close

Вход

Забыли?

вход по аккаунту

?

Microwave plasma reactions on polymer surfaces: Spectroscopic studies

код для вставкиСкачать
/
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UME
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may be
from any type of computer printer.
The q u a lity of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely afreet reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in reduced
form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6” x 9” black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly to
order.
UMI
A Bell & Howell Information Company
300 North Zeeb Road, Ann Arbor MI 48106-1346 USA
313/761-4700 800/521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MICROWAVE PLASMA REACTIONS ON POLYMER SURFACES:
SPECTROSCOPIC STUDIES
A Dissertation
Submitted to the Graduate Faculty
of the
North Dakota State University
of Agriculture and Applied Science
By
Heung Kim
In Partial Fulfillment of the Requirements
for the Degree of
DOCTOR OF PHILOSOPHY
Major Department:
Polymers and Coatings
October 1997
Fargo, North Dakota
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number:
9813175
UMI Microform 9813175
Copyright 1998, by UMI Company. AH rights reserved.
This microform edition is protected against unauthorized
copying under Title 17, United States Code.
UMI
300 North Zeeb Road
Ann Arbor, MI 48103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
North Dakota State University
Graduate School
Title
MICROWAVE PLASMA REACTIONS ON POLYMER SURFACES:
SPECTROSCOPIC STUDIES
By
HEUNG SOO RIM
The Supervisory Committee certifies that this disquisition complies with North Dakota
State University’s regulations and meets the accepted standards for the degree of
DOCTOR OF PHILOSOPHY
SUPERVISORY COMMITTEE:
Dr. Marek. W. Urban
Dr. Mark D. Soucek
llcJt to,
%Z1 fjlZ j
Dr. Michael Page
Dr. Jagdxsh Singh
\l
*
—
Approved by Department Chain
Ob\s\,mi
Date
Signature
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
Kim, Heung, Ph.D., Department of Polymers and Coatings, College o f Science
and Mathematics, North Dakota State University, October 1997. Microwave
Plasma Reactions on Polymer Surfaces: Spectroscopic Studies. Major Professor:
Dr. Marek W. Urban.
The primary objective of these studies was to understand molecular level
processes occurring on polymer surfaces resulting from microwave plasma
reactions. A particular emphasis was given to attenuated total reflectance
Fourier transform infrared (ATR FT-ER) spectroscopy and atomic force
microscopy (AFM) as analysis tools.
These studies showed that imidazole reaction mechanisms in the plasma state
depend on the plasma reactors, discharge gases, and substrates employed for
surface reactions. ATR FT-IR spectroscopy and AFM measurements revealed
that, under closed Ar microwave plasma reactor, multilayers of imidazole rings
were formed on the polydimethylsiloxane (PDMS) surface through hydrogen
abstraction of the N-H bonds of imidazole entities. In an open flow Ar
microwave plasma reactor, imidazole rings opened to form CsN species on the
PDMS surface. Analysis of ATR FT-IR spectroscopic data indicated that
reactive discharge gases altered microwave plasma reaction mechanisms of
imidazole molecules. 0 2 and C 02 discharge gases resulted in the formation of
Si-0-CH2 and CH2-N linkages on the PDMS surface for imidazole reactions,
respectively. A new method for reacting imidazole molecules on polyurethane
(PU) surfaces was developed. PU specimen was immersed in imidazole
containing CH2Cl2 solution and exposed to microwave plasma. ATR FT-IR
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
analysis showed that an imidazole reaction occurred on the PU surface through
the C=N opening o f imidazole entities. Surface reactions on polyvinylchloride
(PVC) depend on a thermal history o f the substrate. Imidazole did not react to
hot-pressed PVC. For the solvent-cast PVC with lower surface crystalline
content, the imidazole reaction occurred through a C=C opening o f imidazole
entities. Heparin (HP) immobilization on the PVC surface was studied using
polarized ATR FT-IR spectroscopy. These studies showed that HP was
perpendicularly oriented to form covalent linkages on the PVC surface when
immobilized by direct solution exposure (dipping). When spin-coated, HP was
parallelly orientated to form ionic complexes on the PVC surface. It was also
shown that by changing shear rates o f deposition, it is possible to control ionic
and covalent nature of HP bonding.
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGMENTS
I would like to express my sincere thanks to Dr. Marek W. Urban for his
guidance, encouragement, and financial support throughout my years at North
Dakota State University. I would also like to thank Dr. Mark D. Soucek, Dr.
Michael Page, and Dr. Jagdish Singh for serving on my graduate committee and
all of my colleagues who have helped through their cooperation and friendship.
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEDICATION
This dissertation is dedicated to my parents and my son, Dong.
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
Page
APPROVAL PAGE............................................................................................. ii
ABSTRACT........................................................................................................ iii
ACKNOWLEDGMENTS................................................................................... v
DEDICATION.................................................................................................... vi
LIST OF TABLES............................................................................................. xii
LIST OF FIGURES.......................................................................................... xiii
INTRODUCTION................................................................................................ 1
References...................................................................................................... 4
CHAPTER 1. REACTIONS ON POLYMER SURFACES INDUCED
BY PLASMONS.......................................................................... 6
1.1 Introduction........................................................................................... 7
1.2 Non-film Forming Plasma Reactions of Siloxane Elastomers............ 11
1.2.1 Non-reactive Gas Plasma Reactions.....................................................12
1.2.2 Reactive Gas Plasma Reactions............................................................14
1.2.3 Solid Phase Plasma Reactions............................................................. 17
1.3 Film Forming Plasma Reactions of Siloxane Elastomers................. 22
1.4 Hydrophobic Recovery of Plasma Reacted Siloxane Elastomers
23
1.5 References........................................................................................... 25
CHAPTER
2.
MICROWAVE
PLASMA
REACTIONS
OF IMIDAZOLE ON POLYDIMETHYLSILOXANE
ELASTOMER
SURFACES:
A
SPECTROSCOPIC
STUDY........................................................................................ 31
2.1
Introduction......................................................................................... 32
2.2 Experimental....................................................................................... 33
vii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.2.1 Substrate Preparation.......................................................................... 33
2.2.2 Surface Reactions............................................................................... 34
2.2.3 Spectroscopic Measurements............................................................. 36
2.3
Results and Discussions..................................................................... 37
2.3.1 Analysis o f Imidazole Plasma Reacted PDMS..................................37
2.3.2 Discharge Time Effect.......................................................................40
2.3.3 Silica Inhibition Effect.......................................................................43
2.3.4 Polarization Effect.............................................................................48
2.3.5 Closed and Open Flow Reactors........................................................ 51
2.4
Conclusions........................................................................................ 55
2.5
References.......................................................................................... 56
CHAPTER 3. REACTION SITES ON POLYDIMETHYLSILOXANE
ELASTOMER
PLASMA
SURFACES
REACTIONS
IN
WITH
MICROWAVE
GASEOUS
IMIDAZOLE: A SPECTROSCOPIC STUDY......................... 58
3.1
Introduction........................................................................................ 59
3.2
Experimental...................................................................................... 60
3.2.1 Substrate Preparation......................................................................... 60
3.2.2 Surface Reactions............................................................................... 60
3.2.3 Spectroscopic Measurements..............................................................61
3.3
Results and Discussions......................................................................61
3.3.1 Reaction Sites on PDMS Surfaces......................................................61
3.3.2 Formation of Si-CH2 Linkages...........................................................62
3.3.3 Imidazole Reaction Mechanisms on PDMS Surface......................... 64
3.4
Conclusions.........................................................................................70
3.5
References...........................................................................................72
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER
4.
MICROWAVE
PLASMA
REACTIONS
OF IMIDAZOLE ON POLYDIMETHYLSILOXANE
ELASTOMER SURFACES: ATTENUATED TOTAL
REFLECTANCE
FT-IR
AND
ATOMIC
FORCE
MICROSCOPIC STUDIES.........................................................73
4.1
Introduction......................................................................................... 74
4.2
Experimental....................................................................................... 75
4.2.1 Substrate Preparation.......................................................................... 75
4.2.2 Surface Reactions................................................................................ 76
4.2.3 Spectroscopic Measurements.............................................................. 77
4.2.4 Atomic Force Microscopy................................................................... 77
4.3
Results and Discussions...................................................................... 77
4.3.1 Microwave Plasma Reactions o f Imidazole on PDMS Surface
77
4.3.2 Silica Effect on Surface Morphology..................................................82
4.3.3 Imidazole Reaction Effect on Surface Morphology.......................... 87
4.4
Conclusions......................................................................................... 92
4.5
References........................................................................................... 94
CHAPTER
5.
THE
ON
EFFECT
OF
MICROWAVE
DISCHARGE
PLASMA
GASES
REACTIONS
OF IMIDAZOLE ON POLYDIMETHYLSILOXANE
(PDMS) SURFACES: QUANTITATIVE ATR FT-IR
SPECTROSCOPIC ANALYSIS...........................................96
5.1
Introduction......................................................................................... 97
5.2
Experimental....................................................................................... 98
5.2.1 Substrate Preparation.......................................................................... 98
5.2.2 Surface Reactions................................................................................ 99
5.2.3 Spectroscopic Measurements..............................................................99
5.3
Results and Discussions.................................................................... 100
ix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.3.1 Ar Microwave Plasma Reactions..................................................... 100
5.3.2 O2 Microwave Plasma Reactions..................................................... 109
5.3.3 C 0 2 Microwave Plasma Reactions.................................................. 117
5.4
Conclusions.......................................................................................129
5.5
References.........................................................................................130
CHAPTER
6.
OF
MICROWAVE
PLASMA
IMIDAZOLE
ELASTOMER
ON
SURFACES:
REACTIONS
POLYURETHANE
A
SPECTROSCOPIC
STUDY......................................................................................132
6.1
Introduction.......................................................................................133
6.2
Experimental.....................................................................................134
6.2.1 Sample Preparation...........................................................................134
6.2.2 Spectroscopic Measurements........................................................... 134
6.3
Results and Discussions................................................................... 135
6.3.1 Analysis of Imidazole Plasma Reacted PU...................................... 135
6.3.2 Imidazole Reaction Mechanism on PU Surfaces............................. 142
6.3.3 Depth Profiling of Imidazole Reacted P U ....................................... 145
6.3.4 Quantitative Analysis of Imidazole Reacted PU.............................. 147
6.4
Conclusions.......................................................................................152
6.5
References.........................................................................................153
CHAPTER
7.
OF
MICROWAVE
IMIDAZOLE
PLASMA
ON
REACTIONS
POLYVINYLCHLORIDE
SURFACES: A SPECTROSCOPIC STUDY.......................... 155
7.1
Introduction....................................................................................... 156
7.2
Experimental...................................................................................... 157
7.2.1 Substrate Preparation........................................................................ 157
7.2.2 Surface Reactions............................................................................. 157
7.2.3 Spectroscopic Measurements........................................................... 157
x
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7.3
Results and Discussions.................................................................. 158
7.3.1 Analysis of Imidazole Plasma Reacted Hot-pressed PVC.............. 158
7.3.2 Analysis of Imidazole Plasma Reacted Solvent-cast PVC.............. 161
7.3.3 Depth Profiling of Imidazole Reacted PVC..................................... 166
7.3.4 Quantitative Analysis of Imidazole Reacted PV C .......................... 168
7.4
Conclusions..................................................................................... 168
7.5
References....................................................................................... 171
CHAPTER
8.
REACTIONS
OF
MULTI-LAYERED
THROMBRESISTANT
THIN
POLYVINYLCHLORIDE
(PVC)
FILMS
ON
SURFACES:
A SPECTROSCOPIC STUDY............................................ 173
8.1
Introduction..................................................................................... 174
8.2
Experimental................................................................................... 176
8.2.1 Substrate Preparation....................................................................... 176
8.2.2 Surface Reactions............................................................................ 177
8.2.3 Spectroscopic Measurements.......................................................... 178
8.3
Results and Discussions.................................................................. 179
8.3.1
Multi-layered Thin Films Obtained by Direct Solution
Exposure (Dipping)...........................................................................179
8.3.2 Quantitative Analysis...................................................................... 189
8.4
Multi-layered Thin Films Obtained by Spin-coating...................... 202
8.4.1
Orientation of HP Thin Films......................................................... 206
8.4.2
Quantitative Analysis..................................................................... 211
8.5
Conclusions.................................................................................... 216
8.6
References...................................................................................... 218
CONCLUDING REMARKS........................................................................... 220
xi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
Table
Page
7.1 Tentative Band Assignments o f Hot-pressed and Solvent-cast
PVC.................................................................................................... 159
8.1 IR bands and their tentative assignments........................................... 181
8.2 Extinction coefficients of IR characteristic band for PEI, DS,
and H P .............................................................................................. 190
8.3 A schematic representation of relative film thickness of multi­
layered thin films............................................................................. 217
xii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
Figure
Page
1.1 Schematic representation of plasma reactions......................................9
1.2 Proposed reaction mechanisms of reactive gas plasma on PDMS......15
1.3 Proposed reaction mechanism of maleic anhydride and acryl
amide reacted PDMS in the presence of Ar microwave plasma..........19
1.4 Proposed reaction mechanism of imidazole reacted PDMS in the
presence o f microwave plasma........................................................... 20
2.1
A schematic diagram illustrating closed and open flow
microwave plasma reactors................................................................ 35
2.2 ATR FT-IR spectra in the 2300-1300 cm'1 region of imidazole
reacted to PDMS surface under various pressure conditions
using a closed reactor...........................................................................38
2.3
ATR FT-IR spectra in the 3400-2800 cm*1 region of imidazole
reacted to PDMS surface under various pressure conditions
using a closed reactor.......................................................................... 39
2.4
Schematic representation of imidazole reacted to PDMS surface
in microwave plasma reactions........................................................... 41
2.5
ATR FT-IR spectra in the 2300-1300 cm'1 region of imidazole
reacted to PDMS surface under various discharge time
conditions using a closed reactor........................................................ 42
2.6
ATR FT-IR spectra in the 2300-1300 cm'1 region of imidazole
reacted to PDMS surface containing 5% (w/w) silica under
various pressure conditions using a closed reactor............................. 44
xiii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.7
ATR FT-IR spectra in the 2300-1300 cm’1 region of imidazole
reacted to PDMS surface containing 5% (w/w) silica under
various discharge time conditions using a closed reactor................... 46
2.8
ATR FT-IR spectra in the 1900-900 cm’1 region of imidazole
reacted to silica film surface using a closed reactor........................... 47
2.9
Polarized ATR FT-IR spectra in the 2300-1300 cm’1 region of
imidazole reacted to PDMS surface under various pressure
conditions using a closed reactor........................................................ 49
2.10 Polarized ATR FT-IR spectra in the 3400-2800 cm'1 region of
imidazole reacted to PDMS surface under various pressure
conditions using a closed reactor........................................................50
2.11 ATR FT-IR spectra in the 2300-1300 cm’1 region of imidazole
reacted to PDMS surface under various pressure conditions
using an open flow reactor..................................................................52
2.12 ATR FT-IR spectra in the 2300-1300 cm’1 region of imidazole
reacted to PDMS surface under various discharge time
conditions using an open flow reactor................................................54
3.1
ATR FT-IR spectra in the 2300-1300 cm*1 region of PDMS
surface in Ar microwave plasma under various discharge time
conditions using closed reactor...........................................................63
3.2 ATR FT-IR spectra in the 1200-950 cm’1 region of PDMS
surface in Ar microwave plasma under various discharge time
conditions using closed reactor........................................................... 65
3.3 Reaction mechanism of PDMS surface in Ar microwave plasma......66
xiv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.4 ATR FT-IR spectra in the 2300-1300 cm'1 region of PDMS
surface and imidazole reacted to PDMS surface in Ar
microwave plasma under various discharge time conditions
using closed reactor.............................................................................68
3.5
ATR FT-IR spectra in the 1200-950 cm'1 region of PDMS
surface and imidazole reacted to PDMS surface in Ar
microwave plasma under various discharge time conditions
using closed reactor.............................................................................69
3.6
Reaction mechanism of imidazole on PDMS surface in Ar
microwave plasma.............................................................................. 71
4.1
ATR FT-IR spectra in the 2300-1300 cm'1 region of imidazole
reacted to PDMS surface under various discharge time
conditions using a closed reactor........................................................79
4.2
Polarized ATR FT-IR spectra in the 2300-1300 cm'1 region of
imidazole reacted to PDMS surface under various pressure
conditions using a closed reactor........................................................ 80
4.3 ATR FT-IR spectra in the 2300-1300 cm'1 region of imidazole
reacted to PDMS surface containing 5% (w/w) silica under
various discharge time conditions using a closed reactor....................81
4.4
ATR FT-IR spectra in the 1900-900 cm'1 region of imidazole
reacted to silica film surface using a closed reactor............................83
4.5 Surface views of PDMS imaged using Atomic Force Microscope.... 85
4.6 Surface views of PDMS with 5% (w/w) silica imaged using
Atomic Force Microscope.................................................................. 86
4.7 ATR FT-IR spectra in the 2300-1300 cm'1 region of imidazole
reacted to PDMS surface under various pressure conditions
using an open flow reactor.................................................................. 88
xv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.8
AFM image of the surface o f the imidazole reacted on PDMS
under 26.6 Pa for 10 sec discharge time in closed reactor
conditions............................................................................................. 90
4.9
AFM image of the surface of the imidazole reacted on PDMS
under 26.6 Pa for 10 sec discharge time in open flow reactor
conditions............................................................................................. 91
4.10 A schematic representation of PDMS surface structure resulting
from imidazole reaction in the presence of Ar microwave plasma.... 93
5.1a ATR FT-IR in the 3300-2700 cm'1region for imidazole-PDMS
in the presence of Ar microwave plasma under various discharge
times using closed reactor conditions................................................101
5.1b ATR FT-IR spectra in the 1800-1300 cm*1region for imidazole PDMS in the presence of Ar microwave plasma under various
discharge times using closed reactor conditions................................102
5.1c ATR FT-IR spectra in the 1470-1360 cm'1region for imidazolePDMS in the presence of Ar microwave plasma under various
discharge times using closed reactor conditions................................103
5.Id ATR FT-IR spectra in the 1200-950 cm'1for imidazole-PDMS
in the presence of Ar microwave plasma under various discharge
times using closed reactor conditions................................................104
5.2
Proposed reaction mechanisms of imidazole reactions on PDMS
in the presence of Ar microwave plasma........................................... 106
5.3 Band intensity changes of the imidazole ring stretching band at
1663 cm'1plotted as a function of concentration...............................108
5.4 Surface concentration changes of imidazole reacted to the PDMS
surface in the presence of Ar microwave plasma plotted as a
function of discharge times.............................................................. 110
xv i
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.5a ATR FT-IR spectra in the 3300-2700 cm'1region for imidazolePDMS in the presence of O2 microwave plasma under various
discharge times using closed reactor conditions................................I l l
5.5b ATR FT-IR spectra in the 1800-1300 cm'1region for imidazolePDMS in the presence of O2 microwave plasma under various
discharge times using closed reactor conditions................................112
5.5c ATR FT-IR spectra in the 1470-1360 cm'1region for imidazolePDMS in the presence of O2 microwave plasma under various
discharge times using closed reactor conditions................................113
5.5d ATR FT-IR spectra in the 1200-950 cm'1 region for imidazolePDMS in the presence of O2 microwave plasma under various
discharge times using closed reactor conditions................................114
5.6 Surface concentration changes of imidazole reacted to the PDMS
surface in the presence of O2 microwave plasma plotted as a
function o f discharge tim e.................................................................118
5.7
Proposed reaction mechanisms of imidazole on the PDMS
surface in the presence of O2 microwave plasma under various
discharge time conditions..................................................................119
5.8a ATR FT-IR spectra in the 3300-2700 cm'1region for imidazolePDMS in the presence of CO2 microwave plasma under various
discharge times using closed reactor conditions................................120
5.8b ATR FT-IR spectra in the 1800-1300 cm'1region for imidazole PDMS in the presence of CO2 microwave plasma under various
discharge times using closed reactor conditions................................121
5.8c ATR FT-IR spectra in the 1450-1360 cm*1region for imidazole PDMS in the presence of CO2 microwave plasma under various
discharge times using closed reactor conditions................................122
xvii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.8d ATR FT-IR spectra in the 1200-950 cm'1 region for imidazolePDMS in the presence of CO2 microwave plasma under various
discharge times using closed reactor conditions................................123
5.9 Proposed reaction mechanisms of imidazole reactions on the
PDMS surface in the presence of CO2 microwave plasma................126
5.10 Surface concentration changes of imidazole reacted to the PDMS
surface in the presence of CO2 microwave plasma plotted as a
function of discharge time................................................................ 128
6.1 Urethane linkages containing soft and hard segments......................136
6.2
ATR FT-IR spectra in the 4000-2000 cm'1 region of imidazoleabsorbed PU in the presence of Ar microwave plasma under
26.6 Pa in closed reactor..................................................................137
6.3 ATR FT-IR spectra in the 4000-2000 cm'1 region of imidazolePU in the presence of Ar microwave plasma under various
discharge pressure in closed reactor.................................................. 139
6.4
ATR FT-IR spectra in the 1800-1400 cm'1 region of imidazolePU in the presence of Ar microwave plasma under various
discharge pressure in closed reactor.................................................. 141
6.5
ATR FT-IR spectra in the 1400-1000 cm'1 region of imidazolePU in the presence of Ar microwave plasma under various
discharge pressure in closed reactor.................................................. 143
6.6
Reaction mechanism of imidazole reacted to PU in the presence
of Ar microwave plasma................................................................... 144
6.7
ATR FT-IR spectra in the 1800-1400 cm'1 region of imidazolePU in the presence of Ar microwave plasma under closed
reactor conditions using various incidence angle of infrared light... 146
xviii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6.8 ATR FT-IR spectra in the 1800-1400 cm'1 region of imidazolePU in the presence of Ar microwave plasma under open flow
reactor conditions using various incidence angle of infrared light... 148
6.9 Plot of C=C stretching band absorbance for imidazole as a
function of concentration...................................................................149
6.10 Plots of volume concentration for imidazole reacted to PU as a
function of depth of penetration........................................................151
7.1 ATR FT-IR spectra in the 1900-500 cm'1 region of imidazole
reacted to hot-pressed PVC in the presence of oxygen
microwave plasma under various pressure conditions using a
closed reactor..................................................................................... 160
7.2 ATR FT-IR spectra in the 3700-2800 cm'1 region of imidazole
reacted to solvent-cast PVC in the presence of oxygen
microwave plasma under various pressure conditions using a
closed reactor..................................................................................... 162
7.3 ATR FT-IR spectra in the 1850-1000 cm'1 region of imidazole
reacted to solvent-cast PVC in the presence of oxygen
microwave plasma under various pressure conditions using a
closed reactor..................................................................................... 164
7.4 Proposed mechanism of imidazole reactions to solvent-cast PVC
surfaces in the presence o f oxygen plasma....................................... 165
7.5 ATR FT-IR spectra in the 1850-1000 cm'1 region of imidazole
reacted to solvent-cast PVC in the presence of oxygen plasma
under 26.6 Pa/5 sec using various incidence angle of infrared
light................................................................................................... 167
7.6 Plot of absorbance of the imidazole ring stretching band as a
function of imidazole concentrations in KBr powder........................169
xix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7.7
Plots of surface concentration for imidazole reacted to PVC
surfaces as a function of depth of penetration...................................170
8.1
Chemical structures used in surface reactions and schematic
representation of multi-layered structures deposited on PVC.......... 175
8.2 ATR FT-IR spectra in the S-O stretching region for ammonium
persulfate reacted to PVC surface...................................................... 180
8.3
ATR FT-IR spectra in the N-H stretching region for PEI reacted
to PVC surface using direct solution exposure (dipping)................. 183
8.4
ATR FT-IR spectra in the O-H bending region for DS reacted to
PVC surface using direct solution exposure (dipping)..................... 185
8.5
ATR FT-IR spectra in the O-H bending region for HP reacted to
PVC surface using direct solution exposure (dipping)..................... 186
8.6
Proposed reaction mechanism of HP immobilization using direct
solution exposure (dipping)............................................................... 188
8.7a Surface concentration changes of PEI on the PVC surface as a
function of pH obtained by direct solution exposure (dipping)........ 192
8.7b Surface concentration changes of PEI on the PVC surface as a
function of crosslinker content obtained by direct solution
exposure (dipping).............................................................................193
8.7c Surface concentration changes of PEI on the PVC surface as a
function of solution concentration obtained by direct solution
exposure (dipping).............................................................................194
8.8a Surface concentration changes of DS on the PVC surface as a
function of pH obtained by direct solution exposure (dipping)
197
8.8b Surface concentration changes of DS on the PVC surface as a
function
of solution concentration obtained by direct solution
exposure (dipping).............................................................................198
xx
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8.9a Surface concentration changes of HP on the PVC surface as a
function of pH obtained by direct solution exposure (dipping)
200
8.9b Surface concentration changes of HP on the PVC surface as a
function o f solution concentration obtained by direct solution
exposure (dipping)........................................................................... 201
8.10 ATR FT-IR spectra in the O-H bending region for HP reacted
to the PVC surface using spin-coating............................................ 203
8.11 ATR FT-IR spectra in the O-H stretching region for HP reacted
to the PVC surface for various shear rates...................................... 205
8.12 ATR FT-ER spectra in the O-H bending region for HP reacted
to the PVC surface for various shear rates..................................... 207
8.13 Linear absorptivity of N-H deformation band for HP reacted to
the PVC surface for various shear rates.......................................... 208
8.14 ATR FT-IR spectra in the O-H bending region for HP reacted
to the PVC surface........................................................................... 210
8.15 Proposed reaction mechanism of HP immobilization using spincoating............................................................................................. 212
8.16a Surface concentration changes of PEI on the PVC surface as a
function of solution concentration obtained by spin-coating......... 213
8.16b Surface concentration changes of DS on the PVC surface as a
function of solution concentration obtained by spin-coating.......... 214
8.16c Surface concentration changes of HP on the PVC surface as a
function of solution concentration obtained by spin-coating
xxi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
215
INTRODUCTION
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
One of the most widely used biopolymers is silicone elastomers. Although their
chemically inertness and nontoxic nature are desirable for biologically active
environments, surface modification is often necessary to obtain biocompatibility.1'3
Numerous forms of energy are available for conducting surface modifications.
Among them, microwave energy generated plasmons are an effective source for
reacting monomers to elastomeric surfaces. The advantage of microwave plasma
reactions lies in their ability to alter the surface properties while not affecting the
bulk polymer properties.4
Although one of the drawbacks in plasma reactions is the complexity of the
reaction mechanisms, the advantages are overwhelming. For this reason, various
plasma reactions to functionalize crosslinked polydimethylsiloxane (PDMS) surface
have been studied.5'8 Chapter 1 provides a literature review of plasma chemistry and
its application to polymer surface modifications. Chapter 2 examines microwave
plasma reactions of imidazole on the PDMS surfaces using attenuated total
reflectance (ATR) Fourier transform infrared (FT-IR) spectroscopy. Surface
reactivity and orientation of surface species and how closed and open flow
microwave plasma reactions affect imidazole reaction mechanisms are investigated.
While recent studies of plasma reactions focused on detecting newly formed surface
species, there are no data as to the reactive sites of the substrate available for plasma
reactions. Chapter 3 examines the formation of reactive sites on the PDMS surface,
which provides detailed imidazole reaction mechanisms on the PDMS surfaces.
