1 UNDERSTANDING THE SURFACE CHEMICAL AND MECHANICAL PROPERTIES OF HYDROGEL MATERIALS FOR CONTACT LENS APPLICATIONS By YUCHEN HUO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMEN T OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013
2 2013 Y uchen Huo
3 These efforts are dedicated to the loving memory of my mother Hong Tian (1953 2002)
4 ACKNOW LEDGMENTS These past four years have been nothing shor t of an exciting endeavor for me in the quest of truth about science and beyond I am most greatly indebted to my advisor, Professor Scott Perry, for not only providing me with scholarly advice, but al so for modeling to me the attributes of a responsible and passionate scientist. I am g rateful for his support when this journey became bumpy at times and for his guidance when the vision to move forward seemed unclear. I am also honored to have the oppo rtunity working alongside of and learning from a group of dedicated, intelligent colleagues. Specifically, my gratitude goes to Dr. Xueying Zhao for always willing to be the go to person for me when ever I needed and for Kevin Gilley for his constant cama raderie in facing all the instrumental challenges during numerous AFM experiments. Additionally, I would like to extend my appreciation to Professor Greg Sawyer, Prof essor Tommy Angelini, Dr. Alison Dunn, and Juan Uruea for their collaboration in completing the work on the surface mechanical studies of hydrogel materials. My walk as a doctoral student would have been so much more difficult if it were not for the lovin g support of numerous friends. I am warmed by the friendship of Jamie and Howard Hsu, Shelley and Brandon West, and Heather and Hayden Norris, who upheld me with their prayers even when I did not know. I am particularly grateful for Helen Lin, who has sh own me much love and kindness especially during the last days of finishing this dissertation. As for my family near and afar, there is perhaps no diction comprehensive enough that can express my thankfulness to them. I am indebted to my aunt and uncle,
5 Jenny and Bill Wilcox, for their outpouring of love in accepting me into their family when I first arrived in the States. It is the confidence in knowing that their arms are always wide open for me that shaped my idea of a loving family To my parents an d Qinqin Huo, Hua Fang, Ming Tian, and Aiwen Bai I owe the deep affection that has withstood the test of distance and time. To Julian, my soon to be husband, I am most thankful for his decis ion to begin this journey called together with me an d cannot wait to experience all that is in store for us hereafter. And ultimately, to the Savior, who has upheld my faith t hroughout this process, I owe the immeasurable depth of grace in all that He has done for me which is
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF FIGURES ................................ ................................ ................................ ........ 11 ABSTRACT ................................ ................................ ................................ ................... 14 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 16 The Origin and Applications of Surface Properties ................................ ................. 17 Surface Interaction Forces ................................ ................................ ............... 17 Interaction forces between atoms under ideal conditions .......................... 18 Interaction forces between surfaces in aqueous environment ................... 19 Interaction forces between polymeric surfaces in aqueous environments 21 Organic Surface Modifications ................................ ................................ .......... 23 Organic Surface Characterization Techniques ................................ ................. 24 X ray photoelectron spectroscopy ................................ .............................. 24 Atomic force micros copy ................................ ................................ ............ 25 Surface Properties of Contact Lens Materials ................................ ......................... 27 Interfacial Phenomena at Contact Lens Surfaces in vivo ................................ 28 Modifications of Lens Surfaces for Improved Interfacial Properties .................. 28 Scope of the Present Study ................................ ................................ .................... 30 2 INSTRUMENTAL METHODS AND EXPERIMENTAL APPROACH ....................... 39 X ray Photoelectron Spectroscopy ................................ ................................ ......... 40 Fundamentals of XP S ................................ ................................ ...................... 40 Analyzing Hydrogel Samples with XPS ................................ ............................ 43 Quantitative Chemical Analysis with XPS ................................ ......................... 44 Elemental peak identification ................................ ................................ ..... 45 Atomic Force Microscopy ................................ ................................ ........................ 47 Fundamentals of AFM ................................ ................................ ...................... 47 Assembly of an AFM ................................ ................................ .................. 47 Principles of AFM ................................ ................................ ....................... 48 Quantitative Analysis with AFM ................................ ................................ ........ 50 Force distance plots ................................ ................................ ................... 51 Friction load maps ................................ ................................ ..................... 53 Force calibration of cantil evers ................................ ................................ .. 55 Characterizing Hydrogel Materials with Colloidal AFM ................................ ..... 58 Contact pressure consideration ................................ ................................ 58 Indentation measurements on hydrogel materials ................................ ..... 60
7 3 IMPACT OF ETHYLENE OXIDE BLOCK BUTYLENE OXIDE (EO BO) COPOLYMERS ON THE COMPOSITION AND FRICTION OF SILIC ONE HYDROGEL SURFACES ................................ ................................ ....................... 84 Background and Motivation ................................ ................................ .................... 84 Experimental ................................ ................................ ................................ ........... 87 Materials ................................ ................................ ................................ ........... 87 Surface Chemical Composition Analyzed by XPS ................................ ............ 87 Friction Response and Surface Topography Probed by AFM .......................... 88 Results ................................ ................................ ................................ .................... 89 Surface Composition ................................ ................................ ........................ 89 AFM Analysis ................................ ................................ ................................ ... 93 Discussion ................................ ................................ ................................ .............. 94 Concluding Remarks ................................ ................................ ............................... 96 4 SURFACE AND BULK UPTAKE OF ETHYLENE OXIDE B LOCK BUTYLENE OXIDE COPOLYMER BY SILICONE HYDROGEL CONTACT LENS MATERIALS ................................ ................................ ................................ ......... 108 Background and Motivation ................................ ................................ .................. 108 Experimental ................................ ................................ ................................ ......... 110 Materials ................................ ................................ ................................ ......... 110 Bulk Uptake of EO BO Analyzed by UPLC ................................ ..................... 110 Surface Che mical Composition Analyzed by XPS ................................ .......... 111 Results ................................ ................................ ................................ .................. 111 Hydrogel Total Uptake of EO BO ................................ ................................ ... 111 Lens Surface Modification by EO BO ................................ ............................. 112 Discussion ................................ ................................ ................................ ............ 114 Concluding Remarks ................................ ................................ ............................. 118 5 INFLUENCE OF A SURFACE GEL LAYER ON THE ELASTIC MODULUS OF SILICONE HYDROGEL MATERIALS MEASURED WITH COLLOIDAL PROBE AFM ................................ ................................ ................................ ...................... 126 Background and Motivation ................................ ................................ .................. 126 Experimental ................................ ................................ ................................ ......... 128 Materials ................................ ................................ ................................ ......... 128 Methodology ................................ ................................ ................................ ... 129 Results ................................ ................................ ................................ .................. 130 Discussion ................................ ................................ ................................ ............ 132 Concluding Remarks ................................ ................................ ............................. 133 6 AN INVESTIGATION OF THE DEPENDENCE OF SURFACE MECHANICAL PROPERTIES ON WATER CONTENT OF VARIOUS HYDROGEL MATERIALS ................................ ................................ ................................ ......... 138 Background and Motivation ................................ ................................ .................. 138
8 Experimental ................................ ................................ ................................ ......... 141 Hydrogel Synthesis ................................ ................................ ........................ 141 Temperature variant Measurements in AFM ................................ .................. 142 Surface Elastic Modulus Analyzed by Colloidal Probe AFM ........................... 143 Results ................................ ................................ ................................ .................. 144 Surface Elastic Moduli of PAAM, P(HEMA co MAA), and PHEMA in DI Water ................................ ................................ ................................ .......... 144 PAAM ................................ ................................ ................................ ....... 144 P(HEMA co MAA) and PHEMA ................................ ............................... 145 Temperature induced Phase Transition and the Surface Mechanical Properties of PNIPAM ................................ ................................ ................. 146 Change in surface elastic modulus an d adhesion force ........................... 146 Change in surface topography by tapping mode AFM ............................. 147 Discussion ................................ ................................ ................................ ............ 148 Analyzing Surface Elastic Modulus of Various Hydrogel Thin Films with Colloidal Probe AFM ................................ ................................ ................... 148 Temperature induced Phase Transition in PNIPAM Hydrogel .................... 150 Characteristics of Force Indentation Behavior on Hydrogel Surfaces ............ 152 General remarks on analyzing force curves with colloidal probe AFM ..... 152 Change in interaction forces following phase transition of PNIPAM surface ................................ ................................ ................................ .. 153 Concluding Remarks ................................ ................................ ............................. 154 7 CONCLUSIONS AND FUTURE DIRECTIONS ................................ .................... 172 Recapitulation ................................ ................................ ................................ ....... 172 Surface Uptake of Amphiphilic Copolymer by Silicone Hydrogel Materials .... 173 Surface Mechanical Properties of Hydrogel Materials ................................ .... 174 Perspectives and Implicatio ns ................................ ................................ .............. 177 Future Research ................................ ................................ ................................ ... 178 APPENDIX A NORMAL FORCE CONSTANT CALIBARTION OF AFM CANTILEVERS ........... 181 Overview ................................ ................................ ................................ ............... 181 The Thermal Method ................................ ................................ ............................. 181 The Sader Method ................................ ................................ ................................ 182 Cantilever geometry consideration ................................ ................................ 183 Added mass consideration ................................ ................................ ............. 183 Sample Calculations ................................ ................................ ............................. 1 84 The Sader Method ................................ ................................ .......................... 184 The Thermal Method ................................ ................................ ...................... 185 Discussion ................................ ................................ ................................ ............ 185 B ANGLE RESOLVED XPS (AR XPS) ANALYSIS OF SILICONE CONTACT LENS MATERIALS FOLLOWING EO BO SOLUTION TREATMENT .................. 188
9 Overview ................................ ................................ ................................ ............... 188 Comparison of C 1s Spectra ................................ ................................ ................. 188 LIST OF REFERENCES ................................ ................................ ............................. 194 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 206
10 LIST OF TABLES Table page 1 1 Water contact angle measurements of different silicone hydrogel lenses. ......... 38 2 1 Material properties of the probe and the substrat e that contribute to estimating R and of the Hertzian contact shown in Fig. 2 19. ......................... 82 3 1 Details of the three ma jor SH contact lenses with their respective treatment for improving hydrophilicity ................................ ................................ ................. 98 3 2 Surface elemental compositions of the silicone hydrogels investigated by XPS ................................ ................................ ................................ .................... 99 3 3 Comparison of quantitative distribution of deconvoluted carbon 1s species of lenses treated with saline and EO BO ................................ .............................. 101 4 1 Details of the four major silicone hydrogel contact lenses with their respective treatment for improving hydrophilicity ................................ ............................... 119 4 2 Quantitative comparison of the amount of EO BO eluted from different silicon e hydrogel lenses as measured by UPLC. ................................ .............. 120 4 3 Comparison of quantitative distribution of deconvoluted carbon 1 s species of lenses treated with saline and EO BO solution of the highest conc entration. ... 121 4 4 Semi quantitative comparison of the degree of adsorption of EO BO within the upper 10 nm of the different silicone hydrogel lens types.. ......................... 124 4 5 Average concentration of EO BO measured by UPLC and XPS ...................... 125 6 1 Hydrogel synthesis recipe with the final concentration of each ingredient ........ 159
11 LIST OF FIGURES Figure page 1 1 Schematic of interaction potential V between two particles as a function of separation distance r ................................ ................................ ......................... 33 1 2 Schematic of a diffused electric double layer at a charged surface as described by the Gouy Chapman model. ................................ ........................... 34 1 3 Conformational sta tes of grafted chains predicted by the scaling laws. .............. 35 1 4 Structure and i nterfaces in ocular environment ................................ ................. 36 1 5 The mol ecular structure of tris (trimethylsiloxy) silylpropylmethacrylate (TRIS). ................................ ................................ ................................ ................ 37 2 1 Schematic of the X ray excitation of a 1s core electron. ................................ ..... 63 2 2 Photoelectrons generated as a result of d ifferent scattering processes. ............ 64 2 3 Principle of angle resolved XPS (AR XPS) ................................ ........................ 65 2 4 Sample emission profile of carbon taken at various TOAs. ................................ 66 2 5 Example general survey spectrum of a silicone hydrogel contact lens m aterial balafilcon A ................................ ................................ ........................... 67 2 6 Curve fitting of a sample C 1 s spectrum ................................ ............................. 68 2 7 The components of a typical commercial AFM ................................ ................... 69 2 8 A close up schematic of the optical detection system. ................................ ....... 70 2 9 Laser spot displacement on the photodetector corresponds to the nature and magnitude of the interfacial force experienced by the probe tip. ........................ 71 2 10 Principle of alternating contact (AC) mode AFM in probing the local mechanical properties of an inhomogeneous sample surface ............................ 72 2 11 Schematic of the Lennard Jones potential showing the general trend of the dependence of interaction energy on distance ................................ ................... 73 2 12 Anatomy of a typical force distance plot ................................ ............................. 74 2 13 Long range electrostatic repulsion observed in DI water between a silica colloidal probe and a SiO 2 /Si substrate, both with negati ve surface charges induced by plasma treatment. ................................ ................................ ............. 75
12 2 14 Force distance curves exemplifying a 5 surface of a hydrogel material in saline ................................ .............................. 76 2 15 Examples of LFM images of a small chunk of MoS 2 sliding against a MoS 2 (0001) substrate ................................ ................................ ................................ 77 2 16 Friction loop obtained at a given norm al load ................................ ..................... 78 2 17 g Assembly of the hydrogel sample holder for nanomechanical analyses in the AFM ................................ ................................ ................................ .............. 79 2 18 SEM images of a canti lever modified with a 5 .......................... 80 2 19 Application of Hertz contact theory to colloidal probe AFM indentation experiments ................................ ................................ ................................ ........ 81 2 20 Fitting the Hertz model to a set of indentation data obtained on a contact lens material in saline ................................ ................................ ................................ 83 3 1 An example of the deconvolution of C 1 s spectrum obtained from a senofilcon A lens ................................ ................................ .............................. 100 3 2 XPS spectra of C 1 s of the three types of lenses ................................ ............. 102 3 3 Carbon 1 s spectrum of a solid standard of EO BO. ................................ .......... 103 3 4 hydrogel lenses ................................ ................................ ................................ 104 3 5 A repr esentative set of friction load measurements for the three types of hydrogel samples examined in saline ................................ ............................... 105 3 6 Friction between the hydrogel sample and a silica microprobe measured in salin e and EO BO solution ................................ ................................ ............... 106 4 1 C 1 s spectra of the neat and the treated sample for each lens type ................. 122 5 1 A representative s et of indentation data of delefilcon A ................................ .... 135 5 2 Comparison between three types of silicone hydrogel lenses reveals the effect of surface structure on the deformation mechanism and the magnit ude of elastic modulus under compression of a colloidal probe. ............................. 136 5 3 Schematic of the surface gel layer and the underlying substrate ...................... 137 6 1 Schematic of the different types of hydrogel films synthesized in the present study and their respective water content. ................................ ......................... 157 6 2 A schematic of the microscopic phase tran sition at the LCST of PNIPAM. ...... 158
13 6 3 Molecular representation of the monomer unit of each hydrogel studied. ........ 160 6 4 Assem bly of a liquid sample holder with heating capability .............................. 161 6 5 Force distance curves for three compositions of PAAM hydrogels measured i n DI water ................................ ................................ ................................ ........ 162 6 6 Log log plot of the average elastic modulus vs. polymer concentration of the PAAM samples. ................................ ................................ ................................ 163 6 7 Representative force curves for P(HEMA co MAA) (45% 5%), 6 5% PHEMA, and 85% PHEMA measured in DI water ................................ ........................... 164 6 8 Force curves of 12% PNIPAM sample below (e.g. RT) and above (e.g. 37 C) the LCST in DI water. ................................ ................................ .................. 165 6 9 Interaction forces close to the surface of PNIPAMA ................................ ......... 166 6 10 Three random sites were sequentially sampled on the surface of PNIPAM below and above its LCST. ................................ ................................ ............... 167 6 11 AC mode AFM images of the PNIPAM sample at T < LCST ............................ 168 6 12 AC mode AFM images of the PNIPAM sample at T > LCST ............................ 169 6 13 g PEG brush layer with a thickness of L as modeled by the scaling laws ................................ ................. 170 6 14 Typical force distance curve measured on a hydrogel sample ......................... 171 A 1 Calculating normal spring constant using the thermal method ......................... 187 B 1 Spectra comparisons of balafilcon A ................................ ................................ 190 B 2 Spectra comparisons of senofilcon A ................................ ............................... 191 B 3 Spectra comparisons of lotrafilcon B ................................ ................................ 192 B 4 Spectra comparisons of comfilcon A ................................ ................................ 193
14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy UNDERSTANDING THE SURFACE CHEMICAL AND MECHANICAL PROPERTIES OF HYDROGEL MATERIALS FOR CONTACT LENS APPLICATIONS By Yuchen Huo August 2013 Chair: Scott S. Perry Major: Materials Science and Engineering The surface c hemical properties of commercially available silicone lenses were analyzed with respect to monitoring changes following adsorbed block copolymer molecules of ethylene oxide co butylene oxide (EO BO). X ray photoelectron spectroscopic (XPS) data confirmed the presence of physisorbed EO BO on all lens surfaces following solution treatment, and the extent of adsorption greatly differed between the lenses. Friction measurements with atomic force microscope (AFM) employing a colloidal probe in aqueous environm ent corroborated the adsorption of EO BO by demonstrating the reduction of the friction coefficient following EO BO treatment. In a separate study, based on the bulk elution of EO BO from the lenses and the complementary XPS adsorption data, an EO BO mole cular concentration gradient was evidenced for each lens type. These results suggest that the overall interaction between EO BO and a lens material depends upon both the surface and bulk composition, and the tribological behavior of the polymer surface ca n be altered by chemical modifications. The surface elastic moduli of three different silicone hydrogel lenses were examined with colloidal probe AFM. Fitting of the force indentation plots of the lenses
15 to a Hertzian contact model revealed the disparit y in both the magnitude of modulus and the mechanism of deformation as a result of the different surface chemical treatments. gel exemplifies the potential of molec ular level design to achieve depth dependent, tunable surface mechanical properties. Furthermore, to gai n a more fundamental perspective hydrogels of various chemistry and water content were fabrica ted into films and analyzed using AFM. For a given gel composition, the magnitude of surface modulus varied as much as one order of magnitude for a water content difference of 15%. The dependence of modulus on water content is slightly lower at the surface than in the bulk as predicted by scaling laws. For poly( N isopropylacrylamide) (PNIPAM), one order of magnitude of increase in surface elastic modulus was observed when solvated chains collapsed following a characteristic phase transition. These findings underscored the importance of the molecular structu re and dynamic interaction with surrounding medium in predicting elastic modulus of hydrogel materials at surface.
16 CHAPTER 1 INTRODUCTION The field of surface science, a study of surface chemical and physical processes, offers fundamental understanding f or interfacial phenomena such as friction, wear, corrosion, and chemical catalysis in a myriad of material systems [1, 2] The intimate contact between the surface and its surrounding environment renders these interfacial phenomena crucial for the overall performance of the materials. The study of surface science polymeric biomaterials. The basis of such a claim is two fold. First, the molecular details of surfaces directly influence properties such as bio compatibility, chemical reactivity, and interfacial contact mechanics. Second, the surfaces of organic materials are often dynamic in nature, resulting in chemical composition and molecular structure distinctly different from those of the bulk. Such char acteristic s can lead to a myriad of opportunities for material design and modifications tailored to a variety of applications. The body of this doctoral work focuses on understanding the correlation between surface chemical and mechanical properties of hy drogel materials, as well as applying this correlation to novel design and surface modification of soft contact lenses. To begin a detailed discussion of the surface science of hydrogels, it is important to first define what a surface is. The definition o f surface has indeed evolved over the years due to the advancement of various analytical techniques as well as the growing scientific interest in a broad range of materials to which these tools may be applied. For example, it is appropriate to consider th e surface of a single crystal to be comprised of only the top few atomic layers (up to several nanometers) as they directly participate in chemical reactions  However, such a definition is likely to limit the discussion of
17 organic surfaces. This is b ecause the molecular structures responsible for surface below the topmost atomic layer. These different length scales reflect the extent of interfacial activities relevant to various materials and determine the specific tools that may be used for fundamental investigations. Interfacial phenomena occur as a result of atoms (and molecules) at the surface seeking to minimize the total free energy of the system by interacting with their surrounding environment. As a result, the central theme of surface science research is to explore aspects of the nature and pathway of such interactions with the use of different characterization techniques. In the present chapter, effort will be made to showcase fundamental concepts in surface science, particularly with respect to those of polymeric surfaces. Such theoretical discussion will be followed by a survey of experimental characterization techniques applicable to organic materials. Finally, issues crucial to the performance of hydrogel contact lens materials will be emphasized, highlighting the opportunities offered by surface processes for advanced material design and modification. The Origin and Applications of Surface Properties S urface Interaction Forces Molecular interaction forces are the origin of surface properties observed in different material systems. The magnitude of these interaction forces is dependent upon both the chemical composition as well as the separation distanc e between surfaces. Due to the variety and complexity of surface interaction forces known, detailed consideration on this topic will focus on solid surfaces in aqueous environments, where the influence of water molecules are of interest.
18 Interaction force s between atoms under ideal conditions When two surfaces are brought together, the interaction between surface atoms (consider those that make the first contact) can be described by a generic potential energy curve as seen in Fig. 1 1. The overall behavio r of the system depends upon the following components of the potential curve  : The dependence of V the potential energy, on r the inter atomic separation distance. Because V(r) is the sum of the attractive and repulsive potential curves, their respec tive dependence on r directly influences the overall shape and r dependence of the interaction curve. The minimum point on the potential curve. The magnitude of this minimum, V min is a measure of the strength of the interaction at an equilibrium sepa ration distance r *. This is especially prevalent for colloidal systems whose stability is largely governed by the depth of this potential well relative to the thermal energy in the surroundings. The shape of the potential curve. This factor encompasses a spects of the previous two and indicates the tendency of the system to move toward a thermodynamically stable configuration under a given set of conditions. Attractive and repulsive interaction between particles (atoms and molecules) in the vapor phase ca n be expressed quantitatively by the Lennard Jones potential function (Equation 1 1), where the attractive interaction varies as a function of r 6 and the repulsive interaction varies as a function of r 12  (1 1) Th e attractive term in the function represents what is commonly known as the interaction potentials, namely the Keesom orientation, Debye inductive, and London dispersion interac tions. These potential functions share the same r dependence, namely r 6 but differ with respect to the origin of the charge interaction. Note that,
19 although aris ing from instantaneously induced dipole moment, the dispersion force is the major contribu tor to the vdW attractive potential between molecules without permanent charges or strong dipole moment, and is therefore ubiquitous in all systems. The repulsive term in Equation 1 1 is reflective of a short range, exponentially increasing energy functio n as the separation distance becomes comparable to the size(s) of the interacting particles. This phenomenon is the result of overlapping electron clouds when two atoms are brought into close proximity, as explained by the Pauli exclusion principle. Simi lar to the dispersion force, this short range repulsion is not exclusive for species examined in the vapor phase but exists between atoms and molecules of all forms. The core repulsion experienced by atoms at the surface of a probe tip in contact with a s ubstrate gives rise to the working principle of atomic force microscope (AFM). Interaction forces between surfaces in aqueous environment While the discussion of vdW attraction and short range repulsion benefits our fundamental understanding of pair wi se interaction between molecules, other interaction forces, usually long range (relative to vdW attraction and core shell repulsion) in nature, are crucial in explaining interfacial phenomena seen at surfaces immersed in aqueous environment. Many principl es described here have originated from the development of colloidal science, which provides the foundation for understanding the stability of colloidal particles in solution. In water, many surfaces bear charges due to the dissociation of chemical spe cie s or the adsorption of ions  These surface charges set up an electric field and attract oppositely charged ions (i.e. the counterions) to form a diffuse electrical double layer (EDL) according to the Gouy Chapman model (Fig. 1 2). The interaction bet ween EDLs
20 of two approaching surfaces gives rise to long range electrostatic attraction or repulsion that varies as a function of the separation between the EDLs and the ionic strength (i.e. the combined effect of concentration and valence of the counterio ns) of the electrolyte. The characteristic length away from the surface at which the electrical potential drops to 1/e of the surface potential is termed the Debye length, 1/ derived from the Poisson Boltzmann equation (Equation 1 2) (1 2) where c io is the concentration of the counterions, z is the valence of the ions in electrolyte, r is the ratio of the dielectric constant of the medium to that of the vacuum, and T is the temperature [4, 5] This equation suggests that the intera ction forces present at charged surfaces immersed in aqueous environment can be tuned by manipulating the solvent conditions. When the charges at the interacting surfaces are identical, repulsion may also arise from the osmotic pressure established by th e difference in counterion concentration in the volume between the two surfaces and that i n the bulk solution. When the two surfaces are brought closer, counterions are constrained in place and forced to maintain a concentration gradient between surfaces; as a result, osmotic pressure is established in the vicinity, contributing to the overall repulsive potential of the interface in addition to the electrostatic repulsion described earlier  Because osmotic pressure involves the relative chemical activ ities of both solute (e.g. electrolyte) and solvent (e.g. water) molecules, the observed repulsive potential illustrates the contribution of solvent molecules to the overall surface interaction force.
