups are adsorbed onto the polyelectrolyte multilayer assemblies that have a cationic polyelectrolyte as the outermost layer. The kinetic study of the adsorption of colloids (0.35 |nm and I jum) onto the polyelectrolyte multilayers is performed to have an in-depth understanding of the adsorption phenomenon and monitored by atomic force microscopy. The adsorption occurs without any three-dimensional aggregation. The roughness (and hence the wettability) can be precisely controlled by varying both the diameter of the polystyrene colloids and the adsorption time. Solution casting of colloids onto the polyelectrolyte multilayers is also performed. All of the surfaces prepared are then hydrophobized by gold coating and reacting with an alkanethiol to form self-assembled monolayers. The wettability is assessed by dynamic contact angle measurement using water and hexadecane as probe fluids. 86 Polyelectrolytc Multilayer Assnmhli.>Q In the 1960's Ilcr developed a technique for adsorbing oppositely charged colloidal particles such as silica and alumina onto a substrate.' About three decades later, this work was extended by Decher and his coworkers""" to form multilayer assemblies of polyelectrolytes. Layer-by-layer assembly of polyelectrolytcs has been used to make ultrathin organic films. The principle of constructing multilayer assemblies is to adsorb anionic and cationic polyelectrolytes consecutively onto the substrate. The adsorption is driven electrostatic by attraction between the opposite charges of the polyelectrolytes. The film thickness can be controlled precisely because it is linearly proportional to the number of adsorbed polyelectrolytc layers. Depending on the ionic strength of the polyelectrolytc solution, the thickness of the individual layers can also be modified. For instance, in high salt concentrations, the electrostatic repulsions between the segments are screened and the polymer adsorbs with more loops, hence a thicker layer is observed.' The overall adsorption process has been shown to be independent of substrate size and topography. The formation of multilayer assemblies starts with immersing a positively (or negatively) charged substrate in an anionic (or cationic) polyelectrolytc solution. The polyelectrolytc adsorbs onto the substrate, and since a high concentration of polyelectrolytc is used, the surface charge is reversed by the ionic groups remaining at the interface of the substrate and the solution. Subsequently, the substrate is rinsed with water and immersed in a cationic (or anionic) polyelectrolytc solution. The reversal of 87 surface charge occurs when the cationic polyelcctrolyle adsorbs onto Ihc anion.e polyelectrolyte layer. If these steps arc repeated sequentially, n.ulflayer assemblies ar. formed. A wide range of substrates from inorganic to polymeric ones, has been used for multilayer assemblies. For instance, glass can be surface modified by reacting with a silane coupling reagent containing amine groups onto which anionic polyelectrolytes are adsorbed. In the McCarthy laboratory, layer-by-layer deposition was performed by using PMP,^^ 23 p^TFE,'' LDPE2^ and PTFE"''^' as substrates. In this research, the cationic polyelectrolyte, poly(allylamine hydrochloride) (PAH) " and the anionic polyelectrolyte, poly(sodium styrenesulfonate) (PSS) were used since they arc among the most studied systems with well-established results. It is known that twenty minutes is sufficient time for these polyelectrolytes to adsorb irreversibly onto the charged substrate. There is one more important point that should be stated. The complete reversal of surface charge and stratified layers were observed for PSS-PAH multilayer systems after building 10 layers.'^"'' For the experimental work of this research, it was crucial that the outermost layer was a positively charged surface so that the adsorption of carboxylated polystyrene colloids was achieved. 88 Adsorption of ColK)i(ls onto Polyclcctiolvit-^ Solution casting, dip-coaling and laycr-by-laycr deposition arc the methods used in the Uteraturc to make fihns of organized latex '" part.cles.^^ For instance, Krozcr and coworkers made aUernative layers of sihea and alumina using a quartz crystal microbalancc (QCM).'' Meanwhile, Lvov ct al. argued the fact that the use of polyions as intermediate layers or as 'electrostatic glue' was important in the assembly ol rigid particles. They made alternating assemblies of silica particles of 25, 45, 75 nm and poly(diallyldimethylammonium chloride) and followed the mass change during the assembly process by Q(^M.^" 1 lowcver, Ihey dk\n[ investigate the parameters to control the number of Si02 layers adsorbed onto the polyelectrolyte lllm, hence they usually observed three-dimensionally packed spheres. In addition to this, there are other studies where scientists assembled alternating layers of inorganic colloidal nanoparticles of CdS,-^"'"' gold,''^'''-^ and TiOVIMiS'*'''''^ with polyelectrolytcs. Although these have accomplished assemblies in two and three-dimensional arrays, it is not until tiie recent work of Akashi et al. where they attempted to study the adsorption behavior of colloids more extensively.''^' Akashi and coworkers investigated the alternating adsorption of polystyrene nanospheres (84, 548, 780 nm) onto (he surface ol a polyelectrolyte film thai was prepared by layer-by-layer assembly.'''' 'fhey believed that the adsorption was based on the electrostatic interaction and they studied the adsorption behavior by using QCIVI. They, first of all, assembled 7 alternating layers of PAll and PSS as the precursor film. 89 the top layer bc.ng I>AH. The adsorption of the polystyrene nanospheres was done at 10 "C and for up to 60 minutes, since they were concerned that the aqueous dispersion of the colloids would aggregate after 60 minutes or at higher temperatures (eg. 15-20 "C). The maximum surface coverage they achieved was 67% for the colloids with 780 nm and less coverage for smaller diameters. They concluded that even though small diameter colloids adsorb onto the surface, the maximum coverage was -55% for 548 nm and ~15"/o for 84 nm, as indicated in figure 4.1. 1 Figure 4. . Dependence of the surface coverage on adsorption time at a nanosphcre concentration of 3.8 x lO'" mL'' at 10 °C onto (PAH-PSS).rPAH surface that was prepared in the presence of 2 M NaCl: (a) 780 nm; (b) 548 nm; (c) 84 nm.'^'' The interesting observation was that there was a critical film thickness of 4 A of the outermost PAH layer needed for the colloidal adsorption. There was a dependence of the adsorption behavior of the colloids on the preparation conditions of the polyclectrolyte multilayers. For instance, as shown in figure 4.2, the surface coverage 90 increased from < 10% to > 55%, depending on if the polyeleetrolyle solutions were prepared without salt or with 2 M salt, respectively. They also made the point that they believed the adsorbed colloids did not aggregate or move two-dimensionaliy ui the dried state since the electrostatic interactions were very strong between the colloids and the polymer film. Adsorption time / min Figure 4.2. Time courses for polystyrene adsorption (diameter: 548 nm) at 10 "C onto - - (PAl 1 PSS), PAH surfaces prepared in the presence of NaCl: (a) 2 M; (b) 0.2 M; (c) 0.02 M and (d) 0 M.'