|
SURFACE MODIFICATION OF VASCULAR PROSTHESIS AND
INTRACORNEAL LENS POLYMERS
By
CHRISTOPHER WILLIAM WIDENHOUSE
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1996
Copyright 1996
by
Christopher William Widenhouse
This Dissertation is dedicated to all the people who believed
I could do it more than I did myself; especially, Tammy,
whose love and support I could not do without;
and to the memory of Charles Bucaria.
ACKNOWLEDGMENTS
I would like to express my deepest gratitude to my
advisor and doctoral committee chairman, Dr. Eugene P.
Goldberg, for his guidance, encouragement, and patience.
Sincere thanks are also extended for the advice and teaching
of the members of my supervisory committee: Dr. Christopher
D. Batich, Dr. Anthony B. Brennan, Dr. James S. Seeger, and
Dr. Richard Dickinson. I am also grateful to the research
group of Dr. James S. Seeger. The guidance and input of Dr.
Dinesh 0. Shah, an original committee member, is also
appreciated.
Special appreciation is also felt for the assistance and
encouragement from my colleagues during my graduate tenure.
These include Jesse Arnold, David Bennett, Dr. Sameer
Bhatia, Dr. Emmanuel Biagtan, Charles Bucaria, Scott Butler,
Don DePalma, Shannon Eggers, Kirk Foster, Penelope Kao, Dr.
Steven Kuo, Ingrid Leidermooy, Dr. Tung Liang Lin, Lili
Mateo, Dr. Khalid Mentak, Julie Miller, Tom Miller, Dr. Lynn
Peck, Mark Privett, Dr. Jeanne Quigg, Katsabu Rao, Dr.
William Toreki, John Wironen, Dr. Ali Yahiaoui, Stacey Zambo,
and Mike Zamora.
Special thanks are extended to James F. Kirk for his
wise and thoughtful input, Drew P. Amery for assistance in
learning the ropes of analytical techniques, Paul J. Martin
for his lively debates and invaluable microscopy assistance,
Dr. Anthony B. Brennan for always pointing out the not so
obvious and to his openness and fairness to all questions,
James S. Marotta for his enthusiasm and encouragement, and to
my wife, Tammy for her love, support, pessimistic optimism,
and encouragement.
TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ........................................... iv
LIST OF TABLES ............................................. ix
LIST OF FIGURES ............................................xi
ABSTRACT ................................................. xvi
CHAPTERS
1 INTRODUCTION ............................................. 1
1.1 Vascular Grafts ...................................1
1.2 Intracorneal Lenses ...............................8
2 BACKGROUND ...............................................14
2.1 Synthetic Vascular Grafts ........................14
2.1.1 Vessel Replacement Surgery and Vascular
Grafts.....................................14
2.1.2 Properties of the Natural Vessel..........18
2.1.3 Synthetic Vascular Graft Materials and
Properties ................................23
2.1.4 Advantages of PMMA, SEMA, and PDMS
Surfaces .................................31
2.1.5 Gamma Radiation Initiated
Polymerization ............................38
2.1.6 Polymer Solution Coatings and
Techniques ...............................45
2.2 Intracorneal Implants ............................48
2.2.1 Refractive Corneal Surgery and
Intracorneal Implants ....................48
2.2.2 Intracorneal Lens Materials and
Designs ..................................49
2.2.3 Surface Properties of Ocular
Biomaterial Implants ......................52
2.2.4 Surface Modification Techniques ...........53
3 MATERIALS AND METHODS .................................55
3.1 Materials .......................................55
3.1.1 Substrates................................55
3.1.2 Monomers and Reagents .....................57
3.2 Methods ..........................................57
3.2.1 Sample Preparation and Substrate
Cleaning..................................57
3.2.2 Surface Modification Methods .............60
3.2.3 Solution Dip Coating of PDMS onto
Dacron* ...................................63
3.2.4 Characterization ..........................65
3.2.5 Diffusion/Flow Cell Testing of ICLs ....... 73
3.2.6 Porosity and Coating Stability Testing
of Vascular Grafts ........................75
3.2.7 Ex Vivo and in Vivo Studies of Vascular
Substrates................................78
4 RESULTS AND DISCUSSION...................................80
4.1 Gamma Radiation Induced Polymerization of
Methyl Methacrylate on PET, PTFE, and PDMS ......80
4.1.1 Swelling of PET, PTFE, and PDMS in MMA
Solutions ................................81
4.1.2 Radiation Grafting of MMA on PET, PTFE,
and PDMS .................................95
4.2 Gamma Radiation Induced Polymerization of
Sulfoethyl Methacrylate on PET .................. 146
4.2.1 Swelling and Gravimetric Analysis of
PET Surface Modified with SEMA ...........146
4.2.2 XPS Analysis of Radiation Grafted SEMA
onto PET .................................147
4.3 Solution Dip Coating and Thermal Curing of
PDMS Coatings on Dacron* PET .................... 148
4.3.1 Conditions for Dip Coating Dacron with
PDMS ..................................... 148
4.3.2 Thermal Curing of PDMS Coatings on
Dacron ..................................149
4.3.3 Gamma Polymerization of MAOP-t-PDMS..... 150
4.3.4 Analysis of PDMS Coated Dacron*..........151
4.4 Results and Discussion of Fenestrated ICLs
with Hydrograft' Surface Modifications .......... 167
4.4.1 Determination of Presoaking and
Grafting Conditions ...................... 167
4.4.2 Analysis of Modified ICLs ................ 174
5 SUMMARY AND CONCLUSIONS ................................179
5.1 Surface Modification of Vascular Graft
Substrates ..................................... 180
5.1.1 Gamma Radiation Induced Polymerization
of MMA on Vascular Graft Substrates ......180
5.1.2 Solution Dip Coating and Thermal
Polymerization of PDMS on Dacron*, with
and without Pre-modification with MAOP-
t-PDMS ...................................184
5.2 Surface Modification of Fenestrated PMMA
Intracorneal Lens Substrates ..................186
5.2.1 Gamma Radiation Induced Polymerization
of NVP on PMMA ICL Substrates ............ 186
6 FUTURE WORK ................................... ......... 189
6.1 Surface Modification of Vascular Graft
Substrates.....................................189
6.1.1 PMMA Modified PET ........................ 189
6.1.2 PMMA Modified PTFE .......................190
6.1.3 PMMA Modified PDMS .......................191
6.1.4 SEMA Modified PET........................ 192
6.1.5 Solution Dip Coatings on PET.............192
6.2 Surface Modification of Intracorneal Lenses .....193
LIST OF REFERENCES ......... ........................ .....195
BIOGRAPHICAL SKETCH ......................... .................. 207
viii
LIST OF TABLES
Table page
2.1 Initial and two week post implant dynamic
compliance values for canine vascular graft
materials...........................................30
3.1 Molecular weight data for PMMA ocular implant
materials...........................................56
4.1 Solubility parameters and H-bonding groups for
selected solvents and polymers .......................81
4.2 Percent weight increases of Mylar* D-1000 films in
selected solvents at room temperature ................82
4.3 Percent weight increase of DMSO-MMA solutions by
PET (Mylar* D-1000) films and woven Dacron"
fabrics at 60C. Swelling time is 24 hours for
all samples................. ........................90
4.5 Percent weight changes of Teflon* PTFE irradiated
in MMA-acetone and MMA-DMSO solutions as a
function of radiation dose and solution
concentration......................................103
4.6 Percent weight changes of GORE-TEX' ePTFE
irradiated in MMA-acetone and MMA-DMSO solutions
to 0.11 Mrad as a function of monomer
concentration ...................................... 103
4.7 Percent weight changes of GORE-TEX* ePTFE
irradiated in 100% MMA as a function of radiation
dose ................................................104
4.8 Contact angle of Teflon PTFE irradiated in MMA-
acetone and MMA-DMSO solutions as a function of
radiation dose and solution concentration ...........111
4.9 Average carbon, oxygen, and fluorine
concentrations for Teflon*, GORE-TEX*, and PMMA as
determined with XPS.................................115
4.10 Weight increase of PDMS irradiated in MMA-DMSO
solutions as a function of monomer concentration
and radiation dose..................................127
4.11 Contact angle data for PDMS irradiated in MMA-DMSO
solutions as a function of solution concentration
(left hand 2 columns) and total radiation dose
(right hand 2 columns) ..............................132
4.12 Percent weight increases of Dacron" dip coated
with PDMS or gamma radiation irradiated in MAOP-t-
PDMS as a function of modification conditions....... 152
4.13 Carbon, oxygen, and silicon atomic concentrations
for Dacron* modified with dip coatings of PDMS and
gamma irradiated in MAOP-t-PDMS as determined with
XPS.................................................154
4.14 Silicon concentrations in solutions from the
vascular graft delamination analysis as measured
by ICP.............................................. 158
4.15 Pressurized porosity analysis of unmodified and
PDMS dip coated Dacron*. Flow rate is reported as
ml/min-cm2 normalized to the sample surface area. ...159
4.16 Platelet counts from ex vivo AV shunt experiments
for unmodified, PDMS dip coated, and MAOP-t-PDMS
modified Dacron* fabric. Platelet counts are
reported as counts/mm2 sample surface area ..........169
4.17 Penetration depths of 10 and 20% NVP into
fenestrated PMMA disks as determined by optical
microscopy.........................................171
4.18 Flow rate of water through modified and unmodified
ICLs at room temperature and constant pressure
(atmospheric pressure plus 1 inch of water). Lens
type column gives fenestration hole size and
surface percentage of fenestrations................. 176
LIST OF FIGURES
Figure page
1.1 Schematic diagram illustrating the anatomy of the
human eye and relavent features .......................8
1.2 Schematic diagram illustrating a cross section of
the cornea with an ICL in place, and diffusion and
flow processes of fluids and nutrients ................9
2.1 Cumulative patency plotted against time for four
types of vascular graft materials ....................17
2.2 Cut away view of an artery and vein showing the
three distinct layers ................................19
2.3 Typical curve for vessel compliance measurements.
Compliance is given as % radial change per
millimeter Hg X 10-2 .................................21
2.4 Possible mechanisms of gamma radiation induced
radiolysis and free radical formation for PET,
PTFE, and PDMS....................................... 44
2.5 Reaction mechanism for thermal curing of Shin-Etsu
PDMS .................................... ............47
2.6 Possible gamma radiation reactions and products
for PMMA............................................54
3.1 Chemical structures of polymer substrates used for
surface modification .................................56
3.2 Chemical structures of monomers used for surface
modification........................................58
3.3 Schematic diagram of the irradiation chamber used
for polymerization ...................................62
3.4 Photograph of motorized carousel used to provide
uniform exposure within the irradiation chamber......62
3.5 Schematic representations of the captive bubble
technique and angles measured for contact angle
goniometry showing (a) sample analysis chamber,
and (b) bubble at interface ..........................67
3.6 Schematic representation of the IR beam, crystal,
and sample for FT-IR/ATR spectroscopy ................69
3.7 Schematic illustration of diffusion and flow cell
used for evaluating permeability of ICL
materials...........................................75
3.8 Schematic illustrations of the setups used for
leakage and stability analysis for vascular
grafts. (a) Leakage analysis setup and (b) flow
system for stability analysis ........................77
4.1 Percent weight increase with time of Mylar* D-1000
PET films in MMA-chloroform solutions as a
function of solution concentration...................83
4.2 Mt/Mm vs. time1/2/1 for MMA-chloroform solution as
a function of solution concentration .................84
4.3 Diffusivity of MMA-chloroform solutions in PET as
a function of solution concentration .................85
4.4 Maximum percent weight uptake by PET of MMA-
chloroform solutions as a function of solution
concentration ......... ...........................87
4.5 Solubility parameter for polyisobutene and
polystyrene as determined by intrinsic viscosity
measurements in a series of solvents................. 88
4.6 Variation of D (diffusion coefficient) and (De)p
(polymer-fixed diffusion coefficient of the
diluent) with volume fraction of benzene for the
natural rubber-benzene system.......................88
4.7 Percent weight increase with time of Mylar* and
Dacron* in MMA-DMSO solutions at 60"C as a
function of solution concentration................... 92
4.8 Cls binding peak of polystyrene showing the n-n*
shake-up peak....................................... 98
4.9 SEM micrograph of Mylar* D-1000 film showing the
dispersion of silica on the surface..................100
4.10 SEM micrograph of Meadox woven Dacron* fabric.......100
xii
4.11 Percent weight increase of GORE-TEX (ePTFE)
irradiated in MMA-acetone and MMA-DMSO solutions
to 0.11 Mrad as a function of monomer
concentration ...................................... 105
4.12 Percent weight increase of GORE-TEX* (ePTFE)
irradiated in 100% MMA monomer as a function of
radiation dose...................................... 107
4.13 Contact angle of Teflon" (PTFE) irradiated in MMA-
acetone and MMA-DMSO solutions to 0.11 Mrad as a
function of solution concentration................. 112
4.14 FT-IR/ATR absorbence spectra of PMMA, GORE-TEX*,
and GORE-TEX* surface modified with MMA.............113
4.15 XPS Cls spectra for Teflon*, GORE-TEX*, and PMMA
showing differences in the chemical shifts due to
various carbon bonds ................................115
4.16 Sample calculation to determine the surface
concentration of PMMA and PTFE on Teflon" and
GORE-TEX' following modification with MMA............116
4.17 Surface concentration of PMMA on GORE-TEX* and
Teflon* irradiated in MMA-acetone and MMA-DMSO
solutions as a function of monomer concentration.
Concentrations of PMMA determined with XPS..........118
4.18 Surface concentration of PMMA on GORE-TEX" and
Teflon" irradiated in 100% MMA to 0.11 Mrad as a
function of radiation dose ..........................119
4.19 XPS Cls spectra for Teflon* irradiated in MMA-
acetone solutions showing changes in the spectra
with increasing surface concentrations of PMMA as
a function of solution concentration ................120
4.20 SEM micrographs of unmodified GORE-TEX* .............122
4.21 SEM micrographs of GORE-TEX* irradiated in 3%
MMA-DMSO to 0.11 Mrad................. ................123
4.22 SEM micrographs of GORE-TEX* irradiated in 20%
MMA-DMSO to 0.11 Mrad. Deformation and stretching
of nodule structure is visible ......................124
4.23 SEM micrographs of GORE-TEX* irradiated in 3%
MMA-acetone to 0.11 Mrad. Breaking of the nodule
structure is typical for samples modified in
acetone solutions..................................125
4.24 Percent weight increase of PDMS irradiated in 10%
MMA-DMSO as a function of radiation dose............128
xiii
4.25 Percent weight increase of PDMS irradiated in MMA-
DMSO to 0.11 Mrad solutions as a function of
solution concentration ..............................131
4.26 Contact angle of PDMS following polymerization of
10% MMA-DMSO solutions as a function of radiation
dose ................................................133
4.27 Contact angle of PDMS irradiated in MMA-DMSO
solutions as a function of monomer concentration
polymerized to 0.10 0.13 Mrad.....................133
4.28 FT-IR transmission spectra for unmodified PDMS and
PMMA, and PDMS irradiated in 10% MMA-DMSO to 0.11
Mrad................................................135
4.29 Surface concentration of PMMA on PDMS as a
function of MMA-DMSO solution concentration
polymerized to 0.11 Mrad............................136
4.30 Surface concentration of PMMA on PDMS irradiated
in 10% MMA-DMSO to 0.11 Mrad as a function of
radiation dose......................................137
4.31 1 Hz frequency storage modulus (E') of PDMS
modified with MMA-DMSO irradiated to 0.11 Mrad as
a function of monomer concentration .................139
4.32 1 Hz frequency loss modulus (E") of PDMS modified
with MMA-DMSO irradiated to 0.11 Mrad as a
function of monomer concentration ................... 140
4.33 1 Hz frequency storage modulus (E') of PDMS
irradiated in 10% MMA-DMSO as a function of
radiation dose......................................141
4.34 1 Hz frequency loss modulus (E") for of PDMS
irradiated in 10% MMA-DMSO as a function of
radiation dose......................................142
4.35 1 Hz frequency tan 6 of PDMS irradiated in MMA-
DMSO to 0.11 Mrad as a function of monomer
concentration ..................... ............... 143
4.36 1 Hz frequency tan 8 of PDMS irradiated in 10%
MMA-DMSO as a function of radiation dose............ 144
4.37 Storage modulus (E') plotted against percent PMMA
by weight of PDMS irradiated in MMA-DMSO as a
function of monomer concentration and radiation
dose ................................................145
4.38 CIs spectra showing, (a) a comparison between 60*C
cured PDMS from chloroform solution and high
temperature cured PDMS, and (b) a comparison
between Dacron* and PDMS dip coated Dacron*......... 156
4.39 SEM micrographs of unmodified DacronO fabric
(Meadox) ............................................162
4.40 SEM micrographs of PDMS dip coated DacronO fabric
(Meadox) ............................................163
4.41 SEM micrographs of PDMS dip coated reinforced
Dacron" vascular prosthesis (Bard).................. 164
4.42 SEM micrographs of unmodified Dacron* fabric
(Meadox) after ex vivo AV shunt analysis ............165
4.43 SEM micrographs of PDMS dip coated Dacron* fabric
(Meadox) after ex vivo AV shunt analysis ............166
4.44 Optical micrographs of a fenestrated ICL after
being surface modified with NVP and stained with
silver nitrate ..................................... 170
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
SURFACE MODIFICATION OF VASCULAR PROSTHESIS AND
INTRACORNEAL LENS POLYMERS
By
Christopher William Widenhouse
May, 1996
Chairman: Dr. Eugene P. Goldberg
Major Department: Materials Science and Engineering
Synthetic vascular replacements are expanded
polytetrafluoroethylene (ePTFE) and polyethylene terephthalate
(Dacron* PET). Poor long term patency of small diameter
vascular prostheses is attributed to platelet adhesion and the
inability of the vascular endothelium to regenerate. Most
attempts to reduce thrombus also reduce endothelial cell
adhesion, and attempts to promote endothelial cell
proliferation simultaneously facilitate thrombus formation.
Surface graft polymerization of polymethyl methacrylate (PMMA)
and sulfoethyl methacrylate (SEMA) onto prostheses substrates
(PET, PTFE, ePTFE, and PDMS) using gamma radiation induced
polymerization (GRIP) in solutions of dimethylsulfoxide (DMSO)
and acetone produced stable surface grafts of PMMA and SEMA.
This was studied as a method to reduce thrombus formation and
encourage healing. Platelet adhesion to PMMA modified PET was
not significantly different than unmodified PET.
Relationships between polymerization reactions and both
radiation dose and monomer solution concentration were
examined. PDMS was solution dip coated onto PET and thermally
polymerized. Bonding of PDMS coatings onto PET was
accomplished by GRIP of methacryloxypropyl terminated PDMS
onto PET prior to dip coating. Both processes produced stable
PDMS coatings on PET, and inhibited platelet adhesion in ex
vivo canine arteriovenous (AV) shunt studies.
Surfaces were characterized by gravimetric analysis,
contact angle goniometry, Fourier transform infrared
spectroscopy (FT-IR/ATR), X-ray photoelectron spectroscopy
(XPS), inductively coupled plasma (ICP), pressurized flow
analysis, optical microscopy (OM), and scanning electron
microscopy (SEM). Blood compatibility was evaluated by ex
vivo AV canine shunt experiments.
The second part of this research involved intracorneal
lenses (ICLs), which are designed to correct myopia,
hyperopia, and astigmatism. Current designs prevent nutrient
migration through the implant, leading to stromal necrosis and
complications. Fenestrated PMMA ICLs were surface modified
with an N-vinyl pyrrolidone (NVP) monomer presoak, followed by
GRIP providing a lens surface of polyvinylpyrrolidone (PVP).
The hydrophilic PVP surface is designed to facilitate nutrient
flow through the lens, and to provide a tissue protective
layer which is less tissue damaging and cell adhesive.
Modified ICLs were analyzed by contact angle goniometry, XPS,
OM, and diffusion and flow analysis.
xvii
CHAPTER 1
INTRODUCTION
1.1 Vascular Prostheses
A vascular prosthesis is a synthetic or natural vessel
used to replace damaged or diseased sections of veins and
arteries. Vascular disease, complications from surgical
interventions, vascular trauma, and disorders which damage
the normal endothelial lining of vessels or impart
thrombogenic complications are often treated with synthetic
or natural vascular grafts (Brody et al., 1972, Epstein,
1988, and Gertler and Abbott, 1992). Currently, the primary
synthetic materials used for vascular grafts are woven or
braided Dacron* polyethylene terephthalate (PET) and expanded
polytetrafluoroethylene grafts, like GORE-TEX*. In the U.S.,
vessel replacements are used over 350,000 times per year
(Ratner, 1993). Improvements in the patency of vascular
grafts and other blood contact devices, such as arterial-
venous (AV) shunts (150,000 per year), heart valves (75,000
per year), and pacemakers (130,000 per year), through the use
of improved materials, would be a major health care advance
(Ratner, 1993).
The success of a vascular graft varies with the inner
diameter of the vessel replacement. Vascular grafts with an
inner diameter greater than 6 mm are considered large
diameter grafts, and those with a diameter less than 6 mm (3-
6 mm) are considered small diameter grafts. The majority of
complications arise from the use of small diameter grafts to
replace coronary and peripheral arteries (Underwood et al.,
1988 and Litwak et al., 1987). Although some intimal
hyperplasia (thickening of the inner vessel layer, especially
at the anastomosis) is observed in the large diameter
prostheses, the strong and sometimes turbulent blood flow
within these prostheses aids in the prevention of total
occlusion, and they remain patent for extended times, often
never necessitating repeat surgical correction (Underwood et
al., 1988 and Litwak et al., 1987). Small diameter grafts,
however, often have up to 50% or more occlusion within the
first 24 months of implantation, resulting in the need for
secondary surgical repairs, or ultimately causing patient
death (Underwood et al., 1988).
The ideal surface for blood contact is the normal
physiological endothelial lining of vessels, and attempts to
mimic this environment through surface modification with
polymers, proteins, cultured cells, and a host of other
approaches, both biologic and synthetic, have been, and
continue to be, explored (Greisler, 1991, Dale, 1978,
Gimbrone, 1987, Ratner et al., 1987, and Hoffman, 1984).
This research attempted to provide a synthetic polymer
surface which would have improved long-term patency and bio-
acceptance, especially for small diameter vascular graft
implants. Long term success depends on two primary factors,
which are often antagonistic. Modifications successful in
reducing platelet cell adhesion and thrombus formation also
discourage endothelial cell adhesion and growth. It is
necessary to inhibit thrombus formation to maintain
prosthesis patency, while endothelialization is necessary for
long term healing. The overall goal of this research is to
develop a vascular prosthesis that has reduced platelet
adhesion and thrombus formation as well as encourages long
term healing and re-endothelialization.
A careful review of vascular prosthesis literature and
experimental results within this University of Florida
research group suggested consideration of polymethyl
methacrylate (PMMA), anionic sulfonic acid containing
polymers such as polysulfoethyl methacrylate (SEMA), and
polydimethyl siloxane (PDMS) surfaces as less thrombogenic
than currently used vascular replacement materials. Surface
modification of currently used materials with PMMA, SEMA, or
PDMS may provide improved blood compatibility for vascular
prostheses. PMMA, SEMA, or PDMS surfaces, also may be
further modified with biological molecules that would serve
to mimic the performance (both chemistry and function) of the
endothelial surface. This second step, although beyond the
scope of the research presented here, is currently being
pursued within this laboratory.
The bulk properties of PMMA, SEMA, and PDMS are
unsuitable for vascular prostheses. PMMA is glassy at body
temperature, and is therefore too rigid--although the idea of
a graft woven from PMMA fibers is intriguing. SEMA is a
hydrogel with poor mechanical properties, and PDMS has a
modulus very close to that of the natural vessel, but has
poor tear strength. Silica fillers are often used to
reinforce PDMS, but the presence of filler decreases
hemocompatibility (Lim et al., 1994). All of these
materials, however, were found to have favorable surface
properties with respect to the vascular blood-contact
environment. This makes these materials excellent candidates
for surface modification onto currently available vascular
graft biomaterials. Surface modification by gamma radiation-
induced polymerization (GRIP) provides a chemically bound
surface layer with desirable physical and chemical properties
on a substrate biomaterial without altering the desirable
substrate bulk mechanical properties. Substrates were
modified in this research with PDMS by thermally curing a
two-part oligomer mixture. This enables surface modification
to be carried out after substrates have been coated with an
uncured PDMS mixture.
Surface modification through GRIP of polymers, monomers,
and other compounds has been studied and used since the early
1940s (Chapiro, 1962). Radiation-induced polymerization of
vinyl functional monomers involves the initiation of free
radicals on both the monomer and the substrate polymer. To
increase the probability of substrate radicals propagating or
terminating with graft and interacting monomer/polymer
radicals (providing true graft polymerization), an intimate
mixture of monomer and substrate is desired. The process is
diffusion controlled, and the use of organic swelling agents
that enhance monomer penetration into the substrate is a
method of achieving desirable monomer-substrate interaction,
and increases the probability of grafting. Another technique
is to increase radical initiation in the substrate by
increasing the radiation dose (Z 0.5 Mrad), and to include an
inhibitor to prevent excessive solution homopolymerization
(e.g., Mohr's salts) in the monomer solution. The addition
of potentially toxic substances, like inhibitors, is not
desirable for production of implantable biomaterials, and
high radiation doses may degrade the mechanical properties of
many polymeric substrates. Therefore, swelling agents were
used to enhance the penetration of the monomer into the
substrate surfaces as described by Yahiaoui's presoakk
method" (Yahiaoui, 1990). Although organic solvents were
used (i.e., chloroform, acetone, etc.), an effort to utilize
more bioacceptable solvents (i.e., dimethyl sulfoxide) was
made.
Simultaneous irradiation of a monomer in solution and a
polymer substrate initiates free radicals in both the
substrate and monomer creating a grafted, branched,
interpenetrating network (IPN) surface region with new
physical properties. Swelling, or presoaking, the polymer
with a monomer solvent mixture prior to irradiation is
helpful to provide a monomer-rich surface region. Swelling
increases the probability of grafting if (i)
homopolymerization of the monomer is favored, (ii) initiation
on the substrate is difficult, or (iii) interactions between
the substrate and monomer are unfavorable for an intimate
mixture. If homopolymerization is favored, the result is
polymerization outside and away from the substrate polymer
chains, resulting in little or no bonding of the new polymer
to the substrate. Presoaking in this situation provides
monomer within the surface of the substrate, which becomes an
IPN or graft polymer when polymerized. If initiation of the
substrate is difficult, an IPN may still be created with pre-
swelling of monomer into the substrate, but little or no
bonding of the IPN to the substrate through graft
polymerization occurs. Solvents also aid in overcoming
repulsive forces between incompatible monomers and
substrates, facilitating the intimate mixture required for
grafting. For example, a hydrophilic monomer in an organic
solvent may wet or swell a hydrophobic substrate better than
the monomer itself. Solvents enhance monomer diffusion to
the surface region and to the active substrate radicals
during polymerization.
It was the goal of the first part of this research to
investigate the surface modification of currently available
vascular prosthesis polymers, namely PET, PTFE, and PDMS, in
an attempt to obtain a more hemocompatible material. The two
major approaches used for surface modification were (i)
simultaneous gamma radiation induced polymerization of
monomer, solvent, and substrate polymer with and without pre-
swelling at relatively low dose (ca. s 0.15 Mrad), and (ii)
solution dip coating of an uncured oligomer mixture followed
by thermally curing the surface polymer.
PDMS is more readily available as a two-component curing
system, and may be polymerized after being formed into almost
any desired device shape. For this reason, solutions of the
oligomer mixture were used to coat substrates by dipping.
Following solvent evaporation, the PDMS was thermally cured
forming a coating on the polymer substrate. To improve the
bonding of the PDMS coating with the substrate, a vinyl-
acrylic functional silicone polymer (methacryloxypropyl
terminated PDMS) was radiation grafted onto the substrate
prior to dip coating. During the curing process, the
silicone components crosslink to each other and to the
surface grafted silicone functional molecules, providing
covalent bonding of the dip-coated surface to the substrate.
The resulting physical, chemical, and mechanical
properties of the modified substrates were characterized by a
variety of techniques including FT-IR/ATR, XPS, SEM, UV-VIS,
optical microscopy, gravimetric analysis, and contact angle
goniometry. Biocompatibility was evaluated using a canine ex
vivo shunt method (c.f., section 3.2.7.2) and in viva
implants (c.f., section 3.2.7.3), in collaboration with the
research group of Dr. James M. Seeger.
1.2 Intracorneal Lenses
In the late 1940s, Jose Barraquer (Barraquer, 1949)
began experiments on the implantation of synthetic materials
within the corneal stroma of the eye to correct numerous
irregular curvatures and damage caused to the cornea.
Investigations have advanced to clinical studies on a variety
of implant designs and materials, the most widely tested
being intracorneal or intrastromal rings and lenses (ICLs).
Figure 1.1 shows a schematic diagram of the human eye.
Figure 1.2 shows a schematic cross-section of the cornea
region illustrating the placement of an ICL, and outlines the
diffusion and flow of fluids and nutrients in the cornea.
Current designs being studied for refractive corrections to
the cornea include PMMA rings, PMMA lenses, fenestrated PMMA
lenses, and polysulfone (PSf) lenses.
Lens Vitreous
Cornea
Retina
Epithelium
Stroma
Endothelium
Anterior
Chamber ~ i
Optic
Nerve
Iris
Figure 1.1 Schematic diagram illustrating the anatomy of the
human eye and relavent features, modified from
Corel Draw computer art program.
I 14 ^ Meabolic Pump L tlo Glucose
W 1r% Water SU* ked If WHr removal
In by ImbIblllon Acid
by OP pressure
Figure 1.2 Schematic diagram illustrating a cross section of
the cornea with an ICL in place, and diffusion
and flow processes of fluids and nutrients.
(Taken from McCarey, 1990).
Hydrogel lens implants and PMMA ring implants do not
offer significant refractive correction, and therefore serve
to correct vision through manipulation of the corneal
curvature (Climenhaga et al., 1988). The major problem with
these implants occurs following implantation. The degree of
correction required is calculated prior to choosing a
specific implant, and following the surgery, the scaring and
healing of the cornea is unpredictable, often causing over or
under correction of the curvature (McDonald et al., 1993).
When solid lens implants are used, this problem is avoided by
providing refractive correction to the cornea instead of
relying on curvature corrections, and corneal healing has a
less dramatic effect on the overall success of the implant.
The refractive index of PMMA (1.49) and PSf (1.63) make
the polymers excellent candidates for lens materials, and
these lenses have had some success in recent research studies
(McCarey, 1990). The widespread use of PMMA as an
intraocular lens (IOL) material also makes it a material of
choice. The refractive index and curvature of the lens
design determine the refractive power, or diopter, of the
lens. PSf lenses have the advantage of being thinner and,
therefore, somewhat more flexible than PMMA lenses, and
because of the higher refractive index, a thin PSf lens may
have the same diopter as a thicker PMMA lens. PMMA lenses
have the advantage of being studied more thoroughly than PSf,
and the compatibility of PMMA in the ocular environment is
exceptional (Amon and Menapace, 1990). Both PMMA and PSf
lenses, however, block the flow of vital nutrients to stroma
anterior to the implant (Climenhaga et al., 1988).
Fenestrated lenses with holes of various dimensions, however,
may allow flow and diffusion of saline, oxygen, glucose,
proteins, and other metabolites to the stromal tissues
surrounding the implant.
As demonstrated previously in our laboratories by Osborn
(1985), Hoffmeister (1988), Goldberg et al. (1988 and 1989),
Yahiaoui (1990), Mentak (1993), and Lin (1995), surface
modification of PMMA with polyvinylpyrrolidone (PVP) provides
a hydrophilic surface with many distinct advantages. The
advantages of PVP modified PMMA (Hydrograft*) for IOL
applications include reduced corneal endothelial cell damage,
reduced iris abrasion, and a reduced adhesion and
proliferation of lens epithelial cells within the ocular
environment (Yahiaoui, 1990, Goldberg et al., 1988 and 1989).
A hydrophilic surface on a fenestrated lens should also
improve surface wetting and diffusion of nutrients to the
tissues surrounding the implant, particularly to the region
anterior to the lens.
This research investigated the surface modification of
fenestrated PMMA intracorneal lenses, and the effects on the
permeablity of the ICL to vital stromal nutrients. The
previously mentioned benefits of Hydrograft* modifications
with respect to tissue and cell damage were not investigated
in detail in view of prior research. Complete descriptions
and analysis of the tissue and cell compatibility of the PMMA
Hydrograft* are provided in Yahiaoui (1990), Goldberg et al.
1988 and 1989, Mentak (1993), and Lin (1995).
The cornea receives nutrients from the aqueous humor and
the limbal blood supply (McCarey and Schmidt, 1990), and the
driving force for the non-turbulent diffusion and flow
through to the anterior corneal tissues of the eye is
primarily the chemical potential of the solutions present.
The flow of water, saline, and glucose solutions through
fenestrated PMMA ICLs was studied by the use of a diffusion
flow chamber. Unmodified PMMA and Hydrografts (GRIP modified
PMMA with PVP) modified PMMA ICLs were compared to each other
and to hydrogel lens polymers. Process conditions suitable
for GRIP modification of the fenestrated ICLs also required
investigation.
A diffusion chamber was used to evaluate the flow of
water, saline, and glucose through fenestrated PMMA ICLs
having different hole dimensions (10 to 80 pm) and surface
area coverage. The solution of interest was placed in one
side, and allowed to flow or diffuse through the
fenestrations of the implant. Differences between the
unmodified ICLs, HydrograftO ICLs, and hydrogel polymers were
observed and recorded.
Although studied in depth previously within our
laboratories, various GRIP modification parameters were
investigated to provide successful Hydrograft* modifications
of the fenestrated PMMA ICLs. The molecular weight of the
PMMA used for these lenses is significantly lower than
materials used previously (80,000 Mw vs. 2-4 million Mw) for
IOLs. Because of this, monomer presoak diffusion times and
temperatures needed to be investigated to determine
conditions which would afford Hydrograft" surfaces without
distorting or crazing the low molecular weight PMMA. The
diffusion of NVP solutions into the low molecular weight PMMA
as a function of time, temperature, and NVP solution
concentration was first investigated to optimize the
presoaking parameters. Once suitable conditions for
Hydrograft modification of the PMMA ICLs were determined,
wettability, graft penetration depth, and saline and glucose
permeability were studied. Finally, hydrogel lenses were
also tested to compare the permeability of fenestrated ICLs
to glucose and saline. The properties of the modified lenses
13
were determined using contact angle goniometry, optical
microscopy, and the diffusion apparatus previously mentioned.
CHAPTER 2
BACKGROUND
2.1 Synthetic Vascular Grafts
2.1.1 Vessel Replacement Suraerv and Vascular Grafts
Atherosclerosis is the progressive deposition of plaque
in the arteries with resulting clogging and blood flow
restriction. It can lead to heart disease and stroke, and is
responsible for over 50% of the deaths in the US (Fox, 1987)
Reduced blood flow through the arteries ultimately results in
ischemic heart disease. Reduced flow through the coronary
arteries is a critical condition requiring immediate
attention to reduce the permanent damage of heart muscle by
reduced oxygen supply. This occlusion may also occur in
other areas of the body as well, and threaten permanent
damage to vital regions through reduced blood flow. The
severely affected patient often requires treatment of
coronary and peripheral arteries. Occluded arteries are
generally treated surgically today in one of two fashions,
either with balloon angioplasty or with bypass surgery
utilizing a vascular prosthesis. Balloon angioplasty
involves insertion of a catheter with a balloon which is
inflated in the occluded region. Inflation of the balloon
re-opens the artery, often leaving the thrombus and plaque
free to circulate as emboli, which may clog capillaries and
can cause stroke.
2.1.1.1 Natural Cardiovascular Prostheses
Natural vascular prostheses are classified based on the
origin of the replacement vessel. Autografts are transplants
within the same individual, homografts or allografts are
transplants between different individuals of the same species
(typically a donor organ or vessel), and xenografts or
heterografts are transplants from different species (the most
common of which is the use of porcine heart valves in
humans).
Autografts and homografts are the most commonly used
natural prostheses, and include those taken from the
saphenous vein, the umbilical vein, and the mammary vein.
Veins are generally used instead of arteries because the body
has an ability to re-route blood flow through veins more
easily than through arteries without causing permanent damage
to vital regions of the body. Veins also have a slightly
different surface chemistry than arteries which render them
less thrombogenic than arteries. The valves inside the veins
typically are stripped away to eliminate "dead spots" in the
flow caused by the valves, which can lead to thrombus
formation.
2.1.1.2 Synthetic Vascular Grafts
The primary synthetic vascular replacements are expanded
polytetrafluoroethylene (ePTFE), like GORE-TEX*, and woven or
braided Dacron* polyethylene terephthalate (PET) fabrics
(Greisler, 1991). The largest supplier of PET fibers in the
world is DuPont, supplying Dacron*, which was initially
developed for the textile industry. The standards regulating
the production of Dacron* fibers is still largely controlled
by textile industry demands. Dacron& fibers contain titanium
dioxide, which gives the fibers a bright white appearance
demanded by the textile industry, and fibers used for
vascular prostheses, therefore, contain titania as well. The
fibers are woven into a number of different patterns by
different manufacturers. The Dacron* used in this research
was either a non-velour woven Dacron* fabric from Meadox or a
reinforced velour woven knit Dacron* vascular graft from
Bard. GORE-TEX* ePTFE is manufactured by a high temperature,
high speed extrusion process which forms the polymer into a
foam like structure. GORE-TEX* is readily available with
different pore sizes depending on the properties required for
the specific application.
The majority of small diameter ePTFE (GORE-TEX) grafts
and woven Dacron* grafts occlude the vessel by more than 50%
within a 3-year period (Whittemore et al., 1981). A study by
Pevec et al. (1992) shows that all small diameter grafts
occlude within 5 to 10 years. Figure 2.1 shows a re-
tabulation of data presented by Abbott (1987) comparing the
success of natural and synthetic small diameter graft
materials.
- Saphenous Vein
--- Umbilical Vein
--0--- PTFE
----- Dacron
Patency (%) 75-
70-
65-
60-
55-
50-
0 2 4 6 8 1012141618202224
Months
Figure 2.1 Cumulative patency as a function of time for four
small diameter vascular graft materials. Data
from Abbott, 1987, and re-plotted for
presentation here.
Surface properties are the primary factors controlling
the acceptance of biomaterials, especially the blood-material
interactions of vascular grafts (Andrade et al., 1991,
Hoffman, 1987, and Mustard et al., 1987). Baier (1969)
showed that the immediate response of the biological
environment to a foreign material involves protein adsorption
within the first few seconds. The initial proteins adsorbed
by the surface and resulting conformational changes affect
the sequence of events that follow. The surface activity of
adsorbed proteins determines which peptide sequences,
clotting factor proteins, and hence, which cells adhere and
attach upon reaching the surface (Ratner, 1993). Control of
these initial events is critical in the development of a
surface which does not activate the complex sequence of
thrombogenic events (Miyauchi and Shionoya, 1988 and De Mol
van Otterloo et al., 1992). The physical and chemical
characteristics of the surface determine how proteins are
initially adsorbed and which factors control the interactions
between other blood proteins and circulating biological
molecules.
Clotting and intimal thickening, or hyperplasia, is
often attributed not only to surface thrombogenicity of the
prosthesis, but also to the increased stresses resulting from
the mismatch in the radial modulus between the synthetic
replacement and the natural vessel at the anastomosis
(Brothers et al., 1990). (Most physicians label the radial
elastic properties of vessels as "compliance." This is not
to be confused with the engineering definition of compliance
which is the inverse modulus.) Changes in the modulus of the
prosthesis may alter the flow pattern of blood within the
vessel, and may sometimes cause turbulent flow which can
promote thrombosis (Hanson and Barker, 1987).
2.1.2 Properties of the Natural Vessel
2.1.2.1 Physical and mechanical structure of the vessel
The natural blood vessel is a composite structure
comprised of three layers. The outermost layer is the tunica
adventitia or externa, the middle layer is the tunica media,
and the inner layer is the tunica intima (Figure 2.2).
AI-rY
Figure 2.2.
Cut away view of an artery and vein showing the
three distinct layers. (Taken from Fox, 1987).
These three layers are composed of an interwoven network of
collagen, elastin, smooth muscle, and other proteins and
cells, with an innermost layer of endothelial cells
(endothelium). Collagen, elastin, and smooth muscle provide
the vessel with necessary mechanical properties, and the
inner endothelial cell lining provides the surface chemistry
and function of a non-thrombogenic surface. The highly
ordered and abundant fibers of collagen in the adventitia
bear the major part of vessel stresses. This provides the
high modulus and tensile strength of the vessel. Collagen
has a modulus of elasticity of 0.1 to 2.9 X 10 9 Pa (145,000
to 420,000 psi) (Abbott and Cambria, 1982). The high
elasticity of blood vessels are due primarily to the elastin,
which has a modulus of 3 to 6 X 10 5 Pa (45 to 90 psi), a
tensile strength of 0.36 to 4.4 X 10 6 Pa (50 to 650 psi) ,
and an elastic strain of over 300% (Abbott and Cambria, 1982
and Abbott, 1987).
The overall properties of the vessel vary with body
position and distance from the heart. The higher pulse wave
regions closer to the heart damp out the energy with a higher
elasticity and lower modulus. Lower pulse regions do not
require as much damping, and have a lower elastin
composition, less elasticity, and higher modulus. Human
arteries have a modulus of about 1 X 10 5 Pa (15 psi)
longitudinally and 1 X 10 6 Pa (150 psi), circumferentially,
with variations according to location (Nichols and O'Rourke,
1986). Veins typically have a higher modulus than arteries
because the mean pressure within a vein is 2 mm Hg and 100 mm
Hg within an artery (Fox, 1987). The vessel components are
assembled in a complex composite structure that yields
anisotropic mechanical properties.
Most evaluations of vascular mechanical properties
involve measurements of compliance rather than tensile
strength or modulus. Compliance is a measure of the dynamic
circumferential elastic properties as determined by the
relationship shown in equation 2.1,
C ( yD.tolic Ddiutli )
C = (ytl PD i(2.1)
( D.tarc ) x (P~ntolico Pdiatolic )
where Da and Dd are the vessel diameters, and Pa and Pd are
the pressures (Mergerman et al., 1986). This is not to be
confused with the engineering definition of compliance, which
is the inverse of the modulus. Vascular compliance is a
dynamic property, and changes with pressure. A typical
compliance-pressure curve is shown in figure 2.3 for a canine
femoral artery (Mergerman et al., 1986). Matching the
mechanical properties of natural vessels and artificial
grafts is important. A mismatch of properties was reported
to be thrombogenic by Baird et al., (1977), Abbott et al.,
(1987), Hasson et al., (1985), and Kelly et al., (1992). The
mismatch of properties arises from the initial use of stiff
materials and changes in modulus resulting from intimal
hyperplasia, both of which may lead to turbulent flow and
coagulation (Hanson and Harker, 1987). A graft material
replacing the natural vessel must have favorable mechanical
properties, and not induce thrombus and intimal hyperplasia
to remain compliant during use.
35
25o 1
0 11
e \
60 80 t0 12 1to 16e
PRESSURE
Figure 2.3 Typical curve for vessel compliance
measurements., where compliance is % radial
change per mm Hg X 10-2 (Mergerman et al., 1986).
2.1.2.2 Surface properties and functionality of the vessel
The surface characteristics (chemical, physical, and
functional) of the endothelial cell lining make it the ideal
blood contact surface. The dynamic and active roles of the
endothelium and sub-endothelium have only recently been
realized (Gimbrone, 1987). The favorable surface properties
of the endothelium are related to both the production of
proteins and the availability of specific binding sites for
anticoagulant agents and factors. The complex homeostasis
within the vascular environment is an ongoing balance between
formation and lysis of fibrin, the insoluble polymer formed
to repair or close off damaged areas within the arteries.
Exposure of the sub-endothelium, damage to the endothelium,
or simple contact of blood with a non-endothelial surface is
sufficient to initiate platelet activation and the
coagulation cascade (Gimbrone, 1987). Activation causes the
release of mitogens from platelets, endothelial cells, and
monocytes stimulating a host of responses including smooth
muscle cell proliferation giving rise to thrombus formation
and intimal hyperplasia in an attempt to repair or close off
the damaged region (Fox, 1987). Following the formation of
fibrin, the endothelium produces plasminogen activators which
cause the conversion of plasminogen to plasmin, causing
fibrinolysis. Production of plasminogen activator inhibitors
allows control over the extent of lysis (Mustard et al.,
1987). For example, specific surface binding sites are
provided for glycosaminoglycans which associate with
antithrombin III, which in turn inhibits thrombin, a protease
which converts fibrinogen into fibrin, as well as factor Xa,
and possibly factors IXa, XIa, and XIIa (Mustard et al.,
1987). Specific binding sites are also provided on the
endothelium for thrombomodulin, which readily binds thrombin
(Mustard et al., 1987). The endothelium of a normal, healthy
individual can typically repair damaged segments of 1 to 2 cm
in length (Greisler, 1991). Neo-endothelialization (re-
growth of the natural endothelium) of a vascular prosthesis
which allows the retention of normal specific activities and
functions of the endothelium would be advantageous to the
success of vascular grafts.
2.1.3 Synthetic Vascular Prosthesis Materials and Properties
The primary synthetic vascular graft materials are PET
and expanded PTFE. PDMS has also been studied. The surface
and bulk properties of these materials contribute to the
success or failure of these materials as vascular prostheses.
The improvement of the surface properties of these substrate
materials through surface modification to make a prosthesis
that is nonthrombogenic and which is capable of healing was
the focus of this research.
2.1.3.1 Properties and modification of PET
PET has been used as a vascular graft material
clinically for over 40 years, and this is the primary reason
it was studied here for surface modification (Brothers et
al., 1990 and Hufnagel, 1955). PET is used for vascular
grafts in the form of woven and knitted fibers. Because of
the higher porosity of some weaves, some grafts require pre-
clotting prior to use. Grafts with porosity greater than 300
to 400 ml/cm2 min at 120 mm Hg require pre-clotting to prevent
leakage (Greisler, 1991).