21 Interaction forces between polymeric surfaces in aqu eous environments In some cases, surfaces that exhibit high affinity for solvent molecules may experience yet another type of short range repulsive force termed the solvation force, or in the case of aqueous solution, the hydration force  For example, hydration forces between approaching amphiphilic bilayer surfaces have been seen to result in a sharp increase in pressure at separation distance less than a few nanometers [4, 6, 7] This repulsive interaction has a more pronounced r dependence compared to the effect of electrostatic repulsion and osmotic pressure. Pashley and Israelachvili et al. in their study of the hydration force between mica surfaces, suggest that the strong repulsive force observed in high concentration, multivalent electrolyte was a result of the additional work required to dehydrate the adsorbed ions at the contacting surfaces [8 10] Hydration force is one example of specific interactions arising from the chemical and structural nature of the contacting species at separatio n distance s comparable to the molecular dimensions of the interface. The se properties of polymer surfaces are largely dictated by the structural interactions relevant to the overall interfacial system that also include the probing counterface and the solv ent environment. The structural contribution to the surface interaction forces of polymer surfaces is perhaps best illustrated by the example of a polymer brush system grafted onto a solid substrate. From a fundamental perspective, such discussion will understanding of critical concepts such as polymer chain dimension, collective chain interaction, and intermolecular interactions with the solvent. From a phenomenological perspective, brush systems have been extensively studied due to their tunable
22 properties important to processes such as aqueous based lubrication, protein resistance in biomedical devices, and colloidal stability [11 15] Scaling laws based on Flory the size and shape, o f a grafted polymer coil is dependent upon its interaction parameter with the solvent as well as the grafting density on the substrate. In his exhaustive study, de Gennes outlined the possible conformational states assumed by grafted chains in a good solv ent at low grafting density (Fig. 1 3 A ) and high grafting density (Fig. 1 3 B ), corresponding to the unstretched and the stretched conformation, respectively [16, 17] The transition between the unstreched and stretched conformation is achieved when the ch ain length becomes longer than the distance a good approximation for predicting the structural dimension and concentration profile of grafted polymer brush layers, and pr ovide the theoretical foundation for elucidating the contribution of structure to the mechanical and tribological properties of polymer bearing surfaces. Since the development of the surface force apparatus (SFA) and later, the AFM, interfacial propertie s of polymer brush layers have been widely studied at the molecular scale, emphasizing the roles of chemical composition, molecular architecture, and solvent interaction in reducing friction at sliding contacts. It has been demonstrated that for neutral, hydrophilic brushes such as poly(ethylene glycol) (PEG), a stretched conformational state introduces steric repulsion (excluded volume effect) and osmotic pressure (lower solvent concentration) between chains upon compression, resulting in lowered frictio n coefficient when slid against a glass colloidal probe in AFM [11, 13]
23 Similarly, frictional forces are reduced at the interface of charged polymer brushes when the swollen chain conformation provides resistance to mutual interpenetration of the brushes In this case, the effect of hydration force must also be accounted for due to the water molecules tightly bound to the ionized brushes  Even though the surface structure of a hydrogel film in aqueous solution differs greatly from that of a grafted polymer brush layer, an attempt will be made in the subsequent chapters to demonstrate the similarity between the mechanical behavior of subjected to compressive stress i n AFM. Organic Surface Modifications The discussion thus far has highlighted the behavior of polymer chains under the influence of molecular design (chain length and grafting density) and solution environment (repulsion due to osmotic pressure and solvatio n force) when the brush bearing surfaces are subjected to compression and shearing. Another aspect of the dynamic nature of polymer surfaces is the rotation of molecular segments in response to interfacial environment in the absence of mechanical stress. The behavior of poly(2 hydroxyethyl methacrylate) (PHEMA) surfaces best illustrates this effect. It has been noted by many that the flexible chains at PHEMA surfaces expose the hydrophobic methyl group to the polymer air interface while maintaining the h ydrophilic hydroxyl group at its aqueous interface  This property is clearly tunable if the solution environment induces more favorable interaction with the methyl groups. Note that while the interfacial molecular forces described earlier dictate t he underlying interaction between individual functional groups and the surrounding environment, the overall behavior of the surface as manifested in properties such as
24 wetting and adhesion depends upon the collective response of the molecular chains. Chan ges in other environmental factors such as pH and temperature can also induce changes in chemical functionality (e.g. neutral to ionic) or thermodynamic stability (e.g. demixing of polymer and solvent) of the polymeric materials, leading to phenomena such as swelling and phase segregation [19 23] It is important to appreciate the fact that while all reactions, both at the surface and in the bulk, are driven by the minimization of total free energy, polymeric materials do so on a time scale comparable to m ost experimental timeframes, offering opportunities to investigate surface modification processes  Organic Surface Characterization Techniques The complexity and elusiveness (relative to the bulk) of the chemical and structural details of polymer surf aces renders any attempt of molecular scrutiny a demanding task. One of the underlying themes of the present work stresses the importance of instrumental control and calibration, which raises the level of confidence in handling complex data obtained in su rface experiments. It is also important to note that many measured surface properties are dependent upon the specific interfaces and testing probes employed; consequently, care must be taken to define the experimental parameters and to interpret results w ithin the appropriate framework. In this body of work, two types of characterization techniques are of particular interest, namely, X ray photoelectron spectroscopy (XPS) and atomic force microscopy (AFM). X ray photoelectron spectroscopy Ultra high vacuu m (UHV) spectroscopic technology has long been used to examine the chemical properties of inorganic surfaces with well defined structures  Examples of these techniques include XPS, secondary ion mass spectrometry (SIMS),
25 and electron energy loss spect roscopy (EELS). As suggested by its other common name, electron spectroscopy for chemical analysis (ESCA), XPS is a powerful tool in analyzing surface elemental composition and local chemical environment s within the topmost 10 nm of a given substrate. T his level of sensitivity allows XPS to be used for investigating surface modifications of inorganic surfaces following adsorption of self assembled monolayers (SAMs), vapor phase molecules, macromolecules, and biological species [24 29] Furthermore, XPS can also provide depth dependent chemical information by varying t he angle of exiting electrons with respect to the substrate  The application of XPS to organic surfaces has been instrumental in providing insight for understanding processes such as p rotein adsorption, chemical modifications, and surface wettability [30 35] More recently, XPS has proven to be an important tool in analyzing hydrogel surface chemical moieties [36, 37] Because this practice is still in its developing stage, a number o f issues, such as the preparation of a hydrated sample for analysis in UHV environment and the interpretation of the spectroscopic data, require further consideration. Consequently, discussion in the following chapters will focus on establishing appropria te procedures for sample drying and quantitative data analysis of hydrogel materials, particular those used for contact lens applications. Atomic force microscopy AFM, a scanning probe microscopy (SPM) technique, has revolutionized our understanding of sur face interaction forces at the atomic and molecular scales. Following the invention of scanning tunneling microscopy (STM), AFM was developed to expand the application of SPM to examining insulating materials in environments other than vacuum [2, 3] Whe reas STM measures the local electron density of the
26 substrate by establishing a tunneling current across the interface, AFM deduces surface topographical, mechanical, and magnetic properties from the physical interaction between a probe tip and the substra te. The nature of such interactions depend s upon the separation distance between the probe and the substrate as shown in Fig. 1 1. Different interaction regimes are achieved by operating the AFM at different modes, with contact and alternating contact (A C) being the most frequently used modes. In contact mode, information such as elasticity, friction coefficient, and topographical features may be readily obtained based on the analysis of interaction occurring in the core shell repulsive regime of the pot ential distance function. For soft materials such as hydrogels, gentle forces operating under AC mode are preferred when the topography of a large area is scrutinized, so that plastic deformation incurred by the probe tip rastering across the surface may be minimized. When examining mechanical properties of soft materials, the use of colloidal probes is commonly seen. There are two major advantages of using colloidal probes for surface force measurements  First, a spherical particle with defined dia meter, usually between 1 and 30 m, allows the contact pressure to be estimated quantitatively by using appropriate contact mechanics models. Second, colloidal probes have the potential to be chemically modified with various molecular ligands whose specific interactions with a surface m ay be of interest. Various contact mechanics theories have been applied to the study of soft materials with uniform structure through thickness are treated with either th e Hertz contact model or the elastic foundation model [22, 39 45] The Hertz model assumes
27 the contact between two non conforming, elastic bodies to be smooth and frictionless, and the applied strain does not exceed the linear regime of elasticity  When the deformation is comparable to the total thickness of the substrate, the elastic foundation theory must be employed, which assumes the substrate to be a bed of springs rather than an elastic half space  For colloidal probe AFM measurements on hydrogel materials, the Hertzian theory best describes their deformation behavior in aqueous environment because the indentation depth resulted from pushing a colloidal probe approximately 5 m in diameter is trivial compared to the total thickness of the sample, which ranges from 100 to 200 m. Surface Properties of Contact Lens Materials One important application of hydrogel materials is in the field of contact lenses. Though the concept o f a lens used in intimate contact with the cornea was first proposed by Leonardo da Vinci in the early 16 th century, it was not until almost four centuries later that the first pair of contact lenses, made of glass, was introduced to the public  Sinc e then, contact lenses have transformed the definition of an optical device to include functions beyond vision correction. This is accomplished through generations of material designs that aim to accommodate the key issues relevant for the interactions be tween the lens and the ocular environment. The latest development of lens design focuses on the use of silicone based hydrogels, which incorporate the exceptional oxygen permeability of silicone elastomers with the mature technology of soft hydrogel mate rials. While silicone hydrogel lenses have the clear advantage over conventional PHEMA based lenses in preventing corneal hypoxia, a number of interfacial challenges must be addressed in order to meet the key
28 criteria for reliable lens performance. To do so, one must first examine these criteria in light of the in vivo conditions relevant upon lens insertion. Interfacial Phenomena at Contact Lens Surfaces in vivo The portion of the anterior eye that directly interacts with a contact lens includes the corn ea, the eyelid, and the tear film. Phenomena occurring at the interfaces between these components maintain the normal physiological activities in the eye and facilitate the performance of the contact lens. Tear film stability at the corneal surface is on e of the most important indicators of a healthy eye, and the lack of such stability remains a major challenge to lens manufacturers as it is often believed to be the cause of contact lens induced dry eye symptoms (CLIDES)  At the surface of healthy corneal epithelia, the tear film is stabilized by a layer of glycocalyx, which is comprised of transmembrane mucin macromolecules and covers the microprojections at the epithelial surface ( Fig. 1 4A ) [49, 50] The affinity between the glycocalyx and the m ucins contained in the aqueous layer of the tear film facilitates the spreading and the movement of the tear film over the cornea. When a contact lens is introduced to the ocular environment, the tear film is split into pre lens (lipid mucous layer) and p ost lens films (mucous aqueous layer), hindering the normal interfacial activities that promote the wetting of corneal surface (Fig. 1 4 B ) [51 53] Furthermore, a host of biological molecules contained in the tear fluid, namely, proteins and lipids, can r eadily adsorb onto the lens surface and alter the surface properties of the lens. The excessive build up of protein is the number one cause of lens replacement  Modifications of Lens Surfaces for Improved Interfacial Properties Conventional PHEMA bas ed lenses rely on water for oxygen transport and are therefore severely limited when the lens surface dehydrates. Silicone based lenses, on
29 the other hand, provide oxygen permeability at the expense of reduced tear wettability at the surface, a result of the preferential accumulation of silicone species such as the tris (trimethylsiloxy) silylpropylmethacrylate (TRIS) shown in Fig. 1 5, at the lens air interface. Recalling the dynamic response of polymer surfaces, it is possible to modify the surface ene rgy of silicone hydrogel lenses for improved wetting and frictional properties. Two modification strategies are commonly employed. The first strategy involves the permanent chemical modification of the surface or incorporation of wetting agent in the bul k. Generally, different surface chemical modifications result in differences in surface energetics of the lenses as exemplified by water contact angle measurements in Table 1 1. Note that contact angle measurements are usually methodology dependent and m surfaces ; therefore, the values presented here reflect the trend displayed by different lens types The second strategy takes into account the periodic application of solution treatments containing surfa ce active agents, which are practiced as a part of a regular lens care routine  The present work focuses on the second strategy, with an emphasis on understanding the effect of an aqueous based block copolymer surfactant on the extent of chemical modi fications of a variety of silicone hydrogel lenses, as well as the resulting implications on the tribological properties of these lenses, examined herein with colloidal probe AFM. F rom a surface chemical and mechanical perspective, a number of properties c an be readily evaluated in vitro and provide sufficient insight into the in vivo interfacial behavior of the lens materials. These include, though are not limited to, wettability, lubricity, modulus, biological deposition, and active ingredient uptake fro m lens care
30 solution  In the present study, surface uptake from lens care solution, microscopic friction response, and surface elastic modulus of commercial hydrogel lenses were evaluated in vitro in order to identify those factors that can potentiall y contribute to the performance of lenses in vivo Scope of the Present Study The content of this doctoral work consists of two parts that together showcase the most up to date understanding of the chemical and mechanical properties of hydrogel materials, especially those that are prevalent in contact lens applicatio ns. The first part, including C hapters 3 and 4, focuses on investigating the uptake of a block copolymer by the most common types of silicone hydrogel lens materials. The samples were first tr eated with solutions containing the copolymer and then dried in vacuum for XPS experiments. Quantitative XPS data analysis is described in great detail to emphasize the complexity of delineating the surface chemical structures of a carbonaceous species ad sorbing onto a carbonaceous substrate. The block copolymer, ethylene oxide co butylene oxide (EO BO), is an aqueous based amphiphilic surfactant that acts to reduce the surface energy of the substrate upon adsorption. It has been shown that EO BO demons trates greater surface activity than its counterparts, ethylene oxide co propylene oxide (EO PO), the basis for surfactants such as Pluronic  In order to determined the applicability of EO BO as a potential surface active agent in lens care solution the effect of adsorbed EO BO molecules on surface properties of lens materials must be examined. In Chapter 3, a correlation is made between the extent of surface adsorption and the reduction of friction coefficient of the lens surfaces when sliding aga inst a silica colloidal probe measured with AFM. Due to the permanent chemical treatments applied to different
31 lenses for circumventing surface hydrophobicity of the silicone components in the hydrogel, disparity in the chemical affinity for EO BO molecul es was reflected in the different friction responses observed. The results in Chapter 3 suggest the important role of permanent chemical treatments in promoting lubricity at the lens surface; however, the possibility of EO BO absorbing into the bulk of the porous lens could not be precluded. Chapter 4 offers the perspective gained from a complementary bulk technique, ultra high performance liquid chromatography (UPLC), which has been used to measure the amount of EO BO eluted from the bulk lens following s imilar solution treatments addressed in Chapter 3. Comparison between the XPS adsorption and UPLC absorption data depict that different models exist for the uptake behavior of EO BO by different lenses, highlighting the influence of both surface and bulk compositions in the interaction with the copolymer surfactant molecules. The second part of this doctoral work reports on the interrogation of surface elastic modulus of hydrogel materials, both commercial lenses and home made samples. General topics in C hapters 5 and 6 include the discussion of the methodology of using colloidal probe AFM to indent soft materials and the application of appropriate contact mechanics model for analyzing the surface elastic modulus. The differentiation of surface modulus from bulk modulus and the implications of surface modulus on the overall performance of the lens are also described in detail. Chapter 5 presents the analysis of delefilcon A, a daily disposable silicone hydrogel lens, whose composition profile varies wit h thickness, presenting a unique gel layer at the surface. A comparison
32 with other silicone hydrogel lenses highlights the effect of surface structure on the nature of deformation and the magnitude of modulus characteristic to lens materials. To explore t he fundamental relationship between experimentally controlled parameters and the measured surface modulus, a series of hydrogels with varying equilibrium water content, ranging from 15 w/w% to 95 w/w%, was synthesized and their surface moduli were measured as discussed in Chapter 6. Specifically, the surface modulus was examined as a function of water content in hydrogels of the same composition and cross linker concentration. Furthermore, an environmentally responsive hydrogel, poly( N isopropylacrylamid e) (PNIPAM), that undergoes a thermodynamic phase transition was analyzed above and below its lower critical solution temperature. As the transition occurs, changes in molecular structure of the hydrogel surface due to the altered interaction with the sol vent is believed to result in the observed changes in surface topography, adhesion force, and elastic modulus. Together, these results illustrate the influence of chemical composition and molecular structure on the surface properties of hydrogel materials. These properties represent major aspects of the key criteria that must be satisfied for acceptable usage of contact lenses in vivo The perceived correlation between surface chemical compositions ( and the modifications thereof) and the mechanical behavi or of contact lens materials offers tremendous insight into the design of functional polymeric materials, whose surface properties dominate the overall performance in their intended operating environment.
33 Figure 1 1. Schematic of interaction potential V between two particles as a function of separation distance r V min reflects the stability of the system at an equilibrium separation distance r*. The exponents 6 and 12 denote the distance dependence of the interaction potential as modeled by the Le nnard Jones equation (Equation 1 1).
34 Figure 1 2. Schematic of a diffused electric double layer at a charged surface as described by the Gouy Chapman model. The electric potential resulting from a distribution of counterions follows an exponential de cay (dotted line) as distance from the surface, z, increases. Adapted from 
35 A B Figure 1 3. Conformational states of grafted chains predicted by the scaling laws A) L oosely packed chains with R F < D where R F represe nts the Flory radius of the chain B) Densely packed chains with R F > D. R F and D represent the Flory radius of the random coil in a good solvent and the distance between graft points, respectively. In the stretched conformation, overlapping between chai ns leads to molecular steric repulsion that is often seen in brush systems, and is believed to give rise to a low friction surface when the polymer solvent interaction is optimized. Adapted from 
36 Figure 1 4. Structure and interfaces in ocular environment. A) The cross sectional schematic of the corneal epithelial tissue, highlighting the components and interfaces relevant to the promotion of tear stability over the corneal surface. B) Upon lens insertion, the native structure of tear film is disrupted, introducing additional interfaces at which the wetting of both the cornea and the lens must be accommodated by lens surface design. A B
37 Figure 1 5. The molecular structure of tris (trimethylsiloxy) silylpropylmethacrylate (TRIS).
38 Tab le 1 1. Water contact angle measurements (mean standard deviation) of different silicone hydrogel lenses. Adapted from . Method Lens Type Sessile Drop Captive Bubble Acuvue Oasys 85.0 9.0 32.4 5.5 O 2 Optix 37.2 8.3 44.3 6.9 PureVision 101.6 6.3 30.1 5.8
39 CHAPTER 2 INSTRUMENTAL METHODS AND EXPERIMENTAL APPROACH In this chapter, various surface chemical and mechanical analytical techniques employed in the present studies are introduced. Following the discussion o f the fundamental principles of these techniques, specific experimental protocols designed for handling and measuring soft hydrogel materials are presented. X ray photoelectron spectroscopy (XPS), an ultra high vacuum (UHV) technique, has long been used to investigate the surface chemical compositions of vacuum compatible substrates. Due to its exceptional surface sensitivity, XPS has also been proven useful in probing the chemical properties of adsorbates species on inorganic materials [ 5 7 5 9 ] The experiments involved in this body of work utilize XPS to analyze solutions treatments containing surface active agents. Drying preparation and charge neutralization procedures are carefully deve loped to ensure consistency in the quantitative analysis. Atomic force microscopy (AFM) is a widely used technique in probing interfacial forces between a small probe tip (at most microns in radius) and the substrate. The versatility of AFM allows mater ials to be examined in various environments, including ambient, solution, and vacuum. In this body of work, various hydrogel materials are examined with colloid modified probe tips in aqueous based solutions, with an emphasis placed on understanding the m echanical properties of these materials.
40 X ray Photoelectron Spectroscopy Fundamentals of XPS photoelectric effect, which relates the kinetic energy ( E KE ) of a given exiting electron to the binding energy ( E BE ) of the electron and the energy of the incident photon ( ) impinging upon the surface (2 1) Here, spectro denotes the work function of a given spectrometer, defined as the energy required to remove an electron from the highest occupied state in a solid to the vacuum level (Fig 2 1). A beam of incident X ray photons with fixed energy (e.g. 1486.7 eV from a monochromatic Al source) penetrates deeply into a specimen, effectively ejecting core level electrons of the atoms within the sample. However, only the photoelectrons in the surface zone that have elastically scattered (primary electrons) contribute to the photoemission peaks, while those that have suffered energy loss (secondary electrons) contribute to the emission background  It is important to note that only the pho toelectrons in the surface zone have sufficient proximity to the surface to escape without inelastic scattering and be detected; dictated by the depth of this zone, which is approximately 10 nm. Photoelectrons at larger depths are accommodated by the bulk during scattering (Fig 2 2). Understanding the process of photoelectron generation offers greater insight into the relationship between measured photoelectron intensity and the depth f rom which it originate s as this information helps better explain the principles of depth dependent chemical analysis with XPS. As mentioned above, primary electrons from the s urface
41 region contribute to the overall intensity of the spectrum collected. The intensity of these el ectrons dampens as the electrons travel through the material and escape from the surface. Similar to the Beer a medium, the measured intensity, I follows a first order exponential decay as a functi on of the distance between the origin of the electrons and the surface of the sample, denoted by d Moreover, such decaying process also depends upon a material and energy specific parameter, the inelastic mean free path (IMFP), defined e distance that an electron with a given energy travels between  These quantities are related by the following equation, assuming the incident beam is normal to the surface of the sample  : (2 2) where I 0 represents the incident beam intensity before it interacts with the sample calculation [ 60 ] For photoelectrons with kinetic energy in the range of 1.5 to 3 ke V, IMFP (in nm) is effectively proportional to E photon 1/2 (in nm) and atomic diameter (in nm), where E photon is the incident beam energy in eV [ 6 1 ] Values of IMPFs of carbon, oxygen, and silicon employed in these studies have been calculated based on dat a provided in reference [ 6 2 ] and are used in deriving the AR emission profiles of carbon in Fig 2 4. Sampling depth, a measure of the surface sensitivity of XPS, is defined as 3 or the distance from which 95% of the detected primary electrons originate s  One practical way of enhancing the surface sensitivity of XPS is to position the incident beam at a grazing angle with respect to the surface, defined as the take off an gle (TOA) (Fig 2 3). With this slight variation, Equation 2 2 becomes
42 (2 3) Note that the distance traveled by a primary electron from a specific depth d is calculated to be As increases, this effective path length  decreases such that only those electrons located immediately beneath the surface may be detected while others are likely to be attenuated while exiting the sample. This principle provides the basis for angle resolved XPS (AR XPS) analysis by which one may obtain the depth dependent chemical information simply by varying the relative position of the sample to the incident beam. Angle depended emission profiles are calculated according to Equation 2 3 and shown in Fig 2 4 for carbon. Unless otherwise note d, The emitted photoelectrons are analyzed by a set of concentric hemispherical plates between which a potential difference is established to only allow electrons with a specific kinetic energy to pass through. As the hemispherical analyzer scans through a range of energ ies all of the photoelectrons are accounted for and the emission intensity is measured as a function of the kinetic energy of the electrons. With the knowledge of the incident photon energy and the kinetic energy measured for the emitted photoelectrons, it is straightforward to determine the binding energy of the electrons according to Equation 2 1. Because each element possesses a characteristic binding energy for electrons a t a given core shell, chemical identification with XPS is accomplished when one compares the measured photoemission peaks with the documented elemental binding energies. Further quantitative analysis can yield information regarding the relative abundance of each element and of each bonding
43 state within a given element. Detailed procedures for spectra analysis will be described in the following sections. The spectrometer employed in the present study is a commercial system (Omicron NanoTechnology, Taunus stein, Germany) with a monochromatic Al source (1486.7 eV) and a hemispherical analyzer EIS Sphera (Omicron NanoTechnology, Taunusstein, Germany). The work function of this spectrometer is 4.5 eV, while the constant pass energy to which all of the photo electrons are retarded before entering the analyzer is 20 eV. Core level scans are collected with a step size of 0.05 eV. Analyzing Hydrogel Samples with XPS Hydrogel samples involved in the present study are stored in aqueous environment s either in a fresh buffered saline or in a testing solution containing a surfactant to be investigated. Contact lens samples are first soaked in buffered saline to remove residual components of packing solution before conducting any surface analysis. This is because packing solutions often contain surface active agents that are low in surface tension [ 6 3 ] Contact angle measurements on silicone hydrogel lens surfaces are consistent with the removal of these surfactants following extensive soaking of the lenses in sa line solution [ 56 ]. To prepare these samples for analysis in an ultra high vacuum environment, proper drying procedure s are mandatory. In general, hydrogel samples are handled minimally to preserve surface cleanliness prior to XPS analysis; however, thos e that have been previously treated with a surfactant in solution are gently dried with tissue to remove excess solution. Such practice prevents unbound surfactant molecules from condensing onto the surface in vacuo Each hydrogel sample is first mounte d on a UHV compatible platen and dried in a load lock with 10 7
44 Torr vacuum for about 12 hours before conducting XPS measurement, which takes place in 2 x 10 10 Torr vacuum environment. As in the case for all insulating materials, the emission of photoe lectrons inevitably leaves the surface positively charged, and hinders more electrons from elastically exiting the surface, i.e., artificially lowers the kinetic energy of these electrons. As a result, insulating surfaces are usually flooded with low ener gy electrons during XPS analysis to prevent charging  However, many have suggested that electrons alone may also lead to the over compensation of the surface, which leads to the opposite effect of charging. Such over compensation may be corrected by introducing low energy ions, commonly Ar + to the surface  The combination of low energy electrons and ions has also shown to improve the energy resolution of the emission peaks, which offers a greater degree of certainty in peak identification and fitting precision. In the present study, the electron beam is set to have an energy of 1.5 eV and an emission current of 5 A, while the argon ion beam is accelerated by an voltage of 20 V at 10 mA of current. Quantitative Chemical Analysis with XPS Quantitative surface chemical information is obtained by analyzing the photoelectron emission intensity profile collected as a f unction of the kinetic energy of the electrons. Conventionally, the kinetic energy is converted to binding energy by the Einstein equation (Equation 2 1) for ease of i dentifying chemical elements. The a nalytical approach for determining chemical composit ion and bonding states includes elemental peak identification from general survey spectrum and peak fitting of the core level scan of a given element.
45 Elemental peak identification A general survey spectrum, usually collected through the entire range of the kinetic energy achievable by the spectrometer, provides information for identifying various elements present at sample surface. Survey spectra have low energy resolution but sufficient for identifying peaks at energy intervals characteristic to a give n element, which often appear as sharp, narrow vertical lines. General survey spectr a are taken at the beginning of each XPS experiment to ascertain the specific elements for which core level scans will be collected. As expected for most organic materi als, hydrogels almost always contain the following elements: carbon, oxygen, nitrogen, and sometimes with traces of sodium and/or chlorine from processing and handling. Silicon is also found in silicone based hydrogel materials. It is important to note t hat, because of the similarity in elemental composition of organic materials, one must rely on detailed analysis of bonding states to fully grasp the chemical signature of each surface. This is especially prevalent when studying surface modification of hy drogel materials by adsorbate molecules that are also hydrocarbon based. As a result, all of the quantitative analyses presented in this study are based on core level spectra of carbon 1 s (C1 s ), oxygen 1 s (O1 s ), nitrogen 1 s (N1 s ), and silicon 2 p (Si2 p ). Shown in Fig 2 5 is a sample survey spectrum of a silicone hydrogel contact lens material, balafilcon A. Peak fitting protocols Each core level spectrum is collected in the range of binding energy vicinal to the peak energy expected for the given element capturing each peak in its entirety. One example of a C 1 s spectrum is given below (Fig 2 6 A ) Note the presence of background emission that ari se s from the inelastically scattered photoelectrons. Such
46 background must be mathematically subtracted in order to accurately calculate the composition of the surface. Though much discussion has been centered on the algorithms by which the subtraction exercise is done in the spectroscopy community, the main method utilized in the data processing software in the present study (Casa Software Ltd.) involves the Shirley background subtraction method. A review of the details of this method can be found elsewhere [66 68] Peak assignment is the next step to completing quantitative chemical analysis of XPS data. Due to the complex molecular nature of organic materials, multiple chemical bonding states are often detected for each element. The varying amounts of different bonding states give rise to the characteristic asymmetry of each elemental peak, as seen in t he C1 s spectrum shown in (Fig 2 6 B ). Multi ple peaks can be modeled to fit the elemental spectrum to represent the different bonding states. The following procedures are iterated until the desired goodness of fit, usually less than 2% uncertainty, is re ached. First, the number of synthetic peaks is estimated according to the asymmetry of the elemental peak. Then, the peaks are assigned in terms of their relative binding energy with the aid of the knowledge of the sample chemistry or of a reference spec trum acquired from a pure standard. Thirdly, the shape of each peak is restrained by a fixed full width at half max (FWHM) value and a modeled Gaussian (90% contribution) and Lorentzian (10% contribution) distribution, which defines the tail shape of each peak. The goodness of fit is calculated as a normalized difference between the sum of the synthetic peaks and the original spectrum, and is optimized computationally through a series of fitting iteration s A set of fitting constraints is empirically rea lized for each element in a given sample and
47 applied thereafter in order to maintain consistency between experiments. An example of the fitted C1 s spectrum is shown in (Fig 2 6 B ). Following peak assignment, the relative abundance of different bonding sta tes of a given element is determined according to the normalized area under each synthetic peak, while the elemental composition of the sample is determined based on the sum of correcting the peak area with the atomic sensitivity factor (ASF) of a given element. transmission function, which takes into account factors such as X ray flux, photoelectri c cross section for a given atomic orbital, efficiency of photoelectric process, and the mean free p ath of the photoelectrons, etc.  In the present study, the ASF of each element analyzed has been normalized to the empirical value of fluorine. Atomi c Force Microscopy Fundamentals of AFM Atomic force microscopy (AFM), unlike XPS, probes the sample surface with a tip of defined geometry as it physically interacts with the surface and experiences forces of attraction or repulsion as a function of tip s ample separation. Depending upon the origin(s) of the inter facial forces experienced mechanical, chemical, and tribological properties of a given surface may be analyzed using AFM. A ssembly of an AFM A typical atomic force microscope consists of a set of piezoelectrically driven actuators, a cantilever holder, a laser source, and an optical detection system that includes a recollimation lens, a reflective mirror, and a segmented photodetector (Fig 2 7).
