You are on page 1of 218

MULTIFUNCTIONAL MEDICAL DEVICES BASED ON PH-SENSITIVE HYDROGELS FOR CONTROLLED DRUG DELIVERY

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in the Graduate School of The Ohio State University

By Hongyan He, M.S.

The Ohio State University 2006

Doctors Examination Committee: Approved by Professor L. James Lee, Adviser Professor Kurt W. Koelling Professor Robert J. Lee ______________________________ Adviser Graduate Program in Chemical Engineering

ABSTRACT

Hydrogels are a desired material for biomedical and pharmaceutical applications due to their unique swelling properties and highly hydrated structure. To better control the synthesized hydrogels for various applications, it is necessary to have a thorough understanding of hydrogel structure and reaction mechanism. In this study, pH-sensitive hydrogel networks consisting of methacrylic acid (MAA) crosslinked with tri(ethylene glycol) dimethacrylate (TEGDMA) were synthesized by free-radical

photopolymerization in the water/ethanol mixture with different ratios under various light intensity. Reaction rate was measured using Photo-Differential Scanning Calorimetry (PhotoDSC) with a modified sample pan designed for handling volatile reagents. A photo-rheometer and a dynamic light scattering (DLS) goniometer were used to follow the changes in viscosity and molecule size of the resin system during

photopolymerization. It was found that the rate of polymerization increased and more compact and less swelling gels would form with a higher water fraction in 50wt% solvent/reactant mixture. This is because the weaker interaction between MAA and solvent gives a higher opportunity for propagation and a higher reaction rate. The hydrophobic TEGDMA and initiator tend to form aggregates in the solution with a higher water content, contributing to the inhomogeneous microgel formation. It was also noted that the rate of polymerization and the MAA conversion were enhanced as the light intensity increased. However, at too high a light intensity, an adverse effect was observed
ii

and the final conversion of MAA decreased to 43% at 24 mw/cm2. The optimal light intensity was about 2.0 mw/cm2 to get the PMAA gels with low residue monomers. The use of the high light intensity significantly shortened the reaction time to reach the macro-gelation and increased the swelling ratio of formed hydrogels, which can be explained by the mechanism of intra- and intermolecular reaction. By using the desired functional hydrogels, several drug delivery systems were developed based on the selected integration of a number of micro-manufacturing modules such as soft-lithography, micro-imprinting, and polymer self-folding, to achieve multi-functionalities such as drug protection, self-regulated oscillatory release, enhanced mucoadhesion, and targeted unidirectional release. To evaluate the device performance, adhesion measurement, dynamic flow testing, and targeted unidirectional release were conducted for trans-luminal delivery of two model drugs, acid orange 8 and bovine serum albumin. The self-folding device first attached to the mucosal surface and then curled into the mucus, leading to enhanced mucoadhesion in the mode of grabbing. Furthermore, the folded layer served as a diffusion barrier, minimizing the drug leakage in the small intestine. The resulting unidirectional release provides improved drug transport through the mucosal epithelium due to localized high drug concentration. The functionalities of the devices have been successfully demonstrated in vitro using a porcine small intestine. The novel delivery devices will be of great benefit to the advancement of oral administration of proteins and DNAs. Since the mucus layer covers many tissues at other specific sites, the devices may be applied for ocular, buccal, vaginal and rectal administrations. The polymer self-folding at the microscale can also be applied as probe arrays for bio/chemical sensing, carriers in cell-based bioreactors, and tissue clamping.
iii

This dissertation is dedicated to my parents

iv

ACKNOWLEDGMENTS

I would like to express my great appreciation to my adviser, Dr. L. James Lee, for his inspiring guidance, encouragement, and support throughout this work. I would

also like to acknowledge with sincere gratitude to the members of my dissertation and candidacy exam committee, Dr. Kurt W. Koelling, Dr. Robert J. Lee, and Dr. James F. Rathman for their valuable suggestions and comments on my work. My gratitude is also expressed to Dr. Paula Stevenson, Paul Green, Karl Scott, and Leigh Evrard for their great help in my research work. Special thanks go to my

fellow colleagues Dr. Xia Cao, Dr. Jingjiao Guan, and all other polymer research group members, for their invaluable help and technical support. Finally, I would like to thank my parents for their forever support through the years of my study and my husband, Zhaohui Ning, for his understanding, support, and encouragement.

VITA

January 2nd, 1974........................................................Born - Taiyuan, Shanxi, P. R. China September 1992July 1997..........................................B.S. Chemical Engineering Tsinghua University Beijing, P. R. China September 1997March 2000......................................M.S. Environmental Engineering Shanghai University Shanghai, P. R. China September 2000December 2004.............................Graduate Research Associate The Ohio State University Columbus, OH June 2005present.....................................................Presidential Fellow The Ohio State University Columbus, OH

PUBLICATIONS 1. H. He, L. Li and L. J. Lee, Photopolymerization and structure formation of methacrylic acid based hydrogels in water/ethanol mixture, Polymers, 47, 1612-1619, 2006. 2. H. He, J. Guan, and L.J. Lee, Oral Delivery Devices Based on Self-folding Hydrogels, Journal of Controlled Release, 110(2), 339-346, 2006. 3. J. Guan, H. He, D.J. Hansford and L. J. Lee, Self-folding Hydrogel Three-Dimensional Microstructures, Journal of Physics Chemistry B, 109(49), 23134-23137, 2005.
vi

4. H. He, J. Guan, D.J. Hansford and L.J. Lee, Hydrogel-Based Multifunctional Delivery Devices for Oral Protein Administration, Abstracts of Papers PMSE-016, 229th ACS National Meeting, San Diego, CA, March 13-17, 2005. 5. H. He, J. Guan, D.J. Hansford and L.J. Lee, Hydrogel-Based Multifunctional Delivery Devices for Oral Protein Administration, Polymeric Materials: Science and Engineering, 92, 28-30, 2005. 6. H. He and L. J. Lee, Poly(lactic-co-glycolic Acid) and Functional Hydrogels for Drug Delivery Applications, Proceedings of Society of Plastics Engineers, 62(3), 3356-3360, 2004. 7. H. He, X. Cao and L. J. Lee, Design of a Novel Hydrogel-based Intelligent System for Controlled Drug Release, Journal of Controlled Release, 95, 391-402, 2004. 8. H. He and J. Wei, Synthesis and Properties of Modified Melamine Resin, Shanghai Huanjing Kexue, 19(9), 432-433, 2000. 9. H. He, J. Wei and G. Zhang, Synthesis of Modified Melamine-Formaldehyde Resin and Property Investigation as a Flocculent, Shanghai Daxue Xuebao, V3, 2000. 10. H. He, Synthesis of Modified Melamine-Formaldehyde Resin and Property Investigation, Master Thesis, Shanghai University, China, 2000. 11. H. He, The Extraction of Glycin from Proteins, Bachelor Thesis, Tsinghua University, China, 1997.

FIELDS OF STUDY

Major Field: Chemical Engineering Minor Field: Polymer Engineering

vii

TABLE OF CONTENTS

Page Abstract...... ii Acknowledgments.......... v Vita..vi Table of contents...viii List of tables..xii List of figures....xiii Chapters:

1.

Introduction and motivation .. 1

2.

Literature review.8

2.1

Overview of pH-sensitive hydrogels......8 2.1.1 2.1.2 Anionic hydrogels ...10 Cationic hydrogels ..12

2.2

Temperature-sensitive hydrogels .14 2.2.1 2.2.2 Negatively temperature-sensitive gels 15 Positively temperature-sensitive gels ..19

2.3

Properties of hydrogels.20 2.3.1 2.3.2 Swelling properties .20 Network structure and characterization .22
viii

2.3.3 2.4

Mechanical properties .30

Application of hydrogels in drug delivery ...33 2.4.1 Peroral drug delivery ...34 2.4.1.1 2.4.1.2 2.4.2 2.4.3 2.4.4 2.4.5 2.4.6 Buccal route..34 Gastrointestinal route36

Nasal route ..42 Ocular route .43 Rectal and vaginal routes ....44 Transdermal route ....45 Trends and perspectives....46

3.

Photopolymerization and structure formation of PMAA hydrogels in water/ethanol

mixture. .49

3.1 3.2

Introduction...50 Experimental.52 3.2.1 3.2.2 3.2.3 3.2.4 3.2.5 3.2.6 3.2.7 Materials and sample preparation...52 Modification of DSC pans .....53 PhotoDSC measurement ...55 Rheological measurement..55 Dynamic light scattering analysis..56 Swelling studies.....57 Scanning electron microscopy characterization.....57

3.3

Results and discussions.....58 3.3.1 3.3.2 3.3.3 3.3.4 Kinetics of MAA/TEGDMA photopolymerization ......58 Viscosity measurement and molecule size analysis ..63 Mechanism for gelation ....67 Swelling ratio and structural characterization....72

3.4

Conclusions..77
ix

4.

Photopolymerization and structure formation of PMAA hydrogels cured under

various light intensities.....78

4.1 4.2

Introduction ..79 Experimental ....81 4.2.1 4.2.2 4.2.3 4.2.4 4.2.5 Materials and sample preparation...81 PhotoDSC measurement ....82 Rheological measurement...83 Dynamic light scattering analysis...83 Swelling studies..84

4.3.

Results and discussion......84 4.3.1 4.3.2 4.3.3 4.3.4 4.3.5 Kinetics of MAA/TEGDMA photopolymerization ......84 Viscosity measurement .....89 Kinetic parameters ....92 Molecular size analysis .97 Integrated analysis.....99

4.4

Conclusions....106

5.

Design of smart devices based on the functional hydrogels...107

5.1 5.2

Introduction ....108 Experimental ......111 5.2.1 5.2.2 5.2.3 5.2.4 5.2.5 Materials...111 Device design and drug loading ...113 In vitro drug release ....115 Diffusion studies .. ..116 Targeted unidirectional release....116

5.3.

Results and discussion........119 5.3.1 5.3.2 Swelling properties of hydrogels .119 Model drug release from entrapped devices ....121
x

5.3.3 5.3.4

Diffusion studies ..126 Model drug release from assembled devices....130 5.3.4.1 Drug protection.130 5.3.4.2 Self-regulated oscillatory release .137 5.3.4.3 Targeted unidirectional release 137

5.4

Conclusions.....141

6.

An oral delivery device based on the self-folding hydrogels.....142

6.1 6.2

Introduction....143 Experimental.......144 6.2.1 6.2.2 6.2.3 6.2.4 6.2.5 Materials......144 Device design and fabrication .....145 Swelling and self-folding studies 150 Mucoadhesion measurement .......151 Delivery performance..........................................................153

6.3

Results and discussion....154 6.3.1 6.3.2 6.3.3 Swelling and self-folding studies .....154 Mucoadhesion measurement 158 Delivery performance ..165

6.4

Conclusions.169

7.

Conclusions and recommendations.170

7.1 7.2

Conclusions.170 Recommendations...172

References177

xi

LIST OF TABLES

Table

Page

5.1

Physical properties of model drugs..................................................................112

5.2

Permeability and diffusion coefficient of model drugs through different membranes......................................................129

xii

LIST OF FIGURES

Figure

Page

1.1

The engineering process applied for pH-sensitive hydrogels .....6

2.1

Structures of anionic pH-sensitive hydrogels ...10

2.2

Structures of negative temperature-sensitive hydrogels ...15

3.1

(A) DSC pan treated with PDMS; (B) Seal of DSC pan...54

3.2

Comparison of PhotoDSC measurements by using a modified and an un-modified pan at UV intensity of 2.0 mw/cm2 in the MAA/TEGDMA system (1.0 mole%TEGDMA, 50 wt.% solvent

mixture of the 1/1 water/ethanol ratio) .60

3.3

(A) Reaction rate and (B) conversion versus reaction time for the isothermal photopolymerization of MAA/TEGDMA mole%TEGDMA, 50 wt.% solvent) with different solvent compositions at 30C and UV intensity of 2.0 mW/cm2......62

xiii

3.4

(A) Reaction rate and viscosity rise as a function of conversion of MAA/TEGDMA (mole% TEGDMA, 50 wt.% solvent) with different solvent compositions cured at UVintensity of 2.0 mW/cm2, (B) Gel time and gel conversion versus water/ethanol ratio in the solvent mixture .....65

3.5

The size distribution of MAA/TEGDMA resin (1.0 %TEGDMA, 50 wt.% solvent) with different solvent ratios of water/ethanol: (A) 1/4 and (B) 9/1 cured at light intensity of 2.0 mW/cm2........66

3.6

The schematic diagram of structure formation of MAA/TEGDMA with different solvent qualities .....68

3.7

The size distribution of MAA/TEGDMA monomer solution (1.0 %TEGDMA, 50 wt.% solvent) with different compositions 70

3.8

Equilibrium swelling ratios of the PMAA (1.0 mole% TEGDMA) hydrogels with

different solvent ratios as a function of pH values .......74

3.9

SEM micrograph of swollen PMAA hydrogels (1.0 mole% TEGDMA, 50 wt.% solvent)

with different swelling ratios (SR) in pH=7.4 buffer solution: (A) 9/1 and (B) 1/4...75

3.10

SEM micrograph of swollen PMAA hydrogels


xiv

(1.0

mole% TEGDMA, 50 wt.% solvent) with

the same swelling ratio (SR=4.3) in different buffer solution: (A) 9/1 in pH=6.2 buffer (B) 1/4 in pH=3.0 buffer......76

4.1

Reaction rate vs. conversion of MAA/TEGDMA in the presence of 1% Irgacure 651 with 50 and 100 wt.% monomer content cured under 5.0 mw/cm2...86

4.2

Effect of light intensity on the polymerization of MAA/TEGDMA system in the presence of 1% Irgacure 651 (A) reaction rate, (B) conversion ...88

4.3

Reaction rate and relative viscosity rise as a function of conversion of MAA/TEGDMA (1.0 mole% TEGDMA, 50 wt.% solvent)

cured under different light intensity: (A) 0.25 and 2.0 mW/cm2, (B) 24 mW/cm2 ......90

4.4

Gel conversion versus light intensity for polymerization of MAA/TEGDMA system (1.0 mole% TEGDMA, 50 wt.% solvent)

in the presence of 1% Irgacure 651........91

4.5

Conversion dependence of the rate constant of propagation k p and termination k t for the polymerization of MAA/TEGDMA system at 2.0 mw/cm2......95

4.6

Conversion dependence of the rate constant of propagation k p and termination k t for the polymerization of MAA/TEGDMA system at 24 mw/cm2........96
xv

4.7

The molecular size distribution of the MAA/TEGDMA system (1.0% TEGDMA, 50 wt.% solvent) cured at (A) 2.0 mw/cm2 and (B) 24 mw/cm2...98

4.8

Changes of reaction rate, viscosity during the photopolymerization of MAA/TEGDMA at light intensity of 2.0 mw/cm2: I initiation; II microgel formation; III cluster formation; IV macro-gelation; V post-gelation.....100

4.9

Changes of reaction rate, viscosity during the photopolymerization of MAA/TEGDMA at light intensity of 24 mw/cm2: I initiation; II microgel formation; III cluster formation; IV macro-gelation; V post-gelation.....101

4.10

Dynamic swelling behavior of the PMAA hydrogels with 1.0% TEGDMA cured at different light intensity and immersed in the different pH buffer solutions..105

5.1

Schematic of the assembled device .118

5.2

Dynamic swelling behavior of hydrogels. Samples were 5.0 mm in diameter and 0.8 mm in thickness: ( ( ( ( ) PMAA hydrogel in pH=7.3 buffer. ) PMAA hydrogel in pH=3.0 buffer. ) PHEMA hydrogel in pH=7.3 buffer. ) PHEMA hydrogel in pH=3.0 buffer...120

xvi

5.3

Acid Orange 8 release to pH 7.3 buffer solution from the entrapped 5.0 mm PMAA samples at 25 C. The samples were 0.8 mm in thickness ..123

5.4

BSA release to pH 7.3 buffer solution from the entrapped 5.0 mm PMAA samples at 25 C. The samples were 0.8 mm in thickness ......124

5.5

AO8 and BSA release from the entrapped 5.0 mm PMAA samples at 25 C. The samples were 0.8 mm in thickness: ( ( ( ) AO8 at pH=3.0. ) AO8 at pH=7.3. ) BSA at pH=7.3.......125

5.6

Permeation of AO8 and BSA through different swollen hydrogel membranes at pH 7.3 and 25 C. ( ( ( ) AO8 through PMAA. ) BSA through PMAA. ) AO8 through PHEMA............128

5.7

AO8 release from the assembled device at pH=7.3 and 25C. The diameter of the device is 5.0 mm. The thickness of bilayered gate is 60 m and the thickness of the drug reservoir is 1.0 mm. (A) Dry assembled device. (B) Releasing at t= 40 minutes. (C) Released at t= 80 minutes. (D) Schematic of AO8 release from assembled device.....132

5.8

AO8 release from the 5.0 mm assembled devices


xvii

with different gates at pH=3.0 and 25C. The gate thickness is 60 m and the reservoir thickness is 1.0mm. ( ( ) PMAA hydrogel gate. ) PHEMA and PMAA bilayered gate.......133

5.9

AO8 and BSA release from the 5.0 mm assembled device at 25C. The thickness of the bilayered gate is 60 m and the thickness of the drug reservoir is 1.0 mm. ( ( ( ) AO8 at pH=3.0. ) AO8 at pH=7.3. ) BSA at pH=7.3 ..135

5.10

Thickness effects of the bilayered gate and reservoir on AO8 release behavior at pH=7.3 and 25 C. ( ( ( ) The gate thickness is 60 m and the reservoir thickness is 0.5 mm. ) The gate thickness is 60 m and the reservoir thickness is 1.0 mm. ) The gate thickness is 90 m and the reservoir thickness is 0.5 mm..136

5.11

The oscillatory release behavior of the assembled device. The gate thickness is 50 m and the thickness ratio for PHEMA to PMAA layer is 4....................138

5.12

The comparison of the targeted uni-directional release with untargeted release: (A) Targeted release. (B) Untargeted release ......140

6.1

Schematic of the 3-layer device from (A) side view and (B) top view, (C) folding on the small intestine surface, (D) a capsule containing devices ....148
xviii

6.2

Fabrication procedure of the miniature devices ..149

6.3

Experimental setup for (A) flowing testing and (B) the detachment force measurement...152

6.4

Dynamic swelling behavior PMAA and PHEMA hydrogels...156

6.5

Optical graphs of a bilayered structure at dried state (A) top view, (B) side view, (C) a curled bilayered structure in a buffer solution. Scale bars=2.0 mm.157

6.6

(A) Number of bound samples and (B) residence time for different samples attached to intestinal mucus in the flow test....159

6.7

Dynamic processes for (A) folding behavior and (B) enhanced mucoadhesion. Buffer pH=6.5 and 25C..161

6.8

Compared attachments for the devices with different contact sides in the flow test. Buffer pH=6.5 and 25C..163

6.9

The detachment force of different samples on the small intestinal surface. Buffer pH=6.5 and 25 C. Error bar = SD, n = 3.164

6.10

The fractional leakage of AO8 from the drug reservoir with different protection layers (thickness=20 m) at pH=6.5 and 25C.
xix

Error bar = SD, n = 3...166

6.11

AO8 transport from different systems across the mucosal epithelium at pH=6.5 and 25C. Error bar = SD, n = 3..167

6.12

BSA transport from different systems across the mucosal epithelium at pH=6.5 and 25C. Error bar = SD, n = 3..168

7.1

Schematic of fabrication of self-foldable microdevices..173

7.2

Schematic of the self-foldable microdevice with enhanced nanotips....174

xx

CHAPTER 1

INTRODUCTION AND MOTIVATION

The U.S. market for advanced drug delivery technology exceeded $10 billion in 1996 and is increasing rapidly [Langer, 1998]. A primary driving force is the fact that many protein- and DNA-based drugs exhibit high sensitivity to the surrounding physiological conditions as a result of their delicate physicochemical characteristics and the susceptibility to degradation by proteolytic enzymes in biological fluids. They need

to be properly protected during administration and their release needs to be precisely targeted and controlled. Most conventional drug delivery systems are based on polymers or lipid vesicles: diffusion of the drug species through a polymer membrane; a chemical or enzymatic reaction leading to cleavage of the drug from the system, and solvent activation through swelling or osmosis of the system. A major limitation of these available delivery devices is that they cannot fully protect the drugs and release them at a controllable rate over a long period of time. Certain disease states, such as diabetes,

heart disease, hormonal disorders, and cancer, require drug administration either repeatedly when needed, at a high release rate during the life-threatening moment, or at a constant release rate during a sustained period of time. Drug delivery technology can be brought to the next level by the fabrication of smart materials into miniature devices that are responsive to the individual patients therapeutic requirements and able to deliver
1

a certain amount of a drug in response to a biological state. Such smart therapeutics should possess one or more properties such as proper drug protection, local targeting, precisely controlled release, self-regulated therapeutic action, permeation enhancing, enzyme inhibiting, imaging, and reporting. This is clearly a highly challenging task and it is difficult to add all of these functionalities in a single device. Currently, there are no commercial products based on the miniaturized responsive drug delivery approach, and only limited research. Such a system would also need to exhibit good biocompatibility as drug delivery carriers [Beyssac et al., 1996; Cohen et al., 1997; Draye et al., 1997; Kitano et al., 1998; McNeill et al., 1984; Miyazaki et al., 1998; Petelin et al., 1998]. Hydrogels are crosslinked polymeric networks that are insoluble in water but swell to an equilibrium size in the presence of excess water or biological fluids [Brannon-Peppas et al., 1990; Peppas et al., 1986]. Research on hydrogels started in the 1960s with a landmark paper on poly(hydroxyethyl methacrylate) [Wichterle et al., 1960]. Due to the unique swelling properties and the biocompatible structure, these materials have been extensively studied for biomedical and pharmaceutical applications, such as contact lenses, membranes for biosensors, linings for artificial hearts, materials for artificial skin and drug delivery devices [Peppas et al., 1994; Walther et al., 1995; Peppas et al., 1997; Peppas et al., 2000]. In nature, polymeric hydrogel is a three-dimensional network comprising interconnected hydrophilic macromolecules, with an inner space partially filled with water molecules. The highly hydrated, non-ionic and good biocompatibility provide the ability of hydrogels to release drug in a regulated mode, which can be achieved by controlling the synthesis conditions, such as the reactant composition, the ratio of crosslinked density, the method of polymerization, and the external environment.
-2-

Hydrogels are often synthesized by UV photopolymerization [Lu et al., 1999; Ward et al., 2001] or redox polymerization [Hassan et al., 1999]. Photopolymerization is favored because hydrogels can be synthesized at temperatures and pH conditions near physiological conditions and even in the presence of biologically active materials. Furthermore, photopolymerization can be easily controlled by adjusting the dosage and intensity of UV light, and the curing temperature. Photo-Differential Scanning Calorimetry is the most widely used technique to characterize the photopolymerization kinetics. A great deal of research has been carried out using this approach for photocurable materials. However, the application of this technique for highly volatile reagents is limited since uncovered sample pans lead to significant sample loss during measurement. Some researchers applied unsealed polyethylene (PE) films over the sample pan to reduce the sample loss [Ward et al., 2001], while others used the sample weight after the reaction to correct for the measurement error resulting from reagent evaporation [Jakubia, 2000]. The results from such treatments are doubtful because sample loss during the reaction is a time-dependent process. When preparing the carriers for drug delivery, solvents like water and ethanol are often used in the synthesis to control the hydrogel structure. Evaporation of highly volatile solvents like ethanol makes it impossible to study the reaction kinetics using the existing approaches. We have recently developed a modified DSC sample pan [Li et al., 2005]. Sample loss during reaction is minimized, and loaded samples are much more uniform over the sample surface. This new method is applied in this study. To better control the synthesized hydrogels for various applications, it is essential to understand how the polymerization conditions, chemical structure of reactants and
-3-

their composition, and solvent type and concentration affect the reaction and the resulting properties of hydrogels. A number of studies have reported that varying curing conditions may achieve different gel structures and swelling properties [Lowman et al., 1997; Elliott et al., 2002; Kwok et al., 2003], and the compatibility between the solvent and the resin may affect inter-molecular and primary cyclization of multi-vinyl monomers during the polymerization, and, consequently, the hydrogel properties. However, there lacks a thorough understanding on the interactions of reaction kinetics, rheological changes, gel formation, and hydrogel structures. Oral delivery of peptides and proteins has become a challenging and attractive task with the enormous market potential in resent years. Typically, the intramuscular or intravenous injection is used for their administration. However, due to the disadvantages, such as the pain, inconvenience and inconsistent pharmacokinetics for this administration, lots of work has been done to pursue alternative administration methods other than the conventional injection approach. Among various potential routes, oral administration could be the most convenient and ideal route since it is known as the most desirable route of drug administration. Although being an ideal non-invasive route of drug administration, the peptides and proteins delivery through the oral route is fraught with difficulties around low bioavailability, which results from the pH fluctuation, proteolytic degradation, low transport, and short residence time. Many possible solutions, such as the inclusion of protection, protease inhibitors, enhancers/promoters, and/or specific adhesion, do help the increased drug bioavailability through oral route. Typical oral delivery systems can be summarized as two categories: conventional systems, such as tablet, capsules and syrup,
-4-

and advanced systems, such as micro/nanoparticles and intestinal patches. For microparticles and nanoparticles, the loaded drugs can be released to all directions due to their symmetric shape. Asymmetric intestinal patches and some microdevices can provide protected unidirectional release. Dorkoosh and coworkers [Dorkoosh et al., 2001; Dorkoosh et al., 2002] designed a novel drug delivery system for site-specific drug delivery of peptide drugs in the intestinal tract using superporous hydrogels (SPH) and SPH composite polymers, which swell very rapidly by absorption of gut fluids. The system attached to the intestinal wall and provided a longer residence time for drug release. However, only a slight decrease in blood glucose levels was observed in animal studies. Shen et al. [2002] reported an intestinal patch design for oral delivery. A longer residence time and unidirectional diffusion were achieved for helping drug diffusion through the intestinal barrier by using a mucoadhesive layer of Carbopol/ pectin. Tao et al. [2004] combined microfabrication techniques with the use of mucoadhesive plant lectins to design a microdevice with a long residence time. iMEDD Inc. developed Oral-MEDDs (microfabricated particles) technology [Cohen et al., 2003] which combined several oral delivery approaches into a single drug delivery system to deliver peptides and proteins. The mucoadhesion for these systems is through surface-to-surface contact. Due to the continuous shedding of surface mucus, these systems have the limited residence time and the drug bioavailability is low. To match the patients needs, further efforts and better solutions are still needed. In this work, we design multi-functional devices based on the hydrogels that can bind to the targeted issue for self-regulated and sustained release. A common process model for engineering is used to show how materials appear likely to break previous
-5-

barriers in the process that ultimately results in applications with potential benefits. This process development can be conveniently represented by the schematic description of pH-sensitive hydrogels for oral drug delivery systems and sensors (Figure 1.1).

Swelling, Kinetics, Rheology

Polymeric Self-folding

Microfabrication

Material Characteristics

Polymeric Materials

Device Design

Modeling

Proper swelling High final conversion Better mechanical properties

Better protection, Long residence time, High transport

In Vitro Release, Targeting, Unidirectional Rel.

Material/Process Design to Improve the Delivery Performance

Animal Studies, Clinic Trials

Products/Applications
(Oral DDS, Biosensor and Bioreactors, Tissue Clamping)

Figure 1.1 The engineering process applied for pH-sensitive hydrogels.

-6-

Some issues need to be addressed are as follows: (1) What factors play the important roles in the synthesis of hydrogels with desired properties? (2) How will the solvent ratio and light intensity affect the structure and properties of hydrogels? (3) How will a multi-functional DDS be designed to integrate all possible solutions to achieve high bioavailability? The objectives of the research are (1) to generate functional hydrogels with desired properties, (2) to develop an intelligent DDS, which is effective for controlled release, drug protection, targeted unidirectional release, high transport, long residence time, as well as a quick response time, and (3) to investigate the relationship between the hydrogel properties and the release performance, and then optimize the device design. Thus, we can extend the functionalities of hydrogels by combining with the fabrication technology to match physiological needs for various pharmaceutical applications.

-7-

CHAPTER 2

LITERATURE REVIEW

2.1

Overview of pH-Sensitive Hydrogels Hydrogels can be classified as neutral or ionic based on the type of repeating units

or the nature of the side chains on the polymer backbone. They can be homopolymer or copolymer networks based on the preparation approach. The most important property for hydrogel is the stimuli-sensitivity depending on the external conditions, which include pH, temperature, pressure, ionic strength, electromagnetic radiation, ultrasonic energy, buffer composition, the concentration of glucose, stress and strain, and photo [Peppas, 1991]. These conditions dramatically affect the swelling behavior, network structure, permeability and mechanical strength of hydrogels. Such intelligent materials open the door for novel applications in the areas of nanotechnology (actuators, substrates), surgical implants and tissue engineering, due to hydrogels unique ability to undergo phase transitions under the influence of small stimuli. The pH-sensitive hydrogels exhibit swelling or deswelling behavior with changes of pH values in the surrounding medium. The swelling behavior may be due to one of the following mechanisms: (1) changes in the hydrophobic-hydrophilic nature of chains; (2)
-8-

inter- and intramolecular complexation by hydrogen bonding, or (3) electrostatic repulsion. All these mechanisms are closely related to the protonation phenomena of the ionizable moieties on the polymer backbone or the side chains. In the first case, ionization makes the hydrophobic polymer network more hydrophilic because the ionized structure usually posses more hydrophilicity which can imbibed more water into the matrix. In the second case, ionization results in the breaking up of the hydrogen bonds that exist in the polymeric matrix in the unionized state, leading to the hydrogel swelling. In the third case, the ionization provides the electrostatic repulsion among charges present on the polymer chain to keep the chains apart and allow more water absorbing into the loose structure. In all these cases, the kinetics of the swelling process and the equilibrium extent of swelling are affected considerably by several factors, such as ionic strength of the medium, buffer composition, presence of salts [Hariharan et al., 1996]. Other factors, such as the crosslinking ratio, solvent quality, chemical structure of monomers, and synthesized conditions also influence the structure formation and the swelling behavior of hydrogels. pH-sensitive hydrogels can be divided into anionic and cationic depending on the nature of pendant groups in the networks, which show sudden or gradual changes in their dynamic and equilibrium swelling behavior as a result of pH changes. Anionic gels often contain carboxylic or sulfonic acid. When the pH value of surrounding medium rises above its pKa, the ionized structure will provide increased electrostatic repulsion between chains and the hydrophilicity of network. Under these conditions, hydrogels are capable of uptaking large amounts of water and forming very loose structure. In contrast, cationic hydrogels usually contain pendant group such as amines. As pH values lower
9

than the pKb, the amine groups change from NH2 to NH3+, resulting in the increased hydrophilicity, strong electrostatic repulsion, and high swelling ratio. 2.1.1 Anionic hydrogels Many researchers have studied the dynamic swelling of anionic pH-sensitive hydrogels, which often contain carboxylic groups. Typical examples of such polymers include poly(acrylic acid) (PAA) and poly(methacrylic acid) (PMAA). Copolymers of PAA and PMAA with poly(ethylene glycol) (PEG), poly(vinyl alcohol) (PVA), and poly(hydroxyethyl methacrylate) (PHEMA) also exhibit the pH sensitivity due to the presence of carboxylic segment. Additionally, incorporating other sensitive groups into the networks of PAA or PMAA will give gels more interesting properties. For example, the copolymer of PAA and PMAA with PNIPAAm can provide the coupling environmental sensitivity of pH and temperature [Tian et al., 2003; Zhang et al., 2000]. Recently, a series of smart biomaterials, such as poly(ethylacrylic acid) (PEAA) and poly(propylacrylic acid) (PPAA), has opened new opportunities for the molecular imaging field because of their sharp pH-sensitivity [Stayton et al., 2005] .

PAA

PMAA

PEAA

PPAA

PBAA

Figure 2.1 Structures of anionic pH-sensitive hydrogels.