Typical mechanical properties of PET depend on the
degree of crystallinity and orientation. PET vascular grafts
are composed of woven fibers which are highly ordered and
contain approximately 35 to 60% crystallinity (Rodriguez,
1982). PET fibers have a glass transition temperature, Tg, of
80'C and a melt temperature, Tm, of 245 to 265C. Typical
mechanical properties of unoriented PET are 4.8 to 6.9 X 107
Pa (7,000 to 10,000 psi) tensile strength, 4.1 to 4.8 X 109
Pa (400,000 to 600,000 psi) tensile modulus, and a strain to
failure of 30 to 300% (Rodriguez, 1982 and Roslink, 1990),
although tensile strength values for oriented fibers can
exceed 100,000 psi. (Unlike the value reported earlier for
elastin, 300% elongation of PET is non-elastic.) Compared to
the natural vessel, PET has exceptional mechanical strength,
and in the form of a woven fabric material, has flexibility
and kink resistance. Radial and longitudinal modulus,
however, remain much greater than the natural vessel. The
need for a non-thrombogenic surface is therefore important to
reduce possible intimal thickening caused by mechanical
mismatch.
PET is a polyester, and polyesters are subject to
hydrolysis of the ester linkage. The rigid nature of the
backbone chain and high crystallinity make PET fibers more
resistant to hydrolysis than many other polyesters, and PET
is reported to be relatively stable in the biological
environment compared to other polyesters (Rodriguez, 1982).
Successful surface modification of PET by gamma
radiation polymerization has been reported (Stannett, 1981,
Rebenfeld and Weigmann, 1978, Kale and Lokhande, 1975, and
Nair et al., 1988). The use of PMMA and similar monomers
such as acrylic acid, di-methyl acrylamide, hydroxyethyl
methacrylate (HEMA), and other vinyl monomers have reportedly
been grafted onto PET. Polyester fiber stability during
gamma radiation has been reviewed by Nair et. al (1988).
Doses of 2.5 Mrad increased the crystallinity from 40 to 44%
and increased the breaking load of the fibers by 2.3%. The
changes, although minor, are attributed to cross-linking and
some degradation. The low doses used in this research (ca. s
0.15 Mrad), however, are not expected to significantly alter
the mechanical properties of PET.
PET has been reported to be an activator of the
complement system, and the addition of a barrier layer such
as a PMMA surface graft or PDMS coating may inhibit this
reaction (Shoenfeld et al., 1988, and Miyauchi and Shionoya,
1988). Figure 3.1 in section 3.2.1.1 shows the chemical
structure of PET as well as the other substrate polymers used
in this research.
2.1.3.2 Properties and modification of ePTFE
Hufnagel (1955) reports the use of expanded PTFE as a
vascular prosthesis as early as the late 1940s. PTFE is also
used in catheters, bone joint prostheses, and soft tissue
implants, and has been studied as a paste material to replace
endoscopic balloon inserts and other devices (Ratner, 1993,
Bonomini et al., 1969, and Atala et al., 1992). PTFE is a
hydrophobic, very low surface energy fluoropolymer (19
dynes/cm) which appears less thrombogenic than PET. PTFE is
generally used for vascular grafts in the expanded form as
ePTFE (GORE-TEX*), and implanted grafts of expanded PTFE and
Dacron" invoke a similar biological response (c.f., figure
1.1). Although porous, the high surface energy and
hydrophobic surface of ePTFE grafts prevent leakage during
implantation, and ePTFE grafts do not require pre-clotting.
The porous surface of the material is reported to be
advantageous, allowing greater flexibility with tissue and
cellular in-growth, and the adhesion and formation of a
neointimal layer on the graft surface (Sprugel et al., 1987).
However, the inability of certain materials to adhere to PTFE
may prevent the necessary interactions between circulating
factors and the substrate, hindering the re-
endothelialization and healing process. The major
manufacturers of ePTFE vascular grafts are Gore and Impra.
The process to form expanded PTFE first patented by W.
L. Gore & Associates to make GORE-TEX* entails extrusion of
the material at high temperature (ca. 390'C) and high strain
rate (ca. 40,000 %/sec), producing a microstructure of
interconnecting nodes of 0.5 to 400 pm and fibrils of 5 to
1,000 pm and a porosity of approximately 85 to 95% by volume
(Gore, 1970 and 1975). A high Tm of 310 to 330'C allows the
expansion process to take place without significant
degradation of the PTFE, providing the material with higher
flexibility. PTFE has a tensile strength of 1.4 to 3.5 X 107
Pa (2,000 to 5,000 psi), a modulus of 4.0 to 5.5 X 108 Pa
(58,000 to 80,000 psi), and an elongation of 200 to 400%
(Rodriguez, 1982 and Roslink, 1990).
Surface modification attempts on ePTFE have often
utilized radio frequency plasma polymerization or oxidation
of the surface as an initial step, but the use of gamma
radiation alone has been shown to be effective. Pre-swelling
the surface with a monomer-solvent mix reportedly yielded
successful gamma induced surface modification with N-
vinylpyrrolidone in pyridine (Sayed et al., 1981). Razzak et
al. (1987) reported successful radiation grafting of N,N-
dimethylacrylamide (DMAA) onto PTFE using ethyl acetate and
acetone as solvents. Pre-irradiation of the substrate to
induce a high concentration of free radicals followed by
addition of monomer has also been used. However, the higher
doses required for this pre-irradiation step often degrade
PTFE. This research utilized very low doses and simultaneous
irradiation of monomer and substrate in an attempt to surface
modify PTFE without significant degradation.
Expanded PTFE has mechanical properties similar to the
natural vessel with respect to modulus and elasticity. A
surface graft to improve the biocompatibility is not expected
to significantly alter this advantage of the material, but to
provide a more compatible surface through the altering of the
surface chemistry.
2.1.3.3 Properties and modification of PDMS
Of the substrate materials studied, PDMS is the only
elastomer, and therefore has the potential for a close match
to the modulus of the natural vessel. Silicones are used as
soft tissue implants, IOLs, skin replacements, burn and wound
dressings, catheters, mammary implant components, and for
other devices (Ratner, 1993, Taylor, 1985, Meaburn et al.,
1978, Tsai et al., 1991, Christ et al., 1989, and Grabow,
1991). PDMS has not been used alone clinically as a vascular
graft material, but has been studied for the application as
early as 1955 (Hufnagel, 1955). PDMS has also been used by
the National Institutes of Health/Heart, Lung, and Blood
Institute (NIH/HLBI) as a low-thrombogenicity reference
polymer.
PDMS is typically used in the form of a crosslinked
elastomer for applications demanding mechanical strength.
(Lower molecular weight and lower cross-link density oils and
gels find other applications.) The tensile and tear strength
of silicone is somewhat lower than other vascular graft
materials, and reinforcement is needed. Reinforcement is
achieved with either silica filler particles dispersed within
the silicone, or continuous fiber reinforcement. Typical
mechanical properties of silica filled and unfilled PDMS
elastomers fall in the range of 2.4 to 6.9 X 106 Pa (100 to
1200 psi) tensile strength, 1 X 105 to 1 X 107 Pa(15 to 1450
psi) modulus, and 20 to 700% elongation. The high elasticity
and low Tg (ca. -123 *C) of silicone elastomers afford
modulus values similar to the natural vessel. Table 2.1
shows compliance measurements made by Abbott and Cambria
(1982) for vascular graft materials as they compare to the
natural vessel before and two weeks after implantation.
However, one major concern for PDMS has been the long-term
changes which may accompany the known affinity of PDMS for
lipids, which results in a more brittle material.
Silicone elastomers used as IOLs and mammary prosthesis
shells could benefit from surface modification as well. The
most successful material used for IOL applications to date
has been PMMA (Joo and Kim, 1992, and Balyeat et al., 1989).
The new small incision techniques developed to insert
foldable IOLs with less invasive surgery has utilized
silicone as the implant material (Christ, 1989 and Grabow,
1991). If PMMA surface properties were to be applied to
silicone elastomers without increased rigidity, the success
of the PMMA implants might be achieved with modified PDMS
using less invasive techniques.
The recent criticism of silicone gel mammary implants
has created a frenzy of research for alternative materials.
Although the major problems are gel bleed of the silicone gel
and low molecular weight oligomers through the elastomer
shell and rupture of the shells due to gel swelling and
possible biodegradation, some research shows adverse tissue
reactions, such as fibrous capsule formation and hardening of
the implant, to be problematic as well (Habal et al., 1991).
A PMMA surface graft may reduce the adverse tissue response
based on the more favorable biocompatibility of PMMA.
Because of the difference in solubility parameters between
PMMA (9.4 (cal/cm3)1/2) and silicone (7.3 (cal/cm3)/2)
reduction of gel bleed in PMMA modified mammary prostheses is
expected (Rodriguez 1982).
Table 2.1 Initial and 2 week post implant dynamic
compliance values for canine vascular graft
materials. Compliance is given as percent radial
change per mm Hg (X 10-2) Data taken from Abbott
and Cambria, 1982.
Graft Implant External Initial 2 Week
Material Diameter Compliance Compliance
Normal Femoral Artery 4.69 0.007 5.86 0.26
Femoral Arterial Graft 4.53 t 0.50 4.41 0.80 5.67 1.1
4 mm ID DacronO 6.41 t 0.16 1.46 0.12 1.19 0.1
PTFE 5.08 t 0.05 5.08 0.05 1.3 0.4
Silastic Rubber 6.06 0.50 5.95 0.09 6.2 0.7
Polyurethane 5.27 0.15 6.10 1.10 2.3 0.4
Successful surface modification of PDMS elastomers using
gamma radiation with hydroxyethylmethacrylate, (HEMA),
ethylene glycol dimethacrylate, and copolymers has been
reported by Heaburn et al. (1978). Hoffman et al. (1983)
also report grafting of HEMA and ethyl methacrylate polymers
and copolymers onto silicones. Lin (1995) has successfully
modified PDMS elastomers, Shin-Etsu silicone in particular,
with PVP.
2.1.4. Advantages of PMMA. SEMA. and PDMS Surfaces
2.1.4.1 Advantages of a PMMA surface
PMMA and other acrylic polymers have been used as
biomaterials with much success. Acrylics are currently used
for intraocular lens (IOL) implants (Amon and Menapace,
1990), hemodialysis membranes (Falkenhagen and Brown, 1991),
dental resins, and bone cements, and are studied for blood
compatibility for general biomaterial applications (Apple et
al., 1984, Feuerstein et al., 1991, Feuerstein et al., 1992,
Lentz et al., 1985, Ito et al., 1992, and Sherman et al.,
1963). A general summary of the recent literature on PMMA
and other acrylics would be to label these materials as
relatively bioacceptable for many invasive applications.
Apple et al. (1984) provide an excellent review of the
studies done by Ridley and others for work with PMMA
concerning IOL implants.
Acrylic chemistry has also been reviewed because of
concerns regarding hydrolysis of the ester-carbonyl bond in
the side groups of polyacrylates such as PMMA. Acrylics are
surprisingly resistant to degradation, however, and acrylic
elastomers such as ethyl acrylate have been used widely
because they are more resistant to oxidation, ultraviolet
light damage and hydrolysis than other traditional elastomers
(Rodriguez, 1982 and Sperling, 1986). Long alkyl pendant
groups provide less stability. Thus, methyl methacrylate is
more stable than ethyl, butyl, and isopropyl acrylate, and
the stability decreases with increasing pendant group size.
The best example of the vascular compatibility of PMMA
was presented many years ago at the 1954 annual meeting of
the Society for Vascular Surgery. Hufnagel (1955) reported
that tubes of methyl methacrylate polymer (Lucite) had been
used as thoracic artery replacements in canines for up to 6
years with complete maintenance of patency. Other materials
tested in this study included polyethylene, Teflon*, Kel-F,
nylon, woven stainless steel mesh, Vitallium, other metals,
and silicone rubber. These materials were tested with and
without silicone coatings, and no material was as successful
with respect to low thrombogenicity as methyl methacrylate
(Hufnagel, 1954). The silicone coated prostheses had poor
compatibility, which Hufnagel attributed to incomplete
removal of acids formed in the curing reactions.
The use of methyl methacrylate as a vascular prostheses
has not been reported in the intervening years, probably
because the need for flexible materials has eliminated
consideration of this rigid polymer. Even though Hufnagel
collected data on over 400 implants of rigid vascular
replacements, he indicated the importance of using a material
which had mechanical properties similar to the natural
vessel. Work reported from the late 1950s to date has
therefore shifted to softer, more flexible, and more
hydrophilic surfaces. The potentially excellent
nonthrombogenic surface properties of PMMA thus, seems to
have been lost in the literature. Since the technology now
exists to graft polymerize very thin glassy polymer surfaces
onto flexible substrates without significant reduction in
elasticity, the surface modification of vascular and blood
contact devices now seems logical and promising.
Current work in our laboratory indicates PMMA has
favorable cell adhesion properties with respect to platelet
and endothelial cell adhesion (Goldberg et al., 1988-1995).
A reduction in the ratio of adhered platelet cells to
endothelial cells is sought for blood contact devices such as
a vascular grafts. Observations in our laboratories of in
vitro platelet and endothelial cell adhesion assays show
unmodified PMMA control samples to have a more favorable
ratio (greater endothelial cell adhesion than platelet
adhesion) than other materials studied, including surface
modified PMMA (Goldberg et al., 1988-1995). The affinity of
PMMA for endothelial cells becomes more apparent when
considering research on PMMA for IOL applications. A common
problem of unmodified PMMA IOL materials is excess epithelial
cell adhesion and growth on the implant (Goldberg et al.,
1988-1995, Yahiaoui, 1990, and Lin, 1995). Assuming that
endothelial and epithelial cells exhibit similar surface
adhesion properties on polymers, IOL studies suggest that
PMMA would favor endothelial cell adhesion. Finally, in
studying a new method to measure the adhesive strength of red
blood cells (RBCs) to biomaterials, Bowers et al. (1989)
found less adhesion of RBCs to control PMMA samples than to
hydrophilic glass, tissue culture grade polystyrene, and PET.
2.1.4.2 SEMA surface advantages
Sulfonated surfaces have been investigated in the
vascular environment primarily because of the strong anionic
surface charge contributed by sulfonic acid functional
groups. Many blood components, including red blood cells and
platelets, have a slightly negative surface charge. It
therefore seems logical, that surfaces with negative charges
will repel these components of circulating blood. Our
laboratory has been investigating the use of sulfonated
monomers for surface modification such as potassium 3-
sulfopropyl acrylate (KSPA), sodium methacrylate (SMA),
styrene sulfonic acid, sodium salt (SSA), and other anionic
sulfur containing monomers since the late 1980s (Goldberg et
al., 1988-1995 and Yahiaoui, 1990). These monomers are
supplied as water soluble sodium or potassium salts. It is
often difficult to obtain an intimate mixture of monomer and
substrate molecules through presoaking in aqueous media,
because of the slightly hydrophobic surface of PET. Pre-
treatment with other monomers such as N-vinyl pyrrolidone
(NVP) is often required to obtain grafting of these monomers.
However, some sulfonated monomers, such as 2-sulfoethyl
methacrylate (SEMA), are available as organic, non-salt
compounds, and are therefore soluble in organic solvents such
as acetone and dimethyl sulfoxide (DMSO). Dissolution in
these solvents increases the swelling of substrates during a
presoak process, and grafting is more easily achieved.
Critical components of the vascular environment include
chondroitin sulfate, heparin sulfate, and dermatan sulfate,
which are all proteoglycans secreted by the endothelium.
Marcum et al. (1986) reported heparin sulfate is capable of
binding with antithrombin, and thus imparts antithrombogenic
properties to the vascular endothelium. Ofosu et al. (1989)
report that increasing the degree of sulfonation on heparin
and dermatan sulfate increases the catalytic effects on
thrombin inhibition. Other compounds with functional sulfate
groups are thought to have favorable reactivity within the
vascular environment. Kishida et al. (1991) studied
polyethylene films surface modified with cationic, anionic
(sulfonated and non-sulfonated), and non-anionic monomers,
and found the in vivo cell adhesion to be related to both the
charge and the presence of sulfonated groups, with sulfonated
polymers having higher HeLa S3 cell attachment and growth
than non-ionic and cationic surfaces, indicating a higher
affinity for binding, attachment, and growth of cells to
anionic surfaces. The incorporation of sulfonated functional
polymers on vascular grafts seems promising in reducing
thrombus formation.
2.1.4.3 PDMS surface advantages
PDMS is a hydrophobic material with a low surface energy
(contact angle ca. 80). The low surface energy prevents
adhesion of many compounds from an aqueous environment, as
these molecules are often repelled by silicone surfaces.
Two key proteins determining the thrombogenicity of a
biomaterial are albumin and fibrinogen. Albumin adsorption
is preferred for a non-thrombogenic surface, and fibrinogen
adsorption usually indicates a thrombogenic surface, as
fibrinogen is converted to fibrin by thrombin. Cooper and
Fabrizius-Homan (1991) found silicone rubber to have a higher
affinity for albumin than fibrinogen in competitive
adsorption studies, and when the albumin was preferentially
absorbed, the thrombogenicity of the material was reduced.
This ratio of albumin to fibrinogen was most favorable for
silicone when compared to polyethylene, polyurethane, and
Teflon*. In a canine ex vivo AV shunt platelet adhesion
study by Ip and Sefton (1991), it was found that SilasticO
(PDMS, Dow Corning) and silica free PDMS (Thoratec) both had
significantly lower platelet cell adhesion than polyethylene.
Norgren et al. (1990) found silicone coated Dacron& to
have a reduced thrombogenicity, and Granke et al. (1993)
found reduced inflammatory reaction as well. In the study by
Granke, only the outer surface of the graft was covered to
create a prosthesis which did not require pre-clotting, but
there was significantly more tissue ingrowth and
endothelialization in the silicone treated samples than in
the control samples. Whalen et al. (1992) evaluated a novel
prosthesis made entirely of silicone. The prosthesis was
porous to allow tissue ingrowth, and after implantation in
canines for 8 weeks, had an overall patency of 86%.
Two studies within the research group of Cooper (Lin et
al., 1994 and Silver et al. 1995), report modification of
silicone elastomer surfaces in an attempt to improve the
biocompatibility. Lin et al. (1994) compared silicone to
polyurethane-silicone copolymers and found the compatibility
of the silicone to be superior with respect to clotting time
and platelet deposition. Silver et al. (1995) modified
silicone surfaces with alkylsiloxane monolayers of various
functionalities following exposure to an oxygen radio
frequency plasma. The materials were evaluated in a canine
ex vivo arteriovenous (AV) shunt, and the untreated silicone
had superior properties with respect to platelet and
fibrinogen deposition. In a study by Morel et al., (1989)
the endothelial cells cultured on thin silicone sheets were
found to have motility and contractility indicating a healthy
environment for cell growth and proliferation. Finally,
numerous studies using a canine AV shunt use Silastic, or
some other silicone elastomer, as the tubing in which samples
being analyzed are placed (Lin et al., 1994, Silver et al.,
1995, Goldberg et al., 1985-1995). The silicone tubing would
not be used in these studies if occlusion occurred.
The recent controversy surrounding the use of silicone
gel filled breast implants will obviously invoke a negative
response to the idea of using silicone as a vascular graft
surface. However, the silicone surfaces discussed and
studied in this research are crosslinked elastomers, and the
toxicity and poor biological response to silicone materials
related to breast implants are for the low molecular weight
oils and gels (Kimitoshi et al., 1990 and Kimitoshi et al.,
1991).
Current literature and research indicate a silicone
surface in the vascular environment could have favorable and
beneficial responses. This review led to the investigation
of PDMS coatings on Dacron' vascular prostheses presented
here.
2.1.5 Gamma Radiation Initiated Polymerization
Gamma-rays are electromagnetic waves of short wavelength
(X= < 0.1 nm) which are emitted from a decaying radioactive
source. One of the most commonly used sources of gamma-ray
energy, or gamma radiation, is the radioactive isotope of
cobalt, cobalt-60 (60Co). Providing two sharp spectral lines
of radiation energy of 1.17 and 1.33 MeV (megaelectron
volts), 60Co is often chosen because of its ease of
preparation (nuclear activation of cobalt-59) and its long
half life of 5.3 years (Chapiro, 1962).
Gamma radiation energy generates free radicals on vinyl
monomers and polymeric substrates, making it an excellent
initiation source for free radical polymerization of monomers
onto polymeric surfaces (Chapiro, 1962). Gamma radiation
initiation polymerization (GRIP) has been studied since the
late 1940s, and is often chosen over other techniques because
of its low cost and cleanliness (does not introduce chemical
initiator molecules). Low cost and cleanliness are two
extremely critical factors in determining the success of many
materials for the biomedical industry.
Surface modification through GRIP involves the formation
of free radicals on monomer molecules leading to free radical
polymerization. The simultaneous exposure of the monomer and
substrate allows both initiation and cleavage of substrate
polymer. The growing homopolymer chains may propagate or
terminate with the available reactive sites created on the
substrate, creating a surface region of grafted, crosslinked,
and interpenetrating network molecules. Swelling the
substrate with monomer in solution presoakk technique)
provides a localized monomer-rich region within the substrate
surface, and the surface region becomes an intimate mixture
of substrate and monomer molecules (Yahiaoui, 1990). Upon
exposure to gamma radiation, this entire region becomes a
surface "graft," referring to a region of grafted and IPN
polymer molecules. The intimate mixture of the swollen
region also facilitates the diffusion of the monomer and
homopolymer chains to the activated substrate sites,
increasing the efficiency of the grafting.
2.1.5.1 Polymer radiolysis and free radical reactions
Chapiro (1962) provides an excellent review of early
studies of radiation effects on polymers and monomers. Upon
exposure to high energy radiation, polymers undergo a series
of reactions including, but not limited to, radiolysis of
side chain atoms and functional groups, free radical
formation, crosslinking, chain scission, and degradation.
Although the exact nature of the mechanism of these reactions
with systems as complex as polymers is still not fully
understood, radiation polymerization following initial
radiation events is fairly well documented, and several
events may occur upon simultaneous exposure of monomer
solutions and polymer substrates.
When a polymer or monomer is exposed to gamma radiation,
cations, anions, and free radicals are created. The ions are
only stable at low temperatures, and usually dissociate to
yield radicals (Chapiro, 1962). Graft polymerization may
take place when a polymer and monomer in intimate contact are
simultaneously exposed to high energy radiation. Initially,
a radical may be formed on the substrate polymer or the
monomer. When a monomer molecule reacts with one of these
two radicals, either a graft polymer or homopolymer forms,
respectively. Graft polymerization is favored if the polymer
substrate has a higher ability to cross-link after chain
scission rather than degrade, and more grafting occurs if
radical yields are higher for the substrate polymer than the
monomer (Chapiro, 1962).
Radicals are also created on the growing homopolymer
chains, giving rise to higher molecular weight homopolymers,
branching, or crosslinking. Radical formation on side groups
of growing homopolymer chains or substrate chains yields
branching, and termination of radicals by combination with
branched chains from different molecules yields crosslinking.
These polymer reactions, if occurring within the substrate,
lead to the formation of an IPN.
Substrate main chains may be cleaved as a result of the
high energy of the radiation. The cleaved chains then form
crosslinks with other cleaved substrate chains; form graft
polymers or cross-links by the addition of monomer or
combination with a growing homopolymer chain; or re-combine
with the original site of cleavage (Yamamoto and Yamakawa,
1980). If cleaved chains terminate by disproportionation or
chain transfer, the molecular weight of the substrate polymer
is reduced, and degradation results.
During the course of the polymerization, the polymer
within the substrate surface begins to gel as higher
molecular weights are reached and as cross-linking and
branching occur. This decreases the mobility of larger
propagating molecules. The concentration of monomer
decreases and the concentration of radicals increases rapidly
as higher conversions are reached (Dob6, 1978), leading to
higher degrees of cross-linking and branching in the surface
polymer than in the surrounding solution polymer, as the
reaction auto accelerates.
The efficiency and stability of radical formation on
various substrates and monomers determine the ease or
difficulty of polymerization of a specific monomer to a
specific substrate polymer surface. The radiolysis and
radical formation on the various substrates and monomers
studied in this research will now be discussed.
2.1.5.2 Radical formation on PET
Charlesby (1953) found PET to crosslink upon exposure to
gamma radiation, whereas Todd (1954) found it to undergo
degradation. Low dose exposures always gave an increase in
modulus of Dacron* fibers when studied by Teszler and
Rutherford (Chapiro, 1962). Radical formation on the phenyl
ring without atomic ejection is possible, but low yields are
expected because of the resonance of the ring (Chapiro,
1962). Cleavage of the ester bond or radical formation on
the main chain with ejection of an H atom are the most likely
reactions, as shown in figure 2.4 (Chapiro, 1962). Radicals
are formed on both the amorphous and crystalline regions of
the polymer, but polymerization is only expected in the
amorphous regions because of the reduced diffusion through
the crystalline phase.
2.1.5.3 Radical formation on PTFE and ePTFE
Main chain cleavage is possible in PTFE, but radiation
more commonly yields free radicals from cleavage of the C-F
bond (Rye, 1988). Charlesby (1952) analyzed gas evolution of
PTFE upon exposure to gamma radiation and found the products
to be carbon tetrafluoride, indicating evolution of both
carbon and fluorine. In the presence of oxygen, doses as low
as 1 to 10 Mrad lead to significant degradation and
embrittlement of PTFE. When oxygen is excluded from the
system, however, much less damage occurs. Possible reaction
mechanisms are shown in figure 2.4.
2.1.5.4 Radical formation on PDMS
Radical formation on silicone polymers was studied by
Charlesby and Omerad (1963), and the polymers were found to
crosslink upon exposure to gamma radiation. The Si-O bonds
are significantly more stable than the Si-C and C-H bonds,
leading to radiolysis of the pendant methyl groups. Possible
reactions are shown in figure 2.4.
2.1.5.5 Radical polymerization of vinyl monomers
Methyl methacrylate monomer readily forms radicals when
exposed to gamma radiation, and is expected to polymerize
well on all of the chosen substrates. The resulting surface
polymer is expected to be an interpenetrating network of
grafted, cross-linked, and branched polymer.
O O
---( --OH *CH--CH--O--
I
0O O
- -C-0 ...)-C-CH-CH2a-I-o-
Radiolysis of PET
F
--CFT-CFZ-CF--- Radiation -C C or --Cr
F F
Radiolysis of PTFE
CH3
S- Gamma
--Si-Of-- Radiationb
CH3
CH2
--
&3 &3
Radiolysis of PDMS
Figure 2.4 Possible mechanisms of gamma radiation induced
radiolysis and free radical formation for PET,
PTFE, and PDMS.
2.1.5.6 Polymerization of vinyl monomers onto substrates
Chapiro (1962) discusses the free radical polymerization
of vinyl monomers induced by gamma radiation. The mechanism
of reaction proceeds via traditional free radical processes,
with the rate of polymerization being proportional to the
monomer concentration and the square root of the dose rate.
The equation for the kinetic chain length (average number of
monomer molecules polymerized per initiating primary radical)
follows that of chemical free radical polymerization as well.
The kinetic chain length for gamma radiation polymerization
is proportional to the square of the monomer concentration
and inversely proportional to gamma radiation dose rate.
2.1.6 Polymer Solution Coatings and Techniques
When a solid material is removed from a liquid, (solution or
pure liquid) the surface attractive forces are typically
strong enough to adhere a monolayer or more of the liquid to
the solid surface. The relative surface energies of the
solid and solution determine the thickness and adhesion of
this surface coating. If a solute is dissolved in the
solution, and the solvent is rapidly evaporated after the
object is removed from solution, there will be a layer of
solute on the object. The relative surface energies of the
solute, solvent, and object as well as the evaporation rate
and solution concentration determine the amount remaining on
the surface. The surface energies of the solute and solid
determine the strength of the adhesive bond between the
surface layer and the object. This general principle of
surface phenomena was utilized in the dip coating of PDMS
onto Dacron* vascular prostheses.
Both PET and PDMS are relatively hydrophobic (contact
angle of PET ca. 60, and contact angle of PDMS ca. 80'), and
both materials swell in the organic solvent chloroform. For
this reason, solutions of the two part PDMS oligomers were
made in chloroform, and Dacron was soaked in this solution.
The swelling of the DacronO by the chloroform is believed to
allow the diffusion of PDMS molecules into the PET surface,
creating a swollen network of PET, PDMS, and chloroform.
Upon removal of the PET from the PDMS solution, the
chloroform evaporates, de-swelling the surface network,
leaving PDMS oligomers on and in the PET surface. The
relatively rapid evaporation of chloroform, and the similar
surface energies of PDMS and PET allow a layer of PDMS
oligomers to remain on the prosthesis after removal from the
solution.
In an attempt to improve the bonding between the PDMS
and Dacron*, a pre-modification step was implemented. MAOP-
t-PDMS was gamma polymerized onto Dacron* to provide a link
between the coating and the substrate. A covalent link
between PMMA-g-PET and the MAOP-t-PDMS is expected by using a
presoak of MMA followed by polymerization of MAOP-t-PDMS.
The incorporation of MMA into the system will provide a
propagating link between the substrate and the MAOP-t-PDMS
thus allowing polymerization of more MAOP-t-PDMS. The dip
coating of PDMS should then polymerize and crosslink with the
MAOP-t-PDMS.
2.1.6.1 Thermal curing of PDMS
The curing of Shin-Etsu PDMS proceeds via vinyl addition
polymerization, and is initiated by a platinum catalyst. The
reaction scheme for this polymerization is shown in figure
2.5.
CH,
----Si---Ca=CH + H-Si -CH
CH3 0 CH3
C13 -Si -H + CH2 ==CH -Si ---+
I I
CH, o c13
III
-Si-CH=CH2 + H-Si--1CH3
CH3
Pt Catalyst
CH,
(--O-Si --- -'---Si ---M,
CH3 0 CH3
CH3-Si -H, -01k -Si -0-0
CH3 0 -13
-(-0-Si -O-r---lr--Si --13
013
Figure 2.5 Reaction mechanism for thermal curing of Shin-
Etsu PDMS.
2.2 Intracorneal Implants
2.2.1 Refractive Corneal Surgery and Intracorneal Implants
Myopia (nearsightedness caused by excessive corneal
curvature), hyperopia (farsightedness caused by reduced
corneal curvature), and astigmatism (irregular corneal
curvature) are disorders affecting the cornea which cause
impair vision. Accidental injuries that abrade, scratch,
deform, or scar the cornea also significantly reduce visual
acuity. These disorders and damages of the cornea are
currently corrected by the use of spectacles (glasses or
contact lenses), a corneal transplant for severe conditions,
or with some of the newer surgical techniques such as radial
keratotomy (RK) or laser keratotomy. Intracorneal lenses or
rings provide still another option, and are being studied by
numerous researchers and ocular device companies.
Glasses and contact lenses will always be an option for
correcting many corneal deformations because they have the
advantage of always being reversible. If a problem occurs or
an error was made in fitting the diopter of the corrective
lens, the device may simply be removed. Corneal transplants,
are a major surgical procedure, and a donor organ is
required, restricting their use to severe cases of complete
corneal damage or damage which can not be repaired by other
means. Radial and laser keratotomy scars into the cornea and
relies on healing to alter the shape and modulus of the
cornea, and is an extremely unpredictable procedure (Waring,
1990). The use of RK and other similar procedures offers
patients a quick procedure with extreme pain and little
guarantee of success, as glasses are often still required.
However, it is often difficult to treat severe
astigmatism with contact lenses, and greater deformations
require large and often cumbersome spectacles, and damage or
deformations too great for other means of corrections often
requires treatment with a non-reversible therapy (Kerry,
1995). Keratoprostheses such as ICLs offer a more permanent
solution to refractive corrections of the cornea.
In the late 1940s, Barraquer began experiments on the
implantation of synthetic materials within the corneal stroma
of the eye to correct irregular curvatures (Barraquer, 1949).
The initial designs were lenses constructed of PMMA and
polysulfone that were surgically implanted within the corneal
stroma to provide both mechanical (corneal reshaping) and
refractive (lens power magnification) correction of the
cornea. More recent designs include hydrogel lenses,
intracorneal rings, and fenestrated PMMA lenses.
2.2.2 Intracorneal Lens Materials and Designs
Polysulfone is an optically clear, amorphous, stiff, but
flexible polymer with a glass transition temperature (Tg) of
190'C and a refractive index of 1.63. Polysulfone ICLs are
less rigid than those made with PMMA, and rely more on the
high refractive index for correction of the cornea. PMMA is
an optically clear, amorphous, glassy polymer with a Tg of
117 "C and a refractive index of 1.49 (Most commercial grades
of PMMA contain a small fraction of other acrylic polymer
such as polyethylacrylate which reduces the overall Tg to
around 105'C.). PMMA ICLs provide both refractive
corrections and mechanical correction, and PMMA intracorneal
rings rely solely on mechanical changes in the corneal shape
to provide correction.
The initial designs of a refractive lens made of
polysulfone and PMMA were proved to be successful at
correcting corneal correction, but implant studies have shown
them to be unsuitable as long term implants thus far (Kerry,
1995 and Lane et al., 1989). Both polysulfone and PMMA
lenses are impermeable to water and aqueous solutions, and
the initial design of these ICLs hinders the diffusion of
water, ions, proteins and other vital nutrients to the
corneal stroma anterior to the implant. The result of these
implants was corneal opacification and necrosis in the
deprived regions, and lipid deposits posterior to the
implants (Climenhaga, 1988).
Although the lens designs are still being studied,
recent studies have focused on a lens made with hydrogel
materials (McDonald et al., 1993). Hydrogel lenses (e.g.,
Permalens*, CooperVision and Lidofilcon A, Allerghan Medical
Optics) are polymers and copolymers containing hydrogels such
as polyhydroxyethyl methacrylate (pHEMA),
polyvinylpyrrolidone (PVP), and polyacrylic acid (pAA). The
hydrogel component in these lenses makes them water
permeable. Claims have been made that since the hydrogel
lenses are water permeable, it is therefore also permeable to
glucose and other metabolites necessary to the corneal stroma
(Werblin and Patel, 1992, Werblin and Peiffer 1992, McCarey,
1990, McCarey, 1981). However, diffusion or permeability of
water does not guarantee the permeability of other molecules,
especially those significantly different in size and
functionality. Hydrogel lenses also have fairly poor
mechanical properties, and tearing or damage to the lens
during handling, implantation, or use is possible (Menapace,
1990).
The intracorneal ring (ICR) design allows passage of
nutrients and offers curvature correction by flattening the
central cornea (Kerry, 1995). The ring is surgically
implanted by sliding and rotating it into an intrastromal
channel created by a radial incision of the stroma lamella
(Kerry, 1995). The design of a PMMA ring provides mechanical
correction of the cornea without hindering the diffusion of
nutrients to the stromal region anterior to the implant.
(The absence of a refractive center in the ring requires
application of more mechanical "pressure" to achieve the same
correction of the lens designs. Calculating the required
ring shape is often difficult, and may be complicated by
unpredictable healing of the cornea following the implant
surgery (Quantock et al., 1995).
The Surgidev intracorneal lens (ICL) is a fenestrated
PMMA lens designed to correct the corneal refractive power
both mechanically (by corneal reshaping) and refractivly
(magnification by the ICL). The holes in the ICL are present
to allow diffusion of stromal nutrients to the anterior
implant region and still provide refractive correction. The
primary component of interest which is vital to stromal
tissues is glucose. Glucose is delivered to the stroma via
the aqueous humor, and its normal concentration is constant
parallel to the stroma, and decreases across the stroma from
880 to 580 pg/ml (88 to 58 mg/dl), posterior to anterior
(McCarey and Schmidt, 1990).
2.2.3 Surface Properties of Ocular Biomaterial Implants
The corneal endothelium plays a critical role in the
balance of fluids and nutrients within the cornea itself as
well as the eye posterior to the cornea. The implantation
and residence of an ICL or ICR in the cornea requires a
minimally damaging procedure because the human adult corneal
endothelium does not regenerate when damaged (Bourne and
Kaufman, 1976). The work in our laboratories has shown that
PMMA placed in contact with the corneal endothelium or
epithelium will strip away vital cells upon removal
(Yahiaoui, 1990, Katz et al., 1977, and Sheets, 1983).
Studies by Andrade (1985) and Absolom et al. (1987) show the
tendency of hydrophobic surfaces (such as PMMA) to strongly
absorb proteins, whereas hydrophilic surfaces tend to readily
desorb proteins following initial adsorption. Likewise, it
was determined by our research that PMMA modified with NVP
(Hydrograft*) provides a surface which will not readily
adhere to these cell layers on contact. The Hydrograft*
surface imparts less damage to surrounding tissues during
implantation than would an unmodified PMMA lens by providing
long term lubrication of the tissues by the hydrophilic
grafted surface. Also reported by Yahiaoui (1990) and
continually observed in current research within this
laboratory, radiation induced graft polymerization with NVP
(Hydrograft* modification) significantly reduces the
adhesion, growth, and spreading of ocular epithelial cells on
PMMA ocular implant materials.
Surface modification of the PMMA ICLs NVP is expected to
provide a hydrophilic surface within the ICL fenestrations,
and increase the permeablity of the lens to vital stromal
nutrients. This research focuses on the determining suitable
surface modification conditions for low molecular weight PMMA
ICLs and resulting changes in permeability.
2.2.4 Surface Modification Techniques
The presoak surface modification techniques discussed
previously for vascular graft modifications will be used for
PMMA ICL modifications. The monomer solutions, however, are
aqueous because of the sensitivity of the ocular environment
and the low chemical resistance of PMMA to organic solvents.
2.2.4.1 Radical formation on PMMA
Figure 2.6 shows several possibilities for the
radiolysis of PMMA The abstraction of a hydrogen from the
main chain may create Radical A, which upon
disproportionation leads to the formation of Radical B and
End Group A (Todd, 1954). Main chain homolysis is also
possible, yielding Radical C and D (Kirsher et al. 1965).
Degradation may occur upon ester group cleavage, yielding
Radical E. Scission of Radical E may then lead to formation
of a new Radical D and Chain End B (Ranby and Rabeck, 1975).
CH3 CH3 C13
H--CH--- -C + C=CH-C--
COOCH3 COOCH3 OOCH3 OOCH3
Radical A Radical B Unsaturated
Endgroup A
CH3 CH3 CH3 CH3 CH3
I I a aI I I
,-C^_ C-CH,._ ^ C^^. + CH-C--
2 Radiation" 2
COOCH3 COOCH3 C COOCHCOOCH3 COOCH3
PMMA I Radical C Radical D
C3 C3 CH3
I I I
--C-CH-C--CH,-- P --C==CH2 + Radical D
COOCH
Radical E Unsaturaded
Endgroup B
Figure 2.6 Possible gamma radiation reactions and products
for PMMA.
CHAPTER 3
MATERIALS AND METHODS
3.1 Materials
3.1.1 Substrates
Surface modification was carried out on several vascular
prosthesis substrate materials. These included PET (Mylar D-
1000 and 700-Dl films from DuPont Electronics), woven PET
(Dacron fabric, Meadox), reinforced velour woven PET
(reinforced Dacron vascular graft, Bard), PDMS (KE-1935 A and
B, Shincor Silicones), ePTFE (sp.# 728-3 GORE-TE* expanded
PTFE, Gore), and PTFE (skived Teflon", Goodfellow). The
chemical structures of polymer substrates used in this
research are presented in figure 3.1.
All intracorneal lens (ICL) substrates for Hydrograft*
modification were PMMA, and were provided by Surgidev as
either PMMA sheets (V-811 low molecular weight PMMA),
fenestrated flat PMMA disks, or ICLs. These ICLs have three
hole sizes and percent surface coverage of holes, 10pm/5%
(209-99), 30pm/5% (220-121), and 80pm/5% (220-131), and all
are 6 mm/-8 diopter. The PMMA used in the manufacture of the
ICLs has a significantly lower molecular weight than PMMA
typically used for ocular applications. For this reason,
different conditions than those studied by Yahiaoui (1990)
were used for modification of ICLs. Table 3.1 shows
molecular weight data for several PMMA ocular materials.
O O
PET -<>-CHr-CH2
CH3
PMMA -(-CH =c
H CH,
I
PDMS i
F F
PTFE
F F n
Figure 3.1 Chemical structures of polymer substrates
subjected to surface modification.
Table 3.1 Molecular weight data for PMMA ocular implant
materials, from Goldberg, et al. 1988-1995).
Name Lot # Mn MW MWD
Perspex CQCV 001544 3.3 X 106 5.1 X 106 1.5
Blue Perspex B# 001710 1.2 X 106 3.6 X 106 2.9
Opticlear P973 1.1 X 106 2.5 X 106 2.4
Low MW PMMA V-811 45,000 86,000 1.9
Nidek IOL PMMA 650,000 2.6 X 106 3.9
Storz IOL PMMA 60,000 110,000 1.8
3.1.2 Monomers and Reagents for Surface Modification
The monomers and polymers used for surface modification
included methyl methacrylate (MMA, Kodak), N-2-vinyl-
pyrrolidone (NVP, Polysciences), 2-sulfoethyl methacrylate
(SEMA), methacryloxypropyl terminated PDMS (MAOP-t-PDMS,
United Chemical Technology, Inc.), and PDMS (KE-1935 A and
B). MMA and NVP were supplied with MEHQ as an inhibitor, and
were purified by vacuum distillation (1-2 mm Hg at 40-60'C).
The Shin-Etsu silicone was supplied in a ready to use two
part oligomer mixture. SEMA was supplied with 5% MEHQ, and
no practical methods of its removal were found. Therefore,
SEMA was used as supplied. The MAOP-t-PDMS was supplied with
no inhibitor, and was used as received. The structure of the
monomers and oligomers used are displayed in figure 3.2.
3.2 Methods
3.2.1 Sample Preparation and Substrate Cleaning
3.2.1.1 Preparation of Shin-Etsu PDMS Films
The KE-1935 PDMS was supplied by Shincor Silicones in
two separate parts. The mechanical properties of the final
polymer may be varied by altering the time and temperature of
the curing reaction. Samples used in this research were
prepared by the following standard procedure unless otherwise
noted. Twenty to twenty-two grams of each component were
poured onto a bordered glass plate (16 cm X 24 cm), mixed
thoroughly with a glass stirring rod, and the surface was
leveled with a glass microscope slide. The plate was
degassed in a vacuum oven (76 cm Hg) at 50'C for 45 minutes,
and then transferred to a preheated, 150 'C ,oven and cured
for one hour in air. The final thickness of the cured sheet
was approximately 1 mm.
CH2=CH CH3 CH3
//o CH2---C CH==C
C=0 =0
I I
CH3 'CHT-CH2----OH
O
NVP MMA SEMA
CH3
CH2J=
I
C=0
I
0 CH3 CH3 CH3
H- CHT-CH 'Ii-O -- i '--- i-CHy-CHr ,
CH3 CH3 CH3 O
0=CH
MAOP-t-PDMS CH3
Figure 3.2 Chemical structures of monomers and reagents
used for surface modification.
3.2.1.2 Substrate cleaning prior to surface modification
All substrates were cleaned prior to use to remove
surface contamination and impurities. Samples referred to as
"controls" have undergone the cleaning process as well,
unless otherwise mentioned. Dacron", Mylar*, PTFE, ePTFE,
and silicone substrates were cleaned by sonication for 10
minutes each in acetone, isopropanol, and Ultrapure" water.
The samples were then rinsed in Ultrapure" water and placed
in a vacuum oven at 50'C to dry for 12 hours (76 cm Hg).
PMMA samples were cleaned by sonication in Ultrapure"
water for 10 minutes followed by a rinse in Ultrapure" water.
Samples provided as lenses were not cleaned prior to
modification because they were received in final manufactured
condition.
3.2.1.3 Solution degassing
The samples for gamma radiation surface modification
were placed into borosilicate glass tubes with the grafting
monomer solution (sample completely submerged in the
solution). Before irradiation, the samples were "degassed"
to remove as much oxygen from the solution as possible using
one of two methods. The samples were degassed by vacuum (ca.
20-30 mm Hg) for 2 to 5 minutes, purged with argon, and
sealed with a polyethylene cap (Tainer Top, Fisher
Scientific). The second method used was to bubble argon
through the solution with agitation to replace the oxygen
with argon in solution. Bubbling was done through a glass
pasteur pipette for 2 to 5 minutes (2 minutes for 1 to 3 ml
solution volumes and 5 minutes for 5 to 10 ml solution
volumes). Bubbling was chosen over vacuum degassing for more
volatile solutions (e.g., methylmethacrylate monomer/acetone
solutions) to avoid evaporating of solution components and
changing solution concentrations. All samples were then
irradiated to the specified dose.
3.2.1.4 Substrate cleaning after modification
Following surface modification with MMA, SEMA, or PDMS,
the residual polymer, monomer and solvent was removed from
the sample tube. The modified substrates (PET, PTFE, ePTFE,
and PDMS) were then placed in acetone to begin washing to
remove the remaining unbound polymer and residual monomer.
The acetone was removed and replaced three times per day, for
three days. The samples then were rinsed in Ultrapure" water
and vacuum oven dried.
PMMA samples modified with NVP were cleaned in the same
manner described, with Ultrapure" water being substituted for
acetone for the washing procedure.
3.2.2 Surface Modification Methods
3.2.2.1 Presoaking
Some samples were subjected to a presoak step prior to
gamma polymerization. The presoak step involves placing the
sample to be modified in a monomer solution, usually at
higher concentration and temperature, to allow monomer and
solvent to diffuse into and swell the substrate surface. The
presoak conditions for a polymer-monomer system were usually
determined by swelling experiments. The samples were placed
in various solution concentrations in sealed borosilicate
tubes, and then placed in an isothermal water bath. The
weight increase of the sample with time was recorded, and the
solution concentration, temperature, and presoak time which
provided significant weight increases (5% or higher) were
chosen as the presoak conditions for that particular system.
Presoaking was also used for the solution dip coating of
PDMS onto Dacron*. A chloroform/PDMS solution was used swell
the PET fibers and allow diffusion of silicone oligomers into
the substrate surface.