48 Depending upon the specifics of the system, tip sample separation is controlled by either moving the tip (and the light source) assembly toward the fixed sample stage or by moving the sample toward the tip, both utilizing piezoelectric actuators. The instrument used in the present study, an MFP 3D BIO system (Asylum Research, Santa Barbara, CA), employs a z piezoelectric actuator that drives the tip assembly with a full range of motion up to 16 m. A positive applied voltage to the piezoelectric actuator drives the tip toward the surface (approach), a nd a negative applied voltage pulls the tip away from the surface (retract). The planar movement of the sample in x and y directions is also made possible with a piezoelectric stage. The x and y actuators assume a stacked configuration and are capable The most common type of AFM probe is an integrated sharp tip on a silicon or silicon nitride cantilever made by microfabrication process es The probe tip typically has a radius of curvature in the range between 1 and 20 nm [ 2] al though ultra sharp and blunt tips are desirable in some experimental practices. The cantilever is characterized by its geometry, V shaped or rectangular, and by its coating, usually reflective to improve the optical detection of the signal. The geo metry of cantilever is of great importance in the determination of force constants of the lever. Details on this topic will be discussed in a later section. Principles of AFM AFM relies on the sensitivity and precision of the optical detection system to monitor intermolecular forces experienced by the probe tip. The system works as such: a super luminescent diode generates a laser beam that is focused onto the back of the cantilever. The laser then passes through a recollimation lens and is reflected b y a mirror onto the position sensitive photodiode (Fig 2 8). The photodetector is
49 segmented into four equal sections and the intensity of the signal detected by each quadrant gives rise to a voltage reading. Converting this voltage reading to a force me asure requires a cantilever and instrumental specific factor and will be discussed in a later section. The motion of the cantilever induces the displacement of the laser spot on the photodiode, which in turn causes a change in the voltage reading of the different sectors While the total intensity of the signal detected by the quadrants remains constant, depending upon the position of the spot with respect to the center of the diode, either a normal or a lateral signal is generated, as depicted in Fig 2 9. A normal signal is indicative of force that causes the cantilever to bend (Equation 2 4), (2 4) while a lateral signal is indicative of force that causes the cantilever to twist (Equation 2 5 ), (2 5) Note t hat cross talk between the normal and lateral channel is common in signal detection. As a result, careful decoupling algorithms are necessary to minimize errors and to obtain an accurate measure of the tip sample interaction. The alignment of the laser spot onto the back of the cantilever is achieved by monitoring its physical position with an inverted optical microscope (a component of the MFP 3D system) and the total intensity on the photodetector simultaneously. The position of the laser spot on the cantilever can potent ially affect the accuracy of measured force This is because the design of the AFM assembly ensures that the slight movement of the cantilever is amplified and captured by a detectable displacement of the laser spot on the photodetect or. Limpoco et al  ha ve reported
50 the marked difference in measured friction response of a Si 3 N 4 tip sliding on a SiO 2 /Si surface when the laser spot is aligned at different positions relative to the long axis of a V shaped cantilever. Furthermore, c alibration methods that rely on the measured properties of the cantilever (e.g. the thermal noise method, as described below) are also sensitive to the position of the laser spot on the cantilever  As a result, the position of the laser spot on each cantilever is noted carefully at the time of initial force calibration, such that the same position is kept for the subsequent experiments to ensure data consistency. Quantitative Analysis with AFM The fundamental equation that governs the operation of AFM relates the interfacial force F to the tip sample separation d by a power exponent x whose value depends upon the nature of the interfacial force: (2 6) Therefore, by operating under either a constant force or a constant height feed back mechanism, aspects of the interfacial force between the tip and the sample can be analyzed. Two of the most typical modes of operation in AFM are contact and alternating contact (AC) mode. When operating in contact mode, the tip is physically brou ght into contact with the surface until a desired force is reached. If the feedback mechanism is switched on, the force experienced by the probe tip is kept constant and the applied z piezo voltage is adjusted according to the height changes of the surfac e as commanded by the feedback loop. If the feedback is switched off, measurements such as indentation and friction can be made.
51 In AC mode, the probe tip is driven by a shaker piezoelectric actuator at a frequency near resonance frequenc y in a given medium (Fig 2 10). When feedback is switched on, the probe tip is maintained at constant dampened amplitude of oscillation as a result of its interaction with the surface T he sample height is then measured similar to the mechanism in conta ct mode. The energy dissipated by the probe tip in interacting with the surface is measured by a phase shift between the measured frequency of the tip after tapping the surface and its driving frequency. The change in the degree of phase shift from one l ocation on the surface to another is constantly surfaces are protected from damages ind uced by rastering a stiff tip across the surface. As a result, this mode is most commonly used in imaging soft materials. In the present study, emphasi s has be en placed on quantitatively measuring the mechanical properties of soft hydrogel mat erials in aqueous environment. Corresponding AC mode imaging of these materials aid s in understanding the treatment. Force distance plots In this type of measurement, the probe i s brought into contact with the surface with the feedback turned off. A positive voltage is applied to the z piezo, causing it to extend toward the sample surface until a triggered deflection value is reached, at which point the z piezo reverses direction and pull away from the sample. The force experienced at the tip is recorded as a function of the tip sample separation (see Equation 2 6). Different components of a force distance curve can be analyzed to
52 understand phenomena such as surface adhesion, l ong and short range interaction, indentation, viscoleastic hysteresis, etc. (Fig. 2 12). In the non contact regimen of the plot, the probe starts at a distance far away from the sample, free of the influence from any possible long range interaction with the surface. Though no force is acting on the cantilever in this regime, the deflection reading is typically non zero, accounting for the offset in the voltage reading on the photodetector as a result of hydrodynamic drifting of the cantilever  The correction for the offset in this regime is integral to the determination of the point of contact. Depending upon the nature of the interaction between the probe and the surface, d the sample are both neutral in charge, the interaction energy can be modeled by the Lennard Jones potential (Fig 2 11). As the separation decreases between the two surfaces, the potential energy of the pair reaches a minimum. This negative potential i s reflective of force at the interface, the tip jumps into contact with the surfac e [72, 73] ( Fig 2 12 ) For curves as such, the point of contact is defined as the point at which the apparent force is zero Newton s following the correction of the non contact regime of the curve. If the probe and the sample have charged surfaces, espe cially in solution environment s an electrostatic interaction is observed (Fig 2 13). Counter ions present in solution are attracted by the surface charges and form an electric double layer (EDL), which effective ly neutralizes the surface  In this ca se, the tip does not jump to contact, but display a long range repulsion until the core shell repulsion occurs.
53 Screening of charged surfaces is prevalent when the interaction between a charged polymeric surface and a blunt tip (e.g. a colloidal probe) wi th surface charges occurs in physiological saline solution. In the contact regime of the approach curve, the Pauli exclusion principle describes the repulsive interaction having an r 12 dependence on separation, assuming no sample deformation has occurre d (Fig 2 12). If deformation is present, which is likely when performing force distance plots on low modulus materials, contact mechanism s such as the Hertzian theory can better explain the behavior in this regime (Fig 2 14). For an incompre ssible int erface, the retract portion of the curve traces that of the approach (in the linear regime of the piezo displacement ). When probing a viscoelastic material such as a hydrogel, the recovery from tip deformation results in a hysteresis bet ween the approach and retract curve, indicative of energy dissipated during deformation As the tip continues to retract from the surface, it experiences an 2 12). In air, the main source of adhesion between many materials is the adsorbed the tip and the surface  As tip continues to retract, the capillary force is eventually overcome and the tip pulls off. For polymeric substrates in solution intermolecular adhesive force s between the tip and the dangling polymer chains at surface can cause multiple pull off events, creating a stepwise retract curve that spreads over a large distance while the tip pulls away from the surface (Fig 2 14) [71, 74, 75] Friction load maps Lateral force microscopy (LFM) is one major application of contact mode AFM. For the commercial instrument employed in the present study, LFM is accomplished by
54 obtaining a lateral force image with varying applied normal force. When collecting a typical contact mode normal force image, the tip is kept at a constant deflection value (i.e. in constant force feedback mode) and programmed to raster across the surface in a defined area. The x and y axes of the image define a planar view of the sample surface while the contrast of the image reflects the change in the height (z axis variations) of the surface features. In LFM, the tip starts scanning horizontally (x axis) at a distance above the surface and gradually moves toward the surface (z axis). This motion is effectively the same as ramping up the applied normal force. Upon contacting the surface, the tip continues to indent while rastering across the surface until it travels the defined scan distance in z, at which point the tip gradually pulls away from the surface and the normal load decreases as a result. During scanning, signals in the normal and lateral channels are collected simultaneously. Furthermore, images from both scan directions, trace (left to right) and retrace (right to left), in ea ch channel are also obtained. The resulti ng images are exemplified in Fig 2 15. The contrast in the normal force images corresponds to the magnitude of the load applied, with the brighter color indicating greater cantilever deflection. The contrast in t direction as a function of applied normal load. Notice that the contrast in the trace and retrace images from the normal force channel are the same, but opposite for th e lateral channel. This is expected because friction force always opposes the direction of motion. Friction response (in V) at a given normal load can then be determined by taking half of the difference between the trace and the retrace scans of the lat eral signal and
55 averag ing over the displacement along x axis excluding the measurements in the static friction regimes [69, 76] This analysis is more easily realized by plotting the lateral axi 2 16). Each loop corresponds to a specific normal load, which can be read from either the trace or retrace image of the normal signal. There are two components characteristic to each loo p: static friction at initial contact and turn around point in scan direction, and kinetic friction in between these events As normal load increases, separation between the trace and retrace portion of the loop increases, indicating that the magnitude of the friction force is proportional to the applied normal load. The proportionality of this relationship, the coefficient of friction is determined by plotting the calibrated friction response (in N) against the applied normal loads. Force calibration of cantilevers Quantitative analysis with AFM requires the cantilever to be calibrated so that the force experienced at the probe ti p can be interpreted accurately. In order to convert a voltage reading on the photodetector to a meaningful force measure ment two parameters are needed. First, the optical lever sensitivity (OLS) of the cantilever must be determined in a given instrumen t as the OLS is both a cantilever and instrument specific quantity. This factor relates the voltage change due to the interfacial force at the tip to the displacement of the cantilever in the form of either bending or twisting. OLS is typically expresse d in V/nm. The normal OLS is simply the slope of the contact portion of a force distance plot generated when pushing the tip on a rigid surface. One straightforward way of measuring the lateral OLS is to calculate the slope of the static portion of a fri ction loop collected by sliding the tip across a rigid surface 
56 The second parameter required is the spring constant of the cantilever. The determination of normal and lateral spring constant of cantilevers has been a major topic of discussion in quantitative AFM methodology because it involves extensive knowledge of the geometry, composition, and processing of the cantilever. Furthermore, derived from the beam theory, Equation 2 7 highlights the sensitivity of spring constant to errors in measur  (2 7 ) The relationship between force and the normal spring constant k N of a cantilever (2 8 ) which states that the applied fo rce F is proportional to the deformation of the cantilever by a factor k N et al. and the thermal noise method are commonly employed by many. The Sader method requires the knowledge of the plan view geometry of the cantilever, the resonance frequency of the cantilever, and the quality factor of the surrounding medium, because the hydrodynamic function is geometry dependent [79, 80] The thermal noise method models the cantilever as a simple harmonic oscillator in equilibrium with its surroundings that fluctuates in response to thermal energy Briefly, this theory suggests that the spring cons tant k N of the oscillating cantilever scales with the thermal energy by its root mean square displacement < q 2 > of oscillation. For both of these methods, the hydrodynamic parameters of the cantilever oscillating free from the influence of surface forces are required. The plan view
57 dimensions of the cantilever must be measured with SEM, which increases the level of difficulty in the quick to do, the thermal noise method introduces a gre ater degree of uncertainty because the environmental factors have a large contribution to the calculated k N In addition, other methods based on indirect measurement of k N and finite element analysis (FEA) are also relevant as discussed by Burnham et al.  The lateral spring constant k L of a cantilever can also be determined by a number of methods. For compliant rectangular cantilevers, the torsional Sader method is applicable, provided that the resonance frequency in the lateral channel can be dete cted  In other cases, methods measuring the lateral force sensitivity (in N/V) are typical. The knowledge of provides direct conversion from voltage signal to lateral et al The approach of this method takes into account the geometric contribution to the lateral load from sliding the tip across a surface with facets of defined angles. By varying the applied normal load, the lateral load is monitored and a coefficient of friction at the interface can be determined. Finally, can be related to the original lateral signal output by the friction coefficient and the relative angles of the facets  Other direct and indirect calibration methods for k L are reviewed by Petter s son et al.  In Appendix A details of two major calibration methods are discussed highlighting their applicability to tips employed in the present study.
58 Chara cterizing Hydrogel Materials with Colloidal AFM When characterizing hydrogel materials with AFM, additional experimental considerations with regard to probe geometry, testing environment, and quantitative analysis methods are integral to the understanding of their surface properties. The MFP 3D commercial AFM employed in the present study is equipped with a specialized liquid sample holder (Fig. 2 1 7 ) The sample holder ensures that both the probe and the sample are fully submerged in testing solution t hroughout the duration of the experiment. It is also equipped with a heating element and sensor such that the temperature of the cell, i.e. the sample and its surrounding medium, can be controlled in the desired fashion. For measurements on contact lens materials, the temperature of the cell is kept at ambient (approximately 25C) while for those made on temperature sensitive hydrogels such as poly( N isopropylacrylamide) (PNIPAM), cell temperature may be cycled below and above critical tran sition point. For aqueous based solution s 30 minutes are allotted for the system to reach thermal e quilibrium before proceeding with experiments at a fixed temperature Contact pressure consideration When performing force distance plot on a flat sheet of hydrogel material using a traditional AFM probe tip, the Hertzian contact theory can be used to predict the contact pressure at the interface by the following equation: (2 9 ) where P max is the maximum pressure resulted from an applied normal force F N The
59 (Equation 2 1 0 ), and 1/ R is a reflection of the geometry of the contact area (Equation 2 1 1 ). (2 1 0 ) (2 1 1 ) For a Si tip (bulk elastic modulus ~ 150 GPa  ) with radius of curvature of 20 nm applying 50 nN of force on a hydrogel s ubstrate (bulk elastic modulus ~ 500 kPa), the estimated contact pressure is 1.8 MPa 0.3 for both materials Such high contact pressure poses two challenges in understanding surface mechanical properties of hydrogel materials, which in general have an E value ranging digit MPa. First, large contact pressure can potentially damage the substrate, disrupting the lo cal network structure. Second because the contact pressure at ocular interfaces during movement is less than 10 kPa  contact pressure in the range of MPa clearly does not demonstrate significant ph ysiological relevance in the investigation of contact lens hydrogel materials. To circumvent the high contact pressure caused by probe geometry (note the R 1 /2 dependence), many have considered the use of spherical colloidal probe s for measuring surface mechanical properties of soft materials and biological tissues [42, 85 88] Typical colloidal probes have radii of curvature on the order of a few microns. If the radius of curvature in the previous example increases to 2 m, then the estimated contact p m (in diameter) silica (SiO 2 ) microsphere is glued to the end of a standard Si cantilever and employed as the probe for measuring the indentation of the hydrogel substrate. A sample SEM image of
60 a m odified cantilever is shown in (Fig 2 1 8 ). The calibrated force constants of a modified cantilever are based on the values of the cantilever without the added colloid. The Sader method for calibrating normal spring constant is not affected by an added m ass whose size is in the micron range  Because the wedge method for calibrating the lateral force sensitivity cannot accommodate probe sizes much larger than the separation between ridges of the facets (approximately 50 nm), one must consider calibra ting the lateral constant of the cantilever prior to performing probe modification. The cleanliness of the microsphere is carefully maintained by rinsing the cantilever with 1.0 M HCl, followed by successive rinses with ethanol and deionized water. The cantilever is then treated with oxygen plasma in a plasma cleaning instrument (Harrick PDC 32G plasma cleaner/sterilizer, Ithaca, NY). The oxygen plasma is intended to remove any hydrocarbon molecules tightly bo und to the surface of the probe; such approa ch has been widely used in preparing glass substrates for adsorption studies [11, 89, 90] This cleaning routine is carried out following each experiment to remove possible material transfer incurred during contact with the polymeric surfaces. Indentat ion measurements on hydrogel materials Defined as the linear proportionality between applied stress and strain, elastic modulus E is a bulk property that is characteristic to a given material. The determination of bulk elastic modulus assumes that the mat erial is homogeneous in composition and structure. For materials that have undergone specific surface chemical treatments, it is reasonable to expect that the elastic modulus at the surface may differ from that of the bulk, as in the case of some silicone hydrogel contact lens materials. Because AFM affords the opportunity to make measurements on the nanometers and nanoNewtons
61 scale, the surface elastic modulus of a hydrogel material can be analyzed and such analysis is derived from a set of indentation d ata obtained from the substrate. Specifically, force distance plots are collected in the default fashion as described earlier with the approach speed of the z the fluid within the network structure ha s sufficient time to squeeze out of the pores, minimizing the effect of fluid support. The penetration depth of the probe into the sample surface is calculated as the difference between the vertical displacement of the z piezo and the deflection of the ca ntilever (Fig 2 19 A ). Multiple force curves are collected at each location and multiple locations are sampled on a given surface to ensure that the calculated surface elastic modulus is representative of the sample. Accurate for most of the samples exa mined in the present study, the curvature in the data collected in the contact portion of the force indentation plot can be explained by a Hertzian contact mechanism, from which the elastic modulus of the surface region can be empirically determined. Hert zian theory predicts that the depth of indentation fro m pushing an infinitely rigid body on a soft substrate is a function of the applied load F N the dimension of each body, as well as the elastic property of the contacting surfaces  : (2 1 2 ) Experimentally, is measured as the difference between the z piezo displacement and the normal deflection of the cantilever from pushing on the substrate (Fig 2 19 B ). Table 2 1 summarizes the material properties of the probe and the hydrogel substrate, which contribute to defining a nd R based on Equations 2 1 0 and 2 1 1
62 The following conditions must be satisfied for a given contact to be considered Hertzian in nature: The surfaces are continuous and non conforming ( i.e. the contact half width a << R ); The contacting surfaces a re considered smooth and frictionless; The strains are small; Each solid can be considered an elastic half space. Care must be taken in ensuring that strains applied to the low modulus hydrogel samples do not exceed the limit defined by the linear theory of elasticity. This is achieved by closely monitoring the depth of indentation and keeping it at a value much less than the radius of the probe. Once the zero point of force and zero point of displacement are determined for a given force indentation pl ot, the surface elastic modulus is determined by fitting the Hertzian model (Equation 2 1 2 ) to the raw data using a least square s method. Essentially, with known F N , and R can be estimated by minimizing the error between the model and raw data (Fig 2 2 0 ).
63 Figure 2 1. Schematic of the X ray excitation of a 1s core electron. Adapted from 
64 Figure 2 2. Photoelectrons generated as a result of different scattering processes. Electrons in the selvedge region with sufficient energy cont ribute to the photoemission peak, while others in the region give rise to the background intensity after each suffering a single inelastic scattering event. Those electrons in the bulk undergo multiple inelastic scattering events and are accommodated by t he bulk material Adapted from 
65 Figure 2 3. Principle of angle resolved XPS (AR XPS). By varying the TOA, the distance traveled by an electron at a given depth is increased, effectively increasing the surface sensitivity of detection AR XPS allows one to estimate the depth from which the majority of the spectral information originates, offering the opportunity for chemical depth profiling. Adapted from  A B
66 Figure 2 4. Sample emission profile of carbon taken at various TOAs.
67 Figure 2 5. Example general survey spectrum of a silicone hydrogel contact lens material balafilcon A. Note the large background intensity that must be accommodated mathematically before calculating composition. Binding Energy (eV) O 1 s C 1 s N 1 s Si 2 p Background intensit y O 2 s
68 Figure 2 6. Curv e fitting of a sample C 1 s spectrum. A) An unfitted C 1 s spectrum showing the characteristic asymmetry due to multiple bonding states present. B) The same spectrum is fitted with four different sub peaks, with defined separation in binding energy. The goodness of fit, which is based on the normalized difference between the original data (red curve) and the sum of all sub peaks (black curve), governs the number of iterations to be performed for the fitting exercise. A B
69 Figure 2 7. The components of a ty pical commercial AFM. Note that the cantilever holder and the optical detection system are housed together and driven in the vertical direction with the z piezo actuator.
70 Figure 2 8. A close up schematic of the optical detection system.
71 A B C D E Figure 2 9. Laser spot displacement on the photodetector corresponds to the nature and magnitude of the interfacial force experienced by the probe tip. A) A cantilever f ree from any surface interaction. B) Upward bending as a result of repulsive normal force. C) D ownward bending indicat ive of attractive normal force. D) and E) Cantilever twisting caused by friction fo rce at the interface, leading to the horizontal displac ement of the laser spot
72 Figu re 2 10. Principle of alternating contact (AC) mode AFM in probing the local mechanical properties of an inhomogeneous sample surface. During AC mode scanning, multiple channels of data acquisition are available, including phase shift (shown here), heigh t, and amplitude. Adapted from 
73 Figure 2 11. Schematic of the Lennard Jones potential showing the general trend of the dependence of interaction energy on distance. Note that the energy E is normalized against the well depth at the potential mi nimum and the separation distance r is normalized against the distance at which E is zero.
74 Figure 2 12. Anatomy of a typical force distance plot. For neutral surfaces, the long range repulsion may not be observed. Adhesion force during retracti on can be accentuated by the use of a blunt tip or colloidal probe. Adapted from 
75 Figure 2 13. Long range electrostatic repulsion observed in DI water between a silica colloidal probe and a SiO 2 /Si substrate, both with negative surface charges i nduced by plasma treatment. The effect of the electrical double layer (EDL) is screened by co ions when the contacting surfaces are submerged in buffered saline.
76 Figure 2 14. Force distance curves exemplifying a 5 m silica probe indenting on the surface of a hydrogel material in saline. With both the approach and retract curves plotted, hysteresis between the two curves and multiple pull off events upon retract are observed.
77 D isplacement Displacement A B Displacement Displacement C D Figure 2 15. Examples of LFM images of a small chunk of MoS 2 sliding against a MoS 2 (0001) substrate A) Trace image of the normal force. B) Retrace image of the normal force. C) Trace image of the lateral force. D) Retrace image of the lateral force Note that the contrast proportional to the magnitude of force, is identical in normal force images but opposite in the lateral images, reflecting th e nature of friction force always opposing to the applied motion. The background contrast represents the non contact portion of the images. Images obtained courtesy of Dr. Xueying Zhao Normal Signal Lateral Signal
78 Figure 2 16. Friction loop obtained at a given normal load ( sou rce data are indicated by the horizontal lines in C) and D) of Fig. 2 15) As the normal load increases, V increases by a proportion that is reflective of the friction coefficient of the substrate.
79 Figure 2 1 7 Assembly of the hydrogel sample holder for nanomechanical analyses in the AFM
80 Figure 2 1 8 SEM image s of a cantilever mod ified with a 5 m silica probe. 50 20
81 Figure 2 19 Application of Hertz contact theory to colloidal probe AFM indentation experiments. A) Calculation of the depth of penetration from indenting a rigid probe on a soft hydrogel material. B) Enlarged det ails regarding the Hertzian contact are shown below. Note that the calculation of R and are obtained from values in Table 2 1. A B
82 Table 2 1. Material properties of the probe and the substrate that contribute to estimating R and of the Hertzian cont act shown in Fig. 2 19 The elastic modulus of silica is obtained from  and that of a typical hydrogel contact lens is based on manufacturer reported values. SiO 2 Probe Hydrogel Surface R 5 m E 73 GPa 500 kPa (average) ~ 0.2 ~ 0.45
83 Figure 2 2 0 Fitting the Hertz model to a set of indentation data obtained on a contact lens material in saline. Visual differences are observed when the estimated modulus varies by 300 kPa and t he goodness of fit is apparent for = 2.1 MPa.