10

Hydrogels made of PAA or PMAA can be used to develop formulations that release drugs in a neutral pH environment [Brannon et al. 1990]. Some researchers [Hassan et al., 1999] focused on the synthesis of anionic pH-sensitive hydrogels and the swelling behavior studies. Of particular interest was the design of a self-regulated release device based on the mechanism of the molecule gate system. An important example of copolymer networks was represented [Lowman et al., 1995] to verify the complexation and decomplexation mechanism. The authors not only explored the influence of factors, such as the solution pH, graft chain molecular weight, and copolymer composition, on network structure and dynamic property of p(MAA-g-EG) hydrogels, but also studied the complexation dependent diffusion coefficients. p(MAA-g-EG) is a promising candidate for oral delivery of peptide and protein drugs through the gastrointestinal tract [Torres-Lugo et al., 2002; Robinson et al., 2002; Kim et al., 2003; Ichikawa et al., 2003]. Peppas group prepared p(MAA-g-EG) micro/nanospheres with relatively narrow size distributions. The effects of various reaction parameters on the particle size and the distribution were investigated. The enhancing effect of p(MAA-g-EG) micro- or nano-particles for salmon calcitonin delivery through intestinal epithelial cells was also evaluated using Caco-2 cell monolayer. Results revealed that the p(MAA-g-EG) hydrogel microparticles could be used as a cytocompatible carrier possessing the transport-enhancing effect on the intestinal epithelial cells. PMAA crosslinked with azoaromatic crosslinkers was developed for colon-specific drug delivery [Ghandehari et al., 1997]. The drug release from such hydrogels in the stomach was very minimal. As the gels passed down the intestinal tract, the extent of swelling increased. But, the azoaromatic crosslinks of the
11

hydrogels were degraded by azoreductase produced by the microbial flora of the colon. It is known that the transition between the swollen and the collapsed state with changes in pH can be moved to higher pH values by increasing the hydrophobicity of the monomers. Tirrell and coworkers [1992] first described the pH-dependent properties of PEAA for membrane-disruptive applications. PEAA is inactive at physiological pH and has a sharp transition around pH of 6.3. To obtain a series of shifted pH profiles, Hoffmans group [Murthy et al., 1999] investigated the pH transition change of sensitive hydrogels by using different monomers with increased methylene units and applied their membrane-disruptive properties in a blood cell hemolysis assay. The PPAA exhibited a shift to the membrane active state at a higher pH and a surprising increase over PEAA in hemolytic efficiency. Further addition of another methylene unit with poly(butyl acrylic acid) shifted the pH profile to the physiological pH. This general shift in pH profiles is consistent with the trend expected for making the alkyl group longer and more hydrophobic [Mourad et al., 2001].

2.1.2 Cationic hydrogels The synthesis and properties of cationic pH-sensitive hydrogles have also been investigated over the past three decades. Hariharan and Peppas [1996] investigated the swelling behavior of cationic hydrogels as carriers for drug delivery. Diethylaminoethyl methacrylate (DEAEM) and diethylaminoethyl acrylate (DEAEA) were used as the cationic monomers copolymerized with HEMA. The equilibrium water uptake was a strong function of the ionic strength of the medium. Podual [1998] provided a cationic hydrogel prepared by the copolymerization of DEAEM and poly(ethylene glycol)
12

monomethacrylate (PEGMA). Not only the effect of crosslinking ratio on the swelling properties was studied, but also the structure of hydrogels and the diffusion coefficients were determined. Traitel et al. [2000] studied the insulin controlled release system based on the cationic hydrogel, PHEMA-co-N,N-dimethylaminoethyl methacrylate

(DMAEMA). The effects of polymer morphology and oxygen availability on hydrogel swelling and insulin release kinetics were studied. Hydrogels without the crosslinking agent were stable in water and their sensitivity to pH was higher than the chemically crosslinked hydrogels. A pH-sensitive hydrogel containing glucose oxidase enzyme is called glucose-sensitive hydrogel due to its responsiveness to ambient glucose concentration [Jung et al., 2000]. These systems are functionalized with enzymes by binding the enzyme into a polymer network during polymerization. Glucose oxidase is probably the most widely used enzyme in glucose sensitivity. It oxidizes glucose to gluconic acid, resulting in a pH change of the medium. Horbetts group [Albin et al., 1985, Kost et al., 1985; Klumb et al., 1992; Klumb et al., 1993] was the first to study systems consisting of immobilized glucose oxidase in a pH responsive polymeric hydrogel, enclosing a saturated insulin solution. Glucose oxidase has been successfully immobilized on a wide variety of polymers, such as poly(MAA-g-EG) [Hassan et al., 1999], poly(HEMA-DMAEMA) [Traitel et al., 2000], poly(DEAEM-g-EG) [Podual, 1998], poly(HPMA-co-DMAEMA) [Jung et al., 2000], polyacrylates [Turmanova et al., 1993], polyethylene [Hsiu et al., 1990], poly(vinyl alcohol) [Kozhukharova et al., 1988]. pH-sensitive hydrogels can serve as drug delivery carriers for oral, buccal, rectal, vaginal, ocular, epidermal and subcutaneous applications. However, hydrogels made of
13

non-biodegradable polymers has to be removed from the body after use. The non-biodegradability is not a problem for oral drug delivery, but it becomes a serious limitation in other applications, such as the development of implantable drug delivery carriers or implantable biosensors. Thus, much attention has been focused on the development of biodegradable pH-sensitive hydrogels. Various formulations were developed to obtain the biodegradable pH-sensitive hydrogels with appropriate properties, such as dextran [Chiu et al., 1999, Franssen et al., 1999], semi-interpenetrating Chitosan-PVA [Wang et al., 2004], PVA-gelatin [Wang et al., 2004], and poly(lactic acid)-poly(ethylene glycol)-poly(lactic acid) hydrogels[Mason et al., 2001].

2.2

Temperature-Sensitive Hydrogels Temperature-sensitive hydrogels have received considerable attention for uses in

bioseparations, drug delivery, and diagnostics due to the ability of hydrogels to swell or shrink as a result of temperature change in the surrounding fluid [Peppas et al., 2000]. Based on the transition mechanism, these hydrogels can be classified into three categories: negatively temperature-sensitive gels, positively temperature-sensitive gels, and thermo-reversible gels [Qiu et al., 2001]. Positive hydrogels have an upper critical solution temperature (UCST). If the temperature is below UCST, the hydrogels contract and release solvent from the matrix. In contrast, the swelling behavior of negative hydrogels is attributed to the lower critical solution temperature (LCST). A temperature above LCST results in a collapsed structure for hydrogels. For the thermo-reversible gels, the polymer chains are not covalently crosslinked and the gels may undergo sol-gel phase transitions, instead of
14

swelling-shrinking transitions.

2.2.1 Negatively temperature-sensitive gels For most polymers, the water solubility increases with the increasing temperature. Negatively temperature-sensitive gels, however, have a critical parameter LCST. That means these gels shrink as the temperature increases above the LCST and swell at the lower temperature. The structures of some of those polymers are shown in Figure 2.2.

PNIPAAm

PDEAAm

P(NIAAm-co-AA)

Figure 2.2 Structures of negatively temperature-sensitive hydrogels.

Some of the earliest work with negatively temperature-sensitive hydrogels was done by the Tanakas group [1978]. Poly(N-isopropylacrylamide) (PNIPAAm) is the best example of a negatively temperature-sensitive hydrogel, which is made of polymer chains containing a mixture of hydrophobic and hydrophilic segments. At lower temperatures, water interacts with the side chains through the hydrogen bonds between water molecules and the hydrophilic parts, CONH. These hydrogen bonds lead to enhanced dissolution and well swelling in water [Zhang et al., 2003]. As the temperature is increased to higher
15

than LCST, the hydrophobic interactions among hydrophobic segments, CH(CH3)2, become stronger, while hydrogen bonds become weaker. These interactions result in the shrinking of the hydrogels due to inter-polymer chain association [Qiu et al., 2001]. Hitotsu et al. [1987] worked with crosslinked PNIPAAm and determined that the LCST of PNIPAAm gel was 34.38C. However, the response rate to external temperature changes of typical PNIPAAm hydrogel is low, which limits its applications. Kabra et al. [1991] synthesized fast temperature-response PNIPAAm gels by using a phase separation technique. Preparation of gels at temperatures above LCST [Wu et al., 1992] or below the freezing point [Zhang et al., 1999] results in an enhanced shrinking rate. Gas blowing [Nakamoto et al., 2001] and radiation [Chen et al., 1999] may produce porous structures leading to fast response. Other successful approaches to achieve a high

temperature-response rate involve using poly(ethylene glycol)s as pore-forming agents [Zhang et al., 2000], interpenetrating poly(vinyl alcohol) within the hydrogel network, using aqueous sodium chloride solution as the reaction medium for gel preparation, and carrying out polymerizations in mixed sucrose solutions. These approaches could significantly increase the response rate since these reaction mediums induce the phase separation of gel system. For example, the fully deswelling time could be reduced to 2 min when the hydrogel polymerizations were carried out in aqueous glucose solutions [Zhang et al., 2003]. LCST is a very important parameter for negatively thermo-sensitive gels. LCST could be increased by mixing a small amount of ionic copolymers in the gels [Yu et al., 1993] or by changing the solvent composition [Suzuki et al., 1996]. In general, as the polymer chains contain more hydrophobic constituents, LCST moves to a lower
16

temperature. Thus, incorporating a hydrophilic monomer, acrylic acid (AAc), into the PNIPAAm backbone is a good approach to modulate the properties of PNIPAAm gels [Zhang et al., 2002]. Copolymerization of NIPAAm with different monomers results in hydrogels with versatile properties. However, an increased hydrophilic content in the copolymer network can reduce its temperature sensitivity [Beltran et al., 1991; Feil et al., 1993]. In order to improve the temperature sensitivity of copolymers, several researchers have prepared PNIPAAm-based copolymers. Okano and coworkers [Kaneko et al., 1998] developed an exquisite method to prepare graft hydrogels of PNIPAAm. Small PNIPAAm molecules were grafted with the main chain of the crosslinked PNIPAAm. Above the LCST, hydrophobic regions in the network structure made the gels dehydrate to a collapse state. At temperatures below LCST, the gels could transform into a fully swollen conformation in less than 20 min, which was much faster than that of comparable gels without graft chains. This group also proposed an incorporating carboxylate method to promote gel shrinking [Ebara et al., 2001]. 2-carboxyisopropylacrylamide (CIPAAm) was incorporated into PNIPAAm gels to induce rapid shrinking in response to small temperature increases. In contrast, P(NIPAAm-co-AAc) copolymer gels lose their temperature sensitivity with the introduction of only a few mole percent of AAc. Zhang et al. [2002] synthesized P(NIPAAm-co-AAc) gels in an alkaline solution to achieve the improved oscillating swelling properties. There has also been significant interest in the synthesis of PNIPAAm-based hydrogels by other methods such as graft-, block- or combcopolymerization. Such systems show promise for rapid and abrupt or oscillatory release of drugs, peptides, or proteins, because their swelling or syneresis process can occur relatively fast [Yoshida et al., 1995; Kaneko et al., 1996; Inoue et al., 1997].
17

Recently, efforts have been made to prepare multifunctional hydrogels responding to more than two stimuli, such as the pH and temperature sensitive hydrogels. Chen and Hoffman [1995] prepared p(NIPAAm-g-AA) gels, which exhibited temperature- and pH-sensitive behavior. These gels were able to respond rapidly to both temperature and pH changes. The temperature- and pH-dependent swelling behaviors were better defined in the graft copolymers than in random copolymers containing similar amounts of components. Tian et al. [2003] developed a hydrogel of p(NIPAAm-co-AAc) modified by a small amount of hydrophobic comonomers in tert-butanol solutions. The hydrogels with a suitable 2-(N-ethylperfuorooctanesulfoamido) ethyl acrylate content showed good pH and temperature sensitivity. Similar work was done by the Peppas group [Zhang et al., 2000]. The interpenetrating gels of PNIPAAm and PMAA exhibited the ability of responding to temperature and pH conditions. Additionally, the transition conditions were determined at a pH value of approximately 5.5 and a temperature range of 31-32C. Negatively temperature-sensitive hydrogels have been studied extensively and these materials can be used in a variety of applications, including controlled drug delivery, immobilized-enzyme reactors, separation process, and biochips. In a monolithic device, an onoff drug release profile could be obtained based on the reversible thermo-sensitivity of hydrogels [Bae et al., 1990; Okano et al., 1990], which involve crosslinked p(NIPAAm-co-BMA), and inter-penetrating PNIPAAm and

poly(tetramethyleneether glycol) (PTMEG). In order to increase the mechanical strength of hydrogels, Okano and coworkers incorporated a hydrophobic comonomer, BMA into NIPAAm gels and investigated the onoff release profile of indomethacin from the matrices in response to a stepwise changing temperature. The hydrophobicity of the
18

comonomer influenced the shrinking process and thus controlled the release behavior of the therapeutic agent dispersed in the matrix [Yoshida et al., 1991]. Negatively temperature-sensitive gels are also utilized for controlled delivery of highly sensitive therapeutic agents, such as peptides and proteins. Peppas et al. [1996] developed a hydrogel of inter-penetrating PNIPAAm and PMAA and studied the release kinetics of bioactive streptokinase. Kim et al. [1996] used an inter-penetrating hydrogel of PNIPAAm and PAA to effectively release the protein drug, calcitonin, in response to changing temperature and pH.

2.2.2 Positively temperature-sensitive gels Certain hydrogels formed by IPNs show positive thermosensitivity. IPNs of PAA and polyacrylamide (PAAm) or P(AAmco-BMA) have positive temperature dependence. IPNs composed of PAA and PAAm may shrink at low temperatures because of the interpolymer complexes formed by hydrogen bonding. The complexes dissociate at higher temperatures due to breaking of hydrogen bonds, and the gels rapidly swell above the UCST [Klenina et al., 1981]. Katono et al. [1991] compared the temperature

dependent swelling behavior of poly(AAm-co-BMA), the IPNs of poly(AAm-co-BMA) with PAA, and the random copolymer gel poly(AA-co-AAm-co-BMA). The IPNs and the random gels showed the distinctly different profiles of temperature dependence, although both had the positive temperature dependence. Only the IPNs showed a sigmoidal alteration with a transition zone. The swelling of those hydrogels was reversible, responding to stepwise temperature changes. This resulted in reversible changes in the release rate of a model drug, ketoprofen, from a monolithic device.
19

Clinical applications of thermosensitive hydrogels based on NIPAAm and its derivatives are limited due to the non-biocompatibility of the monomers and crosslinkers and non-biodegradability of NIPAAm polymers and their derivatives. Further development of new, biocompatible and biodegradable thermoreversible gels, such as PEO-PLA block copolymers, is necessary to exploit the useful properties of thermoreversible gles.

2.3 2.3.1

Properties of Hydrogels Swelling properties The swelling behavior of hydrogels is an important property for a variety of

applications. Generally, the swelling property of polymers is reflected by the weight-swelling ratio, the ratio of the weight of the swollen sample to the weight of the dry matrix. Factors affecting the swelling ratio mainly involve the crosslinking ratio, the solvent concentration and quality, the chemical structure, and the specific stimuli. The crosslinking ratio, the ratio of moles of crosslinking agent to the moles of polymer repeating units, has a dominated effect. The higher the crosslinking ratio, the more crosslinking agent is incorporated in the hydrogel structure. Highly crosslinked hydrogels have a tighter structure, and will swell less compared to the same hydrogels with a lower crosslinking ratio. In many cases the influence of solvent is small. However, it is becoming increasingly evident that solvent effects can be used to control the free radical polymerization of hydrogels, both at the macroscopic and at the molecular levels. The solvent concentration during the polymerization affects the material properties of the
20

polymer by increasing the rate of primary cyclization of multivinyl monomers during the polymerization [Anseth et al., 1996; Elliott et al., 2001]. A primary cycle differs from a crosslink in that the propagating free radical reacts intramolecularly with its own pendant double bonds, which then loses the opportunity to crosslink. The greater the extent of primary cyclization, the less crosslinked the polymer will be and the larger the mesh size. This leads to the increased equilibrium swelling and reduced mechanical strength with the increasing solvent concentration during the polymerization. The effects of solvent concentration on the rate of primary cyclization and gel network formation can be explained by the local dynamics of the propagating radical. For lower solvent concentrations, the double bond concentration surrounding the free radical is relatively high, leading to a faster rate of propagation and less opportunity for the free radical to cycle by reacting with its own pendant double bonds. In addition to solvent concentration, solvent quality also affects the three-dimensional network structure created during the polymerization. For a better solvent, the propagating chain is less likely to cycle and thus has a compact structure. However, the propagating chain is more likely to cycle for a poor solvent, and the rate of primary cyclization is high, leading to a loose network structure [Elliott et al., 2002]. The chemical structure of the polymer may also affect the swelling ratio. Hydrogels containing hydrophilic groups swell to a higher degree compared to those containing hydrophobic groups. Hydrophobic groups collapse in the presence of water, thus minimizing their exposure to the water molecule. As a result, the hydrogels will swell much less compared to hydrogels containing hydrophilic groups. Swelling of environmentally sensitive hydrogels can be affected by specific stimuli. For example,
21

temperature and pH affect the swelling of temperature- and pH-sensitive hydrogels, respectively. There are many other specific stimuli that can affect the gel swelling.

2.3.2

Network structure and characterization The effect of chemical structure on polymer properties is without doubt the most

important aspect of polymer chemistry. Extensive uses of hydrogels in drug delivery systems depend to a large extent on their structures in buffer solution. Based on the work done by many researchers, the most important parameters used to characterize the network structure of hydrogels are the polymer volume fraction in the swollen state (v2,s), molecular weight of the polymer chain between two neighboring crosslinking points (Mc), and the corresponding length or mesh size (). In order to elucidate the structure of hydrogels, the equilibrium swelling theory and the rubber elasticity theory are utilized [Peppas et al., 2000]. The polymer volume fraction in the swollen state is a measure of the amount of fluid imbibed and retained by the hydrogel. The molecular weight between two consecutive junctions is a measure of the degree of crosslinking of the polymer. These junctions may be chemical crosslinks, physical entanglement, crystalline regions, or even polymer complex. It is important to note that only average values of Mc can be calculated due to the random nature of the polymerization process. The correlation distance between two adjacent crosslinks () provides a measure of the space available between the macromolecular chains for drug diffusion. Also, it can be reported only as an average value. These parameters can be determined theoretically or through a variety of experimental techniques.
22

Theoretical approaches Several theories have been proposed to calculate the molecular weight between

crosslinks in a hydrogel matrix. Two theoretical methods, which are prominent among the growing techniques utilized to elucidate the structure of hydrogels, are the equilibrium swelling theory and the rubber elasticity theory. The structure of hydrogels that contain ionic moieties was analyzed by Peppas and Merrill [1977] based on the Flory-Rehner theory [Flory et al., 1943]. This thermodynamic theory states that a crosslinked polymer gel, which is immersed in a fluid and allowed to reach equilibrium with its surroundings, is subject only to three opposing forces, the thermodynamic force of mixing, the retractive force of the polymer chains, and the ionic force. At equilibrium, these forces are equal. Eq. (1) describes the physical situation in terms of the Gibbs free energy.

Gtotal = Gelastic + Gmixing + Gionic

(1)

Here, Gelastic is the contribution due to the elastic retractive forces developed inside the gel, Gmixing is the result of the spontaneous mixing of the fluid molecules with the polymer chains, and Gionic is the contribution due to the ionic nature of the polymer network. Eqations (2) and (3) are expressions that have been derived for the swelling of anionic and cationic hydrogels prepared in the presence of a solvent.
2 v v 2M c K V1 V1 v 2 ,s 2 ( )( pH a ) 2 = [ln(1 v 2 ,s ) + v 2 ,s + 1v 2 )(1 )v 2 ,r [( 2 ,s )1 / 3 ( 2, s )] ,s ] + ( vM c Mn v 2 ,r 4 I v 10 2 v 2 ,r Ka 2 v v 2M c K V1 V1 v 2, s 2 ( )( pH 14b ) 2 = [ln(1 v 2 ,s ) + v 2 ,s + 1v 2 )(1 )v 2,r [( 2,s )1 / 3 ( 2 ,s )] ,s ] + ( vM c Mn v 2,r 4 I v 10 2 v 2 ,r Ka

(2) (3)

23

In these expressions, I is the ionic strength, and Ka and Kb are the dissociation constants for the acid and base, respectively. Hydrogels resemble natural rubbers in their remarkable property to elastically respond to applied stresses. The elastic behavior of hydrogels can be used to elucidate their structure by utilizing the rubber elasticity theory originally developed by Treloar [1958] and Flory [1949]. However, the original theory or rubber elasticity does not apply to hydrogels prepared in the presence of a solvent. Silliman [1972] and Peppas et al. [1977] developed the expressions to analyze the structure of hydrogels prepared in the presence of a solvent.

RT
Mc

(1

2M c 1 1 )( 2 )( 2, s ) 3 Mn 2,r

(4)

In Eq. (4), is the stress applied to the polymer sample, is the density of the polymer. The rubber elasticity theory has been used to analyze chemical and physical crosslinked hydrogels [Mark, 1982; Anseth et al., 1996], as well as hydrogels exhibiting temporary crosslinks due to hydrogen bonding [Lowman et al., 1997]. The primary mechanism of drug release from a hydrogel matrix is diffusion, occurring through the space available between macromolecular chains in aqueous media as a result of environmental stimuli. This space is often regarded as the pore. Depending on the size of these pores, hydrogels can be conveniently classified as macro-porous, micro-porous and non-porous. A structural parameter that is often used in describing the size of the pores is the correlation length () which is defined as the linear distance between two adjacent crosslinks, and can be calculated using the following equation [Torres-Lugo et al., 1999],
24

= Qr1 / 3 ( r02 )1 / 2

(5)

where Qr is the volume swelling ratio of the swollen polymer at equilibrium to the dry polymer, and
( r02 )1 / 2 is the end-to-end distance in the unperturbed state, which can be

calculated by the following equation,


( r02 )1 / 2 = ( 2Cn M c 1 / 2 ) l Mr
(6)

where Cn is the polymer characteristic ratio (14.4 in case of a methacrylate chain), Mc is the molecular weight between crosslinks, l is the carbon-carbon bond length (1.54 ), , and Mr is the molecular weight of the repeating unit. The average values of Mc and are related to each other. The molecular weight between crosslinks can be obtained by the following equation,
3.9 104 Tg Tg 0

Mc =

(7)

where Tg is the transition temperature of the crosslinking polymer, Tg0 is the transition temperature of the uncrosslinking polymer. According to these equations, the parameters characterizing the network structure of hydrogels can be experimentally obtained. Although theoretical characterizations of the network structure are complicated, they and the diffusion studies of model drugs provide an invaluable insight into the very complex structure of gel networks and help in the design for drug delivery carriers [Narasimhan et al., 1997].

Experimental approaches Besides theoretical approaches, simpler and more intuitionistic approaches can be
25

used to investigate the hydrogel structure in buffer solutions. This section presents a brief description of each specific approach followed by its advantages and limitations. B.1 Scanning electron microscopy (SEM) To visually examine the surface and interior morphology of a hydrogel in the swollen state, scanning electron microscopy is commonly used to analyze the pore structure and to observe the three dimensional structure. For example, Kim et al. [2000] reported the visual observation of an unique 3D honeycomb-like network structure in the interior of a swollen dextran-methacrylate hydrogel by SEM. Investigations of the hydrogel structure by SEM lead to valuable results: 3D structure, pore shapes, and the approximate pore size. However, this SEM-based technique suffers from a sever disadvantage because the native state of hydrogels is characterized by the presence of water and the need of dehydration and/or fixation procedures prior to SEM examination inevitably affects the morphology of a hydrogel. The preparation of a hydrogel sample for SEM examination involves critical-point drying and vacuum drying methods. Both drying techniques result in volume shrinkage and significantly morphological alterations of the gels. Other techniques, such as cryofixation, cryofracturing, and freeze-drying, have been used to examine the interior structure of hydrogels because solvent (e.g., water) inside can be easily removed by sublimation with minimal disturbance of structure [Hong et al., 1998; Yang et al., 1983]. B.2 Environmental scanning electron microscopy (ESEM) Although the dehydration and/or fixation procedures aid the swollen gel in SEM testing, some reports indicated a discrepancy between the original hydrogel structure and the deduced images from SEM. ESEM represents an important advance in conventional
26

SEM for hydrogel characterization. Whereas conventional SEM requires a relatively high vacuum in the specimen chamber to prevent atmospheric interference with primary or secondary electrons, an ESEM may be operated with a poor vacuum (up to 10 Torr of vapor pressure, or one seventy-sixth of an atmosphere) in the specimen chamber. In such "wet mode" imaging, the specimen chamber is isolated from the rest of the vacuum system. Water is the most common imaging gas, and a separate vacuum pump permits fine control of its vapor pressure in the specimen chamber. Due to the effect of the electron beam, the water molecules are positively ionized, and thus they are forced/attracted toward the hydrogel samples, serving to neutralize the negative charge produced by the primary electron beam. In order to preserve the sample from dehydration, the water vapor is kept at the saturation condition within the microscope chamber by using a Pertiler cold stage. The field-emission gun produces a brighter filament image and its accelerating voltage may be lowered significantly, permitting nondestructive imaging of fragile specimens, such as swollen gels

[www.itg.uiuc.edu]. Because of these technical improvements, ESEM provides more advantages for characterization of the hydrogel network. No additional sample treatment is performed to avoid any introduction of possible specimen-coating artifacts, or problems involved with either changing samples to a vacuum-friendly state or creating their former replica. The controlled environment in the specimen chamber retains the stable structure. Some accessory and the control valves could extend ESEM applications for dynamic experiments. For instance, the morphology change of the pH sensitive poly(AA-co-AAm) hydrogel swollen in different buffers was studied, and the 3D network and the pore size were clearly observed from ESEM images [Zhou et al., 2003].
27

B.3

Confocal laser scanning microscopy (CLSM) Confocal laser scanning microscopy (CLSM) is another valuable imaging

technique for direct observation of the hydrogel structure as it allows us to observe non-destructive samples under the mild conditions. CLSM has been successfully applied in biological, medical, and geological studies as an alternative investigative method for hydrogel structure analysis. For CLSM, the hydrogel can remain in the aqueous environment, thus avoiding the hazardous dehydration. The only sample treatment prior to examination is the conjugation of a fluorescent dye to the polymer segments in a hydrogel. However, the actual dye concentration can be very low so that the disturbance of biological systems is kept to a minimum. Subsequently, images of the bulk structure can be obtained at different locations without cutting or fracturing the hydrogels and magnified images of any area of interest can be obtained [Fergg et al., 2001]. With the aid of application software, the 3D nature of the hydrogel can be calculated from a series of successful images taken at defined intervals and observed as a movie or as a single stereo pair image. There are a great number of advantages to using CLSM compared to conventional fluorescence microscopy (FM). The two most important ones are the ability to eliminate the out-of focus noise and the greatly increased sensitivity of the machine [www.bioteach.ubc.ca]. CLSM also has some limitations. The first limitation of CLSM is that the resolution is limited by the wavelength of light. Photo-damage is also a limitation in the use of CLSM. The good axial resolution (between two focal planes) is obtained by using two-photon fluorescence microscopy, which provides the possibility to obtain biochemical information about cells or tissues and causes minimal photo-damage due to
28

its inherent 3D resolution and long penetration depth. B.4 Porosimetry The use of SEM, ESEM and CLSM approaches provide morphological details of the interior and surface structure of hydrogels. However, there are needs to examine the structure of hydrogels in a quantitative manner, because pore size, volume, and structure of hydrogels are critical factors to control swelling, drug release behavior, and biological interactions inside the body. The quantitative information of the pore structure of hydrogels under a swollen condition could be obtained by nitrogen absorption and mercury intrusion porosimetry. Mercury intrusion porosimetry (MIP) has provided valuable information about various aspects of pore structure characterization for porous media and powders [Mikijelj et al., 1991; Liu et al., 2000; Gemeinhart et al., 2000]. The theory of all mercury porosimeters is based on the physical principle that a non-reactive, non-wetting liquid will not penetrate pores until sufficient pressure is applied to force its entrance. The relationship between the applied pressure and the pore size is given by the Washburn equation:
D= 4 cos P
(8)

Where P is the applied pressure, D is the diameter, is the surface tension of mercury (480 dyne cm-1) and is the contact angle between mercury and the pore wall, usually near 150. As pressure increases, the instrument senses the intrusion volume of mercury. As the mercury column shortens, the pressure and volume data are continuously acquired and displayed by an attached personal computer.
29

As an analytical instrument, MIP can measure pores of diameters ranging from 3.6 nm to 200 mm. and give the porosity data from the intruded volume. Therefore, MIP would be a good method to quantify pore size and volume of swollen hydrogel. However, like SEM-based techniques, MIP also needs the dehydration and/or fixation procedures prior to the examination, inevitably affecting the morphology and pores structure of a hydrogel. In this section, the theoretical approaches and experimental approaches to characterize the network structure of swollen hydrogels are described. Decisions as to which approach is most appropriate for the loose structure must consider the complexity of sample preparation, gel deformation due to water loss, sample preparation, the instrument operation, and the gel applications

2.3.3 Mechanical properties Mechanical properties of hydrogels are extremely important in selecting a material that is suitable for a specific pharmaceutical application. The theories of rubber elasticity and viscoelasticity are used to understand the mechanical behavior of hydrogels. These theories are based on time-independent and time-dependent recovery of the chain orientation and structure, respectively. The use of these theories makes it possible to analyze the polymer structure and determine the effective molecular weight between crosslinks. Anseth et al.[1996] summarized the dependence of the mechanical properties on various parameters, which mainly include monomer composition, the crosslink density, the degree of swelling, and medium conditions. Altering the composition of comonomers
30

used in preparing hydrogels is the simplest parameter to control the mechanical properties of hydrogels. If the hydrogel is not a homopolymer, increasing the relative amount of physically stronger components leads to an increased mechanical strength of the gels. For instance, replacing acrylates with methacrylates causes the increased stiffness of the polymeric backbone and increased mechanical strength. The change of hydrophilicity of the polymer also alters the mechanical strength of the gels. Some results were reported that the addition of N-vinyl-2-pyrrolidone (NVP) in the copolymer system of HEMA and MMA resulted in a significantly decreased Youngs modulus since the hydrophilic NVP alters the swelling properties of the hydrogel [Lustig et al., 1991; Davis et al., 1989]. Changing the crosslinking density has been utilized to achieve the desired mechanical property of the hydrogel. The higher crosslinking density of the system will result in a stronger gel. However, a higher degree of crosslinking creates a more brittle structure and a lower swelling ratio. Hence, there is an optimum degree of crosslinking to achieve a relatively strong and yet elastic hydrogel [Peppas et al., 2000]. The reaction conditions have the profound effects on the mechanical properties of formed hydrogels. These conditions are summarized as reaction time, temperature, light intensity, and amount and type of solvent. Of most importance are the amount and type of solvent. If a large amount of solvent is used in polymerization, the crosslinking agent prefers to intra-crosslinking than inter- crosslinking, which results in the loose network and the low mechanical strength. The type of solvent or the nature of solvent is also used as the controllable variable for mechanical properties. For example, the ionic strength and pH values alter the reactivity of the monomers, leading to changed mechanical strength. Usually, a highly ionic strength reduces the reactivity of monomers [Baker et al., 1994].
31

Other reaction conditions including reaction time and temperature can be changed to get varied properties. For the photopolymerization, the light intensity and dosage influence the network structure of gels and the mechanical properties [Crump, 2001]. Post-reaction treatment can also work as a variable in manipulating the material strength. Techniques such as the addition of a compound [Philippova et al., 1994] and thermal recycle [Cha et al., 1993] can also be used to change the gel strength. Most previously introduced variables such as monomer composition, crosslinking density, and the reaction conditions, are designed to change the degree of swelling of the hydrogels and thus modulate the mechanical properties. Typically, when the polymers swell in a plasticizing solvent, the glass transition temperature of the mixture decreases and the material becomes weaker. In most hydrogel applications, the swelling conditions are usually predetermined according to the application. If not, the external conditions, such as pH, temperature, ionic strength, pressure, or other swelling moduli, can be controlled to get the desired mechanical properties for specific applications. Common approaches for measuring mechanical properties of hydrogels involve tensile or dynamic mechanical analysis. For most uniaxial tensile testing,

dumbbell-shaped samples are placed between two clamps and one end of the material is pulled away from the other at varying loads and rate of extension. For most cases, hydrogel samples are cut in their equilibrium-swollen state and the sample dimensions must be measured in this swollen state. For tensile testing, hydrogel samples should be immersed in a waterbath that is thermally regulated during the testing. To perform dynamic mechanical testing, the samples are usually prepared in thin strips with square edges and a uniform cross-sectional area through the sample length. Dumbbell-shaped
32

samples are no longer an optimal sample shape. The optimal cross-sectional area of the sample is related to the modulus of the materials. For hydrogel samples, the water loss during the experiment significantly influences the mechanical behavior. With the increase of temperature, water loss becomes more prominent and leads to increased moduli. Water loss can be minimized by coating the hydrogel samples with petroleum gel (effective up to 45C) or silicon vacumm grease (effective up to 85C) [Lustig et al., 1991]. Water loss limits the temperature range for dynamic mechanical testing.