Immediately following the presoak for surface
modification of PMMA with NVP, the samples were quenched with
ice water to reduce temperature and stop diffusion.
3.2.2.2 Gamma radiation induced polymerization
The samples were immediately placed into the gamma
solution (if gamma solutions were different than presoak
solutions) following the presoak and degassed. The solution
concentrations for gamma polymerization were chosen based on
the final viscosity of the solution. That is, solutions
which could be easily removed following polymerization were
more desirable, and concentrations which caused gellation at
a given gamma dose were avoided. The samples were placed
into the gamma source immediately following degassing. Gamma
irradiation was conducted by simultaneously exposing the
substrate and monomer solution to a 600 Curie 60Co point
source. The samples were placed into a circular, motorized
carousel to provide uniform exposure to all samples.
Schematic diagrams of the gamma source chamber and carousel
are shown in figures 3.3 and 3.4, respectively.
Chamber
\ 60Co
Carousel
Figure 3.3 Schematic diagram of the irradiation chamber
used for polymerization.
Figure 3.4 Motorized carousel used to provide uniform
exposure within the irradiation chamber.
The total radiation dose varied from 0.02 to 0.15 Mrad,
and was controlled by time of exposure. The samples were
placed 4 inches from the point source, which provided a dose
rate of approximately 425 rads/min. The dose rate was
determined by measuring the absorbence of dosimeter film (GAF
Chromic) exposed on a calibrated irradiator at the Shands
Hospital Radiation Oncology Center, Gainesville, Florida
(Goldberg, et al., 1988-1995). Twenty-five films were
exposed from 0.025 to 0.200 Mrads and the absorbence at 540
nm was determined by UV/VIS spectroscopy (Perkin Elmer Model
Lambda b UV-VIS) at 540 nm wavelength. A calibration curve
was generated, and used for exposures on the gamma source
used in this research.
After exposure to gamma radiation, the samples were
removed from the grafting solution and systematically washed
according to the washing procedure discussed in section
3.2.1.4.
3.2.3 Solution Dip Coating of PDMS onto Dacron
Dacron samples were coated with silicone as a surface
modification method. Equal amounts of a 10% of KE-1935 A and
KE-1935 B in chloroform (w/w% solution) were mixed in a
borosilicate glass tube with a screw cap (Kimax tubes, Fisher
Scientific). Dacron fabrics or Bard vascular grafts were
placed into the solution, and the caps were placed on the
tubes to avoid evaporation of the chloroform. The samples
remained in the oligomer solution for 4 hours at room
temperature to allow swelling of the PET and some diffusion
of the silicone oligomers into the substrate. Upon removal
from the solution, the samples were suspended (alligator
electrical clips) in a 60'C oven, in air The samples were
allowed to cure for 24 hours. Then vacuum was applied to the
oven (76 cm Hg), and the samples cured under vacuum at 60'C
for another 24 hours. To remove the uncured, low molecular
weight oligomers, the samples were washed in acetone, with
six solvent changes, for 48 hours, followed by washing in
hexane with three solvent changes for 24 hours. The samples
were then placed into a vacuum oven (60'C, 76 cm Hg) for 12
hours to remove the acetone and hexane.
The adhesion of the silicone layer to the substrate was
also studied (c.f. section 3.2.6). To improve the bonding of
the silicone layer to the PET substrate, a pre-dip coating
step was used. The Dacron was subjected to a presoak in MMA
solutions, followed by gamma irradiation in a chloroform
solution with 10% MAOP-t-PDMS and 10% MMA. This step is
believed to swell MMA into the Dacron. Upon exposure to
gamma radiation, the MMA and MAOP-t-PDMS polymerize into and
on the Dacron surface providing a covalent link from Dacron,
to PMMA, to MAOP-t-PDMS. The second step of solution dip
coating and curing PDMS onto this surface will then provide a
covalent bond between the Dacron and the PDMS.
3.2.4 Characterization
3.2.4.1 Gravimetric analysis
Gravimetric analysis was used as a quick, non-
destructive and inexpensive method to provide information
concerning gravimetric yield for swelling presoakk) or
surface modification. All mass increases and decreases were
recorded on a Denver Instruments A-200DS electronic balance
with a precision of 0.02 mg. Percent weight increase (or
decrease) was determined by the percent change in final and
initial weights, and is defined in equation 3.1,
Percent Weight Change = {(Wf Wi)/Wi} 100 (3.1)
where Wi is the initial weight (usually referring to the
unmodified, clean, dry substrate) and Wf is the final weight
(usually referring to the modified, clean, dry substrate).
In some instances, wet weights were used as initial weights
(e.g., swelling experiments, c.f. section 4.1.1.2).
3.2.4.2 Contact angle goniometry
Contact angle goniometry is also non-destructive and
inexpensive, and provide information concerning the relative
wettability of the surface. This technique is dependent on
the outermost few monolayers of the polymer surface, and may
only be used on solid polymer substrates. Contact angle
values for all substrates were measured on a Rame-Hart
contact angle goniometer (Mountain Lakes, NJ) at room
temperature using the captive air bubble technique unless
otherwise mentioned. Some samples were evaluated using the
water drop in air technique.
The captive bubble technique involves suspending the
sample in water (clipped to the underside of an aluminum
block immersed in water), and injecting air bubbles (ca. 0.2
pl) with a microliter gas chromatography syringe, and
allowing them to come to rest underneath the sample. The
contact angles reported are averages of measurements on six
bubbles per sample. The contact angle measured is related to
the solid-vapor (ysv), solid-liquid (Y1), and liquid-vapor
(Y1v) interfacial free energies. The relationship between
these values is Young's equation, as shown in equation 3.2.
cos (8) = (Ysv Yl,)/y1v (3.2)
Schematics of the captive bubble technique and the angles
measured are shown in figure 3.5.
Typical values obtained with the goniometer used in this
research are 105-110', for hydrophobic substrates such as
PTFE, 50-60' for intermediate substrates such as PMMA, and
s20' for hydrophilic hydrogel substrates (Yahiaoui, 1990).
Aluminum
Block
Sample
*Contact Angle
Air Bubble
Ultrapure
Water
Transparent
Acrylic
Chamber
(b)
Figure 3.5 Schematic representations of the captive bubble
technique and angles measured for contact angle
goniometry showing (a) sample analysis chamber,
and (b) bubble at interface.
3.2.4.3 FT-IR/ATR and transmission microscopy FT-IR
Fourier-transform infra-red (FT-IR) spectroscopy
provides valuable information concerning functional group
chemistry and bonding in polymeric systems. In FT-IR, a
laser beam is passed through the sample, and the energy of
the beam is absorbed by the polymer through various phenomena
such as molecular vibrations, rotations, and stretching.
Percent transmission or absorbence (of the IR beam) is
collected for wavelengths of 2.5 to 20 pm (or wavenumbers of
4000 to 400 cm-1).
Attenuated total reflectance (FT-IR/ATR) utilizes a
crystal through which the beam is passed. There are
conditions under which the infrared radiation passing through
the crystal will be totally internally reflected. The sample
is mounted in contact with the crystal, and the evanescent
wave created within the sample is attenuated in the regions
where the sample absorbs energy. (Spectra Tech reference
Manual, Chapter 13) A schematic of the crystal, beam, sample
configuration is shown in figure 3.6. The depth of
penetration, dp, of the beam into the sample, and therefore
the depth of practical analysis is described by the
relationship shown in equation 3.3,
dp = X(3.3)
2 xni (sin8 n2/n)2 (3.3)
where X is the wavelength of the incident IR radiation, ni is
the refractive index of the ATR crystal, n2 is the refractive
index of the sample, 0 is the incident and exit angle of the
IR beam.
The major advantage to FT-IR/ATR is the ability to
provide chemical structure and bonding information on
polymeric surfaces. This allows for identification of
surface polymers present following surface modification.
FT-IR microscopy utilizes a microscope stage for
mounting the sample where the IR beam may be focused on a
specific sample surface area. The sample may be analyzed by
either transmission, where the beam is passed directly
through the sample, or by a special ATR microscope stage for
collecting ATR spectra. The main advantage to the
transmission microscopy FT-IR system is being able to
evaluate specific sample areas, however it requires thin
samples, typically less than 100 pm. Since transmission
spectroscopy is not surface sensitive as is the case with
ATR, information on the bulk composition of the sample is
collected.
FT-IR data were collected using a Nicolet 20SXB FT-IR
spectrometer using a ImW HeNe laser and a parallelogram KBr
crystal with a 60' entrance/exit face angle. Typically 128
scans at a resolution of 4 cm-1 were signal averaged to obtain
individual spectra. Data processing was done on the Nicolet
software provided with the equipment. Spectra presented here
were scaled and printed using OMNIC 1.2 software (Nicolet
Instrument Corporation).
Figure 3.6 Schematic representation of the IR beam,
crystal, and sample for FT-IR/ATR spectroscopy.
3.2.4.4 X-ray photoelectron soectroscopy
X-ray photoelectron spectroscopy (XPS), also referred to
as electron spectroscopy for chemical analysis (ESCA), is a
fairly new technique for surface analysis. Developed during
the 1950s, XPS is an analytical technique which provides the
atomic chemical composition both quantitatively and
qualitatively, making it a powerful tool in the
characterization of surface modified polymers.
Based on the photoelectric effect, XPS bombards sample
surfaces with X-rays, which cause the ejection of core
electrons. The ejected electron emit discrete energy values
related to the binding energy of the electron and the
exciting radiation energy by equation 3.4 (Barr, 1994, and
Hercules and Hercules, 1976) Each atom has a specific
binding energy, and the atoms to which they are bonded cause
discrete shifts in the binding energies. This enables XPS to
measure atomic concentrations of surfaces as well as provide
information concerning the chemical environment of the
surface elements.
Eb = hu Ek (3.4)
In equation 3.4, Eb is the binding energy, hu is the photon
energy (where h is Plank's constant and v is the X-ray
frequency), Ek is the kinetic energy of the electron
(measured value), and 0 is the work function specific for the
instrument (Barr, 1990).
XPS data were collected using a Kratos model XSAM-800
spectrometer with a Mg Ka X-ray source. The X-ray gun was
operated at 12 kV and 19 mA, and the analysis chamber
pressure was maintained at 10-7 to 10-8 torr during analysis.
Quantification of the spectra was performed using DS800
Kratos software on a Digital computer system. Binding
energies were calibrated using Eb = 285.0 eV as the Cls peak
on all spectra, unless otherwise indicated. The depth of
analysis is in the range of 50A.
3.2.4.5 Dynamic mechanical sampling
Dynamic mechanical sampling (DMS) measurements were made
with a Seiko DMS 200. The sample response to dynamic tensile
forces were recorded as a function of time, temperature, and
frequency (for frequencies of 0.1, 0.5, 1.0, 5.0, and 10.0
Hz) from -140 to 250'C. The sample thermocouple was
calibrated using the maximum value of the dynamic loss
modulus (E") at 1 Hz for the glass transition of PMMA
(Tg=117'C) (Feller, 1993). The elastic response was
calibrated with the tensile modulus of PMMA measured at an
elongation rate of 10%/min in uniaxial tension.
The samples analyzed had a gage length of 20 mm, and had
a range of cross-sectional areas of 4.9 to 6.4 mm2.
Information for the storage modulus (E'), loss modulus (E"),
and tan 6 (E'/E") are presented as a function of temperature
for silicone modified with PMMA, as a function of both
radiation dose and monomer concentration.
3.2.4.6 Light/Optical microscopy
Sample inspection, graft thickness measurements, surface
topography, and other visible observations were made using a
Nikon optical microscope. Optical micrographs were taken
using a canon camera mounted on the same microscope.
The microscope was equipped with a graduated eyepiece
used for size measurements. A hemocytometer with precision
markings was used for calibration of the eyepiece for all
magnifications used, providing a resolution a 1 pm.
Organic and inorganic compounds were used for staining
surface modified regions of substrates to facilitate
visualization. A 10% silver nitrate solution was used for
staining the Hydrograft* materials. The modified samples
were placed in the silver nitrate solution for 12 to 24
hours, and then placed in phosphate buffered formalin (10%
formaldehyde) to precipitate silver oxide and silver
phosphate. A saturated crystal violet-acetone solution was
used to stain the PDMS coating on Dacron" and PMMA modified
PTFE.
3.2.4.7 Scanning electron microscopy
High magnification evaluations of surface appearances
(before and after modification and evaluations) were made
using a Jeol 6400 scanning electron microscope (SEM).
samples were coated with a thin layer of gold palladium using
a Hummer V sputter-coater (Technics, Alexandria, VA), unless
otherwise mentioned. Typical accelerating voltages of 1 to 3
kv were used. These lower voltages were used to avoid sample
ablation typically caused by higher voltages (Goldberg et
al., 1988-1995). Micrographs were obtained at various
magnifications to provide representative records of surface
appearance.
3.2.5 Diffusion/Flow Cell Testing of ICLs
The fenestrated ICL design attempts to provide a route
for the eye to supply nutrients to the stroma anterior to the
implant. A custom diffusion chamber provided by Surgidev was
used to evaluate the permeability of the fenestrated
intracorneal lenses. As displayed in figure 3.7, the
apparatus can be used to determine flow rates of liquids and
solutions through the lens, as well as evaluating the
diffusion osmoticc pressure controlled) of solution
constituents such as NaC1, glucose, and proteins. The
apparatus is constructed of PMMA with a PMMA divider in the
center. The ICL is placed in the right side of the chamber
on an o-ring, and the pressure screw cap is tightened to hold
the lens in place and seal the edges. Flow is in the right
to left direction mimicking flow through the lens from
posterior to anterior sides.
Two types of tests were conducted using this apparatus.
The first test was to determine the water flow through the
lens at a constant pressure. One chamber was filled with
water and an inverted, water-filled flask (with an air bleed
port) was placed in one chamber to provide a constant
hydrostatic pressure. The other chamber was left empty. The
volume of water flowing through the lens was measured after
90 minutes had passed. This test was conducted five times
for each sample, and an average value for each lens condition
was determined.
The next test involved analysis of the permeability of
saline and glucose solutions though the lens as driven by
osmotic pressure and chemical potential. One chamber was
filled with the solution of interest, and the other was
filled with Ultrapure water. The solution concentration of
each chamber was evaluated as a function of time.
Osmolarity was measured using an osmometer (Osmette
Micro-osmometer, Precision Systems) to determine saline
concentration. A linear calibration curve (r2 = 0.999) for
solution concentrations of 0% to 0.9% saline was created
using a series of 10 solutions. Initial saline
concentrations used in the diffusion chamber were 0.9%.
A Glucometer* glucose reader (Ames) and glucose sticks
(Ames) were used to determine glucose concentrations. A
calibration curve (r2 = 0.997) for solution concentrations of
380 mg/dl to 38 mg/dl glucose was created using a series of
10 solutions. Initial glucose concentrations used were 380
mg/dl.
Pressure
Screw Cap
Figure 3.7 Schematic illustration of diffusion and flow
cell used for evaluating permeability of ICL
materials.
3.2.6 Leak and Stability Testina of Vascular Grafts
The silicone dipped Dacron vascular grafts may be
rendered non-porous by the silicone coating on the inside and
outside of the substrate. The leak-rate of water through the
pores was evaluated using a pressurized flow system. The
graft was placed in series with a pressure manometer and
water reservoir, and a back pressure of nitrogen. A
schematic of this test set up is illustrated in figure
3.8(a). The volume of water leaking through the graft
surface at 120 mm Hg in one minute was measured. The
reported values for leaking were normalized for the surface
area of the graft, and the leak rates are reported in
(ml)/(cM2.min).
The stability of the silicone coating on the Dacron
substrate was also evaluated to determine if any material
delaminated or flaked off during dynamic flowing conditions.
A continuous flow of water at 300 ml/min at 110-120 mm Hg
through a vascular graft was maintained for up to 48 hours.
Samples of the water were taken and analyzed with inductively
coupled plasma (ICP) to determine the concentration of
silicon atoms in the solution. The sample volume taken for
analysis was 10 ml, to which 0.1 ml of Triton X surfactant
was added to suspend silicone molecules which may be in the
water sample taken. Since the vascular environment contains
both lipophillic and hydrophilic molecules, silicone coated
Dacron grafts were analyzed in the flow system with both
water and a 10% Triton X water solution. This test apparatus
is shown in figure 3.8 (b).
3.2.6.1 Inductively coupled plasma
ICP utilizes an argon plasma torch to ignite a fine solution
mist. The emission spectrum quantized for each element, and
when the spectra of the ionized gas is compared to reference
solutions the atomic concentration of a specific atomic
species may be determined in parts per million (ppm). The
detection limit of silicon atoms at 251.6 nm by the ICP used
in this research is 0.1 ppm, which corresponds to a
solution concentration of t 0.1 pg/ml. A Plasma 40 ICP
(Perkin-Elmer) was used with a window size of 0.1 nm, a photo
multiplier tube voltage of 700 V, and an integration of 680
msec. Each sample was analyzed 3 times, and means and
standard deviations are reported.
-- Vascular Graft
Plug to Force
Leakage Through
Graft Pores
Collection
Beaker
(a)
Water Beaker
(b)
Figure 3.8 Schematic illustrations of the leakage (a) and
stability (b) analysis set up.
3.2.7 Ex Vivo and In Vive Studies of Vascular Substrates
3.2.7.1 Ex vivo AV canine shunt analysis
Using aseptic technique, an arteriovenous (AV) shunt was
constructed between the carotid artery and the jugular vein
of an adult mongrel canine. Samples to be tested were placed
into a section of Silastic* tubing, and were sealed in place
with silicone RTV. Autologus 111Indium labeled platelets were
injected into the dog, and blood flow over the samples
allowed for 60 minutes. The samples were then removed from
the shunt, and counted in a gamma-counter (Auto-logic, Abbott
Laboratories). The counts are normalized to surface area of
the samples, and are reported as counts/mm2.
3.2.7.2 Canine in vivo evaluation of vascular grafts
Under general anesthesia and with heparin infusion
(100U/kg), a 6 mm diameter, 10 cm long vascular graft was
placed end-to-side on the infra-renal abdominal aorta,
proximally, and on the common iliac artery distally. Two
grafts are implanted in each animal, an unmodified control
graft, and the modified graft of interest, forming a
bilateral aorto-iliac bypass on each side. The implants were
allowed to remain in place for 30 days.
Twenty-four hours prior to graft explanation, 111Indium
labeled platelets were injected. The grafts were removed and
first counted for radioactivity, and then sectioned for
79
histological analysis. Counts of the grafts are reported in
counts/mm2 for the total (inner and outer) luminal surface
area.
CHAPTER 4
RESULTS AND DISCUSSION
4.1 Gamma Radiation Induced Polymerization of Methyl
Methacrylate on PET. PTFE. and PDMS
A review of the literature and current studies within
this laboratory led to the belief that PMMA surfaces on
biomaterials for the vascular environment have the potential
advantages of reducing platelet cell adhesion and reactivity,
and increasing endothelialization and healing of vascular
prostheses (Hufnagel, 1954, and Bowers et al., 1989).
Gamma radiation induced polymerization (GRIP) of methyl
methacrylate has been reported in the literature. To create
a stable surface of PMMA on a polymeric substrate with gamma
polymerization, a mixture of monomer and substrate surface
molecules must be created. The first step in achieving this
mixture for all substrates swelling the substrate with a
monomer solution, allowing the monomer to diffuse into the
substrate surface. GRIP is a heterogeneous system, and Odian
reports the rate of graft polymerization (polymerization of
monomer directly to activated substrate molecules) to be
diffusion controlled (Odian, 1981). The solvent, monomer
concentration, temperature, and time were all evaluated to
some degree to determine conditions providing surface
modification of the substrates with MMA.
4.1.1 Swelling of PET. PTFE. and PDMS in MMA Solutions
The choice of solvents for the swelling of the
substrates with MMA solutions was based on solubility
parameter, compatibility with MMA and PMMA, relative
toxicity, and ease of removal. Table 4.1 lists the
solubility parameters for studied substrates and some typical
solvents. Typically, the more similar the solubility
parameters of the solute and solvent, the better the
solubility of the solute in the solvent (Sperling, 1986).
Solvents used must be solvents for PMMA to avoid
precipitating the homopolymer and surface grafted region
during polymerization. This facilitates solvent and un-
incorporated homopolymer removal following irradiation, and
allows swelling and diffusion within the surface region to
continue during polymerization. Complete removal of toxic
solvents is important for biomaterials applications.
TABLE 4.1 Solubility parameters and H-bonding groups for
selected solvents and polymers
Solvent Solubility H-Bonding Polymer Solubility
Parameter group Parameter
Acetone 9.9 a PMMA 9.0-9.5
Chloroform 9.3 p PET 10.7
Cyclohexane 8.2 p PTFE 6.2
DMSO 12.0 m PDMS 7.3
n-Hexane 7.3
THB 9.1 m
Toluene 8.9 _
MMA 8.8 m
Solubility parameter units are (cal/ca3)1/2. H-Bonding group refers to
the strength of hydrogen bonding by the material, where a strongly,
m moderately, and p poorly bonded. (Brandrup and Immergut 1975).
4.1.1.2 Effect of solvent, concentration, and temperature
Swelling of PET. In an attempt to diffuse MMA into PET, a
solvent selection was initially difficult because of the high
solubility parameter (10.7) and crystallinity of Dacronp PET
fibers. Percent weight increase, or weight uptake, was used
to evaluate swelling (c.f. section 3.2.4.1.). Weights were
determined after blotting the samples with Whatman #3 filter
paper. Mylar* PET films were used initially in the swelling
studies for the ease of handling, table 4.2 shows the weight
uptake of selected solvents by PET at room temperature (25 to
28'C) after 24 and 48 hours. After observing these data, the
two solvent systems chosen for PET were chloroform and DMSO.
Figure 4.1 shows a plot of weight uptake of MMA-chloroform
solutions by PET (Mylar* D-1000) with time for various
solution concentrations. Percent crystallinity changes were
not evaluated in this research, but some solvent induced
crystallization of PET by chloroform is beleived to be
possible.
Table 4.2 Percent weight increases of Mylar* D-1000 films in
selected solvents at room temperature.
Percent Weight Uptake Percent Weight Uptake
Solvent at 24 hours at 48 hours
Acetone 1% 1
Cyclohexane < 1 -
DMSO 2% 2
Ethyl Acetate < 1% < 1%
Sexane 2% 2
THF < 1% 2
Toluene < 1% < 1%
Methyl methacrylate < 1% -
Chloroform 22-25% -
S20
S15 -
2 10
V
0
0 300 60000 0 1200 1500 1800
Swelling Time (minutes)
Figure 4.1 Percent weight increase of Mylar D-1000 PET
films with time in MMA-chloroform solutions as a
function of solution concentration.
Figure 4.2 shows the same data presented in figure 4.1
plotted as Mt/M. versus t1/2/1 based on equation 4.1.
t = ( (4.1)
In equation 4.1, Mt is the mass uptake at time, t, Mm is the
maximum mass uptake (mass uptake at infinite time), 1 is the
|
|
PAGE 1
SURFACE MODIFICATION OF VASCULAR PROSTHESIS AND INTRACORNEAL LENS POLYMERS By CHRISTOPHER WILLIAM WIDENHOUSE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1996
PAGE 2
Copyright 1996 by Christopher William Widenhouse
PAGE 3
This Dissertation is dedicated to all the people who believed I could do it more than I did myself; especially, Tammy, whose love and support I could not do without; and to the memory of Charles Bucaria.
PAGE 4
ACKNOWLEDGMENTS I would like to express my deepest gratitude to my advisor and doctoral committee chair1cLan, Dr. Eugene P. Goldberg, for his guidance, encouragement, and patience. Sincere thanks are also extended for the advice and teaching of the members of my supervisory committee: Dr. Christopher D. Batich, Dr. Anthony B. Brennan, Dr. James s. Seeger, and Dr. Richard Dickinson. I am also grateful to the research group of Dr. James s. Seeger. The guidance and input of Dr. Dinesh o. Shah, an original committee member, is also appreciated. Special appreciation is also felt for the assistance and encouragement from my colleagues during my graduate tenure. These include Jesse Arnold, David Bennett, Dr. Sameer Bhatia, Dr. Emmanuel Biagtan, Charles Bucaria, Scott Butler, Don DePalma, Shannon Eggers, Kirk Foster, Penelope Kao, Dr. Steven Kuo, Ingrid Leidermooy, Dr. Tung Liang Lin, Lili Mateo, Dr. Khalid Mentak, Julie Miller, Tom Miller, Dr. Lynn Peck, Mark Privett, Dr. Jeanne Quigg, Katsabu Rao, Dr. William Toreki, John Wironen, Dr. Ali Yahiaoui, Stacey Zambo, and Mike Zamora. Special thanks are extended to James F. Kirk for his wise and thoughtful input, Drew P. Amery for assistance in 1V
PAGE 5
learning the ropes of analytical techniques, Paul J. Martin for his lively debates and invaluable microscopy assistance, Dr. Anthony B. Brennan for always pointing out the not so obvious and to his openness and fairness to all questions, James s. Marotta for his enthusiasm and encouragement, and to my wife, Tammy for her love, support, pessimistic optimism, and encouragement. V
PAGE 6
TABLE OF CONTENTS Page ACKNOWLEDGMENTS . 1V LIST OF TABLES ll LIST OF FIGURES X.l. ABSTRACT ......................................... XVJ. CHAPTERS 1 I"NTRODUCTI ON 1 1.1 vascular Grafts 1 1.2 Intracorneal Lenses 8 2 BACKGROUND 14 2.1 synthetic Vascular Grafts 14 2.1.1 Vessel Replacement Surgery and Vascular 2.2 2.1.2 2.1.3 2.1.4 2.1.5 2.1.6 Grafts 14 Properties of the Natural Vessel 18 synthetic Vascular Graft Materials and Properties 2 3 Advantages of PMMA, SEMA, and PDMS Surf aces 31 Gamma Radiation Initiated Polymerization 38 Polymer Solution Coatings and Techniques 45 Intracorneal Implants 48 2.2.1 Refractive Corneal Surgery and Intracorneal Implants 48 2.2.2 Intracorneal Lens Materials and 2.2.3 2.2.4 Designs 49 Surface Properties of Ocular Biomaterial Implants 52 Surface Modification Techniques 53 3 MATERIALS AND METHODS 55 3.1 3.2 Materials 55 3.1.1 3.1.2 Substrates 55 Monomers and Reagents 57 Methods 5 7 3.2.1 Sample Preparation and Substrate Cleaning 57 V1
PAGE 7
3.2.2 3.2.3 3.2.4 3.2.5 3.2.6 3.2.7 Surface Modification Methods 60 Solution Dip Coating of PDMS onto Dacron 63 Characterization 65 Diffusion/Flow Cell Testing of ICLs 73 Porosity and Coating Stability Testing of vascular Grafts 75 Ex Vivo and in Vivo Studies of Vascular Substrates 78 4 RESULTS AND DISCUSSION 80 4.1 Gamma Radiation Induced Polymerization of Methyl Methacrylate on PET, PTFE, and PDMS 80 4.1.1 Swelling of PET, PTFE, and PDMS in MM.A Solutions 81 4.1.2 Radiation Grafting of MM.A on PET, PTFE, and PDMS 95 4.2 Gamma Radiation Induced Polymerization of Sulfoethyl Methacrylate on PET 146 4.2.1 Swelling and Gravimetric Analysis of PET Surface Modified with SEMA 146 4.2.2 XPS Analysis of Radiation Grafted SEMA onto PET 14 7 4.3 Solution Dip Coating and Thermal Curing of PDMS Coatings on Dacron PET 148 4.3.1 Conditions for Dip Coating Dacron with PDMS 148 4.3.2 Thermal Curing of PDMS Coatings on Dacron 149 4.3.3 Gamma Polymerization of MAOP-t-PDMS 150 4.3.4 Analysis of PDMS Coated Dacron 151 4.4 Results and Discussion of Fenestrated ICLs with Hydrograft Surface Modifications 167 4.4.1 Determination of Presoaking and Grafting Conditions 167 4.4.2 Analysis of Modified ICLs 174 5 SUMMARY AND CONCLUSIONS 179 5.1 Surface Modification of vascular Graft Substrates 180 5.1.1 Gamma Radiation Induced Polymerization of MMA on vascular Graft Substrates 180 5.1.2 Solution Dip Coating and Thermal Polymerization of PDMS on Dacron, with and without Pre-modification with MAOPt-PDMS 184 5.2 Surface Modification of Fenestrated PMMA Intracorneal Lens Substrates 186 5.2.1 Gamma Radiation Induced Polymerization of NVP on PMMA ICL Substrates 186 V.l..l.
PAGE 8
6 FUTU'RE WORK 189 6.1 6.2 Surface Modification of Vascular Graft Substrates 189 189 6.1.1 6.1.2 6.1.3 6.1.4 PMMA Modified PET PMMA Modified PTFE 190 PMMA Modified PDMS 191 SEMA Modified PET 192 6.1.5 Solution Dip Coatings on PET 192 Surface Modification of Intracorneal Lenses 193 LIST OF REFERENCES 195 BIOGRAPHICAL SKETCH 2 07 Vl..l..l..
PAGE 9
LIST OF TABLES Table page 2.1 Initial and two week post implant dynamic compliance values for canine vascular graft materials 30 3.1 Molecular weight data for PMMA ocular implant materials 56 4.1 Solubility parameters and H-bonding groups for selected solvents and polymers 81 4.2 Percent weight increases of Myla~ D-1000 films in selected solvents at room temperature 82 4.3 Percent weight increase of DMSO-MMA solutions by PET (Mylar D-1000) films and woven Dacron fabrics at 60C. Swelling time is 24 hours for all samples 90 4.5 Percent weight changes of Teflon PTFE irradiated in MMA-acetone and MMA-DMSO solutions as a function of radiation dose and solution concentration 103 4.6 Percent weight changes of GORE-TE~ ePTFE irradiated in MMA-acetone and MMA-DMSO solutions to 0.11 Mrad as a function of monomer concentration 103 4.7 Percent weight changes of GORE-TEX ePTFE irradiated in 100% MMA as a function of radiation dose ............................................... 104 4.8 Contact angle of Teflon PTFE irradiated in MMA acetone and MMA-DMSO solutions as a function of radiation dose and solution concentration 111 4.9 Average carbon, oxygen, and fluorine concentrations for Teflon, GORE-TEX, and PMMA as determined with XPS 115 J.X
PAGE 10
4.10 Weight increase of PDMS irradiated in MMA-DMSO solutions as a function of monomer concentration and radiation dose 127 4.11 Contact angle data for PDMS irradiated in MMA-DMSO solutions as a function of solution concentration (left hand 2 columns) and total radiation dose ( right hand 2 columns) 132 4.12 Percent weight increases of Dacron dip coated with PDMS or gamma radiation irradiated in MAOP-tPDMS as a function of modification conditions 152 4.13 Carbon, oxygen, and silicon atomic concentrations for Dacron modified with dip coatings of PDMS and gamma irradiated in MAOP-t-PDMS as determined with XPS 154 4.14 Silicon concentrations in solutions from the vascular graft delamination analysis as measured by ICP 158 4.15 Pressurized porosity analysis of unmodified and PDMS dip coated Dacron. Flow rate is reported as ml/mincm2 normalized to the sample surface area 159 4.16 Platelet counts from ex vivo AV shunt experiments for unmodified, PDMS dip coated, and MAOP-t-PDMS modified Dacron fabric. Platelet counts are reported as counts/mm 2 sample surface area 169 4.17 Penetration depths of 10 and 20% NVP into fenestrated PMMA disks as determined by optical microscopy 1 71 4.18 Flow rate of water through modified and unmodified ICLs at room temperature and constant pressure (atmospheric pressure plus 1 inch of water). Lens type column gives fenestration hole size and surface percentage of fenestrations 176 X
PAGE 11
LIST OF FIGURES Figure page 1.1 Schematic diagram illustrating the anatomy of the human eye and relavent features 8 1.2 Schematic diagram illustrating a cross section of the cornea with an ICL in place, and diffusion and flow processes of fluids and nutrients 9 2.1 Cumulative patency plotted against time for four types of vascular graft materials 17 2.2 Cut away view of an artery and vein showing the three distinct layers 19 2.3 Typical curve for vessel compliance measurements. Compliance is given as% radial change per millimeter Hg X 1 o2 21 2.4 Possible mechanisms of gamma radiation induced radiolysis and free radical formation for PET, PTFE and PDMS 4 4 2. 5 Reaction mechanism for ther111al curing of Shin-Etsu PDMS 4 7 2.6 Possible gamma radiation reactions and products for P.MMA 54 3.1 Chemical structures of polymer substrates used for surface modification 56 3.2 Chemical structures of monomers used for surface modification 58 3.3 Schematic diagram of the irradiation chamber used for polymerization 62 3.4 Photograph of motorized carousel used to provide unifor111 exposure within the irradiation chamber 62 Xl
PAGE 12
3.5 Schematic representations of the captive bubble technique and angles measured for contact angle goniometry showing (a) sample analysis chamber, and (b) bubble at interface 67 3.6 Schematic representation of the IR beam, crystal, and sample for FT-IR/ATR spectroscopy 69 3.7 Schematic illustration of diffusion and flow cell used for evaluating permeability of ICL materials ........................................... 7 5 3.8 Schematic illustrations of the setups used for leakage and stability analysis for vascular grafts. (a) Leakage analysis setup and (b) flow system for stability analysis 77 4.1 Percent weight increase with time of Mylar D-1000 PET films in MMA-chlorofor11t solutions as a function of solution concentration 83 4. 2 Mt/Moo vs. time 1 1 2 /l for MMA-chlorofor1n solution as a function of solution concentration 84 4.3 Diffusivity of MMA-chlorofornt solutions in PET as a function of solution concentration 85 4.4 Maximum percent weight uptake by PET of MMAchloroform solutions as a function of solution concentration 87 4.5 Solubility parameter for polyisobutene and polystyrene as determined by intrinsic viscosity measurements in a series of solvents 88 4.6 Variation of D (diffusion coefficient) and (Ds)p (polymer-fixed diffusion coefficient of the diluent) with volume fraction of benzene for the natural rubber-benzene system 88 4.7 Percent weight increase with time of Mylar and Dacron in MMA-DMSO solutions at 60C as a function of solution concentration 92 4.8 Cls binding peak of polystyrene showing the rr-rr* shake-up peak 98 4.9 SEM micrograph of Mylar D-1000 film showing the dispersion of silica on the surface 100 4.10 SEM micrograph of Meadox woven Dacron fabric 100 Xll
PAGE 13
4.11 Percent weight increase of GORE-TEx (ePTFE) irradiated in MMA-acetone and MMA-DMSO solutions to 0.11 Mrad as a function of monomer concentration 105 4.12 Percent weight increase of GORE-TEx (ePTFE) irradiated in 100% MMA monomer as a function of radiation dose 107 4.13 Contact angle of Teflon (PTFE) irradiated in MMA acetone and MMA-DMSO solutions to 0.11 Mrad as a function of solution concentration 112 4.14 FT-IR/ATR absorbence spectra of PMMA, GORE-TEX, and GORE-TEX surface modified with MMA 113 4.15 XPS Cls spectra for Teflon, GORE-TEX, and PMMA showing differences in the chemical shifts due to various carbon bonds 115 4.16 Sample calculation to determine the surface concentration of PMMA and PTFE on Teflon and GORE-TEX following modification with MMA 116 4.17 Surface concentration of PMMA on GORE-TEX and Teflon irradiated in MMA-acetone and MMA-DMSO solutions as a function of monomer concentration. Concentrations of PMMA determined with XPS 118 4.18 Surface concentration of PMMA on GORE-TEX and Teflon irradiated in 100% MMA to 0.11 Mrad as a function of radiation dose 119 4.19 XPS Cls spectra for Teflon irradiated in MMA acetone solutions showing changes in the spectra with increasing surface concentrations of PMMA as a function of solution concentration 120 4.20 SEM micrographs of unmodified GORE-TEX 122 4.21 SEM micrographs of GORE-TEx irradiated in 3% MMA-DMSO to 0.11 Mrad 123 4.22 SEM micrographs of GORE-TEX irradiated in 20% MMA-DMSO to 0.11 Mrad. Deformation and stretching of nodule structure is visible 124 4.23 SEM micrographs of GORE-TEX irradiated in 3% MMA-acetone to 0.11 Mrad. Breaking of the nodule structure is typical for samples modified in acetone solutions 125 4.24 Percent weight increase of PDMS irradiated in 10% MMA-DMSO as a function of radiation dose 128 Xl.l.l.
PAGE 14
4.25 Percent weight increase of PDMS irradiated in MMA DMSO to 0.11 Mrad solutions as a function of solution concentration 131 4.26 Contact angle of PDMS following polymerization of 10% MMA-DMSO solutions as a function of radiation dose 133 4.27 Contact angle of PDMS irradiated in MMA-DMSO solutions as a function of monomer concentration polymerized to 0.10 0.13 Mrad 133 4.28 FT-IR transmission spectra for unmodified PDMS and PMMA, and PDMS irradiated in 10% MMA-DMSO to 0.11 Mrad 135 4.29 Surface concentration of PMMA on PDMS as a function of MMA-DMSO solution concentration polymerized to 0.11 Mrad 136 4.30 Surface concentration of PMMA on PDMS irradiated in 10% MMA-DMSO to 0.11 Mrad as a function of radiation dose 137 4.31 1 Hz frequency storage modulus (E') of PDMS modified with MMA-DMSO irradiated to 0.11 Mrad as a function of monomer concentration 139 4.32 1 Hz frequency loss modulus (E") of PDMS modified with MMA-DMSO irradiated to 0.11 Mrad as a function of monomer concentration 140 4.33 1 Hz frequency storage modulus (E') of PDMS irradiated in 10% MMA-DMSO as a function of radiation dose 141 4.34 1 Hz frequency loss modulus (E") for of PDMS irradiated in 10% MMA-DMSO as a function of radiation dose 142 4.35 1 Hz frequency tan~ of PDMS irradiated in MMA DMSO to 0.11 Mrad as a function of monomer concentration 14 3 4.36 1 Hz frequency tan~ of PDMS irradiated in 10% MMA-DMSO as a function of radiation dose 144 4.37 Storage modulus (E') plotted against percent PMMA by weight of PDMS irradiated in MMA-DMSO as a function of monomer concentration and radiation dose 145 Xl.V
PAGE 15
4.38 Cls spectra showing, (a) a comparison between 60C cured PDMS from chloroform solution and high temperature cured PDMS, and (b) a comparison between Dacron and PDMS dip coated Dacron 156 4.39 SEM micrographs of unmodified Dacron fabric ( Meadox) 162 4.40 SEM micrographs of PDMS dip coated Dacron fabric (Meadox) 163 4.41 SEM micrographs of PDMS dip coated reinforced Dacron vascular prosthesis (Bard) 164 4.42 SEM micrographs of unmodified Dacron fabric (Meadox) after ex vivo AV shunt analysis 165 4.43 SEM micrographs of PDMS dip coated Dacron fabric (Meadox) after ex vivo AV shunt analysis 166 4.44 Optical micrographs of a fenestrated ICL after being surface modified with NVP and stained with silver nitrate 170 xv
PAGE 16
Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SURFACE MODIFICATION OF VASCULAR PROSTHESIS AND INTRACORNEAL LENS POLYMERS By Christopher William Widenhouse May, 1996 Chairman: Dr. Eugene P. Goldberg Major Department: Materials Science and Engineering Synthetic vascular replacements are expanded polytetrafluoroethylene (ePTFE) and polyethylene terephthalate (Dacron PET). Poor long term patency of small diameter vascular prostheses is attributed to platelet adhesion and the inability of the vascular endothelium to regenerate. Most attempts to reduce thrombus also reduce endothelial cell adhesion, and attempts to promote endothelial cell proliferation simultaneously facilitate thrombus formation. Surface graft polymerization of polymethyl methacrylate (PMMA) and sulfoethyl methacrylate (SEMA) onto prostheses substrates (PET, PTFE, ePTFE, and PDMS) using gamma radiation induced polymerization (GRIP) in solutions of dimethylsulfoxide (DMSO) and acetone produced stable surface grafts of PMMA and SEMA. This was studied as a method to reduce thrombus formation and XVl.
PAGE 17
encourage healing. Platelet adhesion to PMMA modified PET was not significantly different than unmodified PET. Relationships between polymerization reactions and both radiation dose and monomer solution concentration were examined. PDMS was solution dip coated onto PET and thermally polymerized. Bonding of PDMS coatings onto PET was accomplished by GRIP of methacryloxypropyl terminated PDMS onto PET prior to dip coating. Both processes produced stable PDMS coatings on PET, and inhibited platelet adhesion in ex vivo canine arteriovenous (AV) shunt studies. Surfaces were characterized by gravimetric analysis, contact angle goniometry, Fourier transform infrared spectroscopy (FT-IR/ATR), X-ray photoelectron spectroscopy (XPS), inductively coupled plasma (ICP), pressurized flow analysis, optical microscopy (OM), and scanning electron microscopy (SEM). Blood compatibility was evaluated by ex vivo AV canine shunt experiments. The second part of this research involved intracorneal lenses (ICLs), which are designed to correct myopia, hyperopia, and astigmatism. Current designs prevent nutrient migration through the implant, leading to stromal necrosis and complications. Fenestrated PMMA ICLs were surface modified with an N-vinyl pyrrolidone (NVP) monomer presoak, followed by GRIP providing a lens surface of polyvinylpyrrolidone (PVP). The hydrophilic PVP surface is designed to facilitate nutrient flow through the lens, and to provide a tissue protective layer which is less tissue damaging and cell adhesive. Modified ICLs were analyzed by contact angle goniometry, XPS, OM, and diffusion and flow analysis. XVll
PAGE 18
CHAPTER 1 INTRODUCTION 1.1 vascular Prostheses A vascular prosthesis is a synthetic or natural vessel used to replace damaged or diseased sections of veins and arteries. Vascular disease, complications from surgical interventions, vascular trauma, and disorders which damage the nor1ctal endothelial lining of vessels or impart thrombogenic complications are often treated with synthetic or natural vascular grafts (Brody et al., 1972, Epstein, 1988, and Gertler and Abbott, 1992). Currently, the primary synthetic materials used for vascular grafts are woven or braided Dacron polyethylene terephthalate (PET) and expanded polytetrafluoroethylene grafts, like GORE-TEX. In the u.s., vessel replacements are used over 350,000 times per year (Ratner, 1993). Improvements in the patency of vascular grafts and other blood contact devices, such as arterial venous (AV) shunts (150,000 per year), heart valves (75,000 per year), and pacemakers (130,000 per year), through the use of improved materials, would be a major health care advance (Ratner, 1993). The success of a vascular graft varies with the inner diameter of the vessel replacement. Vascular grafts with an 1
PAGE 19
2 inner diameter greater than 6 mm are considered large diameter grafts, and those with a diameter less than 6 mm (36 mm) are considered small diameter grafts. The majority of complications arise from the use of small diameter grafts to replace coronary and peripheral arteries (Underwood et al., 1988 and Litwak et al., 1987). Although some intimal hyperplasia (thickening of the inner vessel layer, especially at the anastomosis) is observed in the large diameter prostheses, the strong and sometimes turbulent blood flow within these prostheses aids in the prevention of total occlusion, and they remain patent for extended times, often never necessitating repeat surgical correction (Underwood et al., 1988 and Litwak et al., 1987). Small diameter grafts, however, often have up to 50% or more occlusion within the first 24 months of implantation, resulting in the need for secondary surgical repairs, or ultimately causing patient death (Underwood et al., 1988). The ideal surface for blood contact is the normal physiological endothelial lining of vessels, and attempts to mimic this environment through surface modification with polymers, proteins, cultured cells, and a host of other approaches, both biologic and synthetic, have been, and continue to be, explored (Greisler, 1991, Dale, 1978, Gimbrone, 1987, Ratner et al., 1987, and Hoffman, 1984). This research attempted to provide a synthetic polymer surface which would have improved long-term patency and bio acceptance, especially for small diameter vascular graft
PAGE 20
3 implants. Long term success depends on two primary factors, which are often antagonistic. Modifications successful in reducing platelet cell adhesion and thrombus formation also discourage endothelial cell adhesion and growth. It is necessary to inhibit thrombus fortcLation to maintain prosthesis patency, while endothelialization is necessary for long ternt healing. The overall goal of this research is to develop a vascular prosthesis that has reduced platelet adhesion and thrombus formation as well as encourages long ter1ct healing and re-endothelialization. A careful review of vascular prosthesis literature and experimental results within this University of Florida research group suggested consideration of polymethyl methacrylate (PMMA), anionic sulfonic acid containing polymers such as polysulfoethyl methacrylate (SEMA), and polydimethyl siloxane (PDMS) surfaces as less thrombogenic than currently used vascular replacement materials. Surface modification of currently used materials with PMMA, SEMA, or PDMS may provide improved blood compatibility for vascular prostheses. PMMA, SEMA, or PDMS surfaces, also may be further modified with biological molecules that would serve to mimic the performance (both chemistry and function) of the endothelial surface. This second step, although beyond the scope of the research presented here, is currently being pursued within this laboratory. The bulk properties of PMMA, SEMA, and PDMS are unsuitable for vascular prostheses. PMMA is glassy at body
PAGE 21
4 temperature, and is therefore too rigid--although the idea of a graft woven from PMMA fibers is intriguing. SEMA is a hydrogel with poor mechanical properties, and PDMS has a modulus very close to that of the natural vessel, but has poor tear strength. Silica fillers are often used to reinforce PDMS, but the presence of filler decreases hemocompatibility (Lim et al., 1994). All of these materials, however, were found to have favorable surface properties with respect to the vascular blood-contact environment. This makes these materials excellent candidates for surface modification onto currently available vascular graft biomaterials. Surface modification by gamma radiation induced polymerization (GRIP) provides a chemically bound surface layer with desirable physical and chemical properties on a substrate biomaterial without altering the desirable substrate bulk mechanical properties. Substrates were modified in this research with PDMS by thermally curing a two-part oligomer mixture. This enables surface modification to be carried out after substrates have been coated with an uncured PDMS mixture. Surface modification through GRIP of polymers, monomers, and other compounds has been studied and used since the early 1940s (Chapiro, 1962). Radiation-induced polymerization of vinyl functional monomers involves the initiation of free radicals on both the monomer and the substrate polymer. To increase the probability of substrate radicals propagating or ter111inating with graft and interacting monomer/polymer
PAGE 22
5 radicals (providing true graft polymerization), an intimate mixture of monomer and substrate is desired. The process is diffusion controlled, and the use of organic swelling agents that enhance monomer penetration into the substrate is a method of achieving desirable monomer-substrate interaction, and increases the probability of grafting. Another technique is to increase radical initiation in the substrate by increasing the radiation dose(~ 0.5 Mrad), and to include an inhibitor to prevent excessive solution homopolymerization (e.g., Mohr's salts) in the monomer solution. The addition of potentially toxic substances, like inhibitors, is not desirable for production of implantable biomaterials, and high radiation doses may degrade the mechanical properties of many polymeric substrates. Therefore, swelling agents were used to enhance the penetration of the monomer into the substrate surfaces as described by Yahiaoui's "presoak method'' (Yahiaoui, 1990). Although organic solvents were used (i.e. chlorofor1c1, acetone, etc. ) an effort to utilize more bioacceptable solvents (i.e., dimethyl sulfoxide) was made. Simultaneous irradiation of a monomer in solution and a polymer substrate initiates free radicals in both the substrate and monomer creating a grafted, branched, interpenetrating network (IPN) surface region with new physical properties. Swelling, or presoaking, the polymer with a monomer solvent mixture prior to irradiation is helpful to provide a monomer-rich surface region. Swelling
PAGE 23
6 increases the probability of grafting if (i) homopolymerization of the monomer is favored, (ii) initiation on the substrate is difficult, or (iii) interactions between the substrate and monomer are unfavorable for an intimate mixture. If homopolymerization is favored, the result is polymerization outside and away from the substrate polymer chains, resulting in little or no bonding of the new polymer to the substrate. Presoaking in this situation provides monomer within the surface of the substrate, which becomes an IPN or graft polymer when polymerized. If initiation of the substrate is difficult, an IPN may still be created with pre swelling of monomer into the substrate, but little or no bonding of the IPN to the substrate through graft polymerization occurs. Solvents also aid in overcoming repulsive forces between incompatible monomers and substrates, facilitating the intimate mixture required for grafting. For example, a hydrophilic monomer in an organic solvent may wet or swell a hydrophobic substrate better than the monomer itself. Solvents enhance monomer diffusion to the surface region and to the active substrate radicals during polymerization. It was the goal of the first part of this research to investigate the surface modification of currently available vascular prosthesis polymers, namely PET, PTFE, and PDMS, in an attempt to obtain a more hemocompatible material. The two major approaches used for surface modification were (i) simultaneous gamma radiation induced polymerization of
PAGE 24
7 monomer, solvent, and substrate polymer with and without pre swelling at relatively low dose (ca. s 0.15 Mrad), and (ii) solution dip coating of an uncured oligomer mixture followed by thermally curing the surface polymer. PDMS is more readily available as a two-component curing system, and may be polymerized after being formed into almost any desired device shape. For this reason, solutions of the oligomer mixture were used to coat substrates by dipping. Following solvent evaporation, the PDMS was therntally cured for111ing a coating on the polymer substrate. To improve the bonding of the PDMS coating with the substrate, a vinyl acrylic functional silicone polymer (methacryloxypropyl ter111inated PDMS) was radiation grafted onto the substrate prior to dip coating. During the curing process, the silicone components crosslink to each other and to the surface grafted silicone functional molecules, providing covalent bonding of the dip-coated surface to the substrate. The resulting physical, chemical, and mechanical properties of the modified substrates were characterized by a variety of techniques including FT-IR/ATR, XPS, SEM, UV-VIS, optical microscopy, gravimetric analysis, and contact angle goniometry. Biocompatibility was evaluated using a canine ex vivo shunt method (c.f., section 3.2.7.2) and in vivo implants (c.f., section 3.2.7.3), in collaboration with the research group of Dr. James M. Seeger.