Although ATR FT-IR spectroscopy is a valuable tool in detecting newly formed
surface species in the plasma state, surface morphology of PDMS and accessibility
2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
for imidazole reactions remained a considerable interest to be solved. Chapter 4
concentrates on the inhibition mechanism of silica in imidazole/PDMS plasma
reactions and morphological changes of PDMS surfaces resulting from imidazole
reactions when closed and open flow plasma reactions are employed. For this
purpose, atomic force microscope (AFM) measurements are utilized. Microwave
plasma reactions of imidazole conducted under Ar, 0 2, or C 02 are expected to yield
significant differences in surface reactions. Chapter 5 focuses on addressing how
discharge gases affect reaction mechanisms of imidazole on the PDMS surfaces.
Furthermore, quantitative algorithm9' 10 of ATR FT-IR spectroscopy is used to
determine the extent of imidazole entities on the PDMS surfaces.
In Chapters 6 and 7, microwave plasma reactions of imidazole on polyurethane
(PU) elastomer and polyvinylchloride (PVC) surfaces are investigated. The primary
interest in conducting these studies comes from the difficulties to modify their
i
|
surfaces for numerical biomedical applications. Chapter 6 illustrates a novel method
i
\
of microwave plasma reaction of imidazole. Imidazole molecules are incorporated
|
into a PU network before plasma reactions. Using quantitative ATR FT-IR
i
spectroscopy, imidazole contents chemically attached to the PU surfaces at various
?
depths from the surface are examined. Microwave plasma reactions of imidazole are
|
inhibited in the hot-pressed PVC surfaces due to a high crystalline content. On the
t
\
|
other hand, imidazole reactions occurs on the solvent-cast PVC surfaces.
Chapter 7 examines the effect o f PVC surface morphology on the imidazole/PVC
plasma reactions.
To obtain thrombresistant surfaces, heparin (HP) is utilized to immobilize on the
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
PVC surfaces.11 Among attempts to increase thrombresistant activity, the formation
of the multi-layered structures on the PVC surfaces are proposed, and it increases
surface charge density and functional groups for HP immobilization.11' 12 Using
quantitative ATR FT-IR spectroscopy, Chapter 8 focuses the effect of reaction
parameters on the formation o f polyethyleneimine (PEI) and dextran sulfate (DS)
layers. Reaction mechanisms responsible for HP immobilization on the PVC
surfaces are also proposed when direct solution exposure (dipping) and spin-coating
are employed as an deposition methods.
References
1. M. Millard, J. J. Windle, and A. E. Pavlath, J. Appl. Polym. Sci., 1973,17,2501.
2. M. Morra, E. Occhiello, R. Marola, F. Garbassi, P. Humphrey, and P. Johnson, J.
Colloid Interface Sci., 1990, 11, 137.
3. M. J. Owen and P. J. Smith, J. Adhesion Sci. Technol., 1994, 8, 1063.
4. H. Yasuda, Plasma Polymerization, Academic Press, New York, 1985.
5. S. R. Gaboury and M. W. Urban, Polym. Commun., 1991, 32(13), 390.
6. S. R. Gaboury and M. W. Urban, Polymer, 1992, 33(23), 5085.
7. S. R. Gaboury and M. W. Urban, Langmuir, 1993,9, 3225.
8. S. R. Gaboury and M. W. Urban, Langmuir, 1994, 10, 2289.
9. M. W. Urban, Attenuated Total Reflectance Spectroscopy o f Polymers Theory and Practice, American Chemical Society, Washington, DC, 1996;
and ref. therein.
10. M. W. Urban, Vibrational Spectroscopy o f Molecules and Macromolecules
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
on Surfaces, Wiley - Intersciences, New York, 1993.
11. O. P. Larin, R. Larsson, and P. Olsson, Biomat., Med. Derv., Art. Org., 1983,
11
, 161.
12. O. P. Larm, L. A. Adolfsson, and K. P. Olson, US Patent 5,049,403,1991.
5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 1
REACTIONS ON POLYMER SURFACES INDUCED BY PLASMONS
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.1 Introduction
Plasma chemistry is concerned with the occurrence of chemical reactions utilizing
partially ionized gases composed of electrons, ions, free radicals, and neutral species
in the ground and excited states. 1'5 The plasma state of matter can be produced
through the action o f either very high temperatures or strong electric discharges. Due
to the fact that electric discharge can be controlled easier and has virtually little
effect on bulk properties, it became the primary method of interest. Because most
organic compounds decompose at high temperatures, low temperature plasmas,
typically less than 500 K, are beneficial for conducting surface chemical reactions
withoutthermaldegradation of bulk properties. 6*9
that exists within the plasmaphase, these
Due to thermal nonequilibrium
types of plasmas are also known as
nonequilibrium plasmas. To obtain a low temperature plasma phase, gases at
reduced pressures are exposed to electromagnetic radiation, usually radio or
microwave frequencies. 10 The energy is transferred from the electric field to the gas
phase species, leading to the formation of ions and radicals. 11
Even for most stable polymers, electric discharge is sufficient enough to create a
high density of radicals on the surface of organic materials. 12*16 Although the initial
step of creating plasma from organic compounds is ionization, all chemical reactions
occur through the formation of free radicals. Formation of radicals from ions can be
represented by the following reactions: 17
R -H
Dissociation of ion:
-►
H* +
Rt
n
n
„ . , n +
Rl-R2----------►
Ri* + R2,
7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In the plasma state, the elimination of hydrogen or methyl groups from the
molecular ion leaving carbonium ion has been suggested. 3,17
Subsequent
neutralization of the carbonium ion leads to the formation of radicals:
R+
+
e ---------- ►
R»
Neutralization of ion:
R-2 t + e ------- ► R2 *
These radicals are able to either react with each other or with other species in the
plasma state.
Among many surface reactions, plasma reactions on polymer surfaces are
particularly attractive because they may alter properties such as hydrophilicity, 18
surface friction, 19' 21 bondability, 22' 24 and others, 25 while maintaining bulk polymer
properties. The most efficient and clean use o f plasma reactions appears to provide
surface modifications, where the most outer macromolecular layers are altered to
create new surface properties. 26"29 For the most part, the effect of plasma reactions is
confined to the first 10 pm from the surface. 3 Because of the surface selectivity,
there are many areas of applications, among which surface modifications for
biocompatibility enhancement have been of a primary interest. 30" 40
When a substrate is expose to plasma, three basic plasma reactions may occur. As
described in Figure 1.1: (1) plasma polymerization, (2) plasma modification, and (3)
plasma ablation or etching. All three plasma reactions may occur simultaneously. It
is possible to control which reaction dominates the process by varying conditions of
discharge, monomer and its concentration, and the substrate. In a typical experiment,
starting monomers are rapidly converted in discharge to many reactive fragments
since the energy levels of plasma are high enough to break any bonds. These
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Monomer Gas
or Vapor
Gas Products
Plasma
Plasma
Modification
Polymer-Forming
Intermediate
Ablation
or
Etching
Vapor Phase
Modified Surface
Solid Phase
Figure 1.1. Schematic representation of plasma reactions.
9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
reactive species may combine during discharge to form a polymer in the gas phase
(plasma polymerization). They may also attack the surface species on the substrate
and create active entities, resulting in surface modification of the substrate (plasma
modification). Plasma etching is the third type of reaction occurring during a
gas-discharge process, which can be used to remove impurities from the surface or
modify other surface properties.
Although numerous studies have utilized plasma reactions to modify polymer
surfaces, the issue of their characterization appears to be a continuous challenge. As
a matter of fact, complexity o f plasma reaction mechanisms appears to be not fully
understandable. 41 Multimolecular or multiatomic reactions and extremely fast
reaction rates make predictions and control of chemical processes that occur in the
plasma state difficult. 42
In view of these considerations, it is useful to utilize surface-sensitive
characterization methods that are selective to the structural changes resulting from
the plasma surface reactions. While X-ray photoelectron spectroscopy (XPS) , 43
secondary ion mass spectrometry (SIMS) , 44 and ion scattering spectroscopy (ISS) 45
are surface techniques yielding molecular level information at an Angstrom level,
they require the use of high vacuum conditions, and their depth of penetration is
limited to the first few monolayers. Other methods, such as scanning electron
microscopy, 46 contact angle measurements, 18 atomic force microscopy, 47 and
adhesion measurements, 20 may provide important information about material
properties that are altered by plasma modifications, but they do not provide
molecular level information about the species formed on or below the surface.
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In contrast, spectroscopic technique that allows non-destructive analysis of
molecular structures resulting from surface reactions is attenuated total reflectance
(ATR) 48"4 9 Fourier transform infrared (FT-IR) spectroscopy. This method does not
require high vacuum conditions, which can often disrupt weakly bonded surface
species, and requires minimal sample preparation, thus being highly attractive in the
studies of polymer surfaces when molecular level information from 0-2.5 pm into
the surface is sought. Although one of the disadvantages of ATR FT-IR
measurements is a wavenumber dependence of the depth of penetration, 50 a recently
developed algorithm minimizes this relationship and allows quantitative analysis of
surfaces. 51,52 Finally, the use of polarized IR light in an ATR experiment makes
possible determination of the orientation o f the surface species. 49
1.2 Non-film Forming Plasma Reactions of Siloxane Elastomers
Due to inertness and other desirable properties, crosslinked siloxane elastomers are
widely used in biomedical applications. As a result, they have been subject of
numerous studies. 53*55 Although siloxane-based elastomers exhibit good mechanical,
thermal, and chemical properties, surface modifications are often necessary to obtain
desirable biocompatibility. Plasma can be used for this purpose; I5,44’56*71 and in this
context, chemically non-reactive and reactive plasmons, depending upon the mode
of consumption of the gases, can be used. 72,73 The term reactive is referred to as a
consumption of discharge gases as a result of being incorporated into a solid phase
through the formation of chemical bonds, and non-reactive plasmons are mainly
composed of monoatomic inert gases, capable of ionizing other molecules, but
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
unable to be consumed during plasma reactions.
1.2.1 Non-reactive Gas Plasma Reactions
The use o f inert gases in plasma reactions can substantially affect surface
properties of polymers, leading to surface rearrangements caused by interactions
with highly energetic inert gas particles. 72 In the case of highly inert polymers, such
as polydimethylsiloxane (PDMS), which often contains an amorphous Si02, it is
desirable to alter their surfaces. There are at least two possible reaction sites on the
PDMS surface which highly energetic plasmas are able to attack:61
CH3 CH3
I
I
- Si - O - Si - O I I
ch3 ch3
Plasma • CH2
I
----------- ► - Si - O - Si • - O I I
ch3
ch3
Ar or He
the formation of Si-CH2« radical results from the hydrogen abstraction, and Si« is
generated by the CH3 cleavage. Thus, non-reactive gas plasmas lead to the formation
of Si - H surface species through the Si-CH3 cleavage, followed by a rearrangement
of the Si« radicals.
When one compares Ar microwave plasma reacted PDMS
surfaces with the unreacted PDMS, a new band is detected at 2158 cm'1, which is
attributed to the Si - H stretching modes. Although Si0 2 exists as another source of
the Si-H formation, the Si-H bond intensity increases at higher PDMS to Si0 2 ratio,
and the PDMS methyl groups are the primary source of the Si-H formation. 62 Under
hydrosilation conditions, ATR F T - I R spectra of Ar microwave plasma reacted to
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
PDMS to show a loss o f the Si-H species and an increase o f a broad absorption
centered at 3360 cm'1, resulting from the formation of the O-H stretching modes of
silanol groups. On the other hand, under the same conditions, unreacted PDMS
shows no silanol formation. Therefore, hydrosilation converts the Si-H groups
formed on the PDMS surface to silanol groups, thus creating reactive sites for
additional grafting.
Although Ar was used for the plasma reactions, such non-reactive gas plasma
reactions lead to the formation of the carbonyl groups on the PDMS surface.64
Various possible sources of 0 2 for carbonyl formation were proposed. When air was
introduced by venting the plasma chamber, atmospheric
0 2
may be reacting with
radicals formed on the PDMS surface. 0 2 may come from residual air in the plasma
chamber, or silicone network contains
0 2
in both the polymer backbone and the
silica filler that may be reacting during the plasma reactions.
A comparison of Ar and He plasma conditions imposed on PDMS surfaces
revealed that, while the chemical changes occurring on the surface are similar, Ar
plasma resulted in much more rapid surface modifications than He. Among
biomedical applications, He plasma has been utilized to improve surface properties
of PDMS used for catheters. 56 Unreacted PDMS showed the expected
silicon/oxygen/carbon ratio of approximately 1/1/2.56 On the other hand, when He
plasma is reacted to the PDMS surface, the surface carbon content decreased and
oxygen content increased, which suggested that
0 2
from water impurities replaces
surface carbon atoms, leading to the formation of surface Si02. It is apparent that the
CH3 groups on the PDMS surface were replaced by oxygen, and the backbone of
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
PDMS was cleaved to yield Si0 2 species on the surface. As a result, increased
surface hydrophilicity and lower surface friction are anticipated.
1.2.2 Reactive Gas Plasma Reactions
Reactive gas plasmas are plasmons that participate in the surface reactions. This
class of surface plasma reactions commonly uses such gases as 0 2, C 02, and NH3 . 73
0 2
plasma generates a substantial amount of surface hydroxyl functionalities on the
PDMS surface, 26'44 which is demonstrated by the presence of the 3425 and
3225 cm' 1 bands due to hydrogen bonded O-H groups. The band detected at
3425 cm' 1 is attributed to intramolecular hydrogen bonding, whereas the 3225 cm' 1
band is due to intermolecular hydrogen bonding of the O-H groups. Several
structural possibilities were illustrated for each bonding situation. It was proposed
that the majority of hydroxyl groups were bonded onto the pendant surface carbon
atoms. It was also proposed that a fraction of hydroxyl groups reacted onto the
silicon atoms after pendant methyl groups were cleaved off. The reactions leading to
the formation of O-H groups are depicted in Figure 1.2A . 3
The results o f XPS analysis indicate that 0 2 plasma reaction of PDMS appeared to
replace surface carbon with oxygen. 44 It was suggested that the methyl groups were
being replaced by oxygen. Eventually, the backbone of the polymer was cleaved to
yield Si0 2 molecules on the surface. 0 2 plasma reactions can be utilized to improve
PDMS bondability to other materials.
A tensile adhesion of less than 10 psi was
obtained for unreacted PDMS bonded between A1 disks using an epoxy adhesive.
However, 0 2 plasma reacted PDMS yielded tensile adhesions of 500 psi. It was
14 -
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
S i—CH3
o2" z
I
I
> Si — CH2 — OH +
Plasma
|
z
Si — OH
|
(A)
S i—CH3
co2
Plasma
I
?
f
* 3
Si — CH2 — O— C — C = C H 2
|
z
z
(B)
S i—CH3
nh3
I.
F
S i— c — NH2
I.
+ Si — CH2 — N
(C)
Figure 1.2. Proposed reaction mechanisms of reactive gas plasma on PDMS: A plasma; B - C 0 2 plasma; and C - NH3 plasma.
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
suggested that surface oxidation was the key factor leading to the large increase in
adhesion.
C 0 2 gas plasma reactions lead to the formation of multiple C=0 and C=C groups
on the PDMS surface, 64 as demonstrated by the appearance of the bands at 1725,
1700, and 1675 cm*1 due to C=0 groups and C=C groups at 1596 cm'1. The band
detected at 1700 cm*1 attributed to hydrogen bonded C=0 groups results from the
formation o f the Si-OH functional groups, which was also detected by a broad band
centered at 3400 cm*1. The band detected at 1596 cm*1 was attributed to the
formation of vinyl groups. These studies suggested that C 0 2 plasma reactions lead to
the formation of the -0 -C0 -C(CH3 )=CH2 species attached to the silicone backbone.
This is illustrated in Figure 1.2B.
Formation of reactive N-H species on a polymer surface can be utilized for further
surface grafting. Thus, numerous studies have focused on the NH3 plasma
reactions. 64 The bands detected at 1664 and 1596 cm*1 were of interest because they
result from the primary amide C=0 stretching and amide N-H deformation modes,
respectively. Primary amides are a major component on the PDMS surface when
NH3 plasma is employed. This is schematically depicted in Figure 1.2C. On the
other hand, when chemically non-bonded molecules are present within PDMS
network, they significantly affect NH3 plasma reactions. 65
The presence of chloro-fiinctional molecules, such as an initiator or residual freon
from cleaning processes, leads to the formation of another surface species in the
plasma state. 3 65 NH3 plasma reacted PDMS indicates the formation of amide
functionalities, demonstrated by the presence of two strong bands at 3150 and
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3053 cm'1. However, these bands are not detectable when NH3 plasma reacted
PDMS surfaces are washed with water. It appears that the material removed from
the PDMS surface is identified as N H /C l'. Therefore, the presence of
chloro-functional molecules leads to the adsorption of NHt+C r species, which
inhibit the development of amide groups on PDMS surface.
1.2.3 Solid Phase Plasma Reactions
A majority of the reactants utilized for plasma reactions on polymer surfaces are
gases. While gaseous monomers are relatively easy to apply, the use of solid
monomers has been limited due to apparent difficulties in evaporating solids into a
gas phase. Therefore, the use of solid monomers has been primarily focused on the
post-plasma reactions, such as bulk monomer polymerization, and grafting of plasma
activated surfaces. It was reported that plasma can create active species on the
surface of water-soluble monomers. 74"76 When the monomer is allowed to dissolve
77
in the water, it can initiate post-plasma reactions. Anion radicals generated at the
surface of monomer is the primary active species, which changes to the free radicals
by reactions with hydrogen ions. The speculated process of the post-plasma reaction
was as follows:
Ionization
Radical Formation
Surface Reaction
e" + M
► M•
Ml + H+
► HM*
HM* + P* ------- ► P - MH
Active species formed in the plasma state cannot initiate post-plasma reactions when
the monomer remains in the solid state.
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A new method of reacting solid monomers utilizing microwave plasma reactions
was developed to chemically bond solid monomers directly to PDMS surfaces. 66"70
PDMS can be surface modified using solid monomer microwave plasma reactions of
maleicanhydride and acrylamide.66 Due to the lack of C=C bond in succinic
anhydride and propionamide, these monomers show no reactivity toward PDMS
surfaces. However, maleic anhydride and acrylamide, both containing C=C bonds,
can be reacted on the PDMS surfaces through the C=C opening using microwave
plasma. Therefore, the factor governing the microwave plasma reactions of
maleicanhydride and acrylamide is the presence o f the C=C bond which reacts with
the PDMS surface. This is shown in Figure 1.3. For microwave reaction times of
less than 15 sec, anhydride or amide functionalities of the monomers are maintained.
On the other hand, the microwave reaction times exceeding 15 sec result in a
cleavage of the surface anhydride or amide species and a conversion of the starting
monomer to maleic acid and polyacrylamide, respectively.
When imidazole is reacted to the PDMS surface in the presence of Ar microwave
plasma under closed reactor conditions, the Si-H groups are not detected on the
PDMS surface.68,69 On the other hand, the ->Si-CH2-N< species are formed.
Imidazole radicals resulting from the hydrogen abstraction of the N-H bonds react
with the SiCH2' radicals to form Si-CH2-Imidazole entities. As a result of these
studies, reaction mechanisms illustrated in Figure 1.4A were proposed. 69 Reactivity
of imidazole reacted to the PDMS surface increases at lower discharge pressures,
and its amount increases up to approximately 20 sec. Discharge times exceeding
20 sec decrease the imidazole content reacted to the PDMS surface. This behavior is
18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Major Species
A
Ar Plasma
o
f
tio
o
o
o
(
73 73
A
O
V
'f
Secondary Species
+
o
PDMS
Mb
\
NH2
Ar Plasma
----------------------
*r
b
PDMS
Figure 1.3. Proposed reaction mechanism of maleic anhydride and acryl amide
reacted PDMS in the presence of Ar microwave plasma.
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
H
I
S ,_ch
3
+
H -* Q
H
H
I
s i - C H 2- * Q
Ar
H
H
H
(A)
H
I
^=N
Si — O — N
I
I
)= \
H
H
H
I
Sj ’
C H3
+
>=N
H -n Q
H
T
02
^ r
H
H
I
^=N
~ — CH2 — "N
SSii — O
|'
I
H
H
( B)
/R
N vV ni
T CH2‘ N T
R
\ h 2
H
I
s i _ CH3
♦
)= N
H - N f l
FT
H
R ~^ \
CO2
I
Si
r F
1
H
—
O
r»
R
-
^=m
NJ
/
CH2
R
d —\
R / N ^
N _____
CH2
R
n U_
5
H2
R
R;CH3,C H 2 *
( C)
Figure 1.4. Proposed reaction mechanism of imidazole reacted PDMS in the
presence of microwave plasma: A - Ar microwave plasma; B - O2 microwave
plasma; and C - C0 2 microwave plasma.
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
attributed to a lack o f ability of imidazole to evaporate to a gas phase and
subsequent removal o f already reacted groups. Orientation of imidazole reacted to
the PDMS surface using TE and TM polarized ATR FT-IR spectroscopy reveals that
the imidazole rings are preferentially oriented parallel to PDMS surface.
Analysis of microwave plasma reactions conducted under open flow reactor
conditions reveals different structures formed on PDMS surface. Imidazole reacts to
a PDMS surface by ring opening, resulting in the formation of the G=N surface
groups. The higher energy state of plasma obtained from open flow reactor
conditions are responsible for the ring opening reaction of imidazole.
Reactive gas may be consumed in plasma reactions and incorporated into the solid
phase through the formation of chemical bonds or participating in reaction
mechanisms. Imidazole reactions occur on the PDMS surface through hydrogen
abstraction of the N-H bonds of imidazole when O2 microwave plasma are
employed. 70 As discharge times increased above 20 sec, the formation of the
Si-0-CH2-imidazole species was detected. On the other hand, below 20 sec, the
formation of the Si-O-imidazole species on PDMS was detected. Proposed reaction
mechanisms responsible for various discharge times are shown in Figure 1.4B.
When C 0 2 is utilized in the presence of microwave plasma, these species form Si-Oimidazole linkages. However, protons on the H-C=C-H entities in the presence of
imidazole are substituted with the CH3 groups. These reactions are followed by the
hydrogen abstraction to form Si-0-imidazole-CH2» radicals, followed by reactions
with subsequent imidazole molecules through the formation of the CH2-N linkages.
This sequence of reactions results in the formation of multilayers of imidazole rings
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
on the PDMS surface. A proposed mechanism responsible for these reactions is
shown in Figure 1.4C.
13 Film Forming Plasma Reactions of Siloxane Elastomers
In the non-film forming plasma reactions, polymer surfaces are modified by
plasmons. However, during film forming plasma reactions, a new polymer is
deposited on the original polymer substrate. This process is also known as plasma
polymerization and produces highly crosslinked, pin-hole free, and good adhesive
films to a substrate. Therefore, films of this type are utilized for adhesion promotors
and microelectronics construction, as well as for altering gas permeability and
surface hydrophilicity.
Plasma polymerized fluorocarbon films are highly crosslinked and have a low
friction coefficient, which can be deposited on numerous substrates, including
silicone elastomers. ESCA analysis of fluorocarbon plasma deposited on PDMS
shows peaks at 295, 293, 288.5, and 286 eV, corresponding to CF3, CF2, CF, and C
structures, respectively. 17 A strong fluorine and a weak oxygen peak are also
detected. Perfluoro-l-methyldecalin plasma deposited on PDMS has a surface
friction coefficient of 0.21 to 0.26.
7ft
Although these values are not as low as the
friction coefficient for polytetrafluoroethylene (PTFE), 0.05-0.1, they are sufficient
as dry lubricants for silicon surfaces.
Plasma polymerization can be also used to deposit new surface films containing
functionalities for further reactions. Silicon surfaces are hydroxyl or amine
functionalized using plasma polymerization of aliyl alcohol or allylamine,
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
respectively. 33 To test reactivity of the deposited hydroxyl groups, the allyI alcohol
plasma films were exposed to trifluoroacetic anhydride. ESCA analysis revealed
15% fluorine on the allyl alcohol deposited silicone surfaces. Similarly, allyl amine
deposited silicone surface showed
10
% fluorine on the surface when exposed to
pentafluorobenzaldehyde.
Hydrocarbon gases are also utilized for silicone surface reactions. The
hydrophobic nature o f silicone surfaces can be reduced by plasma polymerization of
hydrocarbons onto these surfaces. 79 PDMS is a favorable material for contact lenses
because it exhibits high oxygen permeability, softness, and durability. However, the
silicone surface is hydrophobic, which is undesirable for contact lenses. 80,81 Methane
plasma polymerization results in the formation of surface methylene radicals, which
forms surface-bound CH2OH groups in the presence o f 0 2. Acetic and formic acid
plasma polymerization also shows a high ratio of hydroxyl functionalities to the total
number of carbon atoms. Therefore, methane, acetic acid, and formic acid plasma
polymerizations on the silicone surface decrease hydrophobicity, as determined by
the water contact angle measurements. They exhibit contact angle of 15-20°, as
opposed to 60° for unreacted PDMS.
1.4 Hydrophobic Recovery of Plasma Reacted Siloxane Elastomers
For many biomedical applications, the use of plasma reacted silicone elastomers is
limited by the fact that polymer surface dynamics cause partial or even complete
disappearance o f the surface modification effects over time. Although reactive gas
plasma reactions of PDMS functionalize the surface, introducing polar groups, aging
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
in the air induces hydrophobic recovery. 82' 86 O2 plasma reactions of PDMS result in
the formation of the Si-OH groups. Therefore, the presence of hydrophobic recovery
suggests silanol condensation leading to the formation of silica-like structure. The
presence of silica-like network develops microcracks leading to the diffusion of
unreacted low molecular weight silicon chains into the silicon elastomer network.
The other reason for hydrophobic recovery is reorientation of the hydrophilic groups
resulting from O2 plasma reactions. Therefore, hydrophobic recovery contributes to
diffusion of polar groups introduced into the bulk. The diffusion controlled
migration of low molecular weight silicon chains plays an important role in the
hydrophobic recovery process when compared with reorientation of the newly
formed polar groups from the surface toward the bulk. 82"84
Hydrophobic recovery of O2 plasma reacted PDMS was completely suppressed
during storage in liquid nitrogen or water, and only a minor increase of
10
° in water
contact angle was observed. XPS analysis showed that Ar, CO2 , and NH3 plasma
reacted PDMS increases carbon contents at the expense of oxygen and silicone after
storage in water or in liquid nitrogen. On the other hand, the carbon content of 0 2
plasma reacted PDMS decreases during storage in water or in liquid nitrogen, while
its oxygen and silicone contents increase. 85' 86
In summary, siloxane-based elastomers exhibit good mechanical, thermal, and
chemical properties, but surface modifications are often necessary to obtain
desirable biocompatibility and other useful properties. Plasma reactions are
attractive for this purpose, because they alter polymer surfaces without minimal
changes of polymer bulk properties. However, one of the drawbacks of plasma
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
reactions is the complexity of reaction mechanisms. In this dissertation, particular
focus is given to understanding molecular level processes resulting from plasma
reactions on siloxane-based elastomer surfaces when different reaction conditions
are employed.