*'' Self-Assembled Monolayers of Alkanethiols on Gold Self-assembled monolayer (SAM) systems have been extensively studied over the to last decade due to their potential scientific and technological applications from sensors microelectronics.^^ SAMs have the following advantages over the Langmuir-Blodgett be used, and 2) stable films can be prepared. SAMs assembly: 1 ) various solvents can 91 ) can be formed either by chemisorption of the molecules with specific sites onto the substrate, or by physisorption of the molecules driven by the long range forces. The first paper about SAMs of thiols showed that dialkylsulfides (RS-SR) form oriented monolayers on gold surfaces.^^ SAMs of alkanethiols on metal substrates are prepared by the formation of strong bonds between the substrate and the thiol. For instance, the hemolytic bond strength of methanethiolates on Au(l 11) is -44 kcal mol"'.^'' It has not been clear how the mechanism of the reaction of thiols with gold surface occurs. The formation of RS'— is Au^ believed to be either via the loss of the proton as H2 or as RS-H + Au°n RS -Au" + Y2 H2 + Au°n (4. i RS-H + + oxidant Au°n RS"-Au^ + V2 H2O + Au°n (4.2) Chemisorption of long chain alkanethiols, HS(CH2)nX, from solution onto the surfaces of metal substrates such as gold,^' copper,^^'^^ and silver^"* has been a versatile method of preparing water-repellent surfaces by altering the chemical structure of those Air-monolayer interface group Chemisorption at the surface Figure 4.3. Schematic illustration of SAMs of thiols on gold substrate. 92 surfaces. For instance, (he surfaces with SAMs of n-alkancth.ols have very low si.rface free energy 1<) m.l/ni^). ( The nu,noh.yers are densely packed and ordered w.th n,c,r I., group oriented towards the nionolayer-air or monolayer-liquid uiterlace. SAMs lonu smooth and homogeneous surfaces regardless of the underlying surface structure. The evaporation of gold on silicon yields the ( 1 1 phase 1) olgold (Au/Cr/Si)" on which the alkanethiolates self-assemble with an all trans conformation, tilted ^0" IVoni the surface normal.'"' The S— spacings are 4.*)7 A and the S calculated area per molecule is 21.4 A resulting in a ''^ pseudohexagonal two-dimensional symmetry, as shown in figure 4.4." 1 1 ). figure 4.4. STM image of an SAM of dodecanethiol on Au ( 1 Intcrmolecular spacings arc 0.50 nm which is V3 times larger than the atomic spacing of gold atoms (0.288 nm) in Au (1 1 1 ) plancs.^'^ In the literature, there are two different scenarios' to explain the thiol-metal surface interaction resulting in the cry.stalline-like order of alkanethiol monolayers, fhe 93 most likely mechanism is that the alkancthiols (Irst phys.sorb, and then chemisorh on the gold surface after a lateral motion to form gold mercapl.de units. The other one .s that the Au-S bond formed is covalent in nature and the R-S-Au assembling molecules have enough lateral surface mobility. The fact that the longer alkyl chains result in n1 more ordered self-assembly due to the van der Waals attraction forces than the shorter alkyl chains support the first scenario, while the second scenario can contribute to the long time scale ordering. The kinetics of the adsorption of alkancthiols were studied by Bain and coworkers.^' For 1 concentration, mM there were two adsorption steps observed as shown in figure 4.5. The first one was the fast step where the adsorption of an imperfect monolayer took place within a few minutes, f he advancing and the receding contact angles almost reached their limiting values and the thickness was -80-90% of its maximum. In the second step, the adsorption was much slower lasting for several hours. There was a reduction in the defects and enhanced packing was observed. In addition to these, Ulman found that a concentration of 10 niM can be used for SAMs of alkancthiols.^"* There was also the ef fect of chain length on adsorption kinetics such that the longer the alkyl chains, the faster was the adsorption due to stronger van der Waals interactions. 94 30 a • • • • C10 ^ 20 i . • • J 10 O CIO 0 •>y/*-r T r— 1 r 120 • o~ O 100 1 o Hp 80 m CD 60 O 0 40 ft fi D HD 20 0 I I 0 10' 10° 10' 10^ 10^ lO" Time (min) Figure 4.5. Kinetics of adsorption ofoctadecanethiol (solid symbols) and decanethiol (open symbols) from ethanol onto gold, (a) Ellipsometric thickness, (b) Advancing contact angles of water and hcxadecane (HD).''' Experimental Materials Si wafers (<100> orientation, P/B dopant, resistivity from 20 - 40 Q-cm, thickness from 400 - 450 )im) were purchased from International Wafer Service. Hydrogen peroxide (Fisher, 30%), sodium hydroxide (Fisher, Certified ACS Grade), ethanol hydrochloric acid (Fisher, 1 N), sulfuric acid (Fisher, Certified ACS Plus Grade), 95 ' (VWR, 99.5%), (4-aminobulyl)dimcthylmcthoxysilanc (ABDMS) (United Chemical Technologies), methanol (Fisher, I IPi.C grade), toluene (Aldrich, 99.9%), MnC1,.4l 1,0 (Fisher, Certified ACS Grade), poly(allyhnnine iiydroehlor.de) (PAll) (Aldnch, M„ = 500()()-650()()), poly(sodiLim styrcne suUbnale (I'SS) (Aldrich, M„ 7()()()()), 1- dodccancthiol (1-DCT) (Aldrich, 98%) were all used as received. C^arboxylated polystyrene colloids (Polyscicnces, 2.5% aqueous solution) with diameters ofO. 1 , 0.35, 0.5, 1, 2, 3.3 and 4.5 |,im and with concentrations of 3.64 x l()'\4.55x 1()'\ 1.06 x !()' 10", 10'", 3.64X 4.55 X 5.68 X 10", 1 x o", 10« .68 1 4.99 x particles per ml, respectively, as estimated by an equation given in the catalog were used after diluting with water and with no further puriHcalion. The pH values of the polyelectrolyte solutions and the colloidal suspensions were adjusted with small amounts of cither HCl or NaOll using a Fisher 825 MP pll meter. General Methods House purified water (reverse osmosis) was further purified using a Milliporc Milli-Q® system that involves reverse osmosis, ion exchange and llltralion steps (K)"^ ohm/cm). Contact angle measurements were made using a Rame-I lart telescopic goniometer with a 24-gaugc flat-tipped needle; dynamic advancing and receding angles were recorded as the probe fluids, water, purifled as described above, and liexadecane were added to and withdrawn from the drop, respectively. X-ray photoelectron spectra (XPS) were obtained using Perkin-Rlmer-Physical Rlectronics 5100 spectrometer with 5" 75". .sensitivity factors. Si 0.270; Mg Ka excitation at takeoff angles of 1 and The is, 9() O,. C. 0.250; 0.660; N,. 0.352; Au..