PAGE 25
8 1.2 Intracorneal Lenses In the late 1940s, Jose Barraquer (Barraquer, 1949) began experiments on the implantation of synthetic materials within the corneal stroma of the eye to correct numerous irregular curvatures and damage caused to the cornea. Investigations have advanced to clinical studies on a variety of implant designs and materials, the most widely tested being intracorneal or intrastromal rings and lenses (ICLs). Figure 1.1 shows a schematic diagram of the human eye. Figure 1.2 shows a schematic cross-section of the cornea region illustrating the placement of an ICL, and outlines the diffusion and flow of fluids and nutrients in the cornea. Current designs being studied f o r refractive corrections to the cornea include PMMA rings, PMMA lenses, fenestrated PMMA lenses, and polysulfone (PSf) l e nses. Lens Cornea Epithelium--._ .............._J Endothelium Anteri o r_...J Chamber Iris Vitreous Retina Optic Nerve Figure 1.1 Schematic diagram illustrating the anatomy of the human eye and relavent features, modified from Corel Draw c o mputer art program.
PAGE 26
Blood ve,ael d' F Wat, Wat,r sucked driven fn fn by lmblblllon by IOP pr11aur1 9 Evaporation -Metabolic Pump for water removal M --Lactic Gluco Acid Figure 1.2 Schematic diagram illustrating a cross section of the cornea with an ICL in place, and diffusion and flow processes of fluids and nutrients. (Taken from Mccarey, 1990). Hydrogel lens implants and PMMA ring implants do not offer significant refractive correction, and therefore serve to correct vision through manipulation of the corneal curvature (Climenhaga et al., 1988). The major problem with these implants occurs following implantation. The degree of correction required is calculated prior to choosing a specific implant, and following the surgery, the scaring and healing of the cornea is unpredictable, often causing over or under correction of the curvature (McDonald et al., 1993). When solid lens implants are used, this problem is avoided by providing refractive correction to the cornea instead of relying on curvature corrections, and corneal healing has a less dramatic effect on the overall success of the implant. The refractive index of PMMA (1.49) and PSf (1.63) make the polymers excellent candidates for lens materials, and
PAGE 27
10 these lenses have had some success in recent research studies (Mccarey, 1990). The widespread use of PMMA as an intraocular lens (IOL) material also makes it a material of choice. The refractive index and curvature of the lens design determine the refractive power, or diopter, of the lens. PSf lenses have the advantage of being thinner and, therefore, somewhat more flexible than PMMA lenses, and because of the higher refractive index, a thin PSf lens may have the same diopter as a thicker PMMA lens. PMMA lenses have the advantage of being studied more thoroughly than PSf, and the compatibility of PMMA in the ocular environment is exceptional (Amon and Menapace, 1990). Both PMMA and PSf lenses, however, block the flow of vital nutrients to stroma anterior to the implant (Climenhaga et al., 1988). Fenestrated lenses with holes of various dimensions, however, may allow flow and diffusion of saline, oxygen, glucose, proteins, and other metabolites to the stromal tissues surrounding the implant. As demonstrated previously in our laboratories by Osborn (1985), Hoffmeister (1988), Goldberg et al. (1988 and 1989), Yahiaoui (1990), Mentak (1993), and Lin (1995), surface modification of PMMA with polyvinylpyrrolidone (PVP) provides a hydrophilic surface with many distinct advantages. The advantages of PVP modified PMMA (Hydrograft) for IOL applications include reduced corneal endothelial cell damage, reduced iris abrasion, and a reduced adhesion and proliferation of lens epithelial cells within the ocular
PAGE 28
11 environment (Yahiaoui, 1990, Goldberg et al., 1988 and 1989). A hydrophilic surface on a fenestrated lens should also improve surface wetting and diffusion of nutrients to the tissues surrounding the implant, particularly to the region anterior to the lens. This research investigated the surface modification of fenestrated PMMA intracorneal lenses, and the effects on the permeablity of the ICL to vital stromal nutrients. The previously mentioned benefits of Hydrograft modifications with respect to tissue and cell damage were not investigated in detail in view of prior research. Complete descriptions and analysis of the tissue and cell compatibility of the PMMA Hydrograft are provided in Yahiaoui (1990), Goldberg et al. 1988 and 1989, Mentak (1993), and Lin (1995). The cornea receives nutrients from the aqueous humor and the limbal blood supply (Mccarey and Schmidt, 1990), and the driving force for the non-turbulent diffusion and flow through to the anterior corneal tissues of the eye is primarily the chemical potential of the solutions present. The flow of water, saline, and glucose solutions through fenestrated PMMA ICLs was studied by the use of a diffusion flow chamber. unmodified PMMA and Hydrograft (GRIP modified PMMA with PVP) modified PMMA ICLs were compared to each other and to hydrogel lens polymers. Process conditions suitable for GRIP modification of the fenestrated ICLs also required investigation.
PAGE 29
12 A diffusion chamber was used to evaluate the flow of water, saline, and glucose through fenestrated PMMA ICLs having different hole dimensions (10 to 80 m) and surface area coverage. The solution of interest was placed in one side, and allowed to flow or diffuse through the fenestrations of the implant. Differences between the unmodified ICLs, Hydrograft ICLs, and hydrogel polymers were observed and recorded. Although studied in depth previously within our laboratories, various GRIP modification parameters were investigated to provide successful Hydrograft modifications of the fenestrated PMMA ICLs. The molecular weight of the PMMA used for these lenses is significantly lower than materials used previously (80,000 Mw vs. 2-4 million Mw) for IOLs. Because of this, monomer presoak diffusion times and temperatures needed to be investigated to determine conditions which would afford Hydrograft surfaces without distorting or crazing the low molecular weight PMMA. The diffusion of NVP solutions into the low molecular weight PMMA as a function of time, temperature, and NVP solution concentration was first investigated to optimize the presoaking parameters. Once suitable conditions for Hydrograft modification of the PMMA ICLs were determined, wettability, graft penetration depth, and saline and glucose per1neability were studied. Finally, hydrogel lenses were also tested to compare the permeability of fenestrated ICLs to glucose and saline. The properties of the modified lenses
PAGE 30
13 were determined using contact angle goniometry, optical microscopy, and the diffusion apparatus previously mentioned.
PAGE 31
CHAPTER 2 BACKGROUND 2.1 Synthetic Vascular Grafts 2.1.1 vessel Replacement Surgery and Vascular Grafts Atherosclerosis is the progressive deposition of plaque in the arteries with resulting clogging and blood flow restriction. It can lead to heart disease and stroke, and is responsible for over 50% of the deaths in the us (Fox, 1987) Reduced blood flow through the arteries ultimately results in ischemic heart disease. Reduced flow through the coronary arteries is a critical condition requiring immediate attention to reduce the permanent damage of heart muscle by reduced oxygen supply. This occlusion may also occur in other areas of the body as well, and threaten per1rtanent damage to vital regions through reduced blood flow. The severely affected patient often requires treatment of coronary and peripheral arteries. Occluded arteries are generally treated surgically today in one of two fashions, either with balloon angioplasty or with bypass surgery utilizing a vascular prosthesis. Balloon angioplasty involves insertion of a catheter with a balloon which is inflated in the occluded region. Inflation of the balloon 14
PAGE 32
15 re-opens the artery, often leaving the thrombus and plaque free to circulate as emboli, which may clog capillaries and can cause stroke. 2.1.1.1 Natural Cardiovascular Prostheses Natural vascular prostheses are classified based on the origin of the replacement vessel. Autografts are transplants within the same individual, homografts or allografts are transplants between different individuals of the same species (typically a donor organ or vessel), and xenografts or heterografts are transplants from different species (the most common of which is the use of porcine heart valves in humans). Autografts and homografts are the most commonly used natural prostheses, and include those taken from the saphenous vein, the umbilical vein, and the mammary vein. Veins are generally used instead of arteries because the body has an ability to re-route blood flow through veins more easily than through arteries without causing permanent damage to vital regions of the body. Veins also have a slightly different surface chemistry than arteries which render them less thrombogenic than arteries. The valves inside the veins typically are stripped away to eliminate ''dead spots'' in the flow caused by the valves, which can lead to thrombus formation. 2.1.1.2 Synthetic Vascular Grafts The primary synthetic vascular replacements are expanded polytetrafluoroethylene (ePTFE), like GORE-TEx, and woven or
PAGE 33
16 braided Dacron polyethylene terephthalate (PET) fabrics (Greisler, 1991). The largest supplier of PET fibers in the world is DuPont, supplying Dacron, which was initially developed for the textile industry. The standards regulating the production of Dacron fibers is still largely controlled by textile industry demands. Dacron fibers contain titanium dioxide, which gives the fibers a bright white appearance demanded by the textile industry, and fibers used for vascular prostheses, therefore, contain titania as well. The fibers are woven into a number of different patterns by different manufacturers. The Dacron used in this research was either a non-velour woven Dacron fabric from Meadox or a reinforced velour woven knit Dacron vascular graft from Bard. GORE-TE~ ePTFE is manufactured by a high temperature, high speed extrusion process which forms the polymer into a foam like structure. GORE-TEX is readily available with different pore sizes depending on the properties required for the specific application. The majority of small diameter ePTFE (GORE-TE~) grafts and woven Dacron grafts occlude the vessel by more than 50% within a 3-year period (Whittemore et al., 1981). A study by Pevec et al. (1992) shows that all small diameter grafts occlude within 5 to 10 years. Figure 2.1 shows a re tabulation of data presented by Abbott (1987) comparing the success of natural and synthetic small diameter graft materials.
PAGE 34
Patency (%) 100 95 90 85 80 75 70 65 60 55 17 EB 0 il Saphenous Ve i n Umbilical Vein PfFE Dacron 50-+---r--r~---,.--,-,--,--r--,-~-,--, 0 2 4 6 8 10 12 14 16 18 20 22 24 Months Figure 2.1 Cumulative patency as a function of time for four small diameter vascular graft materials. Data from Abbott, 1987, and re-plotted for presentation here. Surface properties are the primary factors controlling the acceptance of biomaterials, especially the blood-material interactions of vascular grafts (Andrade et al., 1991, Hoffman, 1987, and Mustard et al., 1987). Baier (1969) showed that the immediate response of the biological environment to a foreign material involves protein adsorption within the first few seconds. The initial proteins adsorbed by the surface and resulting conformational changes affect the sequence of events that follow. The surface activity of adsorbed proteins determines which peptide sequences,
PAGE 35
18 clotting factor proteins, and hence, which cells adhere and attach upon reaching the surface (Ratner, 1993). Control of these initial events is critical in the development of a surface which does not activate the complex sequence of thrombogenic events (Miyauchi and Shionoya, 1988 and De Mol van Otterloo et al., 1992). The physical and chemical characteristics of the surface deter11tine how proteins are initially adsorbed and which factors control the interactions between other blood proteins and circulating biological molecules. Clotting and intimal thickening, or hyperplasia, is often attributed not only to surface thrombogenicity of the prosthesis, but also to the increased stresses resulting from the mismatch in the radial modulus between the synthetic replacement and the natural vessel at the anastomosis (Brothers et al., 1990). (Most physicians label the radial elastic properties of vessels as ''compliance.'' This is not to be confused with the engineering definition of compliance which is the inverse modulus.) Changes in the modulus of the prosthesis may alter the flow pattern of blood within the vessel, and may sometimes cause turbulent flow which can promote thrombosis (Hanson and Harker, 1987). 2.1.2 Properties of the Natural Vessel 2.1.2.1 Physical and mechanical structure of the vessel The natural blood vessel is a composite structure comprised of three layers. The outermost layer is the tunica
PAGE 36
19 adventitia or externa, the middle layer is the tunica media, and the inner layer is the tunica intima (Figure 2.2). Artery Vein .----T unica inttma ---Ii Valve q I , !, h Tun,ca ex ter na t i: i l : i I ' \ I ' I I f j ' I t I I .. ~ I Serosa I ' I ) l I I ; ' ) Figure 2.2. Cut away view of an artery and vein showing the three distinct layers. (Taken from Fox, 1987). These three layers are composed of an interwoven network of collagen, elastin, smooth muscle, and other proteins and cells, with an innern1ost layer of endothelial cells (endothelium). Collagen, elastin, and smooth muscle provide the vessel with necessary mechanical properties, and the inner endothelial cell lining provides the surface chemistry and function of a non-thrombogenic surface. The highly ordered and abundant fibers of collagen in the adventitia bear the major part of vessel stresses. This provides the high modulus and tensile strength of the vessel. Collagen has a modulus of elasticity of 0.1 to 2.9 X 10 9 Pa (145,000
PAGE 37
20 to 420,000 psi) (Abbott and Cambria, 1982). The high elasticity of blood vessels are due primarily to the elastin, which has a modulus of 3 to 6 X 10 5 Pa (45 to 90 psi), a tensile strength of 0.36 to 4.4 X 10 6 Pa (SO to 650 psi) and an elastic strain of over 300% (Abbott and Cambria, 1982 and Abbott, 1987). The overall properties of the vessel vary with body position and distance from the heart. The higher pulse wave regions closer to the heart damp out the energy with a higher elasticity and lower modulus. Lower pulse regions do not require as much damping, and have a lower elastin composition, less elasticity, and higher modulus. Human arteries have a modulus of about 1 X 10 s Pa (15 psi) longitudinally and 1 X 10 6 Pa (150 psi), circumferentially, with variations according to location (Nichols and O'Rourke, 1986). Veins typically have a higher modulus than arteries because the mean pressure within a vein is 2 mm Hg and 100 nu11 Hg within an artery (Fox, 1987). The vessel components are assembled in a complex composite structure that yields anisotropic mechanical properties. Most evaluations of vascular mechanical properties involve measurements of compliance rather than tensile strength or modulus. Compliance is a measure of the dynamic circumferential elastic properties as determined by the relationship shown in equation 2.1, C = { D systolic D diastolic ) { D diastolic ) x { P systolic P diastolic ) (2.1)
PAGE 38
21 where Ds and Dd are the vessel diameters, and Ps and Pd are the pressures (Merger111an et al., 1986). This is not to be confused with the engineering definition of compliance, which is the inverse of the modulus. Vascular compliance is a dynamic property, and changes with pressure. A typical compliance-pressure curve is shown in figure 2.3 for a canine femoral artery (Mergerman et al., 1986). Matching the mechanical properties of natural vessels and artificial grafts is important. A mismatch of properties was reported to be thrombogenic by Baird et al., (1977), Abbott et al., (1987), Hasson et al., (1985), and Kelly et al., (1992). The mismatch of properties arises from the initial use of stiff materials and changes in modulus resulting from intimal hyperplasia, both of which may lead to turbulent flow and coagulation (Hanson and Harker, 1987). A graft material replacing the natural vessel must have favorable mechanical properties, and not induce thrombus and intimal hyperplasia to remain compliant during use. JS Je 25 \ w ' u ' z 2e 0....
PAGE 39
22 2.1.2.2 Surface properties and functionality of the vessel The surface characteristics (chemical, physical, and functional) of the endothelial cell lining make it the ideal blood contact surface. The dynamic and active roles of the endothelium and sub-endothelium have only recently been realized (Gimbrone, 1987). The favorable surface properties of the endothelium are related to both the production of proteins and the availability of specific binding sites for anticoagulant agents and factors. The complex homeostasis within the vascular environment is an ongoing balance between forntation and lysis of fibrin, the insoluble polymer formed to repair or close off damaged areas within the arteries. Exposure of the sub-endothelium, damage to the endothelium, or simple contact of blood with a non-endothelial surface is sufficient to initiate platelet activation and the coagulation cascade (Gimbrone, 1987). Activation causes the release of mitogens from platelets, endothelial cells, and monocytes stimulating a host of responses including smooth muscle cell proliferation giving rise to thrombus formation and intimal hyperplasia in an attempt to repair or close off the damaged region (Fox, 1987). Following the formation of fibrin, the endothelium produces plasminogen activators which cause the conversion of plasminogen to plasmin, causing fibrinolysis. Production of plasminogen activator inhibitors allows control over the extent of lysis (Mustard et al., 1987). For example, specific surface binding sites are provided for glycosaminoglycans which associate with
PAGE 40
23 antithrombin III, which in turn inhibits thrombin, a protease which converts fibrinogen into fibrin, as well as factor xa, and possibly factors IXa, XIa, and XIIa (Mustard et al., 1987). Specific binding sites are also provided on the endothelium for thrombomodulin, which readily binds thrombin (Mustard et al., 1987) The endothelium of a nor1c1al, heal thy individual can typically repair damaged segments of 1 to 2 cm in length (Greisler, 1991). Neo-endothelialization (re growth of the natural endothelium) of a vascular prosthesis which allows the retention of normal specific activities and functions of the endothelium would be advantageous to the success of vascular grafts. 2.1.3 Synthetic Vascular Prosthesis Materials and Properties The primary synthetic vascular graft materials are PET and expanded PTFE. PDMS has also been studied. The surface and bulk properties of these materials contribute to the success or failure of these materials as vascular prostheses. The improvement of the surface properties of these substrate materials through surface modification to make a prosthesis that is nonthrombogenic and which is capable of healing was the focus of this research. 2.1.3.1 Properties and modification of PET PET has been used as a vascular graft material clinically for over 40 years, and this is the primary reason it was studied here for surface modification (Brothers et al., 1990 and Hufnagel, 1955). PET is used for vascular
PAGE 41
24 grafts in the form of woven and knitted fibers. Because of the higher porosity of some weaves, some grafts require pre clotting prior to use. Grafts with porosity greater than 300 to 400 ml/cm 2 min at 120 mm Hg require pre-clotting to prevent leakage (Greisler, 1991). Typical mechanical properties of PET depend on the degree of crystallinity and orientation. PET vascular grafts are composed of woven fibers which are highly ordered and contain approximately 35 to 60% crystallinity (Rodriguez, 1982). PET fibers have a glass transition temperature, Tg, of 80C and a melt temperature, Tm, of 245 to 265c. Typical mechanical properties of unoriented PET are 4.8 to 6.9 x 101 Pa (7,000 to 10,000 psi) tensile strength, 4.1 to 4.8 X 109 Pa (400,000 to 600,000 psi) tensile modulus, and a strain to failure of 30 to 300% (Rodriguez, 1982 and Roslink, 1990), although tensile strength values for oriented fibers can exceed 100,000 psi. (Unlike the value reported earlier for elastin, 300% elongation of PET is non-elastic.) Compared to the natural vessel, PET has exceptional mechanical strength, and in the form of a woven fabric material, has flexibility and kink resistance. Radial and longitudinal modulus, however, remain much greater than the natural vessel. The need for a non-thrombogenic surface is therefore important to reduce possible intjmal thickening caused by mechanical mismatch. PET is a polyester, and polyesters are subject to hydrolysis of the ester linkage. The rigid nature of the
PAGE 42
25 backbone chain and high crystallinity make PET fibers more resistant to hydrolysis than many other polyesters, and PET is reported to be relatively stable in the biological environment compared to other polyesters (Rodriguez, 1982). Successful surface modification of PET by gamma radiation polymerization has been reported (Stannett, 1981, Rebenfeld and Weigmann, 1978, Kale and Lokhande, 1975, and Nair et al., 1988). The use of PMMA and similar monomers such as acrylic acid, di-methyl acrylamide, hydroxyethyl methacrylate (HEMA), and other vinyl monomers have reportedly been grafted onto PET. Polyester fiber stability during gamma radiation has been reviewed by Nair et. al (1988). Doses of 2.5 Mrad increased the crystallinity from 40 to 44% and increased the breaking load of the fibers by 2.3%. The changes, although minor, are attributed to cross-linking and some degradation. The low doses used in this research (ca. s 0.15 Mrad), however, are not expected to significantly alter the mechanical properties of PET. PET has been reported to be an activator of the complement system, and the addition of a barrier layer such as a PMMA surface graft or PDMS coating may inhibit this reaction (Shoenfeld et al., 1988, and Miyauchi and Shionoya, 1988). Figure 3.1 in section 3.2.1.1 shows the chemical structure of PET as well as the other substrate polymers used in this research.
PAGE 43
26 2.1.3.2 Properties and modification of ePTFE Hufnagel (1955) reports the use of expanded PTFE as a vascular prosthesis as early as the late 1940s. PTFE is also used in catheters, bone joint prostheses, and soft tissue implants, and has been studied as a paste material to replace endoscopic balloon inserts and other devices (Ratner, 1993, Bonomini et al., 1969, and Atala et al., 1992). PTFE is a hydrophobic, very low surface energy fluoropolymer (19 dynes/cm) which appears less thrombogenic than PET. PTFE is generally used for vascular grafts in the expanded for11t as ePTFE (GORE-TEX), and implanted grafts of expanded PTFE and Dacron invoke a similar biological response (c.f., figure 1.1). Although porous, the high surface energy and hydrophobic surface of ePTFE grafts prevent leakage during implantation, and ePTFE grafts do not require pre-clotting. The porous surface of the material is reported to be advantageous, allowing greater flexibility with tissue and cellular in-growth, and the adhesion and formation of a neointimal layer on the graft surface (Sprugel et al., 1987). However, the inability of certain materials to adhere to PTFE may prevent the necessary interactions between circulating factors and the substrate, hindering the re endothelialization and healing process. The major manufacturers of ePTFE vascular grafts are Gore and Impra. The process to form expanded PTFE first patented by w. L. Gore & Associates to make GORE-TEX entails extrusion of the material at high temperature (ca. 390C) and high strain
PAGE 44
27 rate (ca. 40,000 %/sec), producing a m.icrostructure of interconnecting nodes of 0.5 to 400 m and fibrils of 5 to 1,000 m and a porosity of approximately 85 to 95% by volume (Gore, 1970 and 1975). A high 'Im of 310 to 330C allows the expansion process to take place without significant degradation of the PTFE, providing the material with higher flexibility. PTFE has a tensile strength of 1.4 to 3.5 X 107 Pa (2,000 to 5,000 psi), a modulus of 4.0 to 5.5 X 10 8 Pa (58,000 to 80,000 psi), and an elongation of 200 to 400% (Rodriguez, 1982 and Roslink, 1990). Surface modification attempts on ePTFE have often utilized radio frequency plasma polymerization or oxidation of the surface as an initial step, but the use of gamma radiation alone has been shown to be effective. Pre-swelling the surface with a monomer-solvent mix reportedly yielded successful gamma induced surface modification with vinylpyrrolidone in pyridine (Sayed et al., 1981). Razzak et al. (1987) reported successful radiation grafting of N,N dimethylacrylamide (DMAA) onto PTFE using ethyl acetate and acetone as solvents. Pre-irradiation of the substrate to induce a high concentration of free radicals followed by addition of monomer has also been used. However, the higher doses required for this pre-irradiation step often degrade PTFE. This research utilized very low doses and simultaneous irradiation of monomer and substrate in an attempt to surface modify PTFE without significant degradation.
PAGE 45
28 Expanded PTFE has mechanical properties similar to the natural vessel with respect to modulus and elasticity. A surface graft to improve the biocompatibility is not expected to significantly alter this advantage of the material, but to provide a more compatible surface through the altering of the surface chemistry. 2.1.3.3 Properties and modification of PDMS Of the substrate materials studied, PDMS is the only elastomer, and therefore has the potential for a close match to the modulus of the natural vessel. Silicones are used as soft tissue implants, IOLs, skin replacements, burn and wound dressings, catheters, mammary implant components, and for other devices (Ratner, 1993, Taylor, 1985, Meaburn et al., 1978, Tsai et al., 1991, Christ et al., 1989, and Grabow, 1991). PDMS has not been used alone clinically as a vascular graft material, but has been studied for the application as early as 1955 (Hufnagel, 1955). PDMS has also been used by the National Institutes of Health/Heart, Lung, and Blood Institute (NIH/HLBI) as a low-thrombogenicity reference polymer. PDMS is typically used in the for1n of a crosslinked elastomer for applications demanding mechanical strength. (Lower molecular weight and lower cross-link density oils and gels find other applications.) The tensile and tear strength of silicone is somewhat lower than other vascular graft materials, and reinforcement is needed. Reinforcement is achieved with either silica filler particles dispersed within
PAGE 46
29 the silicone, or continuous fiber reinforcement. Typical mechanical properties of silica filled and unfilled PDMS elastomers fall in the range of 2.4 to 6.9 X 106 Pa (100 to 1200 psi) tensile strength, 1 X 10 5 to 1 X 10 7 Pa(lS to 1450 psi) modulus, and 20 to 700% elongation. The high elasticity and low Tg (ca. -123 C) of silicone elastomers afford modulus values similar to the natural vessel. Table 2.1 shows compliance measurements made by Abbott and Cambria (1982) for vascular graft materials as they compare to the natural vessel before and two weeks after implantation. However, one major concern for PDMS has been the long-ter11t changes which may accompany the known affinity of PDMS for lipids, which results in a more brittle material. Silicone elastomers used as IOLs and mammary prosthesis shells could benefit from surface modification as well. The most successful material used for IOL applications to date has been PMMA (Joo and Kim, 1992, and Balyeat et al., 1989). The new small incision techniques developed to insert foldable IOLs with less invasive surgery has utilized silicone as the implant material (Christ, 1989 and Grabow, 1991). If PMMA surface properties were to be applied to silicone elastomers without increased rigidity, the success of the PMMA implants might be achieved with modified PDMS using less invasive techniques. The recent criticism of silicone gel mammary implants has created a frenzy of research for alternative materials. Although the major problems are gel bleed of the silicone gel
PAGE 47
30 and low molecular weight oligomers through the elastomer shell and rupture of the shells due to gel swelling and possible biodegradation, some research shows adverse tissue reactions, such as fibrous capsule for1ctation and hardening of the implant, to be problematic as well (Habal et al., 1991). A PMMA surface graft may reduce the adverse tissue response based on the more favorable biocompatibility of PMMA. Because of the difference in solubility parameters between PMMA ( 9 .4 (cal/cm 3 ) 1 1 2) and silicone ( 7. 3 (cal/cm3) 112) reduction of gel bleed in PMMA modified mammary prostheses is expected (Rodriguez 1982). Table 2.1 Initial and 2 week post implant dynamic compliance values for canine vascular graft materials. Compliance is given as percent radial change per mm Hg (X l02 ) Data taken from Abbott and Cambria, 1982. Graft Implant External Initial 2 Week Material Diameter Cornoliance Comoliance Normal Femoral Artery 4.69 0.007 5.86 0.26 Femoral Arterial Graft 4.53 + 0.50 4.41 + 0.80 5.67 + 1.1 4 nun ID Dacron~ 6.41 + 0.16 1.46 0.12 1.19 0.1 PTFE 5.08 0.05 5.08 0.05 1.3 0.4 Silastic Rubber 6.06 + 0.50 5.95 0.09 6.2 + 0.7 Polvurethane 5.27 0.15 6.10 + 1.10 2.3 + 0.4 Successful surface modification of PDMS elastomers using gamma radiation with hydroxyethylmethacrylate, (HEMA),
PAGE 48
31 ethylene glycol dimethacrylate, and copolymers has been reported by Meaburn et al. (1978). Hoffman et al. (1983) also report grafting of HEMA and ethyl methacrylate polymers and copolymers onto silicones. Lin (1995) has successfully modified PDMS elastomers, Shin-Etsu silicone in particular, with PVP. 2.1.4. Advantages of PMMA, SEMA, and PDMS Surfaces 2.1.4.1 Advantages of a PMMA surface PMMA and other acrylic polymers have been used as biomaterials with much success. Acrylics are currently used for intraocular lens (IOL) implants (Amon and Menapace, 1990), hemodialysis membranes (Falkenhagen and Brown, 1991), dental resins, and bone cements, and are studied for blood compatibility for general biomaterial applications (Apple et al., 1984, Feuerstein et al., 1991, Feuerstein et al., 1992, Lentz et al., 1985, Ito et al., 1992, and Sherman et al., 1963). A general summary of the recent literature on PMMA and other acrylics would be to label these materials as relatively bioacceptable for many invasive applications. Apple et al. (1984) provide an excellent review of the studies done by Ridley and others for work with PMMA concerning IOL implants. Acrylic chemistry has also been reviewed because of concerns regarding hydrolysis of the ester-carbonyl bond in the side groups of polyacrylates such as PMMA. Acrylics are surprisingly resistant to degradation, however, and acrylic
PAGE 49
32 elastomers such as ethyl acrylate have been used widely because they are more resistant to oxidation, ultraviolet light damage and hydrolysis than other traditional elastomers (Rodriguez, 1982 and Sperling, 1986). Long alkyl pendant groups provide less stability. Thus, methyl methacrylate is more stable than ethyl, butyl, and isopropyl acrylate, and the stability decreases with increasing pendant group size. The best example of the vascular compatibility of PMMA was presented many years ago at the 1954 annual meeting of the Society for Vascular Surgery. Hufnagel (1955) reported that tubes of methyl methacrylate polymer (Lucite) had been used as thoracic artery replacements in canines for up to 6 years with complete maintenance of patency. Other materials tested in this study included polyethylene, Teflon~, Kel-F, nylon, woven stainless steel mesh, Vitallium, other metals, and silicone rubber. These materials were tested with and without silicone coatings, and no material was as successful with respect to low thrombogenicity as methyl methacrylate (Hufnagel, 1954). The silicone coated prostheses had poor compatibility, which Hufnagel attributed to incomplete removal of acids for1c1ed in the curing reactions. The use of methyl methacrylate as a vascular prostheses has not been reported in the intervening years, probably because the need for flexible materials has eliminated consideration of this rigid polymer. Even though Hufnagel collected data on over 400 implants of rigid vascular replacements, he indicated the importance of using a material
PAGE 50
33 which had mechanical properties similar to the natural vessel. work reported from the late 1950s to date has therefore shifted to softer, more flexible, and more hydrophilic surfaces. The potentially excellent nonthrombogenic surface properties of PMMA, thus, seems to have been lost in the literature. Since the technology now exists to graft polymerize very thin glassy polymer surfaces onto flexible substrates without significant reduction in elasticity, the surface modification of vascular and blood contact devices now seems logical and promising. Current work in our laboratory indicates PMMA has favorable cell adhesion properties with respect to platelet and endothelial cell adhesion (Goldberg et al., 1988-1995). A reduction in the ratio of adhered platelet cells to endothelial cells is sought for blood contact devices such as a vascular grafts. Observations in our laboratories of in vitro platelet and endothelial cell adhesion assays show unmodified PMMA control samples to have a more favorable ratio (greater endothelial cell adhesion than platelet adhesion) than other materials studied, including surface modified PMMA (Goldberg et al., 1988-1995). The affinity of PMMA for endothelial cells becomes more apparent when considering research on PMMA for IOL applications. A common problem of unmodified PMMA IOL materials is excess epithelial cell adhesion and growth on the implant (Goldberg et al., 1988-1995, Yahiaoui, 1990, and Lin, 1995). Assuming that endothelial and epithelial cells exhibit similar surface
PAGE 51
34 adhesion properties on polymers, IOL studies suggest that PMMA would favor endothelial cell adhesion. Finally, in studying a new method to measure the adhesive strength of red blood cells (RBCs) to biomaterials, Bowers et al. (1989) found less adhesion of RBCs to control PMMA samples than to hydrophilic glass, tissue culture grade polystyrene, and PET. 2.1.4.2 SEMA surface advantages Sulfonated surfaces have been investigated in the vascular environment primarily because of the strong anionic surface charge contributed by sulfonic acid functional groups. Many blood components, including red blood cells and platelets, have a slightly negative surface charge. It therefore seems logical, that surfaces with negative charges will repel these components of circulating blood. Our laboratory has been investigating the use of sulfonated monomers for surface modification such as potassium 3sulfopropyl acrylate (KSPA), sodium methacrylate (SMA), styrene sulfonic acid, sodium salt (SSA), and other anionic sulfur containing monomers since the late 1980s (Goldberg et al., 1988-1995 and Yahiaoui, 1990). These monomers are supplied as water soluble sodium or potassium salts. It is often difficult to obtain an intimate mixture of monomer and substrate molecules through presoaking in aqueous media, because of the slightly hydrophobic surface of PET. Pre treatment with other monomers such as N-vinyl pyrrolidone (NVP) is often required to obtain grafting of these monomers. However, some sulfonated monomers, such as 2-sulfoethyl
PAGE 52
35 methacrylate (SEMA), are available as organic, non-salt compounds, and are therefore soluble in organic solvents such as acetone and dimethyl sulfoxide (DMSO). Dissolution in these solvents increases the swelling of substrates during a presoak process, and grafting is more easily achieved. Critical components of the vascular environment include chondroitin sulfate, heparin sulfate, and dermatan sulfate, which are all proteoglycans secreted by the endothelium. Marcum et al. (1986) reported heparin sulfate is capable of binding with antithrombin, and thus imparts antithrombogenic properties to the vascular endothelium. Ofosu et al. (1989) report that increasing the degree of sulfonation on heparin and dermatan sulfate increases the catalytic effects on thrombin inhibition. Other compounds with functional sulfate groups are thought to have favorable reactivity within the vascular environment. Kishida et al. (1991) studied polyethylene films surface modified with cationic, anionic (sulfonated and non-sulfonated), and non-anionic monomers, and found the in vivo cell adhesion to be related to both the charge and the presence of sulfonated groups, with sulfonated polymers having higher HeLa S3 cell attachment and growth than non-ionic and cationic surfaces, indicating a higher affinity for binding, attachment, and growth of cells to anionic surfaces. The incorporation of sulfonated functional polymers on vascular grafts seems promising in reducing thrombus forntation.
PAGE 53
36 2.1.4.3 PDMS surface advantages PDMS is a hydrophobic material with a low surface energy (contact angle ca. 80). The low surface energy prevents adhesion of many compounds from an aqueous environment, as these molecules are often repelled by silicone surfaces. Two key proteins determining the thrombogenicity of a biomaterial are albumin and fibrinogen. Albumin adsorption is preferred for a non-thrombogenic surface, and fibrinogen adsorption usually indicates a thrombogenic surface, as fibrinogen is converted to fibrin by thrombin. Cooper and Fabrizius-Homan (1991) found silicone rubber to have a higher affinity for albumin than fibrinogen in competitive adsorption studies, and when the albumin was preferentially absorbed, the thrombogenicity of the material was reduced. This ratio of albumin to fibrinogen was most favorable for silicone when compared to polyethylene, polyurethane, and Teflon. In a canine ex vivo AV shunt platelet adhesion study by Ip and Sefton (1991), it was found that Silastic (PDMS, Dow Corning) and silica free PDMS (Thoratec) both had significantly lower platelet cell adhesion than polyethylene. Norgren et al. (1990) found silicone coated Dacron to have a reduced thrombogenicity, and Granke et al. (1993) found reduced inflammatory reaction as well. In the study by Granke, only the outer surface of the graft was covered to create a prosthesis which did not require pre-clotting, but there was significantly more tissue ingrowth and endothelialization in the silicone treated samples than in
PAGE 54
37 the control samples. Whalen et al. (1992) evaluated a novel prosthesis made entirely of silicone. The prosthesis was porous to allow tissue ingrowth, and after implantation in canines for 8 weeks, had an overall patency of 86%. Two studies within the research group of Cooper (Lin et al., 1994 and Silver et al. 1995), report modification of silicone elastomer surfaces in an attempt to improve the biocompatibility. Lin et al. (1994) compared silicone to polyurethane-silicone copolymers and found the compatibility of the silicone to be superior with respect to clotting time and platelet deposition. Silver et al. (1995) modified silicone surfaces with alkylsiloxane monolayers of various functionalities following exposure to an oxygen radio frequency plasma. The materials were evaluated in a canine ex vivo arteriovenous (AV) shunt, and the untreated silicone had superior properties with respect to platelet and fibrinogen deposition. In a study by Morel et al., (1989) the endothelial cells cultured on thin silicone sheets were found to have motility and contractility indicating a healthy environment for cell growth and proliferation. Finally, numerous studies using a canine AV shunt use Silastic, or some other silicone elastomer, as the tubing in which samples being analyzed are placed (Lin et al., 1994, Silver et al., 1995, Goldberg et al., 1985-1995). The silicone tubing would not be used in these studies if occlusion occurred. The recent controversy surrounding the use of silicone gel filled breast implants will obviously invoke a negative
PAGE 55
38 response to the idea of using silicone as a vascular graft surface. However, the silicone surfaces discussed and studied in this research are crosslinked elastomers, and the toxicity and poor biological response to silicone materials related to breast implants are for the low molecular weight oils and gels (Kimitoshi et al., 1990 and Kimitoshi et al., 1991). Current literature and research indicate a silicone surface in the vascular environment could have favorable and beneficial responses. This review led to the investigation of PDMS coatings on Dacron vascular prostheses presented here. 2.1.5 Gamma Radiation Initiated Polymerization Gamma-rays are electromagnetic waves of short wavelength (A=< 0.1 nm) which are emitted from a decaying radioactive source. One of the most commonly used sources of gamma-ray energy, or gamma radiation, is the radioactive isotope of cobalt, cobalt-60 ( 60 co). Providing two sharp spectral lines of radiation energy of 1.17 and 1.33 MeV (megaelectron volts), 60co is often chosen because of its ease of preparation (nuclear activation of cobalt-59) and its long half life of 5.3 years (Chapiro, 1962). Gamma radiation energy generates free radicals on vinyl monomers and polymeric substrates, making it an excellent initiation source for free radical polymerization of monomers onto polymeric surfaces (Chapiro, 1962). Gamma radiation
PAGE 56
39 initiation polymerization (GRIP) has been studied since the late 1940s, and is often chosen over other techniques because of its low cost and cleanliness (does not introduce chemical initiator molecules). Low cost and cleanliness are two extremely critical factors in determining the success of many materials for the biomedical industry. Surface modification through GRIP involves the formation of free radicals on monomer molecules leading to free radical polymerization. The simultaneous exposure of the monomer and substrate allows both initiation and cleavage of substrate polymer. The growing homopolymer chains may propagate or terntinate with the available reactive sites created on the substrate, creating a surface region of grafted, crosslinked, and interpenetrating network molecules. Swelling the substrate with monomer in solution (presoak technique) provides a localized monomer-rich region within the substrate surface, and the surface region becomes an intimate mixture of substrate and monomer molecules (Yahiaoui, 1990). Upon exposure to gamma radiation, this entire region becomes a surface "graft," referring to a region of grafted and IPN polymer molecules. The intimate mixture of the swollen region also facilitates the diffusion of the monomer and homopolymer chains to the activated substrate sites, increasing the efficiency of the grafting.
PAGE 57
40 2.1.s.1 Polymer radiolysis and free radical reactions Chapiro (1962) provides an excellent review of early studies of radiation effects on polymers and monomers. Upon exposure to high energy radiation, polymers undergo a series of reactions including, but not limited to, radiolysis of side chain atoms and functional groups, free radical formation, crosslinking, chain scission, and degradation. Although the exact nature of the mechanism of these reactions with systems as complex as polymers is still not fully understood, radiation polymerization following initial radiation events is fairly well documented, and several events may occur upon simultaneous exposure of monomer solutions and polymer substrates. When a polymer or monomer is exposed to gamma radiation, cations, anions, and free radicals are created. The ions are only stable at low temperatures, and usually dissociate to yield radicals (Chapiro, 1962). Graft polymerization may take place when a polymer and monomer in intimate contact are simultaneously exposed to high energy radiation. Initially, a radical may be formed on the substrate polymer or the monomer. When a monomer molecule reacts with one of these two radicals, either a graft polymer or homopolymer forms, respectively. Graft polymerization is favored if the polymer substrate has a higher ability to cross-link after chain scission rather than degrade, and more grafting occurs if
PAGE 58
41 radical yields are higher for the substrate polymer than the monomer (Chapiro, 1962). Radicals are also created on the growing homopolymer chains, giving rise to higher molecular weight homopolymers, branching, or crosslinking. Radical formation on side groups of growing homopolymer chains or substrate chains yields branching, and termination of radicals by combination with branched chains from different molecules yields crosslinking. These polymer reactions, if occurring within the substrate, lead to the formation of an IPN. Substrate main chains may be cleaved as a result of the high energy of the radiation. The cleaved chains then form crosslinks with other cleaved substrate chains; for111 graft polymers or cross-links by the addition of monomer or combination with a growing homopolymer chain; or re-combine with the original site of cleavage (Yamamoto and Yamakawa, 1980). If cleaved chains terminate by disproportionation or chain transfer, the molecular weight of the substrate polymer is reduced, and degradation results. During the course of the polymerization, the polymer within the substrate surface begins to gel as higher molecular weights are reached and as cross-linking and branching occur. This decreases the mobility of larger propagating molecules. The concentration of monomer decreases and the concentration of radicals increases rapidly as higher conversions are reached (Dob6, 1978), leading to higher degrees of cross-linking and branching in the surface
PAGE 59
42 polymer than in the surrounding solution polymer, as the reaction auto accelerates. The efficiency and stability of radical formation on various substrates and monomers determine the ease or difficulty of polymerization of a specific monomer to a specific substrate polymer surface. The radiolysis and radical formation on the various substrates and monomers studied in this research will now be discussed. 2. 1. 5. 2 Radical for11tation on PET Charlesby (1953) found PET to crosslink upon exposure to gamma radiation, whereas Todd (1954) found it to undergo degradation. Low dose exposures always gave an increase in modulus of Dacron fibers when studied by Teszler and Rutherford (Chapiro, 1962). Radical formation on the phenyl ring without atomic ejection is possible, but low yields are expected because of the resonance of the ring (Chapiro, 1962) Cleavage of the ester bond or radical forntation on the main chain with ejection of an H atom are the most likely reactions, as shown in figure 2.4 (Chapiro, 1962). Radicals are formed on both the amorphous and crystalline regions of the polymer, but polymerization is only expected in the amorphous regions because of the reduced diffusion through the crystalline phase.