1.5 References
1. H. M. Mott-Smith and I. Langmuir, Phys. Rev., 1926,28, 727.
2. W. R. Gombotz and H. S. Hoffinan, Crit. Rev. Biocompatibility, 1987, 4(1), 136.
3. B. Hollahan, Techniques and Applications o f Plasma Chemistry, John Wiley &
Sons, New York, 1974.
4. M. M. Shakin, Reaction Under Plasma Conditions, John Wiley & Sons, New
York, 1971.
5. C. H. Bamford and J. C. Ward, Nature, 1960, 186, 712.
6
. P. Hartek, Z. Phys. Chem., 1929, 54, 881.
7. P. Hartek and U. Kopsch, Z. Phys. Chem., 1931, B 12, 327.
8
. C. Copeland, Phys. Rev., 1930, 36, 1221.
9. C. Copeland, J. Am. Chem. Soc., 1930, 52,2580.
10. F. K. McTarggart, Plasma Chemistry in Electrical Discharge,Elsevier,
New
York, 1967.
11. H. B. Boenig, Plasma Science and Technology, Cornell University Press, New
York, 1982.
12. J. P. Wightman and N. J. Johnston, Adv. Chem. Ser., 1969, 80,2317.
13. D. D. Neisevender, Adv. Chem. Ser., 1969, 80, 338.
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14. F. J. Vastola and J. P. Wightman, J. Appl. Chem., 1964, 14, 69.
15. M. Millard, J. J. Windle, and A. E. Pavlath, J. Appl. Polym. Sci., 1973, 17,
2501.
16. N. MorosofF, B. Crist, T. Hsu, and H. Yasuda, J. Macromol. Sci., Chem. Ed.,
1976, A10(3), 451.
17. H. Yasuda, Plasma Polymerization, Academic Press, New York, 1985.
18. N. Inagaki, S. Tasaka, J. Ohkubo, and H. Kawai, J. Appl. Polym. Sci., Appl.
Polym. Symp., 1990, 46, 399.
19. S. L. Kaplan and P. W. Rose, Plastics Eng., 1988,44, 77.
20. N. J. DeLollis and O. Montoya, J. Adhes., 1970, 3, 57.
21. E. M. Liston, J. Adhes., 1989, 30, 199.
22. K. Shah and C. L. Beatty, Org. Coat. Appl. Polm. Sci. Proc., 1982, 47, 232.
23. H. P. Schreiber, M. R. Wertheimer, and M. Lambla, J. Appl. Polym. Sci.,
1982, 27, 2269.
24. H. P. Schreiber and F. St. Germain, J. Adhes. Sci. Technol., 1990,4, 319.
25. P. M. Triolo and J. D. Andrade, J. Biomed. Mater. Res., 1983, 17, 149.
26. J. R. Hollahan, B. B. Stafford, and R. D. Falb, J. Appl. Polym. Sci., 1969, 13,
807.
27. A. T. Bell, T. Wydeven, and C. C. Johnson, J. Appl. Polym. Sci., 1975, 19,
1911.
28. D. Peric, A. T. Bell, and M. Shen, J. Appl. Polym. Sci., 1977, 21, 2661.
29. K. Doblhoffer and I. Eiselt, Thin Solid Films, 1977, 118, 181.
30. A. S. Hoffman, J. Appl. Polym. Sci., Appl. Polym. Symp., 1988, 42, 251.
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31. Y. Ikada, Adv. Polym.. Sci., 1984, 57, 104.
32. D. Kiaei, A. S. Hoffinan, B. D. Ratner, and T. A. Horbett, J. Appl. Polym.Sci„
Appl. Polym. Symp., 1988,42, 269.
33. W. R. Gombotz and A. S. Hoffinan, J. Appl. Polym. Sci., Appl. Polym.Symp.,
1988,42,285.
34. J. Lub, E. Bruninx, and A. Benninghoven, Polymer, 1989, 30,40.
35. A. Toth, I. Bertoli, and M. Blazso, J. Appl. Polym. Sci., 1994, 52, 1293.
36. M. J. Danilich and K. K. Marchant, J. Biomat. Sci., Polym. Ed., 1992, 3, 195.
37. V. N. Pallassana, J. Biomat. Sci., Polym. Ed., 1994, 6 (2), 181.
38. M. J. Danilich, D. Gervasio, and R. E. Marchant, Biomed. Eng., 1993, 21,
655.
39. S. Yuan, G. Szakalas, and K. K. Merchant, J. Appl. Biomat., 6, 1996, 259.
4 0 .1. Kiu Kang, O. Hyeong Kwon, and Y. Moo Lee, Biomaterials, 1996, 17,
841.
41. H. Yasuda and C. E. Lamaze, J. Appl. Polym. Sci., 1973, 17, 1533.
42. A. R. Westwood, Eur. Polym. J., 1971,363.
43. N. H. Turner, Anal. Chem., 1988, 60, 377.
44. M. Morra, E. Occhiello, R. Marola, F. Garbassi, P. Humphrey, and, P. Johnson,
J. Colloid Interface Sci., 1990, 11, 137.
45. T. G. Vargo, S. A. Gardella, and L. Salvati, J. Appl. Polym. Sci., Polym.Chem.
Ed., 1989, 27, 1267.
46. T. E. Nowlin and D. F. Smith, J. Appl. Polym. Sci., 1980, 25, 1619.
47. S. N. Magonov, J. Appl. Polym. Sci., Appl. Polym. Symp., 1992, 51, 3.
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48. M. W. Urban, Vibrational Spectroscopy o f Molecules and Macromolecules on
Surfaces, Wiley - Interscience Publication, New York, 1993.
49. M. W. Urban, Attenuated Total Reflection Spectroscopy o f Polymers-Theory and
Practice, American Chemical Society, Washington, DC, 1996.
50. N. J. Harrick, Internal Reflection Spectroscopy, Interscience Publishers, New
York, 1967.
51. J. B. Huang and M. W. Urban, Appl. Spectrosc., 1992, 46 (11), 1666.
52. J. B. Huang and M. W. Urban, Appl. Spectrosc., 1993,47 (7), 973.
53. D. G. LeGrand, J. Polym. Sci., 1969, B7, 579.
54. K. Ziegel and F. Enrich, J. Polym. Sci., A 2 ,1970, 8, 2015.
55. C. Yang and G. E. Wnek, Polym. Mater. Sci. Eng., 1990, 62, 601.
56. P. M. Triolo and J. D. Andrade, J. Biomed. Mater. Res., 1983, 17, 129.
57. M. J. Owen and P. J. Smith, J. Adhesion Sci. Technol., 1994, 8, 1063.
58. J. L. Fritz and M. J. Owen, J. Adhesion, 1995, 54, 33.
59. U. W. Gedde, A. Hellebuych, and M. Hedenqvist, Polym. Eng. Sci., 1996, 36,
2077.
60. M. J. Owen and J. L. Stasser, Polym. Eng. Sci., 1997, 38, 1087.
61. H. Kim and M. W. Urban, Langmuir, 1996, 12, 1047.
62. S. R. Gaboury and M. W. Urban, Polymer, 1991, 32 (13), 390.
63. S. R. Gaboury and M. W. Urban, Polymer, 1992,33 (23), 5085.
64. M. W. Urban and M. T. Stewart, J. Appl. Polym. Sci., 1990, 39, 265.
65. S. R. Gaboury and M. W. Urban, J. Appl. Polym. Sci., 1992, 44, 401.
66. S. R. Gaboury and M. W. Urban, Langmuir, 1993, 9, 3225.
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
67. S. R. Gaboury and M. W. Urban, Langmuir, 1994, 10, 2289.
68. H. Kim and M. W. Urban, Langmuir, 1995, 11,2017.
69. H. Kim and M. W. Urban, Langmuir, 1996, 12, 1047.
70. H. Kim and M. W. Urban, Langmuir, 1996, 12, 1051.
71. H. Kim, M. W. Urban, F. Lin, and D. J. Meier, 1996, 12,3282.
72. L. H. Coopes and K. J. Gifkins, J. Macromol. Sci., Chem. Ed., 1982, 17, 217.
73. N. Morosoff, Plasma Deposition, Treatment, and Etching o f Polymers,
Academic Press, San Diego, 1990.
74. A. Odajima, Y. Nakase, and Y. Osada, Am. Chem. Soc. Symp., 1979, 108,263.
75. M. Suzuki, A. Kishida, and H. Iwata, Macromolecules, 1986, 19, 1804.
76. B. C. Simionescu, M. Leanca, and C. I. Simionescu, Polym. Bull., 1980, 3,437.
77. Y. Osada and A. Mizumoto, Macromolecules, 1985, 18, 304.
78. D. L. Cho and H. Yasuda, J. Appl. Polym. Sci., Appl. Polym. Symp., 1984, 38,
65.
79. C. P. Ho and H. Yasuda, Polym. Mater. Sci. Eng., 1987, 56, 705.
80. H. D. Gesser and R. E. Warriner, US Patent 3,925,178.
81. F. J. Kai, US Patent 5,580,606.
82. M. Morra, E. Occhiello, R. Marola, F. Garbassi, P. Humphery, and D. Johnson,
J. Colloid Interface Sci. Technol., 1990, 137, 11.
83. M.J. Owen and P.J. Smith, J. Adhesion Sci. Technol., 1994, 8, 1063.
84. A. Toth, I. Bertoli, M. Blazso, G. Banghegyi, A. Bognar, and P. Szaplonczay, J.
Appl. Polym. Sci., 1994, 52, 1293.
85. J. L. Fritz and M. J. Owen, J. Adhesion, 1995, 54, 33.
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
86. P. E. Emmanuel and J. B. Henk, J. Adhesion Sci. Technol., 1996, 10, 351.
30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 2
MICROWAVE PLASMA REACTIONS OF IMIDAZOLE
ON POLYDIMETHYLSILOXANE ELASTOMER SURFACES:
A SPECTROSCOPIC STUDY
31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.1 Introduction
Surface reactions utilizing plasmas are an attractive means for conducting
chemical reactions, in particular, when localized and short pulses of energy are
required. For that reason, utilization o f microwave plasma leading to modifications
of polymer surfaces and interfaces is quite appealing.1 In this context, one of the
merits of plasma reactions is the ability of plasmas to alter surfaces without altering
bulk polymer properties.2 This approach opens numerous opportunities for
modifying surface properties and creating reactive sites for further surface and
interfacial reactions.3 In contrast to traditional chemical reactions, one of the
drawbacks of the reactions conducted in plasma gas phase is the complexity of
reaction mechanisms.4*5 Multimolecular or multiatomic, often simultaneous and
spontaneous collisions, and extremely fast reaction rates6 make predictions and
control of the chemical processes occurring in plasma environments challenging.
Therefore, considerable difficulties may be encountered when detailed analysis of
the surface species created by plasma reactions is attempted.7
8-11
We
utilized microwave energy to generate plasmas which allowed us to create
new functional groups on crosslinked polydimethylsiloxane (PDMS) elastomers.
Analysis of the surface functional groups on PDMS and their quantitative analysis
resulting from microwave plasma reactions of solid monomers was conducted using
attenuated total reflectance (ATR) FT-IR spectroscopy. In these studies, we
established that the monomers containing C=C bonds, such as acrylamide and
maleic anhydride, react with PDMS surface through the C=C double bond opening
and maintain their original structure when reaction times do not exceed 10 sec.
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
However, extended reaction times lead to limited monomer supply to the reaction
sites due to conversion o f monomers to polyacrylamide or maleic acid. Therefore, a
cleavage of the previously reacted surface amide or anhydride species occurs. Using
ATR FT-IR spectroscopy, quantitative analysis of newly created surface species was
performed. 11
In this study, our efforts will concentrate on surface reactions of imidazole on
PDMS elastomer surfaces using microwave plasma energy. Like previous studies,
formation of surface reacted imidazole molecules on PDMS surface, effects of
microwave plasma parameters on surface reactivity, orientation of the surface
species, and how closed or open flow reactors may affect reaction mechanisms will
be investigated.
2.2 Experimental
2.2.1 Substrate Preparation
Polydimethylsiloxane (PDMS) was prepared from a linear, vinyl terminated
dimethylvinylmethylsiloxane polymer (Mn = 28,000; Huls American Inc.).
Reactions among vinyl groups forming crosslinked PDMS network were initiated by
adding 0.5% (w/w) t-butyl perbenzoate (Aldrich Chemical) to PDMS. PDMS
oligomer and initiator were first premixed for 24 hrs to ensure complete dissolution
of initiator in PDMS. Films of crosslinked PDMS were prepared by pressure
molding the oligomer-initiator solution for 15 min at 149°C and post-crosslinking
for an additional 4 hrs at 210°C. Crosslinked PDMS films containing SiC> 2 were
prepared by adding 5% (w/w) of Aerosil 200 (Degussa Corp.) SiC>2 . An
33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
oligomer-initiator solution prepared in such a way was combined with silica and
mixed in a rolling ball mill for an additional 24 hrs. Crosslinking was accomplished
by pressure molding a specimen under approximately 330 psi for 15 min at 149°C
using Carver Lab. Press, Model C, and post-crosslinking for an additional 4 hrs at
210°C. Potential surface contaminants and residual low molecular weight species
were removed by stirring PDMS films in methylene chloride for 5 hrs. Residual
methylene chloride was removed from the PDMS network by vacuum desiccating
each specimen for 24 hrs at room temperature.
2.2.2 Surface Reactions
Plasma reactions were conducted using closed and open flow reactors which are
schematically depicted in Figure 2.1. In the open flow reactor, reactions were
conducted in a continuous flow of gas under a specific pressure. Crosslinked PDMS
substrate, with approximate dimensions of 50x25x2 mm, and approximately 50 mg
of solid imidazole were placed into a reactor. The reactor was evacuated to 1.3 Pa,
followed by purging it with Ar gas to the desired pressures, until a steady state
pressure of Ar gas flow was reached. At this point, microwave radiation of
approximately 600 W of power with output frequency of 2.45 GHz using microwave
source, KMC Model KMO-24G, to induce plasma reactions was turned on. For
experiments conducted in a closed reactor, after imidazole and PDMS were placed
into the reactor, the chamber was evacuated to approximately 1.3 Pa and returned to
atmospheric pressure by introducing Ar gas. The reactor was evacuated again to the
desired pressures, followed by microwave exposure to induce plasmas. In both
34
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Closed Plasma Reactor
Ar
Open Flow Plasma Reactor
Ar
Figure 2.1. A schematic diagram illustrating closed (A) and open flow (B)
microwave plasma reactors: V - vacuum pump; G - pressure gauge; C - cold trap; S substrate; M - monomer; Ar - argon gas; and N - needle valve.
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
cases, gas plasma reactions on PDMS surface were carried out using imidazole
(Aldrich Chemical) which at 1.3 Pa exhibits partial vapor pressure of 2.6xl0 '6 Pa. 12
Under these conditions, the pressure in the reaction chamber increases continuously
during microwave plasma discharge. However, under the same pressure conditions,
the pressure in the reaction chamber remains constant, and no sorption of imidazole
into the PDMS network was detected without microwave plasma discharge.
2.2.3 Spectroscopic Measurements
ATR FT-IR spectra were collected on a Digilab FTS-14B equipped with a liquid
nitrogen cooled MCT detector. A resolution of 4 cm" I and a mirror speed of
0.3 cm s"l were used. The ATR cell was aligned at a 45° angle of incidence using a
45° angle parallelogram
KRS-5 crystal. To determine orientation of the surface
species, 90° (TE) and 0° (TM) polarized infrared light was used. TE is a transverse
vector of the incidence light polarized at 90° with respect to sample surface, whereas
TM is a transverse magnetic vector polarized at 0° with respect to sample surface.
Other experimental details concerning the setup were published elsewhere. 8
1013
Each spectrum represents 300 coaded scans ratioed against a reference spectrum
obtained from 300 coaded scans of an empty ATR cell. Ail ATR spectra were
corrected for spectral distortions using Q-ATR software.
14
36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.3 Results and Discussions
2.3.1 Analysis of Imidazole Plama Reacted PDMS
Figure 2.2 illustrates ATR FT-IR spectra in the C=N and C=C stretching regions
for imidazole reacted to PDMS surface under various initial discharge pressures. In
this case, experiments were conducted in a closed reactor. For reference purposes,
trace A of Figure 2.2 illustrates the spectrum of unreacted PDMS surface. The
spectra of imidazole reacted to PDMS surface exhibit appearance of the bands at
1603 cnr* and 1559 cm 'l which are attributed to the C=C and C=N stretching
modes of imidazole. It appears that when the reaction is conducted at 106.7 Pa
(trace B), the C=C stretching band at 1603 cm"l and the C=N stretching band at
1559 cm- * are detected. The bands become stronger when initial discharge pressures
are dropped to 53.3 Pa (trace C) and 26.6 Pa (trace D). In addition, when pressures
below 106.7 Pa are used, the C-H deformation region of imidazole reacted to PDMS
surface exhibits a new band at 1393 cm~l attributed to the C-H deformation modes
of the -CH=CH- groups.
Although these observations suggest that imidazole reacts with PDMS surface, the
N-H stretching bands should allow us to confirm these findings if this is indeed the
case. Figure 2.3 illustrates ATR FT-IR spectra in the N-H and C-H stretching
regions of the same specimens and exhibit appearance of two new bands at
3161 cm- 1 and 3121 cm~l attributed to antisymmetric and symmetric stretching
modes of the -CH=CH- entities on the surface reacted imidazole. Again, when initial
discharge pressures are decreased from 106.7 Pa (trace B) to 26.6 Pa (trace D), the
intensity of the C-H stretching bands of the - CH=CH- groups also increases. These
37
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CO
o
in
2200
2000
1800
1600
1400
Wavenumbers (cm"I)
Figure 2.2. ATR FT-IR spectra in the 2300-1300 cm 'l region o f imidazole reacted
to PDMS surface under various pressure conditions using a closed reactor: A unreacted PDMS; B - imidazole reacted to PDMS at 106.7 Pa/10 sec; C - imidazole
reacted to PDMS at 53.3 Pa/10 sec; and D - imidazole reacted to PDMS at
26.6Pa/10 sec.
38
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
co
eo
CM
eo
3300
3200
3100
3000
2900
2800
Wavenumbers (cm* 1)
Figure 2.3. ATR FT-ER spectra in the 3400-2800 cm"* region of imidazole reacted
to PDMS surface under various pressure conditions using a closed reactor: A unreacted PDMS; B - imidazole reacted to PDMS at 106.7 Pa/10 sec; C - imidazole
reacted to PDMS at 53.3 Pa/10 sec; and D - imidazole reacted to PDMS at
26.6 Pa/10 sec.
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
observations indicate that imidazole indeed reacts to the PDMS surface, most likely
by the hydrogen abstraction of the N-H bonds, but without the C=C cleavage and
subsequent ring opening. A possible mechanism of imidazole reaction to the PDMS
surface is shown in Figure 2.4, path A. In this figure, “box” represents PDMS
substrate. Although at this point we do not know the origin of the reaction sites on
PDMS available for surface reactions with imidazole, this issue is under
investigation. 15
It should be also realized that similar spectroscopic observations could be made if
imidazole molecules were physically deposited on the surface of PDMS elastomer.
To resolve this concern, similar to the previous studies, 10 all microwave plasma
reacted PDMS samples were boiled in water for 20 min, followed by removing and
drying specimens under vacuum for 24 hrs. After such treatments, which are
considered to be sufficient to remove all physisorbed molecules, ATR FT-IR spectra
were recorded again. In all cases presented in this study, the spectra before and after
surface reactions were identical, indicating that imidazole is chemically attached and
not physisorbed at PDMS surface.
2.3.2 Discharge Time Effect
Figure 2.5 illustrates ATR FT-IR spectra in the C=C and C=N stretching regions
for imidazole reacted to PDMS surface under closed reactor conditions with a
change of discharge times. Again, for reference purposes, trace A illustrates
unreacted PDMS spectrum. As the discharge times increase from 5 sec (trace B) to
10 sec (trace C), intensities of the C=C and C=N stretching bands at 1603 cm 'l and
40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Path A
»»
v
Closed chamber
\
_
ii
N
I
|b=c
ri'
^
N
ii
/Y \i
PDMS
H
Imidazole
O pen ch am be r
ON
0® N
Path B
i w
H
H
Figure 2.4. Schematic representation of imidazole reacted to PDMS surface in
microwave plasma reactions: Path A - imidazole reacted under closed reactor
conditions; Path B - imidazole reacted under open reactor conditions.
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2200
2000
1800
1600
1400
Wavenumbers (cm‘ 1)
Figure 2.5. ATR FT-IR spectra in the 2300-1300 cm 'l region o f imidazole reacted
to PDMS surface under various discharge time conditions using a closed reactor: A unreacted PDMS; B - imidazole reacted to PDMS at 26.6 Pa/5 sec; C - imidazole
reacted to PDMS at 26.6 Pa/10 sec; D - imidazole reacted to PDMS at 26.6 Pa/20
sec; and E - imidazole reacted to PDMS at 26.6 Pa/30 sec.
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1551 cm 'l increase. However, extended reaction times over 20 sec result in a
decrease of the band intensities. This is illustrated in traces C and D, which
represent ATR FT-IR spectra for the specimens with 20 and 30 sec discharge time.
These relatively long discharge times do not allow residual monomer to evaporate,
most likely due to an increase of the inner pressure at longer discharge times in a
closed reactor. Therefore, the monomer cannot be further supplied to a gas phase,
thus preventing a continuation o f surface reactions.
2.3.3 Silica Inhibition Effect
Silica in a form o f powder particles is commonly used as a PDMS/elastomer
reinforcing agent. Its presence, however, may cause a significant effect on surface
reactions. Our earlier studies
8-9
indicated that the presence of silica may affect
surface reactions. For example, PDMS without SiC>2 exposed to microwave plasmas
in the presence of Ar atmosphere leads to the formation of Si-H surface
functionalities. However, the same reactions conducted on a PDMS containing SiC>2
inhibits the formation of Si-H functionalities. These findings stimulated us to pursue
similar experiments when imidazole is microwave plasma reacted to PDMS.
Figure 2.6 illustrates ATR FT-LR spectra of imidazole reacted to PDMS containing
5% (w/w) of silica at various initial discharge pressures. As the initial discharge
pressures are diminished from 106.7 Pa (trace B) to 26.6 Pa (trace D), intensities of
the C=C and C=N stretching bands at 1601 cm"l and 1551 c n rl increase. However,
the intensity changes are not as pronounced as compared to imidazole reacted to the
PDMS surface without silica. A comparison of the results for initial discharge
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
o m
co in
m
2200
2000
1800
1600
1400
Wavenumbers (cm- 1)
Figure 2.6. ATR FT-IR spectra in the 2300-1300 cm"* region of imidazole reacted
to PDMS surface containing 5% (w/w) silica under various pressure conditions
using a closed reactor: A - unreacted PDMS; B - imidazole reacted to PDMS at
106.7 Pa/10 sec; C - imidazole reacted to PDMS at 53.3 Pa/10 sec; and D imidazole reacted to PDMS at 26.6 Pa/10 sec.
44
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
pressures of 26.6 Pa is shown in Figure 2.7 and indicates that no change of
imidazole reactions for the discharge times ranging from 5 to 30 sec are detected.
This is shown in Figure 2.7, traces B and C, respectively. For comparison purposes,
trace D of Figure 2.7 illustrates a spectrum o f the PDMS specimen without silica
exposed to 26.6 Pa for 30 sec discharge time. Intensities of the C=C and C=N
stretching bands at 1598 cm"l and 1551 cm" I of imidazole reacted to PDMS with
silica are diminished.
To verify that the presence of silica is detrimental to the formation of PDMSimidazole linkages, polycrystalline silica films were microwave plasma treated.
Trace A of Figure 2.8 illustrates ATR FT-IR spectrum of unreacted silica film.
Trace B illustrates the spectrum of the same silica film, but exposed to 26.6 Pa for
10 sec discharge time with imidazole present in the reaction chamber. For
comparison purposes, trace C illustrates ATR FT-IR spectrum of imidazole reacted
to PDMS containing 5% (w/w) of silica exposed to 26.6 Pa for 10 sec discharge
time. Analysis of the spectra shown in Figure 2.8 indicates that no C=C and C=N
stretching modes due to imidazole are detected on silica film. Although it is beyond
the scope of these studies to determine a mechanism of silica inhibition, based on
these findings, it is believed that lower surface reactivity of imidazole on PDMS
containing silica is attributed to the formation of microdomains containing silica on
the PDMS surface, which inhibits surface reactions.
45
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IO
2200
2000
1800
1600
1400
Wavenumbers (cm* 1)
Figure 2.7. ATR FT-ER spectra in the 2300-1300 cm" I region of imidazole reacted
to PDMS surface containing 5% (w/w) silica under various discharge time
conditions using a closed reactor: A - Unreacted PDMS; B - imidazole reacted to
PDMS at 26.6 Pa/5 sec; C - imidazole reacted to PDMS at 26.6 Pa/30 sec; and D imidazole reacted to PDMS without silica at 26.6 Pa/30 sec.
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
o
in
Cl
1800
1600
1400
CM
1200
1000
Wavenumbers (cm^)
Figure 2.8. ATR FT-IR spectra in the 1900-900 cm' 1 region of imidazole reacted to
silica film surface using a closed reactor: A - unreacted silica film; B - imidazole
reacted to silica film at 26.6 Pa/10 sec; and C - imidazole reacted to PDMS
containing 5% (w/w) silica at 26.6 Pa/10 sec.
47
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.3.4 Polarization Effect
One of the advantages of using polarized ATR FT-IR spectroscopy is the ability to
determine orientation of the surface species. 13 If one considers symmetry and
reactivity of 5- and 6 -member rings, these entities have a tendency to take a certain
surface orientation. This stimulated us to further investigate orientation of imidazole
reacted to PDMS surface. Figure 2.9 illustrates ATR FT-IR spectra in the C=C and
C=N stretching regions for imidazole reacted to PDMS surface at 106.7, 53.3, and
26.6 Pa in a closed reactor using transverse magnetic (TM) (traces A, B, and C) and
transverse electric (TE) (traces D, E, and F) polarizations. Definitions of TM and TE
16
polarizations along with the experimental setup are published elsewhere. As shown
in Figure 2.9, the intensities o f the C=C and C=N stretching bands at 1606 and
1599 cm- * are significantly lower for the TM polarization (traces A, B, and C). This
observation indicates that the newly formed species are preferentially parallel to the
surface. This assessment is supported by the results shown in Figure 2.10 for
antisymmetric and symmetric C-H stretching bands at 3160 and 3116 cm“l recorded
in the TE polarization, which are virtually absent when the spectra are recorded
using TM polarization. Again, this observation supports the fact that imidazole rings
are preferentially oriented parallel to the PDMS surface.