„ 4.950 and S,„ 0.540, obtained IVon. ,hc sa.nplcs ofknown composition were used to calculate the XI>S atomic concentrations. The colloids tend to aggregate and this was prevented usnig Branson Soniller 450 ultrasonicator. Atomic force microscopy (AI (Nanoprohe^^) have spring constants ranging from 40 to 66 N/m. Nanoscope section analysis software was used to calculate the roughness ratio of the substrates. Preparation of Silicon Substrates The wafers were cut into 1.5x1 .5 cm pieces, placed in a custom designed (slottal hollow glass cylinder) sample holder and were cleaned by submersion into a mixture of concentrated sulfuric acid and hydrogen peroxide (7:3) overnight. They were then rinsed with copious amounts of purified water and dried in a stream of nitrogen immediately prior to the silani/ation reactions. Pretreatmcnt of Silicon Substrates Reaction of Silico n Substrates with (4-a minobutyl)dimelhylmcth()xysilane. The freshly cleaned silicon substrates were placed in a custom-designed (slotted hollow glass cylinder) sample holder and placed in a dry schlenk tube. 5% solution of (4- aminobutyl)dimethylmelhoxysilane (ABUMS) in anhydrous toluene was then cannulated 97 into the schlenk tube. The reactions were carried out for 3 days at 80 °C. The substrates were then rmsed with toluene aliquots), (2 1 : 1 methanol/toluene (2 ahquots), methanol (2 aliquots) and Milli-Q water (3 aliquots), each for 3 minutes. Layer-by-Layer Deposition of Poly electrolYtes, 0.02 M aqueous solutions of PSS and PAH (concentration based on repeat units) with 1 M MnCh were prepared and their pHs were adjusted to 2.2 and 2.9, respectively, with dilute HCl solution. The ABDMS- modified silicon substrates placed in a custom designed (slotted hollow glass cylinder) sample holder were subsequently immersed into the PSS solution for 30 minutes. After the deposition, the substrates were rinsed with three aliquots of water and immersed into the 0.02 aqueous solution M of PAH for 20 minutes. This was followed by rinsing the substrate with water and immersing it into the PSS solution again. Alternating layers of polyelectrolytes were formed when the substrates were immersed into PSS and PAH solutions consecutively. A total of 10 layers of PSS and PAH was deposited. The first PSS deposition took 30 minutes, and the following depositions took only 20 minutes. The substrates were rinsed with aliquots of Milli-Q water three times between each polyelectrolyte deposition. Adsorption of Polystyrene Colloids onto Polyelectrolyte Multilayers Carboxylated polystyrene colloids with diameters of 0.1, 0.35, 0.5, 1, 2, 3.3 and 4.5 ]xm were used. The latex solutions were diluted using fresh Milli-Q water and pH was adjusted to 8.8 by adding small amounts of dilute NaOH solution. The substrates 98 with polyclcctrolylc multilayers were immersed into the latex solution. The soiulions, was kept all the time in the ice bath and was sonieated every 10 muu.tes lor 10 seeonds. Alter hours ofadsorption, 6 the substrates were rinsed with Mill,-Q water (3 aliquots), eaeh for minutes. 3 They were dried in vacuum overnight. Kinetic Study ol'the Adsorption Rdi;.vi,..- nrr.ji..i.|^ 0.35 ,im and 1 ,im diameter colloids were used to study the kinetics ofadsorption behavior of colloidal particles onto the polyelectrolyte multilayer assemblies. The adsorption of I ^,m diameter colloids was carried out for 1/6, 1/2, 1, 2, 4, 6, 10, 24 hours, and that of 0.35 |Lim for 1/2, 2, 4, 6 and 10 hours. The substrates were rinsed with Milli-Q water (3 aliquots), each for 3 minutes. They were then dried in vacuum overnight. Solution Casting Polystyrene Colloids onto Polyelectrolyte Multilayers 10% (by volume) suspensions of polystyrene colloids (0.35, 0.5, 1,2 and 3.3 uni) in water were prepared, pi! was adjusted to 8.8 by adding small amounts of dilute NaOl I solution. After buikliiig 10 layers of polyeleclrolyles with PAIl as the outermost layer, the suspension was poured onto the substrates dropwise with a pipet. They were dried in air for 3 days. 99 Coating and Self-Assembled Monolayers (S,AM^^ of l-dodecanethiol A gold film with a thickness of ~1 10 A was thermally evaporated onto the substrates at pressures < lO"' Torr with deposition rate of -0.2 nm/sec, usmg Thermiomcs VE-90 vacuum evaporator. The substrates were immediately immersed in a freshly prepared solution of 10 mM l-dodecanethiol (1-DCT) in anhydrous ethanol. The reaction was carried out for 24 hours. After that, the substrates were rinsed with ethanol (3 aliquots) and then were put into ethanol for 2 hours. They were dried in vacuum overnight. Results and Discussions Pretreatment of Silicon Substrates The silicon substrate was modified prior to the polyelectrolyte adsorption. After being cleaned in the 'piranha solution', the silicon substrates were reacted with an amine- terminated silane reagent (ABDMS). A typical XPS survey spectrum of the substrate reacted with ABDMS is shown in figure 4.6 (a). The atomic nitrogen concentration was calculated to be 4.39 at 15° takeoff angle, and 1.65 at 75° takeoff 32". angle, as indicated in Table 4.1. The dynamic water contact angle was 0a / Or = 80° / 100 starting with the adsorption of PSS, 10 alternating layers ofpolyelectrolytes were then built. Figure 4.6 (b) shows a representative XPS survey speetrum of silicon substrate with 1 0 polycleetrolyte layers composed of PSS and PAl 1, with PAl 1 being the outermost layer. Nitrogemsulfur ratio, calculated 75° from takeoff angle XPS atomic concentration data, was ~1 .9. The water contact angles = were Oa / Or 91" / 8". These results were in good agreement with the previous work done by l lsieh^' in the McCarthy group. 1000.0 mo 800.0 700.0 Goo.o mn •loo.o m.o ?oo o loo o BIMDINC INtKCV. tV niNlllNC INIIiCV. cV 5° Figure 4.6. XPS survey of a silicon substrate at 1 takeoff angle (a) reacted with ABDMS for 3 days in toluene; (b) with 10 polycleetrolyte layers, PAIl as the outermost layer. 101 0 .^PS • concentration data for a clean ) silieon snhslrale. an ABOMS- 15" Takeoff Angle 75" Takeoff Angle Substrate %Si %C %0 %N %S %Si %C %0 %N %S Si 22.95 47.29 29.76 52.25 16.10 31.65 Si/ABDMS 24.87 38.48 32.26 4.39 45.50 13.91 38.95 1.65 Si/ABDMS/ PSS-PAH 73.47 16.10 7.45 2.98 69.75 18.35 7.85 4.05 (10 layers) Adsorption of Polystyrene Colloids onto Polyelectrolyte Multilayers Hsieh reported that after building up -10 multilayers, each polyelectrolyte layer became stratified rather than being interpenetrated.''^ In all our experiments, the adsorption of polystyrene colloids was performed onto the multilayer assemblies consisting of 10 polyelectrolyte layers. After treating the silicon wafer with ABDMS, 1 alternating layers of PSS and PAH were deposited. Then, the substrates were immersed into the dilute latex solution with a pH of 8.8. Although the polystyrene colloids were negatively charged, they were observed to form aggregates on the surface if only stirred at room temperature. This was overcome by using an ultrasonicator so that every 10 minutes, the solution in the ice bath was sonicated for 10 seconds. The wcllabilily of these rough surfaces was explored before and alter they were hydrophohi/cd by chemisorption of 1 -dodecancthiol (1-DCT). 102 The surface topography was imaged by performing AFM in TappingMode™. The height images captured at 40 ^m x 40 ^m scanned areas were used in section analysis software to calculate the Wenzel's roughness ratios. When performing such an analysis, a straight line was drawn randomly across the scanned area, and the program gave a graph of surface topography such as the one shown in figure 4.7. It calculated the surface distance from a peak to a valley and the horizontal distance of the line. 8 or more lines were drawn across a scanned area and the ratios of (surface distance)^ / (horizontal distance)' were calculated. The Wenzel's roughness ratio was taken as their mean value. I I 10.0 20.0 30.0 40.0 Surface distance 69.642 mh Horiz distanceCLJ 44,688 mm Figure 4.7. AFM section analysis was performed using AFM height images to calculate the roughness ratios of the substrates. The software provided both the surface distance (peak to valley distance) and the horizontal distance. The roughness ratio was calculated 2 2 using the following formula: r = (surface distance) / (horizontal distance) . 