84 CHAPTER 3 IMPACT OF ETHYLENE OXIDE BLOCK BUTYLENE OXIDE (EO BO) COPOLYMERS ON THE COMPOSITION AND FRICTION OF SILICONE HYDROGEL SURFACE S Background and Motivation In the ten years since the introduction o f silicone hydrogel contact lens  this special class of lens materials has evolved to provide exceptional oxygen permeability. Unlike conventional P HEMA based hydrogels, which rely on water for oxygen transport [94, 95] silicone hydrogels achieve hig h oxygen permeability through the incorporation of modified siloxane monomers  This characteristic has allowed these materials to be used for overnight extended wear with significantly reduced occurrences of corneal hypoxia  Surface properties performance in vivo and silicone hydrogels here again differ from conventional P HEMA material. Most of the interactions between cornea, contact lens, and tear film dictating wear comfort involve the interfacial phenomena of wetting and friction. It is well known that the hydrophobic siloxane moieties in silicone hydrogel materials are not confined to film [98, 99] Low w ettability and high friction cause tear film instability leading to contact lens induced dry eye symptoms [97, 100] As a result, many advances in silicone hydrogel lens technology have focused on improving the hydrophilicity of lens surface by various me ans of chemical treatment. Silicone hydrogel lenses can be categorized according to the nature of their hydrophilic treatments. The first group of materials, represented by PureVision (balafilcon A), Focus Night & Day (lotrafilcon A), and O 2 OPTIX (l otrafilcon B), undergo surface chemical treatment during manufacturing. Balafilcon A, for instance, relies on plasma oxidation of the surface for enhanced wettability [93, 98, 99] The
85 lotrafilcon lenses, on the other hand, have a thin, uniform coating d eposited across the surface through plasma induced polymerization  A second group of materials includes ACUVUE OASYS ( senofilcon A) and ACUVUE ADVANCE (galyfilcon A) which both incorporate poly(vinyl pyrrolidone), or PVP, in their lens chemistr y, resulting in greater similarities between the bulk and the surface compositions [99, 102] A third group of silicone hydrogel lens materials employs an entirely different silicone chemistry, entailing neither an internal wetting agent nor surface plasm a treatment, to achieve the desired combination of high oxygen permeability and hydrophilicity  A number of previously reported studies have sought to understand the in vitro surface properties of conventional and silicone hydrogel materials in ter ms of their surface chemistry and related interfacial properties. For example, contact angle measurements have been reported in several studies and have shown that the advancing angles of hydrogel lenses depend on their exposure environment and the measur ement methodology [103 105] In general, these studies have shown that contact lens hydrogels have surfaces that are not usually at thermodynamic equilibrium in their use environments and, as a result, there can be significant changes in the surface chemi stry over time due to the migration of hydrophobic groups to the air/lens interface. To overcome these dynamic changes, the surfaces have been modified the soaking of lenses in saline solutions containing block copolymers to physically modify the surfaces [93, 98, 99, 101, 102, 105] Recent studies have characterized more closely the surface chemistry of contact lens hydrogels using state of the art techniques such as X ray photoelectron
86 spectroscopy (XPS) [36, 37, 61] time of flight secondary ion mass spectrometry (ToF SIMS)  and sum frequency generation (SFG) [107, 108] The exceptional sensitivity of these techniques enables the analysis of only the topmost mol ecular layers of the surface region. Consequently, chemical information gathered at this level of surface sensitivity can offer fundamental insight into the wetting behavior, tear component deposition, and friction response of lenses, all of which are beli eved to be surface chemistry dependent. In the past, atomic force microscopy (AFM) has proven useful in studying the surface morphology of silicone hydrogel lenses on a submicron scale [93, 102, 109] With appropriate modifications, the AFM can also be u sed to detect frictional forces acting between a probe tip and lens surface under environments simulating in vivo wear conditions. This approach has been applied to the microscopic measurement of friction coefficients for conventional P HEMA hydrogel lense s [107, 110, 111] To date, molecular level measurements of silicone hydrogel lens friction properties have not been reported. In the present study, the surface chemical compositions of three types of silicone hydrogel contact lenses have been assessed t hrough the use of XPS while the surface morphology and frictional properties have been probed with AFM. These measurements were conducted prior to and following treatment in solutions of a diblock copolymer of poly(ethylene oxide) and poly(butylene oxide) (EO block BO, or EO BO for short). Block copolymers are often incorporated in lens care solutions as anti foaming agents. Their adsorption and resulting tribological behavior on lens surfaces can directly influence the extent of the reconditioning of the lens through solution treatment. The aforementioned surface techniques were utilized to understand the
87 compositional and frictional alteration of lens surfaces by EO BO copolymer adsorption, highlighting improvements in lens surface properties via soluti on treatment. Together, the results demonstrate close correlations between the surface chemistry and frictional response of the material systems, as well as the opportunity for solution based modification of silicone hydrogel lenses. Experimental Mater ials Three types of silicone hydrogel lenses were investigated: balafilcon A (PureVision) by Bausch & Lomb, senofilcon A (ACUVUE OASYS ) by VISTAKON of Johnson & Johnson, and lotrafilcon B (O 2 OPTIX) by CIBA VISION . The published composition and hydro philic surface treatment of each lens are detailed in Table 3 1 Prior to analysis or solution treatment, all lenses were soaked in a standard buffered saline solution (Unisol4, Alcon, Fort Worth, TX) for at least 24 hours Unisol4 is a borate buffered saline solution with a pH of 7.4 and is generally used to rinse lenses. Subsequent solution treatments involved soaking the lenses in solutions of 0.1 mg/mL of a diblock copolymer, provided by Alcon Laboratories, Inc., in Unisol4 at room temperature for at least 24 hours. The diblock copolymer consisted of, on average, 45 units of ethylene oxide and 10 units of butylene oxide per chain. EO BO surfactants possess better interfacial and wetting properties than the structurally equivalent ethylene oxide b lock propylene oxide (EO PO) surfactants  Surface Chemical Composition Analyzed by XPS The protocols for preparing hydrogel materials for XPS measurements were developed based upon previous reports on the analysis of organic materials with UHV techniq ues [36, 37, 61, 106] Drying the sample in 10 7 Torr vacuum before transferring
88 it to UHV environment maintained the cleanliness of the sample surface. Core level XPS scans of the oxygen 1 s carbon 1 s nitrogen 1 s silicon 2 p and sodium 1 s regions were acquired under basic instrumental settings, i.e. with a pass energy of 20 eV and a step size of 0.05 eV. Dual beam charge neutralization scheme, as discussed in Chapter 2, was employed. The spectroscopic data were processed using CasaXPS software (Casa So ftware Ltd.). Background subtraction was performed using a sequence of built in mathematic treatments of the software. Sub peaks within each core level spectrum were assigned, fitted, and analyzed with defined constraints according to the procedures afore mentioned. When necessary, spectra were energy shifted such that the core level peak associated with the C H/C C/C Si state appeared at a binding energy of 284.5 eV. The cited degree of uncertainty with respect to composition is attributed to the mathema tical uncertainty of the peak fitting results, and such error was incorporated into each composition calculation. Friction Response and Surface Topography Probed by AFM All friction measurements were conducted at room temperature with a customized AFM uti lizing a commercial control and software system (RHK Technology, Troy, MI). The apex of a lens, entailing an area defined by a diameter of approximately 0.5 cm, was prepared for each measurement and supported on a polymeric base curve. A liquid cell enco mpassed both tip and sample in the liquid of interest throughout the duration of the measurement. Fri ctional forces were measured for the sliding contact of a 5 probe, attached to a microfabricated silicon nitride triangular cantilever, with the hydrogel surface. Typical lateral scan dimensions were 500 nm with applied normal
89 loads ran ging from 0 to 20 nN. Though the maximum force applied was identical on different lens surfaces, one must take note that due to the vastly different elastic modulus values characteristic to these lenses (see reported values in Chapter 5) large disparity is expected in terms of contact pressures applied at the interface while scanning ( see Equation 2 9 ) To explore the dependence of friction on applied load, frictional forces were measured during the lateral reciprocation of the probe over the surface, pr oducing multiple friction loops for a series of load settings. A friction load plot was then generated based on the value of the lateral force at a given normal load, producing a linear relationship whose slope is taken to be the microscopic coefficient o f friction. Calibration of the specific cantilever used yielded a normal spring constant of 0.12 N/m and a lateral spring constant of 3.2 N/m. Friction measurements were acquired by Alexander Rudy. Surface topography of the lenses was probed using a com mercial stand alone AFM (Asylum Research, Santa Barbara, CA). As in lateral force measurements, a specialized liquid sample holder was employed for all measurements. Alternating contact mode (AC) AFM imaging was conducted and images were thereby generate d by plotting deviations of the free, driven amplitude as a function of x y position across the surface. Data were subsequently converted to reflect topographical changes using the IGOR Pro software (WaveMetrics, Portland, OR). The resulting images were processed using SPIP software (Image Metrology A/S, Lyngby, Denmark). Results Surface Composition Following a 24 hour soak in saline and vacuum drying for 12 hours, quantitative elemental compositions of the surfaces of these lenses were derived from integ rated
90 XPS intensities and are presented in Table 3 2 Multiple lens samples as well as multiple areas on the lens surface were evaluated. Although slight variations were detected, no deviations were significant enough that the trends presented in Table 3 2 could not be considered representative. As anticipated, carbon, oxygen, nitrogen, and silicon were detected at the surface of each of the lenses. Trace amount of sodium appeared on several samples, possibly as a result of contamination during sample p reparation and/or the manufacturing process. As these elements comprise all of the surface constituents, the surface composition is reported in terms of percent atomic concentration. The greatest variations in surface composition among the three types of lens materials occur in the N and Si concentrations. As the sampling depth of XPS under the employed conditions is limited to approximately 7 nanometers within the outermost surface [2, 60 112] the measured compositions reflect the species present in t his region following vacuum drying. The measured values represent a quantitative basis for following changes in the surface composition as a function of solution treatment. Further analysis of the C 1 s core level spectra provided greater insight into the nature of chemical species found within the near surface region. A representative C 1 s spectrum obtained from a senofilcon A sample is displayed in Fig. 3 1 The contribution of several different moieties to the spectrum is reflected in the presence of u nderlying peaks as well as peak asymmetry. In XPS data, shifts in binding energy for a given element are understood to arise from atoms existing in different bonding environment. Following background subtraction, these data have been fit to four discrete species by methods described in Chapter 2. The identification of each chemical shift (as labeled in
91 Fig. 3 1 ) is based upon literature reported peak positions of known chemical compositions [36, 37] because XPS is not a molecularly specific technique, some chemical functionalities cannot be fully distinguished. Based on the analysis of the three types of lens materials, the data reported in Table 3 3 reflect the variation in chemical moieties present at the surfaces of these lenses. Whereas the percen t atomic concentration data portray similar amounts of carbon in different materials, the detailed core level analysis reflects relatively fewer quantity of N C=O species present at the surface of the senofilcon A lens with respect to the other two. And a lthough the intensities were low, both the balafilcon A and senofilcon A lenses exhibited O=C O species in the surface region, while little evidence of these was found for the lotrafilcon B lens. Following the characterization of the lenses treated with sa line, additional lens samples were subjected to treatment in solutions containing EO BO diblock copolymers. Using similar preparation and analysis methods, the three types of materials were investigated with respect to potential modification of their surf ace chemical compositions as a result of exposure to these solutions. In clinical settings as well as in common patient care practice, multipurpose solutions comprised of cleaning agents, disinfectants, and/or comfort enhancing agents are used to treat so iled lenses between uses. The intent of the present study was to evaluate the degree of adsorptive uptake of the diblock copolymer EO BO onto the surfaces of the silicon e hydrogels. Here, the less hydrophilic butylene oxide block would be viewed as the m olecular attachment point to the hydrogel surface, thus presenting the more hydrophilic ethylene oxide block to the solution environment.
92 In general, percent atomic concentrations derived from the XPS evaluation of the treated lenses reflected compositiona l modifications of the lens surfaces. These modifications entailed increases in carbon and oxygen intensity observed simultaneously to decreases in silicon and nitrogen intensity ( Table 3 2 ). A greater understanding of the degree of surface modification is reached upon closer inspection of the C 1 s core level spectra acquired for each type of lens. For reference, Fig. 3 2 A displays the spectra of the three lenses before EO BO solution treatment. Fig. 3 2B depicts the chemical changes occurring upon solu tion treatment. Each spectrum has been deconvoluted by the procedures described above, with the results summarized in Table 3 3 XPS analysis of an EO BO standard (Fig. 3 3) identified peak intensity at 284.5 eV (C H/C C/C Si) and 286 eV (C O/C N), consi stent with the peaks whose intensity has changed in the spectra of Fig. 3 2 B While the elemental analysis reflects only modest changes, a result of analyzing the adsorption of a hydrocarbon polymer onto a predominantly hydrocarbon surface, the detailed d ata of Fig. 3 2 B and Table 3 3 demonstrate significant differences in the manner in which the different materials are modified. As primarily indicated through intensity at 286 eV, the greatest degree of adsorption of EO BO copolymer is observed on the lo trafilcon B and balafilcon A samples. Only modest changes are observed for the senofilcon A surface. For each of the treated lenses, decreases in C H/C C/C Si, N C=O, and O C=O integrated intensities following solution treatment are consistent with these signals being attenuated by the presence of the overlaying polymer adsorbates. The relative presence of the copolymer in the near surface region can be realized by using a mathematical model that relates the measured total intensity of such
93 species to the escape distance of photoelectrons of a given elemental species [32, 60 112] This model suggests that the measured intensity of the photoelectrons decreases exponentially when the escape distance increases. As the copolymer adsorbate species was clearl y observed on balafilcon A and lotrafilcon B lenses, one can ascertain the relative EO BO concentrations within the selvedge region. Specifically, the concentration of EO BO within the near surface region of lotrafilcon B is observed to be approximately t wice that found for balafilcon A, following identical treatments. AFM Analysis The surface topography of the silicon hydrogels was probed by AC mode AFM using a silicon nitride probe tip possessing a radius of curvature of < 30 nm. As shown in Fig. 3 4 A the surface of balafilcon A featured micron sized circular depressions approximately 50 nm in depth and narrow trenches running in random directions across the surface. Both senofilcon A (Fig. 3 4B ) and lotrafilcon B (Fig. 3 4C ) albeit exhibiting a de gree of surface texture, appeared significantly more homogeneous than balafilcon A. Although potentially related to the surface treatments aimed at improving hydrophilicity, none of the surface features were observed to influence the friction measurements described below. The tribological properties of the hydrogels were examined on a microscopic scale by sliding a 5 liquid. Frictional forces between the silica probe and the lens surface were m easured as a function of applied load as well as treatment of the surfaces with the diblock copolymer test solution. A plot of the frictional force versus normal load in turn reflects the frictional character of an interface. The slope of a given dataset is related to the microscopic friction coefficient, with higher slope values corresponding to higher friction
94 interfaces. Fig. 3 5 presents a comparison of the frictional properties of the three types of lenses tested in saline solution. When sliding i n contact with a silica microprobe, the balafilcon A and lotrafilcon B surfaces clearly exhibited higher friction values than that observed for senofilcon A. This observation was consistently repeated for multiple samples of each type of material. While the exact molecular origin of the differences in friction response is not evident, these differences can be attributed to the compositional differences documented by the XPS measurements of the present study. Following measurement of the inherent frictiona l properties of the hydrogel surfaces, solution exchanges were performed, introducing the 0.1 mg/mL EO BO diblock copolymer test solution. Microscopic friction measurements were repeated in the presence of the copolymer solution, the results of which are displayed in Fig. 3 6 For each type of lens, a reduction in friction is observed with the introduction of the copolymer; however, the extent of modification was lens dependent. Balafilcon A (Fig. 3 6A ) and lotrafilcon B (Fig. 3 6C ) exhibiting the high est coefficients of friction prior to the treatment, experienced the most significant reduction in friction upon the introduction of EO BO The senofilcon A (Fig. 3 6B ) sample, already exhibiting low friction when in contact with the silica probe prior to treatment, saw only a minor reduction in friction in the pres ence of the EO BO test solut ion. Discussion Although the three lens systems investigated all belong to the class of silicone hydrogels, distinct differences were observed in both the composition al and frictional analyses. In terms of composition, differences between the lenses were observed with respect to the significant variations in Si and N concentrations as well the deconvolution of the C 1 s spectra, which reflected the differences in the d istribution of chemical
95 functionalities present at the surface. Together, these changes are consistent with their respective formulations and surface treatments. For example, the higher concentration of Si at the surface of senofilcon A is consistent wi th the absence of the plasma surface treatment presenting an oxidized or carbon rich top layer, as in the case of both balafilcon A and lotrafilcon B [101, 102, 113] Upon solution treatment with the EO BO diblock copolymer, further changes in compositio behavior were observed. These are understood to arise from the strong physisorption of the copolymer at the lens surface although formal chemical bonds are not formed, the strength of this attractive interaction is su fficient to prevent the removal of the copolymer by gentle rinsing prior to vacuum drying. As noted in the discussion of Fig. 3 2 a greater attenuation of the C H/C C/C Si intensity at 284.5 eV and a greater increase in C O/C N intensity at 286 eV was obs erved for the case of balafilcon A and lotrafilcon B than for senofilcon A. These results are consistent with a lesser presence of the EO BO copolymer at the surface of senofilcon A in comparison with the other two lenses. This finding has two possible origins. First, given that balafilcon A and lotrafilcon B undergo plasma treatments by the manufacturer in an effort to increase hydrophilicity, a greater concentration of hydrophilic sites may exist at the surface for the adsorption of the copolymer; how ever, the evidence for this origin is not strongly reflected in the analysis of surface functionality. Specifically, the data of Table 3 3 do not exhibit significant differences in intensity for C O/C N, N C=O, and O C=O species of the three lenses soaked in saline. Second, a lesser extent of EO BO at the surface of senofilcon A may result from a greater internal free volume of this lens  In this
96 way, EO BO may be absorbed into the lens as opposed to adsorbed onto the surface as in the case of the two plasma treated lenses. In terms of friction, the lenses were clearly differentiated by their load dependent friction response ( Fig. 3 5 ). For the case of lenses soaked in saline, senofilcon A exhibited a markedly lower friction than both balafilcon A and lotrafilcon B. The plasma treatments aimed at improving the hydrophilicity of these latter two lenses may very well result in species that produce a greater frictional response when in contact with a silica probe. Additionally, the internal wetting agent of the senofilcon A lens may contribute to the lower surface friction, though this has not been clearly indicated in the compositional analysis. Upon treatment with the diblock copolymer, the extent of friction reduction is seen to qualitatively cor relate with the relative amount of copolymer adsorption on the surface as detected by XPS. A significantly greater reduction in friction is observed for balafilcon A and lotrafilcon B on which a greater presence of EO BO was detected. In turn, the reduct ion in friction is ascribed to the introduction of highly hydrated ethylene oxide moieties to the sliding interface [11 14, 114, 115] Albeit slight, opposite changes in the intensities of C H/C C/C Si and C O/C N binding states in the senofilcon A sample were observed, indicating some presence of EO BO at this surface as well. Such evidence could explain the minute reduction in friction seen on the senofilcon A lens surface. Concluding Remarks in vivo performa nce depends on its generally attributed to the surface hydrophilicity as expressed by the polar groups on the surface. In this study, differences in the surface chemical com position of three
97 major types of silicone hydrogel contact lens materials were resolved using XPS. Friction measurements with AFM also resolved distinct differences in their behavior, with the two lens types involving a factory plasma treatment exhibiting the highest frictional response. The silicone hydrogel lenses were also treated with a test solution of an EO BO diblock copolymer. XPS data revealed that lenses having undergone previous surface treatments, i.e. balafilcon A and lotrafilcon B, experien ced significant surface chemical modification through copolymer adsorption. Furthermore, upon treatment with the diblock copolymer test solution, the reduction in friction of these two lenses was much more pronounced than in the case of senofilcon A. Tog ether, the results convey a strong correlation between the adsorption of the EO BO diblock copolymer and the reduction in interfacial friction. These findings clearly indicate the opportunity for dynamic modifications to silicone hydrogel contact lens sur faces through solution treatment and a potentially important clinical strategy for enhancing wear comfort in vivo
98 Table 3 1. Details of the three major SH contact lenses with their respective treatment for improving hydrophilicity [36, 97] Trade Name USAN Name Published Composition Hydrophilic Surface Treatment PureVision balafilcon A NVP, TPVC, NCVE, PBVC Plasma oxidation ACUVUE OASYS senofilcon A mPDMS, PVP, siloxane macromer, DMA, HEMA None (internal wetting agent) O2 OPTIX lotrafi lcon B TRIS, siloxane macromer, DMA, HEMA Uniform thin coating via plasma induced polymerization NVP: N vinyl pyrrolidone; TPVC: tris (trimethylsiloxysilyl) propylvinyl carbamate; NCVE: N carboxyvinyl ester; PBVC: poly(dimethysiloxy) di(silylbutanol) b is(vinyl carbamate); mPDMS: monofunctional poly(dimethylsiloxane); PVP: poly(vinyl pyrrolidone); TRIS: trimethylsiloxy silane; DMA: N,N dimethylacrylamide; HEMA: poly(2 hydroxyethyl methacrylate)
99 Table 3 2. Surface elemental compositions of the silicon e hydrogels investigated by XPS Lens Type Treatment Surface Elemental Composition (atomic% error) C O N Si Na* B alafilcon A Saline 62.9 1.0 21.2 0.8 7.4 0.4 7.9 0.4 0.6 0.1 EO BO 64.1 1.2 25.0 0.8 4.7 0.3 6.2 0.2 0 S en ofilcon A Saline 64.6 1.5 20.9 0.9 4.7 0.3 9.8 0.3 0 EO BO 66.3 1.0 21.1 0.7 4.0 0.3 8.7 0.3 0 L otrafilcon B Saline 65.4 1.6 20.0 0.6 12.6 0.4 0.8 0.2 1.1 0.1 EO BO 70.1 1.0 25.1 0.5 3.8 0.3 0.5 0.3 0.4 0. 1
100 Figure 3 1 An example of the deconvolution of C 1 s spectrum obtained from a senofilcon A lens. Note that all the chemical shifts were identified by characteristic peak positions as reported in the literature.
101 Table 3 3 Comparison of quantitative distribution of deconvoluted carbon 1s species of lenses treated with saline and EO BO Lens Type Treatment Carbon 1 s Species Distribution (atomic% error) C H, C C, C Si C O, C N N C=O O C=O B alafilcon A Saline 58. 3 0.9 27.4 0.4 11.9 0.2 2.4 0.1 EO BO 48.0 0.9 42.5 0.8 7.9 0.1 1.6 0.1 S enofilcon A Saline 61.3 1.4 28.9 0.7 8.0 0.2 1.8 0.1 EO BO 57.8 0.9 33.3 0.5 6.8 0.1 2.1 0.1 L otrafilcon B Saline 66.9 1.7 22.0 0.6 11 .1 0.3 0 EO BO 33.5 0.5 62.3 0.9 4.2 0.1 0
102 Figure 3 2 XPS spectra of C 1 s of the three types of lenses A) Treated in saline solution. B) Treated in EO BO test solution. The spectra contain devonvoluted pe aks indicating different chemical shifts within the C 1 s core region. The identification of each chemical shift was based on a reference XPS spectrum (not shown) of the residue of an EO BO test solution following evaporation in an inert environment. L otrafilcon B S enofilcon A B alafilcon A Arbitrary Intensity Unit Binding Energy (eV) B alafilcon A S enofilcon A L otrafilcon B Arbitrary Intensity Unit Binding Energy (eV) A B
103 Fig ure 3 3. Carbon 1 s spectrum of a solid standard of EO BO.
104 Figure 3 4 of the silicone hydrogel lenses. A) Balafilcon A. B) S enofilcon A C) L otrafilcon B. The images have been processed through light shading in order to enhance topographic features. Holes and trenches are apparent across the surface of balafilcon A, while few signature topographical features are apparent on the surface of senofilcon A. Lotrafilcon B exhibits a corrugated topography on this length scale. A C B
105 Figure 3 5 A representative set of friction load measurements for the three types of hydrogel samples examined in saline. The coefficients of friction, defined as the slope of the plot, of balafilcon A (filled circles) and lotrafilcon B (open triangles) are significantly greater than that o f senofilcon A (filed squares).
106 Figure 3 6 Friction between the hydrogel sample and a silica microprobe, before (filled) and after (open) the introduction of a EO BO copolymer test solution A) Balafilcon A. B) S enofilcon A C ) L otrafilcon B. B A
107 Figure 3 6 Continued C
108 CHAPTER 4 SURFACE AND BULK UPTAKE OF ETHYLENE OXIDE BLOCK BUTYLENE OXIDE COPOLYMER BY SILICONE HYDROGEL CONTACT LENS MATERIALS Background and Motivation In the effort to improve the surface hydrophilicity and tear wettabi lity of silicone hydrogel contact lenses, manufacturers have implemented various strategies in the design of the bulk and surface chemical compositions of the hydrogels. A r ecent ce chemistry by the adsorption of surface active agents. Past studies have analyzed changes in surface properties of hydrogel lenses following the adsorption of such surfactant molecules as hydroxypropyl methylcellulose (HPMC) and poloxamine 1107 (BASF) [ 105, 116, 117] These studies concluded that the surface adsorption of these molecules resulted in improved wetting behavior of PHEMA based lenses. Furthermore, subjects wearing the surfactant treated lenses also reported enhanced comfort, underscoring t he potential of this treatment to be used in clinical applications of conventional hydrogel lenses. In addition to HPMC and poloxamine, small amphiphilic block copolymers have long been regarded as effective surface active agents in a host of industrial an d medical applications. Categorically, this nonionic and water soluble class of molecules is especially suitable for modifying surfaces under aqueous conditions. As the reduction of friction and enhancement of comfort at the lens tear film interface is a rising concern in the contact lens materials research field, amphiphilic block copolymers containing poly(ethylene oxide) (EO) as the hydrophilic component balanced by a relatively more hydrophobic component have received much attention as functional surf ace active additives to multi purpose disinfecting solutions (MPDS).
109 Much of the literature on EO based amphiphilic block copolymers is devoted to the study of adsorption and association behaviors of poly(ethylene oxide) poly(propylene oxide) (EO PO) bas ed triblock copolymers having various molecular architectures. The Pluronic (BASF) series of block copolymers include various EO PO based copolymers, primarily used as anti foaming agents in some lens care solutions. It has been suggested, however, that the larger difference in polarity between the EO and the butylene oxide (BO) blocks in EO BO copolymers can lead to greater surface activity, i.e., more effective surface tension reduction as compared to that observed with the structurally equivalent EO P O copolymers [55, 118] Prior work within our group as reported in Chapter 2 has characterized the interfacial friction of hydrated silicone lenses in the presence of an EO BO copolymer solution using AFM  It was shown that in compari son to the neat lens samples, reduction in friction was apparent for some lenses following the exposure to this copolymer. Based on chemical modifications seen with XPS, surface adsorption of EO BO was concluded. The extent of such modifications, however, was lens dependent. Specifically, lenses having undergone surface plasma treatments, e.g., balafilcon A and lotrafilcon B, demonstrated much greater EO BO adsorption, and consequently, friction reduction, in comparison to senofilcon A, which is not surfa ce treated during manufacturing. These preliminary results led to speculation that the mechanism by which the surfactant molecules interact with different silicone hydrogel lenses is a possibility of the absorption of the EO BO copolymer into the bulk of hydrogels was not precluded, thus prompting the quantitative investigation of copolymer net uptake. In the present study,
110 the interaction mechanisms between EO BO copolymers and surface s of four types of silicone hydrogel lenses have been concurrently assessed by XPS and ultra performance liquid chromatography (UPLC), respectively. The resulting correlation between the degree of surface adsorption and a quantitative measure of the amoun t of EO BO copolymer uptake, reflective of bulk absorption further reveals a rich disparity in the nature of interaction between this surfactant and lenses of distinct compositions, potentially offering the pathways to finding clinically relevant solution treatments. Experimental Materials Four types of silicone hydrogel lenses were investigated: balafilcon A (PureVision) by Bausch & Lomb, senofilcon A (ACUVUE O ASYS ) by VISTAKON of Johnson & Johnson, lotrafilcon B (O 2 OPTIX) by CIBA VISION , and com filcon A (Biofinity) by CooperVision. The published composition and hydrophilic surface treatment of each lens are detailed in Table 4 1 which is an extended version of Table 3 1 with the addition of information on comfilcon A Standard lens sample pre paration was conducted, followed by soaking the lenses in solutions containing various EO BO concentrations at room temperature for at least 24 hours. All EO BO treated lenses were analyzed without further solution treatment (e.g. rinsing) but were gently dried with tissue to remove excess solution. Bulk U ptake of EO BO A nalyzed by UPLC Bulk uptake of EO BO was assessed using the extraction from each lens and analyzed by an UPLC system (Waters Corporation, Milford, MA, USA), which consisted of a binary solvent separation module and a charged aerosol universal detector (CAD).