PAGE 60
43 2 .1 5. 3 Radical f or111ation on PTFE and ePTFE Main chain cleavage is possible in PTFE, but radiation more commonly yields free radicals from cleavage of the C-F bond (Rye, 1988). Charlesby (1952) analyzed gas evolution of PTFE upon exposure to gamma radiation and found the products to be carbon tetrafluoride, indicating evolution of both carbon and fluorine. In the presence of oxygen, doses as low as 1 to 10 Mrad lead to significant degradation and embrittlement of PTFE. When oxygen is excluded from the system, however, much less damage occurs. Possible reaction mechanisms are shown in figure 2.4. 2. 1. 5. 4 Radical f O!ltlation on PDMS Radical for111ation on silicone polymers was studied by Charlesby and Omerad (1963), and the polymers were found to crosslink upon exposure to gamma radiation. The Si-0 bonds are significantly more stable than the Si-C and C-H bonds, leading to radiolysis of the pendant methyl groups. Possible reactions are shown in figure 2.4. 2.1.s.s Radical polymerization of vinyl monomers Methyl methacrylate monomer readily forms radicals when exposed to gamma radiation, and is expected to polymerize well on all of the chosen substrates. The resulting surface polymer is expected to be an interpenetrating network of grafted, cross-linked, and branched polymer.
PAGE 61
0 11 -C-----< 0 11 --C--< 44 0 11 >--< -OH CH 2 -CH2-0-0 0 11 11 >---< -O-GI 2 --GI 2 --0 __ ...... G ___ a __ 1r __ ur __ ta _____ 0 11 --C-----< Radiation I 0 11 >--1. -D-CHrCH2-0-Radiolysis of PET H F --cF...-CF-rF2-Gamma 2 Radiation I --CF 2 C----iCFT-or --CF~ I I CH 3 I -(Si 0)I CH 3 F F Radiolysis of PTFE Rad1.at1.on CH 2 I -(Si 0)or I CH 3 Radiolysis of PDMS -{Si 0)I CH3 Figure 2.4 Possible mechanisms of gamma radiation induced radiolysis and free radical formation for PET, PTFE, and PDMS. 2.1.5.6 Polymerization of vinyl monomers onto substrates Chapiro (1962) discusses the free radical polymerization of vinyl monomers induced by gamma radiation. The mechanism of reaction proceeds via traditional free radical processes,
PAGE 62
45 with the rate of polymerization being proportional to the monomer concentration and the square root of the dose rate. The equation for the kinetic chain length (average number of monomer molecules polymerized per initiating primary radical) follows that of chemical free radical polymerization as well. The kinetic chain length for gamma radiation polymerization is proportional to the square of the monomer concentration and inversely proportional to gamma radiation dose rate. 2.1.6 Polymer Solution Coatings and Techniques When a solid material is removed from a liquid, (solution or pure liquid) the surface attractive forces are typically strong enough to adhere a monolayer or more of the liquid to the solid surface. The relative surface energies of the solid and solution determine the thickness and adhesion of this surface coating. If a solute is dissolved in the solution, and the solvent is rapidly evaporated after the object is removed from solution, there will be a layer of solute on the object. The relative surface energies of the solute, solvent, and object as well as the evaporation rate and solution concentration determine the amount remaining on the surface. The surface energies of the solute and solid determine the strength of the adhesive bond between the surface layer and the object. This general principle of surface phenomena was utilized in the dip coating of PDMS onto Dacron~ vascular prostheses.
PAGE 63
46 Both PET and PDMS are relatively hydrophobic (contact angle of PET ca. 60, and contact angle of PDMS ca. 80), and both materials swell in the organic sol vent chlorofor11t. For this reason, solutions of the two part PDMS oligomers were made in chloroform, and Dacron was soaked in this solution. The swelling of the Dacron by the chlorofor11t is believed to allow the diffusion of PDMS molecules into the PET surface, creating a swollen network of PET, PDMS, and chloroforxn. Upon removal of the PET from the PDMS solution, the chlorofor11t evaporates, de-swelling the surface network, leaving PDMS oligomers on and in the PET surface. The relatively rapid evaporation of chloroform, and the similar surface energies of PDMS and PET allow a layer of PDMS oligomers to remain on the prosthesis after removal from the solution. In an attempt to improve the bonding between the PDMS and Dacron, a pre-modification step was implemented. MAOP t-PDMS was gamma polymerized onto Dacron to provide a link between the coating and the substrate. A covalent link between PMMA-g-PET and the MAOP-t-PDMS is expected by using a presoak of MMA followed by polymerization of MAOP-t-PDMS. The incorporation of MMA into the system will provide a propagating link between the substrate and the MAOP-t-PDMS thus allowing polymerization of more MAOP-t-PDMS. The dip coating of PDMS should then polymerize and crosslink with the MAOP-t-PDMS.
PAGE 64
47 2. 1. 6 .1 Ther1ttal curing of PDMS The curing of Shin-Etsu PDMS proceeds via vinyl addition polymerization, and is initiated by a platinum catalyst. The reaction scheme for this polymerization is shown in figure 2.5. Ol3 I + Si ---cH -rn1 H Si ---cH 3 I I Ol3 0 Ol3 I I 0-13 Si H + Ol2==GI Si--0+ I I Ol3 0 GI3 I + I Si -a-J=GI1 H Si --0-1 3 I Ol3 Pt Catalyst Figure 2.5 Reaction mechanism for thermal curing of Shin Etsu PDMS.
PAGE 65
48 2.2 Intracorneal Implants 2.2.1 Refractive Corneal Surgery and Intracorneal Implants Myopia (nearsightedness caused by excessive corneal curvature), hyperopia (farsightedness caused by reduced corneal curvature), and astigmatism (irregular corneal curvature) are disorders affecting the cornea which cause impair vision. Accidental injuries that abrade, scratch, deform, or scar the cornea also significantly reduce visual acuity. These disorders and damages of the cornea are currently corrected by the use of spectacles (glasses or contact lenses), a corneal transplant for severe conditions, or with some of the newer surgical techniques such as radial keratotomy (RK) or laser keratotomy. Intracorneal lenses or rings provide still another option, and are being studied by numerous researchers and ocular device companies. Glasses and contact lenses will always be an option for correcting many corneal defor1nations because they have the advantage of always being reversible. If a problem occurs or an error was made in fitting the diopter of the corrective lens, the device may simply be removed. Corneal transplants, are a major surgical procedure, and a donor organ is required, restricting their use to severe cases of complete corneal damage or damage which can not be repaired by other means. Radial and laser keratotomy scars into the cornea and relies on healing to alter the shape and modulus of the
PAGE 66
49 cornea, and is an extremely unpredictable procedure (Waring, 1990). The use of RK and other similar procedures offers patients a quick procedure with extreme pain and little guarantee of success, as glasses are often still required. However, it is often difficult to treat severe astigmatism with contact lenses, and greater deformations require large and often cumbersome spectacles, and damage or deformations too great for other means of corrections often requires treatment with a non-reversible therapy (Kerry, 1995). Keratoprostheses such as ICLs offer a more permanent solution to refractive corrections of the cornea. In the late 1940s, Barraquer began experiments on the implantation of synthetic materials within the corneal stroma of the eye to correct irregular curvatures (Barraquer, 1949). The initial designs were lenses constructed of PMMA and polysulfone that were surgically implanted within the corneal stroma to provide both mechanical (corneal reshaping) and refractive (lens power magnification) correction of the cornea. More recent designs include hydrogel lenses, intracorneal rings, and fenestrated PMMA lenses. 2.2.2 Intracorneal Lens Materials and Designs Polysulfone is an optically clear, amorphous, stiff, but flexible polymer with a glass transition temperature (Tg) of 190C and a refractive index of 1.63. Polysulfone ICLs are less rigid than those made with PMMA, and rely more on the high refractive index for correction of the cornea. PMMA is
PAGE 67
50 an optically clear, amorphous, glassy polymer with a Tg of 117 c and a refractive index of 1.49 (Most commercial grades of PMMA contain a small fraction of other acrylic polymer such as polyethylacrylate which reduces the overall Tg to around 105C.). PMMA ICLs provide both refractive corrections and mechanical correction, and PMMA intracorneal rings rely solely on mechanical changes in the corneal shape to provide correction. The initial designs of a refractive lens made of polysulfone and PMMA were proved to be successful at correcting corneal correction, but implant studies have shown them to be unsuitable as long term implants thus far (Kerry, 1995 and Lane et al., 1989). Both polysulfone and PMMA lenses are impermeable to water and aqueous solutions, and the initial design of these ICLs hinders the diffusion of water, ions, proteins and other vital nutrients to the corneal stroma anterior to the implant. The result of these implants was corneal opacification and necrosis in the deprived regions, and lipid deposits posterior to the implants (Climenhaga, 1988). Although the lens designs are still being studied, recent studies have focused on a lens made with hydrogel materials (McDonald et al., 1993). Hydrogel lenses (e.g., Permalens, CooperVision and Lidofilcon A, Allerghan Medical Optics) are polymers and copolymers containing hydrogels such as polyhydroxyethyl methacrylate (pHEMA), polyvinylpyrrolidone (PVP), and polyacrylic acid (pAA). The
PAGE 68
51 hydrogel component in these lenses makes them water permeable. Claims have been made that since the hydrogel lenses are water per1rteable, it is therefore also pernteable to glucose and other metabolites necessary to the corneal stroma (Werblin and Patel, 1992, werblin and Peiffer 1992, Mccarey, 1990, Mccarey, 1981). However, diffusion or per11teability of water does not guarantee the permeability of other molecules, especially those significantly different in size and functionality. Hydrogel lenses also have fairly poor mechanical properties, and tearing or damage to the lens during handling, implantation, or use is possible (Menapace, 1990). The intracorneal ring (ICR) design allows passage of nutrients and offers curvature correction by flattening the central cornea (Kerry, 1995). The ring is surgically implanted by sliding and rotating it into an intrastromal channel created by a radial incision of the stroma lamella (Kerry, 1995). The design of a PMMA ring provides mechanical correction of the cornea without hindering the diffusion of nutrients to the stromal region anterior to the implant. (The absence of a refractive center in the ring requires application of more mechanical ''pressure'' to achieve the same correction of the lens designs. Calculating the required ring shape is often difficult, and may be complicated by unpredictable healing of the cornea following the implant surgery (Quantock et al., 1995).
PAGE 69
52 The Surgidev intracorneal lens (ICL) is a fenestrated PMMA lens designed to correct the corneal refractive power both mechanically (by corneal reshaping) and refractivly (magnification by the ICL). The holes in the ICL are present to allow diffusion of stromal nutrients to the anterior implant region and still provide refractive correction. The primary component of interest which is vital to stromal tissues is glucose. Glucose is delivered to the stroma via the aqueous humor, and its normal concentration is constant parallel to the stroma, and decreases across the stroma from 880 to 580 g/ml (88 to 58 mg/dl), posterior to anterior (Mccarey and Schmidt, 1990). 2.2.3 Surface Properties of Ocular Biomaterial Implants The corneal endothelium plays a critical role in the balance of fluids and nutrients within the cornea itself as well as the eye posterior to the cornea. The implantation and residence of an ICL or ICR in the cornea requires a minimally damaging procedure because the human adult corneal endothelium does not regenerate when damaged (Bourne and Kaufman, 1976). The work in our laboratories has shown that PMMA placed in contact with the corneal endothelium or epithelium will strip away vital cells upon removal (Yahiaoui, 1990, Katz et al., 1977, and Sheets, 1983). Studies by Andrade (1985) and Absolom et al. (1987) show the tendency of hydrophobic surfaces (such as PMMA) to strongly absorb proteins, whereas hydrophilic surfaces tend to readily
PAGE 70
53 desorb proteins following initial adsorption. Likewise, it was determined by our research that PMMA modified with NVP (Hydrograft) provides a surface which will not readily adhere to these cell layers on contact. The Hydrograft surface imparts less damage to surrounding tissues during implantation than would an unmodified PMMA lens by providing long ter1rt lubrication of the tissues by the hydrophilic grafted surface. Also reported by Yahiaoui (1990) and continually observed in current research within this laboratory, radiation induced graft polymerization with NVP (Hydrograft modification) significantly reduces the adhesion, growth, and spreading of ocular epithelial cells on PMMA ocular implant materials. Surface modification of the PMMA ICLs NVP is expected to provide a hydrophilic surface within the ICL fenestrations, and increase the permeablity of the lens to vital stromal nutrients. This research focuses on the determining suitable surface modification conditions for low molecular weight PMMA ICLs and resulting changes in permeability. 2.2.4 Surface Modification Techniques The presoak surface modification techniques discussed previously for vascular graft modifications will be used for PMMA ICL modifications. The monomer solutions, however, are aqueous because of the sensitivity of the ocular environment and the low chemical resistance of PMMA to organic solvents.
PAGE 71
54 2. 2. 4 .1 Radical fortttation on PMMA Figure 2.6 shows several possibilities for the radiolysis of PMMA. The abstraction of a hydrogen from the main chain may create Radical A, which upon disproportionation leads to the formation of Radical Band End Group A (Todd, 1954). Main chain homolysis is also possible, yielding Radical c and D (Kirsher et al. 1965). Degradation may occur upon ester group cleavage, yielding Radical E. Scission of Radical E may then lead to formation of a new Radical D and Chain End B (Ranby and Rabeck, 1975). CH 3 CH 3 I I -C--cH-4 C -CH --cH 2 + I I 2 COOCH 3 C OO C H 3 Radical A Radical B CH 3 CH 3 CH 3 CH 3 I I I I -tC-CH 2 --C--cH 2 G~mm~ C -CH 2 --C + I I Rad1.at1.on I I COOCH 3 COOCH 3 COOCH 3 COOCH 3 PMMA Radical C CH 3 CH 3 CH 3 CH 3 I C=CH-C-1 I COOCH 3 COOCH 3 Unsaturated Endgroup A Radical D I I I --C-CHzG-CH,:---C=CH 2 + I Radical D COOCH 3 Radical E Unsaturaded Endgroup B Figure 2.6 Possible gamma radiation reactions and products for PMMA.
PAGE 72
3.1.1 Substrates CHAPTER 3 MATERIALS AND METHODS 3.1 Materials Surface modification was carried out on several vascular prosthesis substrate materials. These included PET (Mylar D1000 and 700-Dl films from DuPont Electronics), woven PET (Dacron fabric, Meadox), reinforced velour woven PET (reinforced Dacron vascular graft, Bard), PDMS (KE-1935 A and B, Shincor Silicones), ePTFE (sp.# 728-3 GORE-TE~ expanded PTFE, Gore), and PTFE (skived Teflon, Goodfellow). The chemical structures of polymer substrates used in this research are presented in figure 3.1. All intracorneal lens (ICL) substrates for Hydrograft modification were PMMA, and were provided by Surgidev as either PMMA sheets (V-811 low molecular weight PMMA), fenestrated flat PMMA disks, or ICLs. These ICLs have three hole sizes and percent surface coverage of holes, lOm/5% (209-99), 30m/5% (220-121), and SOm/5% (220-131), and all are 6 mm/-8 diopter. The PMMA used in the manufacture of the ICLs has a significantly lower molecular weight than PMMA typically used for ocular applications. For this reason, 55
PAGE 73
56 different conditions than those studied by Yahiaoui (1990) were used for modification of ICLs. Table 3.1 shows molecular weight data for several PMMA ocular materials. PET PMMA PDMS PTFE 0 11 n n Figure 3.1 Chemical structures of polymer substrates subjected to surface modification. n Table 3.1 Molecular weight data for PMMA ocular implant materials, from Goldberg, et al. 1988-1995). Name Lot# Mn Mw MWD Perspex CQCV 001544 3.3 X 10 6 5.1 X 106 1.5 Blue Perspex B# 001710 1.2 X 10 6 3.6 X 10 6 2.9 ,-inticlear P973 1.1 X 106 2.5 X 10 6 2.4 Low MW PMMA V-811 45,000 86,000 1.9 Nidek IOL PMMA 650,000 2.6 X 10 6 3.9 Storz IOL PMMA 60,000 110.000 1.8
PAGE 74
57 3.1.2 Monomers and Reagents for Surface Modification The monomers and polymers used for surface modification included methyl methacrylate (MMA, Kodak), N-2-vinyl pyrrolidone (NVP, Polysciences), 2-sulfoethyl methacrylate (SEMA), methacryloxypropyl terminated PDMS (MAOP-t-PDMS, United Chemical Technology, Inc.), and PDMS (KE-1935 A and B). MMA and NVP were supplied with MEHQ as an inhibitor, and were purified by vacuum distillation (1-2 mm Hg at 40-60C). The Shin-Etsu silicone was supplied in a ready to use two part oligomer mixture. SEMA was supplied with 5% MEHQ, and no practical methods of its removal were found. Therefore, SEMA was used as supplied. The MAOP-t-PDMS was supplied with no inhibitor, and was used as received. The structure of the monomers and oligomers used are displayed in figure 3.2. 3.2 Methods 3.2.1 Sample Preparation and Substrate Cleaning 3.2.1.1 Preparation of Shin-Etsu PDMS Films The KE-1935 PDMS was supplied by Shincor Silicones in two separate parts. The mechanical properties of the final polymer may be varied by altering the time and temperature of the curing reaction. Samples used in this research were prepared by the following standard procedure unless otherwise noted. Twenty to twenty-two grams of each component were poured onto a bordered glass plate (16 cm x 24 cm), mixed
PAGE 75
5 8 thoroughly with a glass stirring rod, and the surface was leveled with a glass microscope slide. The plate was degassed in a vacuum oven (76 cm Hg) at 50C for 45 minutes, and then transferred to a preheated, 150 c ,oven and cured for one hour in air. The final thickness of the cured sheet was approximately 1 mm. CH 2 =CH I 0 ,// C NVP CH 3 I CH 2 =C I C=O I CH 3 CH I 3 I CH 2=C I CH2==C I C=O C=O I I 0 0 CH 3 'CHz-CH 2 MMA SEMA 0 CH 3 CH 3 I I CHz-CHrCH~i -o-------"' I CH 3 CH 3 MAOP-t-PDMS 0 11 S-OH 11 0 Figure 3.2 Chemical structur es of monomers and reagents used for surfa ce mod ification.
PAGE 76
59 3.2.1.2 Substrate cleaning prior to surface modification All substrates were cleaned prior to use to remove surface contamination and impurities. Samples referred to as ''controls'' have undergone the cleaning process as well, unless otherwise mentioned. Dacron, Mylar, PTFE, ePTFE, and silicone substrates were cleaned by sonication for 10 minutes each in acetone, isopropanol, and Ultrapure water. The samples were then rinsed in Ultrapure water and placed in a vacuum oven at S0C to dry for 12 hours (76 cm Hg). PMMA samples were cleaned by sonication in Ultrapure water for 10 minutes followed by a rinse in Ultrapure water. Samples provided as lenses were not cleaned prior to modification because they were received in final manufactured condition. 3.2.1.3 Solution degassing The samples for gamma radiation surface modification were placed into borosilicate glass tubes with the grafting monomer solution (sample completely submerged in the solution). Before irradiation, the samples were ''degassed'' to remove as much oxygen from the solution as possible using one of two methods. The samples were degassed by vacuum (ca. 20-30 mm Hg) for 2 to 5 minutes, purged with argon, and sealed with a polyethylene cap (Tainer Top, Fisher Scientific). The second method used was to bubble argon through the solution with agitation to replace the oxygen with argon in solution. Bubbling was done through a glass pasteur pipette for 2 to 5 minutes (2 minutes for 1 to 3 ml
PAGE 77
60 solution volumes and 5 minutes for 5 to 10 ml solution volumes). Bubbling was chosen over vacuum degassing for more volatile solutions (e.g., methylmethacrylate monomer/acetone solutions) to avoid evaporating of solution components and changing solution concentrations. All samples were then irradiated to the specified dose. 3.2.1.~ Substrate cleaning after modificatiqn Following surface modification with MMA, SEMA, or PDMS, the residual polymer, monomer and solvent was removed from the sample tube. The modified substrates (PET, PTFE, ePTFE, and PDMS) were then placed in acetone to begin washing to remove the remaining unbound polymer and residual monomer. The acetone was removed and replaced three times per day, for three days. The samples then were rinsed in Ultrapure water and vacuum oven dried. PMMA samples modified with NVP were cleaned in the same manner described, with Ultrapure water being substituted for acetone for the washing procedure. 3.2.2 Surface Modification Methods 3.2.2.1 Presoaking Some samples were subjected to a presoak step prior to gamma polymerization. The presoak step involves placing the sample to be modified in a monomer solution, usually at higher concentration and temperature, to allow monomer and solvent to diffuse into and swell the substrate surface. The presoak conditions for a polymer-monomer system were usually
PAGE 78
61 determined by swelling experiments. The samples were placed in various solution concentrations in sealed borosilicate tubes, and then placed in an isothermal water bath. The weight increase of the sample with time was recorded, and the solution concentration, temperature, and presoak time which provided significant weight increases (5% or higher) were chosen as the presoak conditions for that particular system. Presoaking was also used for the solution dip coating of PDMS onto Dacron. A chloroform/PDMS solution was used swell the PET fibers and allow diffusion of silicone oligomers into the substrate surface. Immediately following the presoak for surface modification of PMMA with NVP, the samples were quenched with ice water to reduce temperature and stop diffusion. 3.2.2.2 Gamma radiation induced polymerization The samples were immediately placed into the gamma solution (if gamma solutions were different than presoak solutions) following the presoak and degassed. The solution concentrations for gamma polymerization were chosen based on the final viscosity of the solution. That is, solutions which could be easily removed following polymerization were more desirable, and concentrations which caused gellation at a given gamma dose were avoided. The samples were placed into the gamma source immediately following degassing. Gamma irradiation was conducted by simultaneously exposing the substrate and monomer solution to a 600 Curie 60co point source. The samples were placed into a circular, motorized
PAGE 79
62 carousel to provide unifor1n exposure to all samples. Schematic diagrams of the gamma source chamber and carousel are shown in figures 3.3 and 3.4, respectively. Figure 3.3 Support Rod----1__.... --Door handle Door (Shown in opened position) I I ,__ ___ Source Housing --+-+-Gamma Chamber ~--60co ------carousel Schematic diagram of the irradiation chamber used for polymerization. --Carousel ':<4-= -/ ., -, Sample -Test Tube -~ I Motor Figure 3.4 ,_ / ... /, ... ~ Motorized carousel used to provide unifortlt exposure within the irradiation chamber.
PAGE 80
63 The total radiation dose varied from 0.02 to 0.15 Mrad, and was controlled by time of exposure. The samples were placed 4 inches from the point source, which provided a dose rate of approximately 425 rads/min. The dose rate was determined by measuring the absorbence of dosimeter film (GAF Chromic) exposed on a calibrated irradiator at the Shands Hospital Radiation Oncology Center, Gainesville, Florida (Goldberg, et al., 1988-1995). Twenty-five films were exposed from 0.025 to 0.200 Mrads and the absorbence at 540 nm was determined by UV/VIS spectroscopy (Perkin Elmer Model Lambda b UV-VIS) at 540 nm wavelength. A calibration curve was generated, and used for exposures on the ganuna source used in this research. After exposure to gamma radiation, the samples were removed from the grafting solution and systematically washed according to the washing procedure discussed in section 3.2.1.4. 3.2.3 Solution Dip Coating of PDMS onto Dacron Dacron samples were coated with silicone as a surface modification method. Equal amounts of a 10% of KE-1935 A and KE-1935 B in chlorofor1c1 (w/w% solution) were mixed in a borosilicate glass tube with a screw cap (Kimax tubes, Fisher Scientific). Dacron fabrics or Bard vascular grafts were placed into the solution, and the caps were placed on the tubes to avoid evaporation of the chloroform. The samples remained in the oligomer solution for 4 hours at room temperature to allow swelling of the PET and some diffusion
PAGE 81
64 of the silicone oligomers into the substrate. Upon removal from the solution, the samples were suspended (alligator electrical clips) in a 60C oven, in air The samples were allowed to cure for 24 hours. Then vacuum was applied to the oven (76 cm Hg), and the samples cured under vacuum at 6oc for another 24 hours. To remove the uncured, low molecular weight oligomers, the samples were washed in acetone, with six solvent changes, for 48 hours, followed by washing in hexane with three solvent changes for 24 hours. The samples were then placed into a vacuum oven (60c, 76 cm Hg) for 12 hours to remove the acetone and hexane. The adhesion of the silicone layer to the substrate was also studied (c.f. section 3.2.6). To improve the bonding of the silicone layer to the PET substrate, a pre-dip coating step was used. The Dacron was subjected to a presoak in MMA solutions, followed by gamma irradiation in a chlorofor111 solution with 10% MAOP-t-PDMS and 10% MMA. This step is believed to swell MMA into the Dacron. Upon exposure to gamma radiation, the MMA and MAOP-t-PDMS polymerize into and on the Dacron surface providing a covalent link from Dacron, to PMMA, to MAOP-t-PDMS. The second step of solution dip coating and curing PDMS onto this surface will then provide a covalent bond between the Dacron and the PDMS.
PAGE 82
65 3.2.4 Characterization 3.2.4.1 Gravimetric analysis Gravimetric analysis was used as a quick, non destructive and inexpensive method to provide information concerning gravimetric yield for swelling (presoak) or surface modification. All mass increases and decreases were recorded on a Denver Instruments A-200DS electronic balance with a precision of .02 mg. Percent weight increase (or decrease) was determined by the percent change in final and initial weights, and is defined in equation 3.1, Percent Weight Change= {(Wt Wi)/Wi} 100 (3.1) where Wi is the initial weight (usually referring to the unmodified, clean, dry substrate) and Wf is the final weight (usually referring to the modified, clean, dry substrate). In some instances, wet weights were used as initial weights (e.g., swelling experiments, c.f. section 4.1.1.2). 3.2.4.2 Contact angle goniometry Contact angle goniometry is also non-destructive and inexpensive, and provide information concerning the relative wettability of the surface. This technique is dependent on the outer1ctost few monolayers of the polymer surface, and may only be used on solid polymer substrates. Contact angle
PAGE 83
66 values for all substrates were measured on a Rame-Hart contact angle goniometer (Mountain Lakes, NJ) at room temperature using the captive air bubble technique unless otherwise mentioned. Some samples were evaluated using the water drop in air technique. The captive bubble technique involves suspending the sample in water (clipped to the underside of an aluminum block immersed in water), and injecting air bubbles (ca. 0.2 l) with a microliter gas chromatography syringe, and allowing them to come to rest underneath the sample. The contact angles reported are averages of measurements on six bubbles per sample. The contact angle measured is related to the solid-vapor (Ysv), solid-liquid (Ysl), and liquid-vapor (Y1v) interfacial free energies. The relationship between these values is Young's equation, as shown in equation 3.2. cos (0) = (Ysv Ys1)/y1v Schematics of the captive bubble technique and the angles measured are shown in figure 3.5. (3.2) Typical values obtained with the goniometer used in this research are 105-110, for hydrophobic substrates such as PTFE, 50-60 for intermediate substrates such as PMMA, and s20 for hydrophilic hydrogel substrates (Yahiaoui, 1990).
PAGE 84
67 rL.~Lt..Z4.~~~~~~~~H~~L.L. 4 ~u;..r-Aluminum Block Ysl Figure 3.5 water <3----++-----Sample (a) a.1.r solid (b)
PAGE 85
68 such as molecular vibrations, rotations, and stretching. Percent transmission or absorbence (of the IR beam) is collected for wavelengths of 2.5 to 20 m (or wavenumbers of 4000 to 400 cm-1). Attenuated total reflectance (FT-IR/ATR) utilizes a crystal through which the beam is passed. There are conditions under which the infrared radiation passing through the crystal will be totally internally reflected. The sample is mounted in contact with the crystal, and the evanescent wave created within the sample is attenuated in the regions where the sample absorbs energy. (Spectra Tech reference Manual, Chapter 13) A schematic of the crystal, beam, sample configuration is shown in figure 3.6. The depth of penetration, dp, of the beam into the ~ample, and therefore the depth of practical analysis is described by the relationship shown in equation 3.3, (3.3) where A is the wavelength of the incident IR radiation, n1 is the refractive index of the ATR crystal, n2 is the refractive index of the sample, 0 is the incident and exit angle of the IR beam. The major advantage to FT-IR/ATR is the ability to provide chemical structure and bonding information on polymeric surfaces. This allows for identification of surface polymers present following surface modification.
PAGE 86
69 FT-IR microscopy utilizes a microscope stage for mounting the sample where the IR beam may be focused on a specific sample surface area. The sample may be analyzed by either transmission, where the beam is passed directly through the sample, or by a special ATR microscope stage for collecting ATR spectra. The main advantage to the transmission microscopy FT-IR system is being able to evaluate specific sample areas, however it requires thin samples, typically less than 100 m. Since transmission spectroscopy is not surface sensitive as is the case with ATR, information on the bulk composition of the sample is collected. FT-IR data were collected using a Nicolet 20SXB FT-IR spectrometer using a lmW HeNe laser and a parallelogram KBr crystal with a 60 entrance/exit face angle. Typically 128 scans at a resolution of 4 cm1 were signal averaged to obtain individual spectra. Data processing was done on the Nicolet software provided with the equipment. Spectra presented here were scaled and printed using OMNIC 1.2 software (Nicolet Instrument Corporation). Detector Figure 3.6 --~~~=~~==~Q;:---Sample V<:J---Crystal IR Source Schematic representation of the IR beam, crystal, and sample for FT-IR/ATR spectroscopy.
PAGE 87
70 3.2.4.4 X-ray photoelectron spectroscopy X-ray photoelectron spectroscopy (XPS), also referred to as electron spectroscopy for chemical analysis (ESCA), is a fairly new technique for surface analysis. Developed during the 1950s, XPS is an analytical technique which provides the atomic chemical composition both quantitatively and qualitatively, making it a powerful tool in the characterization of surface modified polymers. Based on the photoelectric effect, XPS bombards sample surfaces with X-rays, which cause the ejection of core electrons. The ejected electron emit discrete energy values related to the binding energy of the electron and the exciting radiation energy by equation 3.4 (Barr, 1994, and Hercules and Hercules, 1976) Each atom has a specific binding energy, and the atoms to which they are bonded cause discrete shifts in the binding energies. This enables XPS to measure atomic concentrations of surfaces as well as provide information concerning the chemical environment of the surface elements. Eb = hu Ek cf> (3.4) In equation 3.4, Eb is the binding energy, hu is the photon energy (where his Plank's constant and u is the X-ray frequency), Ek is the kinetic energy of the electron
PAGE 88
71 (measured value), and is the work function specific for the instrument (Barr, 1990). XPS data were collected using a Kratos model XSAM-800 spectrometer with a Mg Ka X-ray source. The X-ray gun was operated at 12 kV and 19 mA, and the analysis chamber pressure was maintained at 107 to 10-e torr during analysis. Quantification of the spectra was perforn~d using DSSOO Kratos software on a Digital computer system. Binding energies were calibrated using Eb= 285.0 ev as the Cls peak on all spectra, unless otherwise indicated. The depth of analysis is in the range of soA. 3.2.4.5 Dynamic mechanical sampling Dynamic mechanical sampling (DMS) measurements were made with a Seiko DMS 200. The sample response to dynamic tensile forces were recorded as a function of time, temperature, and frequency (for frequencies of 0.1, 0.5, 1.0, 5.0, and 10.0 Hz) from -140 to 250 c. The sample ther1rtocouple was calibrated using the maximum value of the dynamic loss modulus (E") at 1 Hz for the glass transition of PMMA (Tg=117C) (Feller, 1993). The elastic response was calibrated with the tensile modulus of PMMA measured at an elongation rate of 10%/min in uniaxial tension. The samples analyzed had a gage length of 20 mm, and had a range of cross-sectional areas of 4.9 to 6.4 mm2. Information for the storage modulus (E'), loss modulus (E"), and tan o (E /E'') are presented as a function of temperature
PAGE 89
72 for silicone modified with PMMA, as a function of both radiation dose and monomer concentration. 3.2.4.6 Light/Optical microscopy Sample inspection, graft thickness measurements, surface topography, and other visible observations were made using a Nikon optical microscope. Optical micrographs were taken using a canon camera mounted on the same microscope. The microscope was equipped with a graduated eyepiece used for size measurements. A hemocytometer with precision markings was used for calibration of the eyepiece for all magnifications used, providing a resolution~ 1 m. Organic and inorganic compounds were used for staining surface modified regions of substrates to facilitate visualization. A 10% silver nitrate solution was used for staining the Hydrograft~ materials. The modified samples were placed in the silver nitrate solution for 12 to 24 hours, and then placed in phosphate buffered formalin (10% formaldehyde) to precipitate silver oxide and silver phosphate. A saturated crystal violet-acetone solution was used to stain the PDMS coating on Dacron~ and PMMA modified PTFE. 3.2.4.7 Scanning electron microscopy High magnification evaluations of surface appearances (before and after modification and evaluations) were made using a Jeol 6400 scanning electron microscope (SEM).
PAGE 90
73 Samples were coated with a thin layer of gold palladium using a Hummer v sputter-coater (Technics, Alexandria, VA), unless otherwise mentioned. Typical accelerating voltages of 1 to 3 kV were used. These lower voltages were used to avoid sample ablation typically caused by higher voltages (Goldberg et al., 1988-1995). Micrographs were obtained at various magnifications to provide representative records of surface appearance. 3.2.5 Diffusion/Flow Cell Testing of ICLs The fenestrated ICL design attempts to provide a route for the eye to supply nutrients to the stroma anterior to the implant. A custom diffusion chamber provided by Surgidev was used to evaluate the permeability of the fenestrated intracorneal lenses. As displayed in figure 3.7, the apparatus can be used to determine flow rates of liquids and solutions through the lens, as well as evaluating the diffusion (osmotic pressure controlled) of solution constituents such as NaCl, glucose, and proteins. The apparatus is constructed of PMMA with a PMMA divider in the center. The ICL is placed in the right side of the chamber on an o-ring, and the pressure screw cap is tightened to hold the lens in place and seal the edges. Flow is in the right to left direction mimicking flow through the lens from posterior to anterior sides. Two types of tests were conducted using this apparatus. The first test was to deterxiaj ne the water flow through the
PAGE 91
74 lens at a constant pressure. One chamber was filled with water and an inverted, water-filled flask (with an air bleed port) was placed in one chamber to provide a constant hydrostatic pressure. The other chamber was left empty. The volume of water flowing through the lens was measured after 90 minutes had passed. This test was conducted five times for each sample, and an average value for each lens condition was determined. The next test involved analysis of the permeability of saline and glucose solutions though the lens as driven by osmotic pressure and chemical potential. One chamber was filled with the solution of interest, and the other was filled with Ultrapure water. The solution concentration of each chamber was evaluated as a function of time. Osmolarity was measured using an osmometer (Osmette Micro-osmometer, Precision Systems) to determine saline concentration. A linear calibration curve (r2 = 0.999) for solution concentrations of 0% to 0.9% saline was created using a series of 10 solutions. Initial saline concentrations used in the diffusion chamber were 0.9%. A Glucometer4 glucose reader (Ames) and glucose sticks (Ames) were used to determine glucose concentrations. A calibration curve (r 2 = 0.997) for solution concentrations of 380 mg/dl to 38 mg/dl glucose was created using a series of 10 solutions. Initial glucose concentrations used were 380 mg/dl.
PAGE 92
75 ICL(Dome-Side Facing Left) I Water or Solution Receiving Chamber / Chamber Figure 3.7 4 I Pressure Screw Cap Schematic illustration of diffusion and flow cell used for evaluating permeability of ICL materials. 3.2.6 Leak and Stability Testing of vascular Grafts The silicone dipped Dacron vascular grafts may be rendered non-porous by the silicone coating on the inside and outside of the substrate. The leak-rate of water through the pores was evaluated using a pressurized flow system. The graft was placed in series with a pressure manometer and water reservoir, and a back pressure of nitrogen. A schematic of this test set up is illustrated in figure 3.8(a). The volume of water leaking through the graft surface at 120 nun Hg in one minute was measured. The reported values for leaking were normalized for the surface area of the graft, and the leak rates are reported in (ml)/(cm 2 -min). The stability of the silicone coating on the Dacron substrate was also evaluated to determine if any material
PAGE 93
76 delaminated or flaked off during dynamic flowing conditions. A continuous flow of water at 300 ml/min at 110-120 mm Hg through a vascular graft was maintained for up to 48 hours. Samples of the water were taken and analyzed with inductively coupled plasma (ICP) to deteLmine the concentration of silicon atoms in the solution. The sample volume taken for analysis was 10 ml, to which 0.1 ml of Triton X surfactant was added to suspend silicone molecules which may be in the water sample taken. Since the vascular environment contains both lipophillic and hydrophilic molecules, silicone coated Dacron grafts were analyzed in the flow system with both water and a 10% Triton X water solution. This test apparatus is shown in figure 3.8 (b). 3.2.6.1 Inductively coupled plasma ICP utilizes an argon plasma torch to ignite a fine solution mist. The emission spectrum quantized for each element, and when the spectra of the ionized gas is compared to reference solutions the atomic concentration of a specific atomic species may be determined in parts per million (ppm). The detection limit of silicon atoms at 251.6 nm by the ICP used in this research is 0.1 ppm, which corresponds to a solution concentration of 0.1 g/ml. A Plasma 40 ICP (Perkin-Elmer) was used with a window size of 0.1 nm, a photo multiplier tube voltage of 700 V, and an integration of 680 msec. Each sample was analyzed 3 times, and means and standard deviations are reported.
PAGE 94
Glass Tube Attached to Nitrogen Cylinder and Manometer 77 PE Tubin Flask Filled with Water Polyethylene Tubin Submerged Graft Water Beaker Plug to Force <.J-t--Leakage Through Graft Pores Collection Beaker ( a) (b) Mercury Manometer Flow Direction Water Pump Figure 3.8 Schematic illustrations of the leakage (a) and stability (b) analysis set up.
PAGE 95
78 3.2.7 Ex Vivo and In Vivo Studies of Vascular Substrates 3.2.7.1 Ex vivo AV canine shunt analysis Using aseptic technique, an arteriovenous (AV) shunt was constructed between the carotid artery and the jugular vein of an adult mongrel canine. Samples to be tested were placed into a section of Silastic~ tubing, and were sealed in place with silicone RTV. Autologus 111 Indium labeled platelets were injected into the dog, and blood flow over the samples allowed for 60 minutes. The samples were then removed from the shunt, and counted in a gamma-counter (Auto-logic, Abbott Laboratories). The counts are normalized to surface area of the samples, and are reported as counts/mm2. 3.2.7.2 Canine in vivo evaluation of vascular grafts Under general anesthesia and with heparin infusion (lOOU/kg), a 6 mm djameter, 10 cm long vascular graft was placed end-to-side on the infra-renal abdominal aorta, proximally, and on the common iliac artery distally. Two grafts are implanted in each animal, an unmodified control graft, and the modified graft of interest, forming a bilateral aorta-iliac bypass on each side. The implants were allowed to remain in place for 30 days. Twenty-four hours prior to graft explantation, lll1ndium labeled platelets were injected. The grafts were removed and first counted for radioactivity, and then sectioned for
PAGE 96
79 histological analysis. Counts of the grafts are reported in counts/mm2 for the total (inner and outer) luminal surface area.
PAGE 97
CHAPTER 4 RESULTS AND DISCUSSION 4.1 Gamma Radiation Induced Polymerization of Methyl Methacrylate on PET, PTFE, and PDMS A review of the literature and current studies within this laboratory led to the belief that PMMA surfaces on biomaterials for the vascular environment have the potential advantages of reducing platelet cell adhesion and reactivity, and increasing endothelialization and healing of vascular prostheses (Hufnagel, 1954, and Bowers et al., 1989). Gamma radiation induced polymerization (GRIP) of methyl methacrylate has been reported in the literature. To create a stable surface of PMMA on a polymeric substrate with gamma polymerization, a mixture of monomer and substrate surface molecules must be created. The first step in achieving this mixture for all substrates swelling the substrate with a monomer solution, allowing the monomer to diffuse into the substrate surface. GRIP is a heterogeneous system, and Odian reports the rate of graft polymerization (polymerization of monomer directly to activated substrate molecules) to be diffusion controlled (Odian, 1981). The solvent, monomer concentration, temperature, and time were all evaluated to some degree to determine conditions providing surface modification of the substrates with MMA. 80
PAGE 98
81 4.1.1 Swelling of PET, PTFE, and PDMS in MMA Solutions The choice of solvents for the swelling of the substrates with MMA solutions was based on solubility parameter, compatibility with MMA and PMMA, relative toxicity, and ease of removal. Table 4.1 lists the solubility parameters for studied substrates and some typical solvents. Typically, the more similar the solubility parameters of the solute and solvent, the better the solubility of the solute in the solvent (Sperling, 1986). Solvents used must be solvents for PMMA to avoid precipitating the homopolymer and surface grafted region during polymerization. This facilitates solvent and un incorporated homopolymer removal following irradiation, and allows swelling and diffusion within the surface region to continue during polymerization. Complete removal of toxic solvents is important for biomaterials applications. TABLE 4.1 Solvent Acetone Chloroform Cvclohexane DMSO n-Bexane TBF Toluene MMA Solubility parameters and H-bonding groups for selected solvents and polymers Solubility B-Bonding Polymer Solubility Parameter qroup Parameter 9.9 8 PMMA 9.0-9.5 9.3 p PET 10.7 8.2 D PTFE 6.2 12.0 m PDMS 7.3 7.3 D 9.1 m 8.9 D 8.8 m Solubility parameter units are (cal/cm 3 )1/2. B-Bonding group refers to the strength of hydrogen bonding by the material, wheres strongly, m s moderately, and p = poorly bonded. (Brandrup and 11,unergut 1975).
PAGE 99
82 4.1.1.2 Effect of solvent, concentration, and temperature Swelling of PET. In an attempt to diffuse MMA into PET, a solvent selection was initially difficult because of the high solubility parameter (10.7) and crystallinity of Dacron~ PET fibers. Percent weight increase, or weight uptake, was used to evaluate swelling (c.f. section 3.2.4.1.). Weights were determined after blotting the samples with Whatman #3 filter paper. Myla~ PET films were used initially in the swelling studies for the ease of handling. table 4.2 shows the weight uptake of selected solvents by PET at room temperature (25 to 28C) after 24 and 48 hours. After observing these data, the two solvent systems chosen for PET were chloroform and DMSO. Figure 4.1 shows a plot of weight uptake of MMA-chlorofor1n solutions by PET (Mylar~ D-1000) with time for various solution concentrations. Percent crystallinity changes were not evaluated in this research, but some solvent induced crystallization of PET by chlorofor1c, is beleived to be possible. Table 4.2 Percent weight increases of Myla~ D-1000 films in selected solvents at room temperature. Percent Weight Uptake Percent Weight Uptake Solvent at 24 hours at 48 hours Acetone 11 1 Cyclohexane < 1% DMSO 2% 2 Ethvl Acetate < 1% < 1% Hexane 2% 2 TBF < 1% 2 Toluene < 1% < 1% Methvl methacrvlate < 1% Chloroform 22-25%
PAGE 100
25 Q,) 0 20 as Q,) 0 C ..c: C) Q,) 3: C Q,) 0 Q,) a. 15 10 5 0 0 83 0 1 OOo/o Chloroform: Second order R 2 =0 997 6 5/o MMA-chloroform: Second order R 2 =0.997 1 Oo/o MMA-chloroform: Second order R2=0 96 15/o MMA-chloroform : Second order R 2 =0 93 v 20/o MMA-chloroform: Second order R 2 =0 93 0 0 0 V V 300 600 900 1200 1500 Swelling Time (minutes) 1800 Figure 4.1 Percent weight increase of Mylar@ D-1000 PET films with time in MMA-chloroform solutions as a function of solution concentration. Figure 4.2 shows the same data presented in figure 4.1 plotted as Mt/Moo versus t 1 1 2 /l based on equation 4.1. (4.1) In equation 4.1, Mt is the mass uptake at time, t, Moo is the maximum mass uptake (mass uptake at infinite time), 1 is the
PAGE 101
84 sample thickness, and Dis the diffusivity. Equation 4.1 shows the relationship between data presented and the diffusivity, where the slope of each line in figure 4.2 is the related to the diffusivity, 0 1 1 2 (Crank and Park, 1968). 8 1 0 0 9 0 8 0 7 0 6 0.5 0 4 0.3 0.2 0 1 0 0 0 Figure 4.2 V V 0 0 300 600 900 0 V ~v V 20 MMA-chloroform V 15/o MMA-chloroform 10 MMA-chloroform 6 5/o MMA-chloroform O Oo/o MMA-chloroform 1200 1500 1800 (time' 112 / I fmln 112 / cm 2 ] Mt/M oo vs. time~ / 2 /1 for MMA-chlorofor1ct solution as a function of solution concentration. The curves have initial slopes which change as a function of concentration, then average slopes become equal for each data set as the data becomes somewhat scattered. The initial permeation is within the amorphous phase of the material, and has a rate dependance on solution concentration. Plotting the initial slopes from figure 4.2 as functions of solution concentration yields the second order curve in figure 4.3.
PAGE 102
C >,. > V, ::, .... .... 0 5e-05 3e-05 i 2e-05 ::, (.) 6 1e-05 85 Second Order R 2 = 0. 98 0 0 0 5 10 15 20 Solution Concentration (% MMA In Chloroform) Figure 4. 3 Diffusivity of MMA-chlorofor11\ solutions in PET as a function of solution concentration. Diffusivities are the slopes of the linear regressions for data presented in figure 4.2. The observation made here is that solutions of increasing concentrations of MMA, have progressively decreasing initial diffusivities. The data presented in figure 4.3 follow a second order plot, approaching a maximum with decreasing MMA concentrations. Following initial swelling of the amorphous phase the slopes approach similar values as permeation into the crystalline regions begins. Diffusion into a crystalline phase has a stronger dependance on kinetics than thermodynamics, and the solution concentration dependence is no longer observed. Although percent crystallinity before
PAGE 103
86 and after swelling was not measured, differences in the rate of permeation into amorphous and crystalline regions is the most logical explanation for this observed behavior. Figure 4.4 shows the maximum swelling values plotted as a function of concentration, also following a second order relationship similar to figure 4.3. The relationship between the diffusivity as a function of concentration is related to the solubility parameter. The solubility parameter is related to the cohesive energy density (CED), or relative affinity between each molecular species. The more similar the solubility parameters, the higher the affinity between molecules, and thus, the higher the degree of swelling and solubility (Sperling, 1986). As the concentration of MMA is increased, the solubility parameter of the solution (using a volume fraction linear rule of mixtures) decreases from the value for chloroform (9.3) toward the value of MMA (8.8), and away from the value of PET (10.7). Thus, as the relative difference in the affinities of the substrate and solvent molecules becomes greater, the diffusivity decreases, and less swelling occurs. The second order relationships shown in figures 4.3 and 4.4 match the curves presented by Sperling (1986) and Fujita (1966). Figures 4.5 and 4.6 show relationships between solution behavior as a function of concentration. Figure 4.5 shows the curves obtained for solubility parameter as a function of solution concentration for polyisobutene and polystyrene, where a maximum in the curve is observed when the solubility parameters of the solution and polymer are
PAGE 104
87 equal. The regions of each curve corresponding to concentrations moving away from the maximums follow a second order relationships, matching the shape of the curves in figures 4.3 and 4.4. Figure 4.6 shows a second order increase in the polymer-fixed diffusion coefficient of the diluent, which is a measure of the per1neation of benzene into the natural rubber. Although the situation presented here is more complicated than a simple binary solvent-polymer system, the diffusion of the solvent solution is controlled by the CED of the solution and substrate, the free volume of the substrate, and the activation energy of diffusion. Cl) Cl) cu Cl) (.) C 20 -15 .c: C) 0 E 10 :::, E 5 Second order curve fit R 2 = 0 999 0 5 10 15 20 Percent MMA In Chloroform {v/v%) Figure 4.4 Maximum percent weight uptake by PET of MMA chlorofor11l solutions as a function of solution concentration.