2.4

Applications of Hydrogels in Drug Delivery Hydrogels, as a desired material, have been extremely useful in biomedical and

pharmaceutical applications due to their unique swelling properties and structures. Based on the hydrogel functionalities, these biomaterials can be an excellent candidate for controlled release devices, bioadhesive or targetable devices, and self-regulated release devices. According to the delivery administration, hydrogel-based devices can be used for oral, nasal, ocular, rectal, vaginal, epidermal and subcutaneous applications [Peppas et al., 2000]. This section first summarizes applications of hydrogels for different administrations, including its challenges and current status of development. Hydrogels for gastrointestinal administration are introduced in detail because of their close relationship with the work in this dissertation. This is followed by the trends and perspectives for drug delivery.

33

2.4.1 Peroral drug delivery Oral drug delivery is the most desirable and preferred method of administering therapeutic agents for their systemic effects. In addition, the oral medication is generally considered as the first avenue investigated in the discovery and development of new drug entities and pharmaceutical formulations, mainly because of patient acceptance, convenience in administration, and cost-effective manufacturing process. Because of its enormous market potential, oral drug delivery using controllable hydrogels has attracted considerable attention in the past 20 years. In peroral administration, hydrogels can deliver drugs to four major specific sites: mouth, stomach, small intestine and colon. By controlling their swelling properties or bioadhesive characteristics in the presence of a biological fluid, hydrogels can be a useful carrier for releasing drugs in a controlled manner at these desired sites. Furthermore, the mucoadhesive hydrogels offer an attractive property for drug targeting at certain specific regions, leading to a locally increased drug concentration, and thus, enhancing the drug absorption at the release site.

2.4.1.1 Buccal route Drug delivery to the oral cavity has versatile applications in the local treatment of diseases of the mouth, such as periodontal disease, fungal and viral infections, and oral cavity cancers. To ensure the long-term adhesion of the delivery carrier at specific site and to improve the drug absorption, many types of bioadhesive hydrogels have been considered in the device design since the early 1980s. The typical delivery carrier for buccal route comprises tablets, patches, and ointment. Some of these are already on the
34

market. For example, a double-layered tablet with a bioadhesive layer made of hydroxypropyl cellulose/PAA and a lactose non-adhesive backing layer was introduced in the market by Nagai et al. [1999] for the treatment of aphthous stomatitis. Bouckaert et al. [1993] tested the buccal tablets of miconazole based on a modified starch-PAA mixture. Although these tables showed different mucoadhesion properties, there was no significant difference in the salivary content of miconazole for human volunteers. Nair and Chien [1996] compared patches and tablets of different polymers and different released drugs. Sustained release of all four compounds from mucoadhesive tablets was observed. For systemic drug administration, new buccal bilayered tablets, comprising two layersa drug-containing mucoadhesive layer of chitosan with polycarbophil and a backing layer of ethylcellulose, were developed by direct compression. The double-layered structure design provided a unidirectional drug delivery towards the mucosa, and minimized the drug leakage. The striking feature of this device would be the utilization of an in-situ crosslinking reaction between the cationic chitosan and the anionic polycarbophil, leading to the controlled swelling, prolonged drug release, and an adequate adhesiveness [Remunan-Lopez et al., 1998]. A hydrogel-based ointment can also be utilized as a drug delivery device or a liposome delivery vehicle for the topical treatment of certain diseases in the oral cavity. Compared with the conventional ointment-drug formulations, liposomal formulations within ointment may provide more desirable properties for topical use, such as the reduction of uncontrolled release of drugs into the blood circulation and certain undesirable side effects. Petelin et al. [1998] investigated the pharmaceutical performance of three different hydrogel-based ointments as possible vehicles for liposome delivery
35

into the oral cavity tissues by electron paramagnetic resonance (EPR). Liposome containing mucoadhesive ointments were prepared by simply mixing multilamellar liposomes with each ointment pre-diluted with phosphate-buffered saline of pH 7.4. An EPR study showed that p(MAA-co-MMA) was the most appropriate ointment in terms of liposomal stability in the ointment, transport of liposome-entrapped molecules from the ointment into the oral soft tissues, and washing-out time from oral mucosa or gingvia.

2.4.1.2 Gastrointestinal route The peroral route represents the most convenient route of drug administration, being characterized by high patient compliance. The mucosal epithelium along the gastrointestinal tract varies. In the stomach, the surface epithelium consists of a single layer of columnar cells. A thick layer of mucus covers the surface to protect against aggressive luminal content. This specific site is of minor interest for drug delivery since the low pH and the presence of proteolytic enzymes make the stomach a rather harsh environment. However, there are examples of hydrogel-based devices specially designed to delivery in the stomach. Patel and Amiji [1996] developed stomach-specific antibiotic drug delivery systems for the treatment of peptic ulcer disease using pH-sensitive cationic hydrogels. The hydrogels were composed of freeze-dried chitosan-poly(ethylene oxide)

interpenetrating network. pH-dependent swelling properties and the release of two common antibiotics, amoxicillin and metronidazole were evaluated in an enzyme-free simulated gastric fluid (pH=1.2) and a simulated intestinal fluid (pH=7.2). The rapid swelling and drug release demonstrated by these hydrogel formulations in the lower pH
36

fluid may be beneficial for site-specific antibiotic delivery in the stomach. Amiji et al. [1997] also reported enzymatically degradable gelatin-PEO semi-IPN with pH-sensitive swelling properties for oral drug delivery. The incorporation of gelatin in the IPN made it possible to swell in the acidic pH of the gastric fluid due to the ionization of the basic amino acid residues of gelatin. The small intestine is characterized by an enormous surface area available for the absorption of nutrients and drugs. The most important structural aspect of small intestine is the means by which it greatly increases its effective luminal surface area by folds of mucosa, fingerlike villi, and microvilli. The microvilli region has been referred as the specialized location since regions of the device can be surfaced-modified to incorporate cell-targeting mechanism that localize the vehicles at the specific site of reaction to ensure that the drug diffuses the shortest distance in one direction towards the intestinal epithelium. At the terminal ileum, the Peyer's patches, a particular specialization of the gut-immune system, contain the M cells, which are specialized in endocytosis and processing luminal antigens. The large intestine (colon) has the same cell populations as the small intestine, and its main function is the absorption of water and electrolytes. Aside from being an ideal non-invasive route of drug administration, the peptide and protein delivery through the GI tract is fraught with difficulties around low bioavailability, which results from the pH fluctuation, proteolytic degradation, low transport, and short residence time. The pH fluctuation greatly influences the drug integrity. For example, the high acidity of the stomach fluid can preclude the stability of proteins. And the bile salt secreted from the gall bladder into the small intestine can compromise the protein stability. Therefore, proper protection is required during oral
37

administration of bioactive molecules. Enteric-coated systems have been used in commercial applications for releasing drugs through oral administration [Brogmann et al., 2001]. The encapsulation of drugs within lipid vesicles also has the potential advantage of protection and high drug loading [Park et al., 1997; Gregoiraidis, 1995]. However, a major limitation is that these systems cannot fully protect the drugs and release them at a targeted area with a precisely controllable rate over a long period of time. The use of microspheres or nanoparticles made of pH-responsive complexation hydrogels to protect drugs for site-specific delivery has been of interest. [Lowman et al.,1999; Morishita et al., 2002]. Lowmans group prepared crosslinked copolymer gels of PMAA with graft chains of polyethylene glycol to protect the insulin in the harsh, acidic environment of the stomach before releasing the drug in the small intestine. The insulin-containing p(MAA-g-EG) microparticles demonstrated strong dose-dependent hypoglycemic effects in in-vivo oral administration studies using both healthy and diabetic rats. For a bioactive macromolecule, it is quickly denatured and degraded by proteolytic enzymes in the GI tract. Much work has been done to protect against enzymatic activity by adding protease inhibitors or coating the drug with liposomes and polymeric film. Carbopol 934P and chitosan gels were tested in vivo for their ability to increase the absorption of the peptide when administered intraduodenally in rats [Luellen et al., 1996]. Both polymers increased the absorption of the peptide significantly, probably due to both permeation enhancing and enzyme-inhibition properties. Akiyama et al. [1996] reported novel peroral dosage forms of hydrogel formulations with protease inhibitory activities using Carbopolw (C934P), which has been shown to have an inhibitory effect on the hydrolytic activity of trypsin, and its neutralized freeze-dried
38

modification (FNaC934P). They demonstrated that two-phase formulations had the most profound effect on trypsin activity inhibition. Ramdas et al. [2000] developed an oral formulation for insulin delivery based on liposome encapsulated alginate-chitosan gel capsules to increase the encapsulation efficiency to preserve the insulin stability through the acidic media in the stomach and the enzyme-actively intestinal barrier. In animal studies, it was reported that variable reductions in blood glucose were dependent on factors including the lipid composition, size, surface charge and the physical state of the phospholipid bilayer employed [Choudhari et al., 1994; Kisel et al., 2001]. Besides liposomal approach, coating insulin with a pH-dependent acrylic based biodegradable polymer and its encapsulation in enteric-coated microspheres has also been tried [Musabayane et al., 2000]. Oral administration of insulin encapsulated in biocompatible self-assembled nanocubicles also appears to be effective in animal studies [Chung et al., 2002]. The drug release at specific sites has received much attention. Based on the surface receptors, various targeting molecules are utilized to achieve the local targeting. For instance, a polymer-drug conjugate with an antibody can be recognized by the cell surface antigen for cancer diagnostics and therapeutics [Jelinkova et al., 1999]. For peptides or proteins through GI tract, the drug delivery system (DDS) can bind specifically to the mucosal layer or cell surface to increase the residence time and improve the drug bioavailability. Residence time is an important factor influencing drug transport through the GI barrier. Several groups developed DDS with site-specific delivery for peptides and proteins by the choice of material characteristics and the combination of advanced manufacturing techniques. Dorkoosh et al. [2001] designed a
39

novel DDS for site-specific drug delivery of peptide drugs in the intestinal tract using superporous hydrogels (SPH) and SPH composite polymers, which swell very rapidly by absorption of gut fluids. Thus, the system attached to the intestinal wall and provided a longer residence time for drug release. Shen et al. [2002] reported an intestinal patch design for oral delivery. A longer residence time and unidirectional diffusion were achieved for better drug diffusion through the intestinal barrier by using a mucoadhesive layer of Carbopol/ pectin. Tao et al. [2003] combined microfabrication techniques with the use of mucoadhesive plant lectins to design a microdevice with a long residence time. These mucoadhesive drug delivery systems (MDDSs) have attracted considerable interest because of their sustained drug release profile at the absorption site and increased drug bioavailability due to the intimate contact with the absorbing tissue. However, a major physiological condition, continuous shedding of the mucus, leads to the limited retention of these conventional mucoadhesive devices that can only attach to the surface layer of mucus due to their relatively large sizes [Ponchel et al., 1998]. It is also generally known that gastrointestinal mucus renews completely within a few hours [Rubinstein et al., 1994], which apparently sets an upper limit on the retention time of a mucoadhesive system. In addition, the mucus layer can hinder the diffusion of drugs or drug carriers from the device to the absorption site [Meaney et al., 1999; Khanvilkar et al., 2001]. The GI mucus is a bilayered structure. One of the two layers is on the lumen side and called the loosely-adherent layer because it can be easily sucked away. The other is on the epithelium side and called the firmly-adherent layer since it is tightly attached to the epithelial cells and is resistant to suction. It was reported that the mucus that experiences full renewal in the generally-regarded turnover time might solely be the loosely-adherent
40

layer, and the firmly-adherent mucus probably has a longer turnover time [Stuma et al., 2001; Brownlee et al., 2003]. As a result, longer retention than a few hours may be achieved if a device can penetrate the loosely-adherent layer and adhere to the firmly-adherent mucus layer. Another typical approach to extend the duration time is to reduce the delivery device to micron-sized or smaller. The microvilli region has been referred as the specialized location. Currently, advanced DDS contain components on the micro- and nanoscale, but the devices themselves remain in the macroscale (>1mm). As the scale decreases, micro-fabricated DDS may be delivered by ingestion (<1mm), injected into tissue (<200 m), inhaled (<100m), or released into the systemic circulation (<10nm). To directly deliver the devices into the microvilli extending the residence time, the device scale is required to be 5 m or less. For hydrophilic and macromolecular compounds such as peptides and proteins, which have to be absorbed preferably through the paracellular route, the tightness of the intercellular junctions of the mucosal epithelia forms a very strong absorption barrier [Luessen et al., 1997]. In an effort to increase intestinal absorption of various macromolecules, permeation enhancers have been found to reversibly open epithelial tight junctions. To date, numerous compounds have been reported to have absorption-promoting activity and many researchers have tried to elucidate the mechanisms by which the absorption can be enhanced [Yeh et al., 1994; Lindmark et al., 1998; Kotze et al., 1999]. Nevertheless, the potential local toxicity of the enhancers themselves has made it difficult to apply them to practical use. Only sodium caprate is used as an absorption-enhancing adjuvant in drug products. Another major disadvantage of permeation enhancers is their lack of specificity, opening the possibility
41

that food-borne pathogens and toxins migrate along with therapeutic compounds [Foraker et al., 2003].

2.4.2 Nasal route The nasal route of drug administration is the most suitable alternative of delivery for poorly absorbable compounds such as peptide or protein drugs. The nasal epithelium exhibits relatively high permeability, and only two cell layers separate the nasal lumen from the dense blood-vessel network in the lamina propria. The respiratory epithelium covered by a mucus layer is the major lining of the human nasal cavity and is essential in the clearance of mucus by the mucociliary system. Various structurally different mucoadhesive polymers were tested for their ability to retard the nasal mucociliary clearance in rats [Zhou et al., 1996]. The clearance was measured using microspheres labeled with a fluorescent marker incorporated into the formulation. The clearance rate of each polymer gel was found to be lower than that of a control microsphere suspension, resulting in an increased residence time of the gel formulations in the nasal cavity. Ilium et al. [1994] evaluated chitosan solutions as delivery platforms for nasal administration of insulin to rats and sheep. They reported a concentration-dependent absorption-enhancing effect with minimal histological changes of the nasal mucosa. Oechslein et al. [1996] studied various powder formulations of mucoadhesive polymers for their efficacy to increase the nasal absorption of octreotide in rats. The chitosan delivery systems can reduce the rate of clearance from the nasal cavity, thereby increasing the contact time of the delivery system with the nasal mucosa and providing the potential for raising the bioavailability of drugs incorporated into these
42

systems. Nakamura et al. [1999] described a microparticulate dosage form of budesonide, consisting of bioadhesive and pH-dependent graft copolymers of PMAA and PEG, resulting in elevated and constant plasma levels of budesonide for 8 h after nasal administration in rabbits.

2.4.3 Ocular route The ocular route is mainly used for the local treatment of eye pathologies. Many

physiological constraints prevent a desired drug delivery to the eye due to its protective mechanisms, such as effective tear drainage, blinking and low permeability of the cornea. Therefore, conventional eyedrops containing a drug solution tend to be eliminated rapidly from the eye, and the drugs administered exhibit limited absorption, leading to poor ophthalmic bioavailability (2-10%). Additionally, their short retention often results in a frequent dosing regimen to achieve the therapeutic efficacy for a sufficiently long duration. These challenges have motivated researchers to develop drug delivery systems to provide a prolonged ocular residence time of drugs. The following types of mucoadhesive formulations have been evaluated for ocular drug delivery: viscous liquids (suspensions and ointments), hydrogels, and solids (inserts). Certain dosage forms, such as suspensions and ointments, can be retained in the eye, although these sometimes give patients an unpleasant feeling because of the characteristics of solids and semi-solids. Due to their elastic properties, hydrogels can also represent an ocular drainage-resistant device. In particular, in-situ hydrogels are attractive as an ocular drug delivery system because of their facility in dosing as a liquid, and their long-term retention property as a gel after dosing. Hui and Robinson [1985]
43

introduced hydrogels consisting of crosslinked PAA for ocular delivery of progesterone in rabbits. These preparations increased progesterone concentrations in the aqueous humor four times over aqueous suspensions. Cohen et al. [1997] developed an in situ gel system of alginate with high guluronic acid contents for the ophthalmic delivery of pilocarpine. This system significantly extended the duration of the pressure-reducing effect of pilocarpine. Carlfors et al. [1998] investigated the rheological properties of the deacetylated gellan gum, which gels upon instillation in the eye due to the presence of cations, and indicated that a high rate of the sol/gel transition of in-situ gels results in long precorneal contact times. An approach to ocular inserts was presented by Chetoni et al. [1998] in a study of cylindrical devices for oxytetracycline, made from mixtures of silicone clastomer and grafted on the surface of the inserts with an interpenetrating mucoadhesive polymeric network of PAA or PMAA. The ocular retention of IPN-grafted inserts was significantly higher than the ungrafted ones. An in-vivo study using rabbits showed a prolonged release of oxytetracycline from the inserts for several days.

2.4.4

Rectal and vaginal routes The rectal and vaginal routes are considered to be suitable for the local

application and absorption of therapeutics, although patient acceptability is a variable due to the discomfort arising from administered dosage forms. The drugs are absorbed from these specific sites and into the circulation directly. Thus, the rectal and vaginal routes are useful for drugs suffering heavy first-pass metabolism. Conventional delivery systems at both sites include tablets, foam gels, suppositories. Typical suppositories hitherto adapted as dosage forms are solids at room temperature, and melt or soften at body temperature.
44

However, an uncontrolled release pattern of drugs leads to short residence time at the specific position, and a variation of the bioavailability of certain drugs. Mucoahesive hydrogels may offer a valuable way to overcome the problem in conventional suppositories, providing a sufficient bioadhesive property. Mucoadhesive gel formulations based on polycarbophil have been reported to remain 34 days at the vaginal tissue, providing an excellent vehicle for the delivery of progesterone and nonoxynol-9 [Robinson et al., 1994]. To improve the propranolol bioavailability, Ryu et al. [1999] added certain mucoadhesive polycarbophil and sodium alginate to poloxamer-based thermally gelling suppositories. The largest mucoadhesive force and the smallest intrarectal migration for the suppositories resulted in the largest bioavailability of propranolol. Miyazaki et al. [1998] investigated the potential application of xyloglucan gels with a sol-gel transition temperature of around 22-27C as vehicles for rectal drug delivery. This thermal gelling property provided easy administration at room temperature and a gel status at body temperature. In-vivo rectal administration of indomethacin showed a well-controlled drug plasma concentration-time profile without reduced bioavailability.

2.4.5 Transdermal route A transdermal route has been considered as a possible site for the systemic delivery of drugs. The possible benefits of transdermal drug delivery include ease of access, applying, and easing the delivery, sustained and steady drug release, reduced systemic side effects, avoidance of drug degradation in the GI tract and first-pass hepatic metabolism. Furthermore, swollen hydrogels with a high water content can provide a
45

better feeling for the skin in comparison to conventional ointments and patches. Versatile hydrogel-based devices for transdermal delivery have been proposed. Sun et al. [1997] prepared composite membranes comprising of crosslinked PHEMA with a non-woven polyester support. Depending on the preparation conditions, the composite membranes could be tailored to give a permeation flux ranging from 4 to 68 mg/cm2 per h for nitroglycerin. Gayet and Fortier [1996] reported the use of the BSA-PEG hydrogels containing high water content over 96% as controlled release devices in the field of wound dressing. However, the skin functions naturally as a barrier to foreign substances, preventing the entrance of the majority of drugs. Therefore, researchers are developing various electrically assisted methods to enhance the drug permeation across the skin. The notable technologies include electroporation, ionophoresis, sonophoresis, and laser irradiation [Bellhouse et al., 2003; Mehier-Humbert et al., 2005; Prausnitz et al., 2004]. Several hydrogel-based formulations are being investigated as vehicles for transdermal iontophoresis to obtain the enhanced permeation of hormone [Chen et al., 1996] and enoxacin [Fang et al., 1999]. A methyl cellulose-based hydrogel was used as a viscous ultrasonic coupling medium for transdermal sonophoresis assisted with an AC current, resulting in an enhanced permeation of insulin and vasopressin across human skin in vitro [Zhang et al., 1996].

2.4.6 Trends and perspectives In this chapter, a number of sensitive hydrogels with various applications have been described as novel drug delivery platforms. These polymers, as useful drug carriers, or as safe absorption enhancers, or as improved mucoadhesive hydrogels, are the recent
46

developments in drug delivery platforms for intestinal absorption of drugs. Another trend observed during the past few years is the new methods of preparation of hydrogels with desirable functional groups that may be used in the future for drug delivery applications. For example, novel biodegradable polymers include polyrotaxanes, which are considered potentially useful for molecular assemblies for drug delivery. In the synthesis, choice of

new functional monomers and adjusting of hydrophobicity/hydrophilicity of copolymers can be used to better control the swelling/deswelling behavior of novel gels. Moreover, graft, block, and comb-like copolymerizations offer better advantages and the produced novel gels have the interesting applications for treatment of diabetes, osteoporosis, cancer or thrombosis. Besides the development of novel materials for drug delivery, applications of functional hydrogels as the promising materials have been extended in biomedical and pharmaceutical fields when combined with the advanced manufacturing techniques, such as micro- and nanoscale machining techniques. Drug delivery technology can be brought to the next level by the fabrication of smart materials into miniature devices that are

responsive to the individual patients therapeutic requirements and able to deliver a


certain amount of a drug in response to a biological state. Bures and Peppas [2001] have prepared gels of controlled structure and large biological functionality by irradiation of PEO star polymers. Combined with the techniques of molecular imprinting. Such highly crosslinked gels with the sending/activation mechanism may lead to a variety of new, and robust biomolecular sensing hydrogel networks for drug delivery. Gene therapy with the broad potential has been heavily investigated during last 15 years. Many types of polymers are specifically designed for gene delivery. Gene therapy
47

requires the identification of a therapeutic gene and the transfer of the gene to target cells with high efficiency and without hazard for the patients. Hydrogels are designed to address a specific intracellular barrier based on their stability/degradability, biocompatibility, and sensitivity. Hoffmans group [Pack et al., 2005] developed one class of hydrogel carriers to reversibly control membrane stability in response to sharp pH changes for delivering proteins and nucleic acids to intracellular compartments in gene delivery. To enhance the transfection efficiency of gene into mammalian cells, a new system of plasmid DNA release with a biodegradable hydrogel is described while the biological activity of a plasmid DNA of hepatocyte growth factor is augmented by the use of the release system [Kushibiki et al., 2004]. All these promising applications demonstrate that the use of functional hydrogels is a powerful strategy to improve the controlled drug delivery and may benefit the human being.

48

CHAPTER 3

PHOTOPOLYMERIZATION AND STRUCTURE FORMATION OF PMAA HYDROGELS IN WATER/ETHANOL MIXTURE

SYNOPSIS Hydrogels are a desired material for biomedical and pharmaceutical applications. To better control the synthesized hydrogels for various applications, it is necessary to have a thorough understanding of hydrogel structure and reaction mechanism. In this study, pH-sensitive hydrogel networks consisting of methacrylic acid (MAA) crosslinked with tri(ethylene glycol) dimethacrylate (TEGDMA) were synthesized by free-radical photopolymerization in the water/ethanol mixture. Reaction rate was measured using Photo-Differential Scanning Calorimetry (PhotoDSC) with a modified sample pan designed for handling volatile reagents. A photo-rheometer and a dynamic light scattering (DLS) goniometer were used to follow the changes in viscosity and molecule size of the resin system during photopolymerization. It was found that the rate of polymerization increased and more compact and less swelling gels would form with a higher water fraction in the 50wt% solvent/reactant mixture. This is because the weaker interaction
49

between monomer and solvent gives a higher opportunity for propagation and a higher reaction rate. And the hydrophobic TEGDMA and initiator tend to form aggregates in the higher water solution, contributing to the inhomogeneous microgel formation. This mechanism is conformed by viscosity measurement, DLS analysis, scanning electron microscopy (SEM) observation, and kinetics analysis.

3.1

Introduction Hydrogels are a desired material for biomedical and pharmaceutical applications

due to their unique swelling properties and structures. The highly hydrated structure and good biocompatibility make them suitable for contact lenses, biosensors, artificial organs, and drug delivery devices [Peppas, 1997; Peppas et al., 2000]. In drug delivery, functional hydrogels may release drugs in an aqueous median at regulated rate by controlling the synthesis conditions such as the method of polymerization, the crosslinking ratio, and the solvent composition. Hydrogels are often synthesized by UV photopolymerization [Lu et al., 1999; Ward et al., 2001] and redox polymerization [Hassan et al., 1999]. Photopolymerization is favored because hydrogels can be synthesized at temperatures and pH conditions near physiological conditions and even in the presence of biologically active materials. Furthermore, photopolymerization can be easily controlled by adjusting the dosage and intensity of UV light, and the curing temperature [Crump, 2001]. Photo-Differential Scanning Calorimetry (PhotoDSC) is the most widely used technique to characterize the photopolymerization kinetics. A great deal of research has been carried out using this approach for photocurable materials. However, the application of this technique for
50

highly volatile reagents is limited since uncovered sample pans lead to significant sample loss during measurement. Some researchers applied unsealed polyethylene (PE) films over the sample pan to reduce the sample loss [Ward et al., 2001], while others used the sample weight after the reaction to correct for the measurement error resulting from reagent evaporation [Jakubiak et al., 2000]. The results from such treatments are doubtful because sample loss during the reaction is a time-dependent process. When preparing the carriers for drug delivery, solvents like water and ethanol are often used in the synthesis to control the hydrogel structure. Evaporation of highly volatile solvents like ethanol makes it impossible to study the reaction kinetics using the existing approaches. We have recently developed a modified DSC sample pan [Li et al., 2005]. Sample loss during reaction is minimized, and loaded samples are much more uniform over the sample surface. This new method is applied in this study. To better control the synthesized hydrogels for various applications, it is essential to understand how the polymerization conditions, chemical structure of reactants and their composition, and solvent type and concentration affect the reaction and the resulting properties of hydrogels. A number of studies have reported that varying curing conditions may achieve different gel structures and swelling properties [Lowman et al., 1997; Anseth et al., 1996; Peppas et al., 1991], and the compatibility between the solvent and the resin may affect inter-molecular and primary cyclization of multi-vinyl monomers during the polymerization, and, consequently, the hydrogel properties [Kwok et al., 2003; Elliott et al., 2002; Elliott et al., 2001]. However, there lacks a thorough understanding on the interactions of reaction kinetics, rheological changes, hydrogel structures, and solvent-resin compatibility. In this chapter, PMAA gels synthesized in a water/ethanol
51

mixture were investigated by using a series of analytical tools including PhotoDSC, photo-rheometry, dynamic light scattering goniometry, and scanning electron microscopy of freeze-dried hydrogels.

3.2 3.2.1

Experimental Materials and sample preparation The monomer, MAA (Sigma-Aldrich) and the crosslinking agent, TEGDMA

(Sigma-Aldrich) were used to prepare pH-sensitive hydrogels. For all reactions, the crosslinking agent was presented at a level of 1.0 mole% based on the total mole of monomers. A photoinitiator, 2,2-dimethoxy-2-phenylacetophenone (Irgacure 651, Ciba Specificity Chemicals), was used at 1.0 wt% of the monomer mixture. The free-radical photopolymerization was carried out in a mixed solvent of distilled water and ethanol with varying ratios. The ratio of monomer to solvent was kept at 50:50 (w/w). All reagents, unless specified, were of anylytical grade and were used without further purification. To prepare hydrogel films for the swelling test and structure analysis, 5.0 grams of MAA were mixed with a proper amount of TEGDMA and initiator. An equal weight of solvent mixture was then added. The solution was transferred to a glove box where it was kept under a nitrogen atmosphere. Nitrogen was bubbled through the solution for 20 minutes. Then the mixture was pipetted between two glass slides separated by a Teflon spacer. The thickness of the spacers was 0.3 mm. The setup was then placed under a UV light for photopolymerization at 2.0 mw/cm2. The cured hydrogels were then rinsed in double deionized water for 5 days to remove unreacted monomer, initiator and sol
52

fraction. Subsequently, the monomer-free films were cut into samples with a 5.0 mm diameter for swelling test.

3.2.2 Modification of DSC pans A poly(dimethyl siloxane) (PDMS) curing kit (Sylgard184 silicone kit, Essex Group Inc.) was prepared and dissolved in hexane to form a 0.05 g/ml PDMS solution. About 10 l PDMS solution was placed in the DSC pan, which quickly spread to the inner corner of the pan by capillary forces. After solvent evaporation, the pan was heated at 60oC for 4 hours to cure the PDMS resin. The cured PDMS formed a thin layer of O-ring-like hydrophobic film inside the pan, as shown in Figure 3.1(a). This PDMS ring can prevent the hydrophilic sample from flowing towards the inner corner during sample loading. Through this treatment, the loaded resin sample can form a thin film with uniform thickness, essential for consistent UV irradiation. To minimize the sample weight loss during measurements, the sample pan was further modified as shown in Figure 3.1(b). The PhotoDSC pan was placed face-down and adhere to a layer of photo-safe, double-sided Scotch tape. A small amount of partially-cured HEMA/DEGDMA/PI solution was applied around the outside edge of the pan, which was then completely cured under the UV light. The cured poly(HEMA) formed an edge around the open pan. The Scotch tape in the center above the original pan was removed by a razor. After loading the sample, the pan was covered with a layer of polyethylene (PE) film and sealed by the double-sided Scotch tape along the edge area.

53

(a)

A layer of cured PDMS

(b)
DSC pan HEMA/DEGDMA Double-sided Scotch tape

(i) apply photocurable material around the pan

hv

hv

(ii) photocure the edge

(iii) remove the Scotch tape in the center

PE film to seal the pan cover

(iv) pan sealed by PE film


Monomer solution

Figure 3.1 (A) DSC pan treated with PDMS; (B) Seal of DSC pan [Li et al., 2005].
54

3.2.3 PhotoDSC measurement The reaction kinetics and heat of reaction of PMAA gels were measured using a PhotoDSC (TA 2920, TA Instruments). A UV light source (Novacure, 100W Hg short-arc lamp, EXFO, Mississaugua, Ont., Canada) was used to cure the samples. In order to prevent the weight loss of volatile MAA and ethanol, the DSC pans were physically and chemically modified by using the technique described elsewhere [Li et al., 2005]. We compared the performance of modified sample pans vs. the ones covered with a layer of PE film. A micropipette was used for PhotoDSC sampling (5~7 l), which controlled the sample weight for each test. All measurements were carried out at 30oC and the light intensity was kept at 2.0 mw/cm2. Each run was conducted by purging the sample with nitrogen gas until reaching equilibrium (around 2 minutes), and then UV irradiation was applied to induce the free-radical polymerization. The DPC measured the heat flow per unit mass as a function of time. The rate of polymerization, Rp, was calculated by dividing the measured heat flow per unit mass by the theoretical enthalpy. The units of Rp were fractional double bond conversion per second. Integration of Rp curve versus time provided the conversion as a function of time. It is assumed that in the polymerization of two monomers, the functional groups have equal reactivity. In other words, the theoretical enthalpy derived for a comonomer mixture is an average of the enthalpies of the individual monomers.

3.2.4 Rheological measurement A photo stress rheometer MCR 300 (Physica, Anton Paar) was used to follow the viscosity change during the isothermal photopolymerization. A UV cell, including a top
55

steel plate with a diameter of 50 mm and a bottom plate made of quartz glass, was utilized in this test. The UV light source (Acticure 4000, EXFO, Canada) was illuminated from the bottom. The light intensity on the sample surface was kept at 2.0 mw/cm2. The gap between the two plates was set at 1.0 mm and the shear rate used was 0.1s-1. The gel point was assumed when the relative viscosity, i.e. viscosity of the reactive resin vs. its initial viscosity, reached 104.

3.2.5

Dynamic light scattering analysis Dynamic light scattering (DLS) measurements at 30C were carried out to

determine

the

molecule

size

and

size

distribution

before

gelation

during

photopolymerization by using a BI-DNDC Differential Refractometer (Brookhaven Instruments) with a 10 mW He-Ne laser beam at a wavelength of 633 nm. A scattering angle was held constant at 90 in the measurement. Before the DLS analysis, the partially reacted sample (around 0.3 ml) was dispersed in 3 ml of ethanol, and the diluted solution was then filtrated through a filtration unit with 0.45-micron pore size (Whatman Puradisc 25TF). Count rates between 10 to 200 kilocounts per second were used to obtain meaningful results by changing the sample concentration and adjusting the laser power. Autocorrelation of the intensity was carried out by the method of cumulate analysis to obtain an average diameter of the molecules and the polydispersity. The molecule size distribution was obtained from the correction function by CONTIN analysis using the standard software BI-DNDCW.