111 1.7 m particle size) with a mobile phase consisting of methanol (solvent A) and 0.1 % trifluoroacetic acid /diamond water (solvent B). A semi gradient chromatographic condition was applied to the mobile phase in order to establish an acceptable baseline for the elution study. Subsequently, standards of EO BO and EO BO extract dissolved in methanol from each tr eated lens were acquired and analyzed using Empower software (Waters Corporation, Milford, MA, USA), resulting in the amount of EO BO uptake measured a s a function of retention time. The UPLC experiments were performed by support scientists in Alcon Labor atories, Forth Worth, TX. Surface C hemical C omposition A nalyzed by XPS Standard protocols for vacuum drying and analyzing the hydrogels in XPS were applied to the samples in the present study. AR XPS was conducted to examine the variation in composition a s a function of depth, particularly for determining possible molecular absorption of EO BO into the near surface region of the hydrogels. The spectroscopic data were processed following the procedures established in Chapter 2. Results Hydrogel T otal U pt ake of EO BO The degree of total EO BO uptake by each lens sample was quantified by measuring the eluted amount of EO BO with UPLC as a function of the surfactant concentration of the testing solution. As shown in Table 4 2 the four lenses displayed ma rkedly different absorption behavior at each level of EO BO concentration. The surface plasma treated balafilcon A retained the largest amount of EO BO. A similar degree of absorption was observed for comfilcon A, a lens absent of any surface hydrophilic treatment. Lotrafilcon B, with a uniform plasma induced polymeric surface treatment, resulted in the least amount of EO BO uptake at all levels of surfactant
112 concentrations, while senofilcon A showed a moderate level of uptake. In light of the known cha racteristics of each lens shown in Table 4 1 these elution data do not indicate a simple correlation between lens surface modification and bulk absorption of EO BO molecules. For example, balafilcon A and lotrafilcon B, both having received surface plasm a treatments and both having shown comparable degrees of surface modifications previously upon copolymer solution treatments (see Chapter 3) displayed contrasting bulk absorption behavior. Although the extent of uptake differed between lenses, the am ount of eluted EO BO for a given lens type was directly correlated to the concentration of EO BO in solution. This trend, consistently observed for all lenses, suggests that the absorption of EO BO molecules into the bulk of a generally porous silicone hy drogel follows a classic diffusion model driven by the concentration gradient of solute across the solution hydrogel interface. The significant variation in the degree of such driving forces seen for the different lens types may be a reflection of the dif ferences in both the surface and the bulk microstructure of the materials. Lens S urface M odification by EO BO Following a 24 hour soak in EO BO solutions (or saline for the control samples) and vacuum drying for 12 hours, quantitative elemental composition s of the lens surfaces were derived from integrated XPS intensities and are presented in Table 4 3 Multiple lens samples as well as multiple areas on the lens surface were evaluated. Although slight variations were detected between samples of the same t ype (cf. Table 3 3 ), no deviations were significant enough that the trends presented in Table 4 3 could not be considered representative. Detailed analysis of the XPS
113 spectra of the carbon 1 s core region through curve fitting exercises provided greater in sight into the nature of chemical species found within the near surface region (Table 4 3). Refer to Fig. 3 1 for an example of a deconvoluted C 1 s spectrum from a senofilcon A sample. Although the hydrogel lens material and EO BO surfactant share common chemical functionalities, evidence of surface modification by EO BO is realized, in part, through the increase in relative intensity of the peak at a characteristic binding energy of 286 eV Notably, as seen in Table 4 3, the greatest increase in intensit y of the C O binding state occurred for both balafilcon A and lotrafilcon B, indicating the adsorption of EO BO within the surface region. Senofilcon A and comfilcon A, on the other hand, each displayed a lesser degree of EO BO adsorption, indicated by th e slight change in the C O signals following solution treatment. Fig. 4 1 illustrates the disparity in the degree of surface modification by EO BO through a comparison of the C 1 s spectrum of each lens surface prior to and following solution treatment wi th the highest EO BO concentration. A difference curve was obtained for each lens type by subtracting the spectrum of a neat sample from that of an EO BO treated sample. Here, all spectra have been normalized to the same integrated intensity, thus making the difference curves across different lens types quantitatively comparable. By attributing the integrated area under the positive portion of each difference curve at ~286 eV to intensity arising from C O species found in EO BO, the extent of adsorption c an be estimated as a percentage of the total carbon intensity generated from within the near surface region (Table 4 4). Noting that all lenses exhibited
114 approximately the same atomic concentration of carbon within this region, these percentages can be us ed as a guide to differentiate the lense s in terms of their extent of surface interaction with EO BO molecules. This approach specifically provides a quantitative method of comparing uptake data that moves beyond the qualitative description concluded in th e previous study (see Chapter 3); the trends observed in these two reports are identical. Discussion The four lenses examined in this study, with their distinctive compositions and surface treatments, are differentiated by the extent of EO BO adsorption to the surface or near surface region and absorption into the bulk. Together, the data portray a range of models describing the distribution of the EO BO surfactant throughout the hydrogel materials. In general, the data indicate the following comparativ e descriptions. Balafilcon A exhibits relatively high EO BO uptake in both the surface region and the bulk. For lotrafilcon B, high uptake was observed only in the surface region with little surfactant being absorbed by the bulk lens. Senofilcon A was sh own to have the lowest percentage of EO BO in the surface region, yet demonstrated a moderate degree of bulk uptake. Finally, comfilcon A exhibited a relatively modest modification of surface composition with significant bulk uptake of EO BO. These data also lead to the following observations: (i) the two lenses receiving plasma surface treatments exhibit an affinity for adsorption of EO BO in the surface region, and (ii) independent of the amount adsorbed in the surface region, lenses exhibit vastly var ying amount of the copolymer in their bulk. The differences in surface and bulk uptake behavior for the
115 different lens types also suggest the presence of gradients of EO BO concentration within the lenses. The presence of concentration gradients in the nea r surface region ( up to the outermost 10 nm) can be qualitatively evaluated by angle resolved XPS (AR XPS). AR XPS has proven to be an effective tool for analyzing the chemical states present at various depths beneath the surface of hydrogel materials [ 6 2 ] which in turn, can provide evidence for delineating whether the molecules have penetrated into the near surface region during solution treatment, or simply remained atop the surface. Models constructed to account for attenuation of photoelectron intens ity as a function of collection angle indicate that 95% of the data measured at 85 TOA corresponds to species found in topmost 2 nm, inclusive of the species adsorbed atop the surface. At a TOA of 65, intensity arises from species within approximately 1 0 nm of the surface plane. Relative differences in the amount of EO BO signature appearing for these two sampling depths would highlight the preferential adsorption of EO BO to the outermost portions of the hydrogel. This was the case for the balafilcon A and the lotrafilcon B samples, both of which have received factory surface treatments for improved wettability of the lenses. Particularly in the case of lotrafilcon B, there was very little evidence suggesting the presence of hydrophilic C O species fo und beneath the outermost 2 nm of the surface region. A second model of distribution of EO BO was observed for the senofilcon A and comfilcon A samples. The C 1s spectra of these lenses, taken at 85 TOA, revealed little, if any, change in intensity for the C O binding energy range as a function of angle, consistent with a uniform EO BO concentration
116 through the outermost 10 nm. A detailed account of the AR XPS analysis can be found in Appendix B. Finally, the disparity between the amount of EO BO adsorb ed onto the surface and absorbed into the bulk can be assessed through calculations of surfactant molecule concentration in the different regions of the hydrogel structures. These calculations will obviously rest upon a number of simplifying assumptions; they are, however, beneficial in benchmarking the molecular nature of interaction between the EO BO copolymer and the various hydrogels. First, from the UPLC data, it is relatively straightforward to calculate an average molecular concentration within the lens from the mass of EO BO extracted from each lens. Based upon the dimensions of the lens, assigning a diameter o f 14 mm and a thickness of 100 m, and the copolymer molecular weight, an average bulk concentration can be calculated. For the purpose of visualizing molecular concentration, the average bulk concentration for the different lenses is presented in units of number of EO BO molecules per 1000 nm 3 a unit volume with dimensions on the scale of the Flory radius of the polymer and the sampling depth of the XPS. As previously discussed, gradients in concentration exist within the bulk as a function of distance from the surface, yet the avera ge values presented in Table 4 5 are useful in defining the order of magnitude in concentration. The calculation of molecular concentration within the surface region is based upon the assumption of a silicone hydrogel density in the dried state (1.11 g/ cm3, ref.  and the atomic concentration s of elements measured via XPS By integrating the area under the curve describing exponential attenuation of photoelectron intensity as a function of depth, it is possible to determine a sampling volume. XPS m easurements of
117 the atomic concentration of carbon allow the determination of the mass of carbon present within this sampling volume. The relative percent of adsorbed EO BO ( Table 4 4) in turn provides the opportunity to calculate the mass of EO BO, which is easily converted to the number of molecules via the molecular weight of the copolymer. Again for the purposes of visualization, the results have been scaled to the number of copolymer molecules found in a 1000 nm 3 volume (Table 4 5). In this case, th e thickness of the volume element is taken to be outermost 10 nm of the surface and does not account for any gradients within this region. Together, these results demonstrate both surface and bulk effects. Lenses with hydrophilic surface treatments showe d evidence of preferential adsorption within the near surface region, as in the cases of balafilcon A and lotrafilcon B. For the case of lotrafilcon B, it is unclear whether the surface treatment inhibits bulk adsorption or the bulk hydrogel exhibits a lo w affinity for copolymer absorption; nonetheless, this lens exhibited the lowest degree of bulk uptake. Variations in the affinity for copolymer uptake are further illustrated for the two lenses with no surface treatments, i.e. senofilcon A and comfilcon A, highlighting the influence of the specific hydrogel polymer compositions on the degree of bulk uptake. The results also depict the nature of EO BO concentration gradients. Although only average concentrations have been calculated for the bulk and the surface regions, the diffusion mediated uptake of the copolymer into the relatively low density hydrogel requires continuity in concentration as a function of depth into the bulk. The resulting gradient would move from a higher concentration at the surfac e, measured in all cases, to a lesser concentration in the bulk. With this in mind, the concentration gradient would be seen as the greatest for the case
118 of lotrafilcon B and the least for senofilcon A; the two other lenses exhibit intermediate gradients. Further kinetic studies would be required to refine this picture and to elucidate depth specific concentrations for the different lenses. Concluding Remarks As the inclusion of amphiphilic surfactants in multi purpose disinfecting solutions becomes increa fundamental interaction between these functional molecular species and the hydrogel lens materials increases in importance. In this study, aqueous solutions of an EO BO diblock copolym er have been used to treat four silicone hydrogel lenses possessing a range of bulk compositions and surface treatments. Following solution treatments, the extent of lens surface modification by EO BO and the degree of bulk uptake have been studied using XPS and UPLC, respectively. The experimental results suggest different interaction models for the lenses, highlighting the influence of both surface and bulk composition. Together, these results depict EO BO molecular concentration gradients as a function of depth into the lens with the concentration levels being lens specific. The results further suggest opportunities for compositional modifications of lenses with the potential for improved performance via solution treatments containing surface active ag ents. One example of such opportunity was evidenced previously where the adsorption of EO BO has led to the reduction in friction coefficient of the lenses and the degree of such reduction was also lens specific. From a broader perspective, physiological conditions are likely to influence and participate in such compositional modifications of lenses, through both surface adsorption and bulk uptake.
119 Table 4 1. Details of the four major silicone hydrogel contact lenses with their respective treatment for improving hydrophilicity [36, 93, 103] Trade Name USAN Name Published Composition Hydrophilic Surface Treatment PureVision B alafilcon A NVP, TPVC, NVA, PBVC Plasma oxidation ACUVUE O ASYS S enofilcon A mPDMS, PVP, siloxane macromer, DMA, HEMA None (PVP internal wetting agent) O 2 OPTIX L otrafilcon B TRIS, siloxane macromer, DMA Uniform thin coating via plasma induced polymerization Biofinity C omfilcon A FM0411M, HOB, IBM, M3U, NVP, TAIC, VMA None NVP: N vinyl pyrrolidone; TPVC: tris (tr imethylsiloxysilyl) propylvinyl carbamate; NVA: N vinyl amino acid; PBVC: poly(dimethysiloxy) di(silylbutanol) bis(vinyl carbamate); mPDMS: monofunctional poly(dimethylsiloxane); PVP: poly(vinylpyrrolidone); TRIS: trimethylsiloxy silane; DMA: N,N dimethyla crylamide; HEMA: poly(2 methacryloyloxyethyl iminocarboxyethyloxypropyl poly(dimethylsiloxy) butyldimethylsilane; HOB: 2 bis(methacryloyloxyethyl iminocarboxy e thyloxypropyl) poly(dimethysiloxane) poly(trifluoropropylmethysiloxane) methoxy poly(ethyleneglycol)propylmethylsiloxane; TAIC: 1,3,5 triallyl 1,3,5 triazine 2,4,6(1H,3H,5H) trione; VMA: N vinyl N methylacetamide
120 Table 4 2. Quantitative compariso n of the amount of EO BO ( standard deviation) eluted from different silicone hydrogel lenses as measured by UPL C. EO BO Concentration Total amount of EO (%wt/%vol) B alafilcon A S enofilcon A L otrafilcon B C omfilcon A 0 (contr ol) 0 0 0 0 0.01 14.43 1.08 9.66 0.34 0 15.57 0.70 0.04 48.20 1.55 34.84 1.45 3.36 0.74 55.94 0.79 0.1 151.77 3.74 77.96 2.44 7.21 1.16 140.24 4.96
121 Table 4 3. Comparison of quantitative distribution (at.% fitting error) of d econvoluted carbon 1 s species of lenses treated with saline and EO BO solution of the highest concentration C H/C C/C Si C O/C N N C=O O C=O B alafilcon A Saline 61.3 1.1 26.2 0.5 9.7 0.2 2.8 0.1 EO BO 53.8 0.7 36.0 0.5 8.7 0.1 1.5 0.1 S enofilcon A Saline 64.2 1.0 25.9 0.4 5.9 0.1 4.0 0.1 EO BO 63.5 0.9 26.5 0.4 6.4 0.1 3.6 0.1 L otrafilcon B Saline 70.1 2.0 20.8 0.6 9.1 0.3 0 EO BO 53.4 0.7 37.2 0.5 9.4 0.1 0 C omfilcon A Saline 65.6 1.2 26.9 0.5 7.5 0.1 0 EO BO 63.2 1.0 29.6 0.4 7.2 0.1 0
122 Figure 4 1 C 1 s spectra of the neat (dotted) and the treated (solid) sample for each lens type A) Balafilcon A. B) Senofilcon A. C) Lotrafilcon B. D) Comfilcon A. Note that in A) and C) intensity arising at highlighted portion of the spectrum (arrows) evidences the preferential surface adsorption of EO blocks. However, samples B) and D) do not produce significant peaks at the same binding energy, suggesting a much lesser degree o f adsorption of EO BO on the surface. The data shown here correspond to treatments with the highest EO BO concentration, i.e. 0.1 %wt/%vol. B A
123 Figure 4 1 Continued. C D
124 Table 4 4. Semi quantitative comparison of the degree of adsorption of EO BO within the upper 10 nm of the different silicone hydrogel lens types. Percent adsorption reflects the percentage of C species within the near surface region attributable to EO BO. Lens type Relative degree of adsorption (%) B alafilcon A 7.1 S enofilcon A 0.7 L otrafilcon B 8.2 C omfilcon A 3.0
125 Table 4 5. Average concentration of EO BO measured by UPLC and XPS Lens type Average concentration in the bulk (molecules/1000 nm 3 ) Average concentration at the surface (molecules/1000 nm 3 ) B alafilcon A 1.6 10.7 S enofilcon A < 1 (0.8) 1.1 L lotrafilcon B << 1 (0.1) 12.1 C omfilcon A 1.5 3.6
126 CHAPTER 5 INFLUENCE OF A SURFACE GEL LAYER ON THE ELASTIC MODULUS OF SILICONE HYDROGEL MATERIALS MEASURED WITH COLLOIDAL PROBE AFM Background and Motivation One of t he key parameters employed to describe the mechanical property of a materials are often d istinguished by their bulk elastic moduli measured under uniaxial tension as a quantification of their mechanical durability  However, it is not difficult to realize that bulk tensile modulus alone cannot sufficiently explain the surface mechanical a nd tribological behavior of lens materials, considering the compressive nature of the loading condition under which lenses are employed. Microtribometry and atomic force microscopy (AFM) experiments performed in the past have demonstrated a positive corre lation between the compressive modulus of a hydrogel lens material and its friction coefficient  Typical hydrogel contact lens materials consist of chemically crosslinked polymer networks with equilibrium water contents (EWC) ranging from on averag e, 4 0% for silicone hydrogel lenses to 60% for PHEMA hydrogel lenses (generally a copolymer of HEMA with a more hydrophilic monomer) in the bulk  The lower EWC of silicone lenses originates from the specific siloxane containing bulk chemical composit ions. Whether incorporated into the polymer chain or included as oligomers, siloxane components are generally hydrophobic but are exceptional for oxygen transport due to the flexible Si O bonds  As a result, a major challenge in silicone hydrogel le ns design is to circumvent the reduced hydrophilicity at lens surface, which leads to tear wetting instability, with specific chemical treatments that can alter the surface energy of
127 the lens or with molecular design that improves the retention of water at the lens surface. When employing a colloidal probe, AFM has been proven useful in measuring the elastic modulus of hydrogel materials, biological tissues, and living cells in the past [41, 43, 85, 86] With the assumption of an appropriate contact mechan ics model, known probe geometries enable the calculation of the applied stress. With the application of forces on the order of nanoNewtons, the micron sized probe can produce relative large contact area in order to achieve physiologically relevant contact pressures, unattainable with sharp probes. The classic Hertz model is frequently used to analyze the contact mechanics of homogeneous soft materials and has been shown to be effective in delineating the elastic moduli of a number of contact lens materia ls (baseline measurements preceding the current study; results not shown). The Hertzian theory assumes the contacting surfaces to be non conforming, frictionless elastic half spaces, and predicts linear elasticity under small strain ( details of the theory may be found in Chapter 2 ) [46 121 ] Fitting the experimental AFM data to the power law based Hertzian relationship (Equation 5 1) between force F N and indentation depth gives rise to the elastic modulus of the hydrogel surface (5 1) This study aims to investigate the surface elastic properties of a novel silicone hydrogel contact lens material. The material, formally known as delefilcon A, consists of a silicone hydrogel core modified with a poly(acrylic acid) co poly(ac rylamide) gel layer
128 via a proprietary chemical coupling process  Unlike most silicone hydrogel lens materials with homogenous EWC below 45%, delefilcon A reportedly has an EWC greater than 80% in the near surface region measuring approximately 6 thickness, and a bulk EWC similar to other silicone lenses. As a previous study by Dunn et al. has shown, this exceptionally high EWC is consistent with a low modulus hydrogel structure capable of providing low friction sliding in aqueous environme nt under boundary lubrication conditions  Indentation results reported in the same study briefly discussed the depth dependent elastic modulus of delefilcon A; here, more detailed considerations will be given to the possible origins of its mechanical behavior, especially in comparison with other typical silicone hydrogel lenses. It is hypothesized that the orders of magnitude of variation in the elasticity of delefilcon A can be correlated with changes in the depth specific molecular structure and t he resulting gradient in water content. As such, modifications to the conventional understanding of hydrogel elasticity are necessary in order to account for the unique, graded structure found for delefilcon A. Experimental Materials The delefilcon A lens (DAILES TOTAL1, Alcon) employed in this study was commercially fabricated and slight variations based on the power of the lens. Senofilcon A (ACUVUE OASYS, Johnson & Johnson) and balafilcon A (PureVision, Bausch & Lomb) lenses designed for extended wear were also analyzed for comparative purpose. These two particular lenses were chosen because they represent two general scheme s of improving surface hydrophilicity of silicone hydrogels (see discussion in Chapter s 3 and 4). Prior to the
129 AFM experiment s each lens sample was soaked in saline (Unisol4, Alcon) for 24 hours in order to eliminate the effect of blister pack solutions. All modulus values were obtained from the anterior surface of the sample. Methodology The surface elastic modulus of each hydrogel lens was measu red at room temperature (25C) using a commercial stand alone AFM (Asylum Research, Santa Barbara, CA). The instrument is equipped with a specialized liquid sample holder ensuring that both the probe and the sample are fully submerged in saline throughout the duration of the experiment. A 5 as the probe, which was affixed on a calibrated cantilever, as imaged by SEM in Fig. 2 18, with a normal force constant of 5.19 N/m. The normal force constant of the ca ntilever used in this study was de termined experimentally by the thermal method detailed in Chapter 2 and Appendix A [ 12 2 ] This method relates the normal spring constant with the motion of the cantilever due to thermal fluctuations in free air, i.e., free of the influence of probe surface interactions. The cleanliness of the microsphere was maintained by oxygen plasma treatments following each experiment to remove possible material transfer from the sample. The approach speed of the z piezo was set to be 1 Multiple force curves and multiple locations on a given surface were sampled to capture the characteristic and representative modulus value of the lens. The elastic modulus of the hydrogel samples was measured in the following manner The probe was engaged with the sample at contact, from which point the increase in force experienced by the deflected cantilever was monitored as a function of the travel distance of the probe toward s the sample. In the present study, each force curve was carefully treated in order to evaluate the point of contact, taking into account
130 The indentation depth of the probe into the surface was subsequently calculated as depicted i n Fig. 2 19 A Upon fitting the force indentation behavior to the appropriate contact mechanics model, the elastic modulus of the sample surface was determined (detailed description of the fitting exercise is provided in Chapter 2). It is important to not e that the exact magnitude of contact pressure and the depth of indentation are dependent upon the surface elastic modulus of the material, and thus vary for each sample. The maximum contact pressure estimated by the Hertzian model for each sample examine d in the study was kept well within the range of elastic deformation, i.e. much lower than the estimated elastic modulus. Results A representative data set of force F N versus indentation depth for delefilcon A is shown in Fig. 5 1 A The resolution of th e piezo displacement (i.e. the depth of indentation) allows one to analyze the elastic modulus of the hydrogel to the precision of nanometers, affording the opportunity to delineate the surface mo dulus at depths relevant to the gel layer Slight variation s in the measured surface modulus are expected for the hydrogel samples due to the random nature of polymerization; as a result, these values should not be considered exact, but should benchmark the range of elastic behavior expected for a given sample. T he F N plot of delefilcon A demonstrates regions of distinct elastic behavior at various depths. Specifically, forces on the scale of picoNewtown were detected in the resolution da ta points in this regime are significantly non zero and signify the onset of the initial probe contact with the outmost portion of the surfa ce gel
131 surface represents a region of constant modulus of approxim ately 12.6 0.2 kPa, as described by the Hertzian model. The error associated with the modulus estimation takes into account the optical sensitivity of the photodetector and the precision in the greater than 0.2 continued to indent. examining a logarithmic transformation of the force plot ( Fig. 5 1 B ). The exceedingly points below 1 nN. The resolution of force readings is compromised at this scale as the measurement is nearing the detection limit of the i nstrument. The subsequent regime is seen to coincide with a linear Hertzian model (slope = 3/2 according to Equation 5 1) fitted with a constant modulus of approximately 12.6 kPa. Further indentation resulted in a sharp deviation from the Hertzian behavi or, consistent with a depth dependent elasticity model. Comparison with two other silicone hydrogel lens materials designed without the attachment of a surface gel layer is illustrated in Fig. 5 2. Note that the data points of delefilcon A have been tru ncated to be displayed on the same scale as the other two lenses. Two observations can be made regarding the behavior of these three distinctly different materials. First, the mechanism of elastic deformation great differs between delefilcon A and the tw o lenses for extended wear. Fitting exercises for senofilcon A and balafilcon A reveal that the Hertzian model is applicable for the entire range of indentation depths probed. Second, at comparable indentation for which the elasticity
132 of delefilcon A is Hertzian in nature, the magnitude of modulus of senofilcon A and balafilcon A, 640 kPa and 2100 kPa respectively, are considerably higher than delefilcon A. Discussion The distinct behavior seen at various indentation depths accurately reflects the designe d gradient in the surface chemistry of the delefilcon A material (Fig. 5 3). Immediately upon contact, the probe senses the extended molecular structure at the surface. This diffuse structure, measuring 0.05 the termination chemistry of the surface gel layer. An exceedingly compliant force response was observed under compression, consistent with dangling chains extending from the surface gel layer that exhibit lower pol ymer concentration. However, the determination of a modulus value is precluded in this case because most of the load is being carried by the fluid phase, which cannot be simply described by an elasticity The underlying 0.15 elastic modulus calculated by the Hertz contact theory. The fitting of the F N behavior with a constant modulus suggests that the deformation of the surface gel in this region displays a linea r elastic proportionality, as true for most elastic materials obeying characterizing this elastic regime is reflective of a gel structure with low polymer content, low crosslinking density, and water content greater than 80% by volume. As the probe indents further, a non linear elasticity regime is experienced in delefilcon A. Such deformation behavior is believed to be a result of the load being supported by the underlying substrate with higher crosslinking density and polymer
133 content, namely, the anchor layer as shown in Fig. 5 3. At larger indentation depth ( > assumption in the Hertzian model, but a structure with f inite thickness attached to a much stiffer substrate, i.e. the anchor layer and the silicone hydrogel core. As such, part of the load may be carried by the substrate, giving rise to an increasing resistance to deformation as seen in Fig. 5 1 The effect of the graded structure exhibited by delefilcon A is further corroborated by a comparison against silicone hydrogels without such treatment. Senofilcon A is characterized by a constant elastic modulus of 640 kPa using the Hertzian model. This value is s imilar to the 710 kPa bulk modulus measured by Dunn et al. using a microindenter [ Dunn, A.C. unpublished data ], reflective of the homogeneous composition of the material. Comparison between the bulk and surface moduli of balafilcon A confirmed the effect of its surface chemical treatment. Balafilcon A was reported to have a bulk modulus of 490 kPa, which is significantly lower than that of the oxygen plasma treated surface. The plasma treatment is believed to result in a much stiffer surface region that effectively accommodates the strain imposed by the applied stress and precludes the possibility of displaying depth dependent deformation behavior. As such, the unique structural gradient present at the surface explains the non linear elasticity of delef ilcon A and clearly differentiates it from other silicone lenses available. Conclu ding Remarks The surface elastic modulus of a novel silicone hyrogel lens material, delefilcon A, was analyzed using colloidal probe AFM. Comparison against two silicone h ydrogel lenses of different surface treatments reveals that the chemically anchored gel layer
134 measuring a few microns in thickness was responsible for the depth dependent elastic behavior in delefilcon A. With an equilibrium water content greater than 80% and an exceptionally low modulus of about 12.6 kPa in the selvage region of the lens, the surface gel layer offers great potential for mitigating many in vivo interfacial challenges typically experienced by silicone hydrogel lenses. From a fundamental pe rspective, understanding the characteristics and implications of a surface construct with a distribution of composition affords the opportunity for designing polymeric materials with tunable chemical and mechanical properties.
135 A B Figure 5 1 A representative set of indentation data of delefilcon A. (A ) Force indentation plot. B) The same data set on the logarithmic scale. The linear regime of elasticity is onl which large deviation from the Hertzian model is observed. The scattered points at F N << 1 nN are indicative of the repulsive interaction between the probe and the outermost, dangling polymer chains.
136 Fi gure 5 2. Comparison between three types of silicone hydrogel lenses reveals the effect of surface structure on the deformation mechanism and the magnitude of elastic modulus under compression of a colloidal probe.
137 Figure 5 3. Schematic of the surfac e gel layer and the underlying substrate. The dimension of the gel layer and the anchor zone are determined by the concentration of the polymer and crosslinking agents present. Note that in reality the transitions between the layers are not sharp interfa ces, but rather demonstrate a continuum of compositions, giving rise to the unique mechanical properties observed.