PAGE 105
Figure 4.5 88 8 9 IQ A 1.0 B 0 8 0 0 6 0 0 0 4 0 2 7 8 9 d Solubility parameter for polyisobutene and polystyrene determined by intrinsic viscosity in a series of solvents (Sperling, 1986) 50 ,--------------...,--,250 40 200 u u cu 30 "' 150~ "' E E u u ,._ ,.._ 0 0 .... X X 100~ Q 20 ... Q 10 50 0 0 ~:::::~0L 2 __ 0 L 4 __ 0.J. 6 ___ 0.L a __ _J1 g Volume fra c t i on of benzene Figure 4.6 Variations of D (diffusivity) and (Ds)p (polymer fixed diffusi o n coeff icient of the diluent) with v ol ume fract io n of benzene for the system natural rubber-benzene at 25C (Fujita, 1966).
PAGE 106
89 Woven Dacron (Meadox fabrics) samples were also evaluated to determine the similarities and differences between the swelling of Mylar and Dacron. The percent solvent uptake values for Dacron were consistently 10 to 15% higher than values for Mylar films, and were attributed to excess solvent being trapped in between the fibers after the sample had been blotted. When the initial dipped value was used as the initial weight in percent increase calculations to allow correction for residual solvent trapped by the weave, the percent uptake values of the Dacron were within 3 to 5% of the Myla~ PET film values. As a result, swelling values reported for Dacron use the 1 minute weight as the initial sample weight (this does not apply to total sample weight gain following surface modification, only for swelling experiments) Temperature increases were not necessary for swelling of PET with MMA-chloroform solutions because of the degree of swelling at room temperatures. Temperature increases increase the rate of uptake and allow shorter times to be used to obtain room temperature swelling values. The heat of solution and the permeability coefficient (the volume of solvent vapor passing through a polymer of given thickness in a specified amount of time) have a temperature dependence as displayed in equations 4.2 and 4.3, = s -Rd lnS d (1/T) P = PS = p e-6E / RT 0 (4.2) (4.3)
PAGE 107
90 where ~His the heat of mixing, ~sis the entropy of mixing, Sis the solubility coefficient (which determines the concentration and pressure gradient for the system), Tis the absolute temperature, and ~Eis the activation energy for permeation of a solvent in a polymer, or the energy required to allow the molecule to move from its current position to a void. Increases in heat for a polymer-solvent system thus increase diffusion by adding energy to both the heat of solution and the permeation Swelling of PET in MMA-DMSO solutions showed similar trends previously discussed for MMA-chloroform solutions, but the total weight increases were much lower. Table 4.3 shows weight uptake by PET in MMA-DMSO solutions at 60C. Mylar~ had weight uptakes of less than 4% in DMSO-MMA solutions at room temperature, indicating a higher activation energy barrier diffusion of DMSO-MMA than for chloroform-MMA solutions in PET. Table 4.3 Percent weight increase of DMSO-MMA solutions by PET (Mylar D-1000) and Dacron at 60 c. Swelling time is 24 hours for all samples Solution Concentration Percent Weight Uptake Percent Weight Uptake MMA in DMSO by Dacron by Mylar 100% DMSO 0.7% 3% 20% MMA 3.9% 7.1% 40% MMA 7.6% 9.3% 50% MMA 4.8% 9.1% 60% MMA 2.8% 8.5% 80% MMA 2.9% 6.3% 100% MMA 0.9% 2.5%
PAGE 108
91 The solubility parameter of PET (10.7) is lower than DMSO (12.0) and higher than MMA (8.8). This indicates there is a solution concentration which would have a solubility parameter equal to that of the PET substrate, and maximum swelling should take place at this concentration. Figure 4.7 shows a plot of this data, and the maximum is observed at 40% MMA-60% DMSO, which has a solubility parameter of 10.7 as calculated as a volume fraction based linear rule of mixtures. Although the data presented shows a second order curve, when a curve is hand drawn through the data, two second order concave-up curves approaching a maximum in either direction become visible. The shape of such a curve matches those presented in figures 4.5 and 4.6. The curve for Dacron only has a regression fit of 0.82, making it even more obvious the overall curve is more complex than a simple second order relationship. As the solubility parameter of the solution mixture approaches the value of 10.7 for PET in either direction, the swelling increases as a second order function, matching the data presented above for MMA chloroform solutions. There remains an immediate and obvious difference between the chloroform and DMSO systems that still has not been explained. First, the 10.7 value of PET is reached with a 40% MMA-60% DMSO solution which has a maximum swelling at 60C of approximately 10% whereas the swelling for 100% chloroform has a maximum swelling of approximately 22% even though the solubility parameter of chloroform is 9.3.
PAGE 109
Q) en cu Q) (.) C .c C) C Q) (.) Q) a.. 10 9 8 7 6 5 4 3 2 1 0 92 Mylar Swelling in MMA DMSO Solutions : 24 hours at aoc : R2= 98 V Dacron Swelling in MMA-DMSO Solutions : 24 hours at aoc : R2= 82 ~-~---A.~ Id 6' / /A Au V 6& \ I~ ~\ I \ I \ ,,,,,~ V \ I "' '' \ / / V \ ,~ / \ ,' V 'V', '6 I \ \ \ \ _. \ 10 20 30 40 50 60 70 80 90 100 Solution Concentration (Percent MMA In DMSO) Figure 4.7 Percent weight increase with time of Mylar and Dacron materials in MMA-DMSO solutions at 60c as a function of solution concentration. It is important to realize data presented here is based on weight and not volume. If the densities are considered, the percent swelling based on solvent volumes become 9.5% (for 40% MMA-60% DMSO) and 15% (for 100% chloroform). Another factor controlling diffusion is the activation energy which must be overcome for a solvent molecule to jump from one position to another, and the activation energy for chloroform in PET is obviously much lower than for DMSO-MMA solutions.
PAGE 110
93 The conditions chosen for presoaking PET prior to gamma polymerization with MMA were (i) presoaking in 10% MMA-90% chloroform for 24 hours at room temperature, and (ii) presoaking in 40% MMA-60% DMSO for 24 hours at 60C. Swelling of PDMS. The choice of solvent system for PDMS was also difficult, but for the opposite reason of PET. PDMS swells well in many solvents. In fact, chloroform and hexane solutions were not feasible because the extent of swelling was great enough to cause tearing of the material as a result of excessive expansion forces induced by the solvents. DMSO was chosen as a solvent because of its relatively low toxicity in the body and the lower swelling compared to chloroform and hexane. Swelling based on weight uptake for PDMS in 100% MMA and MMA-DMSO solutions seems to reach a maximum after about 1 hour (5% weight increase for 10% MMA-DMSO solution), indicating diffusion completely through the material. Irradiation of this system would therefore give a bulk modified sample as opposed to a surface modified sample, and since the diffusion is rapid in this system, it was decided to expose these samples to radiation with no presoak process. The overall weight uptake of the solvent and final polymer following surface modification was used to evaluate the effect of monomer concentration for PDMS substrates (c.f. section 4.1.2.3.1). Swelling of PTFE. Because of a low surface energy and low solubility parameter, no weight uptake or swelling of
PAGE 111
94 PTFE was observed for any solvent system evaluated. Acetone has been reported in the literature as a solvent which slightly swells PTFE (1 to 5 weight percent), but this was not found to be reproducible in our laboratories (Sayed et al., 1981 and Razzak et al., 1987). Acetone, because of its previous use in the literature, and DMSO, because of its relatively low toxicity, were used as solvents for PTFE (Teflon) and ePTFE (GORE-TEX) surface modification with MMA. To increase the probability of graft polymerization, wetting of the surface by the solvent was necessary. Acetone-MMA solutions readily wet the surfaces, whereas DMSO MMA solutions did not, especially with the tendency of ePTFE to trap air within the pores. For this reason, DMSO-MMA solutions were mixed upon exposure to the materials, that is, the sample of concern was first added to pure MMA, and once the surface was wet, the specified amount of DMSO was added to achieve the specified solution concentration. Because no swelling was observed, various solution concentrations were irradiated to determine the conditions which provided suitable surface modification of PTFE with PMMA. Presoaking of PTFE and ePTFE samples was found to have no significant effect other than the brief time required to wet the polymer surface.
PAGE 112
95 4.1.2 Radiation Grafting of MMA on PET, PTFE, and PDMS 4.1.2.1 MMA grafting on PET (Dacron) Grafting of PET with MMA, although reported previously in the literature, presents several challenges (Stannett, 1981). First, the radiation stability of PET itself reduces the probability of true graft polymerization. Second, the difficulty of swelling the monomer into the surface reduces the ability to achieve an interpenetrating network (IPN). Finally, analysis of PMMA-PET systems with contact angle goniometry, FT-IR and XPS all have their own set of challenges. 4.1.2.1.1 Gravimetric analysis of MMA surface modified PET The percent weight changes from initial sample weights to final, grafted sample weights of MMA grafted PET did not show any significant weight increase as a result of adding PMMA. This observation was not surprising when the following parameters were considered. The total weights of the PET samples were in the range of 170 to 250 mg, with a thickness of 0.23 mm. A surface graft on a 0.23 mm thick sample will not have a weight high enough to add significant increases to values between 170 to 250 mg. For example, if a sample weighs 0.25 grams and swells 7.5% in a 10% MMA-chloroform solution, the increase in weight due to MMA (assuming ideal solution behavior) is 1.88 mg. If it is then assumed that all of the MMA becomes incorporated as polymer, the total weight increase due to PMMA will only be 0.75%. Previous
PAGE 113
96 results in radiation graft polymerization by Yahiaoui (1990), show that weight increases of 1 to 10% were obtained only after reaching graft a thickness of 100 to 200m. All data measuring weight increases for graft polymerization of MMA onto Mylar and Dacron PET show values of less than 1%, which is inconclusive with respect to showing the presence of PMMA on the Mylar film and Dacron surfaces. 4.1.2.1.2 Contact angle analysis of MMA surface modified PET Contact angles on Dacron were indeterminable because Dacron fabric is rough on the scale of a microliter bubble. When initial values of PMMA and PET (Mylar D-1000) were evaluated as unmodified surfaces to obtain base values, it was observed that they have similar contact angle values. The contact angle for smooth (optical grade and tumble polished) PMMA is 55 to 60, and the same range of values is obtained for Mylar. PMMA modified PET is not expected to have different values than unmodified PET unless oxidative degradation occurs during radiation exposure. The measured values for modified PET, unmodified PET, and PMMA controls were all within the range of 55 to 60. Thus, contact angle goniometry was inconclusive in showing the presence of PMMA on modified PET films. This observance raised questions about the benefit of a PMMA surface on PET with respect to surface energy. 4.1.2.1.3 FT-IR analysis of MMA surface modified PET The primary functional group used to identify the presence of PMMA is the carbonyl, which has an absorbence l
PAGE 114
97 peak from 1675 to 1740 cm-1 (Silverstein, 1981). However, PET has two carbonyl groups per structural repeat unit, and the carbonyl peak does not positively identify PMMA in a PET FT IR interferogram. There are however, benzene rings in the PET structure which have a stretching absorbence peak at 3100 to 3000 cm1 (Silverstein, 1981). FT-IR shows peak intensity relative to other peaks, bases on the relative concentration of the specific molecular groups. In comparing PET and PMMA modified PET, a decrease in the relative intensity of the benzene peak was expected, but not observed. The concentrations required to show differences in FT-IR/ATR spectra are on the order of 1 to 5% (volume), which explains the difficulty in detecting differences in the spectra of PET with and without the presence of PMMA. 4.1.2.1.4 XPS analysis of MMA surface modified PET XPS measures the presence and atomic concentration of atoms with atomic numbers greater than that of hydrogen. Thus, the atomic percent of carbon and oxygen in PET and PMMA was measured, and compared to the structure of the molecules. The ratio of carbon to oxygen C:O for both PMMA and PET is 5:2, and therefore, the percent carbon and oxygen for both PET and PMMA are the same when analyzed by XPS. The presence of carbonyl and ester groups in both polymers also eliminates differences in the shape of the Cls spectra as well. The benzene ring, however, has an unsaturated resonance structure which gives rise to a specific phenomenon known as the n-n* binding peak, shown in figure 4.8 (Moulder et al.,
PAGE 115
98 1988). The rr-rr* peak, or shake-up peak as it is often named, arises from a discrete energy loss from core photoionization lines from the transition of the unsaturated aromatic ring. This results in a second peak at slightly lower kinetic energy (higher binding energy) (Gardella et al., 1984). 10 rr-rr o c==== =--~==========:::::_ --: ...... ---=::::::= =::. ~ -< = Figure 4.8 Cls binding peak of polystyrene showing the rr-rr* shake-up peak (Moulder et al., 1991). There is, however, a slight indication of the presence of PMMA on the modified Dacron surfaces which was observed during the analysis of modified and unmodified Dacron. Dacron fibers are filled with titanium dioxide as previously mentioned, wh ic h is an autoaccelerant for the degradation of
PAGE 116
99 PMMA, especially by ultraviolet radiation (Ranby and Rabek, 1975). Following XPS analysis, the unmodified Dacron samples remained white whereas the samples modified with PMMA turned slightly yellow after being scanned with X-rays during XPS analysis. The wavelength of X-rays is different than that of ultraviolet rays, but both are capable of inducing photo-degradation. The extend of yellowing was not quantified in this research, but this observation could lead to such a study in future work. 4.1.2.1.5 SEM analysis of MMA surface modified PET SEM micrographs showing the typical appearance of both Myla~ D-1000 and woven Meadox Dacron fabric are shown in figures 4.9 and 4.10, respectively. There was no noticeable difference between modified PET and unmodified PET as viewed with SEM. Initial observations of Mylar films led to the discovery of an additive on the surface of D-1000 films. During film processing, a dispersion of silica is applied to the PET surface to reduce the static adhesion upon film roll up. This silica dispersion is clearly visible in figure 4.9 Mylar films which only had the silica dispersion on one side were then obtained from DuPont (Mylar 700Dl films). Contact angle, FT-IR, and XPS analysis presented was obtained from the non-silica coated side of the Myla~ 700Dl films.
PAGE 117
.. .. ,. ' ... .. .. "' ... l .. ~ .. I .. .. . ., ' .. ... ... .. .. 100 .. ... .. ... ... .. .. .. ' .. .. ... . .. .. .. .. .. .. ~ .,, 11' .... .ii/ .,, Jt:.i . .. .. .. .. t-t,, .. ...... . ..: ., . .......... .. ._r l .. ..q. .. ... ., < ... 'I, . , .. .. ,. C, .. .. C, '4 .. . .. .. ,. ., , ... . .. ... .. . . .. .. ..,, .. .. I ... .. .... .. .. .. .. .. .. .. ... .. ' .. . ... C 4' t!. --. .. .. .. .. . ... ... ... ... .. .. ,.. .. ... .. .. .. "2 !'0KU, .. ... ... -" "' . . ~, .. ... .. 'I 'I .:. -' , '" "" .. It .. \ ' .. .... . .. "'I ... _., .... ~ -.. --~_., ..... .. ;, ... ~' .. .. .. ... ......... .. :' . ... .. ... .. ... .. .. .. .. Figure 4.9 SEM micrograph of Mylar D1000 film showing the dispersion of silica on the surface. Figure 4.10 SEM micrograph of Mead o x woven Dacron fabric
PAGE 118
101 4.1.2.1.6 Ex vivo analysis of MMA modified PET There was no significant difference seen between the platelet adhesion of modified PET and unmodified PET for either Mylar or Dacron. For samples evaluated in the canine AV shunt, Dacron had counts/mm2 of 2841 and 2588, and PMMA modified Dacron had counts/mm 2 of 2073 and 2661. Mylar had counts/mm2 of 34 and PMMA modified Mylare had counts/mm2 of 227. There is a difference between the Dacrone and the Mylar which is attributed to the surface roughness of the two materials, but there is no significant difference between the modified and unmodified materials. These initial results indicate similar reactions between PET and PMMA surfaces. SEM micrographs are presented compared to PDMS dip coated Dacron in section 4.3.4. 4.1.2.2 MMA grafting on PTFE Surface modification of PTFE with MMA also presented challenges. Since no solvents were found to swell PTFE, it was unclear how the grafting behavior would proceed. However, the presence of fluorine atoms and a shift in the binding energy of the Cls peak due to these fluorine atoms allowed extensive analysis with XPS. PTFE did not swell in any solvents used. Therefore, various concentrations of acetone-MMA and DMSO-MMA solutions were used for grafting. Thus, results are presented as a function of concentration, solution, and radiation dose where appropriate, in an attempt to determine the best grafting conditions.
PAGE 119
102 Two types of substrate materials were studied in modification of PTFE, a skived PTFE film and expanded PTFE. These will be referred to as either Teflon (skived PTFE) or GORE-TE~ (ePTFE) in the following discussions. 4.1.2.2.1 Gravimetric analysis of MMA surface modified PTFE No swelling occurred in acetone, DMSO, MMA, or their solutions. Therefore, gravimetric analysis focused on weight changes following polymerization. Table 4.5 shows final weight changes following polymerization for Teflon modified with MMA in 100% MMA, acetone-MMA, and DMSO-MMA solutions, and tables 4. 6 and 4. 7 show the same infor1c1ation collected for GORE-TEX. The weight increases of Teflon are not considered significant to the extent of measuring differences in weight uptakes based on concentration or dose. The weight increases for GORE-TEX, however, are more significant, and a general trend showing higher weight increases for both higher MMA concentration and higher doses is observed. Weight increases are attributed to incorporation of MMA into the polymer structure itself, and not filling up the pores. The samples were analyzed by optical microscopy and SEM to determine if the pore structure remained in tact after modification (c.f. section 4.1.2.2.5.). Open pores were visible following surface modification, indicating weight increases were a result of PMMA incorporation.
PAGE 120
103 Table 4.5 Percent weight changes of Teflon~ PTFE irradiated MMA-acetone and MMA-DMSO solutions as a function of radiation dose and monomer concentration. Solution Concentration Radiation Dose Percent Weiqht Chanqe 1% MMA 99% DMSO 0.11 Mrad 0.7% 3% MMA 97% DMSO 0.11 Mrad 0 84% 5% MMA 95% DMSO 0.11 Mrad 0.14% 7% MMA 93% DMSO 0.11 Mrad 1.26% 10% MMA 90% DMSO 0.11 Mrad 0.28% 15% MMA 85% DMSO 0.11 Mrad -0.56% 20% MMA 80% DMSO 0.11 Mrad 0.56% 25% MMA 75% DMSO 0.11 Mrad 0.70% 50% MMA 50% DMSO 0.11 Mrad 1.26% 75% MMA 25% DMSO 0.11 Mrad 2.10% 25% MMA 75% Acetone 0.11 Mrad 0.33 50% MMA 50% Acetone 0.11 Mrad 0.31% 75% MMA 25% Acetone 0.11 Mrad 0.91 Table 4.6 Percent weight changes of GORE-TEX:' ePTFE irradiated in MMA-acetone and MMA-DMSO solutions to 0.11 Mrad as a function of monomer concentration. Solution % Weight Solution % Weight Concentration Increase Concentration Increase 1% MMA 99% DMSO 4.6% 1% MMA 99% Acetone 3.5% 3% MMA 97% DMSO 5.5% 3% MMA 97% Acetone 0.0% 5% MMA 95% DMSO -0.6% 5% MMA 95% Acetone 0.0% 7% MMA 93% DMSO 7.4% 7% MMA 93% Acetone 5.4% 10% MMA 90% DMSO 7.7% 10% MMA 90% Acetone 8.8% 15% MMA 85% DMSO 17.5% 15% MMA 85% Acetone 41.0% 20% MMA 80% DMSO 19.2% 20% MMA 80% Acetone 43.6% 25% MMA 75% DMSO 23.1% 25% MMA 75% Acetone 60.9%
PAGE 121
Table 4.7 104 Percent weight changes of GORE-TE~ ePTFE irradiated in 100% MMA as a function of dose. Solution Concentration Radiation Dose Percent Weiaht Increase 100% MMA 0.05 Mrad 3.9% 100% MMA 0.06 Mrad 15.21 100% MMA 0.07 Mrad 33.91 100% MMA 0.08 Mrad 50.21 100% MMA 0.09 Mrad 98.4% 100% MMA 0.10 Mrad 154.8% 100% MMA 0.11 Mrad 189.1% Figure 4.11 shows a plot of weight increase as a function of monomer concentration for GORE-TEX~ irradiated in MM.A-acetone and MMA-DMSO solutions. These data were expected to show a second order relationship as the MMA concentration is increased. The solubility parameters of both MM.A-acetone and MMA-DMSO solutions approached the solubility parameter of PTFE (6.2) with increasing MM.A. An increase in the diffusivity should occur as the solubility parameter of the solution approaches the substrate, as seen in the PET systems. However, no curve matched the scattered data. An overall trend of increasing weights attributed to PMMA polymer with increasing concentrations of MMA is apparent, and acetone seems to have a greater effect on weight increases than DMSO.
PAGE 122
105 60 o Gortex in MMA-DMSO o Gortex in MMA-Acetone 0 50 Cl) (I) 0 m 0 .. 40 u C .. .c C) a, 30 3: c G) 0 u .. 20 G) 0 a. 0 0 10 0 0 0 8 0 0 0 0 0 0 5 10 15 20 25 Monomer Concentration (% MMA In Solvent) Figure 4.11 Percent weight increase of GORE-TE~ (ePTFE) irradiated in MMA-acetone and MMA-DMSO to 0.11 Mrad as a function of monomer concentration. Figure 4.12 shows data for PTFE irradiated in 100% MMA plotted as a function of dose. The curve fit shown is a second order concave-up fit with an R2 value of 0.995, which corresponds well with diffusion dependent rate of polymerization for free radical polymerization as the reaction reaches the '' gel point. '' Classical theory for radiation initiated polymerization kinetics follows closely with kinetics of free radical polymerization with chemical initiators. The rate of
PAGE 123
106 polymerization is governed by the concentration of initiator in both cases, and in gamma radiation polymerization of vinyl monomers, the rate of polymerization is proportional to the square root of the dose rate (initiator), and has been found to hold true for the polymerization of 100% methyl methacrylate, as well (Chapiro, 1962). However, the introduction of solvents and a substrate (research here focuses on the reactions within the substrate) complicates the kinetics slightly. The termination reactions in gamma polymerization are largely diffusion controlled, especially during the majority of the propagation steps. This is attributed to the ''gel-effect'' in polymerization where higher conversions are reached, and the mobility of the molecules required for terntlnation is reduced with increasing viscosity (Chapiro, 1962 and Odian, 1981). As termination decreases, the monomer molecules have a higher diffusion rate than the growing polymer chains, and continue to react and propagate polymer. MacKay and Melville (1948) show a second-order reaction rate for systems which have reached a ''critical viscosity,'' and the reactions are largely diffusion controlled. The reactions of interest in this research are occurring within a substrate polymer. The substrate polymer has a much higher viscosity relative to the monomer and propagating polymer in solution, and therefore the diffusion of the propagating and terminating molecules control the overall rate of polymerization.
PAGE 124
107 200 R 2 = 0 995 Second Order Regression 175 150 Q) en (1S Q) 125 C: .c C) 100 Q) 3: C: Q) 75 (,) Q) Cl.. 50 25 0 05 0.06 0.07 0.08 0.09 0.10 0 11 Total Dose (Mrad) Figure 4.12 Percent weight increase of GORE-TEX~ ~(ePTFE) irradiated in 100% MMA monomer as a function of radiation dose. Another explanation of the second order relationship becomes evident when the sequence of reactions is considered. No monomer or solvent initially diffused into Teflon~ or GORE-TE~ prior to exposing the samples to gamma radiation. However, upon initial exposure, free radicals are initiated on the substrate and monomer, and polymerization at the
PAGE 125
108 surface begins. The polymerizing solution is now more soluble in this surface graft of PMMA-g-PTFE than it was in the PTFE alone, as the solubility parameter of the substrate begins to change from 6.2 for PTFE toward a higher value of 9.0 to 9.5 for PMMA. The permeability of the solvent mixture in the substrate increases, as shown earlier (c.f. section 4.1.1.2), as the cohesive energy density of the substrate approaches that of the solution with increased polymerization. The polymer then begins to expand with increased swelling, and more free volume is introduced into the system. Free volume increases give rise to increases in the permeability of the substrate by introducing a higher number of voids available for solvent molecules to enter. The swollen region then polymerizes as an IPN, further altering the CED of the substrate. This process continues until the sample is removed from the radiation source. As more PMMA is grafted and incorporated as an IPN, the solubility parameter and free volume continue to increase, thereby increasing the diffusion rate. The initial reaction at the solvent/monomer diffusion front is believed to be true graft polymerization, and the subsequent reactions within the grafted regions continue as both grafting and homopolymerization providing an IPN. The reaction proceeds as a second order function. The acceleration is attributed to (i) the diffusion controlled propagation reaction, (ii) the increased diffusivity of the substrate induced by compositional changes (solubility parameter becomes closer to
PAGE 126
109 that of the solution) and, (iii) the increase in free volume from increased swelling of the solution in the graft-IPN. The reactions in the GORE-TEX are more pronounced than in the Teflon because GORE-TEXhas a lower modulus and higher elasticity as a result of the expanded structure. This is especially evident in the 100% MMA polymerized samples, where the polymerization seems to ''run away'' exceeding 170% weight increases, and the same reactions in Teflon show insignificant weight increases. The Teflon would most likely have similar results if the overall modulus was lower, allowing swelling deformation to occur. 4.1.2.2.2 Contact angle analysis of MMA surface modified PTFE Contact angle measurements were un-measurable on the GORE-TEXbecause of the porosity. When the samples were placed in water, the contact between the air bubble and the surface was mainly contact between the air bubble and the air trapped in the surface. When the samples were degassed to remove the air, pores became water filled, and the air bubble was resting on mainly water instead of the GORE-TEX. Contact angle values for the Teflon, however, were easily obtained. Table 4.8 and figure 4.13 show average contact angle values as a function of solution concentration and radiation dose for Teflon. As either the radiation dose (for 100% MMA) is increased, or as the MMA concentration is increased, the contact angle approaches a value of that for PMMA, indicating the presence of increasing amounts of PMMA on the Teflon surface. The data plotted in figure 4.10 show
PAGE 127
110 the same concave-up second order curve fits indicating the same trend of a diffusion controlled process where the diffusion rate (based on substrate permeability from increasingly ideal solubility parameter and free volume) and thus polymerization rate increase during the process. There are, however, boundary conditions established for contact angles from the value of PTFE (110) to that of PMMA (55). The values measured which go below that of PMMA are attributed to slight surface porosity of the Teflon surface. When the samples are hydrated, the pores are filled with water, giving a value lower than the theoretical value of PMMA. The possibility of oxidative degradation occurring during exposure to gamma radiation increasing the hydrophilicity of the Teflon surface was eliminated with XPS studies (c.f. section 4.1.2.3.4.).
PAGE 128
111 Table 4.8 Contact angle of Teflon PTFE irradiated in MMA acetone and MMA-DMSO solutions as a function of radiation dose and solution concentration. Solution Dose Averaae Contact Anale TEFLON PTFE CONTROL NONE 105 100% MMA 0.11 Mrad 56 100% MMA 0.13 Mrad 47 100% MMA 0.17 Mrad 38 75% MMA 25% DMSO 0.11 Mrad 60 50% MMA 50% DMSO 0.11 Mrad 61 25% MMA 75% DMSO 0.11 Mrad 76 20% MMA 80% DMSO 0.11 Mrad 64 15% MMA 85% DMSO 0.11 Mrad 57 10% MMA 90% DMSO 0.11 Mrad 55 7% MMA 93% DMSO 0.11 Mrad 67 5% MMA 96% DMSO 0.11 Mrad 66 3% MMA 97% DMSO 0.11 Mrad 76 1% MMA 99% DMSO 0.11 Mrad 77 75% MMA 25% Acetone 0.11 Mrad 54 50% MMA 50% Acetone 0.11 Mrad 46 25% MMA 75% Acetone 0.11 Mrad 65 20% MMA 80% Acetone 0.11 Mrad 64 15% MMA 85% Acetone 0.11 Mrad 78 10% MMA 90% Acetone 0.11 Mrad 77 7% MMA 93% Acetone 0.11 Mrad 84 5% MMA 96% Acetone 0.11 Mrad 82 3% MMA 97% Acetone 0.11 Mrad 84 1% MMA 99% Acetone 0.11 Mrad 103
PAGE 129
112 110 --.------------------------, 100 90 Cl) C) C 80 < () ctS C 70 0 (.) Cl) C) ctS 60 Cl) > < 50 40 30 []El [] Irradiated in Acetone to 0 11 Mrad: R 2 = 0 95 [] Irradiated in DMSO to 0.11 Mrad: R 2 = 0 78 [] [] [] 0 5 10 15 20 25 30 35 40 45 50 Figure 4.13 Solution Concentration (% MMA in Solvent) Contact angle of Teflon (PTFE) irradiated MMA-acetone and MMA-DMSO solutions to 0.11 as a function of concentration. 4.1.2.2.3 FT-IR analysis of MMA surface modified PTFE in Mrad FT-IR/ATR clearly identified the presence of PMMA on the modified surface of GORE-TEX. The absorbence spectra for PTFE is fairly simple, having peaks corresponding to c-c bonds, C-F bonds, and some oxidized regions showing C-0 bonds. PMMA has a strong absorbence as a result of the
PAGE 130
113 c a1 : b o nyl bond, which appears in the 1720-1740 cm-1 region, wh er e PTFE has no significant peaks. Figure 4.14 shows FT-IR absorbence spectra for PMMA, GORE-TEX, and MMA-surface modified-GORE-TEX, clearly showing the carbonyl absorption pe a k at 1730 cm1 indicating the presence of PMMA. A b a 0 r b 8 n C 8 o. 1 2 F R. I E X Co n tro l I . 0. 1 1 0 1 0 0 0 9 : . 0 0 8 ' 0 07 1 0 06 ---, -MMA Modified GO RE( 15% MMA accl o nc l rrad 1a1 ed 100 1 1 M111d1 0 07 0 06 0 0 5 0. 04 0 03 1 0 0 8 t I t: ::::::::: ::;,,,c: ===:; = ==>-,c=::::;:::==;====r=~-:-:-;--:::::. ~ r r 2500 2000 Figure 4.14 FT-IR/ATR absorbence spectra of PMMA GORE-TEx, and GORE-TEx surface modified with MMA.
PAGE 131
114 4.1.2.2.4 XPS analysis of surface MMA modified PTFE The presence of the various C-F bonds and ubond neighbors" (bonds surrounding the C-C and C-F bonds) cause shifts in the carbon binding energies when analyzed with XPS. The combination of these binding energy shifts, and the presence of fluorine in PTFE (and its absence in PMMA) allowed concentration differences to be extensively studied with XPS. Table 4.9 shows the atomic concentration of carbon (Cls), oxygen (Ols), and fluorine (Fls) obtained from unmodified PMMA, GORE-TEX, and Teflon~, and figure 4.15 shows the Cls binding energy peaks for these materials. As seen in figure 4.15, the shape of the Cls peaks indicate the presence of various atomic bonds specific to both PMMA and PTFE. The values reported in table 4.9 are averages of measurements of 3 to 5 different samples. These average values were used as the base concentrations for 100% PMMA, 100% GORE-TEX~, and 100% Teflon~. These base concentrations were used to calculate the relative concentrations of PMMA and PTFE in modified samples using a rule of mixtures formula. A ratio of 100% PMMA and 100% PTFE was calculated to match the concentration of a modified sample, and the calculated ratio was used to report the surface concentration of PMMA on a modified PTFE sample. Figure 4.16 illustrates an example of these calculations.
PAGE 132
Table 4.9 Material PMMA Teflon GORE-TEX en C: ::, >,. ca .c < >,. en C: Q,) C: 115 Average carbon, oxygen, and fluorine concentrations for Teflon, GORE-TEX, and PMMA. Atomic% C Atomic% 0 Atomic IF 78.21% 29.21% o.oo 54.58% 7.27% 38.15% 55.08% 6.22% 38.71% Teflon GORE-Tex PMMA 300 295 290 285 280 275 Binding Energy ( eV) Figure 4.15 XPS Cls spectra for Teflon, GORE-TEX, and PMMA showing differences in the chemical shifts due to various carbon bonds.
PAGE 133
116 If the percent concentrations of carbon, oxygen, and fluorine for Teflon, PMMA, and the sample of interest are: Teflon: PMMA: Sample: 54.58% c, 78.21% C, 60.32% C, 7.27% O, 38.15% F 29.79% O, 00.00% F 8.00% O, 31.68% F Using the definitions: SC= The Carbon fraction for a sample analyzed with XPS, so= the Oxygen fraction, SF= the Fluorine fraction, X = Fraction of PMMA control, Y = Fraction of PTFE control, and Y = 1 X X(PMMAC) + Y(PTFEC) = SC X(PMMAO) + Y(PTFEO) = SO X(PMMAF) + Y(PTFEF) = SF X(0.7821) + Y(0.5458) 0.6032 X(0.2979) + Y(0.0727) 0.0800 X(0.0000) + Y(0.3815) 0.3168 And: X 0.170 y 0.830 The sample analyzed has a surface concentration of 17 atomic percent PMMA and 83 atomic percent PTFE. Figure 4.16 Sample calculation to determine the surface concentration of PMMA and PTFE on surface modified Teflon and GORE-TE~. The surface concentration of PMMA on both GORE-TEX~ and Teflon was evaluated as functions of dose and concentration in acetone and DMSO solutions with XPS. This technique (Figure 4.14) was used to determine the surface concentration of modified Teflon and GORE-TEX samples, and is based on the assumption that increases in carbon and oxygen and decreases in fluorine concentrations are due to the incorporation of PMMA in the substrate surface. Other possible changes in these concentrations could arise from
PAGE 134
117 degradation or oxidation of the PTFE substrate during exposure to gamma radiation in the presence of solvents. Samples of Teflon and GORE-TEX were exposed to 0.13 Mrad in air, degassed acetone, and degassed DMSO, and compositions of these samples matched compositions of the unmodified, control samples used for concentration calculations. The concentration changes in carbon, oxygen, and fluorine are attributed to the presence of PMMA on the surface, and are representative of the first 50A of the surface (depth of analysis of XPS used in this research). The surface concentration of PMMA on GORE-TEX and Teflon was evaluated with XPS as a function of solvent, monomer concentration, and dose. Figure 4.17 shows a plot of PMMA surface concentration of GORE-TEX and Teflon as a function of radiation dose for samples polymerized in 100% MMA, and figure 4.18 shows a plot of PMMA surface concentration of GORE-TE~ and Teflon as a function of solution concentration for samples polymerized to 0.11 Mrad. These data were expected to follow the concave-up second order curve corresponding to the increasing diffusivity in the diffusion controlled process. The data presented in figure 4.17 show a general trend of increasing PMMA surface concentrations with increasing in MMA monomer, but there is too much scatter in the data to match a curve of any order. The data presented in figure 4.18, for surface PMMA concentration of PTFE polymerized with 100% MMA, however, follows the expected curve.
PAGE 135
100 90 ?ft. (.) E 0 80 < 70 C: 0 -; 60 C: Q) (.) C: 0 (.) Q) (.) cu :::, (f) 50 40 30 118 O Gortex Polymerized in MMA-Acetone 0 Gortex Polymerized in MMA-DMSO Teflon Polymerized in MMA Acetone Teflon Polymerized in MMA DMSO 0 0 0 0 0 < :E 20 8 0 0 0 :E a.. 0 0 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 Monomer Concentration (% MMA In solvent) Figure 4.17 Surface concentration of PMMA on GORE-TE~ and Teflon~ irradiated in MMA-acetone and MMA-DMSO solutions as a function of monomer concentration.
PAGE 136
119 100 -----------------------'#. 90 < 80 a. 0 70 C 0 cu 60 C (1) (.) 50 C 0 (.) Cl> 40 (.) cu ::, 30 (J) 20 10 0 0 Gortex Irradiated in 100% MMA : R2 = 0 98 second order regression Teflon Irradiated in 1000/o MMA : R2 = ~5. second order regression I I Q I / , / 0 , / I , , I I I I I I I I I I ,b 0 00 0 02 0.04 0.06 0.08 0.10 0 12 0.14 0 16 0.18 Total Radiation Dose (Mrad) Figure 4.18 Surface concentration of PMMA on GORE-TE~ and Teflon~ irradiated in 100% MMA to 0.11 Mrad as a function of radiation dose. Error bars represent standard deviation of the mean of 3 samples. Figure 4.19 shows Cls spectra for increasing modifications of PMMA. The chemical shift for fluorine atoms gradually disappears with increasing PMMA concentrations, until the final curve (top most curve) shows the spectra matching that of 100% PMMA.
PAGE 137
120 PMMA Control PTFE 75/4 MMA-acetone en C: :::, PTFE 25% MMA-acetone as .0 PTFE 20/4 MMA-actone as PTFE 10% MMA-acetone en C: C1) C: PTFE 5% MMA-acetone PTFE 1 /4 MMA-acetone PTFE Control 300 295 290 285 280 275 Binding Energy ( eV) Figure 4.19 XPS Cls spectra for Teflon modified with MMA acetone solutions showing changes in the spectra with increasing surface concentrations of PMMA as a function of solution concentration. 4.1.2.2.5 SEM analysis of surface MMA modified PTFE The overall appearance of Teflon PTFE was unchanged as viewed with SEM following surface modification. The GORE TE~ ePTFE, however, was visibly changed from the deformation caused by the surface modification. Figures 4.20, 4.21 4.22 and 4.23 show GORE-TEX before (4.20) and after surface
PAGE 138
121 modifications (4.21-4.23). The micrographs in figure 4.21 show a sample modified with 3% MMA-DMSO, and no excessive swelling of the GORE-TEX~ structure. This is typical of the samples which have been modified with less than 10% MMA-DMSO, and they do not appear different from the controls under SEM. Figure 4.22 shows a GORE-TEX~ sample which has excessive swelling of the substrate following modification with 20% MMA-DMSO. Figure 4.23 shows a sample modified with 3% MMA acetone, and the ''broken nodule'' structure shown is typical of samples modified with acetone solutions. This is believed to be caused by excessive swelling of the PMMA-g-PTFE regions by the acetone during polymerization and increased diffusion, but it is not clear why the structure breaks in the acetone samples and not the DMSO samples. A polymerized and washed sample was evaluated for swelling in acetone, DMSO, and MMA acetone and MMA-DMSO solutions. It was found that the acetone-MMA solutions had higher weight increases from swelling than did the others. This supports the mechanism of polymerization, which is described here as diffusion controlled, and is allowed to proceed excessively by swelling the substrate in monomer and solvent during the reaction. It also proves the modified substrate has a solubility parameter which allows the diffusion of the solvent-monomer solutions studied. The swelling of modified samples following washing was not evaluated extensively, and is left as future work.
PAGE 139
122 Figure 4.20 SEM micrographs of unmodified GORE-TEX.
PAGE 140
. 12 ] Figure 4.21 SEM micr ographs of GORE-TEX irradiated in 3% MMA-DMSO to 0.11 Mrad.
PAGE 141
124 Figure 4.22 SEM micrographs of GORE-TEX irradiated in 20% MMA-DMSO to O .11 Mrad. Deforntation and stretching of nodule structure is visible.
PAGE 142
125 Figure 4.23 SEM micrograph of GORE-TEX irradiated in 3% MMA acetone to 0.11 Mrad. Breaking of the nodule structure is typical for samples modified in acetone solutions.
PAGE 143
126 4. 1. 2. 2. 6 Ex vivo analysis of MMA modified pr1 1 tc;e; In vivo graft implants are not studied until success has been shown with ex vivo AV shunt analysis. The conditions providing surface modification of PTFE with PMMA were studied in this research. Modification conditions to be studied in ex vivo AV shunts were determined only after extensive work, and at the time this manuscript was prepared, AV shunt analysis of MMA surface modified GORE-TEXhad not been completed. GORE-TEX samples modified with 1 to 5% MMA-DMSO solutions polymerized to 0.11 Mrad are recommended for ex vivo AV shunt studies in future studies. 4.1.2.3 MM.A grafting on PDMS The observations made in the swelling of PDMS with MM.A and DMSO indicated rapid diffusion of monomer and solutions in to the material. Modification of PDMS with PMMA became a bulk modification process because of this rapid diffusion. Although the focus of this research was surface modification, the bulk polymerization of MM.A in PDMS was studied, and interesting results were obtained. 4.1.2.3.1 Gravimetric analysis of MM.A surface modified PDMS The maximum swelling of PDMS in MM.A solutions was reached within the first hour of exposure (5% weight increase for 10% MMA-DMSO solutions), indicating diffusion completely into the bulk. Upon polymerization, the total weight increases continued beyond the maximum non-irradiated swelling values, as expected based on the results discussed previously for PTFE. The rapid swelling and elasticity of
PAGE 144
127 PDMS provided another model for the mechanism of polymerization of an elastic system as first demonstrated with PTFE. Polymerization beyond initial swelling was expected because of the low modulus and high elasticity of PDMS and swelling of the polymerizing molecules within the substrate. Second order curve fits were anticipated as functions of both solution concentration and radiation dose. Table 4.10 contains the data for weight increases of P DMS following gamma polymerization of MMA. The weight increase was measured immediately following polymerization (prior to removal of residual solvent and monomer} and following washing and drying of the sample. Table 4.10 Weight increase of PDMS solutions as a function and radiation dose Solution Concentration Weight PolYlllerized to 0.13 Mrad) Increase 1% MMA in DMSO 1.3% 3% MMA in DMSO 20.7% 5% MMA in DMSO 10.21 7% MMA in DMSO 26.9% 9% MMA in DMSO 33.9% ( Pol ;,T11er i zed to 0.10 Mrad) 1% MMA in DMSO -0.3% 3% MMA in DMSO 4.1% 5% MMA in DMSO 12.3% 7% MMA 1.n DMSO 17.9% 10% MMA in DMSO 29.0% irradiated of monomer in MMA-DMSO concentration Radiation Dose Weight 10% MMA-DMSO Increase 0.050 4.7% 0.051 2.98% 0.060 10.21 0.061 9.9% 0.070 10.8% 0.071 11.7% 0.080 14.3% 0.081 14.8% 0.090 17.5% 0.093 20.8% 0.096 22.8%
PAGE 145
128 Figure 4.24 shows the weight uptakes of PDMS polymerized in 10% MMA-DMSO solutions as a function of radiation dose, showing the same second order relationship discussed previously. The wet swell weight shows an identical trend to the total (dry) weight increase, indicating a consistent amount of residual solvent and monomer within the samples. The weight increase as a function of tjme with no exposure to gamma radiation is also shown to contrast the polymerization, where a maximum swelling of 5% is reached in the first hour. 40 35 30 Cl) en cu Cl) 25 (.) C: .s::::. 20 C) Cl) 15 C: Cl) (.) Cl) 10 a. 5 0 Weight Increase prior to residual solvent removal : Second order R 2 = 96 Final Weight Increase-samples dried and washed : Second order R 2 =.98 0 No Gamma Exposure : Dose refers to equivalent swelling time -80 00 0 02 0 04 0 06 0 08 0 10 0 12 Dose (Mrad) Figure 4.24 Percent weight increase of PDMS irradiated in 10% MMA-DMSO as a function of radiation dose. Swelling only (no polymerization) is shown for comparison.