One of the issues that is rarely addressed when gas plasma reactions are conducted
is the effect of initial discharge pressures in the plasma reactor and their changes
during plasma reactions. This issue is important because chemical structures
produced by microwave plasma reactions may vary due to inner pressure changes in
48
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CO
o
—• ®
m
2
0*3
OB
2200
2000
1800
1600
1400
Wavenumbers (cm 'l)
Figure 2.9. Polarized ATR FT-IR spectra in the 2300-1300 cm 'l region of imidazole
reacted to PDMS surface under various pressure conditions using a closed reactor: A
- TM polarization o f imidazole reacted to PDMS at 106.7 Pa/10 sec; B - TM
polarization of imidazole reacted to PDMS at 53.3 Pa/10 sec; C - TM polarization of
imidazole reacted to PDMS at 26.6 Pa/10 sec; D - TE polarization of imidazole
reacted to PDMS at 106.7 Pa/10 sec; E - TE polarization of imidazole reacted to
PDMS at 53.3 Pa/10 sec; and F - TE polarization of imidazole reacted to PDMS at
26.6 Pa/10 sec.
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
OS
OS
CM
o
CO
CO
3300
3200
3100
3000
2900
2800
Wavenumbers (cm'*)
Figure 2.10. Polarized ATR FT-IR spectra in the 3400-2800 cm'* region of
imidazole reacted to PDMS surface under various pressure conditions using a closed
reactor: A - TM polarization of imidazole reacted to PDMS at 106.7 Pa/10 sec; B TM polarization of imidazole reacted to PDMS at 53.3 Pa/10 sec; C - TM
polarization of imidazole reacted to PDMS at 26.6 Pa/10 sec; D - TE polarization of
imidazole reacted to PDMS at 106.7 Pa/10 sec; E - TE polarization of imidazole
reacted to PDMS at 53.3 Pa/10 sec; and F - TE polarization of imidazole reacted to
PDMS at 26.6 Pa/10 sec.
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the plasma reactor. While velocity of a gas molecule and ionization of gases in
plasma states are inversely proportional to the discharge pressures during plasma
reactions, chemical structures formed on the surface will vary with the inner
pressure changes. In addition, plasma state is controlled by a so-called energy factor,
which is the amount of microwave energy divided by velocity of a gas molecule.
Therefore, increase of the discharge pressures during plasma reactions results in a
significant decrease of velocity of gas phase molecules, thus lowering their kinetic
energy and consequently making a lower energy state.
2.3.5 Closed and Open Flow Reactors
So far, we discussed microwave plasma reactions conducted under closed reactor
conditions. Before we focus on the open microwave plasma reactions, it should be
realized that the primary difference between the closed and open flow reactors is the
change o f discharge pressures during plasma reactions. Under closed reactor
conditions, discharge pressures increase rapidly with the increase of discharge times
during plasma reactions, resulting in a lower energy state. In contrast, when open
flow reactor condition is used, discharge pressures maintain steady state initial
pressures during plasma reactions, which results in a higher energy state. This higher
energy plasma state under open flow reactor conditions is due to lower steady state
pressures during microwave plasma reactions and results in the ring opening of
imidazole molecules shown in Figure 2.4, path B.
Let us now focus on the surface reactions conducted under open flow
reactorconditions. Figure 2.11 illustrates ATR FT-IR spectra in the C=C and C=N
51
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CO
CO
CO
CM
tO
CO
2200
2000
1800
1600
1400
Wavenumbers (cm'l)
Figure 2.11. ATR FT-IR spectra in the 2300-1300 cnr 1 region of imidazole reacted
to PDMS surface under various pressure conditions using an open flow reactor: A unreacted PDMS; B - imidazole reacted to PDMS at 106.7 Pa/10 sec; C - imidazole
reacted to PDMS at 53.3 Pa/10 sec; and D - imidazole reacted to PDMS at
26.6 Pa/10 sec.
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
stretching regions for imidazole reacted to PDMS surface with a change of the initial
discharge pressures in the open flow reactor. The spectrum of imidazole reacted to
PDMS surface at 106.7 Pa (trace B) exhibits a new band at 1658 cm 'l, which is
attributed to the CH2 =CH- stretching modes resulting from the ring opening of
imidazole molecules. Compared to the spectra obtained on the specimens reacted in
a closed reactor (Figures 2.2 and 2.3), the C=N stretching bands at 1559 cm"l are
not detected. However, the spectrum of imidazole reacted to PDMS surface at
53.3 Pa (trace C) exhibits a new band at 2183 cm" 1, which is attributed to the C=N
groups. These observations indicate that the presence of these species results from
the ring opening reaction of the imidazole ring, which appears to occur only when
open flow reactor conditions are employed.
Higher energy states o f plasma in the open flow reactor are most likely responsible
for the ring opening reaction, which is caused by significantly lower steady state
pressures during the plasma reactions. A mechanism responsible for the C=N
formation is proposed in Figure 4, path B. As the initial discharge pressures decrease
from 53.3 Pa to 26.6 Pa, the intensity of the C=N stretching band at 2183 cm 'l
increases. Figure 2.12 illustrates ATR FT-IR spectra in the C=C and C=N stretching
regions for imidazole reacted to PDMS surface with a change of the discharge times
in the open flow reactor. As discharge times increase, intensities of the CH2 =CHand C=C stretching bands at 1655 cm"l and 1596 cm“l, respectively, increase when
the plasma reactions do not exceed 20 sec. However, extended discharge times over
20
sec result in lower intensities, indicating again that the species that were created
during microwave plasma exposure are being cleaved from the surface.
53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CO
CD
ID
00
CO
E
D
C
B
A
2200
2000
1800
1600
1400
Wavenumbers (cnr h
Figure 2.12. ATR FT-IR spectra in the 2300-1300 cm- * region of imidazole reacted
to PDMS surface under various discharge time conditions using an open flow
reactor: A - unreacted PDMS; B - imidazole reacted to PDMS at 26.6 Pa/5 sec;
C - imidazole reacted to PDMS at 26.6 Pa/10 sec; and D - imidazole reacted to
PDMS at 26.6 Pa/20 sec.; E - imidazole reacted to PDMS at 26.6 Pa/30 sec.
54
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
To summarize the results of surface reactions conducted in the open flow
condition, it appears that the G=N stretching band at 2180 cm"l is detected when the
discharge times are between 10 (trace C) and 20 sec (trace D). Above 20 sec
discharge time, the 2180 cm 'l band is not detected (trace E). This behavior results
from a lack of supply o f a monomer and a removal of the imidazole molecules
reacted to PDMS surface by plasma etching.
2.4 Conclusions
Surface analysis of imidazole monomers reacted to PDMS surface using
ATR FT-IR spectroscopy reveals that imidazole molecules are reacted to the PDMS
surface through a hydrogen abstraction of the N-H bonds. This reaction can be
conducted in a closed reactor using microwave plasma environments. Both the
imidazole ring structure and the PDMS crosslinked network are maintained.
Reactivity of imidazole reacted to the PDMS surface increases at lower discharge
pressures, and its amount increases up to approximately 20 sec. Discharge times
exceeding 20 sec decrease the imidazole content reacted to the PDMS surface. This
behavior is attributed to a lack of ability of imidazole to evaporate to a gas phase and
subsequent removal of already reacted groups. For PDMS network containing
5% (w/w) of silica, the presence of silica microdomains results in a small imidazole
content reacted to PDMS surface. This behavior is attributed to lower surface
reactivity of silica microdomains. Finally, orientation of imidazole reacted to the
PDMS surface using TE and TM polarized ATR FT-ER spectroscopy reveals that the
imidazole rings are preferentially oriented parallel to PDMS surface.
55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Analysis of microwave plasma reactions conducted under open flow reactor
conditions reveals different structures formed on PDMS surface. Imidazole reacts to
a PDMS surface by ring opening, resulting in the formation of the C=N surface
groups. Reactivity of imidazole to form C=N surface species increases at lower
discharge pressures, and the amount of the C=N groups reacted to PDMS surface
also increases with discharge times not exceeding 20 sec. Discharge times exceeding
20 sec are destructive to the C=N group stability, most likely due to a localized
thermal energy input from a microwave source.
2.5 References
1. M. Stewart, E. DiDomenico, and M.W. Urban, US Patent 5,364,662.
2. H. Yasuda, Plasma Polymerization, Academic Press, Orlando, FL, 1985.
3. H.Yasuda and A.K. Shamara, J. Polym. Sci., Polym. Phys. Ed., 1981, 19, 1285.
4. H. Kobayashi, M. Shen, and A.T. Bell, J. Macromol. Sci., Chem., 1974, A 8 ,
1354.
5. H. Yasuda and C.E. Lamaze, J. Appl. Polym. Sci., 1973, 17, 1533.
6
. A.R. Westwood, Eur. Polym. J., 1971, 7, 363.
7. Hans J. Griesser and Ronald C. Chatelier, J. Appl. Polym. Sci., Appl. Polym.
Symp., 1990, 46, 361.
8
. S. R. Gaboury and M. W. Urban, Polym. Commun., 1991,32(13), 390.
9. S. R. Gaboury and M. W. Urban, Polymer, 1992, 33(23), 5085.
10. S. R. Gaboury and M. W. Urban, Langmuir, 1993, 9, 3225.
11. S. R. Gaboury and M. W. Urban, Langmuir, 1994, 10, 2289.
56
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12. H. G. De Wit and J.C. Van Miltenburg, J. Chem. Thermodyn., 1983, 15(7), 651.
13. T. A. Thorstenson, L. K. Tebelius, and M. W. Urban, J. Appl. Polym. Sci., 1993,
49, 103.
14. J. B. Huang and M. W. Urban, Appl. Spectrosc., 1992,46(11), 1666.
15. H. Kim and M.W. Urban, Langmuir, 1996, 12, 1047.
16. M.W. Urban, Vibrational Spectroscopy o f Molecules and Macromolecules on
Surfaces, John Wiley & Sons, New York, 1993.
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 3
REACTION SITES ON POLYDIMETHYLSILOXANE ELASTOMER
SURFACES IN MICROWAVE PLASMA REACTIONS WITH GASEOUS
IMIDAZOLE: A SPECTROSCOPIC STUDY
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.1 Introduction
Modification of polymer surfaces using plasma reactions is a continuously
growing area of basic and applied research. Because highly energetic plasmas can
primarily alter surfaces while still maintaining polymer bulk properties,
considerable efforts have been made to analyze surface species created by plasma
reactions. While the majority of studies1"6 concerning plasma reactions focused on
the fragmentation of reacting monomeric molecules resulting from the plasma
reactions and their chemical attachment to the substrate, in essence, there are no data
as to the nature of the substrate sites available for reactions. This issue seems to be
of fairly significant importance because the surface reactive sites determine if
reactions occur and how stable the newly created surfaces are.
We6 utilized microwave energy to generate plasma that allowed us to react
imidazole molecules to crosslinked polydimethylsiloxane (PDMS) surfaces. Based
on the analysis of ATR FT-IR data, the imidazole molecules are chemically bonded
to the PDMS surface through hydrogen abstraction of the N-H bonds of imidazole.
However, one of the deficiencies of these studies was a lack of understanding of the
surface entities on the PDMS surface responsible for imidazole bonding to the
PDMS surface. In this study, our efforts will concentrate on addressing this issue in
a context of the reaction conditions for the imidazole-PDMS system. Specifically,
closed microwave plasma reaction conditions will be utilized. 6 Like previous
studies, ATR FT-IR spectroscopy7"8 will be used to investigate the formation of
reactive sites on PDMS surface and reaction mechanisms o f imidazole molecules in
the presence of Ar plasma.
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.2 Experimental
3.2.1 Substrate Preparation
Polydimethylsiloxane
(PDMS)
films
were
prepared
from
a
linear
dimethylvinylmethylsiloxane copolymer (Mn = 28,000, Huls America, Inc.) The
reaction among vinyl groups forming the crosslinked PDMS network was initiated
by adding 0.5% w/w tert-butyl perbenzoate (Aldrich Chemical) to the linear PDMS.
The linear PDMS resin and initiator were premixed for 24 hrs to ensure complete
dissolution of the initiator. The crosslinking reactions were accomplished by
pressure molding the oligomer mixture for 15 min at 149 °C and postcuring it for 4
hrs at 210 °C. To eliminate surface contaminants and residual low molecular weight
species before plasma reactions, PDMS films were stirred in methylene chloride for
5 hrs. Methylene chloride was removed from the PDMS substrate by vacuum
dessicating the sample for 24 hrs.
3.2.2 Surface Reactions
Plasma reactions were conducted using a closed reactor.6 In the Ar plasma
reactions, a crosslinked PDMS substrate, with approximate dimensions of
50x25x2 mm, was placed into a microwave plasma closed reactor. The reactor was
evacuated to 1.33 Pa, followed by purging it with Ar gas to an atmospheric pressure,
and reevacuated to a desired experimental pressure, typically about 26.6 Pa. At this
point, microwave radiation to induce plasma reactions was turned on. The same
procedures were utilized for reacting the imidazole (Aldrich Chemical Co.)
monomer. In this case, solid monomer and PDMS substrate were placed into a
60
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
before purging and evacuating the reaction chamber.
3.2.3 Spectroscopic Measurements
ATR FT-IR spectra were collected on a Mattson sirius 100 single beam
spectrometer. A resolution of 4 cm' 1 and a mirror speed of 0.3 cm s' 1 were used. The
ATR cell was aligned at a 45° angle of incidence using a 30° angle parallelgram Ge
crystal. Each spectrum represents 200 coaded scans ratioed against a reference
spectrum obtained from 200 coaded scans of an empty ATR cell. All spectra were
corrected for spectral distortions using Q-ATR software.
"t
3 3 Results and Discussions
3.3.1 Reaction Sites on PDMS Surfaces
Before we begin analysis of the spectroscopic data resulting from the microwave
plasma reactions on PDMS surfaces, one important question, that neither we nor any
previous studies have addressed, is the nature and concentration of the surface
reactive sites on PDMS. PDMS is a crosslinked elastomer containing an amorphous
Si0 2 . 5'6 For a typical PDMS network, a 0.03 pm particle size amorphous Si0 2 is
often used. Because highly energetic plasmas are virtually able to break any bonds,
there are at least two possible reaction sites on the PDMS surface:
CH3 CH3
I
I
- Si - O - Si - O I
Ar Plasma
►
’CH2
I
- Si - O - Si* - O -
I
ch3
I
ch3
ch3
I
ch3
61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
in which the Si-CH2* free radical formation results from the hydrogen abstraction
and Si* is generated by the CH3 cleavage. The formation of SiO* free radicals may
result from the Ar microwave plasma reactions of polycrystalline silica.
3.3.2 Formation of Si-CH2 Linkages
To identify the reactive sites responsible for the plasma reactions, several
specimens were prepared under Ar plasma discharge conditions. Figure 3.1
illustrates ATR FT-IR spectra of the PDMS surface. As a reference point, trace A
illustrates the spectrum of unreacted PDMS, with the strongest band at 1412 cm' 1
due to Si-CH3 groups. ATR FT-IR spectrum of the PDMS surface exposed to Ar
microwave plasma radiation for 5 sec discharge, trace B, exhibits the appearance of
the Si-H stretching band at 2157 cm'1. It appears that the intensity of this band
increases, as the discharge time in the Ar plasma increases from 5 (trace B) to 20 sec
(trace D). However, when the discharge times are extended to 30 sec (trace E), the
Si-H stretching band is not detected. These results agree well with our earlier
findings5 and indicate that when the PDMS surface is exposed to Ar microwave
plasma, the Si' radicals are formed by the methyl abstraction o f the Si-CH3 groups.
This process is followed by the reactions of the Si' radicals with hydrogen radicals to
form Si-H groups on the PDMS surface. Although the content of the Si-H groups
formed on the PDMS surface in Ar microwave plasma increases with the discharge
times up to
20
sec, longer discharge times exceeding
20
sec result in a removal of
the Si-H groups. This is demonstrated by a lack of the 2157 cm' 1 band in Figure 3.1,
trace E. Like previous studies, 6 this observation is attributed to a chemical etching in
62
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LC: . .
;|
'
'I
E
I
I
ID
I
!c
-
-
B
A
2200
2000
1800
1600
1400
W av e nu m be rs
Figure 3.1. ATR FT-IR spectra in the 2300-1300 cm'* region of PDMS surface in
Ar microwave plasma under various discharge time conditions using closed reactor:
A - Unreacted PDMS; B - Ar microwave plasma at 26.6 Pa/5 sec; C - Ar microwave
plasma at 26.6 Pa/10 sec; D - Ar microwave plasma at 26.6 Pa/20 sec; and E - Ar
microwave plasma at 26.6 Pa/30 sec.
63
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the presence of extended Ar microwave environments.
Although these observations suggest that, as a result of the Si-H formation, the Si'
radicals are formed on the PDMS surface, occurrence of the hydrogen radicals
should be also examined to confirm these findings. While trace A of Figure 3.2 is
again a reference spectrum o f unreacted PDMS, traces B, C, and D of Figure 3.2
illustrate ATR FT-IR spectra of the PDMS surface exposed to Ar microwave
plasmas. These spectra, illustrating the results for 5 to 20 sec discharge times,
exhibit the appearance of a new band at 1049 cm*1, attributed to the Si-C vibrational
modes resulting from the presence of the Si-CH2 -CH3 groups. 9 Again, discharge
times exceeding 20 sec (trace E) result in the disappearance of the Si-CH2 -CH3
groups on the PDMS surface.
These observations indicate that Si radicals are formed by the methyl group
abstraction of the Si-CH3 groups from the PDMS surface and that the Si-CH2
radicals are formed by hydrogen abstraction of the Si-CH3 groups. While the
hydrogen radicals resulting from the hydrogen abstraction of the Si-CH3 groups
react with the Si radicals to form Si-H groups, the CH3 radicals, resulting from the
methyl abstraction of the Si-CH3 groups, react with the Si-CH2' radicals to form
Si-CH2 CH3 entities. The mechanism for the SiCH2 CH3 and Si-H formation on the
PDMS surface is proposed in Figure 3.3.
3.3.3 Imidazole Reaction Mechanisms on PDMS Surface
So far, we have addressed the issue of the surface reactive sites on PDMS in the
presence of Ar microwave plasma. Keeping these findings in mind and considering
64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
'
__
1200
------------- D
-------- — '
11.j 0
B
1100
1050
1000
950
W a v e n u m b e rs
Figure 3.2. ATR FT-IR spectra in the 1200-950 cm*1 region of PDMS surface in Ar
microwave plasma under various discharge time conditions using closed reactor: A Unreacted PDMS; B - Ar microwave plasma at 26.6 Pa/5 sec; C - Ar microwave
plasma at 26.6 Pa/10 sec; D - Ar microwave plasma at 26.6 Pa/20 sec; and E - Ar
microwave plasma at 26.6 Pa/30 sec.
65
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rearrangement
012013
Si
H
O
S i -----
Figure 3.3. Reaction mechanism of PDMS surface in Ar microwave plasma.
66
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the fact that in previous studies6 we determined that the imidazole reactions with the
PDMS surface occur by the hydrogen abstraction of the N-H bonds of imidazole, let
us focus on the reactive sites available for imidazole. Figure 3.4 illustrates
ATR FT-IR spectra of the PDMS surface after Ar microwave plasma (traces B and
C) and imidazole reactions at discharge times 5 and 10 sec (traces D and E). As
discharge times increase, the C=C and C=N stretching bands are detected at
1599 cm' 1 and 1549 cm'1, and the intensities of both bands increase. On the other
hand, the Si-H stretching band is not detected when imidazole is attempted to react
for 5 (trace D) and 10 sec (trace E) discharge times under Ar plasma conditions.
Figure 3.5 illustrates ATR FT-IR spectra of the specimens described in Figure 3.4,
but in the 1200-950 cm' 1 spectral region. The band due to the Si-CH2 CH3 species
detected at 1047 cm' 1 is detected on the PDMS surface upon Ar microwave plasma
exposure (traces B and C) and also when imidazole is reacted to the PDMS surface
in the presence of Ar plasma (traces D and E). However, when imidazole is reacted
to the PDMS surface, a new band at 1073 cm' 1 is also detected. This is shown in
Figure 3.5, traces D and E. This band is attributed to the Si-C vibrational modes of
the ->Si-CH2-N< groups. 10 Based on this analysis, we are in a position to identify
imidazole and PDMS sites that are responsible for the ->Si-CH2 -N< formation. The
Si-CH2 radicals of the PDMS surface react with the methyl radicals abstracted from
the PDMS surface and imidazole radicals abstracted from imidazole molecules to
form Si-CH2-CH3 and Si-CH2-imidazole species. The possibility of reactions
between Si and imidazole free radicals is excluded because of the crowding effect
between imidazole ring and neighboring substituents o f the PDMS linkage.
67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
«
_
-S
5TA
5O*c
L
_
Of;
uO
r;
E
I
D
B
A
2200
2000
18 0 0
1600
14 00
Wave n u m b e r s
Figure 3.4. ATR FT-IR spectra in the 2300-1300 cm' 1 region of PDMS surface and
imidazole reacted to PDMS surface in Ar microwave plasma under various
discharge time conditions using closed reactor: A - Unreacted PDMS; B - Ar
microwave plasma at 26.6 Pa/5 sec; C - Ar microwave plasma at 26.6 Pa/10 sec;
D - imidazole reacted at 26.6 Pa/20 sec; and E - imidazole reacted at 26.6 Pa/30 sec.
68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
iC
— E
— D
__ C
— B
= A
1200
1 150
1050
1 100
1000
950
ffavenum bers
Figure 3.5. ATR FT-IR spectra in the 1200-950 cm' 1 region of PDMS surface and
imidazole reacted to PDMS surface in Ar microwave plasma under various
discharge time conditions using closed reactor: A - Unreacted PDMS; B - Ar
microwave plasma at 26.6 Pa/5 sec; C - Ar microwave plasma at 26.6 Pa/10 sec;
D - imidazole reacted at 26.6 pa/20 sec; and E - imidazole reacted at 26.6 Pa/30 sec.
69
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 3.6 illustrates possible reaction mechanisms responsible for the ->Si-CH2-N<
formation.
Previous studies6 showed that imidazole rings are preferentially oriented parallel to
the PDMS surface. However, in this case, there is no preferential orientation of the
Si-CH2 linkages and imidazole entities on the PDMS surface. Our earlier studies5
also indicated that the presence of silica in PDMS network is detrimental to the
formation of Si-H species on PDMS surfaces. In Ar microwave plasma reactions of
imidazole to polycrystalline silica films, we detected no C=C and C=N stretching
modes of imidazole species resulting from surface reactions. Therefore, this
possibility of surface reactions between SiO* and imidazole species during
microwave plasma exposure is excluded.
3.4 Conclusions
Using ATR FT-IR analysis, it is possible to determine reactive sites on the PDMS
surface resulting from the Ar microwave plasma reactions of imidazole. Surface
analysis of the microwave plasma reacted PDMS surfaces reveals that Si« radicals
are formed by the methyl group abstraction of the Si-CH3 groups and also that SiCH2 * radicals result from the hydrogen abstraction of the Si-CH3 groups. The Si
radicals react with the hydrogen radicals resulting from the hydrogen abstraction of
the Si-CH3 groups to form Si-H species. These reactions are followed by the
reactions of the Si-CH2' radicals with methyl radicals, resulting from a methyl group
abstraction of the Si-CH3 groups, to form Si-CH2CH3 species. When imidazole is
reacted to the PDMS surface in the presence of Ar microwave plasma, the Si-H
70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
P lasm a
Figure 3.6. Reaction mechanism of imidazole on PDMS surface in Ar microwave
plasma.
71
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
groups are not detected on a PDMS surface. On the other hand, the ->Si-CH2-N<
species are formed. Imidazole radicals resulting from the hydrogen abstraction of the
N-H bonds react with the SiCH2' radicals to form Si-CH2-Imidazole entities.
3.5 References
1. Y. Ishikawa and S. Sasakawa, Radiat. Phys. Chem., 1992, 9(6), 561.
2. Y. Osada, M. Shen, and A. T. Bell, J. Polym. Sci., Polym. Lett. Ed., 1978, 16,
669.
3. A. Odajima, Y. Nakase, Y. Osada, and A. T. Bell, Am. Chem. Soc. Symp., 1979,
108, 263.
4. S. R. Gaboury and M. W. Urban, Polym. Commun., 1991, 32(13), 390.
5. S. R. Gaboury and M. W. Urban, Langmuir, 1993, 9, 3225.
6
. H. Kim and M. W. Urban, Langmuir, 1995, 11, 2071.
7. J. B. Huang and M. W. Urban, Appl. Spectros., 1992, 46(11), 1666.
8
. M. W. Urban, Vibrational Spectroscopy o f Molecules and Macromolecules on
Surfaces, John Wiley & Sons, New York, 1993.
9. G. Socrates, Infrared Characteristic Group Frequencies, John Wiley & Sons,
New York, 1980.
10. L. J. Bellamy, The Infra-red Spectra o f Complex Molecules, John Wiley &
Sons, New York, 1975.
72
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 4
MICROWAVE PLASMA REACTIONS OF IMIDAZOLE
ON POLYDIMETHYLSILOXANE ELASTOMER SURFACES:
ATTENUATED TOTAL REFLECTANCE FT-IR
AND ATOMIC FORCE MICROSCOPIC STUDIES
73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.1 Introduction
Surface reactions o f gaseous monomers on various substrates using microwave
plasma energy are an attractive means to modify surface properties. Their
attractiveness comes from the fact that microwave plasma reactions are fast, clean,
and do not alter bulk properties of a substrate. 1' 3 Although one could argue, and
perhaps rightfully so, that one of the drawbacks of the reactions conducted in the
plasma gas phase is the complexity of reaction mechanisms, the advantages are
overwhelming. For that reason, significant efforts have been made to analyze
reaction mechanisms conducted by microwave energy. We4 utilized ATR FT-IR
spectroscopy to analyze the imidazole reactions on polydimethylsiloxane (PDMS)
surfaces using closed and open flow reactor conditions. Using a closed reactor, we
managed to chemically bond imidazole rings to the PDMS surfaces through
hydrogen abstraction of the N-H bonds. On the other hand, using an open flow
reactor, imidazole ring opening occurred, resulting in the formation of the -C=N
species. In both cases, the presence of silica filler inhibited imidazole reactions on
the PDMS surface.
Although ATR FT-IR spectroscopy has been invaluable tool in detecting newly
formed species resulting from the microwave plasma surface reactions of imidazole
to PDMS surfaces, surface morphology and, therefore, accessibility for further
reactions have not been addressed. For that reason and to leam more about surface
morphology changes resulting from the microwave plasma reactions, we have
combined ATR FT-IR analysis and atomic force microscope (AFM) measurements. 5
Because AFM has the ability to perform imaging of solid surfaces, 6’7 morphological
74
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
changes of the PDMS surfaces resulting from microwave plasma reactions of
imidazole can be detected and analyzed. In this study, our efforts are concentrated
on the inhibition mechanisms of silica in imidazole/PDMS reactions and on the
morphological changes of PDMS surfaces resulting from the imidazole reactions in
both closed and open flow reactor conditions.
4.2 Experimental
4.2.1 Substrate Preparation
Polydimethylsiloxane (PDMS) was prepared from a linear, vinyl terminated
dimethylvinylmethylsiloxane polymer (Mn = 28,000; Huls American Inc.).
Reactions among vinyl groups to form a crosslinked PDMS network were initiated
by adding 0.5% (w/w) t-butyl perbenzoate (Aldrich Chemical) to PDMS. The PDMS
oligomer and initiator were first premixed for 24 hrs to ensure complete dissolution
of initiator in PDMS. Films of crosslinked PDMS were prepared by pressure
molding the oligomer-initiator solution for 15 min at 149°C and post-curing for an
additional 4 hrs at 210°C. Crosslinked PDMS films containing SiC>2 were prepared
by adding 5% (w/w) of Aerosil 200 (Degussa Corp.) SiC>2 . The oligomer-initiator
solution prepared in this way was combined with silica and mixed in a rolling ball
mill for an additional 24 hrs. Crosslinking was accomplished by pressure molding a
specimen under approximately 330 psi for 15 min at 149°C using a Carver Lab.