103 r Kinetic Study ofthc Adsorption Reh.vinrnf r^ iuj.,^ pi,3t ^,p.,„ ion behavior ofcarboxylatcd polystyrene eolloids onto polyclectrolyle nu.ll.layers with the outermost layer being PAH was explored. 0.35 and 1 ^im diameter polystyrene colloids were used for this purpose. The polystyrene colloids with 1 ^m diameter were adsorbed for 1/6, 1/2, 1, 2, 4, 6, 10 and 24 hours. The Al'M height images were used to calculate the Wen/el's roughness ratio of these surfaces. When the roughness ratios were plotted against the time of adsorption in figure 4.X, it clearly showetl that the roughness ratio increased linearly with time of adsorption until ten hours when it reached a value of ~2. then, it 14 and stayed almost constant. The A1 1 I I ' I I I I -I 1 I I T T—I—r— 2.2 I- 2.1 S 2.0 .X 0 ^ 1 n O 1.7 1.6 1.5 1.4 0 5 10 15 20 25 Adsorplion Time (hrs) Figure 4.8. The change in roughness ratio, calculated from the AFM height images, with adsorption time of polystyrene colloids (1 ^im) onto the polyelectrolyte multilayers. 104 1 0 iiiiinilcs (a) 1 .8 MM 0.9 MH 0.0 MM 40. U 0 MN 40.0 tm Data tupt K«l«ht Uatd 1g|>e Z rangt l.fiO MM 2 ranva Rd.Q da 30 iiiiiiutcs (b) 4(1.0 MM Data tui^a Height Data tupa I'lta^e 7. ranue l.RO MM 2 rantfa fiO.O (le 60 iniiuilcs (c) Data tu|)* Data tui>« 2 ranwa 1 Jill MM 2 ranuif surfaces obtained when (he I'ij^iire 4.9. AI'M height (left) and phase (right) images of the onto polyelectrolyte time of adsorption of polystyrene colloids ( I fini diameter) S, (h) and (i) multilayers was for (a) 10, (b) .10, (c) 60 minutes; (d) 2, (e) 4, (1) 6, (g) 10 24 hours, (continued next page) 105 2 hours (d) 1.8 MH 0.9 0.0 MH • • . 40.0 MM 0 40.0 Data type UM Height Data type Phase 2 range 1.80 MM 2 range 50.0 de 4 hours (e) 40.0 MM 0 40.0 MM Data type Height Data type Phase Z range 1.80 iiM Z range 50.0 de 6 hours (f) 40.0 MM 0 40.0 MM Data type Height Data type Phase Z range 1.80 MM 2 range 50.0 de re 4.9 (continued). 106 8 hours (g) i.8 PH 0.9 MM 0.0 PM 40.0 MM 0 Data type Height 40.0 MM Data type Phase 2 range 1.80 MM Z range 50.0 de 0 hours (h) 40.0 MN Data type Height Data type Phase 2 range 1.80 MM 2 range 50.0 ae 24 hours (i) UM U TlAtA tui>p 2 J'«lbi|fc' l.SO un 2 J cuiyir 50.0 Figure 4.9 (continued). 107 The adsorption of polystyrene colloids witli 0.35 urn diameter was carried out lor 1/2, 2, 4, and 1 hours. In 6 0 contrast to 1 ^m, the adsorption kinetics for colloids w.th diameter 0.35 showed a faster adsorption occurring. The roughness rat.o had a sharp increase with time in the Hrst two hours, and then leveled offal 1.9, as mdicaled ni ngure 4.10. The surface coverage increased in a shorter period of time than observed for I ^m colloidal adsorption, as seen from the AFM images in figure 1 1 . 4. 1 lence, wc observed that the larger the diameter of polystyrene colloids, the more time it needed to achieve a high surface coverage. It was not very clear why the surfaces with I diameter colloids had relatively big gaps between the colloids, even at 24 hour adsorption. We believe that the adsorption behavior was different for 0.35 ^im and 1 [im due to the fad that the adsorption of colloids from the solution to the substrate was 0 2 4 6 8 10 Adsoiplion Tunc (hi s) with Figure 4. 1 0. The change in roughness ratio, calculated from AFM height images, adsorption time of polystyrene colloids (0.35 [im) onto the polyelectrolytc multilayers. 108 difftision-controlled at the early stages. The smaller the diameter, the easier for the negatively charged colloids to difftise from the solution and adsorb onto the positively charged surface. Also, the smaller diameters might have higher surface charge density that resulted in faster adsorption. This difference in the adsorption behavior was also shown, quantitatively, in the analyses of the evolution of roughness ratios with time of adsorption, in figures 4.8 and 4.10. 1 12 hour (a) 600.0 nM 300.0 HM 0.0 HM 40.0 MN 0 40.0 MM Data type Height Data type Phase 2 range BOO RM Z range 60.0 de 2 hours (b) 40.0 MM Data type Phase Z range 60.0 de Figure 4.11. AFM height (left) and phase (right) images of the surfaces obtained when the time of adsorption of polystyrene colloids (0.35 |im diameter) onto polyelectrolyte multilayers was for (a) 1/2, (b) 2, (c) 4, (d) 6 and (e) 10 hours, (continued next page) 109 4 hours (c) 600.0 HH 300.0 nM 0.0 nM 40.0 MM 0 Data type Height 40.0 UN Data type Phase Z range 600 HN Z range 60.0 de 6 hours (d) Data type Height Z range 600 nN 1 0 hours (e) 40.0 MM 0 40.0 MM Data type Height Data type Phase 2 rang* 600 nN Z range 60.0 de ^ure 4.1 1 (continued). 110 After colloidal adsorption, the substrates were dried overnight and then coated with ~1 gold. 10 A These were then immersed into an ethanol solution of 1-DCT that resulted in the formation of self-assembled monolayers of alkanethiols on gold. The substrate with multilayer assemblies was also hydrophobized via this method for - comparison and the XPS atomic concentration data is shown in Table 4.2. Table 4.2. XPS atomic concentration data for an ABDMS-modified silicon substrate and after building 1 polyelectrolyte layers 0 composed of PSS and PAH, PAH as the outermost layer which were then hydrophobized with self-assembled monolavers of 1- DCT. ^ 1 5° Takeoff Angle 75° Takeoff Angle Substrate %Au %C %0 %S %Au %C %0 %S Si/ABDMS 10.54 79.88 7.63 1.95 43.68 52.86 - 3.46 Si/ABDMS/ PSS-PAH 13.55 68.81 14.71 2.94 42.75 54.14 - 3.11 (10 layers) Figure 4.12 shows the dynamic water contact angle versus time of adsorption of 1 |Lim and 0.35 |im colloids after the substrates were hydrophobized. For both surfaces, the advancing contact angle increased while the receding contact angle decreased with adsorption time. However, there was a more gradual increase (decrease) in advancing (receding) water contact angles for the surfaces with 1 |im colloids than for the ones with 0.35 |im. In other words, the hysteresis rose linearly with time in the case of 1 \\.m and reached a maximum value of -122° after 24 hours of adsorption, whereas for 0.35 |im colloids, the hysteresis increased substantially from 77° to 125° in the first two hours, and 111 » stayed almost constant afterwards. What this meant physically was that the water droplets pimied on these surfaces due to the high hysteresis. They did not roll off the surfaces even -80° at tilt angle. Hence, these surfaces were not promising candidates as ultrahydrophobic surfaces although they had very high advancing water contact angles. 