PAGE 146
129 By following the proposed sequence of reaction, the second order relationship may also be explained as changes in the solubility parameter of the substrate during polymerization. The reaction is believed to follow in a similar manner to that discussed earlier for the PTFE systems. Although the PDMS has a fair amount of solvent uptake during the first hour of exposure, unlike PTFE which did not swell at all prior to irradiation, the weight increase continues with further exposure. The monomer which initially diffuses into the PDMS substrate begins to polymerize upon exposure, creating PMMA rich regions within the substrate. The exact nature of this IPN is not clear at this time, but increases in opacity of the samples with increasing dose indicate some phase separation. The initial polymerization in this case is a combination of graft polymerization and homopolymerization, because monomer is initially present within the substrate, leading to an IPN. As more PMMA polymerizes, the solubility parameter of the silicone (7.3) increases toward a value for PMMA (8.8), and the rate of diffusion increases as the monomer solution becomes more soluble in the PMMA-graft-IPN-PDMS matrix (calculated solubility parameter of 11.7). The elastomeric properties of PDMS allows expansion of the substrate upon swelling similar to that which occurred in the GORE-TEX~ system, providing increased free volume and increased diffusion. This sequence of events is believed to contribute to increases in diffusion during the reaction, which gives
PAGE 147
130 rise to a second order relationship. Proof of the initial graft polymerization is not as readily available for PDMS because there is initial swelling, and further increases could be a result of the presence of the IPN in the substrate. Graft polymerization to some extent is, however, believed to occur. There is one slight contradiction to this theory which must be noted, however. As shown in figure 4.24, the amount of monomer solution contained within the substrate at any given time is the difference between the dry and wet weight increase curves. This swelling amount appears to be consistent, indicating there is not an increase in the diffusivity, or total swelling, as more polymerization takes place. This further indicates the polymerization medium in which the monomer solution is swelling (PMMA-g-IPN-PDMS) has a fairly constant composition. If phase separation the PMMA occurs during polymerization, the solubility parameter of the polymerizing PMMA within the substrate remains fairly constant, and increased diffusivity does not occur. This observation suggests the second order relationship is based on a diffusion controlled process during the ngel effect" rather than a strong dependence on changing solubility parameters. The weight increases for PDMS polymerized in MMA-DMSO solutions as a function of monomer concentration are show in figure 4.25, and these results are similar to those discussed previously for other substrates. These data also follow a similar concave-up, second order curve. Again, as the
PAGE 148
131 concentration of MMA is increased, the solubility parameter of the solution goes from 12 (DMSO) toward 8.8 (MMA). The solubility parameter of PDMS is 7.3, and increases in concentrations correspond to an approach toward a maximum solubility parameter with increasing diffusivity. The shape of this curve matches those presented in figures 4.5 and 4.6, showing increases in solubility and diffusivity as the solubility parameter of the solvent approaches that of the polymer. 30 25 ?ft. Q,) 20 Cl) ca Q,) (.) C: 15 .c C) Q,) 3: C: 10 Q,) (.) Q,) a.. 5 PDMS Irradiated in MMA-DMSO Solutions : R2=0 996 / / / / / / / / / / / / / / / / / <:I / / I I / 0 1 2 3 4 5 6 7 8 9 10 Solution Concentration (% MMA In DMSO) Figure 4.25 Weight increase of PDMS irradiated in MMA-DMSO solutions as a function of solution concentration.
PAGE 149
132 4.1.2.3.2 Contact angle analysis of MMA surface modified PDMS Contact angle of PDMS were evaluated as a function of PMMA modifications. The contact angle for unmodified PDMS is 80. The results of measurements made on modified samples are displayed in table 4.11 Table 4.11 Contact angles of PDMS irradiated in MMA-DMSO solutions as a function of radiation dose and solution concentration. Solution Concentration Contact Radiation Dose (10% Contact (0.13 Mrad Total Dose) Anqle MMA-DMSO) Anqle PDMS Control 80 PDMS Control 80 3% MMA-DMSO 75 0.04 Mrad 84 5% MMA-DMSO 67 0.05 Mrad 84 7% MMA-DMSO 65 0.06 Mrad 71 9% MMA-DMSO 69 0.07 Mrad 80 10% MMA-DMSO 61 0.08 Mrad 78 PMMA Control 55 0.09 Mrad 72 0.10 Mrad 58 PMMA Control 55 Figures 4.26 and 4.27 show contact angle data as functions of radiation dose and solution concentration, respectively. The plot for changes in contact angle as a function of monomer concentration (figure 4.26) shows a second order relationships (with the exception of one sample, 9% MMA-DMSO), whereas the plot for the contact angle as a function of radiation dose (figure 4.27) does not follow any explainable relationship.
PAGE 150
86 84 82 80 a> 78 76 < 74 (,) 72 C c3 70 68 66 64 62 133 <> <> <> <> <> 60 _._..,. ________________ ._...........,,.......__. __________ __. 0 04 0 05 0.06 0 07 0 08 0.09 0.10 Dose (Mrad) Figure 4.26 Contact angle of PDMS irradiated in 10% MMA-DMSO as a function of radiation dose. 84 82 80 78 76 C) C < 74 72 70 (.) 68 66 64 62 60 -t-T"TT"'m"TT"'m-rrT"T""l~T"'T"'lr-r-r"T"T""'ll'"'l""T'"'T""T""f""'l"""r"'T"'T""r-T"'9 ......... ..,.._-......'P"W""'I~......, 0 1 2 3 4 5 6 7 8 9 10 Solution Concentration (% MMA In DMSO) Figure 4.27 Contact angle of PDMS irradiated in MMA-DMSO solutions as a function of monomer concentration polymerized to 0.10 Mrad.
PAGE 151
134 4.1.2.3.3 FT-IR analysis of MMA surface modified PDMS Modification of PDMS with PMMA was a bulk modification process as opposed to a surface modification process. FT IR/ATR did not detect the presence of PMMA on the surface. This is attributed to the low surface volume concentration of PMMA, and transmission FT-IR was used to evaluate PMMA modified PDMS. Figure 4.28 shows transmission FT-IR spectra for unmodified PDMS, PMMA, and MMA modified PDMS. The carbonyl is clearly visible in the modified sample indicating the presence of PMMA. 4.1.2.3.4 XPS analysis of MMA surface modified PDMS PDMS has three elements which may be detected by XPS, carbon (Cls peak), oxygen (Ols peak), and silicon (Si2p peak). With only carbon and oxygen being detected in PMMA, the surface concentration of PMMA on PDMS was determined using calculations similar to those presented for Teflon~ and GORE-TE.x. The base concentration for PDMS was found to be 50.25% carbon, 22.60% oxygen, and 27.15% silicon. These values were compared to modified PDMS, and used to calculate the surface concentrations of PMMA on the PDMS. Figure 4.29 shows surface concentrations of PMMA on modified PDMS as a function of solution concentration (polymerized to 0.11 Mrad) and as a function of total radiation dose (10% MMA-DMSO). A second order relationship consistent with data presented thus far, is observed for this data, supporting the theory of increasing diffusivity with monomer concentration increases. Data collected for surface
PAGE 152
135 5 ;f'LJMS Con tn (S h1n l :.Uu C u ~ Lo< 51 J s J O. 2 S 20 I 5 1 0 0 5 I N I I l M MA M cx11 f1 cd P D M S ( 10% MA DM SO I rradiated 1 0 0 08 M raJ 1 2 5 A b a 2 O 0 ( 0 l 5 8 n C l O 8 1 0 0 8 ' 0 6 0 4 0 2 40CX) 3500 3000 Figure 4.28 FT-IR transmission PDMS modified with irradiated to 0.11 I I I ~n I I I I --rT f T 1 t J f f" 2500 2000 1 500 spectra for PDMS, PMMA, and PMMA (10% MMA-DMSO solution Mrad). c o ncentration of PMMA as a function of radiation dose was limited, and no consistent trend was o bserved. The data is presented in figure 4.30 only to be consistent in information presented. The inconsistent trends of some of the PDMS s ample sets is attributed to inconsistent degassing of the
PAGE 153
136 solutions prior to irradiation. Silicone is highly permeable to oxygen, and these solutions were bubbled with argon instead of vacuum degassed to avoid monomer evaporation. The residual oxygen within the silicone substrates therefore becomes important, and may inhibit polymerization. 18 16 C 14 0 cu 12 C Cl) (J C 0 10 (..) Cl) (J cu 8 ::, en < 6 :IE :IE a. 4 2 0 1 R 2 for second order curve fit= 0.97 2 3 4 5 Concentration (% , , , , 6 7 8 MMA in DMSO) 9 , , 10 Figure 4.29 Surface concentration of PMMA on PDMS irradiated in MMA-DMSO solutions to 0.11 Mrad as a function of solution concentration.
PAGE 154
C 0 as C G) (.) C 0 (.) G) (.) as ,._ ::, en < a. 10 9 8 7 6 5 4 3 2 1 137 0.04 0.05 0.06 0.07 0.08 0.09 0.10 Dose (Mrad) Figure 4.30 Surface concentration of PMMA on PDMS as a function of radiation dose polymerized to 0.11 Mrad in 10% MMA-DMSO. 4.1.2.3.5 DMS analysis of MMA surface modified PDMS The increase in modulus of PDMS following polymerization was immediately obvious during sample handling. The extent of this change, and its function of dose and polymerization solution concentration were studied by using DMS. Figures 4. 31 and 4. 32 show E' ( storage modulus) and E'' (loss modulus) at 1 Hz for samples polymerized with 10% MMA DMSO as a function of radiation dose, and figures 4.33 and 4.34 show similar information for samples polymerized to 0.10 Mrad in MMA-DMSO solutions as a function of monomer
PAGE 155
138 concentration. The storage modulus is a measure of elastic energy stored by the sample, whereas the loss modulus is a measure of the energy lost as heat during cyclic loading. The weight increase of the sample is shown for each curve in the legend. There is one data set which does not follow the trend of increasing modulus with increasing MMA concentrations, but the weight increase of the sample places it in the proper order. This irregularity is again believed to be caused by inconsistent degassing of the samples as discussed above. Overall, however, increasing amounts of PMMA within PDMS substrate increases the modulus as expected. An attempt was also made to use the glass transition temperature obtained from the DMS data to determine the extent of grafting versus IPN polymerization. A copolymer, which is considered the most intimate type of blend, will have a merging of glass transitions, showing a single peak. As the degree of mixing is decreased, the two individual glass transitions become visible, moving to the value of each homopolymer for a phase separated system. The Tan 6 plots for these samples are presented in figures 4.35 and 4.36. Tan 6 is E'/E" (the loss tangent), and 6 is the angle between in and out of phase components of the cyclic loading. Ma~ima of E" and tan 6correspond to high loss transitions such as the Tg (Sperling, 1986). Looking at both the tan 6 and the E" data, it is clear there is phase separation in the system. This supports earlier observations concerning the polymerization of PMMA rich regions within the substrate,
PAGE 156
139 where the diffusivity was not increased with increasing radiation dose because swelling and polymerization was taking place in PMMA regions. Increases in the loss peaks for the Tg of PMMA consistently increase with increasing amounts of PMMA, and the Tg remains at approximately the same temperature throughout all samples. 1e+10 1e+09 s 0UJ (/) =, =, 1 e + 0 8 "C 0 Q) C) ca ... 0 en 1 e+07 1 50 1 00 -5 0 ( 5 ) 9% MMA DM S O so lut i on : W ei ght I n c rea se= 33 9% ( 4 ) 7% MMA-DMSO solution : W eig ht Incr ease = 26 9% (2) 3% MMA-DM S O s ol ut io n : We i ght In c r e a se= 20 7/4 (3) 5% MMA-DM SO so l u t io n : We ig ht In cre a se= 1 0 2% (1) 1 % MMA DM SO sol ut io n : W e i gh t I n crease = 1 3% (C) Shi n -Ets u Silicone : Unmodified 5 4 2 3 C 0 50 1 00 150 200 250 Temperature c c) Fi g ur e 4. 3 1 1 H z fr e qu e n cy s to r age m od ul us ( E ) of PD M S irr adiat e d in MMA-DM SO to 0.11 M rad as a fu n c t i on of m o n o m e r c o n ce n tr a tio n.
PAGE 157
1e+09 1e+08 ca a.. s w 1e+07 en :::, :::, "C 0 ::E en en 0 1e+06 _,J 1e+05 -150 3 -100 -50 140 (5) 9%MMA-DMSO solution : Weight Increase= 33 9% (4) 7% MMA-DMSO solution : Weight Increase= 26 9% (2) 3% MMA -DMSO solution : Weight Increase= 20 7% (3) 5% MMA-DMSO solution : Weight Increase= 10 2% ( 1) 1 % MMA-DMSO solution : Weight Increase = 1 3% (C) Shin-Etsu Silicone : Unmodified 5 C 0 50 100 150 200 250 Temperature cc) Figure 4.32 1 Hz frequency loss modulus (E") of PDMS irradiated in MMA-DMSO to 0.11 Mrad as a function of monomer concentration.
PAGE 158
1e+10 1e+09 7 C'O 0.. en ::, 1e+08 ::, -c::::, 0 :E Q) 0, C'O 0 en 1e+07 w 1e+06 -150 -100 141 (11) Polymerized to 0 09 Mrad : Weight Increase= 17 5% (10) Polymerized to 0 08 Mrad : Weight Increase= 14 3% 7 7 7 7 1 1 7 7 1 1 (9) Polymerized to 0 07 Mrad : Weight Increase= 11 7% (8) Polymerized to 0 06 Mrad : Weight Increase= 9 9% (7) Polymer ize d to 0 05 Mrad : Weight Increase = 3 0% (C) Shin-Etsu Silicone : Unmodified 11 1 7 9 C -50 0 50 100 150 200 250 Temperature cc) Figure 4.33 1 Hz frequ ency storage modulus (E') for PDMS modified with 10% MMA-DMSO solutions as a function of radiation dose.
PAGE 159
1e+09 ctS 0.. en :::, 1e+07 :::, "O 0 en en 0 ...J UJ 1e+06 1e+05 -150 ..., '.. .. 100 \ 142 (11) Polymerized to 0 09 Mrad : We ig ht Increase= 17 5% (10) Polymerized to 0 08 Mrad : Weight Increase= 14 '3% (9) Polymerized to 0 07 Mrad : Weight Increase= 11 7% (8) Polymerized to 0.06 Mrad : We ig ht Increase= 9 9% (7) Polymerized to 0 05 Mrad : We igh t Increase= 3 0% (C) Shin-Etsu Silicone : Unmodified ,.. ... ,.. .... ... ... ... ... -50 0 50 100 Temperature (.C) 150 200 250 Figure 4.34 1 Hz frequency loss modulus (E") of PDMS irradiated in 10% MMA-DMSO solutions as a function of radiation dose.
PAGE 160
0 5 0 4 0 3 0 0 C: ro 0
PAGE 161
0.50 0 45 0.40 0 35 0 30 C C'O 0 25 C t; e, 0.20 1 c, c, I C 1 0 15 1 .p I 1 1 0.10 c.P b c~ -/; ti C 0 05 C -150 -100 144 (11) Polymerized to 0 09 Mrad (10) Polymerized to 0 08 Mrad (9) Polymerized to 0 07 Mrad (8) Polymerized to 0 06 Mrad t (7) Polymerized to 0 05 Mrad (C) Shin-Etsu Silicone -00 0 50 100 150 200 250 Temperature ("C) Figure 4.36 1 Hz frequency tan o of PDMS irradiated in 10% MMA-DMSO solutions as a function of radiation dose.
PAGE 162
145 The storage modulus (E') data presented above plotted as a function of weight percent PMMA at 37C shown in figure 4.37 results in a linear relationship. The direct increase in modulus with PMMA indicates there is little mixing, and phase separation occurred during polymerization. 1e+08 cu CL -::5e+07 w (/) ::, -; 3e+07 "C 0 :E Q) C) cu 0 ;E 1e+07 5e+06 3e+06 R 2 = 0 99 Linear / / / 0 / 0 / '/ / / / / / / 9 / @' 0 / 0/ /0 / / / / / /0 0 0 5 10 15 20 25 30 35 Percent Weight Increase from PMMA Figure 4.37 37C, 1 Hz frequency storage modulus of PDMS irradiated in MMA-DMSO solutions as a function of weight percent PMMA. 4.1.2.3.6 SEM analysis of MMA surface modified PDMS The surface of PDMS usually mirrors the mold, or in this case, the glass plate, in which it is cured. The surface as
PAGE 163
146 viewed with SEM before and after surface modification was not noticeably different. 4.1.2.3.6 Ex vivo analysis of modified PDMS In vivo graft jmplants are not studied until success has been shown with ex vivo AV shunt analysis. The conditions providing surface modification of PDMS with PMMA were studied in this research. Modification conditions to be studied in ex vivo AV shunts were determined only after extensive work, and at the time this manuscript was prepared, AV shunt analysis of MMA surface modified PDMS had not been completed. PDMS samples modified with 1 to 5% MMA-DMSO solutions polymerized to 0.11 Mrad are recommended for ex vivo AV shunt studies in future work. 4.2 Gamma Radiation Induced Polymerization of Sulfoethyl Methacrylate on PET Surface modification of Dacron with SEMA was not studied as extensively as the other monomers and substrates. The results for surface modification of SEMA presented here are those of a feasibility study for the radiation grafting of non-salt sulfated monomers onto Dacron~. 4.2.1 Swelling and Gravimetric analysis of PET Surface Modified with SEMA Presoaks of SEMA-DMSO and SEMA-MMA-DMSO solutions at 60C for 24 hours provided a weight increases for Dacron~ which depended on the solution concentration. Presoaks in 40% MMA
PAGE 164
147 70% DMSO (weight increase of 8 to 10%) were also used prior to modification. There were no significant weight changes between initial sample weights and weights following polymerization to 0.11 Mrad, with or without presoaking. This is attributed to a low concentration of grafted polymer, similar to results discussed for MMA-PET modifications (c.f. section 4.1.2.1.1). 4.2.2. XPS analysis of Radiation Grafted SEMA onto PET Surface modification of Dacron with SEMA using gamma radiation with and without presoaks was successful as determined by XPS analysis. Samples irradiated to 0.11 Mrad in 15% SEMA-85% DMSO solutions show a sulfur (S2p) atomic concentration of 1.5%. Samples irradiated to 0,11 Mrad in 15% SEMA-85% DMSO following a 24 hour, 60C presoak in 15% SEMA-85% DMSO have a concentration of 2.0% sulfur, which is not considered significantly different from the non-presoak samples. Samples irradiated in 15% SEMA-85% DMSO solution following a presoak in 40% MMA-60% DMSO at 60C for 24 hours, however, provided 2.3% sulfur. The higher concentration of sulfur for the MMA-DMSO presoak system is attributed to the presence of MMA, and the surface graft is a network of poly(SEMA) and PMMA.
PAGE 165
148 4. 3 Solution Dip Coating and Ther11tal Curing of PDMS Coatings on Dacron The initial conditions chosen for dip coating Dacron with PDMS provided an all-PDMS surface on the woven fabric. Research focused on the characterization and the biological response to PDMS coated Dacron. 4.3.1 Conditions for Dip Coating Dacron with PDMS Shin-Etsu KE-1935 PDMS is supplied as two viscous (450 and 750 P) oligomer components. Upon mixing the two components, an addition reaction is initiated by the combination of a platinum catalyst and a vinyl monomer. The viscosity was reduced to approximately 100 to 200 cps with the addition of chloroform. Ten percent (w/w) solutions of each component in chlorofo:r11t were mixed together, and the Dacron placed in the solution. The solution concentration (10%) was chosen based on appearance of the solution viscosity. Other solution concentrations may be used, and the effect of the concentration on the coating thickness and adherence is recommended for future work. The samples were allowed to remain in the PDMS solution for 4 hours. The time of 4 hours was chosen from swelling data for PET in chloroform, and should provide even coverage of the Dacron by the solution, as well as allow some swelling of the Dacron. Because of the high amount of PDMS solution remaining on the surface, blotting the material and
PAGE 166
149 evaluating the weight did not provide information concerning the uptake of PDMS-chloroform solutions uinside" the Dacron, only total weight increases. 4.3.2 Thermal Curing of PDMS Coatings on Dacron The supplier of the Shin-Etsu silicone reconunends a cure profile of 150C for at least 1 hour. Increases in curing temperatures cure the material more quickly, and longer times allow increased extent of polymerization. Thus, curing at room temperature is possible with extended curing times. Dacron, or PET, has a glass transition temperature of ao 0 c, and curing above this temperature will change the percent crystallinity of the fibers, as well as cause deformation of the overall fabric as the fibers relax above the Tg. For this reason, a curing temperature of 60C was chosen. The total curing time was extended over 60 hours, in three steps, as the solvent and residual low molecular weight oligomers were removed. The samples were first allowed to cure in air at 60c for 24 hours. This step was chosen first to avoid rapid evaporation of chloroform, which could potentially cause bubbles or voids in the coating. Then a vacuum was applied to the sample to continue curing at 60C for 24 hours and facilitate the removal of chloroform. Following washing with acetone, the sample was exposed to a final cure under vacuum for 12 hours at 60c. The total curing time of 60 hours at 60C provides a crosslinked elastomeric surface coating of PDMS on Dacron.
PAGE 167
150 4.3.3 Gamma Polymerization of MAOP-t-PDMS The incorporation of PDMS into the Dacrone fibers during the soaking step was not evaluated, and there is no reason to suspect chemical bonding between the Dacrone substrate and PDMS coating. The adhesive bonding of the coating is therefore a concern. MAOP-t-PDMS was gamma polymerized onto Dacrone to provide a link between the coating and the substrate. This polymerization was done both with and without presoaking in MMA solutions, providing three conditions for the MAOP-t-PDMS modifications. The first condition involved a presoak in 40% MMA-60% DMSO for 24 hours at 60C followed by irradiation in a solution of 10% MMA, 10% MAOP-t-PDMS, and 80% chlorofor111 to 0.11 Mrad. The next condition evaluated for MAOP-t-PDMS modification utilized a presoak in 10% MMA-90% chloroform followed by irradiation in 10% MMA, 10% MAOP-t-PDMS, and 80% chloroform to 0.11 Mrad. A final condition using no presoak was evaluated with samples being irradiated to 0.11 Mrad in a solution of 15% MAOP-t-PDMS, and 85% chlorofor111. After gamma polymerization, the samples were dipped based on the standard dipping procedure used.
PAGE 168
151 4.3.4 Analysis of PDMS coated Dacron 4.3.4.1 Gravimetric analysis of PDMS coated Dacron Table 4.12 shows weight increases for the different modifications of Dacron with PDMS and MAOP-t-PDMS. The dip coating modification obviously provides more PDMS on the material, but the gamma polymerization of the MAOP-t-PDMS is significant as well. Presoaking with MMA-chloroform solutions also provides more MAOP-t-PDMS modification than presoaking with MMA-DMSO solutions. This is a result of one of two factors. Either the incompatibility of chloroform (in the MAOP-t-PDMS solution) with DMSO (in the presoak solution) causes poor interactions between the polymerization solution and the swollen substrate, or the higher amount of MMA in the substrate provides a better substrate for polymerization of MAOP-t-PDMS. Samples polymerized without presoaking in MMA solutions did not show significant weight increases following polymerization and washing. This observation indicated the presence of MMA (and PMMA) within the substrate increases the polymerization of MAOP-t-PDMS to the substrate. The MMA is able to penetrate and swell into the fibers, and upon polymerization, propagate with the MAOP-t-PDMS providing covalent bonding of the substrate to PMMA to MAOP-t-PDMS.
PAGE 169
152 Table 4.12 Percent weight increase of Dacron~ dip coated with PDMS or gamma polymerized with MAOP-t-PDMS as a function of modification conditions. Sample Presoak Polymerization Polymerization Weight Set Conditions Solution Conditions Increase Group B 40% MMA/DMSO 10% MAOP-t-PDMS/ Gamma Radiation to 2.5% 24 hrs-60 C 10% MMA/chlor. 0.13 Mrad Group C 10% MMA/chlor. 10% MAOP-t-PDMS Gamma Radiation to 4.5% 24 hrs-Room T. 10% MMA/chlor. 0.13 Mrad Group D 40% MMA/DMSO 40% MMA/DMSO Gamma Radiation to < 1% 24 hrs-60C 0.13 Mrad Group E 10% Shin-Etsu/ 24 hrs-60C-air Dip chloroform NONE 24 hrs-60C-vacuum 19% Coating 4 hrs-Room T. Acetone wash 24 hrs-60C-vacuum Group K NO PRESOAK 10% MAOP-t-PDMS Gamma Radiation to < 1% 10% MMA/chlor. 0.13 Mrad 4.3.4.2 Contact angle analysis of PDMS coated Dacron Although it is immediately obvious that PDMS coated Dacron is more hydrophobic than Dacron itself, contact angle goniometry using micro bubbles gave similar values for both substrates. Placing water drops on coated and uncoated samples of Dacron~ and observing the drop did, however, provide a relative indication of the surface energies. The drop on the uncoated Dacron was readily ''absorbed'' into the weave (into the weave, not actually absorbed by the fibers) and eventually wet the entire sample. The drop on the PDMS coated Dacron remained on the surface, not wetting the Dacron, until it evaporated. The water porosity of the
PAGE 170
153 samples was evaluated in a pressurized system discussed in section 4.3.4.5. 4.3.4.3 FT-IR analysis of PDMS coated Dacron Obtaining a spectra of woven Dacron fabrics was very difficult (as discussed earlier) because of the poor surface contact of the rough surface with the ATR crystal. The addition of a PDMS coating did not significantly increase the resolution of the spectra, and although peaks corresponding to PDMS were seen in the spectra for the dip coated samples, a clear spectrum for Dacron was not obtained for comparison. 4.3.4.4 XPS analysis of PDMS coated Dacron XPS analysis of the silicone coated Dacron prostheses indicates complete coverage of the surface with PDMS. The atomic composition of Shin-Etsu silicone used in this research was the same as the values determined for the silicone coating on Dacron. The composition of the MAOP-t-PDMS (before dip coating) indicates the presence of less silicon than that for 100% PDMS. The MAOP-t-PDMS surface concentrations of silicon were determined by XPS following the presoak conditions. The results are the most conclusive proof of the presence of PMMA grafting on PET following gamma radiation (which was not determined by other means as discussed previously in section 4.1.2.1.). These results are displayed in table 4.13.
PAGE 171
154 Table 4.13 Carbon, oxygen, and silicon atomic concentrations for Dacron with dip coatings of PDMS and gamma polymerized MAOP-t-PDMS determined with XPS. Samole Set Description Carbon o.AY '-fen Silicon Shin-Etsu PDMS Control 50.25 22.60 27.15 Cured as PDMS Sheet Shin-Etsu PDMS Cured using 49.44 22.45 28.11 from Solution Dio Coatinq Conditions Grau A Dacron Control 72.92 24.02 3.06 Presoak in MMA/DMSO 53.41 21.88 24.71 Group B Irradiated in MMA/ MAOP-t-PDMS/chloroform Presoak in MMA/chloroform 61.10 21.78 17.12 Group C Irradiated in MMA/ MAOP-t-PDMS/chloroform Grouo E Dio Coated Dacron 49.44 22.45 28.11 NO PRESOAK Group K Irradiated in MMA/ 67.40 24.50 8.10 MAOP-t-PDMS/chloroform The increased concentrations of silicon following presoak in MMA solutions and gamma polymerization indicate there is MMA in the substrate which allows more complete polymerization of the MAOP-t-PDMS. The size of the MAOP-t-PDMS molecule (MW 33,000) reduces the diffusion of the functional end groups into the substrate, but a covalent link between PMMA-g-PET and the MAOP-t-PDMS is expected. The incorporation of MMA into the system provides a propagating link between the substrate and the MAOP-t-PDMS thus allowing polymerization of more MAOP-t-PDMS. The dip coating of PDMS should now polymerize and crosslink with the MAOP-t-PDMS. There is also a difference in the XPS data for Shin-Etsu PDMS cured as a sheet (S0C under vacuum for 45 min. followed
PAGE 172
155 by 150c in air for 1 hour) and Shin-Etsu cured from solution. For the solution cure sample, the 10% Shin-Etsu chlorofor1cL solution was placed on a glass slide and subjected to the same curing profile as dip coated Dacron samples. This difference became clear after observing XPS the Cls spectra for these samples. Figure 4.38 shows the Cls binding peaks for Shin-Etsu PDMS (high temperature cure) and 60C cured Shin-Etsu PDMS. The shoulder present on the lower binding energy side (right hand side) is present in the high temperature cure sample, indicating the presence of unsaturated bonds or a higher concentration of CH2 molecules. Both of these possibilities indicate either a lower extent of cure for the high temperature cure compared to the 60C cure, or a more complete removal of low molecular weight oligomers as a result of washing a thinner PDMS layer (compared to the high temperature cured material) in acetone and hexane. Also shown in figure 4.38 is the Cls peak for unmodified Dacron and PDMS dip coated Dacron. The carbonyl bond shoulder is completely absent from the modified sample spectra, indicating a coating of PDMS greater than 50A. Pre-modification of Dacron with MAOP-t-PDMS was successful, and provides a covalent link from the Dacron~ substrate to MAOP-t-PDMS to solution dip coated PDMS, and should increase the adhesion of the coating. The dip coating modification process is either cured more extensively or has a more complete removal of low molecular weight oligomers than PDMS cured according to the manufacturer's recommendations.
PAGE 173
t.n C ::) >CG .c < >t.n C Cl) C 300 295 156 290 285 280 Shin-Etsu Silicone Cured as Sheet Shin-Etsu Silicone Cured as Solution Coatin PDMS Dip Coated Daaon Daaon 275 270 265 Binding Energy ( eV) Figure 4.38 Cls spectra showing a comparison between 60C cured PDMS from chloroform solution and high temperature cured PDMS, and Dacron and PDMS dip coated Dacron. 4.3.4.5 Porosity and ICP stability analysis of PDMS coated Dacron Vascular grafts made with materials used in this research (Dacron fabrics, Meadox and reinforced Bard vascular grafts) require pre-clotting prior to implantation because the per1c1eability allows significant blood leakage. A surface coating of PDMS was found to alter this behavior. Dacron grafts were sealed at one end and pressurized with water to 120 mm Hg. Unmodified grafts were found to have a leak rate
PAGE 174
157 of 320 ml/cm2-min, and PDMS dip coated grafts were found to have a leak rate of 200 ml/cm2min. When the grafts were implanted for in vivo analysis, the grafts did not leak, and pre-clotting was not necessary. The adhesion between the PDMS coating and the Dacron substrate must be sufficient to prevent delamination or '' flaking'' of silicone molecules into the vascular environment. Dynamic flow studies were carried out to determine the amount of silicone in the coating that is released upon exposure to a simulated vascular environment. The water from the flow system was analyzed with ICP to determine the silicon (Si) concentration in solution. The test is sensitive to 0.1 ppm (0.1 g/ml). The studies were carried out in Ultrapure~ water and an aqueous solution of 10% Triton x. The Triton X surfactant solution is believed to more closely mimic the molecules in the vascular environment. The complex surface activities of blood molecules may have better suspending properties of silicone than the water, and the Triton X should immediately suspend any silicone leaving the graft. The results for the flow analysis are presented in table 4.14. There is no silicon in solution following analysis of the silicone coated graft in water, and only 0.3 ppm from the coated graft in Triton X solution. The 0.3 ppm corresponds to 0.3 g/ml. This amount is not believed to be significant. The statistics of a student t test were unable to be applied to this data because the ICP calibration is relative to the standard each time the
PAGE 175
158 experiment was run. If all the solutions for each flow test were analyzed at the same time, then the values could be compared. As the data were collected, however, the number of samples compared for each ICP session was one, and statistical analysis was not possible. Table 4.14 Silicon concentrations in solution from the vascular graft delamination analysis as measured by ICP. Sample Description Silicon Standard Concentration Deviation 100 Silicon Standard 106.1 1.16 P Ultrapure Water 0.4 0.04 ..,um 12 Hours flow with Unmodified Graft 0.6 um 0.02 24 Hours flow with Silicone Coated 0.0 ppm 0.02 ppm Graft Senarate Analvsis of Triton X Solution: Sample Description Silicon Standard Concentration Deviation 100 Standard 101.6 ili:im 0.59 Triton X Solution from Flow System 1.3 ppm 0.06 ppm 24 Hours NO GRAFT Triton X Onlv Triton X solution from Flow System 1.6 ppm 0.07 ppm 48 Hours Silicone Coated Graft 4.3.4.6 Ex vivo, and in vivo analysis of PDMS coated Dacron The biological response to the PDMS coated Dacron prostheses was analysis was much better than initially anticipated. The idea of a silicone coating was generated initially from observations of ex vivo shunt studies, where a PDMS shunt is used with no clotting, suggesting a PDMS coating on substrates to be tested would have a similar response.
PAGE 176
159 Table 4.15 shows ex vivo shunt data for Dacron, PDMS coated Dacron, and MAOP-t-PDMS modified Dacron, giving platelet counts (count of the number of platelets on the surface) following removal. The most interesting observation is the MAOP-t-PDMS modified Dacron has platelet counts similar to those of the control samples, even though there is a significant amount of PDMS on the surface. The ANOVA confidence overlaps indicate no significant difference between Groups A, B, and c, or between Groups D, E, or F, but indicates a significant difference between the two sets (A, B, and c compared to D, E, and F). The favorable results obtained with PDMS coated Dacron in the ex vivo shunt led to its use as a prosthesis for in vivo studies. These results were not available at the completion of this manuscript, Table 4.15 Platelet counts from ex vivo AV shunt experiments for unmodified, PDMS dip coated, and MAOP-t-PDMS modified Dacron fabric. Platelet counts are reported as counts/mm 2 sample surface area. Sample Sample Modification Mean Confidence Set Descriotion Counts OVerlaos Grouo A Unmodified Dacron 2227.2 B, C 40% MMA/60% DMSO presoak Group B Irradiated in 10% MMA/ 2035.8 A, C 10% MAOP-t-PDMS/chloroform 10% MMA/chloroform presoak Group C Irradiated in 10% MMA/ 2523.8 B, A 10% MAOP-t-PDMS/chloroform Grouo E PDMS Din Coated and Cured 549.1 F, G Group F Modify according to Group B 839.5 E, G followed bv Group E Group G Modify according to Group C 593.0 E, F followed bv Grouo E
PAGE 177
160 4.3.4.7 SEM analysis of PDMS coated Dacron SEM analysis of the PDMS coated Dacron shows a dramatic difference from the unmodified material. Figures 4.39, 4.40, and 4.41 show SEM micrographs for an unmodified sample of Dacron fabric (4.39), a sample of PDMS dip coated Dacron fabric (4.40), and a PDMS dip coated vascular graft (4.41), (Bard reinforced Dacron vascular prosthesis). It appears from these micrographs the porosity of the graft is completely eliminated. However, this contradicts the results obtained with the leak test. The coating is clearly visible on the Dacron surface, and the overall appearance of the surface is noticeably different. Figures 4.42 and 4.43 show micrographs of samples following ex vivo shunt analysis. The accumulation of proteins, red blood cells, and platelets is clearly visible on the unmodified samples shown in figure 4.42. The coated samples, however, appear relatively clean, with minor deposits and accumulation of blood components. The low platelet adhesion of the PDMS dipped Dacron does not, however, indicate a non-reactive surface. It is possible platelet aggregation and accumulation is induced and the surface energy prevents adhesion. Compliment assays are therefore recommended as future work for this research. 4.3.4.8 Optical microscopy of PDMS coated Dacron A sample of PDMS dip coated Dacron was mounted in paraffin (Tissue Prep, Fisher Scientific), sectioned with a microtome, and stained with saturated crystal violet-acetone
PAGE 178
161 solution. The PDMS surface coating was visible, but too small to be accurately measured. The resolution of the optical microscope used in this research is approximately 1 m, indicating the PDMS surface coating is less than 1 m.
PAGE 179
162 Figure 4.39 SEM micrographs of unmodified Dacron fabric (Meadox).
PAGE 180
163 Figure 4.40 SEM micrographs of PDMS dip coated Dacron fabric (Meadox).
PAGE 181
164 Figure 4.41 SEM micrographs of PDMS dip coated reinforced Dacron vascular prosthesis (Bard)
PAGE 182
165 Figure 4.42 SEM micrographs of unmodified Dacron fabric (Meadox) after ex vivo AV shunt analysis.
PAGE 183
166 Figure 4.43 SEM micrographs of PDMS dip coated Dacron fabric (Meadox) after ex vivo AV shunt analysis.
PAGE 184
167 4.4 Results and Discussion of Fenestrated ICLs With Hydrograft Surface Modifications 4.4.1 Determining Presoaking and Grafting Conditions Yahiaoui (1990) did extensive work on the gamma radiation polymerization of NVP onto PMMA. By using a presoak method, whereby the monomer is allowed to diffuse into the substrate, a surface region of grafted and interpenetrating molecules was created. The presoaking conditions (time, temperature, and solution concentration) were found to have a large effect on the resulting graft thickness, or the depth of penetration of the PVP network. The initial focus of the surface modification investigations on Surgidev lens materials was to determine the grafting conditions necessary for Hydrograft surface modification. The initial target thickness of the graft was lOm. The PMMA used for the corneal inlays is a lower molecular weight than material typically used for surface modified IOLs, and the it was not known if the Hydrograft process would distort the lenses. The modification of the ICLs also raised the question of whether the holes might be partially occluded by the Hydrograft surface. The residual stresses of the ICLs is a function of the processing, and is the primary factor controlling distortion during modification. Initial studies were done on a low molecular weight PMMA material (V-811) because of the
PAGE 185
168 relatively high cost of the fenestrated ICLs. Tests beyond this initial experiments were conducted using fenestrated disks and a limited number of ICLs provided by Surgidev Corporation. 4.4.1.1 Diffusion of Aqueous NVP Solutions During Presoaking A major controlling factor in the surface modification of PMMA with NVP is the diffusion of monomer into the substrate during the presoak. The V-811 PMMA studied first was subjected to two presoak conditions. The first condition was based on current research in this laboratory for surface modification of IOLs, and involved a presoak in 20% NVP (aqueous) at 60c for 90 minutes. These conditions provided a monomer penetration depth of 30-50 m into the surface as determined by optical microscopy. However, there was ''hazing'' of these samples induced by rapid swelling and relaxation of the polymer. A presoak in 10% NVP at 60c for 90 minutes yielded a monomer penetration depth of 10-15 m, with no optical distortion of the PMMA. Although the penetration depth of 10-15 m by the 10%600C-90 min conditions was within the target range for initial surface modification studies, the effect of time and temperature was studied to determine the presoak conditions which would provide a specific depth of monomer penetration. The effect of each step of the surface modification on the overall graft thickness was also studied. A single sample of a fenestrated disk was broken into four sections, and subjected to the presoak and irradiation
PAGE 186
169 steps of surface modification. Each sample was analyzed at a different point of the process with optical microscopy to determine the penetration depth. Typically, silver nitrate is used to stain modified samples to visualize the graft thickness, but the monomer was visible without staining by using polarized light microscopy. The modification steps proceeded as discussed in section 3.2.2. No significant diffusion during the irradiation step was observed, and the presoak step alone was found to determine the penetration/ diffusion depth of the grafted region. The results for these steps are shown in table 4.16. The penetration of NVP solution into the sample was measured in the region surrounding the fenestrations. Figure 4.44 shows an optical micrograph of a fenestrated disk with NVP diffused into the sample. This sample also shows the preferential etching of the PMMA along the stacking lines caused by relaxation of the PMMA, which will be discussed later. TABLE 4.16 Depth of penetration of 10% NVP (aqueous) into fenestrated PMMA disks during surface modification as determined by optical microscopy. Presoak Conditions Grafting Conditions Penetration Depth 10% NVP:1% AgN03-60C-90 min Quench Presoak onlv 6 ,~ 10% NVP-60 C-90 min Quench Presoak onlv 6 ~ Quench Add 10% NVP 10 % NVP-60 C-90 min No Irradiation 6 m Swell 4 hours 18 min. 10 % NVP-60 C-90 min 10% NVP and Degas 6 m Irradiate to 0.11 Mrad
PAGE 187
170 ()_ Figure 4.44 Optical micrographs of fenestrated ICLs after being surface modified wj th NVP and stained with silver nitrate.
PAGE 188
171 Swelling experiments were also conducted at room temperature, 40c, and 50C with the fenestrated disks. The results of these studies are presented in table 4.17. Table 4.17 Penetration depths of 10 and 20% NVP into fenestrated PMMA disks as determined by optical microscopy. Solution Concentration Solution Swelling Time Depth of Temperature Penetration 10% NVP Room Temo. 60 min 1 ,rm 20% NVP Room Temo. 60 min 1 ,rm 10% and 20% NVP 40C 35 min 1 urn 10% and 20% NVP 40C 24 hours 1-1.5 um 10% NVP 50C 30 min 1-2 ,rm 10 NVP 50C 90 min 4-5 ,rm 20% NVP 50C 120 min 5 urn Previous work in this laboratory indicated an activation energy temperature for aqueous NVP solutions diffusing into PMMA of 50 to 55C (Goldberg et al., 1988-1995). Samples subjected to presoaks within this temperature range had sporadic depths of penetration as a result of 1 c temperature control. These data follow the same correlation, and a 1 to 2 m graft in low molecular weight PMMA is readily obtainable under lower temperature conditions if distortion and crazing remain major concerns. Unlike the results presented previously for modification of PTFE and PDMS with PMMA, PMMA substrates are glassy below 117c, and there is no appreciable expansion of free volume
PAGE 189
172 upon polymerization as is the case for a plastic (PTFE) or elastomeric (PDMS) substrates. Diffusion alone is the primary factor determining the depth of penetration of the grafted region in the Hydrograft modification of PMMA. The presoak facilitates the diffusion of monomer into the substrate, and upon exposure to gamma radiation, grafting and homopolymerization occurs, but there is no increase in the depth of penetration of the monomer during polymerization. Thus, the presoak step determines the graft thickness, or the depth of penetration of the NVP-grafted-IPN-PMMA network. 4.4.1.2 Stability and Optical Quality after Presoaking Subjecting a polymer such as PMMA to a solvent at elevated temperatures may increase the chain mobility of the individual molecules or molecular segments. If there are residual stresses ''frozen'' in the material from processing, this increased mobility may cause distortion, or a return to the previous shape as the residual stresses relax. Solvent swelling has been shown to induce stress cracking in glassy polymers such as PMMA (Ueberreiter, 1968). The processing of the ICLs by Surgidev involves several extrusion steps which may induce residual stresses. First, a square PMMA boule with a ceramic binder filling a cylindrical hole through the boule is extruded to create a longer block with a significantly reduced cross sectional area. These individual blocks are stacked and fused into a new boule. The process is repeated until a block with specified hole size and percent hole coverage is achieved. The ICLs are then
PAGE 190
173 machined from the final boule, and the ceramic binder is dissolved from the fenestrations. The final lens is subjected to an annealing process to reduce the residual stresses induced from the processing, however, removal of all stresses is difficult. It is critical to the optical quality of the lens that the presoak process not induce stress cracking or relaxation. Presoaking the initial lenses induced some relaxation of the polymer through either preferential etching or stress cracking of the lens along the stacking lines. Even room temperature exposure induced some etching in early lenses. An optical micrograph of this phenomena is shown in figure 4.44. When this observation was made, Surgidev altered the annealing process and supplied new lenses. These new ICLs did not have the same etching problem as the initial lenses. Presoaking these lenses with a 10% NVP solution at 60C for 90 minutes produced a depth of penetration of 9-10 m. 4.4.1.3 Surface Modification of ICLs with NVP The surface modification process creates a region of grafted and IPN polymer on and in the surface. The depth of penetration of this region into the surface is controlled by the presoak process. During the polymerization, both grafting and homopolymerization occur via free radical initiation polymerization. The process is diffusion controlled based on the kinetics of propagation during the ''gel effect'' of the reaction. The glassy nature of PMMA, however, prevents significant increases in free volume during
PAGE 191
174 polymerization, and the diffusion occurring during the presoak process predominately determines the depth of penetration of the grafted region. The thickness of the surface layer outside or on the surface, however, has not been characterized in previous work. The modification of fenestrated disks and ICLs provided an unique opportunity to evaluate the thickness of the PVP surface layer. Surface modification of the fenestrated disks and ICLs resulted in a penetration depth of 9-10 m for samples subjected to a presoak of 10% NVP-60C-90 min, followed by irradiation to 0.11 to 0.13 Mrad in a fresh, degassed 10% NVP solution. Close evaluation of both the fenestrated disks and ICLs did not identify the surface PVP layer. Samples were observed both hydrated and dry, and the surface graft was too thin to be measured optically, indicating a thickness less than 1 m. The major concern was the resulting fenestration hole size following surface modification, and these observations show the hole size is not altered by the modification process. 4.4.2 Analysis of Modified ICLs The properties of the modified ICLs were evaluated with respect to wetting and pernteability of the fenestrations to water and aqueous solutions of NaCl and glucose. 4.4.2.1 Contact Angle Goniometry It was extremely difficult to determine the contact angle of the ICLs because of the curvature. It was not
PAGE 192
175 possible to get an air bubble trapped by the surface to be evaluated. The fenestrated disks, however, had a contact angle of 55 prior to surface modification and less than 20 following surface modification, indicating a hydrophilic surface. A water drop-in-air contact angle was obtained by placing a 0.1-0.2 drop of water (using a microliter GC syringe) on the lens surface. For the unmodified V-811/XUS sample, an angle of 79 was measured compared to 55 after modification. This measurement was made while the lens was ''dry'' (hydrated by room humidity only). The water drop remains on the unmodified lens for more than 30 minutes whereas the modified lens absorbs the drop (into the graft and through the fenestrations) within 10 minutes. When the sample is hydrated (placed in water for more than 10 minutes), the water drop goes through both the modified and unmodified lenses within 10 seconds. This observation was not unexpected, because it is impossible to remove all the water on the inside of the lens without dehydrating it, and the excess water may force the water through by capillary action on the unmodified sample whereas the modified sample is actually very hydrophilic. 4.4.2.2 Aqueous Flow Analysis The wettability of the ICLs was further evaluated by observations of the water flow rate through the lenses. The diffusion cell was set up as shown in figure 3.7 (c.f. section 3.2.5). Water was placed in one chamber of the
PAGE 193
176 device, and the amount of water passing though the lens in 90 minutes was measured. An aspirated flask was used to maintain a constant pressure on the water filled chamber side. The results of this evaluation for ICLs of various hole size are shown in table 4.18. There is approximately a 30% increase in the flow rate for surface modified ICLs (SOm/5%). The most interesting observation was for the lens modified without a presoak. Although the lens has a penetration depth of 1-2 m, the flow rate is not altered (30m/5% lenses). Table 4.18 Flow rate of water through modified and unmodified ICLs at room temperature and constant pressure (atmospheric pressure plus 1 inch of water). Lens type column gives fenestration hole size and surface percentage of fenestrations. Lens Type Presoak and Polymerization Depth of Flow Rate Conditions Penetration 220-121 NONE CONTROL N/A 0.62 ml/min
PAGE 194
177 4.4.2.3 Saline and Glucose Solution Flow and Diffusion Filtered and degassed isotonic saline solution was placed in one chamber, and degassed Ultrapure water placed in the other. The concentration of saline in both chambers remained unchanged for all ICLs tested. Both modified and unmodified samples of 30m/5% and BOm/5% were tested, and in a 24 hour period, no saline concentration changes were observed. Filtered and degassed glucose (380 mg/dl) solution was placed in one chamber, and degassed Ultrapure water placed in the other. The concentration of glucose in both chambers remained unchanged for all ICLs tested. Both modified and unmodified samples of 30m/5% and BOm/5% were tested over a 24 hour period. These results were surprising, and the lack of diffusion driven by osmotic pressure and chemical potential is unexplainable. Saline and glucose solutions were each analyzed in the flowing set-up to determine if they were excluded by the lens. Solutions were placed in one chamber with the other chamber being left empty, and allowed to flow through the lens. The final concentration of the solution in each chamber was the same as the initial concentration, indicating salt and glucose molecules were not excluded by the lens. It is beleived that the 24 hour time period is insufficient for detectable diffusion of the initial concentration solutions to take place. Longer time periods and higher chemical potential gradients are recommended for future work.