Press, Model C, and post-crosslinking for an additional 4 hrs at 210°C. Potential
surface contaminants and residual low molecular weight species were removed by
stirring PDMS films in methylene chloride for 5 hrs. The residual methylene
75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
chloride was removed from the PDMS network by vacuum desiccating each
specimen for 24 hrs at room temperature.
4.2.2 Surface Reactions
Plasma reactions were conducted using closed and open flow reactors which have
been schematically depicted.4 In the open flow reactor, reactions were conducted in
a continuous flow o f gas under a specific pressure. The crosslinked PDMS substrate,
with approximate dimensions of 50x25x2 mm, and approximately 50 mg of solid
imidazole were placed into a reactor. The reactor was evacuated to 1.3 Pa, followed
by purging it with Ar gas to the desired pressures, until a steady state pressure of Ar
gas flow was reached. At this point, microwave radiation o f approximately 600 W of
power with an output frequency of 2.45 GHz using a KMC Model KMO-24G
microwave source to induce plasma reactions was turned on. For experiments
conducted in a closed reactor, the same procedure was utilized, except after
imidazole and PDMS were placed into the reactor, the chamber was evacuated to
approximately 1.3 Pa and brought back to atmospheric pressure by introducing Ar
gas. The reactor was evacuated again to the desired pressures, followed by plasma
treatment In both cases, gas plasma reactions on PDMS surface were carried out
using imidazole (Aldrich Chemical) which at 1.3 Pa exhibits partial vapor pressure
of 2.6x1 O'6 Pa. Under these conditions, the pressure in the reaction chamber
increases continuously during microwave plasma discharge. However, without
microwave plasma discharge and under the same pressure conditions, the pressure in
the reaction chamber remains constant, and no sorption of imidazole into
76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the PDMS network was detected.
4.2.3 Spectroscopic Measurements
ATR FT-IR spectra were collected on a Digilab FTS-14B spectrometer equipped
with a liquid nitrogen cooled MCT detector. A resolution of 4 cm 'l and a mirror
speed of 0.3 cm s~l were used. The ATR cell was aligned at a 45° angle o f incidence
using a 45° angle parallelogram KRS-5 crystal. To determine the orientation of the
surface species, 90°(TE) and 0°(TM) polarized infrared light was used. All ATR
spectra were corrected for spectral distortions using Q-ATR software. 8
4.2.4 Atomic Force Microscopy
A TopoMetrix TMX 2000 small stage atomic force microscope (AFM) was used
to image the sample surfaces. All images were obtained on PDMS samples glued to
an AFM sample holder and scanned under ambient conditions. The samples were
imaged in a contact mode, using a force of 10-9 N, except for the soft PDMS
containing no SiC>2 . In this case, a “tapping mode” was utilized.
4.3 Results and Discussions
4.3.1 Microwave Plasma Reactions of Imidazole on PDMS Surface
As our previous studies4 indicated, ATR FT-IR spectra of imidazole reacted onto
PDMS surface exhibit the appearance of the bands at 1598 and 1551 cm'1, which are
attributed to the C=C and C=N stretching modes of imidazole. The bands resulting
from the imidazole reactions to the PDMS surface changed when the initial
77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
discharge pressure and time were altered. Figure 4.1 illustrates ATR FT-IR spectra
in the C=C and C=N stretching regions for imidazole reacted to the PDMS surface
at various discharge times under closed reactor conditions. For reference purposes,
trace A of Figure 4.1 illustrates the spectrum of the unreacted PDMS surface. As the
discharge times increase from 5 sec (trace B) to 10 sec (trace C), intensities of the
C=C and C=N stretching bands at 1603 cm*1 and 1551 cm*1 increase. However,
discharge times exceeding 20 sec (traces D and E) result in a decrease of the band
intensities, which is attributed to an increase of the inner pressure at longer
discharge times in a closed reactor.
Figure 4.2 illustrates ATR FT-IR spectra recorded using transverse magnetic (TM)
and transverse electric (TE) polarization in the C=C and C=N stretching regions for
imidazole molecules reacted to PDMS surface at various discharge pressures in a
closed reactor. Intensities of the C=C and C=N stretching bands at 1606 and
1557 cm*1 in TE polarization increase as the discharge pressures become lower
(traces A, B, and C), and a significant drop of intensities is observed when the
spectra are recorded using polarization (traces D, E, and F). This observation
indicates that newly reacted imidazole rings to the PDMS surface are preferentially
parallel oriented.
Finely divided silica is commonly used as a PDMS/elastomer reinforcing agent.
However, as our earlier studies4,9 indicated, its presence may have a significant
effect on the surface reactions. Figure 4.3 illustrates ATR FT-IR spectra in the C=C
and C=N stretching regions for imidazole reacted on the PDMS surfaces. The
crosslinked PDMS polymer network contains 5% (w/w) of silica with a particle size
78
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
n
2200
2000
1800
1600
*h
1400
Wavenumbers (cm'1)
Figure 4.1. ATR FT-IR spectra in the 2300-1300 cm' 1 region of imidazole reacted to
PDMS surface under various discharge time conditions using a closed reactor: A unreacted PDMS; B - imidazole reacted to PDMS at 26.6 Pa/5 sec; C - imidazole
reacted to PDMS at 26.6 Pa/10 sec; D - imidazole reacted to PDMS at
26.6 Pa/20 sec; and E - imidazole reacted to PDMS at 26.6 Pa/30 sec.
79
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
c
2200 2 100 2000
1900 1800 1700 1600
1500 1400
W avenumbers
Figure 4.2. Polarized ATR FT-IR spectra in the 2300-1300 cm*1 region of imidazole
reacted to PDMS surface under various pressure conditions using a closed reactor: A
- TE polarization o f imidazole reacted to PDMS at 106.7 Pa/10 sec; B - TE
polarization of imidazole reacted to PDMS at 53.3 Pa/10 sec; C - TE polarization of
imidazole reacted to PDMS at 26.6 Pa/10 sec; D - TM polarization of imidazole
reacted to PDMS at 106.7 Pa/10 sec; E - TM polarization o f imidazole reacted to
PDMS at 53.3 Pa/10 sec; and F - TM polarization of imidazole reacted to PDMS at
26.6 Pa/10 sec.
80
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CO
in
2200
2000
1800
1600
1400
Wavenumbers (cm'1)
Figure 4.3. ATR FT-IR spectra in the 2300-1300 cm' 1 region of imidazole reacted to
PDMS surface containing 5% (w/w) silica under various discharge time conditions
using a closed reactor: A - Unreacted PDMS; B - imidazole reacted to PDMS at 26.6
Pa/5 sec; C - imidazole reacted to PDMS at 26.6 Pa/30 sec; and D -imidazole
reacted to PDMS without silica at 26.6 Pa/30 sec.
81
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0.03 jim. However, in the reactions of the PDMS containing silica under 26.6 Pa
pressures, no change of imidazole content reacted on the PDMS surfaces is detected
for the discharge times ranging from 5 (trace B) to 30 sec (trace C). For comparison
purposes, trace D o f Figure 4.3 illustrates a spectrum of the PDMS surface without
silica which was exposed at 26.6 Pa for 30 sec discharge times. A comparison of
these data indicates that the intensities of the C=C and C=N stretching bands at 1598
and 1551 cm*1 of imidazole reacted to the PDMS containing silica are diminished.
Because the presence of silica may significantly alter the extent of microwave
plasma reactions, an effort was made to verify that the presence of silica is
detrimental to the formation of PDMS-imidazoIe reactions by studying imidazole
reactions on polycrystalline silica films. Trace A o f Figure 4.4 illustrates the ATR
FT-ER spectrum o f polycrystalline silica film, and trace B shows the spectrum of the
same silica film, but exposed to a microwave plasma of imidazole under 26.6 Pa for
10 sec discharge time. For comparison purposes, trace C illustrates the ATR FT-IR
spectrum of imidazole reacted on the PDMS containing 5% (w/w) of silica after
exposure to the microwave plasma of imidazole under 26.6 Pa for 10 sec discharge
time. As seen, no C=C and C=N stretching modes due to imidazole reactions are
detected on the polycrystalline silica, thus confirming that Si0 2 does not react with
imidazole.
4.3.2 Silica Effect on Surface Morphology
To establish that imidazole molecules do not react on the PDMS surface in the
presence of silica, let us examine the surface morphological changes on PDMS
82
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
o
o
06
CO
m
CO
CM
CO
in
1800
1600
1400
1200
1000
Wavenumbers (cm*1)
ic
Figure 4.4. ATR FT-IR spectra in the 1900-900 cm"1 region of imidazole reacted to
silica film surface using a closed reactor: A - unreacted silica film; B - imidazole
reacted to silica film at 26.6 Pa/10 sec; and C - imidazole reacted to PDMS
containing 5% (w/w) silica at 26.6 Pa/10 sec.
83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
resulting from the presence of silica by means of atomic force microscope (AFM).
As a first step, let us examine the PDMS surface which does not contain silica.
Figure 4.5 illustrates an AFM image of the PDMS surface and indicates that the only
detected morphological features result from the formation of the film. In this case,
the PDMS surface exhibits a maximum height of 1082 nm. On the other hand, the
AFM image of the silica containing PDMS illustrated in Figure 4.6 exhibits a much
smoother surface, but contains many cracks with a minimum-to-maximum peak
height of about 3500 nm. Since this sample is reinforced with silica, the film rigidity
is increased, and its flow is restricted. Therefore, even though there is a significant
fraction of an elastomeric component, surface cracks form because the film is unable
to relax during its formation.
Another issue is aggregation of the silica. PDMS tends to aggregate and will form
regions that are PDMS-rich and other regions that are silica-rich. When there are
non-homogeneous regions of silica, the more rigid aggregates will not be able to
relax, and cracks will occur as in Figure 4.6. Although at this point, the mechanism
of the crack formation is not quite clear, it is believed that it involves a nonhomogeneous dispersion of the silica particles, thus giving a significant difference
between the highest and lowest surface elevations, i.e., 3546.34 nm. On combination
of this information with the spectroscopic data, the presence of silica islands on
PDMS diminishes the extent of imidazole reactions. In our study, 10 imidazole
radicals resulting from the hydrogen abstraction of the N-H bonds react with the
SiCH2 radicals through hydrogen abstraction of SiCH3 groups on PDMS to form
Si-CH2-Imidazole entities. Silica particles have no reactive sites for reactions with
84
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 4.5. Surface views o f PDMS imaged using Atomic Force Microscope.
85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 4.6. Surface views of PDMS with 5% (w/w) silica imaged using Atomic
Force Microscope.
86
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
imidazole and also inhibit the SiCH2 radical formation on PDMS. Therefore, the
presence of silica particles acts as anti-catalyst for imidazole-PDMS microwave
plasma reaction.
4.3.3 Imidazole Reaction Effect on Surface Morphology
Keeping in mind that the data were obtained on a closed system, let us now focus
on the open flow reactor surface reactions. Figure 4.7 illustrates ATR FT-IR spectra
in the C=C and C=N stretching regions for imidazole reacted on a PDMS surface
containing no silica under 26.5 Pa pressures as a function o f the discharge times in
the open flow reactor. Again, for reference purposes, trace A illustrates the
unreacted PDMS spectrum. The spectrum of imidazole reacted to the PDMS surface
for 5 sec discharge time (trace B) exhibits a new band at 1658 cm- *, which is
attributed to the -CH=CH- stretching modes resulting from the ring opening of
imidazole molecules. Compared to the spectra obtained on specimens reacted in
closed reactor (Figure 4.1), the C=N stretching bands at 1559 cm‘ 1 are not detected.
On the other hand, the spectrum of imidazole reacted to PDMS surface for 10 sec
discharge times (trace C) exhibits a new band at 2183 cm- *, which is attributed to
the C=N groups. These species, however, which result from imidazole ring opening
show no preferential orientation on the PDMS surface, as the TE and TM polarized
spectra exhibit no differences. The presence of these bands indicates that the ring
opening reactions of imidazole ring appear to occur when the open flow reactor
conditions are employed. The higher energy state of the plasma gas in the open flow
reactor is most likely responsible for the ring opening, which is caused by the
87
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
o
CO
E
D
C
B
A
2200
2000
1800
1600
1400
Wavenumbers (cm'1)
Figure 4.7. ATR FT-IR spectra in the 2300-1300 cm' 1 region of imidazole reacted to
PDMS surface under various pressure conditions using an open flow reactor: : A unreacted PDMS; B - imidazole reacted to PDMS at 26.6 Pa/5 sec; C - imidazole
reacted to PDMS at 26.6 Pa/10 sec; D - imidazole reacted to PDMS at
26.6 Pa/20 sec; and E - imidazole reacted to PDMS at 26.6 Pa/30 sec.
88
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
significantly lower steady state pressures during the plasma reactions. 2*4
To identify morphological surface changes resulting from the open flow reactor
conditions on PDMS, AFM images were obtained. Figure 4.8 illustrates AFM
images of imidazole reacted to PDMS surfaces, under 26.6 Pa for 10 sec discharge
time using a closed reactor conditions, while Figure 4.9 illustrates AFM images
obtained from imidazole reacted to PDMS under the same conditions using the open
flow reactor. In both cases, the PDMS elastomer did not contain silica. Based on the
ATR FT-IR data illustrated in Figure 4.1, imidazole rings are chemically attached to
the PDMS surface through hydrogen abstraction of the N-H bonds. The difference
between the highest and lowest surface elevations detected from the AFM images of
imidazole reacted to PDMS surface in closed reactor conditions is 54.79 nm from
the z-axis scale. Again, no significant morphological changes are detected when
compared with the AFM images of unreacted PDMS (Figure 4.5).
In the plasma reactions of gaseous imidazole, there are various possibilities which
may result in monolayer and/or multilayer formation. Based on ATR FT-IR spectral
data (Figure 4.1) and the AFM images for the imidazole reacted to PDMS under
closed reactor conditions (Figure 4.8), it seems that imidazole rings are uniformly
formed on the PDMS surface through hydrogen abstraction of the N-H groups on
imidazole. This was demonstrated by the presence of the 1603 and 1551 cm' 1 bands
due to imidazole (Figure 4.1) and no significant morphological deviations compared
to unreacted PDMS surfaces (Figure 4.5). As the discharge pressure decreases, band
intensities of the C=C and C=N stretching bands at 1606 and 1599 cm' 1 in
Figure 4.2 increase faster, compared to the intensities of the C-H deformation band
89
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 4.8. AFM image of the surface o f the imidazole reacted on PDMS under
26.6 Pa for 10 sec discharge time in closed reactor conditions.
90
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 4.9. AFM image of the surface of the imidazole reacted on PDMS under
26.6 Pa for 10 sec discharge time in open flow reactor conditions.
91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of -C=C-H. Based on these results, in a closed reactor, imidazole rings are formed
on PDMS surface through hydrogen abstraction of N-H bonds, followed by
hydrogen abstraction o f the -C=C-H groups on imidazole rings, to form multilayer
of imidazole rings on PDMS. On the other hand, the AFM images of the imidazole
reacted to the PDMS surface under open flow reactor conditions shown in
Figure 4.9 exhibit elevations reaching 3766.78 nm, and the presence of irregular
blocks is detected. Following the results of ATR FT-IR spectra illustrated in Figure
. , it appears that when the open flow reactor conditions are employed, two types
4 7
of reactive groups are produced: «C=N, and CH2 =CH». They result from the
imidazole ring opening reactions in the presence o f Ar microwave plasma. The
CH2 =CH« species are grafted to the PDMS surface by hydrogen abstraction, to form
the -(CH=CH)n- linkages on the surface. On the other hand, the «C=N species act as
a terminal group in the CH2 =CH' grafting process. This is verified by the presence of
the C=C stretching modes at 1601 cm' 1 (Figure 4.7). Therefore, morphological
changes of imidazole reacted to PDMS under an open flow reactor conditions result
from the grafting of imidazole entities reacting on the PDMS surface through the
CH2 =CH- entities. A schematic representation of the PDMS surface structures
generated using closed and open flow reactor conditions is shown in Figure 4.10.
4.4 Conclusions
Analysis of ATR FT-IR data of imidazole reacted to PDMS surfaces indicates that
hydrogen abstraction from the N-H groups of imidazole occurs under closed reactor
conditions. On the other hand, imidazole rings open to form the CH2 =CH* and
92
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A. Closed reactor
VII
H
Plasma
-N
II
.c
J
H
H
PDMS
Imidazole
B. Open flow reactor
H.
C--------N
II
II
h'
Y
Plasma
------------
— C =N
- (C H = C H )i r H
—( CH= CH>^-C=N
' h
H
PDMS
Imidazole
Figure 4.10. A schematic representation of PDMS surface structure resulting from
imidazole reaction in the presence of Ar microwave plasma: A - closed chamber;
and B - open flow chamber.
93
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
•C=N reactive entities on the PDMS surfaces when open flow reactor conditions are
employed. The presence of silica on the PDMS surfaces inhibits the
imidazole-PDMS surface reactions.
Based on the analysis of the AFM images o f imidazole reacted to PDMS surfaces,
it appears that the silica containing PDMS form PDMS-rich and silica-rich domains.
In the PDMS-rich domains, silica particles are well-dispersed, but the aggregation in
the silica-rich domain results in crack formation. The presence of the silica-rich
regions
inhibits PDMS reactions with imidazole. AFM images of the
imidazole-reacted PDMS surfaces under closed reactor conditions reveal no
significant morphological changes, and the maximum surface roughness is found to
be 54.79 nm. On the other hand, using open flow reactor conditions, the surface
roughness is much larger and results from the occurrence of irregular blocks formed
on the PDMS surfaces. Based on these findings, under closed reactor conditions, a
multilayer of imidazole rings is reacted on the PDMS surfaces through hydrogen
abstraction of the N-H bonds of imidazole, followed by hydrogen abstraction of
-CH=CH- of the imidazole molecules. Under open flow reactor conditions,
imidazole entities of CH2=CH* resulting from imidazole ring opening are grafted to
form -(CH=CH)n- linkages on PDMS surfaces with »C=N species acting as terminal
groups in the grafting process.
4.5 References
1. M. Stewart, E. DiDomenico, M. W. Urban, US Patent 5,364,662.
2. H. Yasuda, Plasma Polymerization, Academic Press, Orlando, FL, 1985.
94
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3. H. Yasuda and A. K. Shamara, J. Polym. Phys. Ed., 1981, 19, 1285.
4. H. Kim and M. W. Urban, Langmuir, 1995, 11,2071.
5. D. Sarid, Scanning Force Microscope, Oxford University Press, New York,
1991.
6
. S. N. Magonov, J. Appl. Polym. Sci., Appl. Polym. Symp., 1992, 51, 3.
7. T. R. Albrecht and S. W. Kuan, J Appl. Phys., 1988, 64(3), 1178.
8
. J. B. Huang and M. W. Urban, Appl. Spectrosc., 1992,46(11), 1666.
9. S. R. Gaboury and M. W. Urban, Polymer, 1992, 33(23), 5085.
10. H. Kim and M. W. Urban, Langmuir, 1996, 12, 1047.
95
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTERS
THE EFFECT OF DISCHARGE GASES ON MICROWAVE PLASMA
REACTIONS OF IMIDAZOLE ON POLYDIMETHYLSILOXANE (PDMS)
SURFACES: QUANTITATIVE ATR FT-IR SPECTROSCOPIC ANALYSIS
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.1 Introduction
There are numerous forms of energy available for conducting surface and
interfacial reactions. 1'4 Among them, microwave energy generated plasmons appear
to be an effective source for reacting monomeric molecules to elastomeric
surfaces.5' 8 In particular, the opportunity of controlling surface reactions by
changing reaction conditions are appealing methodologies.9' 11 In the previous
studies, 12*15 we developed closed and open flow plasma reactors which allowed us to
chemically bond imidazole molecules to crosslinked polydimethylsiloxane (PDMS)
surfaces when Ar microwave plasmons were employed. Using attenuated total
1A17
reflectance Fourier transform infrared (ATR FT-IR) spectroscopy,
quantitative
analysis of imidazole reactions under open and closed reactor conditions was
performed. When the reactions were conducted under the open flow plasma reactor
conditions, imidazole molecules were reacted to the PDMS surface by ring opening
to form CsN surface species. CH2 ==CH» radicals resulting from the imidazole ring
opening are grafted to form the -(CH=CH)n- linkages on the PDMS surface, and the
generated «C=N radicals acted as terminal groups in the grafting process. 15 In
contrast, when closed reactor conditions were utilized, multilayers o f imidazole
rings were reacted to the PDMS surface through hydrogen abstraction of the N-H
entities of imidazole, followed by hydrogen abstraction of H-C=C-H.
Although previous studies12,15 showed that by changing reaction conditions one
can significantly alter surface products, it became apparent that one of the
deficiencies of these studies was a lack of understanding of the effect of discharge
gases on microwave plasma reactions. For example, reactions conducted under Ar,
97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0
2, or C 0 2 microwave plasma are expected to yield significant differences in
surface reactions. For that reason, this study will focus on addressing the issue of
how reaction mechanisms o f imidazole on the PDMS surface will be affected by the
presence of these gases. Like the previous studies, 14 we will attempt to quantify the
surface reactions between imidazole and PDMS at surface depths of 1.3 pm, when
different microwave plasma conditions are utilized.
5.2 Experimental
5.2.1 Substrate Preparation
Polydimethylsiloxane (PDMS) was prepared from a linear, vinyl terminated
dimethylvinylmethylsiloxane polymer (Mn = 28,000; Huls American Inc.).
Reactions among vinyl groups forming crosslinked PDMS networks were initiated
by adding 0.5% w/w of t-butyl perbenzoate (Aldrich Chemical Co.) to PDMS.
PDMS oligomer and the initiator were first premixed for 24 hrs to ensure complete
dissolution of initiator in PDMS. Films of crosslinked PDMS were prepared by
pressure molding the oligomer-initiator mixture for 15 min at 149°C, and post­
crosslinking for an additional 4 hrs at 210°C. Crosslinked PDMS films containing
SiC>2 were prepared in a similar way by adding 5% w/w of Aerosil 200 (Degussa
Corp.) SiC>2 . Before crosslinking, specimens were mixed in a rolling ball mill for
24 hrs, and crosslinking was accomplished by a pressure molding under
approximately 330 psi for 15 min at 149°C using Carver Lab. press, Model C, and
post-crosslinking for an additional 4 hrs at 210°C. Surface contaminants and
residual low molecular weight species were removed by stirring PDMS films in
98
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
methylene chloride for 5 hrs. To remove any residual methylene chloride from
PDMS specimens, each specimen was vacuum desiccated in 1.3 Pa for 24 hrs at
room temperature.
5.2.2 Surface Reactions
Plasma reactions were conducted using closed reactor conditions. 12 Crosslinked
PDMS substrate, with approximate dimensions of 50x25x2 mm, and approximately
50 mg of solid imidazole were placed into the reactor, which was evacuated to
1.3 Pa, and brought back to atmospheric pressure by introducing a desired discharge
gas. The reactor was evacuated again, followed by purging it with the discharge gas
to a specific pressure. For all gases, the pressure was 26.6 Pa. At this point, a
microwave radiation of 600 W of power with an output frequency of 2.45 GHz
(KMC Model KMO-24G) was turned on to induce plasma reactions. In this case,
gas plasma reactions on PDMS surface were carried out using imidazole which, at
1.3 Pa, exhibits partial vapor pressure of 2.6 x 10*6 Pa. 18 Under these conditions, a
reaction chamber pressure increases continuously during microwave plasma
discharge. Under the same pressure conditions, the pressure in the reaction chamber
remains constant, and no sorption of imidazole into the PDMS network is detected
without microwave plasma discharge.
5.2.3 Spectroscopic Measurements
ATR FT-IR spectra were collected on a Mattson Sirius 100 single beam
spectrometer. A resolution of 4 cm' 1 and a mirror speed of 0.3 cm s’ 1 were used. The
99
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ATR cell was aligned at a 60° angle of incidence using a 45° angle parallelogram
KRS-5 crystal. Such an experimental setup allows surface analysis at 1.3 pn from
the surface.9 Each spectrum represents 200 coaded scans ratioed against a reference
spectrum obtained from 200 coaded scans of an empty ATR cell. All spectra were
corrected for spectral distortion using Q-ATR software. 16,17
5.3 Results and Discussion
5.3.1 Ar Microwave Plasma Reactions
Figures 5.1a-d illustrate a series of ATR FT-IR spectra for imidazole reacted to the
PDMS surface in the presence of Ar microwave plasma. While each figure
represents a spectrum of the same specimen in a different spectral region, each trace
is the spectrum of the PDMS surface conducted using various microwave plasma
conditions. While trace A in Figures 5.1a-d illustrates the spectrum of unreacted
PDMS in different spectral regions, trace B is the spectrum of imidazole reacted to
PDMS at 10 sec discharge time. Appearance of a new band at 3149 cm"l is detected
in trace B of Figure 5.1a, which is attributed to the C-H stretching modes of the
H-C=C-H entities of imidazole. When discharge times increase from 10 sec
(trace B) to 30 sec (trace D), the band intensities due to the H-C=C-H groups
increase. However, no N-H stretching bands due to imidazole molecules are
detected. These observations indicate that imidazole reacts to the PDMS surface,
most likely by hydrogen abstraction of the N-H bonds. Trace B of Figure 5.1b
illustrates the same ATR FT-IR spectrum as in Figure 5.1a, that is, the spectrum of
imidazole reacted to the PDMS surface at 10 sec discharge times, but in the
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2965
3149
D
C
B
3200
3100
3000
2900
Wavenumbers (cm 1)
2800
A
2700
Figure 5.1a. ATR FT-IR in the 3300-2700 cm' 1 region for imidazole-PDMS in the
presence of Ar microwave plasma under various discharge times using closed
reactor conditions: A
- unreacted PDMS; B
- imidazole-PDMS under
26.6 Pa/10 sec; C - imidazole-PDMS under 26.6 Pa/20 sec; and D imidazole-PDMS under 26.6 Pa/30 sec.
101
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13pl
1413
1663
1448,
D
C
B
A
1800
1700
1600
1500
1400
Wavenurabers (a n ')
1300
Figure 5.1b. ATR FT-IR spectra in the 1800-1300 cm' 1 region for imidazole-PDMS
in the presence of Ar microwave plasma under various discharge times using closed
reactor conditions: A - unreacted PDMS; B - imidazole-PDMS under
26.6 Pa/10 sec; C - imidazole-PDMS under 26.6 Pa/20 sec; and D imidazole-PDMS under 26.6 Pa/30 sec.
102
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1380
1415
1448
D
C
B
A
1460
1440
1420
1400
1380
Wavenumbers (cm1)
Figure 5.1c. ATR FT-IR spectra in the 1470-1360 cm' 1 region for imidazole-PDMS
in the presence of Ar microwave plasma under various discharge times using closed
reactor conditions: A
- unreacted PDMS; B - imidazole-PDMS
under
26.6 Pa/10 sec; C - imidazole-PDMS under 26.6 Pa/20 sec; and D imidazole-PDMS under 26.6 Pa/30 sec.
103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1016
D
C
B
A
1200
1150
1100
1050
1000
950
W&venuntes (cm1)
Figure 5.Id. ATR FT-IR spectra in the 1200-950 cm' 1 for imidazole-PDMS in the
presence of Ar microwave plasma under various discharge times using closed
reactor conditions: A
- unreacted PDMS; B - imidazole-PDMS
under
26.6 Pa/10 sec; C - imidazole-PDMS under 26.6 Pa/20 sec; and D imidazole-PDMS under 26.6 Pa/30 sec.
104
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1800-1300 cm' 1 region. It appears that the appearance of new bands at 1603 and
1556 cm' 1 is detected, which are attributed to the C=C and C=N stretching modes of
imidazole, and their intensities become stronger when discharge times increase from
10 sec (trace B) to 30 sec (trace D). Analysis of the spectral regions illustrated in
Figure 5.1c for the same reactions shows that a new band at 1393 cm~l attributed to
the C-H deformation modes of the H-C=C-H entities in imidazole is detected. These
results indicate that imidazole reacts to the PDMS surface through the hydrogen
abstraction of the N-H bonds, but without the C=C cleavage and subsequent ring
opening.