56

3.2.6

Swelling studyies The swelling tests were performed at various pH values ranging from 2.6 to 7.4 to

characterize the swelling behavior for synthesized pH-sensitive hydrogels. The buffer solutions with different pH values were prepared by mixing the citric acid with appropriate amounts of sodium phosphate solution. Sodium chloride was used to adjust the ionic strength of all solutions to I=0.1M, which is the near-physiological condition. The dried hydrogel samples were weighed and placed in the buffer solution at room temperature (25C). The samples were taken out of the solution at pre-selected time intervals. After the extra water on the surface was removed by laboratory tissue, the weight of the wet hydrogels was measured. The weight-swelling ratio was calculated by the weight of the swollen sample to the weight of the dried sample. The samples were blotted and weighed until the weight change is less than 0.1 mg over a 24-hour period.

3.2.7 Scanning electron microscopy characterization To visually examine the surface and interior morphology of hydrogels in the swollen state, a Hitachi Model S-4300 SEM was used to analyze the pore structure. The samples cured under UV radiation were first swollen to reach equilibrium in buffer solutions for 24 hours, and then quickly frozen below its freezing point using liquid nitrogen. The sample containers were transferred to a freeze dryer (Labconco 75150, Labconco Inc. Kansas City, MI) and freeze-dried for 48 hours until all solvent was sublimed. The freeze-dried samples were loaded on the surface of an aluminum SEM specimen holder and sputter coated with gold for 40 s (Pelco Model 3 Sputter Coater) before observation. A working distance about 8-10 mm, an accelerating voltage of 10 KV, and a chamber
57

pressure of 10-8 Torr were found to be suitable for obtaining high-resolution images of hydrogel samples. The magnification in this study varied from 2000 to 20,000 depending on the network structure.

3.3

Results and Discussions An important feature of this curing system was the formation of heterogeneous

structure in different solvent compositions, which influenced not only the reaction kinetics and rheological changes of the resin, but also the swelling behavior and network structure of the formed gels.

3.3.1 Kinetics of MAA/TEGDMA photopolymerization To minimize the sample weight loss during DSC measurements, the sample pan was physically and chemically modified. The advantage of such treatment was demonstrated via the photopolymerization of the MAA/TEGDMA system. The measured heat flow by using both modified and un-modified pans is shown in Figure 3.2. With a modified sample pan, an equilibrium state was reached in about 1-2 minutes, and the measurement started at a level close to the zero heat flux. While, with a regular sample pan covered with a layer of PE film, there was a continuous endotherm due to the evaporation of monomers and solvents, leading to a negative starting point for heat flux. Additionally, a longer time was needed to reach equilibrium, which would inevitably cause more weight loss. For systems containing highly volatile MAA and ethanol, a strong competition occurred between sample evaporation and chemical reaction. Consequently, a complete change in the reaction rate profile was observed with the use of
58

an un-modified DSC pan. The sample weights before and after the test showed that there was less than 5% weight loss using a modifies pan, compared to about 40% loss using an un-modifies pan (the data represents the mean of six samples). It is clear that the modified pans have to be used in the DSC kinetic analysis of volatile monomers and solvents. Using the modified pans, the effect of solvent composition on the reaction kinetics of MAA/TEGDMA was investigated. Figure 3.3(A) illustrates the reaction rate

versus reaction time for the isothermal photopolymerization of MAA/TEGDMA (1.0 mole%TEGDMA, 50 wt.% solvent) with different solvent compositions at 30 C and a UV intensity of 2.0 mW/cm2. As can be seen, the solvent composition had a great influence on the reaction kinetics of the photocurable MAA/TEGDMA system. With an increase of the ethanol content in the solvent mixture, the polymerization rate decreased correspondingly, and multiple exothermic peaks were observed on the reaction rate profiles for all cases. A peak occurred at the very early stage of polymerization, followed with a stronger second peak. A higher ethanol content delayed and broadened the first peak and substantially reduced the second peak. It is also noted from the conversion profiles shown in Figure 3.3(B) that the higher ethanol content delayed the time to achieve a high conversion.

59

4 un-modified modified 2 Heat Flow (mw)

-2

-4 0 5 10 Time (min) 15 20

Figure 3.2 Comparison of PhotoDSC measurements by using a modified and an un-modified pan at UV intensity of 2.0 mw/cm2 in the MAA/TEGDMA system (1.0 mole%TEGDMA, 50 wt.% solvent mixture of the 1/1 water/ethanol ratio).

60

The multiple peaks observed in the free radical crosslinking polymerization have been reported for several mono- and divinyl monomers [Jakubiak, 2000; Li et al., 2005; Lai et al., 1997; Horie et al. 1975; Cook, 1993; Anseth et al., 1994]. Horie and coworkers postulated that the double maxima in the reaction rate of MMA/EGDM systems were caused by microgel formation. They attributed the first peak to the Trommsdorff effect in the bulk material while the resin mixture was still homogeneous, and the second one to the Trommsdorff rate acceleration in the microgels. As the polymerization proceeded further, the system viscosity limited propagation and the autodeceleration in the reaction rate occurred, as monomer could not diffuse to the relatively immobile radicals. Such hypothesis has also been used to interpret the occurrence of multiple reaction peaks in the acrylic acid (and N-vinylpyrrolidone) copolymerization with TEGDMA [Jakubiak, 2000], in the photopolymerization of HEMA/glycerin [Horie et al. 1975], in the photopolymerization of a series of oligo(methylene) oxide and oligo (ethylene oxide) dimethacrylates [Cook, 1993], and in the reaction between multifunctional methacrylate and acrylate monomers [Anseth et al., 1994]. Although our kinetics results show a similar trend, the viscosity and molecule size analysis presented in the next section, however, show a different mechanism.

61

(A)
0.006
c

0.005

Water to ethanol weight ratio: 9/1


4/1 1/1

Reaction Rate(1/s)

0.004
b

1/4

0.003

a c

0.002

0.001

0 0 5 10 Tim e (m in) 15 20

(B)
1

0.8

Conversion

0.6 Water to ethanol weight ratio: 9/1 0.4 4/1 1/1 1/4 0.2

0 0 5 10 Time(min) 15 20

Figure 3.3 (A) Reaction rate and (B) conversion versus reaction time for the isothermal photopolymerization of MAA/TEGDMA (1.0 mole%TEGDMA, 50 wt.% solvent) with different solvent compositions at 30C and UV intensity of 2.0 mW/cm2.

62

3.3.2 Viscosity measurement and molecule size analysis In order to evaluate the effect of solvent composition on the polymeric structure formation, rheological and DLS measurements were carried out to follow the viscosity change and the growth of molecule size during photopolymerization. Figure 3.4(A) displays both the relative viscosity and reaction rate as a function of double bond conversion for PMAA gels with different solvent compositions. Approaching the gel point, there was the steep increase of relative viscosity (104). For the gels with the water/ethanol ratio of 1/4, the macrogelation occurred at 9 minutes or around a conversion of 78%. With an increase of water content, the curves of relative viscosity shifted to a higher conversion. Figure 3.4(B) presents the corresponding gel time and gel conversion versus water content based on the weight of solvent mixture. The gelation time was linearly decreased and the gel conversion was increased with the increasing water content. For the system with the highest water content (90 wt.%), it only took around 5.5 minutes to reach the gel point. However, its gel conversion could reach 88%. Figures 3.5(A) and (B) summarize the size distribution of polymers formed during the photopolymerization of MAA/TEGDMA in ethanol. For MAA/TEGDMA with the 1/4 solvent ratio, the double bond conversion was around 78% at the gel point. The macromolecules formed at a conversion of 23% (point a, the first maxima of reaction rate in Figures 3.3A and 3.4A) exhibited a narrow unimodal size distribution, ranging from 5 to 80 nm. The intensity reached the maximum value at 17.5nm. With the reaction progressed to a conversion of 45% (point b, onset of the second autoacceleration in Figures 3.3A and 3.4A), the peak was shifted to 64 nm. In addition, a bimodal size distribution occurred, which contains a very narrow peak (13-32 nm) with the same
63

maximum value at 17.5 nm and a broader and larger size distribution (40-164 nm). A further increase in the conversion to 76% (point c, before macrogelation) showed very large clusters with the size distribution from 83 to 223 nm, while the intensity ratio of smaller molecules decreased significantly. Apparently, most small molecules had converted into larger clusters. Compared with the system with the 1/4 solvent ratio, the size distribution curves for the system with the 9/1 solvent ratio exhibited a similar shape and trend. Increasing the water content in the solvent mixture shifted the polymer size distribution to a larger size. For example, the formed polymer showed a unimodal size distribution at the same conversion of 23%, point a, and a bimodal size distribution around the onset of the second autoacceleration, point b, except that the molecule clusters were large. At a conversion of 86%, point c which was close to the gel conversion, the peak for larger molecules moved to 204 nm and the width of the distribution spread from 136 to 304 nm. Obviously, the resin system with a higher water/ethanol ratio formed larger polymer clusters under the same UV radiation when the reaction approached macrogelation.

64

(A)
0.01
Water to ethanol ratio: 9/1 4/1 1/1 1/4

12000

0.008 Reaction Rate(1/s)

0.006

II

III

IV

V
6000

0.004
a

c b

0.002
a b c

3000

0 0 0.2 0.4 0.6 0.8 1 Conversion

(B)
10 100

Gel Time (min)

90 6 80 4

1/4

1/1

4/1

9/1

70

W ater to Ethanol Ratio

Figure 3.4 (A) Reaction rate and viscosity rise as a function of conversion of MAA/TEGDMA (1.0 mole% TEGDMA, 50 wt.% solvent) with different solvent compositions cured at UV intensity of 2.0 mW/cm2, (B) Gel time and gel conversion versus water/ethanol ratio in the solvent mixture.

65

Gel Conversion (%)

Relative Viscosity

9000

(A)
120
3.00min, 23% (a) 5.50min, 45% (b)

90 Intensity

8.80min, 76% (c)

60

30

0 0 100 200 Diameter(nm) 300 400

(B)
120
2.09min, 23% (a) 4.00min, 51% (b) 5.40min, 86% (c)

90 Intensity

60

30

0 0 100 200 Diameter(nm) 300 400

Figure 3.5 The size distribution of MAA/TEGDMA resin (1.0 %TEGDMA, 50 wt.% solvent) with different solvent ratios of water/ethanol: (A) 1/4 and (B) 9/1 cured at light intensity of 2.0 mW/cm2.
66

3.3.3 Mechanism for gelation It is well known that free-radical polymerization of multifunctional monomers forms heterogeneous polymer networks, leading to microgel formation [Hsu et al., 1993; Chiu et al., 1995; Sun et al., 1997]. Such entities are a result of strong intramolecular crosslinking of the growing macroradicals. Eventually, intermolecular reactions among microgels form the network structure. The relative rates of intra- and intermolecular reactions depend on the initial monomer composition, as well as other reaction conditions. The solvent composition is a major factor influencing the gelation kinetics. According to the experimental results shown in the previous section, the photopolymerzation of MAA/TEGDMA system can be described in five stages: initiation, microgel formation, cluster formation, macro-gelation, and post-gelation. The schematic diagram of structure formation in the MAA/TEGDMA photopolymerization describing the first four stages is given in Figure 3.6. In the first stage, all reactants are mixed together and UV radiation initiates initiator decomposition to form radicals (shown as filled dots). In the MAA/TEGDMA

system with a good solvent, such as the one with a high-ethanol content (ethanol is a good solvent for both hydrophilic MAA and hydrophobic TEGDMA and Irgacure 651 due to its participation in both hydrogen bonding and hydrophobic interactions), a homogeneous solution is formed with uniform distribution of all reactants. While in a poor solvent with a high water content, TEGDMA tends to form a micelle-like structure due to the amphiphilic properties. Its hydrophilic ends prefer to be in contact with the water phase by hydrogen bonding while the hydrophobic area is located in the center,

67

Good solvent (high ethanol content)


=O

Poor solvent (high water content) = =


C C C= C

2nm

=O

C=CC C

=O

C= C C C

C= C C C

=O

= =

=O

C=CC C

= =C

CC=C C

=O

= II) Microgel formation =

= = = = = = = =

= = = = = =

= III) Cluster formation = = = == =

= =

IV) Macrogelation =

=O

MAA

Free radical

=O

C=CC C

CC=C C

Figure 3.6 The schematic diagram of structure formation of MAA/TEGDMA with different solvent qualities.
68

=O

=O

CC=C C

=O

=O

C=CC C

I) Initiation

=O

= =
100nm

TEGDMA

C=CC C

=O C C= C C

=O

=O

C=O C= C C

CC=C C C=CC C CC=C C C=CC C CC=C

=O =O

= C=CC

=O

=
=O

C=CC C

=O

=
C C= C

= =
C C= C C
=O
=O
C C C= C

CC=C C

=O CC=C C

=O CC=C C

20nm

90nm

where most Irgacure 651 molecules are located. This initial structure is verified by the DLS measurement of MAA/TEGDMA mixtures without UV radiation shown in Figure 3.7. In the MAA/TEGDMA system with the 1/4 solvent ratio, no particles were observed in the DLS analysis. On the other hand, in the system with the 9/1 solvent ratio, a peak about 6 nm was observed with or without Irgacure 651, supporting the complex formation by amphiphilic TEGDMA. After initiation, radicals react with monomers to produce monomeric radicals. Because of the presence of multifunctional monomers, the monomeric radicals have chances to link with these molecules to form the growing macroradicals with pendant double bonds, leading to the cyclization or ring formation through intramolecular reactions. The intramolecular reactions consume vinyl groups, but do not contribute to the increase of molecule weight and macroscopic network formation. This internal crosslinking on the primary polymer chains leads to the formation of microgels [Dusek et al., 1980]. Inside the microgels, the Trommsdorff effect may occur because termination is largely hindered due to immobilized polymerical radicals, while the propagation rate is less affected since small MAA monomers are still mobile. This leads to a small peak or shoulder in the early stage of the reaction profiles. However, the relative viscosity remains nearly unchanged. The greater extent of intramolecular cyclization means less intermolecular crosslinking. This leads to larger mesh size in formed hydrogels, and the weaker mechanical properties. This mechanism of intramolecular cyclization has been used to explain the network formation influenced by the light intensity [Li et al., 2005], the solvent concentration [Elliott et al., 2001], the solvent quality [Kwok et al., 2003; Elliott et al., 2002], and the curing temperature [Chiu et al., 1995].
69

120

6nm
90 Intensity

MAA/TEGDMA (9/1, no Irgacure 651) MAA/TEGDMA (9/1, Irgacure 651) MAA/TEGDMA (1/4, Irgacure 651)

60

30

0 0 20 40 60 Diameter(nm) 80 100

Figure 3.7 The size distribution of MAA/TEGDMA monomer solution (1.0 %TEGDMA, 50 wt.% solvent) with different compositions.

70

In the solvent mixture, it is favorable for ethanol to participate in the formation of hydrogen bonding with MAA molecules. Thus, more ethanol indicates stronger interaction with the MAA molecules. According to the theory of complex [Henrici-Olive et al., 1965], the propagating macroradicals continually interacts with the surrounding medium (i.e. monomer and solvent). The stronger the interaction between the MAA and the solvent, the lower the overall rate of polymerization since the propagation can only take place if the propagating macroradical is in the vicinity of the monomer molecules. Therefore, the high ethanol content in good solvent system gives a less opportunity for propagation and a lower reaction rate under the UV radiation. Additionally, the uniform distribution of TEGDMA and radicals increase the distance between radicals and free vinyls or pendant vinyls, resulting in a high extent of intramolecular cyclization and smaller microgels with loose structure. On the other hand, there is a higher reaction rate of adding monomers onto the growing radicals and a fast microgel formation in the poor solvent. And the localized TEGDMA and radicals leads to a high extent of intermolecular crosslinking and larger microgels with smaller mesh size. The solvent composition has little effect on the solution viscosity at this stage since microgel formation does not significantly affect bulk properties in the solution. During the cluster formation stage (stage III), the reactive microgels with pendant double bonds may react with free monomers and other microgels to form larger clusters, resulting in a bimodal molecular size distribution. The Trommsdorff effect in the clusters leads to the second autoacceleration in the reaction profiles. At the later part of this stage, the presence of a larger number of clusters and the inter-connection of some clusters lead to an increase of solution viscosity.
71

As a macroscopic polymeric network is formed by chemical or physical crosslinking, the resin system reaches the gel point in stage IV. Approaching the gel point, most small microgels have converted to the larger clusters and intermolecular reactions among these clusters finally lead to macrogelation. For the transition from microgels to macrogels, intermolecular crosslinking reactions require the displacement of neighboring solvent molecules from the vicinity of the microgels. In the system with a higher water content, the microgels can easily form larger aggregates at a higher reaction rate due to the weaker interaction between the microgels and solvent mixture. Therefore, the MAA/TEGDMA with the 9/1 solvent ratio exhibited the shortest gel time and the highest gel conversion as shown in Figure 3(B). While the uniformly distributed smaller microgels in a system with a higher ethanol content have less chance to connect with each other, taking longer time to reach the gel point. As the system entered the post-gelation stage (V), the reaction rate abruptly decreased since both propagation and termination became diffusion limited.

3.3.4 Swelling ratio and structural characterization Figure 3.8 compares the equilibrium swelling ratio (SR) in different pH buffer solutions for hydrogels synthesized with various solvent compositions. In all cases, the hydrogel samples swelled more at higher pH due to the electrostatic repulsion between the ionized forms of the carboxylic segments, as well as the dissociation of hydrogen bonds between the carboxylic acid groups of MAA and the oxygen of the ether groups of TEGDMA and the hydrophilicity of ionized molecules. Below a pH of 6.0, the swelling ratio drastically decreased, indicating the hydrogel was in a relatively collapsed state
72

mainly due to the formation of hydrogen bonding.

It is also interesting to note that the

gels with the highest ethanol content had the highest swelling ratio for a specific pH value and its value reached approximately 33 at a pH of 7.3. SEM technique is useful to reveal hydrogel structure, although the pre-treatment of dehydration and/or fixation procedures for SEM examination may affect the morphology of a hydrogel [Hong et al., 1998]. As shown in Figure 3.9, the pore structures of the swollen interior of PMAA hydrogels are different depending on the solvent composition. Figure 3.9(A) presents the SEM micrograph of PMAA hydrogel with the 9/1 solvent ratio. In a pH=7.4 buffer solution, this hydrogel (SR=10.0) exhibited mostly circular and elliptical pores with smaller pores. Its pore size varies from very small to very large pores, which may be a result of inhomogeneous reaction during photopolymerization. On the other hand, the swollen gel with the 1/4 solvent ratio in the same buffer solution showed larger and more uniform pores as shown in Figure 3.9(B). Figure 3.10 shows the different morphology of swollen PMAA gels with the same swelling ratio (SR=4.3) in the freeze-dried state. To obtain the same swelling ratio, the gels with the solvent ratios of 1/4 and 9/1 were immersed in buffer solutions with the pH values of 3.0 and 6.2, respectively. The gel with a higher water content displayed

smaller pores and much thicker pore walls at the same SR value. These results are consistent with the solvent effect discussed in the previous section. The localized reactants contribute to the formation of highly crosslinked network structure in the poor solvent, leading to the smaller pores with thicker wall, while the uniformly distributed reactants in a good solvent lead to a looser network structure, forming larger pores with thinner wall.
73

35 Weight Swelling Ratio (g / g) 30 25 20 15 10 5 0 2 3 4 5 pH 6 7 8 Water to ethanol 1/4 ratio: 1/1 4/1 9/1

Figure 3.8 Equilibrium swelling ratios of the PMAA (1.0 mole% TEGDMA) hydrogels with different solvent ratios as a function of pH values.

74

(A)

SR=10.0

25 m

(B)

SR=33.0

25 m
18Jul 05 s13 1.8k 25 um

Figure 3.9 SEM micrograph of swollen PMAA hydrogels (1.0 mole% TEGDMA, 50 wt.% solvent) with different swelling ratios (SR) in pH=7.4 buffer solution: (A) 9/1 and (B) 1/4.

75

(A)

SR=4.3

5 m

(B)

SR=4.3

5 m

Figure 3.10 SEM micrograph of swollen PMAA hydrogels (1.0 mole% TEGDMA, 50 wt.% solvent) with the same swelling ratio (SR=4.3) in different buffer solution: (A) 9/1 in pH=6.2 buffer (B) 1/4 in pH=3.0 buffer.

76

3.4 Conclusion This work clarified the role of the solvent composition in the photopolymerization of hydrogels. The solvent composition has a great influence on the reaction kinetics of photocurable MAA/TEGDMA system. With the increase of the ethanol content in the solvent mixture, the photopolymerization rate and the gel conversion decreased, while the gel time and the swelling ratio of PMAA hydrogels increased. This can be explained by the solvent compatibility and interaction with the reactants and the initiator. A less ethanol content indicated less compatibility of TEGDMA and initiator and weaker interaction between MAA and solvent. This weaker interaction led to a higher reaction rate and faster gel formation. The less compatibility resulted in localized TEGDMA and initiator distribution. Since the localized TEGDMA contributed to more highly crosslinked microgels, the resulting hydrogel had a lower swelling ratio and less uniform pore distribution. This mechanism has been confirmed by viscosity measurement, dynamic light scattering analysis, and SEM observation.

77

CHAPTER 4

PHOTOPOLYMERIZATION AND STRUCTURE FORMATION OF PMAA HYDROGELS CURED AT VARIOUS LIGHT INTENSITIES

SYNOPSIS

Hydrogels are a desired material for biomedical and pharmaceutical applications due to their unique swelling properties, the highly hydrated structure and good biocompatibility. To better control the properties of synthesized hydrogels, it is necessary to have a thorough understanding of hydrogel structure and reaction mechanism. The solvent effect on the reaction kinetics and structure formation of pH-sensitive hydrogel networks comprising a PMAA backbone crosslinked by TEGDMA has been discussed in the previous chapter. In this chapter, the effect of light intensity on the reaction kinetics and structure formation is addressed. A series of analytical tools including PhotoDSC, photo-rheometry, and DLS goniometry were used for this study. The kinetics-gelation mechanism based on the concept of microstructure formation is also discussed.

78

4.1

Introduction Hydrogels with the highly hydrated structure and good biocompatibility have

been employed as contact lenses, artificial organs, and drug delivery devices [Peppas, 1997]. The volumetric shape memory capability makes hydrogels an ideal choice as actuator, fluid pump, and valves in microfluidic devices [Osada et al., 1993; Seigel et al., 1991]. In an aqueous environment, hydrogels will undergo a reversible phase transformation that results in dramatic volumetric swelling and shrinking upon exposure and removal of a stimulus, such as pH value. Typically, pH-sensitive hydrogels contain carboxylic groups capable of uptaking a large amount of water above its pKa. Such polymers mainly include poly(acrylic acid) (PAA) and poly(methacrylic acid) (PMAA). Copolymers of PAA and PMAA with poly(ethylene glycol) (PEG), poly(vinyl alcohol) (PVA), and PHEMA also exhibit the pH sensitivity due to the presence of carboxylic segments. Additionally, incorporating other sensitive groups into the networks of PAA or PMAA may give gels more interesting properties. For example, the copolymer of PAA and PMAA with PNIPAAm can provide the environmental sensitivity of both pH and temperature [Tian et al., 2003; Zhang et al., 2000]. Recently, a series of smart biomaterials such as poly(ethyl acrylic acid) (PEAA) and poly(propyl acrylic acid) (PPAA), has opened new opportunities for applications in the molecular imaging field because of their sharp pH-sensitivity [Mourad et al., 2001]. PH-sensitive hydrogels exhibit swelling or deswelling behavior with changes of pH values due to one of the following mechanisms: (1) changes in the hydrophobic-hydrophilic nature of chains, (2) inter- and intramolecular complexation by hydrogen bonding, or (3) electrostatic repulsion. All these mechanisms are closely related
79

to the protonation phenomena of the ionizable moieties on the polymer backbone or the side chains. The kinetics of the swelling process and the equilibrium extent of swelling are affected considerably by several factors, such as ionic strength of the medium, buffer composition, and the presence of salts [Hariharan et al., 1996]. Other factors such as the crosslinking ratio, solvent quality, chemical structure of monomers, and reaction conditions during the photopolymerization also influence the structure formation and hydrogel swelling properties. Photopolymerization is a widely used technique to synthesize polymers and hydrogels due to its distinct advantages of rapid cure, low curing temperature, in-line production, low energy consumption, and easy process control. A great deal of research has been carried out to investigate the effect of light intensity on the reaction kinetics of UV-curable materials with the use of PhotoDSC, in which the hydrogel matrix is loaded into an aluminum pan and then exposed to UV irradiation [Cook, 1993; Ward et al., 2001; Li et al., 2005]. In the experiment, evaperation of volatile solvent or reactants may cause significant measurement errors. Recently, Li et al. [2005] reported a technique of modifying the DSC sample pan to minimize the sample loss and improve the accuracy for volatile systems. PH-sensitive hydrogels has the unique swelling/deswelling behavior with changes of pH values in the surrounding medium. The structure formation and hydrogel swelling properties are affected considerably by several factors, such as the properties of the monomer solution (composition, solvent quality, chemical structure), synthesized conditions during the photopolymerization, and the conditions of medium (ionic strength, composition, pH values). A number of studies have reported that varying curing
80

conditions may achieve different gel structures and swelling properties [Lowman et al., 1997; Anseth et al., 1996; Peppas et al., 1991; Kwok et al., 2003; Elliott et al., 2001; Elliott et al., 2002]. The UV light intensity is one of the most important factors that affect the reaction kinetics of the resin systems and the properties of the formed gels. It was reported that an increase of the intensity led to a higher maximum polymerization rate of the acrylate resin systems. The maximum was achieved more rapidly after the start of the reaction and the induction period slightly decreased [Lovell et al., 1999; Scherzer et al., 1999]. The effect of light intensity on the hydrogel system becomes more complex due to the solvent influence. There lacks a thorough understanding on the interactions of reaction kinetics, rheological changes, gel formation, and hydrogel structures as a result of the UV radiation with different light intensities. In this study, PMAA gels synthesized in a water/ethanol mixture are investigated by using a series of analytical tools including PhotoDSC, photo-rheometry, and dynamic light scattering goniometry. The effects of light intensity on the reaction kinetics and structural properties are addressed.

4.2 4.2.1

Experimental Materials and sample preparation The monomer, MAA (Sigma-Aldrich) and the crosslinking agent, TEGDMA

(Sigma-Aldrich) were used to prepare pH-sensitive hydrogels. For all reactions, the crosslinking agent was presented at the level of 1.0 mole% based on the total mole of

monomers. A photoinitiator, 2,2-dimethoxy-2-phenylacetophenone (Irgacure 651, Ciba Specificity Chemicals), was used at 1.0 wt% of the monomer mixture. The free-radical photopolymerization was carried out in a mixed solvent of distilled water and ethanol
81

with the 1/1 ratio. The ratio of monomer to solvent was kept at 50:50 (w/w). All reagents, unless specified, were of anylytical grade and were used without further purification. To prepare hydrogel films for the swelling test and structure analysis, 5.0 grams of MAA were mixed with a proper amount of TEGDMA and initiator. An equal weight of solvent mixture was then added. The solution was transferred to a glove box where it was kept under a nitrogen atmosphere. Nitrogen was bubbled through the solution for 20 minutes. Then the mixture was pipetted between two glass slides separated by a Teflon spacer. The thickness of the spacers was 0.3 mm. The setup was then placed under a UV light for photopolymerization at 0.25~24 mw/cm2. The cured hydrogels were then rinsed in double deionized water for 5 days to remove unreacted monomer, initiator and sol fraction. Subsequently, the monomer-free films were cut into samples with 5.0 mm diameter for swelling test.

4.2.2 PhotoDSC measurement The reaction kinetics and heat of reaction of PMAA gels were measured using a PhotoDSC (TA 2920, TA Instruments). A UV light source (Novacure, 100W Hg short-arc lamp, EXFO, Mississaugua, Ont., Canada) was used to cure the samples. In order to prevent the weight loss of volatile MAA and ethanol, the DSC pans were physically and chemically modified by using the technique described elsewhere [Li et al., 2005]. A micropipette was used for PhotoDSC sampling (5~7 l), which controlled the sample weight for each test. All measurements were carried out at 30oC and the light intensity was varied from 0.25 to 24 mw/cm2. Each run was conducted by purging the sample with nitrogen gas until reaching equilibrium (around 2 minutes), and then UV irradiation was
82

applied to induce the free-radical polymerization. To obtain the kinetic parameters, a series of unsteady state polymerizations was performed. At a given time, the light source was extinguished and the dark polymerization was continuously monitored by the DSC. Along with an expression for the steady state polymerization, an expression for the unsteady state polymerization was used to determine the kinetic parameters as a function of conversion. The details of this experimental technique are available in the literature [Lovell et al., 1999].

4.2.3 Rheological measurement A photo stress rheometer MCR 300 (Physica, Anton Paar) was used to follow the viscosity change during the isothermal photopolymerization. A UV cell, including a top steel plate with a diameter of 50 mm and a bottom plate made of quartz glass, was utilized in this test. The UV light source (Acticure 4000, EXFO, Canada) was illuminated from the bottom. The light intensity on the sample surface was kept at 2.0 mw/cm2. The gap between the two plates was set at 1.0 mm and the shear rate used was 0.1s-1. The gel point was assumed when the relative viscosity, i.e. viscosity of the reactive resin vs. its initial viscosity, reached 104.

4.2.4

Dynamic light scattering analysis Dynamic light scattering (DLS) measurements at 30C were carried out to

determine

the

molecule

size

and

size

distribution

before

gelation

during

photopolymerization by using a BI-DNDC Differential Refractometer (Brookhaven Instruments) with a 10 mW He-Ne laser beam at a wavelength of 633 nm. A scattering
83

angle was held constant at 90in the measurement. Because the formed polymer swells more in water than in ethanol, the ethanol (3ml) was used as a solvent to dilute the partially reacted sample (around 0.3 ml). The diluted solution was then filtered through a filtration unit with 0.45 micron pore size (Whatman Puradisc 25TF) before measurement. Count rates between 10 to 200 kilocounts per second were used to obtain meaningful results by changing the sample concentration and adjusting the laser power. Autocorrelation of the intensity was carried out by the method of cumulate analysis to obtain an average diameter of the molecules and the polydispersity. The molecule size distribution was obtained from the correction function by CONTIN analysis using the standard software BI-DNDCW.

4.2.5

Swelling study The swelling tests were performed in a pH=4.2 (or 7.3) buffer solution to

characterize the swelling behavior of synthesized pH-sensitive hydrogels. The buffer solutions with specfic pH values were prepared by mixing the citric acid with appropriate amounts of sodium phosphate solution. Sodium chloride was used to adjust the ionic strength of all solutions to I=0.1M, which is the near-physiological condition. The dried hydrogel samples were weighed and placed in the buffer solution at room temperature (25C). The samples were taken out of the solution at pre-selected time intervals. After the extra water on the surface was removed by laboratory tissue, the weight of the wet hydrogels was measured. The weight-swelling ratio was calculated by the weight of the swollen sample to the weight of the dried sample. The samples were blotted and weighed until the weight change was less than 0.1 mg over a 24-hour period.
84

4.3

Results

4.3.1 Kinetics of MAA/TEGDMA photopolymerization Using the modified DSC sample pan, the effects of monomer content and UV irradiation intensity on the reaction kinetics of the MAA/TEGDMA resin system were investigated. Figure 4.1(A) shows the polymerization rate versus conversion for MAA/TEGDMA (100/1 mol.%) with 50 or 100 wt.% monomer cured at a light intensity of 5.0 mw/cm2. As expected, decreasing the monomer content diluted the reactant concentration, hence slowed down the polymerization rate. The addition of solvent in the monomer solution significantly changed the reaction profiles. For the bulk resin system, a large exothermic peak was observed, while the resin system with 50% solvent had multiple exothermic peaks on the reaction profile. The first peak (or shoulder) occurred at the very early stage of polymerization. Regardless of solvent addition, the reaction rate vs. conversion profile followed nearly the same path in the beginning. In other words, changing the solvent content had little influence on the early reaction. It was also noted that the addition of solvent allowed the polymerization to achieve a higher final conversion as compared to the bulk condition (conversion of 99% vs. 61%). This is because the resin system with much higher monomer content reacted faster, leading to more buried monomer and consequently lower double bond conversion. To study the effect of light intensity on the reaction kinetics, isothermal reactions were carried out at 30C for MAA/TEGDMA (100/1 mol.%) with 50 wt.% solvent mixtures. The light intensities varied from 0.25 to 24 mw/cm2. Results are shown in Figures 4.2(A) and (B). As the light intensity was raised, the initiation rate and the
85

0.008

0.006 Reaction Rate(1/s)

100% 50%

0.004

0.002

0 0 0.2 0.4 0.6 0.8 1 Conversion

Figure 4.1 Reaction rate vs. conversion of MAA/TEGDMA in the presence of 1% Irgacure 651 with 50 and 100 wt.% monomer content cured under 5.0 mw/cm2.