138 CHAPTER 6 AN INVESTIGATION OF THE DEPENDENCE OF SURFACE MECHANICAL PROPERTIES ON WATER CONTENT OF VARIOUS HYDROGEL MATERIALS Background and Motivation As depicted by the previous experiments embodied in this work, the surface mechanical properties of silicone hydrogel contact lens materials are largely influenced by their surface chemical compositions and structures. Briefly, in the case of copolymer adsorbed surfaces, coefficients of friction can be five times greater at the plasma treated surfaces than at non treated surface. In terms of surface elastic modulus, silicone hydrogel with loosely cross linked surface gel layers demonstrated a modulus value two orders of magnitude lower than the plasma treated surface. While these experimental results have shed much light upon the measured mechanical properties with known surface chemistry, a fundamental understanding is still needed for correl ating controllable parameters, such as water content, with elastic modulus at the surface. The aim of the present study is to investigate the surface elastic modulus of different types of hydrogel materials with specific values of water content, which is related to the amount of polymer incorporated in the hydrogel at equilibrium swelling. Four types of hydrogels will be synthesized to achieve a broad range of water content from 15% to 95% by weight (Fig. 6 1). Note that all quantities that describe bulk concentration are represented in terms of w/w% hereafter. Due to synthetic limitations such as original monomer precursor concentration (e.g. PAAM stock solution used in the present study contains 40% monomer ) and phase separation at characteristic polym er water mixing ratios (e.g. PHEMA precipitates out of an aqueous solution when water content exceeds 40% see [ 12 3 ] ) only a range of polymer concentration is
139 attainable with a given type of polymer. The experimental approach involves fabricating these h ydrogels samples in house, rather than obtaining them from commercial sources, thus eliminating the number of variables that must be considered. The hydrogel s investigated in the present study are widely used in the field of biomaterials, and some are si gnificant in the contact lens industry. PHEMA for example, constitutes the first generation of soft hydrogel contact lens material and has been studied extensively from a fundamental perspective due to its unique ability to solution environment  Poly(acrylamide) (PAAM) hydrogel is employed in applications such as ophthalmic surgery, drug delivery, and tissue implants due to its excellent biocompatibility [ 12 4 12 5 ] Hydrogels composed of copolymers of HEMA and methacry lic acid (MAA) respond to external pH changes by swelling and de swelling, and as a result have found many uses in site specific drug delivery systems, artificial muscles, and biosensors applications [ 12 6 ] When poly( N isopropylacrylamide) (PNIPAM) sample s undergo a thermodynamic phase transition at approximately 32 C [23, 12 7 1 30 ] its lower critical solution temperature (LCST), changes in bulk mechanical properties are known to occur as a result of segregation between a polymer rich phase and a solvent rich (DI water in this case) phase (Fig. 6 versatile applications in targeted drug delivery, molecular separation process, catalytic reaction control, and microactuators, etc. [4 5, 12 9 ] Extensive work has been done in developing suitable methodologies for
140 discussion concerning the measurements of surface elastic modulus will be reviewed. Rep orts on this topic may be categorized by the experimental approaches employed in AFM measurements, namely, the use of different probe shapes. While many have adapted colloidal probe for measuring cell mechanics [85, 86] uses of sharp probes or cone shape d probes are still commonplace for hydrogel materials [40, 41, 107, 13 1 ] nm) on a flat sample of hydrogel is orders of magnitude greater than that exerted by a colloidal insufficient [44, 46] Alternatively, in light of the contact pressures and elastic p roperties of biological tissues relevant in ocular environment, the use of colloidal probes allows reasonable delineation of surface elastic modulus of soft hydrogel materials suitable for contact lens applications. The first half of the investigation foc uses on the synthesis of a variety of hydrogels to achieve a broad range of polymer concentration. The surface elastic moduli of these hydrogels were then measured with colloidal probe AFM, an experimental approach identical to that described in Chapter 5 The remaining portion of the study aims to understand the changes in the surface elastic modulus of PNIPAM as it experiences phase transition at its LCST. The force indentation curves collected from the AFM were analyzed to elucidate the structural cha nges occurring during composition to the measured surface elastic modulus in aqueous environment, as well
141 as the interaction between the solvent and the polymer ne twork in contributing to the contact mechanism at the probe polymer interface. Experimental Hydrogel Synthesis Four types of hydrogels, including PAAM, PHEMA, P(HEMA co MAA), and PNIPAM, were synthesized by aqueous based solution polymerization (Fig. 6 3 ). In addition to stock monomer precursor, each polymer solution mixture contained specific concentrations of initiators and cross linkers. Each gel mixture was balanced with the appropriate amount of water. Because the combined amount of crosslinker a nd initiators was minute for each mixture, the equilibrium water content (EWC) was approximated by the amount of polymer contained. Note that some gels, e.g. PAAM and PNIPAM, incorporated the same types of initiators and cross linkers at different concent rations. Table 6 1 describes the ingredients involved in each solution mixture. Stock solutions of methylenebisacrylamide (BIS, CAS no. 110269, Sigma Aldrich), ammonium persulfate (APS, CAS no. 7727540), and sodium metabisulfite (SMBS, CAS no. 7681 574) were prepared at concentrations of 2%, 10%, and 10%, respectively. All monomer precursors were available from manufacturer in solution form except for NIPAM, for which a 14% aqueous stock solution was prepared. As shown in Table 6 1, APS was paired with either tetramethylethylenediamie (TEMED, CAS no. 110189) or SMBS to initiate the polymerization through re dox reaction. Also, tetraethylene glycol dimethacrylate (TEGDMA, CAS no. 109171) was utilized as the crosslinking agent based on the report by Kou et al.  All chemicals were obtained from Sigma Aldrich and were used without further purification. The recipe for each
14 2 hydrogel was adapted from multiple reports, and in some cases, the recipe was modified to meet the need of producing thin films su itable for AFM testing. The procedures for preparing hydrogel films are identical for each polymer and are described as the following. The ingredients were carefully mixed with micropipette in a cell culture tube, avoiding formation of air b ubbles, and slowly pipetted between two plasma cleaned glass slides (see discussion in Chapter 2) spaced apart by microscope cover slips. The polymerization process was initiated immediately following the addition of the initiators. The mold containing th e solution mixture was set to cure at ambient conditions overnight, with exception of the P(HEMA co MAA) gel, which was cured at 60 C in the oven for 12 hours. The gel was then released from the mold by immersing it in DI water. The resulting stand alon soaked in fresh DI water for 24 hours prior to AFM measurements. Note that soft gels were made to be thicker to facilitate sample handling and mounting. Temperature variant Measurements in AFM Temperatur e control in the AFM liquid cell is achieved by a specialized sample holder integrated with an integrated heating stage and a thermistor (Fig. 6 4 B ). The overall assembly of the sample holder (Fig. 6 4 A ) is similar to that shown in Fig. 2 1 7 The heating stage is connected to a master controller that is capable of maintaining a specific temperature set point via electronic mechanism. In order to monitor changes in surface mechanical properties of PNIPAM, force analyses were carried out at room temperature (approximately 25 C) and at 37 C. Sufficient time was allowed for the phase transition to reach equilibrium upon heating or cooling, a practice suggested by Sui et al 
143 Surface Elastic Modulus Analyzed by Colloidal Probe AFM The surface elastic modulus of each hydrogel sample was analyzed with colloidal probe AFM as detailed in C hapter 5 Protocols including plasma cleaning the colloidal probe and glass support substrate of the liquid cell were closely followed between indentation ex periments (Note: the probe was not cleaned before and after transition at the LCST for the PNIPAM sample). The same colloidal probe/cantilever assembly was employed for all of the AFM measurements in the present study, and the normal force constant of the cantilever onto which the probe was glued was calibrated to be 0.451 N/m using the Sader method (see discussion in Chapter 2 and the Appendix A ). The procedures for collecting and analyzing force curves measured with colloidal AFM follow those described in Chapter 2. The indentation depth of the probe into the sample is experimentally calculated as the difference between piezo displacement and cantilever deflection (see Fig. 2 19 ) in the approach portion of the force curves. The surface modulus was t hen derived from a Hertzian contact model by least square fits. Here, conditions that satisfy the Hertzian theory, as discussed in Chapter 2, are assumed. For a given sample, at least five locations were examined, with multiple force curves analyzed at e ach location and compared between different locations to inspect consistency in surface elastic modulus. In addition, AC mode AFM topographic investigations were conducted to explore any microscopic morphological changes associated with the phase transiti on for PNIPAM.
144 Results Surface Elastic Moduli of PAAM, P(HEMA co MAA), and PHEMA in DI Water PAAM PAAM hydrogels are high water content hydrogels (typically greater than 60%). Here, samples with compositions at 95%, 85%, and 65% water content (correspondi ng to 5%, 15%, and 35% polymer) were synthesized and their surface elastic moduli were examined using AFM. In order to explore the relationship between modulus and polymer content, the concentration of crosslinking agent was fixed at 0.03%. Following the analytical protocols described earlier, representative force indentation data sets for each sample are plotted in Fig. 6 5 A for comparison. The modulus values were calculated to one tenth of a kPa, reflecting the resolution in force and displacement meas urement of the microscope. Due to the random nature of the polymerization process, slight variations in numeric values of surface elastic modulus are expected; therefore, the modulus values reported here should not be considered exact, but should signify the range of stiffness expected for each sample under a given set of conditions. For the sample containing 5% PAAM, a surface elastic modulus of approximately 1.4 kPa was measured (Fig. 6 5 B ), in g ood agreement with a prior study [ 13 2 ] Engler et al. [ 13 3 ] based on their microindentation study, reported the relationship between bulk elastic modulus and polymer concentration to be a 3 rd order polynomial function at a given cross linker concentration. It is reasonable to believe that, due to the known dis parity in polymer density at the surface and the bulk of a hydrogel sample [ 13 4 ] a different relationship should exist between the measured surface elastic modulus and the PAAM concentration. Indeed, as shown in Fig. 6 6, when an averaged value based on three measurements of a given PAAM sample is plotted against polymer
145 concentration, that is, the volume ratio of polymer in the gel mixture, in logarithm axes, the resulted linear function suggests that with the correlation is to the 1.93 th power, small er than that modeled for the bulk. Note that the error bars included in Fig. 6 6 reflect the variations in the fitted modulus value between different locations of the same sample. As detailed in Appendix A, errors in the measurements of force distance da ta with AFM also include factors such as instrumental resolution (especially at low force values) and uncertainty in the force calibration of the cantilever. These errors are not incorporated in this particular figure as they are common to all measurement s in this set of experiments. P(HEMA co MAA) and PHEMA Similar to the PAAM samples, the force indentation data of P(HEMA co MAA) and PHEMA samples were measured and are presented in Fig. 6 7 A As seen in Fig. 6 7 B though non zero force readings were pres ent in the first 60 nm of indentation, the curve does not reflect a typical Hertzian force indentation relationship. As a result, an offset along the indentation axis was applied to each data set in order to estimate the elastic modulus. The low force re ading is attributed to the long range repulsive interaction between the probe and the loose polymer chains at the surface. A detailed analysis of the forces at the interface is given in the Discussion section. The elastic modulus values estimated for th e HEMA samples did not differ between the samples to the same extent as those observed for the PAAM samples with much lower polymer concentration, i.e. higher water content. Furthermore, the HEMA MAA copolymer demonstrated slightly higher elastic modulus at surface, irrespective of its higher water content compared to the HEMA hydrogels. It is reasonable to believe that the dependence of the elastic modulus on polymer concentration varies for different
146 monomer chemistry as well as for the amount of polyme r incorporated. The copolymer required much higher cross linker concentration to gel compared to the homopolymer, which could be a contributor to the slightly higher surface modulus. For HEMA homopolymer samples with the same cross linker concentration, surface elastic modulus increased as the polymer concentration increased. Temperature induced Phase Transition and the Surface Mechanical Properties of PNIPAM The effects of temperature on surface elastic modulus, adhesion force, and topographical features were investigated on PNIPAM samples in DI water below and above its LCST at 32 C. Change in surface elastic modulus and adhesion force As seen in Fig. 6 8, force curves on PNIPAM surface at room temperature (both before and after the phase transition) and at 37 C are plotted and the elastic modulus for each data set was estimated by a Hertzian model. The following observations can be made: The elastic modulus estimated for the contact portion of the force curves increased by an order of magnitude abo ve the LCST; The effect of the phase transition on the measured surface modulus was reversible; The nature of the force interaction at the onset of contact was drastically changed as a result of the phase transition. At T < LCST, the 12% PNIPAM sample poss essed a modulus on the order of 10 kPa, which, is comparable to the measured modulus of the15% PAAM sample of similar cross linker concentration (Fig. 6 5 A ). After heating the sample to 37 C and allowing for 30 minutes of equilibration, force curves were sampled on the surface and the modulus was observed to be on the order of 100 kPa, indicative of increased polymer
147 chain interaction experienced during compression due to phase segregation between the polymer and the solvent. Low elastic modulus was obse rved again after the temperature was reduced to below the LCST, suggesting the surface mechanical properties have been fully recovered when the hydrogel again entered the swollen state. As described in Chapter 2, an AFM probe can jump to contact when the gradient of vdW attractive force exceeds the gradients of both the force constant of the cantilever and the repulsive force(s) present at the interface, leading to the occurrence of short range negative force in the force curve (Fig. 6 9B)  Such beh avior was evident in force curves measured at T > LCST, but not apparent below the transition temperature (Fig. 6 9 A ). The drastic difference seen here suggests that the phase transition lessened the magnitude of repulsive forces between the probe and the hydrogel surface, which was evidenced by positive, albeit low, forces observed at the interface for the initial ~ 50 nm of indentation in all of the force curves (Fig. 6 9 A ). The extent of surface adhesion upon pull off was also evaluated as a function of temperature. Three different sites were sampled on the surface and the adhesion force was calculated to be the difference between the minimum force on the force curve and that within the non contact region of the dataset, or the zero force line (this i s typically a small non zero value). At T < LCST, the adhesive force at the probe/hydrogel interface was much smaller compared to that at T > LCST, and the effect was reversible (Fig. 6 10). Change in surface topography by tapping mode AFM In order to e stablish constant osmotic pressure in both the polymer rich and solvent rich phases, the bulk volume of the hydrogel undergoes significant reduction as
148 the solvent molecules mobilize out of the hydrogel network. This volume transition is observed macrosco pically when temperature is raised above LCST. To gain further insight into the morphological changes occurring during the transition, AC mode AFM images were acquired. As shown in Fig. 6 11 A vir tually featureless as the z height contrast is only on the order of a few nanometers. The corresponding phase image (Fig. 6 11B) also displays small contrast, again indicating that the surface is homogeneous at this length scale. In comparison, at T > LC ST, more recognizable features were detected and the height contrast was on the order of hundreds of nanometers, with regions of varying roughness visible (Fig. 6 12). Due to the random nature of the phase separation process, quantitative roughness analys is is precluded here; only qualitative comparison s are made to highlight the changes induced by the phase separation. When the temperature is reduced below the LCST, images similar to Fig. 6 11 were again observed (results not shown here), further support ing the claim that the effect of phase transition is reversible. Discussion Analyzing Surface Elastic Modulus of Various Hydrogel Thin Films with Colloidal Probe AFM In the case of PAAM samples of varying polymer concentrations, a power law relationship w ith an exponent of approximately 1.93 was modeled between the elastic modulus and polymer concentration (Fig. 6 6). An order of magnitude of difference in surface modulus was observed between the lowest polymer concentration (5%) and the highest. de Genn E ) to polymer concentration ( C ) states that
149 (6 1) where k is a proportionality constant and is a function of temperature and crosslinking density [ 13 5 ] The term describes the compatibility between the polymer phase and the solvent environment and is usually between 2.25 (in a good solvent) and 3 (in a theta solvent). A comparison with Equation 6 1 suggests that, for elastic modulus at the surface modifications must be made to accommodate the structural differences expected between the surface and the bulk. A smaller exponent value observed at the surface indicates a weaker dependence of modulus on polymer concentration, possibly due to the decreased polymer de nsity at the surface, resulting lower resistance to deformation. The concept of decreased molecular density at the surface was elucidated by Heuberger et al. lysine) graft poly(ethylene glycol) (PL L g PEG) brush layer with extended surface forces apparatus (eSFA) [ 13 4 ] Analysis of the surface force isotherms measured on PLL g PEG brush indicates that a finite polymer density exists beyond the film thickness predicted by scaling laws. Such conclus ion has led to the modeling of a polymer density profile as a function of the distance away from the grafting site (Fig. 6 13). Although the structure of the loosely associated chains at a hydrogel surface does not resemble a brush configuration, the appl ication of this density distribution model to the present system is valid because the origin of load support at the gel surface, namely the increased interaction between chains under large compression, is very similar to the steric factors that contribute to the interaction forces in a brush system.
150 For the P(HEMA co MAA) and PHEMA samples, smaller changes (e.g. all within the same order of magnitude) in surface modulus were seen with varying polymer concentrations. Furthermore, the copolymer, though with lower polymer concentration than the PHEMA samples, demonstrated slightly higher modulus. This result suggests that polymer concentration alone does not fully predict the elastic modulus at the surface, and that one must also take into consideration the specific components present in the network to understand the surface mechanical properties. Note that the analogy for bulk modulus is encapsulated in the term in Equation 6 1, given the same solvent environment. Temperature induced Phase Transition in PNIPAM Hydrogel The surface elastic modulus of PNIPAM below and above its LCST has been examined by a number of groups, although no consistent values were obs erved across different reports. This is partly due to the generic application of Hertzian theory to many different configurations of PNIPAM employed in these studies. These configurations include chemically immobilized and cross linked thin films [45, 12 8 12 9 ] grafted brushes  or chemisorbed microgels on solid substrates such as silicon wafer or glass [ 12 7 13 6 ] Depending upon the thickness of a given PNIPAM sample, the geometry and the indentation depth of the probe, the elastic modulus describe d by the Hertzian relationship (Equation 2 13) varied significantly between these experiments. As some have suggested, the effect of the solid support cannot be excluded from the contact mechanics analysis, especially when the indentation depth is relativ ely large with respect to the total thickness of the samples [23, 44] If the substrate effect is indeed significant, the elastic foundation model is needed to better estimate the elastic modulus of the hydrogel sample  In the present study, the in dentation depth of the
151 probe was carefully controlled with respect to the size of the probe and the thickness of the sample (~ 1% of the total thickness of the film) in order to minimize the support of the solid substrate and to keep the analysis relevant within the linear elasticity regime of the samples. The change in surface elastic modulus of a PNIPAM sample undergoing phase transition at 32 C is illustrated in Fig. 6 8. Through the AFM measurements presented, the highly solvated surface was estima ted to have an elastic modulus on the order of 10 kPa. The extent to which the hydration of the polymer chains influences surface modulus is better realized after the system was heated above its LCST, which resulted in an order of magnitude of increase in modulus. The thermodynamic process of phase transition is explained as follows. Below the LCST, hydrogen bonding between water molecules and the amide group of NIPAM is favored, leading to the formation of an organized hydration layer around the polymer chains. This structural arrangement contributes favorably to the enthalpy of mixing at the expense of the entropy of the system. As the temperature of the system is increased to above the LCST, increased entropy of mixing dominates the free energy funct ion and matrix [ 13 5 ] The polymer rich phase consists of an interconnected network of locally densified polymer chains (Fig. 6 2). Microscopically, changes in surface topography and morphology were apparent following the phase transition. AC mode AFM ima ges confirmed that in the swollen state, the hydrogel surface was virtually featureless, possibly due to the fact that the probe was making intermittent contact with the hydration layer surrounding the polymer chains at the surface (Fig. 6 11). This hydra tion layer consists of water molecules
152 closely associated with the hydrophilic component of NIPAM by hydrogen bonding [ref, Cho]. As a result, the possibility of the probe making contact with water cannot be precluded. For T > LCST, regions of polymer r ich and solvent rich phases form spontaneously, leading to distinct surface features as depicted in Fig. 6 12. However, surface roughness on the collapsed gel was not evaluated quantitatively due to the fact that such phase segregation is random in nature and that quantitative measurements cannot predict the mechanism by which phase transition occurs. As such, qualitative descriptions are sufficient in showcasing the changes induced as the sample is heated above LCST. Characteristics of Force Indentation Behavior on Hydrogel Surfaces General remarks on analyzing force curves with colloidal probe AFM Recalling the major components of a typical force distance curve measured on a hydrogel sample (Fig. 6 14), important information can be derived regarding the molecular details of the surfaces that result in specific, measurable interfacial forces. Without the occurrence of the probe jumping to contact, the determination of the point of contact in Fig. 6 14 (inset) is complicated by a long range repulsive intera ction common to many force curves discussed in the present study. This interaction is characterized by small, but steadily increasing, positive forces over a range of 50 to 60 nm in distance. The relationship between force and indentation within this ran ge of interaction cannot be described by Hertzian theory; thus, a different mechanism must be considered. One possible reason is that this repulsive force originates from the increased load support as the volume fraction of polymer chains under the probe increases upon compression. Long range interaction forces at polymer surfaces arising from molecular structure changes has been extensively studied and thought to influence
153 friction force at polymer brush grafted surfaces probed by AFM [11, 13, 15, 114, 1 15] The range of this interaction is positively correlated with the chain length at the surface [ 13 7 ] As the probe continues to indent, the applied force overcomes the repulsion and tation depth by a Hertzian relationship. Change in interaction forces following phase transition of PNIPAM surface As shown in Fig. 6 8, though the cantilever experienced repulsive force before contact on both the swollen and the collapsed gel, the probe j umped to contact only on the surface of the collapsed gel. Clearly, the interplay between the repulsive and attractive forces upon contact was altered as a result of the phase transition. Many have suggested that the presence of water molecules tightly b ound to ionized surfaces leads to strong repulsion when the separation distance is within a few nanometers [8 10, 13 8 ] This repulsive interaction is likely to arise from the work required to remove the hydration layer surrounding the surfaces and may dom inate the vdW attractive potential in some cases [6, 72] It is possible that the repulsion experienced at low force together with the absence of vdW attraction observed on PNIPAM surface at T < LCST originates from a similar phenomenon, considering that the formation of a stable hydration layer (analogous to bound water at ionized surfaces) is thermodynamically favored in this temperature regime. As the temperature was increased to above the LCST, however, positive free energy of mixing favors the separa tion of water from the polymer chains at the surface. As a result, less work is required to bring the probe to contact with the polymer substrate, effectively increasing the contribution of the vdW attractive force that causes the probe to jump to contact
154 Change in adhesion force upon pull off as a result of phase transition is evident in a representative set of data shown in Fig. 6 10. Because adhesion varies with the area of contact, results in the present study cannot be directly compared with previ ous reports. However, the trend represented here is in good agreement with prior work and is reversible upon cooling to below the LCST  Note that the magnitude of adhesion force reported here is not normalized against the contact area. With the app lied force F being constant for the force curves analyzed in Fig. 6 10, the contact area, characterized by its half width a must be smaller for the surface with higher modulus (i.e. at T > LCST), given by the Hertzian theory below (Equation 6 2). (6 2) If the measured adhesion force, uncorrected for the contact area, was larger at T > LCST, then it is reasonable to believe that the force of adhesion per unit area is much larger compared to that at T < LCST. Recalling the the rmodynamic process by which the phase transition occurs, large adhesion force at the interface above the LCST is likely a result of the exclusion of water from the polymer network, leading to increased interaction between the polymer and the probe. This d ehydration process also gives rise to topographical features and increased surface roughness not seen when the gel surface was highly solvated at T < LCST. Conclu ding Remarks In the present study, the surface mechanical properties of various types of hydro gel thin films was examined, with an emphasis on correlating controllable physical materials, namely, PAAM, PHEMA, P(HEMA co MAA), and PNIPAM, were chosen
155 because they can be synthesized to a specific range of water content desirable and have been used widely in the field of biomaterials. Toward this end, appropriate synthesis methods for fabricating stand alone thin films of hydrogels suitable for surface analysis in AFM were developed, and a set of protocols were established for the quantitative measurements of force indentation behavior using colloidal probe AFM. The results confirmed the hypothesis that for a given gel and at a specific crosslinking density, the elast ic modulus at surface, similar to the bulk, has a power law dependence on the polymer concentration, i.e. inversely related to the water content. This mathematical relationship differs from that predicted by the scaling laws concerning bulk modulus, highl ighting the role of differences in molecular structure and density between the bulk and the surface. Furthermore, comparison between a HEMA MAA copolymer and HEMA homopolymer concludes that chemical composition, in addition to polymer concentration, may a lso influence the measured elastic modulus at surface. To further understand the contribution of molecular structure to the surface elastic modulus of hydrogel films, PNIPAM hydrogel was studied with respect to its environmentally responsive properties as sociated with a characteristic thermodynamic phase transition. Phase segregation and structural changes were observed by comparing the surface elastic modulus, adhesion force, and topographical images of the PNIPAM sample below and above its LCST. Detail ed force curve analysis suggests that the contribution of different interfacial forces is altered when the molecular configuration of the surface changes as a result of thermodynamic phase segregation. Together, the experimental approach developed in the p resent study allowed systematic investigation of the surface mechanical properties of various hydrogel
156 materials. Fundamental relationship between polymer concentration and surface elastic modulus was realized. In addition, study of temperature induced p hase transition of a PNIPAM system lends further insight into the molecular interaction between polymer network and solvent environment, offering future opportunities in probing interfacial phenomena of hydrogel systems in aqueous environment using colloid al probe AFM.
157 Figure 6 1. Schematic of the different types of hydrogel films synthesized in the present study and their respective water content.
158 Figure 6 2. A schematic of the microscopic phase transition at the LCST of PNIPAM. At T > LCST, p olymer chains collapse to form an interconnected structure, and the solvent molecules diffuse out of the original network.
159 Table 6 1. Hydrogel synthesis recipe with the final concentration of each ingredient listed in w/w%. Multiple monomer concentrati ons were prepared for PAAM and PHEMA as an effort to explore the effect of various % WC on surface modulus of hydrogels of the same chemistry. Adapted and modified from [19, 22, 12 3 12 4 13 9 1 40 ] Hydrogel Monomer Initiators Crosslinker APS TEMED SM BS BIS TEGDMA PAAM 5, 15, 35 0.05 0.05 N/A 0.03 N/A P(HEMA co MAA) 45(H) 5(M) 0.5 N/A 0.5 N/A 3 PHEMA 65, 85 0.5 N/A 0.4 N/A 0.1 PNIPAM 12 0.2 0.13 N/A 0.07 N/A
160 poly(acrylamide) poly( 2 hydroxy e thylmethacrylate ) poly( 2 hydroxy ethylmethacrylate co methacrylic acid ) poly( N isopropyl acrylamide) Figure 6 3. Molecular representation of the monomer unit of each hydrogel studied.
161 A B Figure 6 4 Assembly of a liquid sample holder with heating capability ( A ) Side view of the holder indicating the multiple components that ensure proper sealing of the liquid. B) A detailed top view of the h eating and temperature control components Note that t he sample, the thermistor, and the heating element are all immersed in the liquid environment during AFM measurement.
162 A B Figure 6 5. F orce distance curves for three compositions of PAAM hydrogels measured in DI water ( A ) A compilation of for ce curves for the three compositions B) The force curve of the 5% PAAM sample in detail. The difference between 5% and 35% polymer content is reflected by an order of magnitude change in elastic modulus values as described by a Hertzian model. An enlar ged figure of force curve for 5% PAAM reveals the range of forces relevant in measuring surface elastic modulus of a highly solvated hydrogel
163 Figure 6 6. Log log plot of the average elastic modulus vs. polymer concentration of the PAAM samples. The data cor relate well to a 1.93 th power, which is less than the dependence of bulk modulus on polymer concentration. Each error bar indicate s the standard deviation of the fitted modulus values for a given PAAM composition.
164 A B Figure 6 7. Repres entative force curves for P(HEMA co MAA) (45% 5%), 65% PHEMA, and 85% PHEMA measured in DI water. A) A compilation of force curves of the three samples. B) Example data offset applied to allow Hertzian analysis. The Hertzian fit was applied to each set of data for indentation depth beyond the point indicated by the arrow. The point chosen for Hertzian fitting at ap law relationship between F N and in the cont act portion of the force curves
165 Figure 6 8. Force curves of 12% PNIPAM sample below (e.g. RT) and above (e.g. 37 C) the LCST in DI water. Si milar modulus values observed for RT_1 and RT_2 indicate that the effect of the phase transition is reversible.
166 A B Figure 6 9. I nteraction forces close to the surface of PNIPAM A A) A close up illustration of Figure 6 8 for B) Ana tomy of a typical force plot acquired with a colloidal probe. Note the distinct vdW jump to contact behavior following long range repulsion of the force curve acquired at 37 C. A modified version of Fig. 2 12 takes into account phenomena typical to inte rfacial forces on polymeric surfaces. For a neutral surface immersed in DI water, the long range repulsion has a different origin from that described in  which is electrostatic repulsion occurring prior to contact. Here, the probe is in contact with loosely associated polymer chains at the surface and the range of this repulsion is 50 60 nm in indentation.