To further substantiate these conclusions, the spectral region between 1200 and
950 cm*1 was analyzed. This is illustrated in Figure 5.Id, which exhibits the
appearance of two new bands at 1069 and 1042 cm' 1 resulting from imidazole
reactions. The new band detected at 1042 cm' 1 comes from the formation of the
Si-CH2 CH3 species on the PDMS surface. 13 On the other hand, the band at 1073 cm'
1
is attributed to the Si-C vibrational modes of the ->Si-CH2 -N< groups. 13 Based on
this analysis, we are in a position to identify imidazole and PDMS sites that are
responsible for the ->Si-CH2-N< formation. The Si-CH2» free radicals formed on
the PDMS surface react with the methyl radicals abstracted from PDMS, followed
by abstraction o f imidazole radicals from imidazole to form Si-CH2-CH3 and
Si-CH2-imidazole linkages. Figure 5.2 illustrates proposed reaction mechanisms
responsible for the formation of these species on the PDMS surface.
While one of the objectives of many surface studies is determination of a chemical
makeup of a surface, another objective is to be able to quantify newly formed
105
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reactive Species in Ar
^
I
Si— CH3
Microwave Plasma
Ar
I
► Si— CH2 * + Si<
I
plasma
1
P n- h
N=/
plasma
N=^
I
+
*H
H
Rearrangement
I
Ar
I
| i —CH2# + CH3 * pbqTW*‘ | i —CH2 CH3
I
V H
at
I
%
Figure 5.2. Proposed reaction mechanisms of imidazole reactions on PDMS in the
presence of Ar microwave plasma.
106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
species. For this reason, we performed quantitative analysis of imidazole reactions
on the PDMS surface at different discharge time conditions. The imidazole I band of
the ring stretching vibration at 1663 cm*1 will be used to measure the extent of the
surface reactions. Since quantitative infrared analysis requires a calibration curve,
Figure 5.3 illustrates absorbance of the imidazole I band plotted as a function of
known imidazole concentrations. The spectra that allow us to determine this
calibration curve were recorded using a transmission mode of detection. Based on
this plot, the extinction coefficient of the imidazole I band was determined to be
67.80 1/mole-cm.
The next step requires the use of the Beer-Lambert law, but to accomplish
quantitative surface analysis, it is necessary to correct all ATR spectra for optical
effects. 16 For that reason, we employed a well-established algorithm 16 which allows
an absorption index spectrum to be refined by an iterative process that minimizes the
difference between the true and calculated reflectivity resulting from optical effects.
At the same time, the Kramers-Kronig relation between absorption (k) and refractive
(n) indices is maintained. 16,17 This iterative process is used in conjunction with the
numerical double Kramers-Kronig transformation (KKT) method to obtain ATR
spectra suitable for quantitative analysis. Based on the KKT corrected spectra, linear
absorbtivity is obtained to allow calculations of surface concentrations from the
Beer-Lambert law (P = e c; where
8
is the extinction coefficient, c is the surface
concentration, and P is the linear absorptivity). Further details involved in
applications and the use of this algorithm which accounts for distortions of strong
and weak bands as well as quantitative ATR measurements are documented in the
107
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0.70
Extinction coeff. “ 67.8080
0.56
uu
aes
■VO.
O
XJ
<
0.42
CA
0.28
0.14
0.00
0.00
0.16
0.32
0.48
0.64
0.80
(E-2)
Concentration (mole-cm/1)
Figure 5.3. Band intensity changes of the imidazole ring stretching band at
1663 cm' 1 plotted as a function of concentration.
108
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
literature. 16,17
Using these approaches, surface concentrations of imidazole reacted to the PDMS
surface under Ar microwave plasma conditions are plotted in Figure 5.4. As the
discharge times increase from 5 sec to 20 sec, the surface concentration of imidazole
increases from 0.23xl0'7 to 4.64xl0'7 mole/1. On the other hand, it decreases from
4.64xl0‘7 to 4.32xl0'7 mole/1, when discharge times increase from 20 sec to 30 sec.
Based on these data, it is apparent that the discharge times increase, exceeding 20
sec results in diminished imidazole surface concentrations due to an increase of the
inner pressure in a microwave reactor under closed reactor conditions. As a result,
solid monomer imidazole cannot be supplied to a gas phase, thus not allowing a
continuation of the surface reactions. 8,12
5.3.2 O 2 Microwave Plasma Reactions
Figures 5.5a-d illustrate a series of ATR FT-IR spectra for imidazole reacted to the
PDMS
surface in the presence of 0 2 microwave plasma. Again, for reference
purposes, traces A in Figures 5.5a-d illustrate the spectrum of unreacted PDMS
surface recorded in different regions. The bands detected at 2963 and 2905 cm"1 of
trace A in Figure 5.5a are assigned to antisymmetric and symmetric C-H stretching
modes, respectively, of the Si-CH3 groups on PDMS. 19 Trace B in Figure 5.5a is the
spectrum of imidazole reacted to PDMS at 10 sec discharge times showing different
spectral regions. In contrast to the Ar microwave plasma experiments, no N-H and
C-H stretching bands due to imidazole molecules are detected. When discharge
times increase up to 20 sec (trace C), the appearance of new bands at 2925 and
109
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
.00
4.80
_u
o
3.60
G
6
s
oo
2.40
1.20
0.00
0
7
14
21
28
35
Discharge time (sec.)
Figure 5.4. Surface concentration changes of imidazole reacted to the PDMS surface
in the presence of Ar microwave plasma plotted as a function o f discharge times.
110
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2963
2924
1905
2856
3091
3200
3100
3000
2900
2800
2700
Wavenumbers (cm'1)
(a)
Figure 5.5a. ATR FT-IR spectra in the 3300-2700 cm' 1 region for imidazole-PDMS
in the presence of O2 microwave plasma under various discharge times using closed
reactor conditions: A - Unreacted PDMS; B - imidazole-PDMS
under
26.6 Pa/10 sec; C - imidazole-PDMS under 26.6 Pa/20 sec; and D imidazole-PDMS under 26.6 Pa/30 sec.
Ill
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16Q3
1661
1558
1384
1412
1452
D
C
B
A
1700
1600
1500
1400
1300
Wavenumbers (cm'1)
(b)
Figure 5.5b. ATR FT-IR spectra in the 1800-1300 cm*1 region for imidazole-PDMS
in the presence of O2 microwave plasma under various discharge times using closed
reactor conditions:
A - Unreacted PDMS; B - imidazole-PDMS under
26.6 Pa/10 sec; C - imidazole -PDMS under 26.6 Pa/20 sec; and D imidazole-PDMS under 26.6 Pa/30 sec.
112
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
i
1460
i---------------------1-------------------- 1--------------------- 1------
1440
1420
1400
W rvoinfcers (a ti1)
A
1380
(c)
Figure 5.5c. ATR FT-ER spectra in the 1470-1360 cm' 1 region for imidazole-PDMS
in the presence of O2 microwave plasma under various discharge times using closed
reactor conditions: A
- Unreacted PDMS; B - imidazole-PDMS under
26.6 Pa/10 sec; C - imidazole-PDMS under 26.6 Pa/20 sec; and D imidazole-PDMS under 26.6 Pa/30 sec.
113
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1012
1260
1350
1087
1250
1150
1050
950
W&vaurbers (cm 1)
(d>
Figure 5.5d. ATR FT-IR spectra in the 1200-950 cm' 1 region for imidazole-PDMS
in the presence of O2 microwave plasma under various discharge times using closed
reactor conditions: A - Unreacted PDMS; B - imidazole-PDMS under
26.6 Pa/10 sec; C - imidazole-PDMS under 26.6 Pa/20 sec; and D imidazole-PDMS under 26.6 Pa/30 sec.
114
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2854 cm' 1 is detected, and their intensities increase for 30 sec discharge times (trace
D). These bands are attributed to the C-H stretching region and result from the
formation of newly formed CH2 linkages on the PDMS surface.
A spectral region from 1800-1300 cm' 1 of the same reactions is shown in
Figure 5.5b. As seen,
appearance of new bands at 1603 cm' 1 and 1558 cm' 1
(trace B) is detected, which are attributed to the C=C and C=N stretching modes of
imidazole molecules. In addition, imidazole I band due to imidazole ring stretching
vibrations is detected at 1661 cm'1, but no N-H deformation bands due to imidazole
molecules are present. Based on these data, it appears that imidazole molecules react
with the PDMS surface through hydrogen abstraction of the N-H bonds of imidazole
and the formation of the CH2 linkages, but without the C=C cleavage and imidazole
ring opening. These observations suggest that when 0 2 microwave plasma
conditions are employed, formation of the CH2 linkages on PDMS should occur. If
this is the case, analysis of the C-H deformation region should allow us to confirm
these findings. Figure 5.5c, trace A, which represents the spectral region from
1500-1350 cm' 1 for the same reactions, exhibits the band at 1415 cm' 1 in unreacted
PDMS. This band is attributed to the C-H deformation modes of the Si-CHj groups
in PDMS.
Another band of interest is the band at 1448 cm'1, which is attributed to the C-H
deformation modes of the Si-CH2 groups on PDMS, resulting from the crosslinking
reactions of the vinyl groups in PDMS. For the imidazole reactions conducted on
PDMS at 10 sec discharge times, the spectrum shown in trace B indicates the
presence of a new band at 1382 cm*1. This band, which increases at extended
115
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
reaction times, is attributed to the C-H deformation modes of the H-C=C-H entities
in imidazole. When discharge times increase up to 20 sec (trace C), another band at
1457 cm' 1 is detected, and its intensity increases with extended discharge times up to
30 sec. This is illustrated in Figure 5.5c, trace D. Since this band is attributed to the
formation of the Si-0 -CH2 species, it is reasonable to expect that these linkages
result from the 0 2 microwave plasma reactions of imidazole on the PDMS surface.
This conclusion is supported by the fact that, at the same time, the intensity of the
C-H deformation modes detected at 1415 cm' 1 due to Si-CH3 groups on the PDMS
surface decreases with the increased discharge times. Based on these findings, for
discharge times below 20 sec, imidazole molecules react with the PDMS surface
through methyl group abstraction of the Si-CH3 species to form Si-O bonds. On the
other hand, above 20 sec discharge times, formation of the Si-0-CH2 linkages is
detected, and their concentrations increase with extended discharge times.
If this is the case, let us analyze the Si-O stretching region of the imidazole reacted
to the PDMS surface. This is presented in Figure 5.5d. Trace A, which is an ATR
FT-IR spectrum of unreacted PDMS surface, illustrates the band at 1012 cm' 1
assigned to the asymmetric stretching modes of the Si-0 species on PDMS. As
discharge times increase from 10 sec (trace B) to 30 sec (trace D), the band intensity
o f the Si-0 stretching modes at 1012 cm' 1 also increases, indicating that the increase
of the Si-O stretching band above a 20 sec discharge time results from the formation
of the Si-0-CH 2 entities. On the other hand, below 20 sec, increase of the Si-O
stretching band intensities can be attributed to the formation of the Si-O-imidazoIe
species on PDMS.
116
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Using the calibration curve shown in Figure 5.3, imidazole concentration can be
determined. Concentration of imidazole molecules reacted to the PDMS surface in
the presence of O2 microwave plasma conditions is plotted in Figure 5.6. As the
discharge times increase from 5 sec to 10 sec, surface concentration of imidazole
molecules increases from 0.58xl0*7 to 2.27xl0'7 mole/I, followed by its further
increase from 2.27xl0‘7 to 9.73xl0 ' 7 mole/1 for discharge times from 20 sec to
30 sec. However, when the discharge times vary from 10 to 20 sec, the imidazole
concentration remains unchanged.
Based on these data, it appears that when discharge times do not exceed 10 sec,
microwave plasma reactions result in the formation of the Si-O-imidazole species on
the PDMS surface. On the other hand, up to 20 sec, the CH3* and Si-O* radicals are
rearranged to form Si-0-CH 3 linkages. Therefore, no increase of the imidazole
concentration is detected. Above 20 sec discharge times, hydrogen abstraction of the
newly formed Si-0-CH 3 species occurs to form the Si-0-CH2* radicals, which are
capable of further reactious with imidazole to form Si-0-CH2-imidazole entities on
the PDMS surface. Proposed reaction mechanisms responsible for various discharge
times are depicted in Figure 5.7.
5.3.3 C 02 Microwave Plasma Reactions
Figures 5.8a-d illustrate a series of ATR FT-IR spectra for imidazole reacted to
the PDMS surface in the presence of C 0 2 microwave plasma. For reference
purposes, traces A and B illustrate the spectra of imidazole reacted to the PDMS
surfaces under 30 sec discharge times when Ar and 0 2 microwave plasma are
117
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
c*iw
X
•s
6
(j
a
o
w
«
,<a
bS
GO
_L
14
21
28
35
Figure 5.6. Surface concentration changes of imidazole reacted to the PDMS surface
in the presence of
0 2
microwave plasma plotted as a function of discharge time.
118
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I
Si— CH3
I
Reactive Species in 0 2
o2
I
, z »
plasma
Si — O • + *CH3
|
Microwave Plasma
P n- h -22—
P n«
plasma
N=^
• H
H
Discharge tim e
Reaction mechanism
0sec.
I
S i— O *
+
H
^=N
* N j
H
---------►
H
I
>=N
S i - O - N l
H
H
H
10 sec.
I
I
S i — O * + *CH3
---------►
S i — O — CH3
20 sec.
I
S i — O — CH3
—H •
1
S i — O — CH2 #
1
---------► S i - 0 - C H 2 - N
— ►
V
30 sec.
1
S i — O — CH2 *
%j
H
H
Figure 5.7. Proposed reaction mechanisms of imidazole on the PDMS surface in the
presence of O2 microwave plasma under various discharge time conditions.
119
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2965
3152
3200
3027
2905
2856
3100 3000
2900
Wavenumbers (cm"1)
2800
2700
Figure 5.8a. ATR FT-IR spectra in the 3300-2700 cm ' 1 region for imidazole-PDMS
in the presence of C 0 2 microwave plasma under various discharge times using
closed reactor conditions: A - imidazole - PDMS under Ar microwave plasma; B imidazole-PDMS under 0 2 microwave plasma; C - imidazole-PDMS under
26.6 Pa/10 sec; D - imidazole-PDMS under 26.6 Pa/20 sec; and E imidazole-PDMS under 26.6 Pa/30 sec.
120
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1409
1661
1600
1553
1800
1700
1600
1500
1400
Wavenumbers (cm-1)
1300
Figure 5.8b. ATR FT-IR spectra in the 1800-1300 cm' 1 region for imidazole-PDMS
in the presence of CO2 microwave plasma under various discharge times using
closed reactor conditions: A - imidazole-PDMS under Ar microwave plasma; B imidazole-PDMS under O2 microwave plasma; C - imidazole-PDMS under
26.6 Pa/10 sec; D - imidazole-PDMS under 26.6 Pa/20 sec; and E imidazole- PDMS under 26.6 Pa/30 sec.
121
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1415
1401
1457
1446
1381
E
D
C
B
A
1460
1440
1420
1400
1380
Wavenumbers (cm-1)
1360
Figure 5.8c. ATR FT-IR spectra in the 1450-1360 cm' 1 region for imidazole-PDMS
in the presence of CO2 microwave plasma under various discharge times using
closed reactor conditions: A - imidazole-PDMS under Ar microwave plasma; B imidazole-PDMS under 0 2 microwave plasma; C - imidazole-PDMS under
26.6 Pa/10 sec; D - imidazole-PDMS under 26.6 Pa/20 sec; and E imidazole-PDMS under 26.6 Pa/30 sec.
122
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1014
109Q
107
E
D
C
B
1200
1150
1100
1050
1000
Wavenumbers (cm-1)
A
950
Figure 5.8d. ATR FT-IR spectra in the 1200-950 cm' 1 region for imidazole-PDMS
in the presence of CO2 microwave plasma under various discharge times using
closed reactor conditions: A - imidazole-PDMS under Ar microwave plasma; B imidazole-PDMS under 0 2 microwave plasma; C - imidazole-PDMS under
26.6 Pa/10 sec; D - imidazole-PDMS under 26.6 Pa/20 sec; and E - imidazolePDMS under 26.6 Pa/30 sec.
123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
employed. In Figure 5.8a, trace A, the band detected at 3152 cm*1 is attributed to the
H-C=C-H stretching modes of the Si-CH2 -imidazole species on the PDMS surface.
The band at 2856 cm*1 shown in trace B is attributed to the H-C=C-H stretching
modes of the Si-0 -CH2 -imidazole entities on the PDMS surface. When CO2 is
employed in the microwave plasma reactions from 10 sec (trace C) to 30 sec
discharge times (trace E), no bands due to the formation o f the H-C=C-H entities of
imidazole are detected. Because no N-H stretching bands of imidazole are detected,
these observations indicate that imidazole reactions in the presence of the C 0 2
microwave plasma environment occur not by a hydrogen abstraction of the N-H
bonds, but by the hydrogen abstraction of the H-C=C-H entities on imidazole.
With these findings in mind, let us now examine the C=C and C=N stretching
regions for imidazole reacted to the PDMS surface. This is illustrated in Figure 5.8b.
Trace B of Figure 5.8b illustrates ATR FT-IR spectrum of imidazole reacted to the
PDMS surface at 10 sec discharge times. This spectrum exhibits the appearance of
new bands at 1600 and 1555 cm*1 which are attributed to the C=C and C=N
stretching modes of imidazole. When discharge times increase up to 30 sec, the
intensities of these bands increase. However, similar to the 0 2 microwave plasma
conditions (Figure 5.5b), no N-H deformation bands due to imidazole are detected.
In Figure 5.8c, traces A and B, the band at 1381 cm*1 is detected, and it is
attributed to the C-H deformation modes of the H-C=C-H entities of imidazole
molecules. However, when CO2 microwave plasma are employed for 10 sec
(trace C) to 30 sec (trace E), no bands due to the H-C=C-H entities of imidazole are
detected. On the other hand, the band at 1401 cm*1 is detected. This is shown in trace
124
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
C, and this band is attributed to the C-H deformation modes of newly formed
CH3 -C=C-CH3 entities of imidazole on the PDMS surface. These observations
indicate that imidazole reactions occur through hydrogen abstraction of the N-H and
H-C=C-H entities of imidazole, followed by the formation of the CH3 -C=C-CH3
entities, without a cleavage of the C=C and C=N bonds on an imidazole ring.
Figure 5.8d, trace A, illustrates ATR FT-IR spectrum of the Ar microwave plasma
reactions on the PDMS surface. The bands at 1075 and 1045 cm' 1 are attributed to
the formation of the Si-CH2-N and Si-CH2 -CH3 linkages, respectively. When C 0 2
microwave plasma are employed in the same imidazole reactions, the appearance of
a new band at 1061 cm' 1 is detected (trace C), and the band intensity increases when
discharge times range from 10 sec (trace C) to 30 sec (trace E). Since this band is
attributed to the CH2-N stretching modes, these observations indicate that the
CH3-C=C-CH3 entities of imidazole, which already reacted to the PDMS surface,
are now being hydrogen abstracted to form CH3 -C=C-CH2« radicals. These radicals
react with subsequent imidazole molecules through hydrogen abstraction of the N-H
bonds, which form CH2-N linkages between imidazole rings.
A proposed mechanism responsible for these reactions at the PDMS surface is
shown in Figure 5.9, Schemes 1-4. The Si-0« radicals react with imidazole
molecules through hydrogen abstraction of the N-H bonds, which form the
Si-O-imidazole entities on the PDMS surface (Figure 5.9-1). Hydrogens on the HC=C-H groups of the Si-O-imidazole species are substituted with the CH3 groups
(Figure 5.9-2), followed by their hydrogen abstraction to form Si-0-imidazole-CH2«
radicals
(Figure 5.9-3). These radicals react with subsequent imidazole
125
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1
S i— CH3
vh
1 N -H
N=^
H
( 1)
co2
plasma
+ «ch 3
VH w
• h 2c
co2
+
plasma
H
1
^i — 0 •
1
^i — 0 •
n
H
H.
+
NK
1 N -H
N=<
H
=<N
H
1
co2 ^
(2)
H
(3)
°
H
I
1
+ *CH3
%=N
H
N
>4
H
%
i
C02
—
plasma
r - o ~ NM
H
H
C02
plasma
S i— 0 — N
H>= n
plasma
H
I
>=N
ji-O -N
1
h
I
- - - ■ »•
CH3
%=N
Si— O — N
]
CH3
H
CH2 «
yR
ch 2 - n
n
> -N
H
(4)
^ i— 0 — n J ^
n.
H
R
C02
N=<
H
CH2 <
J.
?p
H
ru,
S=N
R
I> = ^
/ CH2
T
C^ 2
CM,
^ 2
*CU *X>
R -y ^
R; CH3, CH2 *
Figure 5.9. Proposed reaction mechanisms of imidazole reactions on the PDMS
surface in the presence of C 0 2 microwave plasma.
126
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R
molecules through the formation of the CH2 -N linkages. They likely form
multilayers of imidazole rings on the PDMS surface, which are depicted in Figure
5.9-4. Surface concentration of imidazole reacted to the PDMS surface in the
presence of CO2 microwave plasma is plotted in Figure 5.10. As discharge times
increase from 5 sec to 30 sec, surface concentrations o f imidazole increases linearly
from 0.05xl0*7to 0.42xl0'7 mole/1.
Quantitative analysis of imidazole reactions in the presence of 0 2 plasma shows
the highest yields, namely 9.73xl0 *7 mole/1. As shown in Figure 5.7, the formation
of two different lengths of reactive sites, S i-O and Si-O-CHo* radicals, in the
presence of O2 microwave plasma reduces the occurrence of crowding effects with
neighboring imidazole rings on the PDMS surface. Therefore, formation of two
reactive sites is responsible for the highest yields o f these reactions. On the other
hand, the CO2 microwave plasma reactions show the lowest yields of imidazole,
0.42x10‘7 mole/1, resulting from the formation o f the CH2 -N linkages with
imidazole. This is illustrated in Figure 5.9. At the discharge times above 20 sec
under 0 2 and C 0 2 microwave plasma conditions, the imidazole concentrations vary
linearly with the discharge times. However, Ar microwave plasma reactions result in
a decrease of the imidazole concentrations on the PDMS surface. Although there is a
limited supply of the monomer above
20
sec discharge times, chemically active O2
and CO2 are incorporated in the plasma reactions to form reactive sites for further
reactions. On the other hand, in the presence of Ar microwave plasma, surface
ablasion is predominant at the discharge times above 20 sec. Our studies using
atomic force microscopy confirmed this conclusion. 15
127
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0.50
0.40
«
o
B
oc
uo
0.30
0.20
0.10
0.00
0
7
14
21
28
35
Discharge time (sec.)
Figure 5.10. Surface concentration changes of imidazole reacted to the PDMS
surface in the presence of CO2 microwave plasma plotted as a function of discharge
time.
128
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.4 Conclusions
Based on ATR FT-IR spectroscopic analysis, which allowed us surface analysis at
1.3 pm from the surface, when Ar microwave plasma are utilized, imidazole
molecules react to the PDMS surface to form Si-CH2-imidazoIe species through
hydrogen abstraction of the N-H bonds. However, when 0 2 or C 0 2 microwave
plasma are employed, reactive discharge gases alter imidazole reactions.
0 2
microwave plasma reactions reveal that the imidazole radicals react with Si-0» and
Si-0-CH2* species, which form Si-O-imidazole and Si-0-CH2-imidazole entities on
the PDMS surface. When C0 2 is utilized in the presence of microwave plasma,
these species form Si-O-imidazole linkages. However, protons on the H-C=C-H
entities in the presence of imidazole are substituted with the CH3 groups. These
reactions are followed by the hydrogen abstraction to form Si-0-imidazole-CH2«
radicals. These radicals react with subsequent imidazole molecules through the
formation o f the CH2-N linkages, which result in the formation of multilayers of
imidazole rings on the PDMS surface.
Quantitative ATR FT-IR analysis shows that the highest yields of the imidazole
reactions occur when the 0 2 microwave plasma conditions are employed. Ar
microwave plasma reactions result in a decrease of the imidazole concentrations
above 20 sec discharge times. It is believed that the diminished supply of imidazole
into the gas phase lowers the imidazole content of the PDMS surface. In contrast,
concentration levels of imidazole on the PDMS surface in the presence of the 0 2 and
C 0 2 microwave plasma increase linearly for the discharge times above 20 sec.
129
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.5 References
1. H. Yasuda, Plasma Polymerization, Academic Press, New York, 1985.
2. A. S. Hoffinan, Adv. Polym. Sci., 1984, 57, 141.
3. L. T. Nguyen, N. H. Sung, and N. P. Suh, J. Polym. Sci., Polym. Lett. Edn., 1980,
18, 541.
4. S. L. Kaplan and P. W. Rose, Plastics Eng., 1988, 44, 77.
5. T. Kasemura, S. Ozawa, and K. Hattori, Adhesion, 1990, 33, 33.
6
. J. P. Badey and E. Urbaczewski-Espuche, Polymer, 1994, 35, 2472.
7. S. R. Gaboury and M. W. Urban, Polym. Commun., 1991, 32(13), 390.
8
. S. R. Gaboury and M. W. Urban, Langmuir, 1993, 9, 3225.
9. D. F. O’Kane and D. W. Rice, J. Macromol. Sci. Chem., 1976, A10, 567.
10. G. P. Lopez and B. D. Ratner, J. Appl. Polym. Sci., Appl. Polym. Symp., 1990,
46,493.
11. G. P. Lopez and B. D. Ratner, Langmuir, 1991, 7, 766.
12. H. Kim and M. W. Urban, Langmuir, 1995, 11, 2071.
13. H. Kim and M. W. Urban, Langmuir, 1996, 12, 1047.
14. H. Kim and M. W. Urban, Langmuir, 1996, 12, 1051.
15. H. Kim, M. W. Urban, F. Lin, and D. J. Meier, Langmuir, 1996, 12, 3282.
16. M. W. Urban, Attenuated Total Reflectance Spectroscopy o f Polymers-Theory
and Practice, American Chemical Society, Washington, DC, 1996.