86

polymerization rate increased. The light intensity significantly influenced the reaction rate profiles (i.e. the size and shape of the exothermic peaks). Under a low light intensity, the first peak was small. It gradually became larger and took place at an earlier time with an increased light intensity. However, the second peak tended to become smaller at a higher light intensity. When the sample was cured at a light intensity larger than 5.0 mw/cm2, the first peak dominated and the second one became a shoulder. A further increase in the light intensity caused the size of the second peak to become even smaller. From the conversion versus time curves presented in Figure 4.2(B), one can see that an increase in the light intensity generally reduced the time required to achieve a high conversion. For example, to reach a conversion of 40%, the time required was shortened from 10.8 to 3.4 minutes when the light intensity increased from 0.25 to 5.0 mw/cm2. However, if the sample was cured at a light intensity larger than 5.0 mw/cm2, a higher reaction rate was observed at the early stage, but the reaction rate became lower later than that at a low light intensity at a later time. Consequently, the time to reach 40% conversion at a light intensity of 24 mw/cm2 was as long as 4.3 minutes. This indicates that too high a light intensity has an adverse effect on the photopolymerization of the resin system. The multiple peaks observed in the free radical polymerization can be explained by microgel formation, which may affect the onset of macrogelation and the curing behavior. Horie et al. [1975] has postulated this hypothesis to explain the occurrence of double maxima in the reaction rate of MMA/EGDM systems: the first peak attributes to the Trommsdorff effect in the bulk material and the second one to the Trommsdorff effect in the microgels.
87

(A)

0.004
2 24mw/cm2 24 mw/cm 2 5.0 mw/cm 5.0mw/cm2 2 2.0 mw/cm 2.0mw/cm2 2 0.25 mw/cm 0.25mw/cm2

0.003
Reaction Rate(1/s)

0.002
a c b b c

0.001

0
0 10 Time(min) 20 30

(B)
1

0.8

Conversion

0.6
2 24mw/cm2 24 mw/cm 2 5.0mw/cm2 5.0 mw/cm 2 2.0mw/cm2 2.0 mw/cm 2 0.25 mw/cm 0.25mw/cm2

0.4

0.2

0 0 10 Time(min) 20 30

Figure 4.2 Effect of light intensity on the polymerization of MAA/TEGDMA system in the presence of 1% Irgacure 651 (A) reaction rate, (B) conversion.

88

4.3.2 Viscosity measurement In order to evaluate the effect of light intensity on the polymeric structure formation, a rheometer equipped with a UV cell was used to follow the viscosity change during the reaction. Figures 4.3(A) and (B) display both the relative viscosity and reaction rate as a function of double bond conversion for MAA/TEGDMA (100/1 mol.%) cured at 0.25, 2.0, and 24 mw/cm2. Approaching the gel point, there was a steep increase of the relative viscosity. At a low light intensity of 0.25 mw/cm2, macrogelation occurred before the maximum of the second peak. As the intensity increased to 2.0 mw/cm2, the gelation point reached the maximum of the second peak. While for a high intensity of 24 mw/cm2, macrogelation occurred near the end of the first peak. Combining these two figures shows that the on-set of macrogelation shifted to a higher conversion when the light intensity increased from 0.25 to 2.0 mw/cm2. Approaching an optimal intensity, the gel conversion reached the maximum. However, if the light intensity was larger than 2.0 mw/cm2, a decreased gel conversion was observed. Figure 4.4 presents the gel conversion versus light intensity. The gel conversion was only 71% when cured at 0.25 mw/cm2, but rose to around 80% at 2.0 mw/cm2, after which the gel conversion significantly decreased. According to this figure, an optimal intensity (2.0mw/cm2) can be used for curing the PMAA hydrogels as drug delivery carriers to minimize the negative effect of residue monomers.

89

(A)
0.004
0.25 mw/cm 2 2.0 mw/cm
2

10000

Reaction Rate(1/s)

6000 0.002 4000 0.001

2000

0 0 0.2 0.4 0.6 0.8 1 Conversion

(B)
0.005
24 mw/cm
2

12000

0.004 Reaction Rate(1/s) Relative Viscosity 9000 0.003 6000 0.002 3000 0.001

0 0 0.2 0.4 0.6 0.8 1 Conversion

Figure 4.3 Reaction rate and relative viscosity rise as a function of conversion of MAA/TEGDMA (1.0 mole% TEGDMA, 50 wt.% solvent) cured at different light intensity (A) 0.25 and 2.0 mW/cm2, (B) 24 mW/cm2.

90

Relative Viscosity

0.003

8000

100

Gel Conversion (%)

80

Optimal

60

40

20 0 6 12 18 24 30
2 Intensity (mw/cm (mw/cm2) Intensity )

Figure 4.4 Gel conversion versus light intensity for polymerization of MAA/TEGDMA system (1.0 mole% TEGDMA, 50 wt.% solvent) in the presence of 1% Irgacure 651.

91

4.3.3 Kinetic parameters Polymerization is often described as a chain reaction with a set of rate constants of elementary reactions among which the most important ones are the rate constants of propagation ( k p ) and termination ( kt ). Photoinitiation is a useful process for determining the kinetic rate constants in free radical polymerization. By monitoring the rate of polymerization during UV-exposure and afterwards in the dark, one can evaluate the rate constants k p and kt . The ratio k p k t0.5 is calculated from rate measurements under steady-state irradiation conditions using the following equation [Decker, 1998]:
1/ 2 d [M ] k p (1) = 0.5 [M ] I o 1 e [PI ]l dt ktb Here, the rate of propagation ( R p ) is directly related to the incident light intensity ( I 0 ),

Rp =

( [

])

sample thickness (l), the absorptivity ( ), concentration of the photoinitiator ([PI] ), and the quantum yield of initiation ( ) (number of initiating species produced per photon absorbed). During the dark polymerization, no more radicals are produced and the rate equation becomes [Tryson et al., 1979]:

[ M ]t k [ M ]i = tb t + (d [ M ] / dt )t k p ( d [ M ] / dt )i

(2)

where i and t refer to the monomer concentration and the rate of polymerization at the onset of the dark reaction and after a given time, respectively. The linear time
0.5 dependence allows one to evaluate the ratio ktb k p . Together with the k p k tp ratio, the

individual values of k p and kt can be determined.


92

Figures 4.5 and 4.6 show the variation of k p and kt with the degree of conversion for the MAA/TEGDMA resin system cured at 2.0 mw/cm2 and 24 mw/cm2, respectively. Under a low light intensity, the propagation and termination processes were reaction controlled at the very beginning of the polymerization in the solvent mixture, so
k p and kt remained relatively constant in Figure 4.5. Above a conversion of 10%, kt

started to decrease gradually. When the reaction reached a conversion of 46%, the termination rate curve leveled off. In the corresponding process, the k p value kept increasing. This phenomenon can be explained by the theory of complex [Henrici-Olive et al., 1962 &1965]. The essence of the theory is the assumption that the propagating macroradicals continually interact with the surrounding medium. In the solution polymerization, the propagating macroradical is surrounded by monomers as well as the solvent molecules. Since the propagation can only take place if the propagating macroradical is in the vicinity of the monomer molecules, the local concentration of monomer molecules influences the rate of solution polymerization and the rate constant for propagation. This hypothesis has been used to explain the variation of the rate constant for propagation in systems containing monomers, such as acrylamide, methacrykaminde, acrylic acid, methyacrylic acid, and their derivatives. To explain the variation of k p and kt at low light intensity, we also need consider the microgel formation. Above the conversion of 10% in this case, the microgel entanglements started to form, although the viscosity of the bulk system showed little change. Inside the microgels, the motion of the macroradicals was restricted due to increased diffusional limitations, leading to a decrease in the overall value of the termination kinetic constant.
93

This entanglement formation made it possible for propagating macroradicals in the bulk system to be surrounded by more monomer molecules. Consequently, the propagation constant gradually increased and the first autoacceleration of the polymerization rate occurred in Figures 4.2(A) and 4.3(A). A further increase in the conversion close to

80% induced a dramatic increase of bulk viscosity. At this point, the propagation rate dropped rapidly since k p also became controlled by diffusion due to the increasing mobility restriction in bulk materials. In contrast, a high light intensity provided more energy for initiator to activate, leading to more free radicals. Because there were so many reactive molecules in the system and the polymerization reacted so fast, both the propagation and termination processes were controlled by the diffusion even at the very beginning of the polymerization. The values of k p and kt dramatically reduced, although the bulk viscosity maintained a relative constant. After the macrogelation (above a conversion of 42%), the values of k p and kt varied with the increasing monomer conversion.

94

1000

Kp (Kt) ( l/mol's)

100

2 kt, 2.0mw/cm2 2.0 mw/cm 2 2.0 mw/cm kp, 2.0mw/cm2

10

1 0 0.2 0.4 0.6 0.8 1 Conversion

Figure 4.5 Conversion dependence of the rate constants k p and k t for the polymerization of MAA/TEGDMA system at 2.0 mw/cm2.

95

100
2 kt, 24 mw/cm2 24 mw/cm 2 24 mw/cm kp, 24 mw/cm2

Kp (Kt) ( l/mol's)

10

0.1

0.01 0 0.2 0.4 0.6 0.8 1 Conversion

Figure 4.6 Conversion dependence of the rate constants k p and k t for the polymerization of MAA/TEGDMA system at 24 mw/cm2.

96

4.3.4 Molecular size analysis

Figure 4.7(A) summarize the molecular size and its distribution of polymers formed during the photopolymerization of MAA/TEGDMA cured at 2.0 mw/cm2. Under this condition, the gel conversion was around 80%. The macromolecules formed at a conversion of 23% (point a, the first maximum of reaction rate in Figure 4.2A) exhibited a narrow unimodal distribution, ranging from 6 to 45 nm. The intensity reached the maximum value at 18 nm polymer diameter. With the reaction progressed to a conversion of 39% (point b, onset of the second autoacceleration in Figure 4.2A), the peak was shifted to 62 nm. In addition, a bimodal size distribution occurred, which contained a relatively narrow peak (11~22 nm) and a larger size distribution (44~87nm). A further increase in the conversion to 78% (point c, before macrogelation) induced a broad size distribution from 116 to 303 nm, while the intensity ratio of smaller molecules decreased significantly. This suggests that most small molecules have converted into larger clusters. The growth of hydrogel particles under UV radiation of 24 mw/cm2 was investigated and is shown in Figure 4.7(B). The size distribution curves exhibit similar shape under this condition. Increasing the light intensity shifted the polymer size distribution to a smaller size. For example, the formed particles showed a unimodal size distribution at the conversion of 9% (point a), and a bimodal size distribution at the conversion of 40% (point b), except that the molecule clusters were small. At a conversion of 42% (point c), which was close to the gel conversion, the peak for larger molecules was at 123 nm and the width of the distribution was from 54 to 212 nm. The resin system cured at a lower light intensity formed larger polymer clusters when the reaction approached macrogelation.
97

(A) 2.0mw/cm2
120

2.50min, 22% (a) 4.23min, 39% (b) 7.80min, 78% (c)


90

Intensity

60

30

0 0 70 140 210 280 350

Diameter(nm)

(B) 24mw/cm2
120

90

0.51min, 9% (a) 3.93min, 40% (b) 5.00min, 42% (c)

Intensity

60

30

0 0 70 140 210 280 350 Diameter(nm)

Figure 4.7 The molecular size distribution of the MAA/TEGDMA system (1.0 mole% TEGDMA, 50 wt.% solvent) cured at (A) 2.0 mw/cm2 and (B) 24 mw/cm2.
98

Batzilla and Funke [1987] used poly(4-vinyl styrene) monomer to synthesize highly crosslinked microgels under different conditions. The viscosity of the reactive system decreased and then increased during polymerization. The initial viscosity decrease was due to the intramolecular cyclization in the beginning of the reaction. As the reaction proceeded, the viscosity increased due to intermolecular crosslinking. Although our overall reaction kinetics followed a similar trend, the viscosity, the kinetic parameters, and molecule size analysis, showed a different mechanism.

4.3.5

Discussion

For the chain crosslinking polymerization, the existence of multifunctional monomers leads to the formation of pendant double bonds on the growing macro-radicals. The pendant double bonds can react with propagation radicals through intramolecular reactions to form cycles, and may also react through intermolecular reactions to form network structures. Therefore, the network formation may coexist with the microgel formation during polymerization. Based to the reaction kinetics, the changes of viscosity, and the corresponding particle formation discussed in the previous sections, the curing process of MAA/TEGDMA system can be described in five stages: initiation, microgel formation, cluster formation, macrogelation, and post-gelation. The schematic diagram of the structure formation in the MAA/TEGDMA photopolymerization at different light intensities is described in Figures 4.8 and 4.9 for the first four stages. In the first stage, all reactants are mixed together and UV radiation initiates initiator decomposition to form free radicals (shown as filled dots). MAA/TEGDMA system with 50 wt.% solvent mixture, a homogeneous solution is
99

In the

Reaction Rate(1/s)

Intermolecular crosslinks

MAA Free radical a c b

0%

20%

40%

60%

80%
V

Conversion
V

Figure 4.8 Changes of reaction rate, viscosity during the photopolymerization of MAA/TEGDMA at light intensity of 2.0 mw/cm2: I initiation; II microgel formation; III cluster formation; IV macrogelation; V post-gelation.

100

Relative Viscosity

Reaction Rate(1/s)

Intermolecular crosslinks

MAA Free radical

c b

0%

20%

40%

60%

80%

Conversion

Figure 4.9 Changes of reaction rate, viscosity during the photopolymerization of MAA/TEGDMA at light intensity of 24 mw/cm2: I initiation; II microgel formation; III cluster formation; IV macro-gelation; V post-gelation.

101

Relative Viscosity

formed with uniform distribution of all reactants since ethanol is a good solvent for both hydrophilic MAA and hydrophobic TEGDMA (Irgacure 651). This is verified by the DLS measurement of MAA/TEGDMA mixtures without UV radiation. According to the measurement, no particles were observed in the DLS analysis. The initiation step of the radical polymerization may be divided into the radical formation and the addition of a monomer to the radical. Since the rate constant for the addition of a monomer to the radical is usually several orders of magnitude higher than the value for the radical formation (primary radicals), the decisive step of the initiation process is the formation of primary radicals. A high light intensity provides more energy for initiator to activate, leading to more formed primary radicals in solution. Therefore, more filled dots are distributed in the proposed diagram in the first stage (Figure 4.9). After the formation of monomeric radicals, the monomeric radicals may link with multifunctional monomers to form the growing macroradicals with pendant double bonds, leading to the cyclization through intramolecular reactions. This internal crosslinking on the primary polymer chains leads to the formation of microgels [Dusek et al., 1980]. Simultaneously, the pendent double bonds may react through intermolecular reaction to form a network structure. The relative rates of the intra- and intermolecular reactions are strongly affected by the monomer composition, solvent concentration and quality, and the curing conditions, such as the temperature and the intensity of incident light. Here, we focused on the influence of light intensity. A high light intensity leads to a faster initiation, more radicals and more pendant vinyls in the system. The concentration of active radicals is relatively high, leading to a faster polymerization rate and a higher possibility for the polymeric radical to cycle by
102

reacting with its own pendant double bonds. Consequently, cyclization may dominate from the beginning of the reaction. The greater extent of intramolecular cyclization means less intermolecular crosslinking, resulting in larger mesh and smaller size of formed particles (Figures 4.7B and 4.9), and the weaker mechanical properties. The propagation rate decreased with the reaction progress due to the comsumption of bulk monomers. However, the Trommsdorff effect inside the microgels may occur because termination is largely hindered due to immobilized macroradicals, Therefore, a large peak was shown in the early stage of the reaction profile (Figures 4.3B and 4.9). On the other hand, at a low light intensity, less radicals are fromed and the reaction rate is low at the beginning. Due to the microgel formation, the Trommsdorff effect may occur becuase termination is diffusion controlled, while the propagating process is still in the reaction-controlled stage in the bulk system. Thus, a small shoulder was observed in the early stage of polymerization (Figures 4.2A and 4.8). In addition, the active radicals prefer to intermolecularly react with the double bonds. Therefore, the formed molecules are generally larger in size with a more compact structure. During the cluster formation stage (III), the reactive microgels with pendant double bonds may react with free monomers and other microgels to form larger clusters, resulting in a bimodal molecular size distribution. At the later part of this stage, the presence of a larger number of clusters and the inter-connection of some clusters lead to an increased viscosity. Approaching the gel point in stage IV, most small microgels have converted to the larger clusters and intermolecular reactions among these clusters finally lead to macrogelation. For the transition from microgels to macrogels, intermolecular
103

crosslinking reactions require the displacement of neighboring solvent molecules from the vicinity of the microgels. In the system cured at a higher light intensity, the dominant intramolecular reaction can form many microgel particles. These microgels can easily form large aggregates and quickly reach the gel point. In contrast, the distributed microgels in a system with a lower light intensity have less chance to connect with each other, taking a longer time to reach the gel point. As the system enters the post-gelation stage (V), the reaction rate abruptly decreases since both propagation and termination become diffusion limited. Obviously, the high light intensity facilitates the cyclization, thus playing a significant role in the overall structure of formed gels. One of the most important physical properties characterizing the hydrogels structure is the weight swelling ratio. Figure 4.10 illustrates this property of PMAA hydrogels cured under different light intensities and immersed in different pH buffer solutions. When the light intensity increased from 2.0 to 24 mw/cm2, the swelling ratio of cured hydrogels only rose from 5.3 to about 5.7 after immersing in a pH=4.2 buffer for 4 hours. In a higher pH buffer (pH=7.3), the difference of the swelling ratio became very significant and increased from 21.4 to 32.8. The structure difference of formed PMAA gels is more easily characterized in higher pH buffer solutions due to the electrostatic repulsion between the ionized forms of the carboxylic segments, as well as the dissociation of hydrogen bonds between the carboxylic acid groups of MAA and the oxygen of the ether groups of TEGDMA. These swelling results are consistent with the particle size and integrated analysis discussed in the previous section.

104

35 24mw/cm2, pH 7.3 24mw/cm2, pH 4.2 28 Weight Swelling Ratio(g/g) 2.0mw/cm2, pH 7.3 2.0mw/cm2, pH 4.2 21

14

0 0 40 80 120 Time(min) 160 200 240

Figure 4.10 Dynamic swelling behavior of the PMAA hydrogels with 1.0% TEGDMA cured at different light intensity and immersed in the different pH buffer solutions.

105

4.4

Conclusions

This work studied the effect of the light intensity in the photopolymerization of hydrogels. The copolymerization of photocurable MAA/TEGDMA system was enhanced as the light intensity increased, especially at the low light intensity range and low conversion. At too high a light intensity, an adverse effect was observed and the final conversion of MAA decreased to 43% at 24 mw/cm2. The optimal light intensity was about 2.0 mw/cm2 to get the PMAA hydrogels with low residue monomers. The use of the high light intensity significantly shortened the reaction time to reach macrogelation and increased the swelling ratio of formed hydrogels, which can be explained by the mechanism for the relative rates of intra- and intermolecular reactions. With a high light intensity, more free radicals and more intramolecular reactions led to a higher reaction rate and faster gel formation. Since the intramolecular reaction contributed to less crosslinked microgels, the resulting hydrogels had a higher swelling ratio.

106

CHAPTER 5

DESIGN OF SMART DEVICES BASED ON THE FUNCTIONAL HYDROGELS

SYNOPSIS

This chapter focused on the design of an assembled drug delivery system (DDS) to provide multifunctions, such as drug protection, self-regulated oscillatory release, and targeted uni-directional delivery by a bilayered self-folding gate and simple surface mucoadhesion. In this device, a pH-sensitive hydrogel together with a poly(hydroxyethyl methacrylate) (HEMA) barrier was used as a gate to control drug release. In addition, PHEMA coated with poly(ethylene oxide) / poly(propylene oxide) / poly(ethylene oxide) (PEO-PPO-PEO) surfactant was utilized to enhance mucoadhesion on the device surface. The release profiles of two model drugs, acid orange 8 (AO8) and bovine serum albumin (BSA) were studied in this assembled system, which compared with the conventional drug-entrapped carriers and enteric-coating systems. Furthermore, targeted unidirectional release was demonstrated in a side-by-side diffusion cell. In conclusion, for such an assembled device, the PHEMA layer not only affects the folding direction but also serves
107

as a barrier to protect the model drugs. The release time can be controlled by the thickness of the bilayered gate and the drug reservoir. Due to the reversible swelling behavior of PMAA gels, the bilayered gate can sense the environmental pH change and achieve an oscillatory release pattern. Moreover, the local targeting and uni-directional release have been successfully demonstrated in vitro.

5.1

Introduction

It would be most desirable for drug release to match a patients physiological needs at the proper time and/or proper site. This is why there is a great interest in the development of controlled delivery systems [Qiu et al., 2001]. Drug delivery technology can be brought to the next level by the fabrication of smart materials into a single assembled device that is responsive to the individual patients therapeutic requirements and able to deliver a certain amount of drug in response to a biological state. Such smart therapeutics should possess one or more properties such as proper drug protection, local targeting, precisely controlled release, self-regulated therapeutic action, permeation enhancing, enzyme inhibiting, imaging, and reporting. This is clearly a highly challenging task and it is difficult to add all of these functionalities in a single device. The objective of this study is to develop an intelligent system for drug protection, self-regulated oscillatory release, and targeted uni-directional release based on hydrogels. Such a system would need to exhibit [Park et al., 1993], serving as drug delivery

carriers for oral, buccal, rectal, vaginal, ocular, epidermal and subcutaneous applications [Petelinet al., 1998; Kitano et al., 1998; Miyazaki et al., 1998; McNeill et al., 1984; Cohen et al., 1997; Draye et al., 1997; Beyssac et al., 1996].
108

Proper protection is required during administration of bioactive molecules. Enteric-coated systems have been used in commercial applications for releasing drugs through oral administration [Brogmann et al., 2001]. The encapsulation of drugs within lipid vesicles also has the potential advantage of protection and high drug-loading [Park et al., 1997; Gregoiraidis 1995]. However, a major limitation is that these systems cannot fully protect the drugs and release them at a targeted area with a precisely controllable rate over a long period of time. The use of microspheres or nanoparticles to protect drugs for site-specific delivery has been of interest [Lowman et al., 1999; Horak et al., 2001; Morishita et al., 2002]. In order to avoid periodic insulin injection, Lowman et al [1999]. prepared p(MAA-g-EG) hydrogel microparticles containing insulin for in vivo oral administration. The hydrogel protects the insulin in the acidic condition of the stomach. However, protein instability resulting from exposure to an organic solvent during loading is a major problem [Li et al., 2000; Sah et al., 1999]. The applications are also limited by organic solvent residues, the complexity of the process, and the need to sterilize the microspheres. Besides proper protection, controlled release and self-regulation of drug delivery are highly desirable in many applications. Self-regulated devices can be classified into substrate-specific and environment-specific devices [Heller 1996]. Makino et al. [1990] developed a sugar-insulin conjugate, which was complexed with the protein Concanavalin A (Con A). Such a device could deliver insulin in response to a change in blood glucose concentration. In order to adjust the release of insulin by a molecule gate system, Hassan et al. [1999] synthesized glucose-oxidase containing gels to convert the pH-sensitivity to glucose-sensitivity. These substrate-specific devices are still under
109

development. In environment-specific devices, devices can directly respond to changes in pH, temperature, ion strength, electromagnetic radiation, ultrasound, and photo or pressure stimulation [Neuberger 2002; Peppas 1991]. Using functional hydrogels as a switch or gate for controlled drug delivery has been explored recently by several researchers [Kaetsu et al., 1999; Cao et al., 2001]. However, these devices either had a very long response time or could not completely stop drug diffusion in non-delivery conditions. The release of drugs at specific sites has received much attention lately. Based on the surface receptors, various targeting molecules are utilized to achieve the local targeting. For instance, a polymer-drug conjugate with an antibody can be recognized by the cell surface antigen for cancer diagnostics and therapeutics [Jelinkova et al., 1999]. For peptides or proteins through the gastrointestinal (GI) tract, the DDS can bind specifically to the mucosal layer or cell surface to increase the residence time and improve the bioavailibity of drugs. Residence time is an important factor for drug transport through the GI-tract barrier. Dorkoosh et al. [Dorkoosh et al., 2001] designed a novel DDS for site-specific drug delivery of peptide drugs in the intestinal tract using superporous hydrogels (SPH) and SPH composite polymers, which swell very rapidly by absorption of gut fluids. Thus, the system attached to the intestinal wall and provided a longer residence time for drug release. Shen et al. [2002] reported an intestinal patch design for oral delivery. A longer residence time and uni-directional diffusion were achieved for better drug diffusion through the intestinal barrier by using a mucoadhesive layer of Carbopol/ pectin. Tao et al. [2003] combined microfabrication techniques with the use of mucoadhesive plant lectins to design a microdevice with a long residence time.
110

The present work focuses on the design of an assembled DDS that can integrate multiple functions in a single system. Specifically, the drug protection and self-regulated oscillatory release were demonstrated by using a bilayered self-folding design of hydrogel. PHEMA coated with a PEO-PPO-PEO surfactant was utilized to enhance mucoadhesion on the device surface for targeted uni-directional release. The release profiles of two model drugs, acid orange 8 (AO8) and bovine serum albumin (BSA) were studied in this assembled system. The results were compared with the conventional hydrogel entrapped with drugs and enteric-coating systems.

5.2

Experimental

5.2.1 Materials

The monomer, methyacrylic acid (MAA) (Aldrich), and a crosslinking agent, tri(ethylene glycol) dimethacrylate (TEGDMA) (Aldrich), were used to prepare pH-sensitive hydrogels [Zhang et al., 2000]. HEMA (Aldrich) was used to prepare neutral hydrogels, while diethylene glycol dimethacrylate (DEGDMA, Aldrich, Milwaukee, WI) was the crosslinking agent [Lu et al., 1999]. Both hydrogels contained 0.01~0.02 mol of crosslinking agent/mol of monomer. A photoinitiator,

2,2-dimethoxy-2-phenylacetophenone (Irgacure 651, Aldrich), was used at around 1 wt.% of the monomer mixture. The swelling tests were performed at pH=3.0 and 7.3 to characterize the swelling behavior of hydrogels. The buffer solutions with different pH values were prepared by mixing the citric acid solution and appropriate amounts of sodium phosphate solution. Sodium chloride was used to adjust the ionic strength of all solutions to I=0.1M. For the swelling test, the dried hydrogel samples were weighed and
111

placed in the buffer solution at room temperature. The hydrogels were taken out of the solution at pre-selected time intervals. After the extra water on the surface was removed by laboratory tissue, the weight of the wet hydrogels was measured. The weight swelling ratio was calculated by the weight of the swollen sample to the weight of the dried sample. The agent used to enhance mucoadhesion was a surfactant, Pluronic F127 Prill (BASF Corporation). The major component of this surfactant is a tri-block polymer PEO-PPO-PEO. Mucin (type III) was obtained from Sigma-Aldrich. Enteric coating materials were prepared from MAA monomer using a low level of crosslinking agent. Two hydrophilic model drugs, acid orange 8 (AO8) and bovine serum albumin (BSA), were purchased from Sigma-Aldrich. Their molecular weights and physical properties are listed in Table 5.1.

Solute AO8 BSA

MW(Da) 386.4 65000

&) Stokes radius (

Solubility in water at 25 C (mg/ml) 1 40

3.4a 34.8

Table 5.1 Physical properties of model drugs.

Note: a

The stokes radius of AO8 is approximately calculated based on the stokes radius

of three different model drugs[Zhang et al., 2000].

112

5.2.2

Device design and drug loading

For most conventional delivery systems, drugs are either entrapped in a polymeric matrix or encapsulated by a protective coating. Besides these simple systems, more complicated DDS can be developed to control the drug release. Decisions as to which type of device is most appropriate for an intended application must consider the need for response time, drug release pattern, cost, safety, and therapeutic uses.
A Entrapped devices

The hydrogel matrix with the entrapped drugs was prepared as follows. First, the hydrogel matrix was prepared by free-radical photo-polymerization at room temperature. 5.0 grams of MAA, together with TEGDMA (crosslinking ratio 0.01) and 1.0 wt% Irgacure 651, were mixed at the ambient temperature. The monomer mixture was diluted with a solvent mixture of 50 wt% double deionized water and ethanol to make a 50 wt% monomer solution. The monomer solution was then injected between two glass slides separated by teflon spacers with 0.8 mm in thickness and exposed to a low intensity 365 nm UV light at a light intensity of 1.8 mw/cm2 for 20 minutes under nitrogen flow. The cured hydrogels were then rinsed in double deionized water for 5 days to remove unreacted monomer, initiator and sol fraction. Subsequently, the monomer-free disks were cut into samples with a 5 mm diameter and 0.8 mm thickness. These hydrogel disks were placed in a 10 ml buffer solution with pH of 7.3 and AO8 concentration of 0.3 wt% for 24 hours to load the model drug, then dried to a constant weight in a vacuum oven at 37C. In addition to AO8, bovine serum albumin was selected as a model protein drug with a large molecular size. A dried and weighed hydrogel sample was placed in 10 ml of 2.0 wt% BSA solution and allowed to swell for 2 days at 2~4C under gentle shaking.
113

The swollen hydrogel sample was wiped dry using laboratory tissue and weighed, then dried to a constant weight at room temperature.
B Assembled devices

The assembled device consists of two parts: a drug reservoir with targeting function and a bilayered hydrogel gate. The drug reservoir was made of PHEMA gels, which were prepared by the same approach described in the previous section. The gate (5.0 mm in diameter and 60 m in thickness) was made of two partially cured layers using different hydrogels, PHEMA and PMAA. The bilayered gate and the drug reservoir loaded with drug were bonded together by photo-polymerization of the residual monomer in the bilayered gate under UV light as shown in Figure 4.1. For BSA loading, a photomask was used to cover the area loaded with drug to prevent protein denaturing by UV light. In order to completely remove the residual monomers, the loaded area was totally cured using a large dose of high intensity light before bonding with the reservoir, while the circle area of bilayered gate was masked. After loading, the residual monomers within the area of the bilayered gate and the reservoir were cured with a large dose to ensure the complete conversion of hydrogels. By using photo-differential scanning calorimetry, it was found that the conversion for HEMA monomer is higher than 98% at a light intensity of 3.2 mw/cm2 for 10 minutes. The concentration of the solvent during polymerization determines the homogeneous or heterogeneous structure of the gel produced. In this study, the pH-sensitive hydrogels, PMAA, were synthesized with 50% distilled water to achieve a good balance between high mechanical strength and a high swelling response to pH changes. PHEMA hydrogels were prepared with 40% distilled water to ensure the optical
114

transparency of homogeneous hydrogels [Lu et al., 1999]. For comparison, an enteric-coating was also used as the drug release gate in the DDS. MAA was mixed with a very small amount of crosslinking agent TEGDMA at a concentration of 0.3 mol%. Irgacure 651 was added around 1wt% of the monomer mixture. The monomer mixture spread on a microscopic slide was exposed to UV light for 20 minutes under nitrogen flow. The film was quickly washed by DI water several times to remove the unreacted monomer and then dissolved in a pH=7.3 solution to form a homogeneous solution. The solution was poured in a petri dish and dried in a vaccum oven at 37 C overnight to form an enteric-coating layer. This layer was bounded to the DDS following the same procedure as that used for the bilayered gate.

5.2.3 In vitro drug release

AO8 release from the hydrogel systems was measured by monitoring its absorbance at 495 nm using a UV-vis Spectrophotometer (Varian Cary UV-Visible Spectrophotometer). Drug release tests were performed in a buffer solution with pH values of 3.0 and 7.3. Hydrogel devices with 5 mm diameter were placed in 30 ml of buffer solution at room temperature (25C) and subjected to constant shaking. At pre-selected time intervals, 2.5 ml buffer solution was taken out of the vials for the UV test, then placed back into the vials. The concentration of AO8 in the buffer solution was obtained from a calibration curve, and the amount of AO8 release at time t (Mt) was calculated from accumulating the total AO8 release up to that time. The fractional drug release, Mt/M0, could then be calculated. Here M0 is the amount of initially loaded AO8. For the BSA release experiment, protein concentrations were measured by monitoring
115

their absorbance at 270 nm by the same UV-vis Spectrophotometer. Due to the reversible swelling response of PMAA to pH changes in the aqueous environment, the bilayered gate can offer self-regulated release. To demonstrate this function, the assembled device was immersed in a buffer solution with pH=3.0 at 25 C. UV-vis Spectrophotometer monitored the absorbance changes of the buffer solution. After 10 minutes, the device was transferred to a pH=7.3 buffer for 10 minutes. This cycle was repeated three times.