PAGE 195
178 The benefits of Hydrograft modified ICLs, thus far, appear to be the reduced tissue damage and cell adhesion previously reported for Hydrograft PMMA modifications (Yahiaoui, 1990), and the increased flow rate of water through the lens fenestrations.
PAGE 196
CHAPTER 5 SUMMARY AND CONCLUSIONS Surface modification of vascular graft substrates and ICLs using a variety of techniques and materials was completed. The primary goal of this work was to create more compatible materials for each of the two applications studied. Vascular prosthesis substrates which were less platelet adhesive (non-thrombogenic) and more conducive to endothelial cell adhesion and vascular healing were sought the research on surface modification of vascular prostheses. Creating materials which are more hydrophilic, less tissue damaging, and more permeable to aqueous solutions by gamma radiation induced polymerization was the goal of the ICL surface modification research. Each part of this research presented challenges, and the results presented are pronu.sing. The two primary methods studied for surface modification of vascular prostheses were simultaneous gamma irradiation of substrate and methyl methacrylate (MMA) monomer solutions, with and without presoaks, and solution dip coating and thermal curing of polydimethyl siloxane coatings on Dacron substrates. The primary method studied for surface modification of ICL materials was the Hydrograft presoak 179
PAGE 197
180 method developed by Yahiaoui (1990), and surface modification effects on low molecular weight PMMA ICL materials. 5.1 surface Modification of vascular Graft Substrates 5.1.1 Gamma Radiation Induced Polymerization of MMA on vascular Graft Substrates Yahiaoui (1990) demonstrated the importance of intimate contact between monomer and polymer substrate for successful graft polymerization during simultaneous substrate and monomer solution irradiation. The diffusion of monomer solutions into polymeric substrates was studied as the initial step in this research. The factors controlling diffusion determine the degree of swelling and success of surface modification. These factors include activation energy for diffusion, relative cohesive energy density of substrate, monomer, solvent, and polymer molecules, heats of mixing, and diffusivity. In this research, it was found that the degree of swelling during the presoak is strongly dependent on monomer solution concentration, as related to the relative solubility parameters of the solution and substrate. The diffusivity was observed to change with concentration, following a second order relationship. The polymerization yields, as determined by gravimetric analysis, contact angle goniometry, XPS, and DMS, have a dependence on both the total radiation dose and the monomer solution concentration. Polymerizations as a
PAGE 198
181 function of radiation dose demonstrate a second order, diffusion controlled reaction which mimics free radical polymerization during the ''gel effect,'' because the polymerization reactions studied are taking place within a viscous substrate (Odian, 1981 and Chapiro, 1962). Polymerization reactions studied as a function of monomer concentration demonstrate the diffusion dependence on the relative cohesive energy density of the reaction components, and show increased diffusivity and permeation during the reaction as a second order function. The progression of gamma radiation induced polymerization of a monomer within a polymer substrate begins with graft polymerization initiated by free radical initiation on both the substrate and monomer. As propagation proceeds, the monomer solution becomes more soluble in the grafting matrix giving rise to an increased diffusivity. If the substrate modulus is sufficiently low, the matrix swells, provides an increase in free volume, and polymerization of an IPN occurs. The research presented provides evidence for this phenomena by mapping the reactions as a function of both radiation dose and monomer concentration. The conclusions for the gamma radiation induced polymerization of MMA and SEMA on vascular prosthesis polymers are summarized below. Conclusions for surface modification of PET with MMA 1. Polymethyl methacrylate was polymerized onto the surfaces of PET Mylar films and Dacron fabrics using a
PAGE 199
182 presoak method followed by simultaneous gamma radiation of the monomer solution and polymer substrate. 2. The surface energies of PET and PMMA modified PET are similar, with a contact angle of 55 to 60, indicating a similar response will be expected upon exposure to adsorbing materials. 3. Canine ex vivo AV shunt platelet accumulation of PET and PMMA modified PET was found to be the same, as explained by the surface energies of the two polymers. 4. The ex vivo AV shunt platelet accumulation of Mylar films is significantly lower than Dacron fabrics, supporting previous literature citations concerning surface topographical effects on platelet adhesion. Conclusions for surface modification of PTFE with MMA 1. Polymethyl methacrylate was polymerized onto the surfaces of PTFE Teflon and ePTFE GORE-TE~ polymers by simultaneous gamma radiation of the monomer solution and polymer substrate without the use of a presoak. 2. Polymerization proceeds initially with grafting, followed by polymerization of an IPN as the structure expands with increased swelling. 3. The rate of polymerization follows a second order relationship related to diffusion during the "gel effect," and as the relative solubility parameters of the solutions and substrate change as polymerization proceeds.
PAGE 200
183 4. Polymerization in increasing concentrations of MMA give an increase in polymerization based on the solubility parameter changes of the solution. s. GORE-TEx, having a lower modulus, has higher degrees of polymerization than Teflon. 6. MMA-acetone polymerization solutions cause breaking of the nodules in the macro-structure of GORE-TEX, whereas MMA-DMSO solutions do not. 7. Polymerization of monomer solutions in DMSO with concentrations in excess of 10% caused deformation of the macro-structure of GORE-TEX. Conclusions for surface modification of PDMS with MMA 1. Polymethyl methacrylate was polymerized onto of Shin Etsu PDMS polymer substrates by simultaneous gamma radiation of the monomer solution and polymer substrate without the use of a presoak. 2. The modification of PDMS with MMA is a bulk process, resulting in a new graft copolymer of PMMA and PDMS. 3. The polymer may be tailored to have a range of surface properties from those of PMMA to those of PDMS. The materials have a higher modulus than PDMS (up to a 1200% increase) while retaining flexibility. This increased modulus is proportional to the weight fraction of PMMA within the PDMS. 4. Polymerization as a function of radiation dose and monomer concentration both follow second order relationships.
PAGE 201
184 Conclusions for surface modification of PET with SEMA 1. Polysulfoethyl methacrylate was polymerized onto the surfaces of PET Mylar films and Dacron fabrics using a presoak method followed by simultaneous gamma radiation of the monomer solution and polymer substrate. 2. Surface modification of SEMA is enhanced by the addition of MMA during the presoak and polymerization, eluding to the presence of PMMA in the final graft. 3. Surface modification of sulfonated monomers (non-salt functionalities) using a presoak method was successfully completed using non-aqueous organic solvents. 5.1.2 Solution Dip Coating and Thermal Curing of PDMS on Dacron, with and without Pre-Modification with MAOP-t-PDMS Silicone materials have been used extensively in blood contact applications without obvious complications. Use of Silastic tubing in canine ex vivo AV shunt studies provides an excellent example of the low thrombogenicity of PDMS tubing (Goldberg et al., 1988-1995, Lin et al., 1994, and Silver et al., 1995). Dip coating uncured PDMS oligomers from chloroform solutions followed by evaporation of the chloroform and thermal curing of the oligomers provided a PDMS layer covering the Dacron surface. Variations in dipping and curing parameters was not studied, but the properties of the coated material were examined extensively. Concerns over delamination and flaking of the coating led to the
PAGE 202
185 investigation of a coupling agent, MAOP-t-PDMS. MAOP-t-PDMS is a silicone oligomer (Mn= 33,000) with vinyl acrylic functional groups. Presoaks in MMA and MAOP-t-PDMS-MMA solutions of acetone and chloroform, followed by gamma radiation induced polymerization, and then dip coating with PDMS provided covalent bonding between the Dacron substrate, the MAOP-t-PDMS, and the dip coating of PDMS. The surface chemistry, coating thickness, adhesion stability, water permeability, and platelet reactivity of the modified surfaces were studied. A summary of the conclusions is listed below. Conclusions for solution dip coating PET with PDMS 1. The dip coating procedure provides a stable, cross linked, PDMS coating less than 1 m thick. 2. Vascular prostheses modified with this method showed no delamination or flaking of silicone from the surface when analyzed by a pressurized (140 mm Hg and a flow rate of 300 cc/min) for 48 hours. 3. Under a pressurized flow of 120 mm Hg, the leak rate of the PDMS dip coated prostheses was reduced from 320 ml/cm 2 -min to 200 ml/cm 2 -min, and the prostheses did not require pre-clotting prior to implantation in canines. 4. Coated Dacron had significantly reduced platelet adhesion (549.1 counts/mm2) compared to unmodified Dacron (2227.2 817.5 counts/mm2) samples as determined by a canine ex vivo AV shunt model.
PAGE 203
186 5. MAOP-t-PDMS was polymerized onto PET Dacron fabrics simultaneous gamma radiation of the monomer solution and polymer substrate with and without the use of a presoak in a methyl methacrylate solution. A covalent link between PET, PMMA, and MAOP-t-PDMS is assumed based on the results presented. 6. Non-PDMS dip coated, MAOP-t-PDMS modified Dacron modified Dacron did not have significantly reduced platelet adhesion compared to unmodified Dacron, indicating a more significant coating of PDMS is required for reduced adhesion. 5.2 Surface Modification of Fenestrated PMMA ICL Substrates 5.2.1 Gamma Radiation Induced Polymerization of NVP on PMMA ICL Substrates The Hydrograft presoak method studied by Yahiaoui (1990) was applied to low molecular weight, fenestrated, PMMA ICLs. The benefits of a hydrophilic polymer (especially Hydrograft) have been studied extensively (Bourne, 1976, Osborn, 1985, Goldberg et al., 1988, Goldberg et al., 1989, Hoffmeister, 1988, Yahiaoui, 1990, and Mentak, 1993). Therefore, the focus of this part of the research was to determine presoaking and grafting conditions which would provide a Hydrograft surface on low molecular weight PMMA ocular materials and to evaluate the permeability of modified fenestrated lenses to aqueous solutions.
PAGE 204
187 The process Surgidev Corporation uses to fabricate fenestrated ICLs induces residual stresses. It was found that a new annealing process was necessary for reduction of these stresses prior to surface modification. Surface modification conditions which provide a 6 to lOm graft penetration and a contact angle less than 20 were determined. A summary of the conclusions for surface modification of fenestrated PMMA ICLs via gamma radiation polymerization of aqueous NVP solutions using the presoak method is listed below. Conclusions for surface modification of PMMA ICLs with PVP 1. Polyvinylpyrrolidone was polymerized onto the surfaces of PMMA fenestrated ICLs and disks using a presoak method followed by simultaneous gamma radiation of the aqueous monomer solution and polymer substrate, producing a hydrophilic surface graft. 2. Modification of the low molecular weight PMMA devices is strongly dependent on the manufacturing process, and conditions for favorable modifications were identified. 3. The depth of penetration of monomer and graft polymer during and after surface modification is dependent on the presoak process, with no significant monomer penetration occurring after the presoak and during polymerization. 4. Observations of lenses with 20, 30, and 80m fenestrations following surface modification indicate no occlusion of the holes by the Hydrograft modification
PAGE 205
188 process. The thickness of PVP on (as opposed to in) the surface was not determined. 4. Surface modification of an 80m/5% fenestrated ICLs increased the water pe1.1cLeability from 3.5 to 4.5 ml/min, a 28% increase. 5. Diffusion of aqueous solutions of saline (0.9%) and glucose (380 mg/dl) as driven by chemical potential and osmotic pressure does not occur across the fenestrated lens in a 24 hour period.
PAGE 206
CHAPTER 6 FUTURE WORK 6.1 Surface Modification of vascular Prostheses The overall results of the surface modifications which have been presented indicate successful modification and some beneficial results with respect to the overall goal of achieving a non-thrombogenic prosthesis capable of re endothelialization. However, there are key experiments for each aspect of whose results would greatly complement the presented data and conclusions. 6.1.1 PMMA modified PET The modification of PET with PMMA was successful, as demonstrated by indirect qualitative methods. However, quantitative analysis of PMMA concentrations in and on the surface of PET were not deterrrilned. Morphological changes in PET are indicated based on swelling data which may be identified with volumetric and crystallinity analysis as a function of swelling. Also, with respect to platelet adhesion, there was no immediately evident benefit of the modified surfaces. Critical experiments to corroborate these observations are listed below. 189
PAGE 207
190 Next steps for surface modification of PET with MM.A 1. Determine a method for quantitative analysis of PMMA concentrations in modified PET samples. 2. Closely measure dimensional changes and volumetric expansion during swelling and evaluate percent crystallinity using differential scanning calorimeter following swelling. Compare the results of changes between Mylar and Dacron samples. 3. Evaluate complement activity of modified and unmodified Dacron and Mylar to determine possible benefits of PMMA surface modified PET with regards to vascular exposure and reduced thrombogenicity. 6.1.2 PMMA modified PET The modification of PTFE (Teflon) and ePTFE (GORE-TEX) with PMMA was successful, and clear relationships for the polymerization as a function of dose and monomer concentration were identified. The morphology or microstructure of PMMA within PTFE was not identified, and the overall change in mechanical properties remains unevaluated. Platelet reactivity and adhesion studies also have not yet been completed. Critical experiments to for the continued investigation of PMMA modified PTFE are listed below. Next steps for surface modification of PTFE with MMA 1. Evaluate the mechanical properties of modified GORE-TEx using DMS. Information regarding the mixing of PMMA and
PAGE 208
191 PTFE. Determine changes in mechanical properties due to radiation induced degradation and the addition of PMMA. 2. Begin ex vivo studies to evaluate platelet adhesion. 3. Evaluate modified and unmodified samples with complement activation assays. 6.1.3 PMMA modified PDMS Modification of PDMS with PMMA proved to be a bulk process, creating a new copolymer network. The dynamic mechanical properties have been evaluated, but mechanical improvements to PDMS were not fully investigated. The poor tear and tensile properties of PDMS should be compared to the new polymer. Applications of this material to other areas such as ocular biomaterials is also interesting, and studies related to these applications are necessary. The success of PDMS ex vivo coupled with the hypothesis that PMMA surfaces on vascular biomaterials should have beneficial properties warrants ex vivo studies on PMMA modified PDMS. Experl.lllents necessary fo r the continued investigation of PMMA modified PDMS are listed below. Next steps for surface modification of PDMS with MMA 1. Evaluate tensile and tear strengths of modified PDMS samples as a function of PMMA content. 2. Evaluate properties critical to IOL applications, such as optical clarity and ocular tissue compatibility (in vitro epithelial adhesion studies, iris abrasion and adhesion tests, and in vivo implants).
PAGE 209
192 3. Begin canine ex vivo shunt studies to determine the platelet reactivity of PMMA modified PDMS. 4. If canine ex vivo studies show promise, apply the PMMA modification to PDMS dip coated Dacron~ samples, and evaluate the resulting properties. 6.1.4 SEMA modified PET The primary goal of SEMA modification of PET was to demonstrate surface modification with a sulfonic acid containing monomer using a presoak in non-aqueous, organic solvents. Next steps in this research include improvements in the surface yield of SEMA and biological testing. Next steps for surface modification of PTFE with MMA 1. Increase SEMA concentrations during presoaking and polymerization to achieve higher surface concentrations of polysulfoethyl methacrylate. 2. Evaluate the stability of the SEMA surface using the flow analysis system and ICP. 3. Evaluate the leak rate and porosity of SEMA modified Dacron prostheses. 4. Investigate ex vivo platelet reactivity and adhesion using the canine shunt model. 6.1.5 Solution Dip Coating of PDMS onto PET work on solution dip coating of PET with PDMS was, by far, the most successful surface modification method at reducing platelet adhesion studied in this research. The
PAGE 210
193 modification parameters were not extensively investigated, and, although the MAOP-t-PDMS was successfully polymerized onto PET, the added adhesive strength was not evaluated. Critical experiments to for the continued investigation of PDMS coated vascular prostheses are listed below. Next steps for surface modification of PTFE with MMA 1. Continue prosthesis healing studies using the canine in vivo model. 2. Evaluate the effect of solution concentrations and dipping times as functions of final coating thickness and concentration. 3. Apply the coatings to Myla~ films and use peeling tests to determine the adhesive strength of the coating with and without MAOP-t-PDMS pre-modifications. 4. Apply the dip coating process to other vascular biomedical device materials such as PTFE, stainless steel, tantalum, and titanium, to name a few. 6.2 Surface Modification of Intracorneal Lenses The surface modification research presented for Hydrograft modified ICLs is preliminary work on a new biomedical device. The overall results of the surface modifications identified important parameters with respect to successful modifications. The overall goal of achieving a hydrophilic surface which has improved transport properties within the stroma was accomplished in part, and further evaluations are required to clarify the results obtained.
PAGE 211
194 The diffusion experiments will be repeated using an improved chamber design. The biological response to the modified materials also requires evaluation for determination of the potential success of these devices. Critical experiments to corroborate these observations are listed below. Next steps for surface modification of PMMA ICLs with PVP 1. Design a diffusion apparatus which has stacked chambers and re-evaluate diffusion of aqueous solutions. 2. Calculate the theoretical diffusion rate based on the total porosity of the lens, and compare theoretical calculations to the actual presented results. 3. Begin biological testing on modified lenses, including in vitro epithelial cell adhesion tests, iris tissue abrasion tests, and endothelial touch tests which are designed to evaluate the adhesive strength of ocular tissues to biomaterials. 4. Design experiments to evaluate the in vivo or in vitro diffusion and flow of metabolites through the lens during implantation. Evaluation of in vivo animal implants would indicate the improvements of the modified devices. Flow analysis using an in vitro tissue model where an ICL is implanted into a dissected cornea would also provide vital information.
PAGE 212
LIST OF REFERENCES Abbott, w. M., Seminar at the March 1987 Annual Conference of the National Institute of Health/Heart, Lung, and Blood Institute. Abbott, w. M., and Cambria, R. P., Control of physical characteristics (elasticity and compliance) of vascular grafts, in Biologic and synthetic vascular prostheses, J. c. Stanley, ed., Grune & Stratton, New York, 189-220. (1982). Abbott, W.M., Merger1rtan, J., Hasson, J.E., L'Italien, G. J., and Warnock, D. F., Effect of compliance mismatch on vascular graft patency, J. vase. Surg., 5(2), 376-382, (1987). Absolom, D.R., Zing, w., and Newmann, A. w., Protein adsorption to polymer particles: role of surface properties, J. Biomed. Mater. Res., 21, 161, (1987). Amon, M., and Menapace, R., Cellular invasion on hydrogel and poly(methyl methacrylate) implants. An in vivo study, J. Cataract Refract. Surg., 17, 774-779, (1990). Andrade, J. n., Principles of protein adsorption, in Surface and interfacial aspects of biomedical polymers, Vol. 2, Andrade, ed., Plenum Press, New York, (1985). Andrade, J. D., Coleman, D. L., Didisheim, P., Hanson, s. R., Mason, R., and Merrill, E., Panel Conference: Blood materials interactions 20 years of frustration, Trans. Am. Soc. Artif. Inter. Organs, 27, 659-662, (1991). Apple, D.J., Mamalis, N., Loftfield, K., Googe, L.M., Novak, L.C., Kavka-van No1rtan D., Brady, S.E., and Olson, Complications of intraocular lenses. A historical and histo-pathological review, Survey of Ophthalmology, 29, 1-54, (1984). Atala, c. A. Peters, A. B. Retnik, and J. Mandell, Endoscopic treatment of vesicoureteral reflux with a self-detachable balloon system, J. Urology, 148, 724-728, (1992). 195
PAGE 213
196 Baier, R. E., Initial events in interactions of blood with a foreign surface, J. Biomed. Mater. Res., 3, 191-206, (1969). Baird, R. N., Kidson, I. G., L'Italien, G. J., and Abbott, W. M., Dynamic compliance of arterials grafts, Am. J. Physiol., 233(5), H568-H572, (1977). Balyeat,H.D., Nordquist, R.E., Lerner, M.P., and Gupta,A., Comparision of endothelial damage produced by control and surface modified poly(methyl methacrylate) intraocular lenses, J. Cataract Refract. Surg., 15, 491-494, (1989). Barr, T. L., Modern ESCA: the principles and practice of ray photoelectron spectroscopy, CRC Press, Boca Raton, Florida, (1994). Barraquer, J. L., Quertoplastia refractiva: estudios e infor1c1aciones, Ophthalmologica, 2, 10, ( 1949). Bonomini, v. Mioli, A. Albertazzi, and A. Vangelista, Aorto cava percutaneous teflon catheters system and teflon shunt for repeated haemodialysis in severe acute renal failure, Urol. Int., 24, 30-36, (1969). Bourne, w. M., and Kaufman, H. E., Endothelial damage associated with intraocular lenses, Am. J. Ophthalmol., 81, 482, (1976). Bowers, V.M., Fisher, L.R., Francis, G.W. and Williams, K.L., A micromechanical technique for monitoring cell-substrate adhesiveness: Measurements of the strength of red blood cell adhesion to glass and polymer test surfaces, J. Biomed. Mater. Res., 23, 1453-1473, (1989). Brandup, J. and Immergut, E. H., eds., Polymer Handbook, third edition, John Wiley & Sons, New York, (1989). Brody, w. R., Kosek, J. c., Angell, w. w., and Shumway, N. E., Changes in vein grafts following aorta-coronary bypass induced by pressure and ischemia, J. Thoracic and Cardiovascular Surgery, 64(6), 847-854, (1972). Brothers, T. E., Stanley, J. c., Burkel, w. E., and Graham, L. M., Small-caliber polyurethane and polytetra fluoroethylene grafts: a comparative study in a canine aortoiliac model, J. Biomed. Mater. Res., 24, 761-771, (1990). Chapiro, A. Radiation chemistry of polymeric systems, Wiley Interscience, New York, (1962).
PAGE 214
197 Charlesby, A., Effect of high-energy radiation on long-chain polymers, Nature, 171, 167, (1954). Christ, F.R., Fencil, D.A., Van Gent, s., and Knight, P.M., Evaluation of the chemical, optical, and mechanical properties of elastomeric intraocular lens materials and their clinical significance, J. Cataract Refract. Surg., 15, 176-184, (1989). Climenhaga, H., MacDonald, J.M., Mccarey, B. E., and waring, G. o., Effect of diameter and depth on the response to solid polysulfone intracorneal lenses in cats, Arch Ophthalmol, 106, 818-824, (1988). Cooper, S.L. and Fabrizius-Homan, D.J.,A comparison of the adsorption of three adhesive proteins to biomaterial surfaces, J. Biomater. Sci. Polymer Edn., 3(1), 27-47, (1991). Crank, J., and Park, G. s., Diffusion in polymers, Academic Press, New York, (1968). Dale, w. A., Arterial Grafts: 1900-1978, in Graft materials in vascular surgery, H. Dardik, ed., Symposia Specialists Inc., Chicago, 3-14 (1978). De Mol Van Otterloo, J. c. A., van Bockel, J. H., Ponfoort, E. D., Brommer, E. J.P., Hermans, J., and Daha, M. R., The effects of aortic reconstruction and collagen impregnation of Dacron prostheses on the complement system, J. vase. Surg., 16(8), 774-783, (1992). Dob6, J., Notes and communications: Comment on the radical initiation of chain reactions, Radiat. Phys. Chem., 11, 151-152, (1978). Epstein, s. E., Influence of stenosis severity on coronary collateral development and importance of collaterals in maintaining left ventricular function during acute coronary occlusion, Am. Jour. Cardiol., 61, 866-868, (1988). Falkenhagen, D., and Brown, G. s., Clinical evaluation of three haemodialysers containing PMMA membranes of different surface area, with special attention to thrombogenicity. Nephrol Dial Transplant Suppl., 2, 4348, (1991). Feller, F., The physical aging behavior of poly(etherimide), Master's Thesis, University of Florida, (1993).
PAGE 215
198 Feuerstein, I.A., and Mcclung, W.G., Platelet adherence and detachment with adsorbed fibrinogen: A flow study with a series of hydroxyethyl methacrylate-ethyl methacrylate copolymers using video microscopy, J. Biomed. Mater. Res., 26, 221-237, (1992). Feuerstein, I.A., Buchan, S.M., Horbet, T.A., and Hauch, K.D., Platelet adherence and detachment: A flow study with a series of hydroxyethyl methacrylate-ethyl methacrylate copolymers using video microscopy, J. Biomed. Mater. Res., 25, 185-198, (1991). Fox, S.I., Human Physiology, 2nd edition, Wm. c. Brown Publishers, Dubuque, 346-422, (1987). Fujita, H., Diffusion, in Encyclopedia of polymer science and technology, Volume 5, John Wiley & Sons, New York, (1966). Gertler, J.P., and Abbott, w. M., Current research Review: Prothrombotic and fibrinolytic function of normal and perturbed endothelium, J. Surg. Res., 52, 89-85, (1992). Gimbrone, M.A., Vascular endothelium: nature's blood compatible container, Annals NY Acad. Sci., 516, 5-11, (1987). Goldberg, E. P., Hofmeister, F. M., Yalon, M., Ida, S., Stacholy, J., Martin, P., Yahiaoui, A., and Sheets, J. w., "Hydrophilic surface modifications of IOL implants to reduce surgical and post operative ocular tissue trauma," paper presented at the Academy of Surgical Research Meeting, Boston, MA, (1988). Goldberg, E. P., Hofmeister, F. M., Yalon, M., Sheets, J. w., Stacholy, J., Martin, P., Ito, A., Yahiaoui, A., and Robinson, R., "Tissue protective properties of permanent hydrophilic surface modification of PMMA: Iris abrasion and endothelium damage studies," paper presented at the 2nd American-International Congress on Cataract, IOL and Refrative Surgery Meeting, Washington, DC, (1989). Goldberg, E.P., Unpublished Data from the research groups of Goldberg, E.P., (Dept. of Materials Science and Engineering, University of Florida) and Seeger, J.M, (Dept. of Cardiovascular Surgery, Veterans Administration Hospital), Gainesville, Florida, (1988-1995). Gore, R.w., Porous poly(tetrafluoroethylene) products, U.S. Patent #39,753, (1970). Gore, R.W., Porous poly(tetrafluoroethylene), u.s. Patent #369,814, (1975).
PAGE 216
199 Grabow, H.B., Early results of 500 cases of no-stitch cataract surgery, J. Cataract Refract. Surg., 17, 726730, (1991). Granke, K., Ochsner, J. L., McClugage, s. G., and Zdrahal, P., Analysis of graft healing in a new elastomer-coated vascular prosthesis, Cardiovascular surgery, 1(3), 254261, (1993). Greisler, H.P., Chapter 2 in New biologic and synthetic vascular prostheses, R. G. Landes Company, Austin, Texas, (1991). Habal, M. B., Luria, L. w. Carter, w. L., Kent, K., and Mician, M., Regarding the safety of silicone implants, J. Fla.-Med. Assoc., 78(6), 349-350, (1991). Hanson, s. R., and Harker, L.A., vascular thrombus formation, Annals NY Acad. Sci., 516, 653-661, (1987). Hasson, J.E., Mergerman, J., and Abbott, w. M., Increased compliance near vascular anastomoses, J. vase. Surg., 2(3), 419-423, (1985). Hebeish, A., Shalaby, S. E., and Bayazeed, A. M., Graft copolymerization of 2-methyl-5-vinyl pyridine to poly(ethylene terephthalate) fibres using a post radiation technique, J. App. Polym. Sci., 22, 3335-3342, (1978). Hercules, s. H., and Hercules, D. M., Surface characterization by electron spectroscopy for chemical analysis (ESCA}, Chapter 13 in Characterization of solid surfaces, Kane and Larrabee, eds., Plenum Press, New York, (1976). Hoffman, A. s., Synthetic polymeric biomaterials, Chapter 2 in Polymeric materials and artifical organs, American Chemical Society, Washington, DC, 13-29, (1984). Hoffman, A. s., Modification of material surfaces to affect how they interact with blood, Annals New York Academy of Sciences, 516, 96-101, (1987). Hoffman, A.S., Cohn, D., Hanson, S.R., Harker, L.A., Horbett, T.A., Ratner, B.D., and Reynolds, R.O., Applications of radiation-grafted hydrogels as blood-contacting biomaterials, Radiat. Phys. Chem. 22(1-2), 267-283, (1983). Hoffmeister, F. M., Biosurface properties of ocular implant polymers, Master's Thesis, University of Florida, (1988).
PAGE 217
200 Hufnagel, c. A., The use of rigid and flexible plastic prostheses for arterial replacement, Surgery, 37(2), 165174, (1955). Ip, w.F., and Sefton, M.V., Platelet consumption by NHLBI reference materials and silastic, Journal of Biomedical Materials Research, 25, 1321-1324, (1991). Ito, Y., Liu, S.Q., Nakabayashi, M. and Imanishi, Y., Cell growth on immobilized cell-growth factor. II. Adhesion and growth of fibroblast cells on poly(methyl methacrylate) membrane immobilized with proteins of various kinds, Biomaterials, 13, 789-794, (1992). Joo, C.K. and Kim, J.H., Compatibility of intraocular lenses with blood and connective tissue cells measured by cellular deposition and inflammatory response in vitro, J. Cataract Refract. Surg., 18, 240-246, (1992). Kale, P. D. and Lokhande, H. T., Grafting on polyester fibers, J. App. Polym. Sci., 19, 461-480, (1975). Katz, J., Kaufman, H. E., Goldberg, E. P., and Sheets, J. w., Prevention of endothelial damage from IOL insertion, Trans. Am. Acad. Ophthalmol. Otolaryngol., 83, 204, (1977). Kelly, R. P., Tunin, R., and Kass, D. A., Effect of reduced aortic compliance on cardiac efficiency and contractile function of in stitu canine left ventricle, Cirr. Res., 71(3), 490-502, (1992). Kerry, K. A., Barrett, A. M., Fouraker, B. D.,and Schanzlin, D. J., One-year results of the intrastromal corneal ring in nonfunctional human eyes, Arch. Ophthalmol., 113, 159167, (1995). Kircher, J. F., Sliemers, FA., Markle, R. A., Gager, w. B., and Leininger, R. I., On the degradation and electron spin resonance spectra of irradiated methacrylate polymers, J. Phys. Chem., 69, 189, (1965). Kishida, A., Iwata, H., Tamada, Y., and Ikada, Y., Cell behaviour on polymer surfaces grafted with non-ionic and ionic monomers, Biomaterials, 12(8), 786-792, (1991). Lane, s. s., and Lindstrom, R. L., Polysulfone intracorneal lenses, in Corneal surgery: theory, technique, and tissue, Part 4, Refractive corneal surgery, second edition, Brightbill, Ed., Mosby, Boston, (1990).
PAGE 218
201 Lentz, A.J., Horbet, T.A., Hsu, L., and Ratner, B.D., Rat peritoneal macrophage adhesion to hydroxyethyl methacrylate-ethyl methacrylate copolymers and hydroxystyrene-styrene copolymers, J. Biomed. Mater. Res., 19, 1101-1115, (1985). Levine, A. w., Bandyk, D. F., Bonier, P.H., and Towne, J. B., Lessons learned in adopting the in situ saphenous vein bypass, J. vase. Surg., 2(1), 145-153, (1985). Lim, F., Yang, c.z., and Cooper, S.L., Synthesis, characterization and ex vivo evaluation of polydimethyl siloxane polyurea-urethanes, Biomaterials 15(6), 408-416, (1994). Lin H.B., Hergenrother, R.W., Silver, J.H., Lin, J.C., Ljm, F. Brodhagen, T.W., & Cooper, S.L., Ex-Vivo blood compatibility of silicone-containing biomaterials, Journal of Materials Science: Materials in Medicine, 5 207-213, (1994). Lin, T.L., Hydrophilic surface modification of biomaterials, Ph.D. Dissertation, University of Florida, 1995. Litwak, P., ward, R. s., Robinson, A. J., Yilgor, I., and Spatz, c. A., Development of a small diameter, compliant, vascular prostheseis, Annals NY acad. Sci., 516, 1-6, (1987). Marcum, J.A., Atha, D.H., Fritze, L.M.S., Nawroth, P., Stern, D., & Rosenberg, R.D, Cloned bovine aortic endothelial cells synthesize anticoagulantly active heparan sulfate proteoglycan, J. Biol. Chem. 261, 7507-7517, (1986). Mccarey, B. E., Synthetic Keratophakia, Chapter 36 in Corneal surgery: theory, technique, and tissue, Part 4, Refractive corneal surgery, second edition, Brightbill, ed., Mosby, Boston, (1990). Mccarey, B. E. and Andrews, D. M., Refractive keratoplasty with intrastromal hydrogel lenticular implants, Invest. Ophthalmol. Vis. Sci., 21, 107-115, (1981). Mccarey, B. E. and Schmidt, F. H., Modeling glucose distribution in the cornea, Current eye research, 9(11), 1025-1039, (1990). McCoy, T.J., Wabers, H.S., and Cooper, S.L., Series shunt evaluation of polyurethane vascular graft materials in chronically AV-shunted canines, J. Biomed. Mater. Res., 24, 107-129, (1990).
PAGE 219
202 McDonald, M. B., Mccarey, B. E., Storie, B., Beuerman, R. w., Salmeron, B., van Rij, G., and Knight, P. M., Assesment of the long-ter111 corneal response to hydrogel intrastromal lenses implanted in monkey eyes for up to five years, J. Cataract Refract. Surg., 19, 213-222, (1993). Meaburn, G.M., Hosszu, J.L., and Coles C.M., Radiation grafting of methacrylates onto silicone rubber: Prototype burn wound dressing, Int. J. Appl. Radiat. and Iso., 29, 233-241, (1978). Mentak, K., Tissue-protective hydrophilic polymer solutions and surface modifications, Ph.D. Dissertation, university of Florida, (1993). Mergerntan, J., Hasson, J.E., Warnock, D. F., L'Italien, G. J., and w. M. Abbott, w. M., Noninvasive measurements of nonlinear arterial elasticity, Am. J. Physiol., 250, H181-Hl88, (1986). Miyauchi, M., and Shionoya, s., Complement activation by vascular prostheses and its role in progression of arteriosclerotic lesions, Angiology The Journal of vascular Diseases, 881-890, October, (1988). Morel, N. M. L., Dodge, A. B., Patton, w. F., Herman, I. M., Hechtman, H.B., and Shepro, D., Pulmonary microvascular endothelial cell contractility on silicone rubber substrate, J Cell. Phys., 141, 653, (1989) Moulder, J. F., Salvati, L., and Grobe, G. L., Surface characterization of polymer-based biomaterials, Medical Device and Diagnostic Industry, reprint, April, (1991). Mustard, J. F., Groves, H. M., Kinlough-Rathbone, R. L., and Packham, M.A., Thrombogenic and nonthrombogenic biological surfaces, Annals New York Acad. Sci., 516, 1221, (1987). Nair, P. D., Sreenivasan, K., and Jayabalan, M., Multiple gamma radiation sterilization of polyester fibres, Biomaterials, 9, 335-338, (1988). Nakamura, K., Refojo, M. F., Crabtree, D. v., and Leong, F. L., Analysis and fractionation of silicone and flurosilicone oils, Investigative Ophthalmology and Visual Science, 31 (10), 2059, (1990).
PAGE 220
203 Nakamura, K., Refojo, M. F., Crabtree, D. v., Pastor, J., and Leong, F. L., Ocular toxicity of low-molecular-weight components of silicone and flurosilicone oils for intraocular use, Investigative Ophthalmology and Visual Science, 32 (12), 3007, (1991). Nichols, w. w., and O'Rourke, M. F., eds., McDonald's blood flow in arteries: Theoretic, experimental and clinical principles, third edition, Lea & Febiger, New York, (1990). Nojiri, c., Park, K. D., Grainger D. w., Jacobs, H. A., Okano, T., Koyanagi, H., and Kim, s. w., In vivo nonthrombogenicity of heparin immobilized polymer surfaces, ASAIO Transactions, 36, Ml68-M172, (1990). Norgren, L., Hallberg, E., Jawien, A., Krantz, P., Parsson, H., Ribbe, E., and Thorne, J., Experimental evaluation of polymerized dacron grafts in the iliac position of pigs, Annals of vascular Surgery, 4, 575-579, (1990). Odian, G., Principles of Polymerization, 2nd ed., John Wiley & Sons, New York, (1981). Ofosu, F.A., Buchanan, M.R., Anvari, A., Smith, L.M., & Blajchman, M.A., Plasma anticoagulant mechanisms of heparin, heparin sulfate, and dermatan sulfate, Annals NY Acad. Sci. 556, 123-31, (1989) Omerod, M. G., and Charlesby, A., The radiation chemistry of some polysiloxanes: An ESR study, Polymer, 4, 459, (1963). Osborn, D. c., Intraocular lens implant polymer: biocompatibility and physical-chemical properties, Ph.D. Dissertation, University of Florida, (1985). Pevec, w. c., Darling, c., L'Italien, G. J., and Abbott, w. M., Femoropopliteal reconstruction with knitted, nonvelour Dacron versus expanded polytetrafluoroethylene, J. vase. Surg., 16(1), 60-65, (1992). Quantock, A. J., Kincaid, M. c., and Schanzlin, D. J., Stromal healing following explantation of an ICR (intrastromal corneal ring) from a nonfunctional human eye, Arch. Ophthalmol., 113, 208-209, (1995). Ranby, B. and Rabek, J. F., Photodegradation, photo-oxidation and photostabilization of polymers: Principles and applications, Wiley & Sons, New York, 1975.
PAGE 221
204 Ratner, B. D., society for Biomaterials 1992 Presidential Address: New ideas in biomaterials science a path to engineered biomaterials, J. Biomed. Mater. Res., 27, 837850, (1993). Ratner, B. D., Johnston, A. B., and Lenk, T. J., Biomaterial surfaces, J. Biomed. Mater. Res., 2l(Al), 59-90, (1987). Razzak, M.T., Otsuhata, K., and Tabata, Y., Radiation-induced grafting of N N-Dimethylacrylamide onto polytetra fluoroethylene, J. Appl. Polym. Sci., 33, 2345-2351, (1987). Rebenfeld, Y. L., and Weigmann, H. D., The in situ polymerization of vinyl monomers in polyester yarns, J. App. Polym. Sci., 22, 125-147, (1978). Rodriguez, F., Principles of Polymer Systems, second edition, Hemisphere Publishing Corporation, New York, (1982). Roslink, J., Modern Plastics Encyclopedia, McGraw-Hill, Inc., New York, (1990). Ruckenstein, I., and Gourisankar, s. v., A surface energetic criterion of blood compatibility of foreigh surfaces, J. Coll. Inter. Sci., 101(2), 436-451, (1984). Rye, R.R., Radiation hardening of PTFE against chemical ethcing, J. Polym. Sci. Polym. Phys. Ed., 26, 2133, (1988). Sayed, E., Hegazy, A., Ishigaki, I., and Okamoto, J., Radiation grafting of acrylic acid onto fluorine containing polymers. I. Kinetic study of preirradiation grafting onto poly(tetrafluoroethylene), J. Appl. Polym. Sci., 26, 3117-3121, (1981). Sheets, J. w., Hydrophilic polymer coatings to prevent tissue adhesion, Ph.D. Dissertation, University of Florida, (1983). Sherman, R.T., Jackson, T.M., Nielson, C.A., and Wilson H., Reconstruction of the common bile duct with and acrylate amide prosthesis, Annals of Surgery, 158(3), 420-430, (1963). Shoenfeld, N. A., Connolly, R., Ramberg, R. K., Baleri, c. R., Eldrup-Jorgensen, J., and Callow, A. D., The systemic activation of platelets by Dacron grafts, Surgery, Gynecology, & Obstetrics, 166, 454-457, (1988).
PAGE 222
205 Silver, J.H., Hergenrother, R.W., Lin, J.C., Lim, F., Lin H.B., Okada, T., Chaudhury, M.K., & Cooper, S.L. Surface and blood-contacting properties of alkylsiloxane monolayers supported on silicone rubber, Journal of Biomedical Materials Research, 29, 535-548, (1995). Sperling, L.H. Introduction to physical polymer science, John Wiley and Sons, Inc., New York, (1986). Sprugel, K. H., McPherson, J.M., Clowes, A. w., and Ross, R., Effects of growth factors in vivo. I. Cell ingrowth into porous subcutaneous chambers, Am. J. Pathol., 129, 601-613, (1987). Stannett, v., Grafting, Radiat. Phys. Chem., 18(1-2), 215222, (1981). Taylor, D.E., Comparative biotolerance of polyacrylamide agarose gel, silicone rubber, and microporous PTFE as soft tissue implants, Biomaterials, 7, 277-282, (1985). Todd, A., Pile irradiation of polyethylene terephthalate ('Terylene'), Nature, 174, 613, (1954) Tsai, c.c., Dollar, M.L., Constantinescu, A., Kulkarni, P.V., and Rberhart, R.C., Performance evaluation of hydroxylated and acylated silicone rubber coatings, Trans. Am. Soc. Artif. Inter. Organs, 37, M192-M193, (1991). Underwood, c. J., Tait, w. F., and Charlesworth, D., Design considerations for a small bore vascular prosthesis, The International Journal of Artificial Organs, 11(4), 272276, (1988). Waring, G. o., Classification and terminology, Chapter 35, in Corneal surgery: theory, technique, and tisue, second edition, Brightbill, ed., Mosby, Boston, (1990). werblin, T. P., Patel, A. A., and Barraquer, J. I., Initial human experience with Permalens myopic hydrogel intracorneal lens implants, Refractive and Corneal Surgery, 8, 23-26, (1992). Whalen, R.L., Cardona, R.R., & Kantrowitz, A., A new, all silicone rubber small vessel prosthesis, ASAIO Journal, 38, M207-M212, (1992). Whittemore, A. D., Clowes, A. w., Couch, N. P., and Mannick, J. A., Secondary femoropopliteal reconstruction, Ann. Surg., 193(1), 35-42, (1981).
PAGE 223
206 Wilson, J.E., and Jones, c. E., Thromboresistant polymers: a new approach, Biomat. Artif. Cells, Art. Org., 17(4), 437-445, (1989). Yahiaoui, A., Surface modification of intraocular lens polymers by hydrophilic graft polymerization for improved ocular implant biocompatibility, Ph.D. Dissertation, University of Florida, (1990). Yamamoto, F., and Yamakawa, s., Surface grafting of polyethylene by mutual irradiation in methyl acrylate vapor. VI. Kinetic analysis by quartz heliz microbalances, J. Polym. Sci: Polym. Chem. Ed., 18, 2257-2265, (1980).
PAGE 224
BIOGRAPHICAL SKETCH Christopher William Widenhouse was born in Wilmington, North Carolina to Gary and Marianne Widenhouse on September 26, 1967. He lived in Wilmington until graduating from John T. Hoggard High School in 1985. Chris attended North Carolina State University in Raleigh, North Carolina, where he pursued a Bachelor of Science degree in materials science and engineering with an emphasis on polymeric materials. During the course of his studies at NCSU, he experienced his first exposure to research by working in three different summer internship programs. The summer of 1986 was spent at Ames Laboratories at Iowa State University in Ames, Iowa, studying hydrogen embrittlement in niobium under the guidance of Dr. William Spitzig. He then worked for GTE Laboratories in Waltham, Massachusetts, in the 1987 Industrial Undergraduate Research Program (IURP) where he studied the processing parameters of Lanthana Doped Yttria under the guidance of Dr. Marina Pascucci. The summer of 1989 was spent working with The Procter & Gamble Company in Cincinnati, Ohio on the mechanical properties of stereolithography resins under the supervision of Bruce Johnson and Dr. Andy Wnuk. 207
PAGE 225
208 While at NCSU, Chris was a member of the varsity Fencing team, earning varsity letters in 1988 and 1989, and was named the ''Rookie of the Year'' in 1988. After graduating from NCSU in December of 1989, Chris was employed by The Procter & Gamble Company in the Corporate Packaging Development Division from January to August of 1990, again under the supervision of Dr. Wnuk. While at P&G, he did work which led to several patent applications in the area of microwave interactive packaging. In August of 1994, he was named as co-inventor on the patent entitled ''Microwave susceptor incorporating a coating material having silicate binder and an active constituent.'' In August of 1990, Chris left P&G to begin his graduate studies under the supervision of Dr. Eugene P. Goldberg at the University of Florida in Gainesville, Florida. While working in the area of polymeric biomaterials, he received a Master of Science degree in May 1995. While at UF, Chris also became a YMCA, CMAS, and UFADP scuba instructor, and taught scuba and fencing courses for the Department of Exercise and Sport Sciences in the College of Health and Human Performance. In his final year of graduate work, Chris married Tamara Shea Vetro on August 12, 1995.
PAGE 226
I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholar ly presentation and is fully adequate, in scope and quality, as a dissertation for the degr e e of Doctor of Pu=so y. Eugene v P. Gol berg, c~~rman Professor of Materials Science and Engineering I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. c_._;__ c~~-~ l :1. Christopher D. Batich Professor of Materials Science and Engineering I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of c or of Philo ophy. Antho Brennan Assoc a e Professor of Materials Science and Engineering I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctou.-~f Philoso~y. Richard B. Dickinson Assistant Professor Chemical Engineering
PAGE 227
I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Jcu,1 s M. Seego-t-, Professor of Cardiovascular Surgery This dissertation was submitted to the Graduate Faculty of the College of Engineering and to the Graduate School and was accepted as partial fulfillment of the requirements for the dissertation of Doctor of Philosop~h'i--, May, 1996 Winfred M. Phillips Dean, College of Engineering Karen A. Holbrook Dean, Graduate School
PAGE 228
LD 1780 199 ,,, UNIVERSITY OF FLORIDA 11 11 1 I Il l I 1 11 11 11111 11 11 11111 I 3 1262 08554 9011
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ED99ZQE8B_8QJCGP INGEST_TIME 2013-01-23T15:22:45Z PACKAGE AA00012925_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
|