17. M. W. Urban, Vibrational Spectroscopy o f Molecules and Macromolecules on
Surfaces, Wiley-Intersciences, New York, 1993.
18. H. G. De Wit and J. C. Van Miltenburg, J. Chem. Thermodyn., 1983, 15(7), 651.
130
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19. A. L. Smith and D. R. Anderson, Appl. Spectrosc., 1984, 38(6), 822.
131
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 6
MICROWAVE PLASMA REACTIONS OF IMIDAZOLE
ON POLYURETHANE ELASTOMER SURFACES:
A SPECTROSCOPIC STUDY
132
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6.1 Introduction
Due to desirable various chemical and physical properties, polyurethane (PU)
elastomers are often used in countless applications. However, to achieve necessary
surface properties, their surfaces may require modifications. Among many surface
reactions, gas plasma reactions are widely used due to their ability to alter surface
properties, such as surface friction, 1 adherence, 2,3 and biocompatibility, 4,5 while
maintaining bulk properties. This approach opens numerous opportunities for
modifying surface properties and also creating reactive sites for further surface and
interfacial reactions.
We6-7 utilized microwave energy to generate plasmas which allowed us to create
new surface species of imidazole on crosslinked polydimethylsiloxane (PDMS)
elastomers under closed and open flow microwave plasma reactor conditions.
Analysis of the imidazole molecules chemically attached to PDMS surfaces and
their reactive sites resulting from Ar microwave plasma reactions was conducted
using attenuated total reflectance (ATR) FT-IR spectroscopy. In these studies, we
established that imidazole radicals resulting from the hydrogen abstraction of the
N-H bonds react with the SiCH2» radicals, to form Si-CH2-imidazole entities in Ar
microwave plasma reactions under closed reactor conditions. On the other hand,
under open flow reactor conditions, imidazole molecules react to PDMS surfaces
through ring opening, resulting in the formation of C=C surface groups.
Although we used the same experimental setup of closed and open flow reactor
conditions to obtain imidazole entities on PU surfaces, we could not detect any
newly created species resulting from Ar microwave plasma reactions of imidazole.
133
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Therefore, we used Ar microwave plasma reactions on imidazole, which were
incorporated into a PU surface before microwave plasma reactions. Using ATR
FT-IR spectroscopy, formation of surface reacted imidazole molecules on PU, effect
of microwave plasma parameters on surface reactivity was investigated.
Furthermore, we quantified the amount of imidazole molecules chemically attached
to PU surfaces at various depths from the surface.
6.2 Experimental
6.2.1 Sample Preparation
Imidazole was absorbed into the PU by swelling its soft segments in methylene
chloride. In a typical experiment, 0.015 mole of imidazole (Aldrich Chemical Co.)
was dissolved in 50 ml of methylene chloride (Aldrich Chemical Co.), and PU
(EN-20, Conap Co.) was immersed in such a solution for 15 min at room
temperature. After removal, to clean the PU surface, PU was washed with methanol.
Such prepared PU specimens were evacuated and storred in a vacuum desiccator.
Microwave plasma reactions were conducted using the experimental setup. 6
6.2.2 Spectroscopic Measurements
ATR FT-IR spectra were collected on a Mattson Sirius 100 single beam
spectrometer. A resolution of 4 cm' 1 and a mirror speed of 0.3 cm s' 1 were used. The
ATR cell was aligned at various angles of incidence using a 30° angle parallelogram
Ge crystal and 45° angle parallelogram KRS-5 crystal. Each spectrum represents 200
coaded scans ratioed against a reference spectrum obtained from
200
134
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
coaded scans
of an empty ATR cell. All spectra were corrected for spectral distortion using
Q-ATR software. 8
6.3 Results and Discussions
Before we analyze PU surfaces and determine potential reactions with imidazole,
let us realize that, although urethane linkage shown in Figure 6.1 is a primary
component o f the PU structure, there are “soft” and “hard” segments in a PU
network which may have a different affinity for reactions with other species. This is
why, despite our efforts to react imidazole to PU, surfaces using a traditional
experimental setup so successfully utilized to the PDMS/imidazole reactions1
appeared to create a problem. Several efforts resulted in unsuccessful reactions; and,
for that reason, imidazole was incorporated into the surface of PU by immersing it in
imidazole-containing methylene chloride solution. Details concerning procedures
involved are described in Section 6.2. After PU specimens were immersed into the
solution, they were exposed to a microwave radiation to facilitate reactions of
imidazole and PU linkages.
6.3.1 Analysis of Imidazole Plasma Reacted PU
Trace A in Figure 6.2 illustrates ATR FT-IR spectrum of the PU elastomer
surface. This spectrum illustrates the N-H and C-H stretching regions of the spectra,
with the band at 3335 cm' 1 due to N-H stretching modes and the C-H stretching
bands of antisymmetric and symmetric C-H modes at 2930 and 2857 cm '1. Trace B
of Figure 6.2 illustrates ATR FT-IR spectrum of a PU specimen in which imidazole
135
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
o
C
h
h
N
R'
o
C
N
O
R
Polyurethane linkage
Where R \ hard segment; R, soft segment
Figure 6.1. Urethane linkages containing soft and hard segments.
136
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
o
W av e nu m be rs
Figure 6.2. ATR FT-IR spectra in the 4000-2000 cm' 1 region of imidazole-absorbed
PU in the presence of Ar microwave plasma under 26.6 Pa in closed reactor: A Unreacted PU; B - imidazole-PU; and C - imidazole-PU under 26.6 Pa/10 sec.
137
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
was allowed to absorb into the soft segments of PU, but the specimen was not
exposed to microwave radiations. As seen, characteristic bands due to imidazole are
not detected on the surface of PU. On the other hand, trace C of Figure 6.2 illustrates
ATR FT-IR spectrum of imidazole-absorbed PU exposed to Ar microwave plasmas.
It appears that this spectrum exhibits the appearance of new bands at 3127 cm '1,
attributed to C-H stretching bands of the imidazole, and the 2623 cm' 1 band,
attributed to the -NH N<— ionic interactions between hydrogen of N-H groups in
imidazole, and a lone pair electrons of nitrogen in PU .9 Based on these findings,
imidazole molecules absorbed in PU react into the PU in the presence of Ar plasma.
With these data in mind and our previous studies on PDMS, 6 let us consider how
discharge pressures affect the efficiency o f reactions. Figure 6.3 illustrates
ATR FT-IR spectra of imidazole-absorbed PU in the presence of Ar plasma under
various discharge pressure conditions. As discharge pressures decrease from 106.7
Pa
(trace B) to 26.6 Pa (trace E), the intensity of the C-H stretching bands due to
the C=C entities in imidazole at 3132 cm' 1 increases. As the pressure decreases from
53.3 Pa (trace D), a new band at 3387 cm' 1 attributed to the N-H stretching modes in
imidazole develops, and its intensity reaches a maximum when 26.6 Pa pressure is
used (trace E). At the same time, the intensity of the band at 2639 cm' 1 due to the
NH"'N ionic interactions between imidazole and PU increases. This is illustrated in
traces D and E. Based on these findings, it appears that the imidazole molecules are
attached to the PU surface when microwave plasmas are employed. Furthermore, the
amount of imidazole molecules detected at the PU surface increases, when lower
discharge pressures are employed.
138
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
— ---
— ■E
' ----- ~
•
~
.B
.
4000
3500
3000
D
2500
2000
ffavenumbers
Figure 6.3. ATR FT-IR spectra in the 4000-2000 cm*1 region o f imidazole-PU in the
presence of Ar microwave plasma under various discharge pressure in closed
reactor: A - Unreacted PU; B - imidazole-PU under 106.7 Pa/30 sec; C - imidazolePU under 79.8 Pa/30 sec; D - imidazole-PU under 53.3 Pa/30 sec; and E imidazole-PU under 26.6 Pa/30 sec.
139
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Although these data provide some evidence for the imidazole-PU reactions, at this
point, it would be highly speculative to assess if imidazole is chemically bonded to
the PU surface. For that reason, we conducted a series o f experiments in which the
effect of discharge pressures of Ar plasma on imidazole-PU was examined. While
trace A of Figure 6.4 illustrates the spectrum of unreacted PU, traces B and C of
Figure 6.4 illustrate ATR FT-IR spectra of imidazole-PU reacted in the presence of
Ar plasma under 106.7 and 79.8 Pa pressures. It appears that traces B and C are
similar to the spectrum of unreacted PU (trace A). In contrast, trace D of Figure 6.4
illustrates ATR FT-IR spectrum of imidazole-PU in the presence of Ar plasma under
53.3 Pa. As a result of microwave radiation, the appearance of a new bands at
1584 cm' 1 attributed to the C=C bond stretching modes and the 1441 cm' 1 band
attributed to the C-H deformations of the C=C bonds of imidazole is detected. As
discharge pressures decrease to 26.6 Pa (trace E), not only an increase of the band
intensity at 1441 cm-1 is detected, but also a decrease of the amide I band intensity
of the C=0 stretching at 1731 cm' 1 and the amide II band intensity of N-H
deformation at 1532 cm' 1 due to amide groups in PU are observed. On the other
hand, the C=N stretching band due to imidazole is not detected in this spectral
region. Based on the analysis of these results, it appears that imidazole reacts to PU
surface under Ar plasma pressures lower than 79.8 Pa, and the reaction sites are the
C=N bond of imidazole and the amide groups on PU.
140
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
»V
"*
uC
-■=’
*
'E
-
~
.D
- '
_
~
____________
1750
1700
B
- „__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
1650
1600
1550
1500
.C
~~ 1450
A
1400
Waven u m b e r s
Figure 6.4. ATR FT-IR spectra in the 1800-1400 cm' 1 region of imidazole-PU in
the presence of Ar microwave plasma under various discharge pressure in closed
reactor: A - Unreacted PU; B - imidazole-PU under 106.7 Pa/30 sec; C - imidazolePU under 79.8 Pa/30 sec; D - imidazole-PU under 53.3 Pa/30 sec; and E imidazole-PU under 26.6 Pa/30 sec.
141
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
63.2 Imidazole Reaction Mechanism on PU Surfaces
Although the results presented in Figure 6.4 identify reaction sites on PU and
imidazole entities, to establish a detailed mechanism responsible for these reactions,
a different spectral range needs to be analyzed. For that reason, traces B and C of
Figure 6.4 illustrate the 1300-1000 cm' 1 region. It appears that the characteristic
features detected in traces B and C are similar to that detected in trace A. As the
discharge pressures of Ar plasma decrease from 53.3 Pa (trace D) to 26.6 Pa (trace
E), the intensity of the amide IV band of the (C=0)-N stretching band at 1222 cm"1
of PU decreases. On the other hand, the intensity of the C-O-C stretching band at
1063 cm' 1 increases.
Based on these data, one can deduce how the C-O-C linkages between imidazole
and PU are being formed. As illustrated in Figure 6.5, traces D and E exhibit the
appearance of a new band at 1095 cm' 1 attributed to the C-N stretching modes. This
band results from the reactions of imidazole through C=N bond opening to the PU
surface. The reaction mechanism responsible for imidazole-PU in the presence of Ar
plasma is proposed in Figure 6 .6 . The C=N bond of imidazole and the amide C=0
bond of PU opens up to create the C-O-C linkage. As we recall, this reaction occurs
in the presence of Ar microwave generated plasmas. Furthermore, hydrogen
abstraction of the N-H bonds in PU results in the formation of the C-N linkage
through the C=N bond opening of imidazole. The interactions of the NH'"N
molecules between the PU linkages and chemically attached imidazole molecules
were already discussed in conjunction with Figure 6.3 (traces D and E).
142
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
W a v e n u m b e rs
Figure 6.5. ATR FT-IR spectra in the 1400-1000 cm' 1 region of imidazole-PU in
the presence of Ar microwave plasma under various discharge pressure in closed
reactor: A - Unreacted PU; B - imidazole-PU under 106.7 Pa/30 sec; C - imidazolePU under 79.8 Pa/30 sec; D - imidazole-PU under 53.3 Pa/30 sec; and E imidazole-PU under 26.6 Pa/30 sec.
143
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Polyurethane
+
H
"C
-N
II
,c
H
"
' S
.
H
Imidazole
Ar plasma
H
H
\
=
c /
N— H
N
H
H
0
H
1
C
N
R’ ---- N
H
C
C H-----O ----- R ----- O ----H
N
/
H— N>
H
H
Figure 6 .6 . Reaction mechanism of imidazole reacted to PU in the presence of Ar
microwave plasma.
144
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission
6.3.3 Depth Profiling of Imidazole Reacted PU
So far, we determined that a 26.6 Pa Ar pressure and 30 sec discharge times in Ar
microwave plasma result in high surface yields of imidazole reactions on the PU
surface. If this is the case, we would like to establish the amount of imidazole
reacted to the PU surface, including quantitative analysis as a function of the depth
of penetration. Although this can be accomplished by changing an angle of
incidence in an ATR setup or using ATR crystal with various refractive indicies, 8,10
the band distortions resulting from optical effects should also be considered.
Figure 6.7 illustrates ATR FT-IR spectra, corrected using Q-ATR algorithm, of
imidazole reacted to PU recorded at various incidence angles. The reactions were
conducted under 26.6 Pa and 30 sec discharge times conditions in Ar microwave
plasma. As the angle of incidence of infrared light entering into the ATR crystal
changes from 40° (KRS-5 - trace A) to 60° (Ge - trace F), the intensities of the C=C
stretching band at 1581 cm*1 and the C-H stretching band due to C=C bond at
1447 cm*1 increase. On the other hand, the intensities o f the amide C=0 stretching
band at 1732 cm*1 and N-H deformation band of PU at 1533 cm*1 decrease. These
results indicate that the amount of imidazole reacted to the PU surface decreases
with the increasing depth of penetration.
Although at this point we are in a position to quantify the results shown in
Figure 6.7, let us temporarily postpone this analysis and realize that the experiments
were conducted using a closed system. 1 As our previous studies1 indicated, reaction
mechanisms of imidazole to polydimethylsiloxane (PDMS) can vary, depending
upon using open and closed reaction chamber conditions. For that reason, we
145
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 6.7. ATR FT-IR spectra in the 1800-1400 cm' 1 region of imidazole-PU in the
presence of Ar microwave plasma under closed reactor conditions using various
incidence angle of infrared light: A - 40° angle of KRS-5 crystal; B - 50° angle of
KRS-5 crystal in closed reactor; C - 60° angle of KRS-5 crystal; D - 40° angle of Ge
crystal; E - 50° angle of Ge crystal; and F - 60° angle of Ge crystal.
146
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
analyzed the effect of open flow microwave plasma reactor on the imidazole
reactions with PU surfaces. Figure
6 .8
illustrates ATR FT-IR spectra of imidazole
reacted to PU using open flow microwave plasma reactor. In contrast to the PDMS
studies, the reaction mechanisms o f imidazole for the open and closed microwave
plasma reactors result in the appearance o f the C=C bond stretching band of
imidazole at 1588 cm' 1 and the disappearance o f amide C=0 stretching band of PU
at 1732 cm'1.
6.3.4 Quantitative Analysis of Imidazole Reacted PU
While the objective of many surface studies is determination of the chemical
makeup of the surface species, today’s applications require quantitative knowledge.
However, one of the drawbacks o f the infrared analysis is the necessity of obtaining
a calibration curve. Typically, such a curve represents a band intensity plotted as a
function of concentration, and its slope is equal to an extinction coefficient. In our
case, the C=C bond stretching band will be used as a measure of surface reactions
on PU. Figure 6.9 shows the plot of absorbance of the C=C stretching band of
imidazole, plotted as a function of imidazole concentrations in KBr powder. Based
on this plot, the extinction coefficient of the C=C bond of imidazole was determined
to be 86.49 1/mole-cm. Since the next step of analysis requires volume
concentrations, the volume concentration of imidazole reacted to polyurethane can
be obtained using Beer-Lambert law
P= e c
147
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
•F
.E
•D
('
.
B
A
1750
1700
1650
1600
1550
1500
1450
1100
Wa veil u m b e r s
Figure 6 .8 . ATR FT-ER spectra in the 1800-1400 cm' 1 region of imidazole-PU in
the presence of Ar microwave plasma under open flow reactor conditions using
various incidence angle of infrared light: A - 40° angle of KRS-5 crystal in open
flow reactor; B - 50° angle of KRS-5 crystal; C - 60° angle of KRS-5 crystal; D 40° angle of Ge crystal; E - 50° angle of Ge crystal; and F - 60° angle o f Ge crystal.
148
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Extinction cooff. - 80.40 (l/mote-cm)
A bsorbance
0.56
0.42 -
0.28 -
0.14 -
0.00 --------------------------------------------------------------------------------------------0.00
0.16
0.32
0.48
0.64
0.80
(E-2)
C oncentration (mole-cm/1)
Figure 6.9. Plot of C=C stretching band absorbance for imidazole as a function of
concentration.
149
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
where e is the extinction coefficient, c is the concentration, and p is the linear
absorptivity. While these quantities are necessary to quantify the data, one issue that
should be also mentioned is that to use ATR FT-IR spectra for quantitative purposes,
they should be corrected for optical effects. 8,10 For that reason, we developed an
algorithm which we utilize in all ATR studies. In this algorithm, the absorption
index spectrum is refined by an iterative process that minimizes the difference
between the true and calculated reflectivity resulting from optical effects, while
maintaining the exact Kramers-Kronig relation between absorption (k) and refractive
a
index (n ).'
iA
This iterative process was used in conjunction with numerical
Kramers-Kronig transformation (KKT) method to obtain ATR spectra suitable for
quantitative analysis using Beer-Lambert law.
A
The volume concentration of imidazole reacted to polyurethane is plotted in
Figure 6.10 and changes with the depth o f penetration. Under closed reactor
conditions, as the depth of penetration increases from 0.30 pm to 1.31 pm, the
volume
concentration
of imidazole decreases
from 7.51xl0'3 mole/1 to
0.30x10‘3 mole/1. On the other hand, the volume concentration of the imidazole
reacted to polyurethane under open flow reactor conditions decreases from
2.54xl0 *3 mole/1 to 1.51xl0‘3 mole/1. Based on the results illustrated in Figure 6.10,
it is apparent that the amount of imidazole reacted to the PU surface varies with the
surface depth and the conditions of the surface reactions. Furthermore, as the depth
of penetration changes, reactions under closed conditions lead to a significant
change while going from 0.3 to 1.25 pm. In contrast, only small concentration
changes are detected for the open system.
150
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8 .0 0
c~■
>
w
X
-?
C
o
a
c
<o
o
c
o
U
6 .4 0
4 .8 0
3 .2 0
1.60
0.00
0.00
—
0.30
0.60
0.90
1.20
1.50
D epth o f p e n e tra tio n (p m )
Figure 6.10. Plots o f volume concentration for imidazole reacted to PU as a function
of depth o f penetration: # - closed microwave plasma reactor; ▲ - open flow
microwave plasma reactor.
151
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The major differences between open and closed microwave plasma reactors are
resident times of a gas phase monomeric molecule and the pressure changes during
the plasma reactions. In a closed microwave plasma reactor, the resident time of the
gas phase is longer, compared to that in the open flow reactor. This phenomenon is
responsible for larger quantities of imidazole reacted to the PU surface in the open
flow reactor. As the depth of penetration into the PU increases in a closed
microwave plasma reactor, concentration of gas phase imidazole molecules
decreases due to increase of the pressures during plasma discharge, which results in
the lower plasma energy1 and subsequent decrease of the volume concentration of
imidazole reacted to the PU surface. However, for the open flow reactor, the
pressures are constant during plasma discharge. Therefore, concentration of the gas
phase imidazole molecules remains constant, thus resulting in relatively uniform
volume concentrations of imidazole at various depths.
6.4 Conclusions
To facilitate the imidazole-PU reactions, PU specimens were immersed in a
methylene chloride-imidazole solution, followed by exposure to the microwave
radiation using closed and open flow microwave plasma conditions. Physisorbed
imidazole molecules incorporated into PU elastomer networks react with PU
surfaces in the presence of Ar plasma, which creates chemically bonded imidazole
entities to the PU surfaces through the C=N opening of imidazole and C=0 opening
of PU, resulting in the C-O-C linkages between PU and imidazole. High surface
yields of imidazole reacted to the PU surfaces is obtained at 26.6 Pa Ar pressure
152
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and 30 sec discharge times.
Analysis of microwave plasma reactions conducted under closed and open flow
reactor conditions reveals the same structures, but concentrations o f imidazole
reacted to PU surfaces varies with depth. Under closed reactor conditions, as the
depth of penetration increases, the volume concentration of imidazole entities
reacted to PU surfaces decreases. On the other hand, the volume concentration of
imidazole entities reacted to PU surfaces under open flow reactor conditions remains
constant with depth penetration. The decrease of imidazole concentration reacted to
PU surfaces with depth of penetration under closed reactor conditions results from a
decrease of the gas phase imidazole molecules due to the increased pressure during
microwave plasma discharge. For the open flow reactor, the pressure remains
constant during Ar microwave discharge, which results in the uniform concentration
of imidazole entities reacted to the PU surfaces.
6.5 References
1. D.L. Cho and H. Yasuda, J. Appl. Polym. Sci., Appl. Polym. Symp., 1988, 42,
139.
2. N.J. DeLollis and O.J. Montoya, Adhes., 1971, 3, 57.
3. E.M. Liston and P.W. Rose, Int. Symp. Plasma Chem., 7th, 1985.
4. H. Yasuda and M. Gazick, Biomaterials, 1982,3,68.
5. W.R. Gombotz and A.S. Hoffman, Crit. Rev. Biocompt., 1987,4(1), 1.
6
. H. Kim and M. W. Urban, Langmuir, 1995, 11, 2071.
7. H. Kim and M. W. Urban, Langmuir, 1996, 12, 3282.
153
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8
. J.B. Huang and M. W. Urban, Appl. Spectrosc., 1992, 46(11), 1666.
9. S.R. Gaboury and M. W. Urban, Langmuir, 1994, 10, 2289.
10. J.B. Huang and M. W. Urban, Appl. Spectrosc., 1993, 47(7), 973.
154
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 7
MICROWAVE PLASMA REACTIONS OF IMIDAZOLE
ON POLYVINYLCHLORIDE SURFACES:
A SPECTROSCOPIC STUDY
155
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7.1 Introduction
We developed a new method for reacting gaseous monomers to various substrates.
This process involves the use of microwave plasma energy. 1-6 Its attractiveness
comes from the fact that microwave plasma reactions are fast and clean and usually
do not alter bulk polymer properties. Although one could argue, and perhaps
rightfully so, that one of the drawbacks of the reactions conducted in the plasma gas
phase is a complexity of the reaction mechanisms, advantages are obvious. We4-6
utilized microwave energy to generate plasmas that allowed us to react imidazole
molecules to polydimethylsiloxane (PDMS) and polyurethane (PU) surfaces.
Furthermore, during these studies, we developed two experimental approaches that
may significantly alter the nature of the produced surface species. 4 For example,
using closed reactor conditions, imidazole rings are chemically bonded to the PDMS
surface through a hydrogen abstraction of the N-H bonds. On the other hand, when
the open reactor conditions are employed, imidazole rings open up, forming the C=N
species on the PDMS surfaces.
In this study, our efforts will concentrate on imidazole reactions to
polyvinylchloride (PVC) surfaces. The primary interest in conducting the studies on
the PVC surfaces comes from its usefulness as an implant for numerous biomedical
applications. In this case, we will employ closed microwave plasma reactions and
focus on the surface reactivity and morphology of PVC as well as on the formation
of the reactive surface sites, reaction mechanisms, and quantitative analysis of the
surface species at various depths. Like previous studies, ATR FT-IR spectroscopy
will be used for the surface analysis of the imidazole reactions.
156
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7.2 Experimental
7.2.1 Substrate Preparation
Polyvinylchloride (PVC) films were prepared from
PVC powder (Aldrich
Chemical Co.) using two methods. In the first method, approximately 3 g of PVC
powder were placed in a hot pressure plate at 17.9 kPa for 3 min at 185 °C and
allowed to cool to room temperature. In the second method, 2 g of PVC powder
were
dissolved
in
10
mL
of
tetrahydrofuran
(THF),
cast
onto
a
polytetrafluoroethylene (PTFE) plate, and allowed to dry for 24 hrs.
7.2.2 Surface Reactions
Plasma reactions were conducted using closed reactor conditions. 4
In the Ar
plasma surface reactions, a PVC substrate, with approximate dimensions of
20x20x0.5 mm, was placed into a microwave plasma closed reactor. The reactor
was evacuated to 1.33 Pa, followed by purging it with Ar and 0 2 gases to an
atmospheric pressure and reevacuated to a desired experimental pressure, typically
about 26.6 Pa. At this point, microwave radiation to induce plasmas was turned on.
The same procedures were utilized for reacting imidazole (Aldrich Chemical Co.)
monomer. In this case, before purging and evacuating the reaction chamber, a solid
monomer and PVC substrate were placed into a reactor.
7.2.3 Spectroscopic Measurements
ATR FT-IR spectra were collected on a Mattson Sirius 100 single beam
spectrometer. A resolution of 4 cm' 1 and a mirror speed of 0.3 cm s*1 were used. The
157
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ATR cell was aligned at a 45° angle o f incidence using a 45° angle parallelogram
KRS-5 crystal. Each spectrum represents 200 coded scans ratioed against a reference
spectrum obtained from 200 coaded scans of an empty ATR cell. All spectra were
corrected for spectral distortions using Q-ATR software. 7
7.3 Results and Discussion
Although our earlier studies4"6' 8"u indicated that microwave plasma can be
successfully used for reactions of monomers to elastomeric surfaces, surface
reactions on thermoplastic polymers may be influenced by the surface morphology
of a polymer. For example, if the amounts of crystalline and amorphous components
vary near the surface, these variations will influence surface reactions. For that
reason, we prepared two specimens of PVC which crystallized from melt and casted
from a solvent. While the former is anticipated to have a higher crystalline content,
the latter will contain more amorphous network due to a plasticizing effect of a
solvent. To react imidazole to PVC surfaces, several attempts utilizing various Ar
plasma pressure conditions were made. In all cases, regardless of the experimental
conditions, spectral analysis showed that no imidazole reactions occurred. For
reference purposes, Table 7.1 provides a list of the IR bands observed in hot-pressed
and solvent-cast PVC ATR FT-IR spectra and their tentative assignments.
7.3.1 Analysis of Imidazole Plasma Reacted Hot-pressed PVC
When oxygen gas was used in the presence of microwave energy, the situation
changed. Figure 7.1 illustrates ATR FT-IR spectra of imidazole-reacted surface of
158
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 7.1. Tentative Band Assignments of Hot-pressed and Solvent-cast PVC
Band (cm'1)
PVC
Solvent-cast
Hot-pressed
1722
-C=0 stretching
~
1573
-C=N- stretching
—
1430
-CH2- deformation
-CH2- deformation
1339
-CH- deformation
-CH- deformation
1255
-CH2- wagging
-CH2- wagging
1071
-C-C- skeletal vib.
-C-C- skelatal vibration
960
-CH2- rocking vib.
-CH2- rocking vibration
695
-CH - rocking vib.
-CH- rocking vibration
620
-CCl stretching
-CCl stretching
„ T
159
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
.