5.2.4 Diffusion studies

A side-by-side diffusion cell made by CNC machining was used to measure the permeability and the diffusion coefficient of AO8 and BSA through hydrogel layers. The hydrogel layers were swollen in pH=7.3 buffer solutions until reaching an equilibrium state, then cut into a disc shape 2.2 cm in diameter and placed between the two cells (the effective diffusion area was 2.83 cm2). Subsequently, 8 ml of 0.3 mg/ml AO8 (or 5 mg/ml BSA) solution was injected into the donor cell (Cell A), while 8 ml buffer solution without any model drug was simultaneously injected into the receptor cell (Cell B). The cells were subjected to constant shaking at room temperature (25C). At predetermined time intervals, 2.5 ml buffer solution was taken from Cell B for UV Spectrophotometry test [Zhang et al., 2000].

5.2.5 Targeted unidirectional release

Besides drug protection, the carrier design for many gene-, vaccine-, and proteinbased drugs must offer local targeting. There are many bioadhesive agents for
116

site-specific targeting. As an example, a mucoadhesive agent was used to simulate the targeting function for the assembled device. The same principle can be applied for other bioadhesive agents. Conventional emulsification or enteric-coating techniques provided a similar targeting function. However, drugs tend to release in all directions after targeting. In contrast, this assembled device can provide targeted uni-directional release because only the releasing surface of the device is modified. Enhanced mucoadhesion was achieved by UV-curing of HEMA with 5wt% PEO-PPO-PEO surfactant as shown in Figure 5.1. The targeted unidirectional release of a food dye AO8 was measured by video in a side-by-side diffusion cell. 25mg mucin was gently blended with 500mg distilled water to form a homogeneous solution, which was then evenly spread over a 25mm diameter millipore membrane and allowed to dry at room temperature to create a mucin-coated membrane. Prior to testing, a digital camcorder was set to record the drug targeting and release. Subsequently, the device was placed in the donor cell and the mucus-coated membrane was then placed between the two cells. 8 ml of pH=7.3 buffer solution was simultaneously injected into both cells and the set-up was subjected to constant shaking at 25C.

117

Preparation of release gate

Surface targeting Poly(HEMA)

UV exposure (partial cure)

Glass slide MAA HEMA/5% PEO-PPO-PEO

UV exposure (partial cure)

HEMA P(MAA-g-EG) Bilayered gate


PEO PEO

PEO

PEO

Loaded drug

PEO

PEO

Device assembling

Photomask

Figure 5.1 Schematic of the assembled device.

118

5.3

Results and Discussion

5.3.1. Swelling properties of hydrogels

The delivery device is based on the swelling properties of different hydrogels. The dynamic swelling behaviors of the two hydrogels in different buffer solutions are shown in Figure 5.2. As can be seen, the dried hydrogels swell at all pH conditions due to the adsorption of water into the porous structure. However, compared with PHEMA hydrogels, PMAA hydrogels have a much more sensitive response. In the high pH buffer solution, the PMAA hydrogels swell rapidly and can achieve a much higher equilibrium swelling ratio than PHEMA hydrogels. This is because ionization of the carboxyl groups (the pendent group of MAA) occurs as the solution becomes less acidic, resulting in dissociation of the hydrogen bonds between the carboxylic acid groups of MAA and the oxygens of the ether groups of TEGDMA. The dissociation of hydrogen bonds,

combined with the electrostatic repulsion force, causes the hydrogel network to swell quickly, thus more water is imbibed into the hydrogels and a higher swelling ratio is obtained. On the other hand, PHEMA is a neutral hydrogel, which has no ionizable groups on its side chain. With a change of pH values, this material exhibits very small swelling in buffer solutions. In addition, since the solvent content in the HEMA monomer solution (40 wt.%) was less than that in the MAA solution (50 wt.%), PHEMA hydrogels should have a more compact structure than PMAA gels with the same crosslinking ratio. Although DEGDMA has a shorter chain than TEGDMA, its contribution could be neglected when considering the low amounts of crosslinker.

119

24 Swelling Ratio (g/g) 20 16 12 8 4 0 0 30 60 90 120 Time (min) 150 180

Figure 5.2 Dynamic swelling behavior of hydrogels. Samples were 5.0 mm in diameter and 0.8 mm in thickness. ( ) PMAA hydrogel in pH=7.3 buffer. ( ) PMAA hydrogel in pH=3.0 buffer. ( ) PHEMA hydrogel in pH=7.3 buffer. ( ) PHEMA hydrogel in pH=3.0 buffer.

120

5.3.2

Model drug release from entrapped devices

There are two general loading methods for entrapped hydrogels as drug carriers. In one method, the mixture of monomer, initiator, crosslinking agent, and model drug was cured by free-radical photopolymerization to form a hydrogel matrix with uniform entrapment of the model drug. However, two major drawbacks limit this methods application. One is the UV adsorption of model drug, which inhibits the hydrogel polymerization, thus limiting the amount of loaded drug. The other drawback is drug instability. The carried drugs, such as peptides and proteins, become unstable under the UV light. Therefore, a different method is usually adopted, which overcomes the disadvantages of the direct curing method. In this method, cured hydrogels are allowed to swell to an equilibrium state in a drug solution, and then dried to obtain the drug-loaded hydrogel matrix. However, the long drug loading time is the major drawback of this method. In this experiments, the second approach was used to make the entrapped samples. In order to investigate the effect of gel structure on the drug release, such model drugs as AO8 and BSA were entrapped into the gel matrix in pH 7.3 buffer solutions. For small molecular AO8, 24-hour loading time was long enough to reach an equilibrium state and homogeneous distribution. While for large molecule, even 48-hour loading time is not long enough to get a homogeneous distribution. Based on the picture of confocal Microscopy (not shown here), the closer to the surface the distance, the more entrapped BSA. Figure 5.3 represents the AO8 release from 5mm entrapped samples with different crosslinking agent. According to the data, it was concluded that the AO8 entrapped into the lower crosslinking agent could fast release AO8, which corresponds to the swelling
121

behavior of PMAA. The experimental results in Figure 5.4 illustrate the effect of crosslinking agent on the BSA release pattern. As expected, the gels with lower crosslinking agent swell quickly and release the BSA with a fast release rate. Moreover, all curves show a typical first-order release behavior: an initial high release rate followed by a declining drug release rate. To compare the release behavior of drugs with different sizes, AO8 and BSA release from 5mm entrapped samples are presented in Figure 5.5. As can be seen, under the acidic condition (pH=3.0), AO8 is released very slowly. The concentration gradient drives AO8 release from the polymeric matrix to the buffer solution. At neutral condition, the pH-sensitive hydrogel is capable of imbibing a large amount of water, enlarging the mesh size and causing AO8 to be easily released. As shown in Fig. 5.5, about 40% AO8 can be released after 150 minutes at 25C. In the experiment, the BSA loading concentration was about 6 times higher than that of AO8 in order to be easily detected by UV spectroscopy. This figure also shows the BSA release profile from the entrapped hydrogels in pH=7.3 buffer solutions. The BSA release profile can be divided into two stages: an initial fast release for 60 minutes, followed by a slow release. This release profile may be explained as follows. In the first stage, due to the compact structure of swelling hydrogels, the predominant transport is due to the movement of hydrogel chains. Therefore, BSA and AO8 have a similar release profile in the first 60 minutes. After 60 minutes, the drug size becomes the major factor dominating the drug release rate. The BSA release rate becomes about 43% that of AO8 because large molecules usually diffuse slower than small molecules.

122

1 0.75%TEGDMA 1.00%TEGDMA 2.00%TEGDMA 0.6

AO8 Fractional Release

0.8

0.4

0.2

0 0 30 60 Time(min) 90 120

Figure 5.3 Acid Orange 8 release to pH 7.3 buffer solution from the entrapped 5.0 mm PMAA samples at 25 C. The samples were 0.8 mm in thickness.

123

1 BSA Fractional Release 0.8 0.6 0.4 0.2 0 0 30 60 Time(min) 90 120 0.75%TEGDMA 1.00%TEGDMA 2.00%TEGDMA

Figure 5.4 BSA release to pH 7.3 buffer solution from the entrapped 5.0 mm PMAA samples at 25 C. The samples were 0.8 mm in thickness.

124

1 0.8 0.6 0.4 0.2 0 0 30 60 90 120 Time(min) 150 180

Figure 5.5 AO8 and BSA release from the entrapped 5.0 mm PMAA samples at 25 C. The samples were 0.8 mm in thickness. ( ) AO8 at pH=3.0. ( ) AO8 at pH=7.3. ( ) BSA at pH=7.3.

Fractional Release

125

5.3.3

Diffusion studies

To further investigate the transport behavior of model drugs with different sizes in the hydrogel matrix, permeation experiments of AO8 or BSA across the hydrogel layers were carried out as described. Permeability can be calculated by the following equation [Schwarte et al., 1998]:

ln(1

2Ct 2A )= Pt C0 V

(1)

Here, Ct is the solute concentration in Cell B at time t, C0 is the initial solute concentration of Cell A, V is cell volume, A is the effective area of permeation, and P is the membrane permeability coefficient. By plotting (V/2A)*ln[1 - 2(Ct/C0)] versus time t, the slope is the permeability coefficient. The diffusion coefficient can be obtained from the permeability P, the solute partition coefficient Kd, and the membrane thickness L in the swollen state. Their relationship is shown in the following equation:

Dm =

PL Kd

(2)

To determine the diffusion coefficient, the solute partition coefficient, Kd, needs to be calculated from the experimental data by the following equation:

Kd =

Cm C V = ( 0 1) 0 + 1 Cs Cs Vm

(3)

Here, Cm is the concentration in the membrane at equilibrium, Cs is the concentration in the surrounding solution at equilibrium, C0 is the initial concentration in the surrounding solution, V0 is the initial solution volume, Vm is the solution volume in the membrane at
126

equilibrium, and Kd is a measure of the solubility of the solute in the membrane. A low value of Kd means that a solute molecule is not easily soluble in the membrane. A high value of Kd indicates that there may be binding between the solute and the polymer, thus the solute molecule can be easily soluble in the membrane phase. The permeability is defined as a particular solute through a particular membrane. The solute size, membrane mesh size, pH, temperature, and the affinity of the solute with the membrane may affect the permeation of the solute. In this experiment, the temperature and pH were maintained constant. Two model drugs with significantly different sizes were used in the experiment. The hydrodynamic radius of BSA is about 10 times larger than that of AO8. Figure 5.6 shows the solute permeation of AO8 and BSA through swollen PMAA and PHEMA membranes at 25C in pH=7.3 buffer solution. As can be seen, (V/2A)*ln[1 - 2(Ct/C0)] increases linearly with time. The slope of each linear curve represents the permeability for a particular solute. As expected, for PMAA membrane, the permeability of AO8 is higher than that of BSA. And, for a solute like AO8, a PMAA membrane with a larger mesh size in the swollen state has a much higher permeability than a PHEMA membrane.

127

0.15 0.12

-V/2A*ln(1-2Ct/ C0)

0.09

0.06 0.03 0 0 50 100 150 Time(min) 200 250 300

Figure 5.6 Permeation of AO8 and BSA through different swollen hydrogel membranes at pH 7.3 and 25 C. ( ) AO8 through PMAA. ( ) BSA through PMAA. ( ) AO8 through PHEMA.

128

Based on the permeability and partition coefficient, the diffusion coefficient can be calculated as listed in Table4.2. The diffusion coefficient of AO8 (MW= 386.4g/mol) within the PMAA film matrix is 2.03 10-6 cm2/s. Zhang et al. [2000] investigated the release kinetics of oxprenolol HCl (MW= 302g/mol) from a swollen

poly(MAA-g-NIPAA) hydrogel (weight swelling ratio=18.2) at 25C in pH=7.3 buffer solution. The reported diffusion coefficient was 4.68 10-6 cm2/s. Therefore, this measured AO8 diffusion coefficient in the PMAA film matrix is reasonable. The BSA diffusion coefficient in the PMAA film matrix at 25C was 8.00 10-7 cm2/s, which is close to the BSA diffusion coefficient estimated by Mariah et al. [2001]. The diffusion coefficient of BSA within the PMAA hydrogel matrix is about 40% that of AO8. This agrees with the measured results that the BSA release rate is 43% that of AO8 through the same hydrogel matrix.

Permeability P Membrane PMAA PMAA PHEMA Model drug AO8 BSA AO8 10 5 (cm / s ) 2.83 0.33 0.17

Partition coefficient Kd 0.99 0.35 0.99

Diffusion coefficient 10 7 (cm 2 / s ) 20.03 8.00 0.67

Table 5.2 Permeability and diffusion coefficient of model drugs through different membranes.

129

The mesh size of the polymer matrix also affects the diffusion coefficient. As shown in Table 4.2, the diffusion coefficient of AO8 in the PMAA film matrix is about 30 times larger than that of AO8 in the PHEMA film matrix. This is due to significantly different polymeric structures at equilibrium. At pH=7.3, the PMAA hydrogel has a much looser structure than the PHEMA hydrogel and the model drug can diffuse quickly and easily in the matrix. The large BSA molecule is not easily soluble in the membrane, while AO8 can be entrapped in the hydrogel matrix easily.

5.3.4 Model drug release from assembled devices

The entrapped device based on pH-sensitive hydrogels can control the drug release rate under different conditions. However, it has the disadvantages of low drug-loading efficiency and a long loading time. Small molecules can be entrapped in a hydrogel matrix easily and quickly due to their stable structure, high solubility, and large diffusion coefficient. However, macromolecular drugs such as proteins cannot be easily entrapped. In addition, these molecules are very sensitive to the environment. Therefore, an assembled device was designed to solve these problems.

5.3.4.1 Drug protection

In this design, a controlled release was achieved by self-folding of the bilayered hydrogel. The PHEMA layer is a major factor to control the drug release. First, its swelling property influences the folding direction with increasing pH values. In pH=3.0 medium, PMAA hydrogels have a similar swelling response as PHEMA, thus the bilayered gate would not open for drug delivery. With increasing pH, the swelling ratio of
130

the PMAA hydrogel layer increases significantly, while the PHEMA layer has a relatively constant swelling ratio independent of pH values. Therefore, the bilayered gate folds outward until the bonding between the gate and the reservior breaks. As a result, the model drug can be released quickly. Figure 5.7 describes the AO8 release from the assembled device at pH=7.3. In this system (Fig. 5.7A), AO8 particles were loaded in the reservoir. After the device was placed in the buffer solution, the bilayered gate started to fold outward due to water imbibing into the device and a small amount of AO8 was released from a small interstice (Fig. 5.7B). With increasing time, the interstice became larger and larger. After 80 minutes, the swelling properties of hydrogels caused the gate to fold like a roll (Fig. 5.7C). Figure 5.7D shows the schematic of AO8 release from the side view. Since the PHEMA layer has a much lower permeability to the model drug than the PMAA layer, it also serves as a barrier to protect the proteins through the stomach. Figure 5.8 presents the AO8 release from the 5.0 mm assembled device with different gates at pH=3.0 and 25C. As can been seen, in pH=3.0 medium, AO8 fractional release was nearly zero after 4 hours for the bilayered gate. Actually, after 24 hours, AO8 could not be released in pH 3.0 buffers based on the experiment. However, for PMAA gate, the gate did not have sufficient mechanical strength to resist the enlarged volume of drug solution, such that the PMAA gate was broken in the center and AO8 started to release quickly from the device at 160 minutes. This protection function can protect biomolecules from gastric acids and enzymes in the stomach.

131

5.0 mm

5.0 mm

5.0 mm

(A)

(B)

(C)

(D)

Figure 5.7 AO8 release from the assembled device at pH=7.3 and 25C. The diameter of the device is 5.0 mm. The thickness of bilayered gate is 60 m and the thickness of the drug reservoir is 1.0 mm. (A) Dry assembled device. (B) Releasing at t= 40 minutes. (C) Released at t= 80 minutes. (D) Schematic of AO8 release from assembled device.

132

0.8 Fractional Release

0.6

0.4

0.2

0 0 30 60 90 120 150 180 210 240 time(min) Time(min)

Figure 5.8 AO8 release from the 5.0 mm assembled devices with different gates at pH=3.0 and 25C. The gate thickness is 60 m and the reservoir thickness is 1.0mm. ( PMAA hydrogel gate. ( ) PHEMA and PMAA bilayered gate.

133

Figure 5.9 shows AO8 and BSA release from the 5 mm assembled device at room temperatures. In pH=3.0 medium, there was no drug release for 2 hours. In pH= 7.3 buffer solution, it took about 40 minutes to open the device at 25C and then reached 90% drug release quickly. Compare with the AO8 release, BSA had a similar release pattern, which confirms that the release mechanism of this device is based on hydrogel folding and is independent of the model drug size. The thickness of the bilayered gate and the thickness of the drug reservoir also influence the release time as shown in Figure 5.10. When the thickness of the gate was reduced from 90 m to 60 m, the open time would be reduced to about 20 minutes. Decreasing the thickness of the drug reservoir would reduce the bonded area between the bilayered gate and the drug reservoir. Thus, the model drugs would be released more quickly. Because the controlling mechanism is based on the hydrogel swelling behavior, not the height of the drug reservoir, the height can be varied to adjust the amount of drug loading.

134

0.8 Fractional Release

0.6

0.4

0.2

0 0 30 60 Time(min) 90 120

Figure 5.9 AO8 and BSA release from the 5.0 mm assembled device at 25C. The thickness of the bilayered gate is 60 m and the thickness of the drug reservoir is 1.0 mm. ( ) AO8 at pH=3.0. ( ) AO8 at pH=7.3. ( ) BSA at pH=7.3.

135

0.8 Fractional Release

0.6

0.4

0.2

0 0 30 60 time(min) Time(min) 90 120

Figure 5.10 Thickness effects of the bilayered gate and reservoir on AO8 release behavior at pH=7.3 and 25 C. ( ) The gate thickness is 60 m and the reservoir thickness is 0.5 mm. ( ) The gate thickness is 60 m and the reservoir thickness is 1.0 mm. ( ) The gate thickness is 90 m and the reservoir thickness is 0.5 mm.

136

5.3.4.2 Self-regulated oscillatory release

Since PMAA gels possess unique swelling properties, the device can sense the environmental pH change and provide oscillatory release behavior. To demonstrate the oscillatory regulation, the device was tested with a varying pH field. Figure 5.11 presents the oscillatory release behavior of this device. It is evident from the graph that a pulsatile release rate was obtained when the pH was increased from 3.0 to 7.3 due to self-folding of the bilayered gate. When the pH decreased from 7.3 to 3.0, the small value of the release rate indicates that the bilayered gate has reversed to its flat shape and blocked the drug release. In contrast, the assembled device with an enteric gate can only provide a one-time irreversibly pulsatile release profile. This gate design has the limitation that the response time is in minutes. By controlling the chemical structure of hydrogels, gate thickness, and the bilayer ratio, the response time can be reduced to seconds. By using various stimuli-sensitive hydrogels, assembled devices can be activated by pH, temperature, pressure, ionic strength, electromagnetic radiation, buffer composition or the concentration of glucose [Peppas 1991].

5.3.4.3 Targeted unidirectional release

Spatial localization of the therapeutic payload in the target regions is very important for high bioavailability of the administrated drug for therapeutic uses. Different targeted molecules can be attached to the surface of delivery devices by covalent or non-

137

0.02

12

Release Rate (mg/min)

0.015

0.005

0 0 10 20 30 Tim e (m in) 40 50 60

Figure 5.11 The oscillatory release behavior of the assembled device. The gate thickness is 50 m and the thickness ratio for PHEMA to PMAA layer is 4.

138

Buffer pH

0.01

covalent binding to improve the device bioadhesion. Typical examples of bioadhesion include mucoadhesive hydrogels for mucosal route of delivery, plant lectins for mucosal route, and carbohydrate antibody for cell surface receptors. The mucoadhesion was demonstrated in this study by modifying the targeted area with photo-curing of HEMA/5% PEO-PPO-PEO surfactant. Pappes has proposed to the enhancement of mucoadhesion by tethered chains of poly(ethylene glycol) (PEG) grafted on a polymer backbone. Along the polymeric structure of this surfactant, each domain plays a specific role in the resulting surface function: the hydrophobic PPO backbone prefers to interpenetrate in the PHEMA hydrogels, while the hydrophilic tethered chains of PEO act as adhesion promoters to enhance mucoadhesion due to tether diffusion. The addition of bioadhesive polymer chains increases the entanglement between the polymer and mucus network, resulting in strong interaction binding [Ascentiis et al., 1995]. This surface modification prolongs the residence time at delivery sites and improves drug absorption. Figure 5.12 compares the targeted unidirectional release with an untargeted release in the side-by-side diffusion cell. As shown in Figure 5.12A, the device can attach on the mucin-coated membrane due to the mucoadhesive modification on the device surface. When the bilayered gate self-folded, the imbibed water in the reservoir pushed the dissolved drugs from the donor cell to the receptor cell through the membrane. On the other hand, the unmodified device could not attach to the membrane surface and the released AO8 was in the donor cell as shown in Figure 5.12B.

139

(A)

(B)

Figure 5.12 The comparison of the targeted uni-directional release with untargeted release: (A) Targeted release. (B) Untargeted release.

140

5.4

Conclusions

There is considerable interest in the development of smart therapeutics DDS for bioactive drugs. Such a desirable carrier needs to offer multiple functions in a single device. An assembled DDS was demonstrated in this study to achieve multifunctions such as drug protection, self-regulated oscillatory release, and targeted uni-directional delivery by a bilayered hydrogel design and simple surface mucoadhesion. A PHEMA layer not only affects the folding direction but also serves as a barrier to protect the model drug. A cylindrical drug reservoir design provides easy loading of large amount of drugs. The release time can be controlled by the thickness of the bilayered gate and the thickness of the drug reservoir. Due to the reversible swelling behavior of PMAA gels, the bilayered gate can sense the environmental pH change and achieve an oscillatory release pattern. Surface modification with PEO chains can act as adhesion promoters to enhance the device mucoadhesion. The local targeting and uni-directional release have been successfully demonstrated in vitro. Based on the self-folding mechanism, optimization of gate and device design, as well as the proper choice of hydrogel materials, the DDS described in this study has the potential to provide the desired release pattern for a broad range of therapeutic uses. For biomolecular delivery used in inter- and intra-vascular applications, the device need be reduced to micron-sized or smaller. Current work focuses on the design of miniaturized DDS by using polymer micro-fabrication and integration techniques.

141

CHAPTER 6

AN ORAL DELIVERY DEVICE BASED ON SELF-FOLDING HYDROGELS

SYNOPSIS

A self-folding miniature device has been developed to provide enhanced mucoadhesion, drug protection, and targeted unidirectional delivery. The main part of the device is a finger like bilayered structure composed of two bonded layers. One is a pH-sensitive hydrogel based on crosslinked poly(methyacrylic acid) (PMAA) that swells significantly when in contact with body fluids, while the other is a non-swelling layer based on poly(hydroxyethyl methacrylate) (PHEMA). A mucoadhesive drug layer is attached on the bilayer. Thus, the self-folding device first attaches to the mucus and then curls into the mucus due to the different swelling of the bilayered structure, leading to enhanced mucoadhesion. The non-swelling PHEMA layer can also serve as a diffusion barrier, minimizing any drug leakage in the intestine. The resulting unidirectional release provides improved drug transport through the mucosal epithelium. The functionality of this device is successfully demonstrated in vitro using a porcine small intestine.
142

6.1

Introduction

Many protein- and DNA-based drugs exhibit high sensitivity to the surrounding physiological conditions as a result of their delicate physicochemical characteristics and the susceptibility to degradation by proteolytic enzymes in biological fluids. They need to be properly protected during administration and their release needs to be precisely targeted and controlled. Typically, the intramuscular or intravenous injection is used for the administration of peptides and proteins. However, due to the undesirable nature of this method, such as pain, inconvenience and inconsistent pharmacokinetics, other routes have been considered. They include pulmonary, oral, nasal, buccal, rectal, ocular, vaginal, and transdermal delivery [Kopecek et al., 1998], among which oral administration is the most convenient and ideal route. Although oral administration is a non-invasive route of drug delivery, peptides and proteins delivery through the gastrointestinal (GI) tract remains a highly challenging task because of their low bioavailability resulting from the pH fluctuation, proteolytic degradation, low transport efficiency, and short residence time. Enteric-coated systems have been commercially used for releasing drugs through oral administration [Brogmann et al., 2001]. The encapsulation of drugs within lipid vesicles also has the potential advantage of drug protection and high drug loading [Gregoiraidis, 1995]. The inclusion of enhancers/promoters, protease inhibitors, and/or specific adhesion may help the diffusion of large molecules across the epithelial membrane. However, a major limitation is that these systems cannot fully protect the drugs and release them in a targeted area with a precisely controllable rate over a long period of time. Mucoadhesive drug delivery systems (MDDSs) have attracted considerable
143

interest because of their sustained drug release profile at the absorption site and increased drug bioavailability due to the intimate contact with the absorbing tissue. MDDSs typically present in the form of symmetric micro- and nano-spheres or asymmetric patches. Mucoadhesion occurs through surface-to-surface contact. Micro-/nano-particles prepared by phase separation, microemulsion and spray drying have been successfully used as drug delivery carriers. [Jain, 2000; Langer, 2000; Li, 2000]. These particles usually have polydisperse sizes and relatively simple structures. Additionally, the symmetric shape leads to drug release to all directions. Recently, several research groups have made efforts to design patch-like asymmetric delivery devices with functionalities such as drug protection and targeted unidirectional release [Dorkoosh et al., 2002; Shen et al., 2002; Whitehead et al., 2003; Tao et al., 2004; He et al., 2004]. However, the surface-to-surface adhesion for all these systems leads to the limited residence time due to the continuous shedding of surface mucus. In this study, a novel particulate-like miniature device is developed based on the integration of a number of micro-manufacturing modules such as soft-lithography, micro-imprinting, and polymer self-folding. Approaches that are able to improve oral bioavailability, such as protective coating, mucoadhesive binding and mechanical grabbing are also applied in the device design.

6.2 6.2.1

Experimental Materials

The pH-sensitive hydrogel was prepared from the monomer, methyacrylic acid (MAA, Sigma-Aldrich), and a crosslinking agent, tri(ethylene glycol) dimethacrylate
144

(TEGDMA, Sigma-Aldrich). Hydroxyethyl methacrylate (HEMA, Sigma-Aldrich) crosslinked with diethylene glycol dimethacrylate (DEGDMA, Sigma-Aldrich) was used to prepare the non-swelling hydrogel. Both hydrogels contained 0.01mol of crosslinking agent/mol of monomer. A photoinitiator, 2,2-dimethoxy-2-phenylacetophenone (Irgacure 651, Aldrich), was used at 1 wt% of the monomer mixture. The free-radical photopolymerization of MAA/TEGDMA system was carried out in a water/ethanol mixture ( 1vs.1 ratio). The ratio of monomer to solvent during synthesis was 50:50 (w/w). The HEMA/DEGDMA system was polymerized in a water solution with a 40 wt.% solvent ratio. Poly(dimethylsiloxane) (PDMS) resin was purchased from Dow-Corning. A degradable poly( -caprolactone) (PCL) and a water-soluble poly(vinyl alcohol) (PVA) were purchased from Sigma-Aldrich. Carbopol 934 was purchased from BF Goodrich (Cleveland, OH). All reagents, unless specified, are of analytical grade and were used without further purification. Two hydrophilic model drugs, acid orange 8 (AO8) and bovine serum albumin (BSA) were also purchased from Sigma-Aldrich. Fresh porcine small intestines were collected from The Ohio State University Lab Animal Resource.

6.2.2

Device design and fabrication

The device mainly consists of three functional layers: a backing layer, a foldable bilayer (a swelling layer/a non-swelling layer), and a mucoadhesive layer entrapped with drugs (shown in Figure 6.1A). The swelling bilayer was made of MAA crosslinked by TEGDMA and the non-swelling layer was HEMA crosslinked by DEGDMA. Soft-lithographic techniques were used to produce hydrogel bilayered microstructures. The devices were fabricated following the procedures shown in Figure 6.2. A PDMS
145

mold with a desirable surface pattern was made by casting a prepolymer and a curing agent at 10:1 weight ratio onto a complementary relief structure from the standard photolithographic process [Guan et al., 2005; Xia et al., 1998]. The HEMA monomer solution was brushed onto the PDMS mold with an applicator. The solution was trapped in the discrete wells due to discontinuous dewetting. After being subjected to UV radiation for 10 minutes, the MAA monomer solution was brushed onto the cured PHEMA layer to prepare a bilayered structure under another 15-minute UV radiation. A high light intensity and large dosage were applied to ensure high monomer conversion (around 99%). Our experimental observations showed no loosening or separation between these bilayers. This is because the MAA solution diffused into the PHEMA layer before the PMAA layer was solidified. To remove the residue monomer and unreacted initiator, distilled water was used to continuously wash the cured structures covered by a 10 m-thick isopore membrane in the wells for 2 hours. To take out the bilayered structures, the PDMS mold was placed on a PHEMA film covered on a glass slide by briefly exposing the film to water vapor generated from a hot water bath. A solid weight (50g/cm2) was placed on the PDMS mold for 10 minutes. The mold was then removed with the bilayered structures stuck to the PHEMA/glass slide. In this study, the model drug was mixed with Carbopol 934 and PVA to form a drug/mucoadhesive layer. A homogeneous solution of these materials in distilled water (1:1:1, 10wt.%) was brushed onto the PDMS mold. Water was allowed to evaporate and the drug/mucoadhesive layers were formed in the wells. The PDMS mold was then aligned and placed onto the bilayered structures. A solid weight (around 500g/cm2) was placed on the PDMS mold so the sticky drug/mucoadhesive layer would adhere to the bilayered structures due to the
146

compression force. After the drug/mucoadhesive layers were totally dried out in 10 minutes, the PDMS mold was removed. By using this simple approach, we can make both micro- and millimeter sized devices (240 m 4 mm). The typical dimensions of

device used in this study are shown in Figures 6.1(A) and (B). When the device is conveyed into the small intestine, it may directly target onto the small intestine surface due to the Carbopol mucoadhesion. Then the bilayered structures may fold into the mucosa in a grabbing manner, resulting in better drug protection and enhanced mucoadhesion (Figure 6.1C).

147

2.0mm Thin non-adhesive layer (PHEMA, 1~10m) 0.2mm 4.0mm

Swelling layer (PMAA, 50m) Non-swelling layer (PHEMA, 50m) Drug/Mucoadhesive layer (PVA, Carbopol, Drug, 300m)

(A)

(B)

4mm

Small intestine
(C) (D)

Figure 6.1 Schematic of the 3-layer device from (A) side view and (B) top view, (C) folding on the small intestine surface, and (D) a capsule containing devices.

148

1. Non-swelling monomer solution (HEMA/DEGDMA, 60 wt.%) 2. UV partial curing and drying

PDMS Mold

3. Swelling monomer solution (MAA/TEGDMA, 50 wt.%)

4. UV curing

5. Stamping of bilayered structures

Thin PHEMA layer Drug/mucoadhesive layer

6. Stamping of prepared drug layer

Folding bilayer Thin PHEMA layer

Figure 6.2

Fabrication procedure of the miniature devices.

149

6.2.3 Swelling and self-folding studies

To prepare hydrogel samples for the swelling test, a monomer solution was transferred to a glove box under a nitrogen atmosphere. Nitrogen was bubbled through the solution for 20 minutes, then the mixture was pipetted between two glass slides separated by a Teflon spacer. The thickness of the spacer was 0.3mm. The setup was then placed under a UV light for photopolymerization at 2.0 mw/cm2. The cured hydrogels were then rinsed in double deionized water overnight to remove unreacted monomer, initiator and sol fraction. Subsequently, the monomer-free disks were cut into disk samples with 5.0 mm in diameter. Swelling tests were performed at various pH values ranging from 3.0 to 7.0 to characterize the hydrogel behavior in the GI tract. The buffer solutions with different pH values were prepared by mixing the citric acid with appropriate amounts of sodium phosphate solution. Sodium chloride was used to adjust the ionic strength of all solutions to I=0.1M, which is the near-physiological condition. For the swelling test, the dried hydrogel samples were weighed and placed in the buffer solution at room temperature (25C). The hydrogels were taken out of the solution at pre-selected time intervals. After the extra water on the surface was removed by laboratory tissue, the weight of the wet hydrogels was measured. The weight-swelling ratio was calculated by the weight of the swollen sample to the weight of the dried sample. Self-folding of the hydrogel bilayers was observed and recorded in a buffer solution and on the porcine small intestine. All animal procedures were performed based on the institutional protocols.