80 (a) I 60 140 h 120 o 100 2 80 60 40 n I I J 1 1 I u 0 5 10 15 20 25 Adsorption Time (hrs) (b) o -4— Uo 0 2 4 6 8 Adsorption Time (hrs) Figure 4.12. The advancing () and the receding () water contact angle with time of adsorption of (a) 1 |Lim, (b) 0.35 |im colloids after the substrates were hydrophobized with self-assembled monolayers of 1-DCT. 112 The dynamic contact angles of the substrates with 1 ^im and 0.35 mii colloids were measured using water and hexadecane as probe fluids, and they are given in Table 4.3 and 4.4. Table 4.3. Dynamic contact angles for surfaces with 1 |am colloids, before and after hydrophobizing with self-assembled monolayers of 1-DCT, using water and hexadecane (HD) as probe fluids, (time = 0 is for polyelectrolyte multilayers with PAH as the outermost layer) Au/ 1-DCT Au/ 1-DCT Time Or Oa Or Oa Or (hours) water water water water HD HD 91° 8° 0 118° 92° 49° 17° 1 92° 8° 118° 90° 49° 17° 2 91° 10° 128° 82° 46° 9° 4 99° 9° 138° 69° 45° 10° 6 106° 10° 146° 59° 40° 9° 10 107° 9° 155° 57° 37° 9° 24 109° 7° 163° 42° 26° 8° Table 4.4. Dynamic contact angles for surfaces with 0.35 |im, before and after hydrophobizing with self-assembled monolayers of 1-DCT, using water and hexadecane (HD) as probe fluids. . Au/l-DCT Au/l-DCT Time Ba Or Oa 9a 9r (hours) water water water water HD HD 1 81° 9° 139° 62° 35° 7° 2 99° 6° 159° 34° 18° 6° 4 102° 8° 163° 38° 25° 7° 6 105° 8° 161° 35° 26° 7° 10 100° 7° 161° 37° 23° 7° 113 Summary ol the Rouah Surfaces Prepared Our aim was to vary the roughness of the substrates by simply using eollo.ds of different diameters and adsorbing them onto the polyeleetrolyte multilayers. The ehange in surface roughness would ultimately affect the wettability of surfaces. The colloids with 0.1, 0.35, 0.5, 1, 2, 3.3 and 4.5 ^im diameters were adsorbed onto the multilayer assembly for 6 hours. The AFM height and phase images are shown in figure 4. 1 3. The surfaces consisting of 0. and 1 0.35 ^m colloids had a full surfiice coverage. 0. 1 ^m colloids showed 3-dimensional aggregation on the surface. As the diameter increased to 0.5 [im and higher, the surfiice coverage decreased. A possible explanation for the decrease in surface coverage with increasing diameter is that the adsorption process is diffusion controlled and the smaller colloids diffused faster from the solution and adsorb onto the PAH surface layer faster than the larger ones. Hence, it would take longer than 6 hours for the larger colloids to have a full coverage on the surface, as observed for the case with 1 )iim colloids. These rough surfaces were further coated with gold and reacted with l-DCT to impart hydrophobicity. Table 4.5 lists the dynamic water contact angles for the surfaces prepared by using polystyrene colloids with sub-micron and micron diameters. There was a clear difference in the wettability behavior between the surfaces that had less than 1 ^m colloids and the ones with greater than 1 ^im. When the diameter of the colloids high advancing were between 0.1 pm and 1 pm, the hydrophobi/cd surfaces had very water contact angles (-160"), except for 0.5 pm where a good surface coverage was not 1 14 achieved. These surfaces exhihiled remarkahle hysteresis with h.gh advancing hut low receding water contact angles. When these surfaces were tdted, even at SO", the droplets stayed pinned. There was great resistance to rolling across the surface due to the high hysteresis. When the diameter was greater than 1 ^m, the surface coverage of the colloids decreased remarkably as can be seen from the AhM images. They did not alTcct the wettability of the surfaces. Hence, the dynamic contact angles of these surfaces after hydrophobi/ation were very similar to that of a smooth surface consisting of only polyelectrolyte multilayers ~ which exhibited Oa / Or 118"/ 92". Table 4.5. Dynamic contact angles for surfaces prepared by adsorbing polystyrene colloids with the indicated diameters, before and after hydrophobi/ing with self- assembled monolayers of dodecanethiol (Au/l-Dd ), using water and hexadecane (ill)) as probe (luids. r is the roughness ratio calculal(ul iVom the A1<'M height image with 40 |j,m X 40 |im scanned area. Au/1 -Dcr Au/1 -DCl Diameter Oa Or Ba Or Oa r 0R (^m) water water water water IID 111) 0 1 91° 8° 118° 92° 49° 17° 90 63" 0.1 1.81 75" 1 95" 45" 90 0.35 1 .93 105° 8° 161° 35° 26° 70 6° 0.5 1 .36 72" 135" 85" 46" 11" 90 1 1.85 1 06° 10° 146° 59° 40° 2 1.61 70" 5° 124" 82" 47" 11" 3.3 1.9 78" 5° 117" 95" 46" 23" 4.5 1.47 81" 6" 118" 96" 45" 17" 115 0.1 |„im (a) 250.0 HM 125.0 HH 0.0 HM 0.35 |im (b) 600.0 RH 300.0 HM •hi 0.0 nH .0 LIH 60.0 ae 0.5 |im (C) 850.0 HM 425.0 RM 0.0 HM Phase BU.O dw 1 urn (d) 1 .8 MM 0.9 MM 0.0 iJM using Figure 4. 1 3. AFM height (left) and phase (right) images of the surfaces obtained (a) 0.1, (b) 0.35, (c) 0.5, (d) 1, (e) 2, (f) 3.3 and (g) 4.5 [im diameter polystyrene colloids that were adsorbed onto the polyelectrolyte multilayers for 6 hours, (continued next page) 116 2 f.im (e) 3.5 MM 1.8 UN 0.0 MM 40.0 Aw 3.3 |Lim (0 5.0 MM 2.5 UM 0.0 MM 4.5 |im (g) 5.0 MM 2.5 MM 0.0 MM •10.0 MM Data tu|i< h.OO MM 7 r-4(iiiti HO.O (!• Figure 4.13 (conlinucd). 117 Solution Casting Polystyrene Colloids onto PolvelectrolytP Mnltil.yers A full coverage could not be achieved with the surfaces prepared by adsorbing carboxylated polystyrene colloids onto polyelectrolyte multilayers. Hence, solution casting technique was used to have a full coverage of colloids on the surfaces. 0.35, 0.5, 1, 2 and 3.3 ^im colloidal suspensions were spread onto multilayer assemblies forming 0.5 |im o«t« tup* 2 rangr* 1 |im 2 \im 0 40.0 UM 0 40.0 UM 0 UH 0 40.0 UH 2 r«n«* 2.00 UM Z r*r>«« £0,0 im ^ rang* 2. SO iim 2 r«ng« 60.0 d» 3.3 |im Figure 4. 14. AFM height (left) and phase (right) images of the surfaces obtained by solution-casted using 0.35, 0.5, 1, 2 and 3.3 urn diameter polystyrene colloids that were onto polyelectrolyte multilayers, and dried for 3 days. 118 surfaces as shown in figure 4.14. These surfaces were further hydrophobized to explore their wettability. Table 4.6 gives the dynamic contact angles (for water and hexadecane) for these surfaces after hydrophobizing them with SAMs of thiols. The water contact angle hysteresis of these surfaces were less, but still was not low enough to impart ultrahydrophobicity. Table 4.6. Dynamic contact angles for surfaces prepared by solution casting polystyrene colloids and hydrophobizing with self-assembled monolayers of 1-dodecanethiol (Au/1- DCT), using water and hexadecane (HD) as probe fluids, r is the roughness ratio calculated from the AFM height image with 40 \im x 40 |im scanned area. Diameter Ga Gr Ga r Gr (Mm) water water HD HD 0.35 1.83 159° 115° 26° 7° 0.5 1.34 124" 72" 37" 8" 1 2.1 161° 116° 24° 40 2 2.31 150° 118° 29° 8° 3.3 1.94 161° 119° 24° 11° 119 Conclusions Submicron and micron scale rough surfaces were prepared by adsorbing carboxylated polystyrene colloids onto polyelectrolyte multilayer assembly consisting of 10 alternating layers of anionic polyelectrolyte, PSS, and cationic polyelectrolyte, PAH. The negatively charged colloidal particles were efficiently adsorbed onto the outermost cationic polyelectrolyte surface, showing no aggregation as indicated in the AFM images. Using various diameters of colloids enabled us to develop a better understanding of their adsorption behavior onto the polyelectrolyte multilayers. A kinetic study of adsorption was performed using 0.35 |Lim and 1 |^m diameter colloids. AFM analyses showed that the smaller the diameter, the faster the adsorption. An almost full surface coverage was achieved using colloids with submicron diameter. Also, the flexibility of altering the roughness of the surface was gained by varying the diameter of the colloids. The wettability was explored before and after the rough surfaces were hydrophobized by chemisorption of an alkanethiol. The advancing water contact angle increased and the receding contact angle decreased as the surface coverage (and the roughness) increased for both 0.35 |im and 1 \im colloidal surfaces. The advancing water contact angle was as high as -163° for the surface with 1 ^im colloids after 24 hours of adsorption. However, these surfaces exhibited a remarkable hysteresis of -122" which resulted in the pinning of the water droplet on the surface. For instance, the water droplets were not rolling off the surface even at high tilt angles. This was an observation along the lines of our discussion about wettability and hysteresis in the first chapter, such that even though the advancing contact angle was very high, the hysteresis was the real determining factor for 120 hydrophobicity rather than the absokite values of water contact angles. From this study, we concluded that hydrophobic surfaces could not be achieved by making rough surfaces by colloidal adsorption. 121 . . . Notes and References (1 Saiki, Y.; Nakao, M.; ) Ono, M. Kogyo-Zairyo 1996, 44, 52. (2) Tokuumi, A.; Hiromatsu, K.; Kumai, S.; Mihara, H. Toso-to-Toryo 1998, 571, 37. (3) Takai, K.; Saito, H.; Yamauchi, G. Proceedings of the Composites: Design for Performance mi, 22Q. (4) Welters, W. J.; Fokkink, L. G. Langmuir 1998, /4, 1 535. (5) Sasaki, H.; Shouji, M. Chem. Lett. 1998, 293. Shouji, M.; Sasaki, H.; (6) Kawashima, K. J. Chem. Soc. Jpn. 1998, 12, 837. (7) Murase, H.; Nanishi, K.; Kogure, H.; Fujibayashi, T.; Tamura, K.; Hariita N J. Appl. Polym. Sci. 1 994, 54, 205 1 (8) Dcttre, R.H.; Johnson, R. R. Jr. Adv. Chem. Ser. 1963, 43, 136. (9) Her, R. J. Colloid Interface Sci. 1 966, 21, 569. (10) Decher, G.; Hong, J.-D. Ber. Bunsen-Ges. Phys. Chem. 1991, 95, 1430. (11) Decher, G.; Hong, J.-D.; Schmitt, J. Thin Solid Films 1992, 210, 83 1 (12) Decher, G.; Schmitt, J. Progr. Colloid Polym. Sci. 1992, 89, 160. (13) Lvov, Y.; Decher, G.; Mohwald, H. Langmuir 1993, 9, 481. (14) Sukhorukov, G. B.; Schmitt, J.; Decher, G. Ber. Bunsenges. Phys. Chem. 1996, 100, 948. (15) Korneev, D.; Lvov, Y.; Decher, G.; Schmitt, J.; Yarodaikin, S. Physica B 1995, 213, 954. (16) Decher, G. Multilayer Films (Polyelectrolytes), in The Polymeric Materials Encyclopedia: Synthesis, Properties and Applications; Salamone, J. C, Ed., Vol. 6, CRC Press: Baco Raton, 1996, p 4540. J. ./. Phys. Chem. 1 984, <^<*^, 666 1 ( 1 7) Van dcr Schee, H.A.; Lyklema, 2717. (18) Stockton, W.B.; Rubner, M.F. Macromolecules 1997, 30, 122 (19) Ferreira, M.; Cheung, J.H.; Rubner, M.F. Thin Solid Films 1994, 244, 806. (20) Schmitt, J.; Grunewald, G.; Decher, G.; Pershan, P.S.; Kjaer, K.; Losche M Macromolecules 1993, 26, 7058. (21) Watanabe, S.; Regan, S.L. J. Am. Chem. Soc. 1994, 1 16, 8855. (22) Levasalmi, J.-M.; McCarthy, T. J. Macromolecules 1997, 30, 1752. Chen, (23) W.; McCarthy, T. J. Macromolecules 1997, 30, 78. Phuvanartnuruks, V.; (24) McCarthy, T. J. Macromolecules 1998, 57,1 906. Hsieh, (25) M. C. Ph.D. Thesis, University of Massachusetts Amherst, 1999. Hsieh, M. Farris, (26) C; R. J.; McCarthy, T. J. Macromolecules 1997, 30, 8453. (27) Ferreira, M.; Rubner, M.F. Macromolecules 1995, 28, 71 07. (28) Denkov, N.D.; Velev, O.D.; Kralchevsky, P.A.; Ivanov, I.B.; Yoshimura, H.; Nagayama, K. Langmuir 1992, 5, 3 1 83. (29) Gao, M.; Zhang, X.; Yang, Y.; Yang, B.; Shen, J. J. Chem. Soc, Chem. Commun. 1994, 2777. (30) Kotov, N.A.; Dekany, I.; Fendler, J.H. J. Phys. Chem. 1995, 99, 13065. (31) Dimitrov, A.S.; Dushkin, CD.; Yoshimura, H.; Nagayama, K. Langmuir 1994, JO, 432. (32) Feldheim, D.L.; Grabar, K.C.; Natan, M.J.; Mallouk, T.E. J. Am. Chem. Soc. 1996, 118, 7640. (33) Dimitrov, A.S.; Nagayama, K. Langmuir 1996, 12, 1303. (34) Li, J.; Meier, D.J. Polymer Preprints 1996, 37, 591. (35) Johnson, C.A.; Lenhoff, A.M. J. Colloid and Interface Sci. 1996, 1 79, 587. (36) Ariga, K.; Lvov, Yu; Onda, M.; Ichinose, I.; Kunitake, T. Chem. Lett. 1997, 125. 479. (37) Krozer, A.; Nordin, S.-A.; Kasemo, B. J. Colloid Interface Sci. 1995, 1 76, 6117 (38) Lvov, Y.; Ariga, K.; Ichinose, I.; Kunitake, T. J. Am. Chem. Soc. 1995, 117, 25. (39) Ariga, K.; Onda, M.; Lvov, Y.; Kunitake, T. Chem. Lett. 1997, 123 . . ' (40) Lvov, Y.; Ariga, K Onda, .; M .; Ichinosc, I.; Kunitakc, T. Langmmr 1997 / ? ()1 95. ' ^ (41) Kotov, N. A.; Dckany, I.; Fcndler, J. H. Adv. Mater. 1996, 637. (42) Schmitt, J.; Dcchcr, G.; Geer, R.E.; Drcssick, WJ.; Calvert, .1. M ^Hn.iviaicr.Adv Malcr 1997,9,61. (43) Yonczawa, T.; Onouc, S.; Kunitake, T. Adv. Mater. 1998, /O, 414. (44) Sun, Y.; Hao, E.; Zhang, X.; Yang, B.; Gao, M.; Shen, J. J. Chem. Soc Chem " Commun. 1996, 2381. ' (45) Sun, Y.; Hao, E.; Zhang, X.; Yang, B.; Gao, M.; Shcn, J.; Chi, L.; Fuchs H Langmuir ]997, 13, 5\6^. Serizawa, T.; (46) Takcshita, H.; Akashi, M. Langmmr 1998, 14, 4088. (47) Allara, D. L., Biosensors and Bioelectronics 1995, 10, (48) Nuzzo, R. G.; Allara, D. L. ./ Am. Chem. Soc. 1983, 105, 4481 (49) Dubois, L.H.; Nuzzo, R.G. Annu. Rev. Phys. Chem. 1992, 43, 437. (50) Abraham Ulman, in An Introduction to Ullrathin Organic Films from Langmuir- Blodgetl to Self-Assembly, Academic Press, Inc. San Diego, CA, 1991, p 288. (51) Bain, CD.; Troughlon, E.B.; Tao, Y.-T.; Evall, J.; Whitesides, G.M.; Nuzzo, R. G.J. Am. Chem. Soc. 1989, ///, 321. (52) Stewart, K. R.; Whitesides, G. M.; Godfricd, H. P.; Siilvera, I. F. Surf. Sci. 1986, 57,1381. (53) Blackman, L. C. F.; Dcwar, M. .T. S. J. Chem. Soc. 1957, 7, 171 (54) Ulman, A. J. Mat. Ed. 1989, //, 205. (55) Abraham Ulman, in An Introduction to Ultrathin Organic Films from Langmuir Blodgett to Self-Assembly, Academic Press, Inc. San Diego, CA, 1991, p 286. (56) Fenter, P.; Eisenberger, P.; Liang, K.S. Phys. Rev. Lett. 1993, 70, 2447. (57) Strong, L.; Whitesides, G.M. Langmuir 1988, 4, 546. (58) Chidsey, C.E.D.; Loiacono, D.N. Langmuir 1990, 6, 709. 124 (59) Dubois, L.H.; Zegarski, B.R.; Nuzzo, R.G. J. Chem. Phys. 1993, 98, 678. (60) Chidsey, C.E.D, Liu, G.-Y .; Rowntree, P, Scoles, G. J. Chem. Phys. 1989, 97, Alves, (61) C.A.; Smith, E.L.; Porter, M.D. J. Am. Chem. Soc. 1992, 1 14, 1222. (62) Delamarche, E.; Michel, B.; Gerber, C; Anselmetti, D.; Gunthelrodt H -1 • Wolf H.; Ringsdorf, H. Za«gmw/r 1994, 70, 41 16. (63) Abraham Ulman, m An Introduction to Ultrathin Organic Films from Langmuir- Blodgett to Self-Assembly, Academic Press, Inc. San Diego, CA, 1991, p 296. (64) Ulman, A.; Tillman, N. Langmuir 1989, 5, 1418. 