167 Figure 6 10. Three random sites were sequentially sampled on the surface of PNIPAM below and above its LCST. This representative set of d ata reflects clearly the increase in adhesion following phase transition and the reversible effect after the system was cooled to RT.
168 Figure 6 11. AC mode AFM images of the PNIPAM sample at T < LCST A) The h eight image. B) The correspond ing phase image. C) A 3D rendition of the height image z contrast: 7.4 nm A B C
169 Figure 6 12. AC mode AFM images of the PNIPAM sample at T > LCST A) The height image. B) The corresponding phase image. C) A 3D rendition of the height image. Compared to Fi g. 6 11A, the surface roughened to have a z contrast on the order of 100 nm The phase image here suggests possible variations in mechanical properties in different regions of the surface, a result of the random phase segregation process occurred at LCST. z contrast: 269.0 nm A B C
170 Figure 6 13. Schematic of polymer density profile (D) of a PLL g PEG brush layer with a thickness of L as modeled by the scaling laws. Polymer density gradually decreases as D, the distance away from the grafting site, increases. Adapted from [ 13 4 ]
171 Figure 6 14. Typical force distance curve measured on a hydrogel sample. The point of contact is not clearly defined due to the combination of repulsive interaction and the lack of vdW jump to contact behavior (inset).
172 CHAPTER 7 CONCLUSIONS A ND FUTURE DIRECTIONS Recapitulation The central theme of this doctoral work focuses on uncovering the crucial factors that guide the chemical and physical processes occurring at polymer surfaces, particularly those of the hydrogel materials employed for so ft contact lens applications. A major thrust of research effort in the development of contact lens materials is driven by a number of design challenges that involve the interfacial phenomena of wetting and friction in vivo Particularly, the ability for the lens surface to retain water and maintain a lubricious slidi ng interface is considered the holy grail of the contact lens community. Therefore, the present doctoral research attempted to address these interfacial challenges via the investigation of qu antifiable surface properties using state of the art characterization techniques. Due to their exceptional oxygen permeability achieved at the cost of reduced hydrophilicity (i.e. tear wettability), silicone hydrogel contact lenses demand specific surfac e modifications for their adequate performance in vivo As a result, interrogating the effects of surface modification on various interfacial processes of silicone hydrogel lenses offers perspectives for understanding other similar material systems. Two major aspects of surface properties were examined and reported in this dissertation. They are, namely, the surface chemical composition and the surface elastic modulus of silicone hydrogel lenses. Furthermore, fundamental study involving homogeneous hydr ogel films (i.e. not commercial lens materials) underscored the importance of the molecular structure and dynamic interaction with surrounding medium in predicting elastic modulus at the surface.
173 Surface Uptake of Amphiphilic Copolymer by Silicone Hydrogel Materials The surface chemical properties of commercially available silicone lenses were analyzed with respect to monitoring the changes following the adsorbed block copolymer molecules. Three types of lenses with different surface chemical compositions were examined and the amphiphilic block copolymer, ethylene oxide co butylene oxide (EO BO), was of interest due to its potential usage as a surfactant molecule for improving lens surface wetting and lubrication. X ray photoelectron spectroscopy (XPS) wa s utilized to examine the surface chemical signatures of the lenses prior to and following a solution treatment containing EO BO. Spectroscopic data confirmed the presence of physisorbed EO BO on all lens surfaces; however, the extent of adsorption greatl y differed between the lenses. Such differences were ascribed to the various surface chemical compositions of the lenses. Specifically, those that have undergone plasma treatments, i.e. balafilcon A and lotrafilcon B, designed to improve the hydrophilici ty of silicone materials displayed a greater affinity for EO BO. Nanoscopic tribological testing with the atomic force microscope (AFM) in aqueous environment also corroborated the adsorption of EO BO. It was observed that, following EO BO treatment, the degree of reduction in coefficient of friction measured at the sliding contact between the lens and a 5 m silica probe scaled with the extent of adsorption realized by the XPS data. The strong correlation between the degree of adsorption and the friction reduction implies the potential for modifying lens surface properties via simple practices such as dai ly soaking in lens care solutions. Furthermore this set of experiments offers insight into the possible tribological effects of other molecules relevant to the ocular
174 environment, whether biological (e.g. proteins and lipids) or synthetic (e.g. other sur factant), that may have the ability to alter the surface energy of the lenses. An other in depth study complemented by the analysis of the bulk uptake of EO BO further illustrated the specific interaction model between the copolymer and the lenses. Elution data obtained by the ultra high performance liquid chromatography (UPLC) revealed that the amount of bulk uptake of EO BO differed between lenses in a fashion that was less apparent than the surface adsorption However, together with the XPS data (simila r results as previously described with one additional lens tested), an EO BO molecular concentration gradient was realized for each lens type, with the specific level of concentration being a differentiating factor between different lenses. The ramificat ions of the results are two fold. First, the combined XPS and UPLC findings suggests that the overall interaction between EO BO and a lens material depends upon both the surface and bulk composition. Second, certain surface chemical treatments, such as t he plasma polymerized surface of lotrafilcon B, may lead to a relatively impermeable surface for the penetration of molecules with sizes similar to EO BO. Surface Mechanical Properties of Hydrogel Materials The second part of the dissertation focuses on i nvestigating the correlation between surface molecular structure and the mechanical properties measured with to deformation within the topmost sub micron region, is a s eldom concern for many because elastic modulus is generally considered a bulk property. However, the surface molecular structure of polymeric materials in solution has been shown to be distinctively different from the bulk in terms of parameters such as p olymer concentration [ 13 4 ]
175 giving sufficient reason to believe that disparity may also exist in mechanical properties between the surface and the bulk of the hydrogel The study of silicone hydrogel contact lenses with different surface chemistry and str ucture serves as a springboard for the fundamental discussion of hydrogel surface mechanics. The surface elastic modulus of three types of silicone hydrogel lenses, each with a specific manufacturer applied hydrophilic treatment, were examined with colloi dal probe AFM. The Hertzian contact model was used to determine the modulus of each lens surface under the compression of a silica spherical probe. The comparison of the force indentation plots of delefilcon A, senofilcon A, and balafilcon A demonstrated disparity in both the magnitude of modulus and the mechanism of deformation as a result of the different surface chemical treatments. Particularly, a constant Hertzian behavior was observed for the lenses with homogeneous composition at the micron scale (i.e. senofilcon A) or with a much stiffer surface region (i.e. balafilcon A). Delefilcon A, on the other hand, displayed a depth dependent resistance to deformation, owing to its unique surface gel layer chemically immobiliz ed onto a silicone hydrogle bu lk material The gel layer, along with its underlying anchoring zone, crosslinking density that gives rise to a continuously increasing elastic modulus. The exceedingly low modulus, about two orders of magnitude lower than that of balafilcon A, in the topmost 150 nm of the surface reflects water content greater than 80% by volume. Further indentation causes water to be locally displaced out of the gel network and the effect of underlying stiff substrate (anchoring zone and the hydrogel core) starts to contribute to the load bearing, leading to increased modulus.
176 level design in augmenting the surface properties of the material. Furthermore, such correlation offers opportunities for manufacturing polymeric materials with tunable surface properties for applications other than contact lens. To gain a better understanding of the relationship between synthesis parameters such as water content (i.e. polymer content at equilibrium condition) and the surface elastic modulus, hydrogels of various chemistry and water content were fabricated into films and analyzed with AFM. Not surprisingly, due to the homogeneous structure (at the length scale of AFM indentation experiments) of the films, the Hertzian contact model provided good estimation of the elastic behavior of these hydrogel samples. The magnitude of surface modulus, however, varied as much as one order of magnitude for a water content difference of 15% for a given composition. The dependence of modulus on water content is slightly lower at the surface than in the bulk as predicted by scaling laws, which may be attributed to the difference i n the molecular structure (i.e. lightly crosslinked at the surface vs. heavily crosslinked in the bulk). The influence of the polymer solvent (water in this case) interaction on the measured surface modulus was elucidated by an environmentally responsive hydrogel, poly( N isopropylacrylamide) (PNIPAM). As PNIPAM undergoes phase transition at its lower critical solution temperature (LCST), the polymer phase segregates and forms locally densified polymer chains surrounded by solvent molecules. One order of magnitude of increase in surface elastic modulus was measured when solvated chains collapsed following the phase transition. This finding agrees with a myriad of previous
177 studies, albeit performed at various length scales. Interestingly, changes in inter facial forces experienced by the silica probe upon contacting the surface were also observed. One possible explanation ascribes this phenomenon to the changes in the specific interaction between the polymer and the solvent (i.e. hydrogen bonding between t he amide group in PNIPAM and water molecules), as a result of the changed thermodynamic stability at elevated temperature above the LCST. Due to a lack of previous reports on this behavior, a thorough understanding demands additional experiments that expl ore the possible pathway(s) of such dynamic behavior. Perspectives and Implications Surface chemical and mechanical analyses of silicone hydrogels demonstrated the engineering strategies that can be employed to improve the wetting and friction propertie s of these materials for contact lens applications. Two strategies have been proven to be effective by works included in this dissertation. The first strategy entails the alteration of surface energy by adsorption of an amphiphilic block copolymer via so lution treatment. The adsorption of the copolymer was also seen to facilitate the lubricious sliding of a silica probe across lens surfaces, confirming the effect of surface modification. The second strategy encompasses grafting a high water content, surf ace grafted gel layer to a silicone hydrogel lens resulting in an exceedingly low surface elastic modulus The continuous distribution of chemical composition correlates well with a gradient of stiffness observed within the topmost 1 m of the surface, w arranting the feasibility of producing depth specific properties at the hydrogel surface. The various aspects of the surface properties investigated in this doctoral research, though mostly pertaining to the in vitr o performance of a contact lens,
178 together offer a fundamental depiction of the crucial characteristics of polymeric surfaces relevant in other material systems: The surface of a polymeric material in solution is permeable to the penetration of small molecules, and specific surface and bulk chemi stry can affect the degree of such permeability; At the length scale relevant to the sensitivity of the applied surface techniques, the structure of a polymeric surface is largely diffuse in nature, enabling the manipulation of depth dependent properties; Molecular segments at the surface lack the same degree of crosslinking as the bulk, and thus are free to interact with surrounding environment. The example of interaction with the solvent environment, which subsequently led to the changed surface mechanical property. Future Research The work included in this dissertation presents the most current understanding of the surface chemical and mechanical properties of hydrogel materials. Ther e still remains, however, a number of interesting scientific problems to be investigated using the methodology and experimental approaches established in this dissertation. The bulk uptake behavior of EO BO discussed in Chapter 4 suggests that the retent ion of surface active molecules by the hydrogel is non trivial and may potentially contribute to the conditioning of the lens surface in vivo However, such claim s must be corroborated by detailed analysis of the release mechanism of the molecules over an extended period of time. One experimental approach may be to utilize the UPLC for measuring the amount of EO BO that leaches out of a lens during a given amount of time. Recall that the measurements presented in Chapter 4 resulted from sonicating the le nses to force the release of the EO BO. From a surface science perspective, XPS analysis can be carried out following mechanical treatments under simulated wear conditions to monitor the changes in the amount of the C O species present at lens
179 surface. T he study of the release mechanism of surfactants currently being incorporated in lens care solutions can offer insight into correlating the presence of these molecules to the performance of the lenses in vivo at a time scale relevant to clinical settings. As pointed out at the beginning of the discussion of surface characterization techniques, many surface properties strongly depend upon the specific counterface and testing conditions employed at the time of experiment, with friction coefficient being an ex ample. Most interfaces in vivo do not include the contact of a stiff material such as silica. As a result, it is natural to speculate the changes in friction and contact mechanics of a hydrogel surface when the counterface is more like itself in terms of chemistry and structure Furthermore, a persistent issue encountered in measuring surface mechanical properties of hydrogel materials is the occurrence of interfacial adhesion. A s a problem under constant scrutiny adhesion can confound the calculation of contact forces as it introduces hysteresis to the force displacement plots [71 73] Functionalization of AFM colloidal probe presents a myriad of opportunities to address these counterface related issues. The versatilities offered by chemically grafte d functional groups, or even a thin layer of hydrogel film, can reduce the surface energy of the probe and/or influence the structural contribution to frictional or adhesive contact [11, 13 4 ] In accordance with the depiction of a hydrogel surface consis ting of dangling chains interacting with the solution environment, the discussion of mechanical properties of the surface cannot preclude the influence of this topmost structure. In fact, data presented in Chapter s 5 and 6 revealed a region of surface meas uring
180 approximately 50 nm in thickness that exhibited repulsive interaction with the probe under compression. In the PNIPAM system, changes in the topmost surface structure as a result of the phase transition contributed to the observed disparity in i nter action forces (i.e. from repulsive to jump to contact) upon first contact. Moreover, the extraction of unpolymerized chain s at the surface of a PHEMA sample has been observed to reduce the friction response measured with AFM  It is, therefore reas onable to believe that more in depth provide understanding of the mechanical properties of a hydrogel surface when subjected to low contact pressure (i.e. small indentation depth), which is relevant for interfa ces pr esent in the ocular environment This dissertation has highlighted several important aspects of the surface properties pertinent to the performance of hydrogel materials employed in contact lens applications. Chemical and structural contributions to interfacial phenomena such as adsorption, friction, elasticity, and adhesion at polymeric material surfaces have been deduced with quantitative surface analyses using XPS and AFM. Together, these findings underscore the importance of designing and engi neering functional polymeric surfaces tailored to the specific applications for whic h the materials are employed.
181 APPENDIX A NORMAL FORCE CONSTANT CALIBARTION OF AFM CANTILEVERS Overview The precise calibration of force constants of AFM cantilever rests at the foundation of quantitative force analysis in AFM. This is a demanding task and often susceptible to large errors; as a result, force calibration remains as an active area of research in the AFM community. Two major calibration methods employed in this doctoral work, also represen ting two of the most frequently used methods, are the thermal method, first reported by Hutter and Bechhoefer  and the Sader method [79 81] Here, the theoretical basis, the calculation procedure, and the limitations of each method will be discussed. The two methods will also be compared in terms of the ease of implementation and the degree of uncertainty. The Thermal Method The thermal method models the AFM tip as a simple harmonic oscillator with a spring constant k N which is equivalent to the stiffness of the cantilever. The theory assumes the oscillator fluctuates in response to the thermal noise present. The Hamiltonian of such a system describes the oscillative motion of the cantilever at a given frequency b y the displacement from its equilibrium position q and its momentu m A s such, the spring constant k N is found to scale with these parameters by the following relationship quantified at the resonance frequency of the oscillation f o : where < q 2 > is the rm s displacement of the system, and k B T describes the thermal energy giving rise to the oscillation.
182 In the MFP 3D, the value of < q 2 > is measured as a function of the rms deflection of the cantilever at resonance in free air (i.e. free from the influence of surface interaction forces), which also requires the knowledge of the inverse optical lever sensitivity invOLS (in nm/V) as described in Chapter 2: Here, < 2 of the cantilever as a result of thermal oscillation, which is a function of the quality factor Q f and the resonance frequency. These two values are determined by a power spectrum, which plots the deflection volts against the measured frequency of the oscillation (Fig. A 1). 2 is an instrument determined correction factor. The Sader Method The Sader method determines a rectangular on the following parameters: the plan view dimension, the unloaded resonance frequency, and the quality factor o f the surrounding medium, as suggested by the equation below: where f is the fluid density b is the width of the cantilever L is length of the cantilever Q f is cantilever quality factor measured by the power spectrum, i is the imaginary component o f the hydrodynamic function (which is a function of the radial frequency, f ). The hydrodynamic function depends only on the Reynolds number ( R e ) :
183 where is the viscosity of the fluid. Having calculated the Reynolds number, the corresponding value of the imaginary part of the hydrodynamic function can be obtained based on numeric calculations detailed in [ 14 1 ] Cantilever geometry consideration Due to the geometric constraint of the hydrodynamic function, the Sader method for normal force calibration a s outlined above is only applicable to rectangular cantilevers with aspect ratios (i.e. L/b ) greater than 3. For V shaped cantilevers, therefore, the normal force constants may be calculated by one of two methods. First, if located on the same chip as a rectangular lever, the force constant of a V shaped cantilever can be calculated theoretically based on parameters such as the E and lever thickness h realized from the known value of k N of the rectangular lever via  : For a stand al one V shaped lever, theoretical or experimental values derived from functions involving the Reynolds number are required  Added mass consideration The effect of an added mass at the end of a cantilever on the overall force constant (in both normal and lateral directions) is an important issue to be considered, especially for colloidal probe microscopy technique employed in this doctoral work. Ensuing investigations by Sader et al. suggest that a silica sphere 7 m in diameter virtually did not alter t he measured spring constant of a rectangular cantilever  As a result, for the 5 m silica sphere utilized in the present work, no quantitative adjustment was deemed necessary for the measured spring constant.
184 S ample Calculation s The following calcula tions are based on the dimensions and parameters of an uncoated, rectangular silicon nitride cantilever shown in Fig. 2 18 This particular cantilever has a nominal spring constant of 2.7 N/m as reported by the manufacturer f (air) = 1.18 kg/m 3 (air) = 1.86e 5 kg/m/s b L Q f (air) = 170.7 (resonance frequency of th e cantilever in air) = 143.8 kHz f (radial frequency in air) = 1.868e4 kHz i nvOLS (air) = 108.33 nm/V The Sader Method The corresponding value of calculated with equations presented in [79, 140] is approximately 0.768. Hence, Because the force is usually not applied at the end of the cantilever but at a point off the end where the tip is located, indicated by
185 The Thermal Method Note that this value is different from that reported in Chapter 5 (5.19 N/m) which was also calibrated using the thermal method Deviations in the calculation of force constant using the thermal method are common because the measurement of InvOLS is usually susceptible to instrumental uncertainty. This shortcoming is less apparent with the Sader method because t he values of Q f and f o are more consistent when measured with the power spectrum Discussion Each of the two methods presented clearly demonstrates a set of advantages and disadvantages. For example, the therma l method is indeed quick and simple to use, but is susceptible to the effect of the size and the position of the laser spot, which inevitably affects the measurements of invOLS  On the other hand, the Sader method only relies on the dimensions of th e lever without the need for additional measurements that may be subjected to instrumental noise. However, due to the sensitive dependence (3 rd power) of the spring constant on dimension, cantilever dimensions should be taken with the highest level of pre cision possible, such as with a scanning electron microscope. The overall degree of uncertainty associated with force measurements compounds the uncertainty involved in the calculation of k N as well as the regime of the force of interest relative to the d etection limit of the instrument. For example, in the sub specific process of analog to digital conversion employed by the instrument ( see Fig. 5
186 1 B ). In such a case, the instrumental resolution dominates the uncertainty of the force measurements. As a result, the discussion of uncertainty must incorporate both the inherent detection capability of the instrument and the specific force regimes of interest, with the uncerta inty in the determination of cantilever force constant being a dominating factor in the higher force regime.
187 A B Figure A 1. Calculating normal spring constant using the thermal method. A) A sample power spectrum of a canti lever oscillating in free air. B) An e nlarged figure of the simple harmonic oscillation (SHO) peak. The peak indicates the resonance frequency f o whereas the quality factor Q f is calculated as the peak amplitude at f o divided by the FWHM of a Lorentzian fit to the SHO peak (fitting not shown). The horizontal red line represents the background of the peak.
188 APPENDIX B ANGLE RESOLVED XPS (AR XPS) ANALYSIS OF SILICONE CONTACT LENS MATERIALS FOLLOWING EO BO SOLUTION TREATMENT Overview The AR XPS results reported in Table 4 4 ar e illustrated graphically in detail here. Emphasis is placed on the changes in relative composition observed at approximately 286 eV, corresponding to the C O chemical shift. These changes were analyzed as a function of EO BO treatment at various depths beneath the surface. The depth sensitivity is controlled by the take off angle (TOA), which is defined as the angle between the surface normal and the position of the analyzer. As discussed in Chapter 2, the large r the TOA, the greater the surface sensi tivity of the XPS analysis. Specifically, a TOA of 65 degrees corresponds to a depth of approximately 10 nm beneath the surface, whereas a TOA of 85 correlates to approximately 2 nm of the outermost surface region. The quantification of the relative degr ee of adsorption was discussed in detail in Chapter 4 and will not be repeated here. Comparison of C 1s Spectra Two sets of comparison are made for each lens sample. The first is the comparison between the control and the treated sample at 65 degrees TO A, and the second is the comparison between the control and the treated sample at 85 degrees TOA. If the increase in C O species compared with the control sample is not apparent at 85 degrees, then penetration into the topmost 10 nm of the surface region and the bulk material is evident. If, compared with the control sample, the relative increase in C O signature measured at 85 degrees is similar to that is observed at 65 degrees, then the adsorption is only concentrated within the topmost 2 nm of the sur face. It is
189 important to note here that, as explained in Fig. 2 3, the chemical information derived at 65 degrees TOA is inclusive of that derived at 85 degrees. As such, spectra comparisons for balafilcon A (Fig. B 1) and lotrafilcon B (Fig. B 3) ref lect a concentrated adsorption of EO BO molecules in the topmost 2 nm of the surface, indicative of limited penetration. Senofilcon A (Fig. B 2) and comfilcon A (Fig. B 4) samples, on the other hand, demonstrate the presence of EO BO beneath the 2 nm dete cted at 85 degrees TOA, reflective of the diffusion of EO BO molecules into the surface region and likely the bulk material beneath These results together with those from the UPLC study, give rise to the prediction of the overall interaction m odel betwe en EO BO and each lens type.
190 A B Figure B 1. Spectra comparison s of balafilcon A. A). Control vs. EO BO treated sample measured at 65 TOA. B). Control vs. EO BO treated sample measured at 85 TOA.
191 A B Figure B 2. Spectra comparison s of senofil con A. A). Control vs. EO BO treated sample measured at 65 TOA. B). Control vs. EO BO treated sample measured at 85 TOA.
192 A B Figure B 3. Spectra comparison s of lotrafilcon B. A). Control vs. EO BO treated sample measured at 65 TOA. B). Control vs EO BO treated sample measured at 85 TOA.
193 A B Figure B 4. Spectra comparison s of comfilcon A. A). Control vs. EO BO treated sample measured at 65 TOA. B). Control vs. EO BO treated sample measured at 85 TOA.
194 LIST OF REFERENCES 1. Garbassi, F. Morra, M., Occhiello, E.: Polymer S urfaces: F rom P hysics to T echnology. Wiley, Chichester, England (1998). 2. Attard, G., Barnes, C.: Surfaces. Oxford University Press, USA (1998). 3. J. C. Vickerman: Surface A nalysis: T he P rincipal T echniques. Wiley Chichester, England (1997). 4. Stokes, R.J., Evans, D.F.: Fundament als of Interfacial Engineering. Wiley VCH, New York (1997). 5. Butt, H. J., Graf, K., Kappl, M.: Physics and C hemistry of I nterfaces. Wiley VCH (2003). 6. Raviv, U., Klein, J.: Flui dity of b ound h ydration l ayers. Science. 297, 1540 1543 (2002). 7. Cowley, A.C., Fuller, N.L., Rand, R.P., Parsegian, V.A.: Measurement of repulsive forces between charged phospholipid bilayers. Biochemistry (Mosc.). 17, 3163 3168 (1978). 8. Pashley, R ..: Hydration forces between mica surfaces in aqueous electrolyte solutions. J. Colloid Interface Sci. 80, 153 162 (1981). 9. Pashley, R.., Israelachvili, J..: Dlvo and hydration forces between mica surfaces in Mg2+, Ca2+, Sr2+, and Ba2+ chloride solutio ns. J. Colloid Interface Sci. 97, 446 455 (1984). 10. Israelachvili, J.N., Pashley, R.M.: Molecular layering of water at surfaces and origin of repulsive hydration forces. Nature. 306, 249 250 (1983). 11. Perry, S.S., Yan, X., Limpoco, F.T., Lee, S., M M., Spencer, N.D.: Tribological p roperties of p oly(l lysine) graft poly(ethylene glycol) f ilms: Influence of p olymer a rchitecture and a dsorbed c onformation. Acs Appl. Mater. Interfaces. 1, 1224 1230 (2009). 12. Mller, M.T., Yan, X., Lee, S., Per ry, S.S., Spencer, N.D.: Preferential s olvation and i ts e ffect on the l ubrication p roperties of a s urface b ound, b rushlike c opolymer. Macromolecules. 38, 3861 3866 (2005). 13. Mller, M.T., Yan, X., Lee, S., Perry, S.S., Spencer, N.D.: Lubrication p roper ties of a b rushlike c opolymer as a f unction of the a mount of s olvent a bsorbed within the b rush. Macromolecules. 38, 5706 5713 (2005).
195 14. Drobek, T., Spencer, N.D.: Nanotribology of s urface g rafted PEG l ayers in an a queous e nvironment. Langmuir. 24, 1484 1488 (2008). 15. Raviv, U., Giasson, S., Kampf, N., Gohy, J. F., Jerome R., Klein, J.: Lubrication by charged polymers. Nature. 425, 163 (2003). 16. De Gennes, P.G.: Conformations of p olymers a ttached to an Interface. Macromolecules. 13, 1069 1075 (1 980). 17. De Gennes, P.G.: Polymers at an interface; a simplified view. Adv. Colloid Interface Sci. 27, 189 209 (1987). 18. Holly, F.J., Refojo, M.F.: Wettability of hydrogels I. Poly(2 hydroxyethyl methacrylate). J. Biomed. Mater. Res. 9, 315 326 (197 5). 19. Kou, J., Amidon, G., Lee, P.: pH d ependent s welling and s olute d iffusion c haracteristics of p oly( h ydroxyethyl m ethacrylate co m ethacrylie a cid) h ydrogels. Pharm. Res. 5, 592 597 (1988). 20. De, S.K., Aluru, N.R., Johnson, B., Crone, W.C., Beebe D.J., Moore, J.: Equilibrium swelling and kinetics of pH responsive hydrogels: models, experiments, and simulations. J. Microelectromechanical Syst. 11, 544 555 (2002). 21. Sinniah, S.K., Steel, A.B., Miller, C.J., Reutt Robey, J.E.: Solvent e xclusio n and c hemical c ontrast in s canning f orce m icroscopy. J. Am. Chem. Soc. 118, 8925 8931 (1996). 22. Matzelle, T.R., Ivanov, D.A., Landwehr, D., Heinrich, L.A., Herkt Bruns, C., Reichelt, R., Kruse, N.: Micromechanical p s g s tudies by s canning f orce and s canning e lectron m icroscopy of PNIPAAm. J. Phys. Chem. B. 106, 2861 2866 (2002). 23. Sui, X., Chen, Q., Hempenius, M.A., Vancso, G.J.: Probing the c ollapse d ynamics of p oly(N isopropylacrylamide) b rushes by AFM: Effects of c o nonso lvency and g rafting d ensities. Small. 7, 1440 1447 (2011). 24. Schaefer, J.A., Stucki, F., Frankel, D.J., Gopel, W., Lapeyre, G.J.: Adsorption of H, O, and H2O at Si(100) and Si(111) surfaces in the monolayer range: A combined EELS, LEED, and XPS study. J. Vac. Sci. Technol. B. 2, 359 365 (1984). 25. Wagner, M.S., Horbett, T.A., Castner, D.G.: Characterizing multicomponent adsorbed protein films using electron spectroscopy for chemical analysis, time of flight secondary ion mass spectrometry, and radiol abeling: capabilities and limitations. Biomaterials. 24, 1897 1908 (2003).