150

6.2.4

Mucoadhesion measurement

The detachment between the device and a segment of porcine small intestine was measured in a flow trough and a microbalance (shown in Figure 6.3). First, a sacrificed small intestine was longitudinally cut into small pieces (2cm 3cm), sliced lengthwise to spread flat, exposing the lumen side, bonded on the trough bottom by super glue, then washed with 50 ml phosphate buffer saline (PBS) solution. Before the pump drove the buffer solution through the trough, the sample was gently dropped on the intestinal surface. The buffer solution with a high viscosity was prepared by mixing 0.2wt% Xanthan Gum (CP Kelco, Wilmington, DE) in a pH=6.5 buffer for a solution viscosity of 87.9 cp. By controlling the flow rate, the residence time of samples on the intestinal surface was determined through the microscope observation. To prevent the acidic degradation in the stomach, the devices can be loaded in an enteric capsule (shown in Figure 6.1D), so they can maintain the shape until the enteric capsule is dissolved in the small intestine. The flow experiments were carried out to evaluate the device adhesion in the small intestine. Briefly, a 15cm long porcine intestine segment was placed horizontally on a bench top to form a flow channel and one end was connected to a tube so that the lumen could be filled with a pH=6.5 buffer solution at a volumetric flow rate of 1 ml/min. A capsule containing three devices shown in the following figure was placed near the entrance of the tube and pushed into the intestine channel. After 20 minutes, the flow test was stopped and a longitudinal incision was carried out in the intestine to observe the device attachment. The experimental temperature was maintained near 37C.

151

(A)

Microscope

Sample

Pump Buffer Collector

(B)

Buffer solution around the tested sample DCA

Small intestine

Figure 6.3 Experimental setup for (A) flowing testing and (B) the detachment force measurement.

152

The detachment forces were also quantitatively measured by a microbalance attached to a dynamic contact angles analyzer (Cahn DCA-322). A 3.0 cm section of intestine was cut and bonded on a beaker bottom as in the flow test, and covered with pH=6.5 PBS solution at room temperature. The beaker was then placed in the microbalance enclosure and fixed on the stage. A cylindrical sample (the bottom area: 2 mm 2 mm) or a miniature device, mounted on a clamp and hung from the sample loop of the microbalance, was brought in contact with the tissue by moving up the stage. The polymeric sample was left in contact with the tissue for three minutes with an applied force of approximately 100 mN and then pulled vertically away from the tissue sample by moving down the stage while recording the required force for detachment. The mucoadhesion force was normalized by the contact area.

6.2.5 Delivery performance

To evaluate whether the self-folded device has any improved effect on drug protection and transport, targeted unidirectional release was conducted for transepithelium delivery of two model drugs in a side-by-side diffusion chamber. Having rinsed with PBS buffers, the jejunum part of the intestine was cut into a disc shape of 2.2 cm in diameter and placed on a support between the two chambers (the effective diffusion area was 2.83 cm2). Before the experiment, the prepared device (the dimension 4mm4mm, shown in Figures 1A and B) was placed onto the jejunum surface in the donor chamber. Subsequently, 8 ml of pH=6.5 buffer solution was simultaneously injected into both the donor chamber and the receptor chamber at room temperature (25C). The setup was subjected to constant shaking at 180 rpm. At predetermined time
153

intervals, 0.15 ml buffer solution was taken from the receptor chamber for concentration test. To maintain a constant volume, 0.15 ml fresh PBS buffer was added after each sample was withdrawn. AO8 release was measured by monitoring its absorbance at 490 nm using a microplate reader (GS Spectra MAX250). The concentration of AO8 in the buffer solution was obtained from a calibration curve, and the amount of AO8 release at time t (Mt) was calculated from accumulating the total AO8 release up to that time. The fractional drug release, Mt/M0, could then be calculated. Here M0 is the amount of initially loaded AO8. For the BSA release experiment, 0.1 ml samples were taken and replaced by fresh buffer. After accounting for dilution caused by previous measurements, protein concentrations were measured with a Bio-Rad protein assay using the microplate assay protocol. The color change of the dye in response to the concentration change was monitored by measuring the absorbance at 595 nm on the same microplate.

6.3 6.3.1

Results and Discussion Swelling and self-folding studies

The pH-sensitive hydrogel, PMAA has been studied extensively as a promising candidate for oral delivery of peptide and protein drugs through the gastrointestinal tract because of its unique swelling property. Figure 6.4 exhibits the dynamic swelling behavior of the hydrogels in different buffer solutions. As can be seen, the dried hydrogels swelled at all pH conditions due to the adsorption of water into the porous structure. In the high pH buffers, PMAA hydrogels swelled rapidly and achieved a much higher weight-swelling ratio. This was because ionization of the carboxyl groups (the
154

pendent group of MAA) occurred as the solution become less acidic, resulting in dissociation of the hydrogen bonds between the carboxylic acid groups of MAA and the oxygen of the ether groups of TEGDMA. The dissociation of hydrogen bonds, combined with the electrostatic repulsion force, caused the hydrogel network to swell quickly and greatly under an osmotic pressure. Below a pH of 6.5, the swelling ratio drastically decreased to a small value. This implied that the hydrogel was in a relatively collapsed state. On the other hand, PHEMA is a neutral hydrogel, which has no ionizable groups on its side chain. With a change of pH values, this material exhibited very little swelling in buffer solutions. In addition, since the solvent content in the HEMA monomer solution (40 wt.%) was less than that in the MAA solution (50 wt.%), PHEMA hydrogels should have a more compact structure than PMAA gels with the same crosslinking ratio. Although DEGDMA has a shorter chain than TEGDMA, its contribution could be neglected when considering the low amounts of crosslinker. Due to different swelling of the two layers, the bilayered structures would curl in the buffer solutions. To demonstrate the self-folding function, a dried bilayer is shown in Figures 6.5(A) and (B), respectively. The dried bilayer consisted of a PHEMA layer at the top and a PMAA layer at the bottom. The bilayers represent a convex curvature after becoming completely dried out. Figure 6.5(C) shows the folded bilayer in a buffer solution (pH=6.5). It was observed that this structure folded like a fist.

155

24 20
PMAA in pH=7.3 PMAA in pH=6.5

Weight Swelling Ratio (g/g)

16 12 8 4 0 0 30 60

PMAA in pH=3.0 PHEMA in pH=7.3

90 120 Time (min)

150

180

Figure 6.4 Dynamic swelling behavior PMAA and PHEMA hydrogels.

156

Figure 6.5 Optical graphs of a bilayered structure at dried state (A) top view, (B) side view, (C) a curled bilayered structure in a buffer solution. Scale bars=2.0 mm.

157

6.3.2

Mucoadhesion measurement

The layered shape of the device maximizes its contact area with the intestinal wall, while the thin side areas minimize its exposure to the liquid flow through the intestine. Additionally, since the bilayers curl into the mucus in the mode of grabbing, it is expected to provide more resistance to mucus shedding than conventional mucoadhesion. Thus, the residence time can be significantly increased due to the combination of the grabbing adhesion of the folding bilayers and the conventional adhesion of the mucoadhesive layer. This enhanced performance was demonstrated in the flow test. At 5cm height, samples with similar dimensions were randomly dropped on the intestinal surface using tweezers without external force and the flow rate was gradually adjusted from 4.0 to 5.5 ml/s. Figure 6.6(A) summarizes the number of bound samples remaining on the mucus surface as a function of the flow time. For each case, the initially bounded samples were the same. Within three minutes, all samples with a PHEMA surface were washed away at a flow rate of 4.0 ml/s. For the PCL patches (i.e. the drug layer adhered onto a PCL layer) and the folded devices, all samples still stayed on the mucosal surface after 60 minutes. A higher flow rate (5.5 ml/s) was then used in the measurement. According to the Figure 6.6(B), the average residence time for the PCL patch was around 72 minutes. The folded devices showed the longest average residence time, around 103 minutes. To visually demonstrate the folding behavior and enhanced mucoadhesion, a folding device tinted with blue dyes was placed on the mucus surface and a digital camcorder recorded its folding process from the side view. Figure 6.7(A) shows the folding behavior of a bilayered device with each layer having a thickness of 10 m.
158

(A)
4 Number of Bound Samples

0 0 30 60 Time(min) 90 120

(B)
120 100 Residence Time (min) 80 60 40 20 0 PHEMA PCL Patch Folded Device

Figure 6.6

(A) Number of bound samples and (B) residence time for different samples attached to intestinal mucus in the flow test.

159

In the beginning, the device adhered on the mucus surface. Around 2 minutes later, the bilayered structure started to fold into the mucus and at 4 minutes the structure completed the folding. Temperature is a very important factor, which may influence the swelling ratio of gels, response time of folding bilayer, and residence time of the folded device. At the typical body temperature 37C, the swelling ratio of PMAA in pH=6.5 buffer was increased from 10.39 to 11.01 and the response time was improved to 2 minutes as a result of temperature increase. The residence time of the folded device also increased due to the increased extent of folding. Snapshots shown in Figure 6.7(B) describe the device attachment in the flow test. As a control, a PCL patch was also placed on the mucosal surface. At the beginning, both devices attached onto the surface tightly in the flow field. After 65 minutes, the PCL patch started to detach from the surface. Around 70 minutes, the patch was completely washed away from the mucosal surface. Due to the combined effect of mucoadhesion and self-folding, the folded device could stay on the mucus for a longer time. It started to detach at 82 minutes and was finally washed away at approximately 108 minutes. The detachment was due to the mucus shedding, not the unfolding of the bilayered structure. PMAA is a typical mucoadhesive material with a strong detachment force. To ensure that only the drug/Carbopol layer would stick on the mucosal surface, a thin PHEMA layer was added onto the PMAA side. Since the compression pressure from a solid weight (50g/cm2) was weak, this layer could be delaminated from the folded bilayer after the device was immersed in the buffers. The presence of the thin PHEMA layer also offered a delay time for device folding. Figure 6.8 compares the attachment of two devices with different contact sides on the porcine intestine. For the left one (S1), the
160

Folded device

Folded device
4mm

Time (min)

Folded device

Patch 3mm

65

82

108 Time (min)

Figure 6.7 Dynamic processes for (A) folding behavior and (B) enhanced mucoadhesion. Buffer pH=6.5 and 25C.

161

PHEMA-side contacted with the mucus, while the right one (S2) showed the drug/Carbopol layer in contact with the mucosal surface. S1 was washed away immediately at a flow rate of 4.0 ml/s, while S2 stayed on the surface. This thin PHEMA layer was completely peeled off in several minutes (about 10 minutes for this case) when the bilayered arms curled into mucus. When the enteric capsule dissolved in the flow experiment, the devices were able to adhere to the mucos and fold. This experiment was repeated three times. Eight out of nine devices were found adhered to the lumenal wall by the drug/Carbopol-side. The enhanced mucoadhesion of the self-folding device was also revealed in the detachment force measurement. As shown in Figure 6.9, the one-layer PCL and PHEMA samples exhibited very weak adhesion. The major component of Carbopol is acrylic acid, which is a mucoadhesive material. To prepare the sample for the detachment measurement, Carbopol 934, PVA and the model drug were mixed to form a homogeneous solution in distilled water (1:1:1, 10.0wt.%), which was then poured into a petri dish. Water was allowed to evaporate and a drug/mucoadhesive layer was formed. Samples of 2 mm2 mm dimensions were cut for the detachment measurement. The Carbopol/PVA/Drug sample showed a much stronger detachment force, which could be explained by the formation of hydrogen bond due to the carboxylic acid groups [Peppas et al., 1996]. The strongest force was observed for the folded device. These results agree with what was observed in the flow test.

162

2mm
S2 S1

S1: PHEMA-side

S2: Carbopol-side

Time (min) 0

0.5

10

Figure 6.8 Compared attachments for the devices with different contact sides in the flow test. Buffer pH=6.5 and 25C.

163

8 Detachment Force (mN/cm2)

0 PCL PHEMA Carbopol / PVA /Drug Folded Device

Figure 6.9 The detachment force of different samples on the small intestinal surface. Buffer pH=6.5 and 25 C. Error bar = SD, n = 3.

164

6.3.3 Delivery performance

A side-by-side diffusion chamber was used for drug release studies. When the device was attached to the intestinal surface, the drug concentration change in the donor chamber indicated the leakage in the small intestine. Figure 6.10 compares AO8 leakage of delivery systems with different protection layers. Due to good mucoadhesion, a simple PMAA layer could adhere to the mucus surface tightly and the leakage was very low in the beginning. After 60 minutes, the high swelling of PMAA hydrogel, however, led to a very large permeability resulting in severe drug leakage through the protection layer. For the PCL layer, the drug could gradually leak into the donor chamber from the edge of the patch. For the bilayered structure, since the PHEMA protection layer has a lower permeability than the PMAA layer, it served as a barrier to provide protection from drug leakage. Furthermore, the folded structure prevented the leakage from the edges. Consequently, the total leakage from the folded device was very low, less than 30% of loaded drugs after 2 hours. For in vitro drug transport across the mucosal epithelium, we separated the mucosal membrane from the serosal compartment of the small intestine. The isolated mucosal membrane was loaded in the side-by-side diffusion chamber for the diffusion measurement. The drug concentration in the receptor chamber indicates the transferred drugs. Figure 6.11 compares the AO8 transport from different systems across the mucosal epithelium. The squares indicate the homogeneous solution loaded into the donor chamber. The triangles and the circles are for the PCL patch system and the folded device, respectively. All three systems had an equal amount of loaded drug. The figure shows

165

1
PMAA PCL PMAA and PHEMA

Fractional Leakage (Mt/Mo)

0.8 0.6 0.4 0.2 0 0 20 40 60 Time(min) 80 100 120

Figure 6.10 The fractional leakage of AO8 from the drug reservoir with different protection layers (thickness=20 m) at pH=6.5 and 25C. Error bar = SD, n = 3.

166

0.6 Fractional Release (Mt/Mo)

0.4

Folded device PCL Patch Solution

0.2

0 0 30 60 Time(min) 90 120

Figure 6.11 AO8 transport from different systems across the mucosal epithelium at pH=6.5 and 25C. Error bar = SD, n = 3.

167

0.4

Fractional Release 0.3

Folded device Solution

0.2

0.1

0 0 30 60 Time(min) 90 120

Figure 6.12 BSA transport from different systems across the mucosal epithelium at pH=6.5 and 25C. Error bar = SD, n = 3.

168

that only about 12% AO8 in the solution was delivered through the mucosal epithelium in 120 minutes, while 20% AO8 loaded in the patch system could transfer across the intestinal membrane. The self-folded device showed the highest drug transport fraction (33%) due to its localized high drug concentration. To compare the release behavior of drugs with different sizes, BSA was also used as a model drug. In the experiment, the BSA loading concentration was about 3 times higher than that of AO8 in order to provide easy detection by UV spectroscopy. Figure 6.12 shows the BSA transport profile from a folded device and the homogeneous solution at room temperature. As can be seen, the self-folded device exhibited an improved BSA transport fraction. Compared with Figure 6.11, the transport of large molecules across the mucosal epithelium was much more difficult than small molecules.

6.4

Conclusions

A self-folding miniature hydrogel device has been developed based on the integration of a number of micro-manufacturing modules. They demonstrated multi-functionalities such as enhanced mucoadhesion, lower drug leakage, and improved unidirectional delivery. The enhanced mucoadhesion due to self-folding increased the residence time at the target site, and led to improved drug transport. The PHEMA layer served as a diffusion barrier to provide good drug protection and prevented the drug leakage.

169

CHAPTER 7

CONCLUSIONS AND RECOMMENDATION

7.1

Conclusions

This work determined the roles of the solvent composition and light intensity in the photopolymerization of the MAA/TEGDMA resin system. It was found that the rate of polymerization increased and more compact gels would form with a higher water fraction in the 50wt% solvent/reactant mixture. This is because the weaker interactions between MAA and solvent molecules give a higher opportunity for propagation and a higher reaction rate. The hydrophobic TEGDMA and initiator tend to form aggregates in the higher water solution, contributing to the inhomogeneous microgel formation. It was also conlcuded that the rate of polymerization was enhanced as the light intensity increased, especially at the low light intensity range and low conversion. At too high a light intensity, a reduced MAA conversion was obtained. Additionally, the high light intensity significantly shortened the reaction time to reach the macro-gelation and increased the swelling ratio of formed hydrogels, which can be explained by the mechanism of intra- vs. intermolecular reaction. With a high UV intensity, more free
170

radicals and more possibility for intramolecular reaction lead to a higher reaction rate and faster gel formation. Since the intramolecular reaction contributes to less crosslinked microgels, the resulting hydrogels have a higher swelling ratio. By using these desired functional hydrogels cured under the optimal polymerization conditions, an assembled and a self-folding DDS were developed based on the selected integration of a number of micro-manufacturing modules to achieve multi-functionalities such as drug protection, self-regulated oscillatory release, enhanced mucoadhesion and targeted unidirectional release. The self-folding device first attached to the mucosal surface and then curled into the mucus, leading to enhanced mucoadhesion in the mode of grabbing. Furthermore, the folded layer served as a diffusion barrier, minimizing the drug leakage in the small intestine. The resulting unidirectional release provides improved drug transport through the mucosal epithelium due to the localized high drug concentration. The functionalities of the devices have been successfully demonstrated in vitro using a porcine small intestine. The novel delivery devices will be of great benefit to the advancement of oral administration of proteins and DNAs. Since the mucus layer covers many tissues at other specific sites, the devices may be applied for ocular, buccal, vaginal and rectal administrations. The polymer self-folding phenomena at the microscale can also be applied as probe arrays for bio/chemical sensing, carriers in cell-based bioreactors, and tissue clamping.

171

7.2

Recommendation

The developed self-folding device has significantly enhanced the mucoadhesion and extended the residence time for drug transport. To further improve the mucoadhesion, it would be desirable if a device can penetrate the loose-adherent layer and adhere to the firmly adherent mucus layer such that longer retention than a few hours may be achieved. This objective can be realized by reducing the device scale and adding the nanotips. The device scale should be reduced to 5 m or less such that they can move into the microvilli for a longer residence time. Traditional fabrication protocols, such as phase separation, microemulsion and spray drying, have been successfully used for the production of micro-/nano-particles for drug delivery [Jain, 2000; Langer, 2000]. However, the resulting particles are usually polydisperse and relatively simple structurally due to the surface-driven manufacturing process of these methods. To obtain an ideal delivery vehicle, a series of methods for making micron-sized polymeric layered structures has been developed in our laboratory using a soft lithography micro-transfer molding technique [Guan et al., 2005]. PDMS molds with an array of micron-sized wells can be made by the standard soft lithography technique. Figure 7.1 presents the self-foldable microdevices for drug delivery. These soft lithographic techniques can produce microparticles with similar structures but are simpler and of lower cost. Compared to the conventional microspheres for drug delivery, the microfabricated capsules are more uniform in size and shape, have a higher drug loading capacity, and may be absent of the burst effect that is typically associated with microspheres prepared by conventional methods. This basic fabrication operation has been successfully demonstrated in our laboratory and by other researchers. The remaining challenges are to
172

extend the technique to smaller particle sizes (i.e. 1-10 m and nanoscale) with different shapes, to extend the imprinting area, and to be adopted to the high precision manufacturing platform for mass production.

Figure 7.1 Schematic of fabrication of self-foldable microdevices. Optical micrographs of (a-c) bilayered microdevices with different curvatures controlled by the composition of the primary swelling layer; (d) a self-folded microdevice in water, and (e) several microdevices folded into the mucus of porcine intestine [Guan et al., 2005].

173

Controlling particle size and size distribution is most important for drug delivery applications. However, this alone is insufficient for the increase of delivery efficiency. Although the folding structure of the microvillis spatially restricts the mobility of small delivery devices, it has a high possibility for the delivery devices moving out these fine structures due to the peristalsis of the intestinal wall. The miniature device with flat and layered shape maximizes the contact area with the intestinal wall. The thin side areas minimize the exposure to the flow of liquids in the intestine. To extend a long duration time, the self-foldable finger-like arms and enhanced nanotips are considered in the device design. Figure 7.2 shows the schematic of a proposed device from the side view and the top view. By using a novel low-cost sacrificial template imprinting (STI) process developed by our group [Wang et al., 2004], the nanotips can be introduced on the drug layer. The enhanced nanotips not only help the device adhere to the firmly adherent mucus layer, but also may mechanically open the local tight junctions for improved permeability.

Figure 7.2 Schematic of the self-foldable microdevice with enhanced nanotips.


174

Furthermore, the choice of other functional copolymers can be used to better control the response performance and delivery behavior of DDS for various applications. The designed devices are mainly used for the delivery of proteins and DNAs through the GI tract. However, the physiological characteristics of each segment in small intestine changes a lot. For example, the pH value in the duodenum is around 5.5, while this value increases to 7.0 in the ileum. Other factors such as food compositions also influence the pH values. To deliver the device at a more specific site, the transition range of hydrogels between the swollen and the collapsed state with a pH change needs to be very narrow. Although it is possible to localize a device within each part of small intestine, the attainment of site-specific delivery in the rectum (pH=7.0) is even easier than in the small intestine [Kim et al., 2002]. The monomer composition is adjusted to match the requirement for pH-sensitivity of functional hydrogels at different specific sites. It is known that the transition range becomes sharper for more hydrophobic hydrogels and shifts to a higher pH for gels with the longer alkyl group. The transition range of PMAA is around the pH of 6.1. With the addition of a single methylene unit, poly(ethylacrylic acid) exhibited a sharper transition at the pH of 6.3. The addition of another methylene unit with poly(propyl acrylic acid) (PPAA) shifted the pH profile even further and PPAA displayed a much sharper transition close to the physiologic pH [Stayton et al., 2005]. Since the mucus layer covers many tissues at various specific sites, the device may be applied for ocular, buccal, vaginal and rectal administrations. Variation of the transition range for acrylic polymers with similar molecular weight provides a series of potential candidates for these applications. Delivery systems developed in this study are likely to enhance the oral
175

bioavailability of proteins and DNAs. The major market could be for improving the delivery of existing therapeutic agents with established markets such as protein drugs insulin, human growth hormone, and interfereon-alpha, and nucleic acid drugs such as antisense oligonucleotides (e.g., anti-bcl-2 oligo Genesence). Recent advances in biomedical research have yielded many novel therapeutic candidates that are based on proteins or nucleic acid, which have tremendous clinical potential but minimal oral bioavailability. The development of this technology can lead to significant benefits to improve patient compliance and cost savings, in addition to the reduction in pain and inconvenience associated with parenteral administration.

176

REFERENCES

Akiyama, Y.; Lueben, H.L.; de Boer, A.G.;Verhoef, J.C.; Junginger, H.E., Novel peroral dosage forms with protease inhibitory activities. II. Design of fast dissolving poly(acrylate) and controlled drugreleasing capsule formulations with trypsin inhibiting properties, Int. J. Pharm. 1996, 138,13-23. Albin, G.; Horbet, T.A.; Ratner, B.D., Glucose-sensitive membranes for controlled delivery of insulin: insulin transport studies, J. Control. Release 1985, 3, 153-164. Anseth, K.S.; Wang, C.M.; Bowman, C.N., Reaction behavior and kinetic constants for photopolymerizations 3243-3250. Anseth, K.S.; Bowman, C.N.; Brannon-Peppas, L., Mechanical properties of hydrogels and their experimental determination, Biomaterials 1996, 17, 1647-1657. Ascentiis, A.D.; deGrazia, J.L.; Bowman, C.N.; Colombo, P.; Peppas, N.A., Mucoadhesion of poly(2-hydroxyethyl methacrylate) is improved when linear poly(ethylene oxide) chains are added to the polymer network, J. Control. Release1995, 33, 197-201. Bae, Y.H.; Okano, T.; Kim, S.W., Onoff thermo-control of solute transport. Part 1. temperature dependence of swelling of N-isopropylacrylamide networks modified with hydrophobic components in water, Pharm. Res. 1991, 8, 531-537. of multi(meth)acrylate monomers, Polym.
1994,

35(15),

177

Bain, A.D.; Eaton, D.R., Line broadening in the carbon-13 NMR spectra of crosslinked polymers, Macromolecules 1989, 22, 3561-3564. Batzilla, T.; Funke, W.E., Formation of intra- and intermolecular crosslinks in the radical crosslinking of poly(4-vinylstyrene), Makromol. Chem. Rapid Commun. 1987, 8(6), 261-268. Bellhouse, B.J.; Kendall, M.A.F., Dermal powderject device, in: M.J. Rathbone, J. Hadgraft, M.S. Roberts, (Eds.), Modified-Release Drug Delivery Technology, Marcel Dekker, New York, 2003, pp. 607-617. Beltran, S.; Baker, J.P.; Hooper, H.H.; Blanch, H.W.; Prausnitz, J.M., Swelling equilibria for weakly ionizable, temperature-sensitive hydrogels. Macromolecules 1991, 24, 549-551. Beyssac, E.; Bregni, C.; Aiache, J-M.; Gerula, S.; Smolko, E., Hydrogel implants for methotrexate obtained by ionizing radiation, Drug. Dev. Ind. Pharm. 1996, 22, 439-444. Bouckaert, S.; Lefebvre, R.; Remon, J.P., In Vitro/in vivo correlation of the bioadhesive properties of a buccal bioadhesive miconasole tablet, Pharm. Res. 1993, 10, 853-856. Brannon-Peppas, L.; Peppas, N.A., Dynamic and equilibrium swelling behavior of pH-sensitive hydrogels containing 2-hydroxyethyl methacrylate, Biomaterials 1990, 11, 635-644. Brannon-Peppas, L., Preparation and characterization of crosslinked hydrophilic networks, in: L. Brannon-Peppas, R.S. Harland (Eds.), Absorbent Polymer Technology, Elsevier, Amsterdam, 1990, 45-66. Brogmann, B.; Beckert, T.E., Enteric targeting through enteric coating, in: H. Schreier (Ed.), Drugs and the Pharmaceutical Sciences, Vol. 115, Marcel Dekker Inc., New York,
2001, 1-29.
178

Brownlee, I.A.; Havler, M.E.; Dettmar, P.W.; Allen, A.; Pearson, J.P., Colonic mucus: secretion and turnover in relation to dietary fibre intake, Proc. Nutr. Soc. 2003, 62, 245-249. Bures, P.; Huang, Y.; Oral, E.; Peppas, N.A., Surface modifications and molecular imprinting of polymers in medical and pharmaceutical applications, J. Control. Release
2001, 72, 25-33.

Cao, X.; Lai, S.; Lee, L.J., Design of a self-regulated drug delivery device, Biom. Microdevice 2001, 3(2), 109-117. Carlfors, J.; Edsman, K.; Petersson, R.; Jornving, K., Rheological evaluation of Gelrite in situ gels for ophthalmic use, Eur. J. Pharm. Sci. 1998, 6, 113-119. Cha, W.; Hyon, S.; Graiver, D.; Ikasa, Y., Sticky poly(vinyl alcohol) hydrogels, J. Appl. Polym. Sci. 1993, 47, 339-343. Chen, G.H.; Hoffman, A.S., Graft copolymers that exhibit temperature induced phase transitions over a wide range of pH, Nature 1995, 373, 49-52. Chen, J.; Park, H.; Park, K., Synthesis of superporous hydrogels: hydrogels with fast swelling and superabsorbent properties, J. Biomed. Mater. Res. 1999, 44, 53-62. Chen, L.L.H.; Chien, Y.W., Transdermal iontophoretic permeation of luteinizing hormone releasing hormone: characterization of electric parameters, J. Control. Release 1996, 40, 187-198. Chetoni, P.; Colo, G.D.; Grandi, M.; Morelli, M.; Saettone, M.F.; Darougar, S., Silicone rubber/hydrogel composite ophthalmic inserts: preparation and preliminary in vitro/in vivo evaluation, Eur. J. Pharm. Biopharm. 1998, 46,125-132. Chiu, H.C.; Hsiue, G.H.; Lee, Y.P.; Huang, L.W., Synthesis and characterization of
179

pH-sensitive dextran hydrogels as a potential colon-specific drug delivery system, J. Biomater. Sci. Poly. Ed. 1999, 10, 591-608. Chiu, Y.Y.; Lee, L.J., Microgel formation in the free radical crosslinking polymerzation of ethylene glycol dimethacrylate (EGDMA). I. Experimental, J. Polym. Sci. Part A: Polym. Chem. 1995, 33, 257-267. Choudhari, K.B.; Labhasetwar, V.; Dorle, A.K., Liposomes as a carrier for oral administration of insulin: effect of formulation factor, J. Microencapsul. 1994, 11, 319-325. Chung, H.; Kim, J.; Um, J.Y.; Kwon, J.C.; Jeong, S.Y., Self-assembled nanocubicle as a carrier for peroral delivery, Diabetologia 2002, 45, 448-451. Cohen, M. H.; Melnik, K.; Boiarski, A.; Ferrari, M.; Martin, F.J., Microfabrication of silicon-based nanoporous particulates for medical applications, Biomed. Microdevices
2003, 5(3), 253 - 259

Cohen, S.; Lobel, E.; Trevgoda, A.; Peled, T., A novel in situ-forming ophthalmic drug delivery system from alginates undergoing gelation in the eye, J. Control. Release 1997, 44, 201-208. Crump, S., UV curing gel coat technology for composites application, Composites 2001 Convention and Trade Show, October 3-6, 2001. Davis, T.P.; Huglin, M.B., Studies on copolymeric hydrogels of N-vinyl-2-pyrrolidone with 2-hydroxyethyl methacrylate, Macromolecules 1989, 22(6), 2824-2829. Decker, C., The use of irradiation in UV polymerization, Polym. Inter., 1998, 45, 133-141. Dorkoosh, F.A.; Verhoef, J.C.; Borchard, G.; Rafiee-Tehrani, M.; Junginger, H.E.,
180

Development and characterization of a novel peroral peptide drug delivery system, J. Control. Release 2001, 71, 307318. Dorkoosh, F.A.; Verhoef, J.C.; Borchard, G.; Rafiee-Tehrani, M.; Verheijden, H.H.M.; Junginger, H.E.; Intestinal absorption of human insulin in pigs using delivery systems based on superporous hydrogel polymers, Inter. J. Pharm. 2002, 247, 47-55. Draye, J.P.; Delaey, B.; Bulcke, A.; Bogdanov, B.; Voorde, A.; Schacht, E., Dextran dialdehyde crosslinked gelatin hydrogel: a biocompatible drug delivery system, Proc. Int. Symp. Control. Release Bioact. Mater. 1997, 24, 467-468. Dusek, K.; Spevacek, J., Cyclization in vinyl-divinyl copolymerization, Polym. 1980, 21(7), 750-756. Ebara, M.; Aoyagi, T.; Sakai, K.; Okano, T., The incorporation of carboxylate groups into temperature-responsive poly(N-isopropylacrylamide)-based hydrogels promotes rapid gel shrinking, J. Polym. Sci. Part A: Polym. Chem. 2001, 39(3), 335-342. Elliott, J.E.; Anseth, J.W.; Bowman, C.N., Kinetic modeling of the effect of solvent concentration on primary cyclization during polymerization of multifunctional monomers, Chem. Eng. Sci. 2001, 56, 3173-3184. Elliott, J.E.; Bowman, C.N., Effects of solvent quality during polymerization on network structure of cross-linked methacrylate copolymers, J. Phys. Chem. B 2002, 106(11), 2843-2847. Fang, J.Y.; Hsu, L.R.; Huang, Y.B.; Tsai, Y.H., Evaluation of transdermal iontophoresis of enoxacin from polymer formulations: in vitro skin permeation and in vivo microdialysis using Wistar rat as an animal model, Int. J. Pharm. 1999, 180, 137-149. Feil, H.; Bae, Y.H.; Feijen J.; Kim, S.W., Effect of comonomer hydrophilicity and ionization on the lower critical solution temperature of N-isopropylacrylamide
181

copolymers, Macromolecules 1993, 26, 2496-2500. Fergg, F.; Keil, F.J.; Quader, H., Investigations of the microscopic structure of poly(vinyl alcohol) hydrogels by confocal laser scanning microscopy, Colloid. Polym. Sci. 2001, 279, 61-67. Flory, P.J.; Rabjohn, N.; Shaffer, M.C., Dependence of elastic properties of vulcanized rubber on the degree of crosslinking, J. Polym. Sci. 1949, 4, 225-245. Flory, P.J.; Rehner, J., Statistical mechanics of cross-linked polymer networks. II. swelling, J. Chem. Phys. 1943, 11, 521-526. Foraker, A.B.; Walczak, R. J.; Cohen, M.H.; Boiarski, T.A.; Grove, C.F.; Swaan, P.W., Microfabricated porous silicon particles enhance paracellular delivery of insulin across intestinal Caco-2 cell monolayers, Pharm. Res. 2003, 20, 110-116. Franssen, O.; Vandervennet, L.; Roders, P.; Hennink, W.E., Degradable dextran hydrogels: controlled release of a model protein from cylinders and microspheres, J. Control. Release 1999, 60, 211-221. Ganapathy, S.; Badiger, M.V., Dynamic response to hydration in a superabsorbing polymer by 4255-4263. Gayet, J.C.; Fortier, G., High water content BSA-PEG hydrogel for controlled release device: evaluation of the drug release properties, J. Control. Release 1996, 38, 177-184. Gemeinhart, R.A.; Park, H.; Park, K., Pore structure of superporous hydrogels, Polym. Adv. Technol. 2000, 11, 617-625. Ghandehari, H.; Kopeckova, P.; Kopecek, J., In vitro degradation of pH-sensitive hydrogels containing aromatic azo bonds, Biomaterials 1997, 18, 861-872.
182
13

C NOE and spin-lattice relaxation times, Macromolecules 1992, 25,

Gregoiraidis, G., Engineered liposomes for drug delivery: progress and problems, Trends Biotechnol. 1995, 13, 527-537. Guan, J.; He, H.; Hansford, D.J.; Lee, L.J., Self-folding of three dimensional hydrogel microstructures, J. Phys. Chem. 2005, 109(49), 23134-23137. Hariharan, D.; Peppas, N.A., Characterization, dynamic swelling behavior and solute transport in cationic networks with application to the development of swelling-controlled release systems, Polymer 1996, 37, 149-158. Hassan, C.M.; Peppas, N.A., Novel ethylene glycol-containing pH-sensitive hydrogels for drug delivery application: molecular gates for insulin delivery, American Chemical Society, 1999, 54-69. He, H.; Cao, X.; Lee, L.J., Design of a novel hydrogel-based intelligent system for controlled drug release, J. Control. Release 2004, 95, 391402. Heller, J., Drug delivery systems, in: Ratner, B.D.(Ed.), Biomaterials Science: an Introduction to Materials in Medicine, Academic Press, San Diego, 1996, pp. 346-356. Henrici-Olive, G.; Olive, S., Solvent effects in radical polymerization. I. Initiation rate of the polymerization of styrene with azodiisobutyronitrile as the initiator, Makromol. Chem.
1962, 58, 188-194.