125 BIBLIOGRAPHY Abraham Ulman, in An Introduction to Ultrathin Organic Films from Langmuir-Blodgett to Selj-Assembly, Academic Press, Inc. San Diego, C 1 99 A, 1 , p 286. Abraham Ulman, in An Introduction to Ultrathin Organic Films from Langmuir-Blodgctt to Self-Assembly, Academic Press, Inc. San Diego, CA, 1991, p 288. Abraham Ulman, in An Introduction to Ultrathin Organic Films from Langmuir-Blodgett to Self-Assembly, Academic Press, Inc. San Diego, CA, 1991, p 296. Adams, J. H. J. Polym. ScL, Part A- J 1970, 1279. Allara, D. L., Biosensors and Bioelectronics 1995, 10, 771. Alves, C.A.; Smith, E.L.; Porter, M.D. J. Am. Chem. Soc. 1992, 1 14, 1222. Amaratunga, G. A. J.; Welland, M. E. J. Appl. Phys. 1990, 68, 5140. Andrade, J. D. X-ray Photoelectron Spectroscopy; Andrade, J. D., Ed.; Plenum Press' New York, 1985, Vol. l,p 105. Andrade, J.D. in Surface and Interfacial aspects ofBiomedical Polymers, Andrade, J.D., Ed., Plenum Press, New York, 1985; Vols. 1 and 2. Aouroux, j.; Arefi, F. Polym. Mater. Sci. Eng. 1990, 62, 418. Arefi, F.; Montazer-Rahmati, P.; Andre, V.; Amouroux, J. J. Polym. Sci., Appl. Polvm. Symp 1990, 46,?>. Ariga, K.; Onda, M.; Lvov, Y.; Kunitake, T. Chem. Lett. 1997, 25. Ariga, K.; Lvov, Yu; Onda, M.; Ichinose, I.; Kunitake, T. Chem. Lett. 1997, 125. Bain, CD.; Troughton, E.B.; Tao, Y.-T.; Evall, J.; Whitesides, G.M.; Nuzzo, R. G. J. Am. Chem. Soc. 1989,777,321. Bartell, F. E.; Shepard, J. W. J. Phys. Chem. 1953, 57, 211. Barthlott, W.; Neinhuis, C. Planta 1997, 202, 1. Bassett, D. C., in Principles ofPolymer Morphology, Cambridge University Press, Cambridge, 1981. Bell, A. T. Solid State Technol. 1978, 27, 89. 126 Binning, G.; Quate, C. F.; Gerber, C. H. Phys. Rev. Lett. 1986, 56, 930. Blackman, L. C. F.; Dewar, J. M. S. J. Chem. Soc. 1957, 7, 171. Blais, P.; Carlsson, D. J.; Wiles, D. M. J. Polym. ScL, PartA-1 1973, 10, 1077. Blais, P.; Carlsson, D. J.; Csullog, G. W. Wiles, D. ; M. J. Coll. Interface Sci 1974 47 . , 636. , Bowden, M. J. in Introduction to Microlithography, 2"^* ed., Thompson, L. F.; Willson, C. G.; Bowden, M. J., Eds; American Chemical Society, Washington DC, 1994, Chapter 2. Brewis, D.M.; Birggs, D. Polymer 1981, 22, 7. Briggs, D.; Brewis, D. M.; Konieczo, M. B. J. Mater. Sci. 1976, 77, 1270. Briggs, D.; Brewis, D. M.; Comyn, J.; Dahm, R. H. Green, M. A.; Konieczko, M. B. Surf. Interface Anal. 1980, 2, 107. Briggs, D. J. Adhes. 1982, 13, 287. Carley, J. F.; Kitze, P. T. Polym. Eng. Sci. 1980, 11, 113. Cassie, A.B.D.; Baxter, S. Trans. Faraday Soc. 1944, 40, 546. Cassie, A. B. D. Discuss. Faraday Soc. 1948, 3, 1 1. Chaudhury, M. K.; Whitesides, G. M. Langmuir 1991, 7, 1013. Chen, W.; McCarthy, T. J. Macromolecules 1997, 30, 78. Chen, W.; Fadeev, A. Y.; Hsieh, M.; Oner, D.; Youngblood, J.; McCarthy, T. J. Langmuir 1999, 75, 3395. Chidsey, C.E.D.; Liu, G.-Y.; Rowntree, P.; Scoles, G. J. Chem. Phys. 1989, 91, 4421. Chidsey, C.E.D.; Loiacono, D.N. Langmuir 1990, 6, 709. Clark, D. T.; Thomas, H. R. J. ofPolym. Sci. \911, 15, 2843. Claudy, P.; Letoffe, J. M.; Gaget, C; Morell, D.; Serpinet, J. J. Journal of Chromatography 1987, 395, 73. Cooper, G.D.; Prober, M. J.Polym. Sci. 1960, 44, 379. 127 Davidson, R. S.; Meek, R. R. Eur. Polym. J. 1975, 1 1, 85. Decher, G.; Hong, J.-D. Ber. Bunsen-Ges. Phys. Chem. 1991, 95, 1430. Decher, G.; Hong, J.-D.; Schmitt, J. Thin Solid Films 1992, 210, 831. Decher, G.; Schmitt, J. Progr. Colloid. Polym. Sci. 1992, 89, 160. Decher, G. Multilayer Films (Polyelectrolytes), in The Polymeric Materials Encyclopedia: Synthesis, Properties and Applications; Salamone, .T C Ed Vol 6, CRC Press: Baco Raton, 1996, p 4540. Delamarche, E.; Michel, B.; Gerber, C; Anselmetti, D.; Gunthelrodt, H.-J.; Wolf, H.; Ringsdorf, H. Langmuir 1994, 70, 41 16. Denkov, N.D.; Velev, O.D.; Kralchevsky, P.A.; Ivanov, I.B.; Yoshimura, H.; Nagayama K. Langmuir \992, 8, ?>\%2>. Derco, J.; Lodes, A.; Lapcik, L.; Blecha, J. Chem. Zvesti 1984, 38, 463. Dettre, R.H.; Johnson, R. R. Jr. Adv. Chem. Ser. 1963, 43, 136. Dettre, R.H.; Johnson, R.E. in Wetting; S.C.I. Monograph No. 25, Society of Chemical Industry, London, 1967, p 144. Dimitrov, A.S.; Dushkin, CD.; Yoshimura, H.; Nagayama, K. Langmuir 1994, 10, 432. Dimitrov, A.S.; Nagayama, K. Langmuir 1996, 12, 1303. Drelich, J.; Miller, J. D. Part. Sci. Technol. 1992, 3, 11. Drelich, J.; Miller, J. D. Langmuir 1993. 9, 619. Drelich, J. Ph.D. Dissertation, University of Utah, 1993. Drelich, J.; Miller, J. D.; Kumar, A.; Whitesides, G. M. Colloids Surf., A: Physicochem. Eng. Aspects 1994, 93, 1. Drelich, J.; Wilbur, J. L.; Miller, J. D.; Whitesides, G. M. Langmuir 1996, 12, 1913. Dubois, L.H.; Nuzzo, R.G. Annu. Rev. Phys. Chem. 1992, 43, 437. Dubois, L.H.; Zegarski, B.R.; Nuzzo, R.G. J. Chem. Phys. 1993, 98, 678. Fadeev, A. Y.; McCarthy, T. J. Langmuir 1998, 14, 5586. 128 Fadeev, A. Y.; McCarthy, T. J. Langmuir 1999, 15, 3759. Fadeev, A. Y.; McCarthy, T. J. Langmuir 1999, 1 5, 7238. Fadeev, A. Y.; McCarthy, T. J. Langmuir 2000, 76, 7268. Feldheim, D.L.; Grabar, K.C.; Natan, M.J.; Mallouk, T.E. J. Am. Chem. Soc. 1996. 118, Fenter, P.; Eisenberger, P.; Liang, K.S. Phys. Rev. Lett. 1993, 70, 2447. Ferguson, G.S.; Whitesides, G.M., in Modern Approaches to Wettability, M. E Schrader and G. I. Loeb, Eds., Plenum, New York, 1992, p 143. Ferreira, M.; Cheung, J.H.; Rubner, M.F. Thin Solid Films 1994, 244, 806. Ferreira, M.; Rubner, M.F. Macromolecules 1995, 28, 7107. Foerch, R.; Mclntyre, N.S.; Hunter, D. H. J. Polym. Sci., Polym. Chem Ed 1990 28 ' 803. ' Furmidge, C. G. L. J. Colloid Sci. 1962, 17, 309. Fumeaux, G. C; Ledbury, K. J.; Davis, A. Polym. Degrad. Stab. 1981, J, 431. Gao, M.; Zhang, X.; Yang, Y.; Yang, B.; Shen, J. J. Chem. Soc, Chem. Commun 1994 2777. Garino, T. J.; Bowen, H. K. J. Am. Ceram. Soc. 1987, 70, C315. Gill, G. in Current Topics in Materials Science, Kaldis, E., ed., Amsterdam/New York, North Holland, 1985, 12:420, p 72. Hall, J.R.; Westerdahl, C.A.L.; Devine, A.T.; Bodnar, M. J. J. Appl. Polym. Sci. 1969, 13, 2085. Hansen, R.H.; Pascale, J. V; De Bendictis, T.; Rentzepis, P. M. J. Polym. Sci. 1965, A3, 2205. Haridoss, S.; Perlman, M. M. J. Appl. Phys. 1984, 55, 1332. Heacock, J. F.; Mallory, F. B.; Gay, F. P. J. Polym. Sci. PartA-1 1968, 6, 2921. Holloway, F.; Cohen, M.; Westheimer, F. H. Amer. Chem. Soc. 1973, 9151, 65. 129 Holmes-Farley, S R^; Reamey, R. H, McCarthy, T. J, Deutch, J, Wh.tes.des, G M Langmuir 1985, /, 725. Hozumi, A.; Takai, O. Thin Solid Films 1997, 303, 222. Hsieh, M. C. Ph.D. Thesis, University of Massachusetts Amherst, 1999. Hsieh, M. C; Farris, R. J.; McCarthy, T. J. Macromolecules 1997, 30, 8453. Her, R. J. Colloid Interface Sci. 1966, 21, 569. Inagaki, N.; Tasak, S.; Kawai, H.; Kimura, Y. J. Adhes. Sci. Technol. 1990, 4, 99. Ingall, M.D.K.; Honeyman, C.H.; Mercure, J. V.; Bianconi, P.A.; Kunz R R J Am Chem. Soc. 1999, 121, 3607. Johnson, C.A.; Lenhoff, A.M. J. Colloid and Interface Sci. 1996, 179, 587. Johnson, R. E., Jr.; Dettre, R. H. Adv. Chem. Ser. 1963, 43, 136. Johnson, R.E.; Dettre, R.H. Surface Coll. Sci. 1969, 2, 85. Kato, K. J. App. Polym. Sci. 1977, 21, 2735. Kendall, K. J. Phys. D. Appl. Phys. 1971, 4, 1 186. Komeev, D.; Lvov, Y.; Decher, G.; Schmitt, J.; Yarodaikin, S. Physica B 1995, 213, 954. Kotov, N.A.; Dekany, I.; Fendler, J.H. J. Phys. Chem. 1995, 99, 13065. Kotov, N. A.; Dekany, I.; Fendler, J. H. Adv. Mater. 1996, 637. Krozer, A.; Nordin, S.-A.; Kasemo, B. J. Colloid Interface Sci. 1995, 176, 479. Kumar, A.; Whitesides, G. M. Appl Phys. Lett. 1993, 63, 2002. Kumar, A.; Biebuyck, H. A.; Whitesides, G. M. Langmuir 1994, 10, 1498. Kunugi, Y.; Nonaku, T.; Chong, Y.B.; Watanabe, N. J. Electroanal. Chem. 1993, 353, 209. Lanauze, J. A.; Myers, D. L. J. Appl Polym. Sci. 1990, 40, 595. Letterier, Y.; Andersons, J.; Pitton, Y.; Manson, J.