196 26. Verdaguer, A., Sacha, G.M., Bluhm, H., Salmeron, M.: Molecular s tructure of Water at i n anometer s cale. Chem. Rev. 106, 1478 1510 (2006). 27. Kiss, ., Samu, J., Tth, A., Bertti, I.: Novel w ays of c ovalent a ttachment of p oly(ethylene oxide) onto p m odification and c haracterization by XPS an d c ontact a ngle m easurements. Langmuir. 12, 1651 1657 (1996). 28. Wang, L. Q., Baer, D.R., Engelhard, M.H., Shultz, A.N.: The adsorption of liquid and vapor water on TiO2(110) surfaces: T he role of defects. Surf. Sci. 344, 237 250 (1995). 29. Hendewerk M., Salmeron, M., Somorjai, G.A.: Water adsorption on the (001) plane of Fe2O3: An XPS, UPS, Auger, and TPD study. Surf. Sci. 172, 544 556 (1986). 30. Briggs, D., Beamson, G.: Primary and secondary oxygen induced C1s binding energy shifts in x ray phot oelectron spectroscopy of polymers. Anal. Chem. 64, 1729 1736 (1992). 31. Briggs, D.: Surface analysis of polymers by XPS and static SIMS. Cambridge University Press (1998). 32. Ratner, B.D., Castner, D.G.: Electrospectroscopy for chemical analysis. Su rface Analysis The Principal Techniques. pp. 43 98. John Wiley & Sons (1997). 33. Schnyder, B., Lippert, T., Ktz, R., Wokaun, A., Graubner, V. M., Nuyken, O.: UV irradiation induced modification of PDMS films investigated by XPS and spectroscopic elli psometry. Surf. Sci. 532 535, 1067 1071 (2003). 34. Ton That, C., Shard, A.G., Teare, D.O.H., Bradley, R.H.: XPS and AFM surface studies of solvent cast PS/PMMA blends. Polymer. 42, 1121 1129 (2001). 35. Sabbatini, L., Zambonin, P.G.: XPS and SIMS surf ace chemical analysis of some important classes of polymeric biomaterials. J. Electron Spectrosc. Relat. Phenom. 81, 285 301 (1996). 36. McArthur, S.L., McLean, K.M., St. John, H.A.W., Griesser, H.J.: XPS and surface MALDI MS characterisation of worn HEM A based contact lenses. Biomaterials. 22, 3295 3304 (2001). 37. Karlgard, C.C.S., Sarkar, D.K., Jones, L.W., Moresoli, C., Leung, K.T.: Drying methods for XPS analysis of PureVision(TM), Focus Night&Day(TM) and conventional hydrogel contact lenses. Appl Surf. Sci. 230, 106 114 (2004). 38. Butt, H. J., Cappella, B., Kappl, M.: Force measurements with the atomic force microscope: Technique, interpretation and applications. Surf. Sci. Reports. 59, 1 152 (2005).
197 39. Kim, S., English, A.E., Kihm, K.D.: Su rface elasticity and charge concentration dependent endothelial cell attachment to copolymer polyelectrolyte hydrogel. Acta Biomater. 5, 144 151 (2009). 40. Domke, J., Radmacher, M.: Measuring the e lastic p roperties of t hin p olymer f ilms with the a tomic f orce m icroscope. Langmuir. 14, 3320 3325 (1998). 41. Kurokawa, T., Gong, J.P., Osada, Y.: Substrate e ffect on t opographical, e lastic, and f rictional p roperties of h ydrogels. Macromolecules. 35, 8161 8166 (2002). 42. Dunn, A.C., Uruea, J.M., Huo, Y., Perry, S.S., Angelini, T.E., Sawyer, W.G.: Lubricity of s urface h ydrogel l ayers. Tribol. Lett. 49, 371 378 (2012). 43. Lin, D.C., Horkay, F.: Nanomechanics of polymer gels and biological tissues: A critical review of analytical approaches in the Hertzian regime and beyond. Soft Matter. 4, 669 (2008). 44. Dimitriadis, E.K., Horkay, F., Maresca, J., Kachar, B., Chadwick, R.S.: Determination of e lastic m oduli of t hin l ayers of s oft m aterial u sing the a tomic f orce m icroscope. Biophys J. 82, 2798 2810 (2002) 45. Cheng, X., Canavan, H.E., Stein, M.J., Hull, J.R., Kweskin, S.J., Wagner, M.S., Somorjai, G.A., Castner, D.G., Ratner, B.D.: Surface c hemical and m echanical p roperties of p lasma p olymerized N i sopropylacrylamide. Langmuir. 21, 7833 7841 (2005). 46 Johnson, K.L.: Contact Mechanics. Cambridge University Press (1987). 47. Efron, N.: Contact L ens P ractice. B utterworth Heinemann/Elsevier, Oxford (2010). 48. Agarwal, A.: Dry Eye: A Practical Guide to Ocular Surface Disorders and Stem Cell Surgery. Slack Incorporated (2006). 49. Tsilimbaris, M.K., Lesniewska, E., Lydataki, S., Le Grimellec, C., Goudonnet, J.P., Pallikaris, I.G.: The u se of a tomic f orce m icroscopy for the o bser vation of c orneal e pithelium s urface. Invest Ophthalmol Vis Sci. 41, 680 686 (2000). 50. Nichols, B., Dawson, C., Togni, B.: Surface features of the conjunctiva and cornea. Invest Ophthalmol Vis Sci. 24, 570 576 (1983). 51. Dohlman, C.H., Friend, J., Kalevar, V., Yagoda, D., Balazs, E.: The glycoprotein (mucu s) content of t ears from normal and dry eye patients. Exp. Eye Res. 22, 359 365 (1976). 52. Moore, J.C., Tiffany, J.M.: Human ocular mucus. Origins and preliminary characterisation. Exp. Eye Res. 29, 291 301 (1979).
198 53. Chao, C. C.W., Vergnes, J. P., Brown, S.I.: Frac tionation and partial characterization of macromolecular components from human ocular mucus. Exp. Eye Res. 36, 139 150 (1983). 54. Gebelein, C.G., Dunn, R.L.: Progress in Biomedical Polymers. Springer (1990). 55. Nace, V.M.: Contrasts in the surface ac tivity of polyoxypropylene and polyoxybutylene based block copolymer surfactants. J. Am. Oil Chem. Soc. 73, 1 8 (1996). 56. Maldonado Codina, C., Morgan, P.B.: In vitro water wettability of silicone hydrogel contact lenses determined using the sessile dr op and captive bubble techniques. J. Biomed. Mater. Res. A. 83A, 496 502 (2007). 57 Li, S., Cao, P., Colorado, R., Yan, X., Wenzl, I., Shmakova, O.E., Graupe, M., Lee, T.R., Perry, S.S.: Local p acking e nvironment s trongly i nfluences the f rictional p rope rties of m ixed CH3 and CF3 t erminated a lkanethiol SAMs on Au(111). Langmuir. 21, 933 936 (2005). 58 Frantz, P., Didziulis, S.V., Merrill, P.B., Perry, S.S.: Spectroscopic and scanning probe studies of oxygen and water on metal carbide surfaces. Tribol. Lett. 4, 141 148 (1998). 59 Shon, Y. S., Lee, S., Perry, S.S., Lee, T.R.: The adsorption of u nsymmetrical s piroalkanedithiols onto g old a ffords m ulti c omponent i nterfaces that a re h omogeneously m ixed at the m olecular l evel. J. Am. Chem. Soc. 122, 1278 1281 (2000). 60 Seah, M.P., Dench, W.A.: Quantitative electron spectroscopy of surfaces: A standard data base for electron inelastic mean free paths in solids. Surf. Interface Anal. 1, 46 55 (1979). 61 Clark, D.T., Abu Shbak, M.M., Brennan, W.J.: Ele ctron mean free paths as a function of kinetic energy: A substrate overlayer investigation of polyparaxylylene ray source. J. Electron Spectrosc. Relat. Phenom. 28, 11 21 (1982). 62 Fakes, D.W., Newton, J.M., Watts, J.F., E dgell, M.J.: Surface modification of a contact lens co polymer by plasma discharge treatments. Surf. Interface Anal. 10, 416 423 (1987). 63. Menzies, K.L., Jones, L.: In v itro a nalysis of the p hysical p roperties of c ontact l ens b lister p ack s olutions. O ptom. Vis. Sci. 88, 493 501 (2011).
199 64. Larson, P.E., Kelly, M.A.: Surface charge neutralization of insulating samples in x ray photoemission spectroscopy. J. Vac. Sci. Technol. Vac. Surfaces Films. 16, 3483 3489 (1998). 65. Briggs, D., Grant, J.T.: Su rface a nalysis by a uger and X Ray p hotoelectron s pectroscopy. IM Publications (2003). 66. Repoux, M.: Comparison of background removal methods for XPS. Surf. Interface Anal. 18, 567 570 (1992). 67. Vgh, J.: The analytical form of the Shirley type back ground. J. Electron Spectrosc. Relat. Phenom. 46, 411 417 (1988). 68. Tougaard, S., Jansson, C.: Comparison of validity and consistency of methods for quantitative XPS peak analysis. Surf. Interface Anal. 20, 1013 1046 (1993). 69. Limpoco, F.T., Payne, J.M., Perry, S.S.: Experimental Considerations when Characterizing Materials Friction with Atomic Force Microscopy. Tribol. Lett. 35, 3 7 (2009). 70. Proksch, R., Schffer, T.E., Cleveland, J.P., Callahan, R.C., Viani, M.B.: Finite optical spot size and position corrections in thermal spring constant calibration. Nanotechnology. 15, 1344 (2004). 71. Heinz, W.F., Hoh, J.H.: Spatially resolved force spectroscopy of biological surfaces using the atomic force microscope. Trends Biotechnol. 17, 143 150 (199 9). 72. Butt, H. J.: Measuring electrostatic, van der Waals, and hydration forces in electrolyte solutions with an atomic force microscope. Biophys. J. 60, 1438 1444 (1991). 73. Burnham, N.A., Dominguez, D.D., Mowery, R.L., Colton, R.J.: Probing the su rface forces of monolayer films with an atomic force microscope. Phys. Rev. Lett. 64, 1931 1934 (1990). 74. Dammer, U., Popescu, O., Wagner, P., Anselmetti, D., Gntherodt, H. J., Misevic, G.N.: Binding strength between cell adhesion proteoglycans measur ed by atomic force microscopy. (1995). 75. Willemsen, O.H., Snel, M.M.E., Cambi, A., Greve, J., De Grooth, B.G., Figdor, C.G.: Biomolecular i nteractions m easured by a tomic f orce m icroscopy. Biophys. J. 79, 3267 3281 (2000). 76. Perry, S.S.: Scanning pr obe microscopy measurements of friction, MRS Bulletin 29, 478 483 (2004).
200 77. Cain, R.G., Biggs, S., Page, N.W.: Force c alibration in l ateral f orce m icroscopy. J. Colloid Interface Sci. 227, 55 65 (2000). 78. Burnham, N.A., Chen, X., Hodges, C.S., Mate i, G.A., Thoreson, E.J., Roberts, C.J., Davies, M.C., Tendler, S.J.B.: Comparison of calibration methods for atomic force microscopy cantilevers. Nanotechnology. 14, 1 6 (2003). 79. Sader, J.E., Chon, J.W.M., Mulvaney, P.: Calibration of rectangular atom ic force microscope cantilevers. Rev. Sci. Instrum. 70, 3967 3969 (1999). 80. Sader, J.E., Pacifico, J., Green, C.P., Mulvaney, P.: General scaling law for stiffness measurement of small bodies with applications to the atomic force microscope. J. Appl. P hys. 97, 124903 124903 7 (2005). 81. Green, C.P., Lioe, H., Cleveland, J.P., Proksch, R., Mulvaney, P., Sader, J.E.: Normal and torsional spring constants of atomic force microscope cantilevers. Rev. Sci. Instrum. 75, 1988 1996 (2004). 82. Ogletree, D .F., Carpick, R.W., Salmeron, M.: Calibration of frictional forces in atomic force microscopy. Rev. Sci. Instrum. 67, 3298 3306 (1996). 83. Pettersson, T., Nordgren, N., Rutland, M.W., Feiler, A.: Comparison of different methods to calibrate torsional sp ring constant and photodetector for atomic force microscopy friction measurements in air and liquid. Rev. Sci. Instrum. 78, 093702 093702 8 (2007). 84. Microelectromechanical Syst. 19, 229 2 38 (2010). 85. Straehla, J.P., Limpoco, F.T., Dolgova, N.V., Keselowsky, B.G., Sawyer, W.G., Perry, S.S.: Nanomechanical p robes of s ingle c orneal e pithelial c ells: s hear s tress and e lastic m odulus. Tribol. Lett. 38, 107 113 (2010). 86. Berdyyeva, T.K., Woodworth, C.D., Sokolov, I.: Human epithelial cells increase their rigidity with ageing in vitro: direct measurements. Phys. Med. Biol. 50, 81 92 (2005). 87. Rico, F., Roca Cusachs, P., Sunyer, R., Farr, R., Navajas, D.: Cell dynamic adhesion and elas tic properties probed with cylindrical atomic force microscopy cantilever tips. J. Mol. Recognit. 20, 459 466 (2007). 88. Lin, D., Shreiber, D., Dimitriadis, E., Horkay, F.: Spherical indentation of soft matter beyond the Hertzian regime: numerical and e xperimental validation of hyperelastic models. Biomech. Model. Mechanobiol. 8, 345 358 (2009).
201 89. Pasche, S., De Paul, S.M., Vrs, J., Spencer, N.D., Textor, M.: Poly(l lysine) graft poly(ethylene glycol) a ssembled m onolayers on n iobium o xide s urfaces: A q uantitative s tudy of the i nfluence of p olymer i nterfacial a rchitecture on r esistance to p rotein a dsorption by ToF SIMS and in s itu OWLS. Langmuir. 19, 9216 9225 (2003). 90. Kenausis, G.L., Voros, J., Elbert, D.L., Huang, N., Hofer, R., Ruiz Taylor, L., Textor, M., Hubbell, J.A., Spencer, N.D.: Poly(l lysine) g p oly(ethylene glycol) l ayers on m etal o xide s urfaces: a ttachment m echanism and e ffects of p olymer a rchitecture on r esistance to p rotein a dsorption. J. Phys. Chem. B. 104, 3298 3309 (2000). 91. Garcia, R., Magerle, R., Perez, R.: Nanoscale compositional mapping with gentle forces. Nat Mater. 6, 405 411 (2007). 92. Scott, W.W., Dehaemer, M.J.: Metals Handbook: Engineered Materials Handbook. ASM International (2001). 93. Guryca, V., Hobzov, R., Prdn, M., Sirc, J., Michlek, J.: Surface morphology of contact lenses probed with microscopy techniques. Contact Lens Anterior Eye. 30, 215 222 (2007). 94. Lpez Alemany, A., Compa, V., Refojo, M.F.: Porous structure of Purevision (TM) versus Foc us Night&Day (TM) and conventional hydrogel contact lenses. J. Biomed. Mater. Res. 63, 319 325 (2002). 95. Gonzlez Mijome, J.M., Lpez Alemany, A., Almeida, J.B., Parafita, M.A., Refojo, M.F.: Microscopic observations of superficial ultrastructure of unworn siloxane hydrogel contact lenses by cryo scanning electron microscopy. J. Biomed. Mater. Res. B Appl. Biomater. 76B, 419 423 (2006). 96. Subbaraman, L.N., Glasier, M. A., Senchyna, M., Sheardown, H., Jones, L.: Kinetics of i n v itro l ysozyme d eposi tion on s ilicone h ydrogel, PMMA, and FDA g roups I, II, and IV c ontact l ens m aterials. Curr. Eye Res. 31, 787 796 (2006). 97. Fonn, D.: Targeting c ontact l ens i nduced d ryness and d iscomfort: What p roperties w ill m ake l enses m ore c omfortable. Optom. Vis. S ci. 84, 279 285 (2007). 98. Santos, L., Rodrigues, D., Lira, M., Oliveira, M.E.C.D.R., Oliveira, R., Vilar, E.Y. P., Azeredo, J.: The influence of surface treatment on hydrophobicity, protein adsorption and microbial colonisation of silicone hydrogel con tact lenses. Contact Lens Anterior Eye. 30, 183 188 (2007). 99. Nicolson, P.C.: Continuous w ear c ontact l ens s urface c hemistry and w earability. Eye Contact Lens Sci. Clin. Pr. January 2003. 29, (2003).
202 100. Daniels, K.: Contac lens induced dry eye. A Pra ctical Guide to Ocular Surface Disorders and Stem Cell Surgery. p. 324. Slack Incorporated, Thorofare, NJ (2006). 101. Weikart, C.M., Matsuzawa, Y., Winterton, L., Yasuda, H.K.: Evaluation of plasma polymer coated contact lenses by electrochemical impedan ce spectroscopy. J. Biomed. Mater. Res. 54, 597 607 (2001). 102. Teichroeb, J.H., Forrest, J.A., Ngai, V., Martin, J.W., Jones, L., Medley, J.: Imaging p rotein d eposits on c ontact l ens m aterials. Optom. Vis. Sci. 85, 1151 1164 (2008). 103. Cheng, L., Mul ler, S.J., Radke, C.J.: Wettability of silicone hydrogel contact lenses in the presence of tear film components. Curr. Eye Res. 28, 93 108 (2004). 104. Svitova, T.F., Lin, M.C.: Tear l ipids i nterfacial r heology: Effect of l ysozyme and l ens c are s olutions. Optom. Vis. Sci. 87, 10 20 (2010). 105. Ketelson, H.A., Meadows, D.L., Stone, R.P.: Dynamic wettability properties of a soft contact lens hydrogel. Colloids Surf. B Biointerfaces. 40, 1 9 (2005). 106. Maldonado Codina, C., Morgan, P.B., Efron, N., Canry J. C.: Characterization of the s urface of c onventional h ydrogel and silicone hydrogel contact lenses by time of flight secondary ion mass spectrometry : Optometry & Vision Science. Optom. Vis. Sci. 81, 455 460 (2004). 107. Opdahl, A., Koffas, T.S., Amit ay Sadovsky, E., Kim, J., Somorjai, G.A.: Characterization of polymer surface structure and surface mechanical behaviour by sum frequency generation surface vibrational spectroscopy and atomic force microscopy. J. Phys. Condens. Matter. 16, R659 R677 (2004 ). 108. Kim, S.H., Opdahl, A., Marmo, C., Somorjai, G.A.: AFM and SFG studies of pHEMA based hydrogel contact lens surfaces in saline solution: adhesion, friction, and the presence of non crosslinked polymer chains at the surface. Biomaterials. 23, 1657 1 666 (2002). 109. Gonzlez Mijome, J.M., Lpez Alemany, A., Almeida, J.B., Parafita, M.A., Refojo, M.F.: Microscopic observation of unworn siloxane hydrogel soft contact lenses by atomic force microscopy. J. Biomed. Mater. Res. B Appl. Biomater. 76B, 412 418 (2006). 110. Rennie, A.C., Dickrell, P.L., Sawyer, W.G.: Friction coefficient of soft contact lenses: measurements and modeling. Tribol. Lett. 18, 499 504 (2005). 111. Kim, S.H., Marmo, C., Somorjai, G.A.: Friction studies of hydrogel contact lenses using AFM: non crosslinked polymers of low friction at the surface. Biomaterials. 22, 3285 3294 (2001).
203 112. Bain, C.D., Whitesides, G.M.: Attenuation lengths of photoelectrons in hydrocarbon films. J. Phys. Chem. 93, 1670 1673 (1989). 113. Lorentz, H., R ogers, R., Jones, L.: The i mpact of l ipid on c ontact a ngle w ettability. Optom. Vis. Sci. 84, 946 953 (2007). 114. Brady, M.A., Limpoco, F.T., Perry, S.S.: Solvent d ependent f riction f orce r esponse of p oly(ethylenimine) graft poly(ethylene glycol) b rushes i nvestigated by a tomic f orce m icroscopy. Langmuir. 25, 7443 7449 (2009). 115. Raviv, U., Frey, J., Sak, R., Laurat, P., Tadmor, R., Klein, J.: Properties and i nteractions of p hysigrafted e nd f unctionalized p oly(ethylene glycol) l ayers. Langmuir. 18, 7482 7495 (2002). 116. Thai, L.C., Tomlinson, A., Simmons, P.A.: In vitro and in vivo effects of a lubricant in a contact lens solution. Ophthalmic Physiol. Opt. 22, 319 329 (2002). 117. Stephen Tonge, Lyndon Jones, Scott Goodall, Brian Tighe: The ex vivo wet tability of soft contact lenses Curr. Eye Res. 23, 51 59 (2001). 118. Yang, Y. W., Deng, N. J., Yu, G. E., Zhou, Z. K., Attwood, D., Booth, C.: Micellization of d iblock and t riblock c opolymers in a queous s olution. New r esults for o xyethylene/ o xybutylene c opolymers E38B12 and E21B11E21. Comparison of o xyethylene/ oxybutylene, o xyethylene/ o xypropylene, and o xyethylene/ a lkyl s ystems. Langmuir. 11, 4703 4711 (1995). 119. Huo, Y., Rudy, A., Wang, A., Ketelson, H., Perry, S.S.: Impact of e thylene o xide b utylen e o xide c opolymers on the c omposition and f riction of s ilicone h ydrogel s urfaces. Tribol. Lett. 45, 505 513 (2012). 120. Fornasiero, F., Krull, F., Prausnitz, J., Radke, C.: Steady state diffusion of water through soft contact lens materials. B omaterials 26, 5704 5716 (2005). 121. Lee, S.J., Bourne, G.R., Chen, X., Sawyer, W.G., Sarntinoranont M.: Mechanical characterization of contact lenses by microindentation: Constant velocity and relaxation testing Acta Biomater. 4, 1560 1568 (2008). 12 2 Hutter, J.L., Bechhoefer, J.: Calibration of atomic force microscope tips. Rev. Sci. Instrum. 64, 1868 1873 (1993). 12 3 Dalton, P.D., Flynn, L., Shoichet, M.S.: Manufacture of poly(2 hydroxyethyl methacrylate co methyl methacrylate) hydrogel tubes for use as ne rve guidance channels. Biomaterials. 23, 3843 3851 (2002).
204 12 4 Yeung, T., Georges, P.C., Flanagan, L.A., Marg, B., Ortiz, M., Funaki, M., Zahir, N., Ming, W., Weaver, V., Janmey, P.A.: Effects of substrate stiffness on cell morphology, cytoskeletal stru cture, and adhesion. Cell Motil. Cytoskeleton. 60, 24 34 (2005). 12 5 Christensen, L.H., Breiting, V.B., Aasted, A., J??rgensen, A., Kebuladze, I.: Long t erm e ffects of p olyacrylamide h ydrogel on h uman b reast t issue. Plast. Reconstr. Surg. 111, 1883 1890 (2003). 12 6 Khare, A.R., Peppas, N.A.: Swelling/deswelling of anionic copolymer gels. Biomaterials. 16, 559 567 (1995). 12 7 Burmistrova, A., Richter, M., Uzum, C., Klitzing, R. v: Effect of cross linker density of P(NIPAM co AAc) microgels at solid sur faces on the swelling/shrinking 624 (2011). 12 8 Junk, M.J.N., Berger, R., Jonas, U.: Atomic f orce s pectroscopy of t hermoresponsive p hoto c ross l inked h ydrogel f ilms. Langmuir. 26, 7262 7269 (2010). 12 9 Muniz, E.C., Geuskens, G.: Compressive e lastic m odulus of p olyacrylamide h ydrogels and s emi IPNs with p oly(N isopropylacrylamide). Macromolecules. 34, 4480 4484 (2001). 1 30 Winnik, F.M., Ringsdorf, H., Venzmer, J.: Methanol water as a co no nsolvent system for poly(N isopropylacrylamide). Macromolecules. 23, 2415 2416 (1990). 13 1 Koffas, T.S., Amitay Sadovsky, E., Kim, J., Somorjai, G.A.: Molecular composition and mechanical properties of biopolymer interfaces studied by sum frequency gener ation vibrational spectroscopy and atomic force microscopy. J. Biomater. Sci. -Polym. Ed. 15, 475 509 (2004). 13 2 Boudou, T., Ohayon, J., Picart, C., Tracqui, P.: An extended relationship for the of tunable polyacrylamide gels. Biorheology. 43, 721 728 (2006). 13 3 Engler, A., Bacakova, L., Newman, C., Hategan, A., Griffin, M., Discher, D.: Substrate c ompliance versus l igand d ensity in c ell on g el r esponses. Biophys. J. 86, 617 628 (2004). 13 4 H euberger, M., Drobek, T., Spencer, N.D.: Interaction f orces and m orphology of a p rotein r esistant p oly(ethylene glycol) l ayer. Biophys. J. 88, 495 504 (2005). 13 5 Shibayama, M., Morimoto, M., Nomura, S.: Phase s eparation i nduced m echanical t ransition of p oly(N isopropylacrylamide)/ water i sochore g els. Macromolecules. 27, 5060 5066 (1994).
205 13 6 Crowther, H.M., Vincent, B.: Swelling behavior of poly N isopropylacrylamide microgel particles in alcoholic solutions. Colloid Polym. Sci. 276, 46 51 (1998). 13 7 Kidoaki, S., Nakayama, Y., Matsuda, T.: Measurement of the Interaction Forces between Proteins and Iniferter Based Graft Polymerized Surfaces with an Atomic Force Microscope in Aqueous Media. Langmuir. 17, 1080 1087 (2001). 13 8 LeNeveu, D.M., Rand, R.P ., Parsegian, V.A.: Measurements of forces between lecithin bilayers. Nature. 259, 601 603 (1976). 13 9 Pinchuk, L., Eckstein, E.C., Van De Mark, M.R.: Effects of low levels of methacrylic acid on the swelling behavior of poly(2 hydroxyethyl methacrylate) J. Appl. Polym. Sci. 29, 1749 1760 (1984). 1 40 Cook, A.D., Sagers, R.D., Pitt, W.G.: Bacterial adhesion to poly(HEMA) based hydrogels. J. Biomed. Mater. Res. 27, 119 126 (1993). 14 1 Sader, J.E.: Frequency response of cantilever beams immersed in visc ous fluids with applications to the atomic force microscope. J. Appl. Phys. 84, 64 76 (1998).
206 BIOGRAPHICAL SKETCH Yuchen elementary and middle schools ited States in 2002. She then settled in New Port Richey, Florida, where she comp leted her secondary education at J.W. Mitchell High School in 2005. In the same year, she began her study at the University of Florida majoring in materials science and eng ineering (MSE). She explored a number of research topics during her four years of undergraduate career, including single crystal silicon material heat treatment for semiconductor applications and the fundamental study of collagen fibrils for in vitro biom ineralization process. In 2007, she spent a summer at a n REU program hosted by the Georgia Institute of Technology, and investigate d the dynamic mechanical properties and processing of bacteria derived polymeric fibers. Two year s later, she received her Bachelor of Science degree in MSE with the honor of summa cum laude She began her doctoral work in late 2009 at the Department of MSE at UF under the supervision of Professor Scott Perry. Since then, her work has primarily focuse d on interrogating the surface chemical, mechanical, and tribological properties of contact the Association for Academic Women at UF, an honor given to a total of two fema le students campus wide for their excellence in doctoral research.