Henrici-Olive, G.; Olive, S., Solvent effects in radical polymerization. III. Electron donor-acceptor complexes between polymer radicals and solvent molecules and their effect on the kinetics, Z. Phys. Chem. (Muenchen, Germany) 1965, 47, 286-298. Hirotsu, S.; Hirokawa, Y.; Tanaka, T., Volume-phase transitions of ionized N-isopropyl acrylamide, J. Chem. Phy. 1987, 87, 1392-1395. Hoffman, A.S., Applications of thermally reversible polymers and hydrogels in
183

therapeutics and diagnostics, J. Control. Release 1987, 6, 297-305. Hong, P.D.; Chen, J.H., Network structure and chain mobility of freeze-dried polyvinyl chloride/dioxane gels, Polym. 1998, 39, 5809-5817. Horak, D.; Rittich, B.; Safar, J.; Spanova, A.; Lenfels, J.; Benes, M.J., Properties of RNase A immobilized on magnetic poly(2-hedroxyl methacrylate) microspheres, Biotechnol. Prog. 2001, 17, 447452. Hsiu, G.; Wang, C., Glucose oxidase immobilized polyethylene-g-acrylic acid membrane for glucose oxidase sensor, Biotech. Bioeng. 1990, 36, 811-815. Hsu, C.P.; Lee, L.J., Free-radical crosslinking copolymerization of styrene/unsaturated polyester resins: 3. Kinetics-gelation mechanism, Polym. 1993, 34(21), 4516-4523. Hui, H.W.; Robinson, J.R., Ocular delivery of progesterone using a bioadhesive polymer, Int. J. Pharm. 1985, 26, 203-213. Ichikawa, H.; Peppas, N.A., Novel complexation hydrogels for oral peptide delivery: in vitro evaluation of their cytocompatibility and insulin-transport enhancing effects using Caco-2 cell monolayers, J. Biomed. Mater. Res. 2003, 67, 609-617. Ilium, L.; Farraj, N.F.; Davis, S.S., Chitosan as a novel nasal delivery system for peptide drugs, Pharm. Res. 1994, 11, 11861189. Inoue, T.; Chen, G.; Nakamae, K.; Hoffman, A.S., Temperature sensitivity of a hydrogel network containing different LCST oligomers grafted to the hydrogel backbone, Polym. Gels Network 1997, 5, 561-575. Jain, R.A., The manufacturing techniques of various drug loaded biodegradable poly(lactide-co-glycolide) (PLGA) devices, Biomaterials 2000, 21(23), 24752490.
184

Jakubiak, J., Crosslinking photocopolymerization of acrylic acid (and N-vinylpyrrolidone) with triethylene glycol dimethacrylate initiated by camphorquinone

/ethyl-4-dimethylaminobenzoate, J. Polym. Sci. Part A: Polym. Chem. 2000, 38(5), 876-886. Jung, D.Y.; Magda, J.J., Catalase effects on glucose-sensitive hydrogels, Macromolecules
2000, 33, 3332-3336.

Kabra, B.G.; Gehrke, S.H.; Hwang, S.T.; Ritschel, W.A., Modification of the dynamic swelling behavior of poly(2-hydroxyethyl methacrylate) in water, J. Appli. Polym. Sci.
1991, 42(9), 2409-2416.

Kaetsu, I.; Uchida, K.; Shindo, H.; Gomi, S.; Sutani, K., Intelligent type controlled release systems by radiation techniques, Radiat. Phys. Chem. 1999, 55, 193-201. Kaneko, Y.; Nakamura, S.; Sakai, K.; Kikuchi, A.; Aoyagi, T.; Sakurai, Y.; Okano, T., Deswelling mechanism for comb-type grafted poly(N-isopropylacrylamide) hydrogels with rapid temperature responses, Polym. Gels Network 1998, 6, 333-345. Kaneko, Y.; Saki, K.; Kikuchi, A.; Sakurai, Y.; Okano, T., Fast swelling/deswelling kinetics of comb-type grafted poly(N-isopropyl acrylamide) hydrogels, Macromol. Symp.
1996, 109, 41-53.

Katono, H.; Maruyama, A.; Sanui, K.; Okano, T.; Sakurai, Y., Thermo-responsive swelling and drug release switching of interpenetrating polymer networks composed of poly(acrylamideco-butyl methacrylate) and poly(acrylic acid), J. Control. Release 1991, 16, 215227. Khanvilkar, K.; Donovan, M.D.; Flanagan, D.R., Drug transfer through mucus, Adv. Drug Deliv. Rev. 2001, 48, 173-193. Kotze, A.F.; Thanou, M.M.; Lueben, H.L.; de Boer, A.G.; Verhoef, J.C.; Junginger, H.E.,
185

Enhancement of paracellular drug transport with highly quaternized N-timethyl chitosan chloride in neutral environments: In vitro evaluation in intestinal epithelial cells (Caco-2), J. Pharm. Sci. 1999, 88, 253-257. Kopecek, J.; Kopeckova, P.; Akala, E.O.; Yeh, P.; Ulbrich, K., pH-sensitive hydrogel with adjustable swelling kinetics for colon-specific delivery of peptides and proteins, PCT Int. Appl. WO 9801421, 1998. Kim, B.; Peppas, N.A., Poly(ethylene glycol)-containing hydrogels for oral protein delivery applications, Biomed. Microdevices 2003, 5, 333-341. Kim, S-H; Chu, C-C, Synthesis and characterization of dextranmethacrylate hydrogel and its structural study by SEM, J. Biomed. Mater. Res. 2000, 49, 517-527. Kisel, M.A.; Kulik, L.N.; Tsybovsky, I.S.; Vlasov, A.P.; Vorob, M.S.; Kholodova, E.A., Liposomes with phosphatidylethanol as a carrier for oral delivery of insulin: studies in the rat, Int. J. Pharm. 2001, 216, 105-114. Kitano, M.; Mitani, Y.; Takayama, K.; Nagai, T., Buccal absorption of golden hamster cheek in vitro and in vivo of 17-estradiol from hydrogels containing three types of absorption enhancers, Int. J. Pharm. 1998, 174, 19-28. Klenina, O.V.; Fain, E.G., Phase separations in a poly(acrylic acid)-polyacrylamide-water system, Polym. Sci. 1981, 23, 1439-1445. Klumb, L.A.; Horbett, T.A., Design of insulin delivery devices based on glucose-sensitive membranes, J. Control. Release 1992, 18, 59-80. Klumb, L.A.; Horbett, T.A., The effect of hydronium ion on the transient behavior of glucose-sensitive membranes, J. Control. Release 1993, 27, 95-114. Kost, J.; Horbett, T.A.; Ratner, B.D.; Singh, M., Glucose-sensitive membranes containing
186

glucose oxidase: activity, swelling, and permeability studies, J. Biomed. Mater. Res. 1985, 19, 1117-1133. Kozhukharova, A.; Popova, Y.; Kirova, N.; Klissurski, D.; Simeonov, D.; Spasov, L., Properties of glucose oxidase immobilized in gel of poly(vinyl alcohol), Biotech. Bioeng.
1988, 32, 245-248.

Kushibiki, T.; Tomoshige, R.; Tabata, Y., Controlled release of plasmid DNA by cationized gelatin hydrogel, Ensho. Saisei
2004, 24, 634-641.

Kwok, A.Y.; Qiao, G.G.; Solomon D.H., Synthetic hydrogels. 1. Effects of solvent on poly(acryamide) networks, Polym. 2003, 44, 6195-6203. Langer, R., Drug delivery and targeting, Nature, 1998, 5, 392(6679 Suppl). Langer, R., Biomaterials in drug delivery and tissue engineering: one laboratorys experience, Acc. Chem. Res. 2000, 33(2), 94101. Li, L.; Lee, L.J., Photopolymerization of HEMA/DEGDMA hydrogels in solution, Polymer 2005, 46(25), 11540-11546. Li, X.; Zhang, Y.; Yan, R.; Jia, W.; Yuan, M.; Deng, X.; Huang, Z., Influence of process parameters on the protein stability encapsulated in poly-DL-lactide-poly(ethylene glycol) microspheres, J. Control. Release 2000, 68, 4152. Lindmark, T.; Kimura, Y.; Artursson, P., Absorption enhancement through intracellular regulation of tight junction permeability by medium chain fatty acids in Caco-2 cells, J. Pharmacol. Exp. Ther. 1998, 284, 362-369. Liu, Q.; Hedberg, E.L.; Liu, Z.; Bahuleker, R.; Meszlenyi, R.K.; Mikos, A.G., Preparation of macroporous poly(2-hydroxyethyl methacrylate) hydrogels by enhanced phase separation, Biomaterials 2000, 21, 2163-2169.
187

Live, D.; Kent, S., Fundamental aspects of chemical applications of crosslinked polymer, ACS Series 1982, 193, 501-515. Lovell, L.G.; Newman, S.M.; Bowman, C.N., The effects of light intensity, temperature, and comonomer composition on the polymerization behavior of dimethacrylate dental resins, J. Dent. Res. 1999, 78(8), 1469-1476. Lowman, A.M.; Peppas, N.A., Analysis of the complexation/de-complexation phenomena in graft copolymer networks, Macromolecules 1997, 30, 4959-4965. Lowman, A.M.; Morishita, M.; Kajita, M.; Nagai, T.; Peppas, N.A., Oral delivery of insulin using pH-responsive complexation gels, J. Pharm. Sci. 1999, 88, 933-937. Lu, S.; Anseth, K.S., Photopolymerization of multilaminated poly(HEMA) hydrogels for controlled release, J. Control. Release 1999, 57, 291-300. Luellen, H.L.; De Leeuw, B.J.; Langemeyer, M.W.; DeBoer, A.G.; Verhoef, J.C.; Junginger, H.E., Mucoadhesive polymers in peroral peptide drug delivery. VI. carbomer and chitosan improve the intestinal absorption of the peptide drug buserelin in vivo, Pharm. Res. 1996, 13, 1668-1672. Luessen, H.L.; Rentel, C.O.; Kotze, A.F.; Lehr, C.-M.; de Boer, A.G.; Verhoef, J.C.; Junginger, H.E., Mucoadhesive polymers in peroral peptide drug delivery. IV. Polycarbophil and chitosan are potent enhancers of peptide transport across intestinal mucosae in vitro, J. Control. Release 1997,45(1), 15-23. Lustig, S.R.; Caruthers, J.M.; Peppas, N.A., Dynamic mechanical properties of polymer/fluid systems: characterization of PHEMA and P(HEMA-co-MMA) hydrogels, Polym. 1991, 32, 3340-3354 Machida, Y., Nagai, T., Bioadhesive preparations as topical dosage forms. Bioadhesive Drug Delivery System; Mathiowitz, E., Chickering, D.E., Eds.; Marcel Dekker Inc.: New
188

York, 1999, 641-657. Makino, K.; Mack, E.J.; Okano, T.; Kim, S.W., A microcapsule self-regulating delivery system for insulin, J. Control. Release 1990, 12, 235-239. Mariah, N.M.; Andrew, T.M.; Christopher, N.B.; Peppas, N.A., Predicting controlled release behavior of degradable PLA-b-PEG-b-PLA hydrogels, Macromolecules 2001, 34, 46304635. Mark, J.E., The use of model polymer networks to elucidate molecular aspects of rubberlike elasticity, Adv. Polym. Sci. 1982, 44, 1-26. Mason M.N.; Metters A.T.; Bowman C.N.; Anseth K.S., Predicting controlled-release bahavior of degradable PLA-b-PEG-b-PLA hydrogels, Macromolecules 2001, 34, 4630-4635. McNeill, M.E.; Graham, N.B., Vaginal pessaries from crystalline/rubbery hydrogels for the delivery of prostaglandin E2, J. Control. Release 1984, 1, 99-117. Meaney, C.; ODriscoll, C., Mucus as a barrier to the permeability of hydrophilic and lipophilic compounds in the absence and presence of sodium taurocholate micellar systems using cell culture models, Eur. J. Pharm. Sci.1999, 8,167-175. Mehier-Humbert, S.; Guy, R.H., Physical methods for gene transfer: improving the kinetics of gene delivery into cells, Adv. Drug Deliv. Rev. 2005, 57, 733-753. Mikijelj, B.; Varela, J.A.; Whittemore, O.J., Equivalence of surface areas determined by nitrogen adsorption and by mercury porosimetry, Ceram. Bull. 1991, 70, 829-831. Miyazaki, S.; Suisha, F.; Kawasaki, N.; Shirakawa, M.; Yamatoya, K.; Attwood, D., Thermally reversible xyloglucan gels as vehicles for rectal drug delivery, J. Control. Release 1998, 56, 75-83.
189

Morishita, M.; Lowman, A.M.; Takayama, K.; Nagai, T.; Peppas, N.A., Elucidation of the mechanism of incorporation of insulin in controlled release systems based on complexation polymers, J. Control. Release 2002, 81, 25-32. Mourad, P.D.; Murthy, N.; Porter, T.M.; Poliachik, S.L.; Crum, L.A.; Hoffman, A.S.; Stayton, P.S., Focused ultrasound and poly(2-ethylacrylic acid) act synergistically to disrupt lipid bilayers in vitro, Macromolecules, 2001, 34, 2400-2401. Murthy, N.; Robichaud, J.R.; Tirrell, D.A.; Stayton, P.S.; Hoffman, A.S., The design and synthesis of polymers for eukaryotic membrane disruption, J. Control. Release 1999, 61, 137-143. Musabayane, C.T.; Munjeri, O.; Bwititi, P.; Osim, E.E., Orally administered, insulin-loaded amidated pectin hydrogel beads sustain plasma concentrations of insulin in streptozotocin-diabetic rats, J. Endocrinol. 2000, 164, 1-6. Nair, M.K.; Chien, Y.W., Development of anticandidal delivery systems. 2. mucoadhesive devices for prolonged drug delivery in the oral cavity, Drug Dev. Ind. Pharm. 1996, 22, 243253. Nakamoto, C.; Motonaga, T.; Shibayama, M., Preparation pressure dependence of structure inhomogeneities and dynamic fluctuations in poly(N-isopropylacrylamide) gels, Macromolecules 2001, 34, 911-917. Nakamura, K.; Maltani, Y.; Lowman, A.M.; Takayama, K.; Peppas, N.A.; Nagai, T., Uptake and release of budesonide from mucoadhesive, pH-sensitive copolymers and their application to nasal delivery, J. Control. Release 1999, 61, 329-335. Narasimhan, B.; Peppas, N.A., Molecular analysis of drug delivery systems controlled by dissolution of the polymer carrier, J. Pharm. Sci. 1997, 86, 297-304. Neuhaus, D.; Williamson, M.P., The nuclear overhauser effect in structural and
190

conformational analysis, New York, VCH, 1989. Neuberger, W., Device and method for photoactivated drug therapy, U.S. Patent 6, 397, 102, July 6, 2002. Oechslein, C.R.; Fricker, G.; Kissel, T., Nasal delivery of octreotide: absorption enhancement by particulate carrier systems, Int. J. Pharm. 1996, 139, 25-32. Okano, T.; Bae, Y.H.; Jacobs, H.; Kim, S.W., Thermally onoff switching polymers for drug permeation and release, J. Control. Release 1990, 11, 255-265. Osada, Y.; Rossmurphy, S.B., Intelligent gels, Sci. American 1993, 268(5), 82-87. Pack, D.W.; Hoffman, A.S.; Pun, S.; Stayton, P.S., Design and development of polymers for gene delivery, Nat. Rev. Drug Discovery 2005, 4(7), 581-593 Park, K.; Shalaby, W.S.W.; Park, H., Biodegradable hydrogels for drug delivery, Lancaster-Basel: Technomic Publishing, Lancaster, PA, 1993. Park, J.W.; Hong, K.; Kirpotin, D.; Papahadjopoulos, D.; Benz, C.C., Immunoliposomes for cancer treatment, Adv. Pharmcol. 1997, 40, 399-435. Patel, V.R.; Amiji, M.M., Preparation and characterization of freeze-dried

chitosan-poly(ethylene oxide) hydrogels for site-specific antibiotic delivery in the stomach, Pharm. Res. 1996, 13, 588-593. Peppas, N.A.; Merrill, E.W., Crosslinked poly(vinyl alcohol) hydrogels as swollen elastic networks, J. Appl. Polym. Science 1977, 21,1763-1770. Peppas, N.A.; Mikos, A.G., Preparation methods and structure of hydrogels, in: N.A. Peppas (Ed.), Hydrogels in Medicine and Pharmacy, CRC Press, Boca Raton, FL, 1986, 1-27.
191

Peppas, N.A., Physiologically responsive gels, J. Bioact. Compat. Polym. 1991, 6, 241-246. Peppas, N.A.; Langer, R., New challenges in biomaterials, Science 1994, 263, 1715-1720. Peppas, N.A.; Sahlin, J.J., Hydrogels as mucoadhesive and bioadhesive materials: a review, Biomaterials 1996, 17, 15531561. Peppas, N.A., Hydrogels and drug delivery, Curr. Opin. Coll. Int. Sci. 1997, 2, 531-537. Peppas, N.A.; Bures, P.; Leobandung, W., Hydrogels in pharmaceutical formulations, European J. Pharm. and Biopharm. 2000, 50, 27-46. Petelin, M.; Sentjurc, M.; Stolic, Z.; Skaleric, U., EPR study of mucoadhesive ointments for delivery of liposomes into the oral mucosa, Int. J. Pharm. 1998, 173,193-202. Philippova, O.E.; Karibyants, N.S.; Stardubtzev, S.G., Conformational changes of hydrogels of poly(methacrylic acid) induced by interactions with poly(ethylene glycol), Macromolecules 1994, 27, 23982401. Podual, K., Glucose-sensitive cationic hydrogels for insulin release, a thesis submitted to the Faculty of Purdue University, December 1998. Ponchel, G.; Irache, J., Specific and non-specific bioadhesive particulate systems for oral delivery to the gastrointestinal tract, Adv. Drug Deliv. Rev.1998, 34, 191-219. Prausnitz, M.R.; Mitragotri, S.; Langer, R., Current status and future potential of transdermal drug delivery, Nat. Rev. Drug Discov. 2004, 3, 115-124. Qiu, Y.; Park, K., Environment-sensitive hydrogels for drug delivery, Adv. Drug Delivery Rev. 2001, 53, 321-339.

192

Ramadas, M.; Paul, W.; Dileep, K.J.; Anitha, Y.; Sharma, C.P., Lipoinsulin encapsulated alginate-chitosan capsules: intestinal delivery in diabetic rats, J. Microencapsul. 2000, 17, 405-411. Remunan-Lopez, C.; Portero, A.; Vila-Jato, J.L.; Alonso, M.J., Design and evaluation of chitosan/ethylcellulose mucoadhesive bilayered devices for buccal drug delivery, J. Control. Release 1998, 55, 143-152. Robinson, D.N.; Peppas, N.A., Preparation and characterization of pH-responsive poly(methacrylic acid-g-ethylene glycol) nanospheres, Macromolecules 2002, 35, 3668-3674. Robinson, J.R.; Bologna, W.J., Vaginal and reproductive-system treatments using bioadhesive polymer, J. Control. Release 1994, 28, 87-94. Rubinstein, A.; Tirosh, B., Mucus gel thickness and turnover in the gastrointestinal tract of the rat: response to cholinergic stimulus and implication for mucoadhesion, Pharm. Res.
1994, 11, 794-799.

Ryu, J.M.; Chung, S.J.; Lee, M.H.; Kim, C.K.; Shim, C.K., Increased bioavailability of propranolol in rats by retaining thermally gelling liquid suppositories in the rectum, J. Control. Release 1999, 59, 163-172. Sah, H., Protein instability toward organic solvent/water emulsification: implications for protein microencapsulation into microspheres, PDA J. Pharm. Sci. Technol. 1999, 53, 3-10. Schaefer, J., High resolution pulsed carbon-13 nuclear magnetic resonance analysis of some crosslinked polymers, Macromolecules 1971,4, 110-112. Scherzer, T.; Decker, U., Kinetic investigations on the UV-induced photopolymerization of a diacrylate by time-resolved FTIR spectroscopy: the influence of photoinitiator
193

concentration, light intensity and temperature, Radia. Phys. Chem. 1999, 55, 615-619. Schwarte, L.M.; Peppas, N.A., Novel poly(ethylene glycol)-grafted, cationic hydrogels: preparation, characterization and diffusive properties, Polymer 1998, 39, 60576066. Scranton, A.B., Structure of hydrophilic polymer networks formed by copolymerization /crosslinking reactions, Ph.D. thesis, Purdue University, 1990. Seigel, R.A., Implantable, self-regulating mechanochemical insulin pump, The Regents of the University of California, Berkeley, CA, 1991. Serres, A.; Baudys, M.; Kim, S.W., Temperature and pH-sensitive polymers for human calcitonin delivery, Pharm. Res. 1996, 13, 196-201. Shen, Z.; Mitragotri, S., Intestinal patches for oral drug delivery, Pharm. Res. 2002, 19(4), 391-395. Silliman, J.E., Network hydrogel polymers-application in hemodialysis, Sc.D. Thesis, Massachusetts Institute of Technology, Cambridge, MA, 1972. Scranton, A.B.; Bowman, C.N.; Peiffer, R.W., Photopolymerization fundamentals and application, ACS Symposium Series 673, Washington DC, 1996. Stayton, P.S.; Hoffman A.S.; EL-Sayed M.; Kulkarni S.; Shimoboji T.; Murthy N.; Bulmus V.; Lackey C., Intelligent biohybrid materials for therapeutic and imaging agent delivery, Proc. IEEE, 2005, 93, 726-736 Stuma, C.; Strugala, V.; Allen, A.; Holm, L., The adherent gastrointestinal mucus gel layer: thickness and physical state in vivo, Am. J. Physiol. Gastrointest. Liver Physiol.,
2001, 280, G922-G929.

Sun, Y.M.; Huang, J.J.; Lin, F.C.; Lai, J.Y., Composite poly(2-hydroxyethyl methacrylate)
194

membranes as rate-controlling barriers for transdermal applications, Biomaterials 1997, 18, 527-533. Sun, X.D.; Chiu, Y.Y.; Lee, L.J., Microgel formation in the free radical crosslinking copolymerzation of methyl methacrylate (MMA) and ethylene glycol dimethacrylate (EGDMA), Ind. Eng. Chem. Res. 1997, 36, 1343-1351. Suzuki, Y.; Tomonaga, K.; Kumazaki, M.; Nishio, I., Change in phase transition behavior of an NIPA gel induced by solvent composition: hydrophobic effect, Polym. Gels Netw.
1996, 4, 129142.

Tanaka, T., Collapse of gels and the critical endpoint, Phys. Rev. Lett. 1978, 40, 820-823. Tao, S.L.; Lubeley, M.W.; Desai, T.A., Bioadhesive poly(methyl methacrylate) microdevices for controlled drug delivery, J. Control. Release 2003, 88, 215228. Tao, S.L.; Lubeley, M.W.; Desai, T.A., Bioadhesive poly(methyl methacrylate) microdevices for controlled drug delivery, J. Control. Release 2004, 98, 215-228. Thomas, J.L.; Tirrell, D.A., Polyelectrolyte-sensitized phospholipid vesicles, Acc. Chem. Res. 1992, 25, 336342, Tian, Q.; Zhao, X.; Tang, X.; Zhang, Y., Hydrophobic association and temperature and pH sensitivity of hydrophobically modified poly(N-isopropylacrylamide/acrylic acid) gels, J. Appli. Polym. Sci. 2003, 87(14), 2406-2413. Torres-Lugo, M.; Peppas, N.A., Molecular design and in vitro studies of novel pH-sensitive hydrogels for the oral delivery of calcitonin, Macromolecules 1999, 32, 6646-6651. Torres-Lugo, M.; Peppas, N.A., Preparation and characterization of P(MAA-g-EG) nanospheres for protein delivery applications, J. Nanoparticle Res. 2002, 4, 73-81.
195

Traitel, T.; Cohen, Y.; Kost, J., Characterization of glucose-sensitive insulin release systems in simulated in vivo conditions, Biomaterials 2000, 21, 1679-1687. Treloar, R.G., The physics of rubber elasticity, 2nd Edition, Oxford University Press, Oxford, 1958. Tryson, G.R.; Shultz, A.R., A calorimetric study of acrylate photopolymerization, J. Polym. Sci., Polym. Phys. Ed., 1979, 17(12), 2059-2075. Turmanova, S.; Trifonov, A.; Kalaijiev, O.; Kostov, G., Radiation grafting of acrylic acid onto polytetrafluoroethylene films for glucose oxidase immobilization and its application in membrane biosensor, J. Membr. Sci. 1993, 127, 1-7. Vakkalanka, S.K.; Brazel, C.S.; Peppas, N.A., Temperature and pH sensitive terpolymers for modulated delivery of streptokinase, J. Biomater. Sci. Polym. Ed. 1996, 8, 119-129. Walther, D.H.; Sin, G.H.; Blanch, H.W., Pore-size distributions of cationic 2-hydroxyethyl methacrylate (HEMA) hydrogels, Poly. Gels and Net. 1995, 3, 29-45. Wang, S.; Zeng, C.; Lai, S.; Juang, Y.-J.; Yang, Y.; Lee, L.J., Polymeric nanonozzle array fabricated by sacrificial template imprinting, Adv. Mater. 2005, 17, 1182-1187. Wang, T.; Turhan, M.; Gunasekaran, S., Selected properties of pH-sensitive, biodegradable chitosanpoly(vinyl alcohol) hydrogel, Polym. Int. 2004, 53, 911-918. Wang, Y.; Yuan, K.; Li, Q.; Gu, S., Synthesis and swelling properties of biodegradable pH sensitive PVA-gelatin hydrogels, Gongneng Gaofenzi Xuebao, 2004, 17(4), 586-590. Ward, J.H.; Peppas, N.A., Preparation of controlled release systems by free-radical UV polymerizations in the presence of a drug, J. Control. Release 2001, 71, 183-192. Whitehead, K.; Shen, Z.; Mitragotri, S., Oral delivery of mcromolecules using intestinal
196

patches: applications for insulin delivery, J. Control. Release 2003, 88, 37-45. Wichterle, O.; Lim, D., Hydrophilic gels for biological use, Nature 1960, 185, 117-118. Wu, X.S.; Hoffman, A.S., Synthesis and characterization of thermally reversible macroporous poly(N-isopropylacrylamide) hydrogels, J. Polym. Sci., Part A: Polym.Chem. 1992, 30, 2121-2129. Xia, Y.; Whitesides, G.M., Soft lithograph, Annu. Rev. Mater. Sci. 1998, 28, 15384. Yang, Y.C.; Geil, P.H., Morphology and properties of PVC/solvent gels, J. Macromol. Sci. Phys. 1983, 22, 463-488. Yeh, P.Y.; Smith, P.L.; Ellens, H., Effect of medium-chain glycerides on physiological properties of rabbit intestinal epithelium in vitro, Pharm. Res. 1994, 11, 1148-1154. Yoshida, R.; Sakai, K.; Ukano, T.; Sakurai, Y.; Bae, Y.H.; Kim, S.W., Surface-modulated skin layers of thermal responsive hydrogels as onoff switches: I. drug release, J. Biomater. Sci. Polym. Ed. 1991, 3, 155162. Yoshida, R.; Uchida, K.; Kaneko, Y.; Sakai, K.; Kikcuhi, A.; Sakurai, Y.; Okano, T., Comb-type grafted hydrogels with rapid de-swelling response to temperature changes, Nature 1995, 374, 240-242. Yu, H.; Grainger, D.W., Thermo-sensitive swelling behavior in crosslinked

N-isopropylacrylamide networks: cationic, aanionic, and ampholytic hydrogels, J. Appl. Polym. Sci. 1993, 49, 1553-1563. Zhang, I.; Shung, K.K.; Edwards, D.A., Hydrogels with enhanced mass transfer for transdermal drug delivery, J. Pharm. Sci. 1996, 85, 1312-1316. Zhang, J.; Peppas, N.A., Synthesis and characterization of pH- and temperature- sensitive
197

poly(methacrylic acid)/poly(N-isopropylacrylamide) interpenetrating polymeric networks, Macromolecules 2000, 33, 102-107. Zhang, J.T.; Cheng, S.X.; Zhuo, R.X., Preparation of macroporous

poly(N-isopropylacrylamide) hydrogel with improved temperature sensitivity, J. Polym. Sci., Part A: Polym. Chem. 2003, 41, 2390-2392. Zhang, J.T.; Cheng, S.X.; Huang, S.W.; Zhuo, R.X., Temperature-sensitive

poly(N-isopropylacrylamide) hydrogels with macroporous structure and fast response rate, Macromolecular Rapid Comm. 2003, 24, 447-451. Zhang, X.Z.; Yang, Y.Y.; Wang, F.J.; Chung, T.S., Thermosensitive

poly(N-isopropylacrylamide-co-acrylic acid) hydrogels with expanded network structures and improved oscillating swelling-deswelling properties, Langmuir 2002, 18(6), 2013-2018. Zhang, X.Z.; Zhuo, R.X., Preparation of fast responsive, temperature-sensitive poly(N-isopropylacrylamide) hydrogel, Macromolecular Chem. Phy. 1999, 200, 2602-2605. Zhang, X.Z.; Zhuo, R.X., Preparation of fast responsive, thermally sensitive. poly(N-isopropylacrylamide) gel, Eur. Polym. J. 2000, 36, 2301-2303. Zhou, M.P.; Donovan, M.D., Intranasal mucociliary clearance of putative bioadhesive polymer gels, Int. J. Pharm. 1996, 135, 115125. Zhou, X.; Weng, L.; Chen, Q.; Zhang, J.; Shen, D.; Li, Z.; Shao, M.; Xu, J., PH sensitivity of poly(acrylic acid-co-acrylamide) hydrogel, Poly. Inter. 2003, 52, 1153-1157.

198