-A. E. J. Polym. Sci. Part B: Polym. Phys. 1997, JJ, 1463. 130 Levasalmi, J. M.; McCarthy, T. J. Macromolecules 1997, 30, 1752. Leyden D E., Ed., Silcmes, Surfaces, and Interfaces, Gordon and Breach: New York ' I VOO. Li, J.; Meier, D.J. Polymer Preprints 1996, 37, 591. Lisichkin, G.V., Ed., Chemically Modified Silicas in Adsorption, Chromatography and Catalysis, Khimiya, Moscow, 1986. Lvov, Y.; Decher, G.; Mohwald, H. Langmuir 1993, 9, 481. Lvov, Y.; Ariga, K.; Ichinose, 1.; Kunitake, T. J. Am. Chem. Soc. 1995, //7, 6117. Lvov, Y.; Ariga, K.; Onda, M.; Ichinose, L; Kunitake, T. Langmuir 1997, 13, 6195. Miller, .I.D.; Veeramasuneni, S.; Drelich, .1.; Yalamanchili, M.R.; Yamauchi Y PolYin yi«^. 6c/. 1996,36, 1849. Mittal, K. L., Ed., Silane and Other Coupling Agents, VSP, Zeist, 1992. 2'"' Mucha, J. A.; Hess, D. W.; Aydil, E. S. in Introduction to Microlithography, cd., Thompson, L. F.; Willson, C. G.; Bowden, M. J., Eds; American Chemical Society, Washington DC, 1994, Chapter 5. Murase, H.; Nanishi, K.; Kogurc, H.; Fujibayashi, T.; Tamura, K.; Haruta, N. ./. Appl. Polym. Sci. 1994, 54, 2051. Mutcl, B.; Dcssaux, O.; Goudmand, P.; Grimblot, .T.; Crapentier, A.; Szarzynski, S. Rev. Phys. Appl. 1988,23, 1253. Nagl, M. M.; Saunders, R. .T.; Evans, W. T; Hall, D. J. Corros. Sci. 1993, 35, 965. Nir, D. Thin Solid Films 1984, 112, 4L Nuzzo, R. G.; Allara, D. L. J. Am. Chem. Soc. 1983, 105, 4481. Ogawa, K.; Soga, M.; Takada, Y.; Nakayama, 1. Jpn. J. Appl. Phys. Pt. 2 - Lett. 1993, 32, L614. Olde Rickerink, M. B.; Terlingen, J. G. A.; Engbers, G. H. M.; Feijen, .1. Langmuir 1999, 15, 4847. Onda, T.; Shibuichi, S.; Satoh, N.; Tsujii, K. Langmuir 1996, 12, 2125. Owens, K. K. J. Appl. Polym. Sci. 1975, 19, 265. 131 Oner, D.; McCarthy, T. J. Langmuir 2000, /6, 7777. Peeling, J.; Clark, D. T. J. Polym. ScL, Polym. Chem. Ed. 1983, 21, 2047. Pesek, J. J., Leigh, I. E., Eds., in Chemically Modified Surfaces; Royal Society of Chemistry, Cambridge, 1994. Phuvanartnuruks, V.; McCarthy, T. J. Macromolecules 1998, 37, 1906. Plueddemann, E. P. in Silane Coupling Agents; T'^ ed.; Plenum, New York, 1991. Powell, R.A. in Materials Processing Theory and Practices, Vol. 4: Dry etching for Microelectronics, North-Holland Physics Publishing, Netherlands, 1984, p 1 19. Price, G. J.; Keen, F.; Clifton, A. A. Macromolecules 1996, 29, 5664. Rasmussen, J. R.; Stedronsky, E. R.; Whitesides, G. M. J. Am. Chem. Soc. \977 99 4736. S. Wu, in Polymer Interface and Adhesion, Marcel Dekker, New York, 1982, p 279. Saiki, v.; Nakao, M.; Ono, M. Kogyo-Zairyo 1996, 44, 52. Sasaki, H.; Shouji, M. Chem. Lett. 1998, 293. Schakenraad, J.M.; Stokroos, 1.; Bartels, H.; Busscher, H.J. Cells and Materials 1992, 2, 193. Schrnitt, J.; Grunewald, G.; Decher, G.; Pershan, P.S.; Kjaer, K.; Losche, M. Macromolecules 1993, 26, 7058. Schrnitt, J.; Decher, G.; Geer, R.E.; Dressick, W.J.; Calvert, J. M. Adv. Mater. 1997, 9, 61. Schonherr, H.; Vancso, G. J. J. Polym. ScL, Polym. Phys. 1998, 36, 2483. Serizawa, T.; Takeshita, H.; Akashi, M. Langmuir 1998, 14, 4088. Sharma, A.K.; Millich, F.; Hellmuth, E.W. J. Appl. Polym. Sci. 1981, 26, 2205. 19512. Shibuichi, S.; Onda, T.; Satoh, N.; Tsujii, K. J. Phys. Chem. 1996, 100, 837. Shouji, M.; Sasaki, H.; Kawashima, K. J. Chem. Soc. Jpn. 1998, 12, 132 Smith, P. Chun I, Liu G, Dimitrievich, D, Rasbum, J.; Vancso, G. / Polvn, Eng. oci. I vvo, JO, 2129. Stardal, M.; Goring, D. A. I. Polym. Engineering Sci. 1977, 7 7, 38. Stewart, K. R.; Whitesides, G. M.; Godfried, H. P.; Siilvera, I. F. Surf. Sci. 1986. 57, 1381. Stockton, W.B.; Rubner, M.F. Macromolecules 1997, 30, 2717. Strawbridge, A.; Evans, H. E. Engineering Failure Analysis 1995, 2, 85. Strong, L.; Whitesides, G.M. Langmuir 1988, 4, 546. Sukhorukov, G. B.; Schmitt, J.; Decher, G. Ber. Bunsenges. Phys. Chem. 1996, 100, 948. Sun, Y.; Hao, E.; Zhang, X.; Yang, B.; Gao, M.; Shen, J. J. Chem. Soc., Chem. Commuu 1996, 2381. Sun, E.; Y.; Hao, Zhang, X.; Yang, B.; Gao, M.; Shen, J.; Chi, L.; Fuchs, H. Langmuir 1997,73,5168. Tadanaga, K.; Katata, N.; Minami, T. J. Am. Ceram. Soc. 1997, 80, 1040. Tadanaga, K.; Katata, N.; Minami, T. J. Am. Ceram. Soc. 1997, 80, 3213. Takai, K.; Saito, H.; Yamauchi, G. Proceedings of the Composites: Design for Performance 1997, 220. Thoden van Velzen, E. U.; Engbersen, J. F. J.; Reinhoudt, D.N. J. Am. Chem. Soc. 1994, 776, 3597. Thompson, L. F. in Introduction to Microlithography, 2"^ ed., Thompson, L. F.; Willson, C. G.; Bowden, M. J., Eds., American Chemical Society, Washington DC, 1994, Chapter 1. Thouless, M. D. Acta Metall. 1988, 36, 3131. Thouless, M. D. J. Vac. Sci. Technol \99\,A9, 2510. Thouless, M. D. Amm. Rev. Mater. Sci. 1995, 25, 69. Tokuumi, A.; Hiromatsu, K.; Kumai, S.; Mihara, H. Toso-to-Toryo 1998, 571, 37. Ulman, A.; Tillman, N. Langmuir 1989, 5, 1418. 133 . Ulman, A. J. Mat. Ed. 1989, 77, 205. Ulman, A. MRS. Bull. June 1995, 46. Van der Linden, R. Polym. Sci. Technol. 1980, 725, 563. Van der Schee, H.A.; Lyklema, J. J. Phys. Chem. 1984, 88, 6661 Veeramasuneni, S.; Drelich, J.; Miller, J.D.; Yamauchi, G. Prog Org Coat 1997 31 6 , 265. • , Volynskii, A. L.; Voronma, E. E.; Lebedeva, O. V.; Yaminskii, 1. V.; Bazhenov, S. L.;' Bakeev, N. F. Vysokomolekulyarnye Soedineniya Seriya A 1999, 41, 1627. (in Russian) Volynskii, L.; A. Voronina, E. E.; Lebedeva, O. V.; Yaminskii, I. V.; Bazhenov, S. L.; Bakeev, N. F .Vysokomolekulyarnye Soedineniya Seriya A 2000, 42, 262. (in Russian) Volynskii, A. L.; Voronina, E. E.; Lebedeva, O. V.; Yaminskii, I. V.; Bazhenov, S. L.; Bakeev, N. F. Vysokomolekulyarnye Soedineniya Seriya A 2000, 42, 658. (in Russian) Volynskii, A. L.; Lebedeva, O. V.; Bazhenov, S. L.; Bakeev, N. F. Vysokomolekulyarnye Soedineniya Seriya A 2000, 42, 665. (in Russian) Wagner, T.; Neinhuis, C; Barthlott, W. Acta Zool. 1996, 77, 213. Watanabe, T.; Yamaguchi, I. Journal ofPesticide Science 1991, 16, 651. Watanabe, S.; Regan, S.L. J. Am. Chem. Soc. 1994, 116, 8855. Weiss, T.; Scheirbaum, K.D.; Thoden van Velzen, U.; Reinhoudt, D.N.; Gopel, W. Sens. Actuators 1995, B26-27, 203. Welters, W. J.; Fokkink, L. G. Langmuir 1998, 14, 1535. Wenzel, R.N., Ind. Eng. Chem. 1936, 27, 105. Wenzel, R. N., J. Phys. Colloid Chem. 1949, 53, 1466. Wiberg, K. B.; Eisenthal, R. Tetrahedron 1964, 20, 1 151. and W. J. Feast, Eds., Willis, H. A.; Zichy, V. J. I. in Polymer Surfaces, D. T. Clark ' Wiley, New York, 1978, Chapter 15. 134 Winslow, F. H., Matreyek, W.; Trozzolo, A. M. Polym. Prepr., Am. Chem. Soc Div' Polym. Chem. 1969, 70, 1271. ' Winters, H. F. in Plasma Chemistry III; Topics in Current Chemistry Series 94 Sprineer- Verlag, Heidelberg, Germany, 1980, p 69. Yamauchi, G.; Miller, J.D.; Saito, H.; Takai, K.; Ueda, T.; Takazawa, H.; Yamamoto H • Nislhi, S. Coll. and Surf., A: Physicochemical and Engineering Aspects 1996^ Yang, X; Shi, J.; Johnson, S.; Swanson, B. Langmuir 1998, 14, 1505. Yonezawa, T.; Onoue, S.; Kunitake, T. Adv. Mater. 1998, 70, 414. Young, T. Philos. Trans. R. Soc. London 1805, 95, 65. Youngblood, J. P.; McCarthy, T. J. Macromolecules 1999, 32, 6800. 135