1
__
-
-
1800
1(500
-
-
'
-
1400
.E
1200
_. . D
—
~
-
—
1000
_
—
-
- B
_
800
—
A
600
Wavenumbers
Figure 7.1. ATR FT-IR spectra in the 1900-500 cm' 1 region of imidazole reacted to
hot-pressed PVC in the presence of oxygen microwave plasma under various
pressure conditions using a closed reactor: A - Unreacted hot-pressed PVC; B oxygen microwave plasma at 106.4 Pa/10 sec; C - imidazole reacted to PVC at
106.4 Pa/10 sec; D - imidazole reacted to PVC at 26.6 Pa/10 sec; and E - imidazole
reacted to PVC at 13.3 Pa/10 sec.
160
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the hot-pressed PVC exposed to oxygen microwave plasma radiation discharged for
10 sec. As a reference point, trace A illustrates the spectrum of unreacted PVC, with
the bands listed in Table 7.1. It appears that in the hot-pressed PVC, the intensities
of the C-C skeletal normal vibrations at 1073 cm'1, the CH2 deformation modes at
1430 cm'1, and the CH2 rocking modes at 960 cm' 1 increase with the decreasing
initial discharge pressures in the plasma chamber. However, no imidazole reactions
are detected on the PVC surface. These observations indicate that, as the result of
the microwave plasma exposure, the PVC surface contains CH2 linkages
(traces B - E). Like previous studies, 4 lower gas plasma pressures result in the
enhancement of the CH2 deformation bands. Although there is no direct evidence for
the C-Cl bond cleavage in the hot-pressed PVC specimens, plasma reactions lead to
the formation of the C-H radicals abstracted from imidazole monomers to increase
the content of the -CH2- linkages.
7.3.2 Analysis of Imidazole Plasma Reacted Solvent-cast PVC
Figure 7.2 illustrates ATR FT-IR spectra in the C-H stretching region for
imidazole reacted to the solvent-cast PVC surface under various discharge pressures.
For reference purposes, traces A and B of Figure 7.2 illustrate the spectra of
unreacted and oxygen microwave plasma reacted to the PVC surface. The spectrum
of oxygen plasma reacted to the PVC surfaces exhibits the appearance of the -OH
stretching modes at 3410 cm' 1 due to the formation of carboxylic acid. It appears
that when imidazole is reacted to the PVC surface, the C-H stretching bands at 2926
cm' 1 and 2854 cm' 1 are detected, which is attributed to the C-H stretching bands of
161
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3G00
3500
3400
3300
3200
3100
3000
2900
2800
Wavenum beis
Figure 7.2. ATR FT-IR spectra in the 3700-2800 cm' 1 region of imidazole reacted
to solvent-cast PVC in the presence o f oxygen microwave plasma under various
pressure conditions using a closed reactor: A - Unreacted solvent-cast PVC; B oxygen microwave plasma at 106.4 Pa/5 sec; C - imidazole reacted to PVC at
106.4 Pa/5 sec; D - imidazole reacted to PVC at 53.2 Pa/5 sec; E - imidazole reacted
to PVC at 26.6 Pa/5 sec; and F - imidazole reacted to PVC at 13.3 Pa/5 sec.
162
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the -CH2- linkages formed from the C=C bond opening of imidazole molecules.
Like our previous studies, 4 these bands become stronger when initial discharge
pressures decrease from 106.7 Pa (trace C) to 13.3 Pa (trace F).
With this in mind, let us analyze the C=C and the C=N stretching regions, which
are illustrated in Figure 7.3. It appears that the appearance of two new bands at
1658 cm' 1 and 1587 cm*1 attributed to the imidazole I band of the ring and the C=N
stretching of imidazole are detected. On the other hand, although initial discharge
pressures decrease from 106.7 Pa (trace C) to 13.3 Pa (trace F), the C=C bond
stretching and the -CH=C-H deformation bands of imidazole molecules are not
detected. These observations indicate that imidazole reacts to the PVC surfaces
through a C=C bond cleavage, but the imidazole ring remains. When oxygen gas is
present in the plasma environment, the C=0 stretching band at 1724 cm' 1 is detected
(trace B), which is attributed to the formation of carboxylic acid species on the
surface of PVC. When imidazole is reacted to the PVC surface in the presence of
oxygen plasma at the initial discharge pressure of 106.4 Pa (trace C), the intensity of
the C=0 stretching band at 1775 cm' 1 also increases, indicating the formation of
ester linkages between the PVC surface and imidazole molecules. In addition, the
increase of the 1040 cm' 1 band intensity, relative to the C-C skeletal vibrations at
1071 cm' 1 (trace F), is attributed to the presence of ester linkages. These
observations indicate that imidazole reacts at the PVC surface, most likely by the
C=C cleavage in the presence of oxygen plasma. A possible mechanism of
imidazole reaction at the PVC surface is shown in Figure 7.4.
163
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2-
.
F
'
—
—
• •E
-
D
■
.
c
--------------------
•B
-
__
1800
1700
1600
1500
1400
1300
1300
A
1100
1000
W a v e n u m b e rs
Figure 7.3. ATR FT-IR spectra in the 1850-1000 cm*1 region of imidazole reacted
to solvent - cast PVC in the presence of oxygen microwave plasma under various
pressure conditions using a closed reactor: A - Unreacted solvent-cast PVC; B oxygen microwave plasma at 106.4 Pa/5 sec; C - imidazole reacted to PVC at
106.4 Pa/5 sec; D - imidazole reacted to PVC at 53.2 Pa/5 sec; E - imidazole reacted
to PVC at 26.6 Pa/5 sec; and F - imidazole reacted to PVC at 13.3 Pa/5 sec.
164
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A
O2 plasm a
-------------- >>
Closed system
O
II
h - C — OH
PVC
B
>C----- N
+
PVC
1
H
O2 plasma
------------ ►
Closed system
O
II
- c •OH
o
II
c ■
fHTJ
o —
Imidazole
x
T
'h
H
Figure 7.4. Proposed mechanism of imidazole reactions to solvent-cast PVC
surfaces in the presence of oxygen plasma.
165
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7.3.3 Depth Profiling of Imidazole Reacted PVC
Let us now examine how imidazole content may change as a function of the depth
from the PVC surface. Although this can be accomplished by changing an angle of
incidence in an ATR setup, 12 the band distortions resulting from the optical effects
in an ATR experiment also should be considered. For that reason, we7 developed an
algorithm which we utilize for all ATR spectra analyses. In this algorithm, the
absorption index spectrum is refined by an iterative process that minimizes the
difference between the true and calculated reflectivity resulting from optical effects,
while maintaining the Kramers-Kronig relation between absorption (k) and
refractive index (n) components of a spectrum. This iterative process can be used in
conjunction with the double-Kramers-Kronig transformation (KKT) method to
obtain ATR spectra free of distortions and suitable for quantitative analysis using the
Beer-Lambert law. Figure 7.5 illustrates ATR FT-IR spectra of imidazole reacted to
the PVC surfaces recorded at various incidence angles. The reactions were
conducted under 26.6 Pa and 5 sec discharge times in the presence of oxygen
microwave plasma. As the angle of incidence of infrared light into the KRS-5 crystal
changes from 40° (trace B) to 60° (trace D), the intensities of the imidazole I band of
ring structure 1662 cm*1 and the C=N stretching band at 1585 cm' 1 increase. These
results indicate that the amount of imidazole reacted to PVC surfaces decreases with
the increasing penetration depth.
166
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
=
'_
.D
-
’
'
-
,C
•B
- _ _ _ _ _ _ _
1800
1700
1600
I.500
A
1400
1300
1200
1100
1000
W avenum ber s
Figure 7.5. ATR FT - ER spectra in the 1850-1000 cm' 1 region of imidazole reacted
to solvent-cast PVC in the presence of oxygen plasma under 26.6 Pa/ 5 sec using
various incidence angle of infrared light: A - unreacted PVC at 60° angle of KRS-5;
B - imidazole reacted PVC at 40° angle of KRS-5 crystal; C - imidazole reacted at
50° angle of KRS-5 crystal; and D - imidazole reacted at 60° angle of KRS-5.
167
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7.3.4 Quantitative Analysis of Imidazole Reacted PVC
While the objective of many surface studies is determination of the chemical
makeup of the surface species, modem science also requires a quantitative
knowledge. However, one of the drawbacks of the infrared analysis is the necessity
of obtaining a calibration curve. Typically, such a curve represents the band intensity
plotted as a function of concentration, and the slope is equal to an extinction
coefficient. In our case, the imidazole ring stretching band detected at 1662 cm' 1 will
be used as a measure of surface reactions on PVC. Figure 7.6 shows the plot of
absorbance of the imidazole ring stretching band, plotted as a function o f imidazole
concentration in KBr powder. Using these data, the extinction coefficient of the
imidazole ring stretching band is determined to be 67.81 1/mole-cm. This calibration
curve, along with the double KKT analysis, 7 allows us to determine imidazole
concentration on the PVC surface at various depths. The results are shown in
Figure 7.7. As the depth of penetration increases from 0.86 pm to 0.93 pm, the
surface concentration of imidazole reacted to the PVC surface exhibits a decrease,
from 0.37x10^ mole/cm2 to 0.17XKT6 mole/cm2. At surface depths exceeding
0.93 pm, the imidazole concentration decreases considerably. Thus imidazole/PVC
reactions occur largely at the top 0 . 8 pm surface layers.
7.4 Conclusions
In this study, a closed microwave plasma reactor was used to react imidazole
molecules to PVC surfaces. Using ATR FT-IR spectroscopy, quantitative analysis of
newly created surface species was performed. In a hot-pressed PVC, no imidazole
168
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0.70
0.56
u
C
0.42
CB
£>
t- i
O
ca
X>
<
0.28
0.14
0 . 0 0 ----------------------------------------------------------------------------------------------0.00
0.16
0.32
0.48
0.64
0.80
(E-2)
Concentration (mole-cm/1)
Figure 7.6. Plot of absorbance of the imidazole ring stretching band as a function of
imidazole concentrations in KBr powder.
169
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Surface concentration ( mole/cm*)
(E-6)
0.42
0.34
0.25
0.17
0.08
0.00
0.75
0.86
0.97
1.08
1.19
1.30
Depth of penetration ( pm )
Figure 7.7. Plots of surface concentration for imidazole reacted to PVC surfaces as a
function of depth of penetration.
170
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
reactions were detected on the surfaces under argon and oxygen microwave plasma
conditions. However, under oxygen microwave plasma conditions, PVC surfaces
exhibit newly formed CH2 linkages, resulting from the C-Cl bond cleavages, to form
CH2 linkages on PVC surfaces. In a solvent-cast PVC, imidazole molecules react to
the PVC surface through the C=C cleavage of imidazole and the formation of the
ester linkages between PVC and imidazole molecules. Imidazole reactions are
inhibited in the hot-pressed PVC surfaces due to a high crystallinity content. On the
other hand, imidazole reacts to the solvent-cast PVC surfaces through a C=C
cleavage. Although there is no direct evidence that would account for the reactivity
of the solvent-cast PVC, it is believed that the presence of amorphous network due
to a plasticizing effect of a solvent is responsible for imidazole’s reacting to the
PVC surface. Quantitative analysis o f imidazole reactions to the solvent-cast PVC
surfaces at various depths indicates that imidazole reactions occur largely at the top
0.8 pm surface layers. As the depth increases from 0.8 to 1.2 pm, surface
concentration of imidazole reacted to the PVC surfaces decreases from
0.37.10*6 mole/cm2 to 0.17X10-6 mole/cm2.
7.5 References
1. M. Stewart, E. DiDomenico, and M.W. Urban, US Patent 5,364,662.
2. H. Yasuda, Plasma Polymerization, Academic Press, Orlando, FL, 1985.
3. H.Yasuda and A.K. Shamara, J. Polym. Sci., Polym. Phys. Ed., 1981, 19, 1285.
4. H. Kim and M. W. Urban, Langmuir, 1995, 11,2071.
5. H. Kim and M. W. Urban, Langmuir, 1996, 12, 1047.
171
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
. H. Kim and M. W. Urban, Langmuir, 1996, 12, 1051.
7. J.B. Huang and M. W. Urban, Appl. Spectrosc., 1992,46(11), 1666.
8
. S. R. Gaboury and M. W. Urban, Polym. Commun., 1991, 32(13), 390.
9. S. R. Gaboury and M. W. Urban, Polymer, 1992,33(23), 5085.
10. S. R. Gaboury and M. W. Urban, Langmuir, 1993, 9, 3225.
11. S. R. Gaboury and M. W. Urban, Langmuir, 1994, 10,2289.
12. J. P. Kunkel and M. W. Urban, J. Appl. Polym. ScL, 1993, 50, 1217.
172
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 8
REACTIONS OF THROMBRESISTANT MULTI - LAYERED THIN FILMS
ON POLYVTNYLCHLORIDE (PVC) SURFACES:
A SPECTROSCOPIC STUDY
173
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8.1 Introduction
During the last decade, there has been a considerable interest in the development
of biomedical materials that would exhibit thrombresistant surfaces. 1' 3 For that
reason, immobilization of heparin (HP) molecules on artificial surfaces has been
extensively explored as one of the possible means to create thrombresistant
surfaces. 4*7 Nitrous acid degraded HP was often employed to generate these
properties. These studies also indicated that to attach HP to polymeric surfaces, it is
necessary to form several layers that will serve as a buffer between the blood stream
and the surface of an artificial organ. Thus, to succeed using this approach, it is
necessary to create bonding sites available for surface reactions. 8 This issue is
particularly important when thermoplastic polymers, such as polyvinylchloride
(PVC), are used for artificial organs. Although several attempts to directly bond HP
molecules to the PVC surface were made, little success was met because HP
molecules were physisorbed, instead of being covalently or ionically attached to the
surface to provide sufficient thrombresistant activity. Therefore, to increase surface
charge density and the content of functional groups for HP immobilization, multi­
layered structures were proposed. However, the surface and interfacial chemical
structures responsible for thrombresistant activity o f HP are unknown. 5' 9*12
The objective o f this study was to elucidate the molecular origin of structures that
develop in multi-layered films that consist o f polyethyleneimine (PEI), dextran
sulfate (DS), and heparin (HP) attached to PVC surfaces. While chemical structures
of the species involved in surface reactions are shown in Figure 8.1 A, Figure 8 . IB
illustrates multi-layered thin films that will be deposited by direct solution exposure
174
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A. Chemical Structures
Polyethyleneimine (PEI)
Crosslinked polyethyleneimine
H2N-f CH2 CH2 N H ^C t^ CH2 NH2
H2N-eCH2 CH2 N H tnCH2 CH2 NH
CH3CHCH2CH
H2N-(-CH2 CH2 n h ^ c h 2c h 2n
Nitrso acid degraded heparin (HP)
Dextran sulfate (DS)
CH2 SO3 (H3 0 +)
C °2 h_ / q H
H
H'
H H/ H NH
S0 3 (H
30 +
)
S03 (H3CT)
H O
H
B. Multi - layered Thin Films
pvc -
so;
s
PEI
—
nc
S0 3
nh3
so^
so^
nh3
PVC
so^
so^
so^
nh3
so^
NH3
so^
so^
dci
PEI
NHj
nhJ-
DS
PEI
HP
nh3
(so^(H3o^
nh3
(so ^ o * )
S0 3
so^
nhJ
nh3
so^
nh3
CHO
nh3
Nitrso acid degraded heparin
Polyethylene mine
Dextran sulfate
Crosslinked polyethyleneimine
Dextran sulfete
Crosslinked polyethyleneimine
Suifonated polyvinylchbride
Figure 8.1. A - Chemical structures used in surface reactions; and B - Schematic
representation of multi-layered structures deposited on PVC.
175
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(dipping) and spin-coating methods. Although the choice of depositing thin films
using dipping and spin-coating exhibits appealing practical features, it appears that
both application methods may lead to different surface species. As shown in
Figure 8 .IB, the first step involves activation of the PVC surface by chemically
generating negatively charged SCVCNH**) groups through the sulfonation reactions,
followed by the formation of alternating crosslinked PEI and DS thin films. The top
uncrosslinked PEI layer is used to immobilize HP molecules, which forms the
outermost surface. Using attenuated total reflectance Fourier transform infrared
(ATR FT-IR) spectroscopy, these studies investigated the effect of pH of the
solution on the formation of thin films, concentration of the reacting species, and the
effect of crosslinking conditions on the formation of multi-layered thin films,
followed
by
identification of reaction
mechanisms
responsible
for
HP
immobilization.
8.2 Experimental
8.2.1. Substrate Preparation
PVC (Aldrich Chem. Co.; M.W. = 75, 000) was used as a substrate for depositing
multi-layered thin films. 10 w/w % of PVC in tetrahydrofuran (THF) was casted on
polytetrafluoroethylene (PTFE)-coated plate, followed by drying the PVC specimen
at room temperature for 24 hrs. Such prepared PVC films (20X20X1 mm) were
allowed to dry in a vacuum deccicator for an additional 24 hrs to remove residual
low molecular weight species (THF and H2O).
176
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8.2.2 Surface Reactions
Multi-layered thin films were deposited on the PVC surface using solution
exposure (dipping) and spin-coating techniques. While dipping involved an
exposure of PVC to a desired solution for 10 min, spin-coating was accomplished
using Laurell WS-200-4NPP/RV spin-coater using various shear rates from 5.9 to
74.8 s'1.
The S0 3 '(NH4 +) groups were attached by dipping or spin-coating aqueous
solutions of (NH4 )2 S2 0 g (0.1-0.8 g/ml) at 80 °C, followed by surface oxidation
reactions with H3 B 0 3 (0.7-5.9 g/ml).
Aqueous solutions of
PEI (H2N-(CH2 CH2 NH)n-H) (0.005-0.16 g/ml) were
stabilized on the sulfonated PVC through crosslinking reactions with crotonaldehyde
(CH3 CH=CHCHO) at 80 °C. Solution pH was adjusted from
6
to 11 using
4M NaOH and 2M HC1, and crotonaldehyde concentration was changed from
0.0008 to 0.0117 mol/1 at pH 10.
DS (H-(0CH 2 C4 H 3(S0 3')(H 3 0 +)(0H)2 0CH)ri-H) was applied to the PEI-coated
surface at 80 °C. The pH of the solution was varied from 2 to 7 using 4M NaOH and
2M HC1, and DS concentration was varied from 0.005 to 0.16 g/ml. Thereafter,
alternating layers o f crosslinked PEI and DS, followed by a top layer of
uncrosslinked PEI, were reacted at 80 °C. After each step, PVC films were washed
with distilled water and dried in a vacuum deccicator for 24 hrs.
HP (-0 CH2C4 H4 (0 H)20 (S0 3')(H30 +)CH0 ) was immobilized on crosslinked and
uncrosslinked PEI surfaces in the presence of 1.5 M NaCl and 0.39 M sodium
cyanoborohydride (NaCNBH3) at 80 °C. Solution pH was controlled from 2 to 7
177
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
using 4M NaOH and 2M HC1, and the HP concentration was varied from 0.0025 to
0 .1 2
g/ml.
8.2.3 Spectroscopic Measurements
ATR FT-ER spectra were collected on a Nicolet Magna-IR single beam
spectrometer. A resolution of 4 cm' 1 and a mirror speed of 0.3165 cm s' 1 were used.
The ATR cell was aligned at 45° angle of incidence using a 45° angle parallelogram
KRS-5 crystal. To determine orientation of the surface species, 90° (TE) and 0°
(TM) polarized infrared light were used. TE is a transverse electric vector of the
incidence light polarized at 90° with respect to sample surface, whereas TM is a
transverse magnetic vector polarized at 0° with respect to sample surface. Each
spectrum represents 2 0 0 coaded scans ratioed against a reference spectrum obtained
from 200 coaded scans of an empty ATR cell. All spectra were corrected for spectral
distortion using Q-ATR™ software. 13'* 4
To determine the extinction coefficient of hydroxyl groups, various concentration
standards of PEI, DS, and NAD heparin were prepared, and transmission spectra
were obtained. Transmission spectra were collected on a Nicolet Magna-IR single
beam spectrometer at a resolution of 4 cm' 1 and with a mirror speed 0.3165 cm s '1.
Quantitative analysis was performed using Q-DEPTH™ software obtained from
Quan-Spec.
178
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8.3 Results and Discussions
As stated in the Introduction, the objective of these studies was to produce
thrombresistant surfaces by attaching thin films o f PEI, DS, and HP to the PVC
surface. Two methods o f depositing multi-layered structures were explored: solution
exposure (dipping) and spin-coating. Because of the distinctly different nature of
both processes, they will be discussed in separate sections.
8.3.1 Multi-layered
Thin
Films
Obtained
by
Direct
Solution
Exposure (Dipping)
According to a schematic diagram shown in Figure 8 . IB, the first layer is attached
by sulfonating the PVC surface with (N H ^ S ^ g , followed by oxidizing it with
H3BO3. Figure 8.2 illustrates ATR FT-IR spectra of sulfonated PVC surface in the
S-0 stretching region. While trace A is the spectrum of unreacted PVC surface,
trace B represents the spectrum of sulfonated PVC surface. It appears that the new
band at 1023 cm' 1 is detected. As shown in Table 8.1, this band is attributed to the
S-0 symmetric stretching modes, 15 resulting from the formation of the S 0 3 *(NH4+)
groups on the PVC surface. On the other hand, an increase of the band at 1254 cm' 1
attributed to the S-0 asymmetric stretching mode15 of the SO3 XNH4 O groups is
detected. When concentration of the (NH4 )2 S2 0 g solution increases from 0.1 (trace
B) to 0.8 g/ml (trace E), intensities of the S-O stretching bands detected at 1023 cm' 1
and 1254 cm' 1 also increase. These observations indicate that the formation of the
S0 3'(NH4+) groups introduced on the PVC surface is a function of the (NH4 )2 S2 0 g
solution concentration.
179
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
«
in
03
CM
CO
o
o
1250
1200
1150
1100 1050
Wavenumbers
1000
950
900
Figure 8.2. ATR FT-IR spectra in the S-0 stretching region for ammonium
persulfate reacted to PVC surface: A - Ammonium persulfate solution concentration
0.05 g/ml; B - Ammonium persulfate solution concentration 0.1 g/ml; C Ammonium persulfate solution concentration 0.2 g/ml; D - Ammonium persulfate
solution concentration 0.4 g/ml; and E - Ammonium persulfate solution
concentration 0 . 8 g/ml.
180
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 8.1. IR bands and their tentative assignments.
Functional Groups
Sulfonated PVC
—
+
- s o 3 (NH4 )
_.
_
IR Band
cm' *
Assignment
1023
1254
S - 0 symmetric stretching
S - O asymmetric stretching
1650
1560
N - H deformation
C = N stretching
1642
O - H bending
- s o 3 (H3 o+)
1623
O - H bending
-N H SO j(H 3 0+)
3358
O - H stretching
-CH2 S 0 3” (H 3 0+)
3422
0 - H stretching
PEI
-CH2 CH2 NH-CH = N DS
-SO 3 (H3 0+)
HP
181
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The next layer to be reacted is PEI (Figure 8 .IB), which will form a crosslinked
surface network. Figure 8.3 illustrates ATR FT-IR spectra in the N-H deformation
region of PEI reacted to the PVC surface. Trace A illustrates the spectrum of the
first layer of PEI crosslinked to the previously sulfonated PVC surface. It appears
that new bands are detected at 1650 and 1560 cm' 1 and are attributed to the N-H
deformation modes of PEI and C=N stretching modes o f crosslinked PEI, 16
respectively. As the PEI solution concentration increases from 0.005 (trace A) to
0.04 g/ml (trace D), the intensities of both bands increase, indicating that PEI is
attached to the sulfonated PVC surfaces and crosslinked through the C=C opening
of crotonaldehyde to form C=N bonds.
As shown in Figure 8 . IB, after reacting the DS layer, the second layer of PEI is
crosslinked to the DS surface. Trace E represents the spectrum o f PEI crosslinked to
the DS surface. When 0.04 g/ml o f PEI solution concentration is employed, the
intensities of both bands detected at 1650 and 1560 cm' 1 increase, as compared to
the PEI crosslinked to the previously sulfonated PVC surface, indicating that DS
molecules provide a good source for increasing effective concentration of the
SCV ^O *) groups to react with PEI.
After depositing two alternating layers of crosslinked PEI and DS on the
sulfonated PVC surface, the top layer of uncrosslinked PEI is added to increase the
amount of terminal NH2 groups on the surface. Trace F represents the spectrum of
uncrosslinked PEI reacted to the DS surface. The intensity of the N-H deformation
band detected at 1650 cm ' 1 decreases, and the C=N stretching band detected at
1560 cm' 1 disappears due to a lack of the PEI crosslinking reactions. This
182
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
in
eo
£
03
S
*v•>
•H
fl
05
1700
1650
1600
1550
Wavenumbers
1500
1450
Figure 8.3. ATR FT-IR spectra in the N-H stretching region for PEI reacted to PVC
surface using direct solution exposure (dipping): A - First layer of PEI solution
concentration 0.005g/ml; B - First layer of PEI solution concentration 0.01 g/ml; C First layer of PEI solution concentration 0.02 g/ml; D - First layer of PEI solution
concentration 0.04 g/ml; E - Second layer of PEI solution concentration 0.04 g/ml;
and F - Third layer of PEI solution concentration 0.04 g/ml.
183
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
is desirable because the primary reason for the formation of uncrosslinked PEI
layers on the outermost surface is to provide the high content of NH2 species to
enhance HP immobilization.
As shown in Figure 8 .IB, the next layer attached to the crosslinked PEI surface is
DS. Figure 8.4 illustrates ATR FT-IR spectra in the O-H bending region for DS
reacted to crosslinked PEI surfaces. The new band developed at 1642 cm' 1 is
attributed to the O-H bending modes of DS-SOs'f^O*) groups. 17 As DS solution
concentration increases from 0.005 (trace A) to 0.08 g/ml (trace E), the 1642 cm' 1
band intensity also increases. These data indicate that DS is attached to the
crosslinked PEI surface and the thickness of this layer depends on the DS solution
concentration. The spectrum o f the second PEI layer exhibits the significantly higher
intensity of the N-H deformation band, indicating that DS is reacted to the second
PEI layer. Trace F represents the spectrum of DS reacted to the second layer of PEI
surface. A comparison of the spectra of the second layer with the first DS layer
indicates that the band intensity at 1642 cm' 1 due to O-H bending modes increases.
The depth penetration of DR. light reaches about 0.8 pm from the top surface into the
DS layer. In this range, the 1650 cm' 1 band due to the second layer of PEI was not
detected. Therefore, trace F of Figure 8.4, the spectrum of the second layer of DS, is
not affected by the first layer o f DS. These observations indicates that the effective
concentration of the SOsXHaO*) groups of the second DS layer is increased, thus
providing more reactive sites for PEI.
The next step involves reactions of HP to the modified PVC surface. Figure 8.5
illustrates ATR FT-IR spectra in the O-H bending region for HP deposited on the
184
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CM
CO
CO
CM
F
E
D
C
B
A
1800
1700
1600
1500
1400
Wavenumbers
Figure 8.4. ATR FT-IR spectra in the O-H bending region for DS reacted to PVC
surface using direct solution exposure (dipping): A - First layer of DS solution
concentration 0.005 g/ml; B - First layer of DS solution concentration 0.01 g/ml; C First layer o f DS solution concentration 0.02 g/ml; D - First layer of DS solution
concentration 0.04 g/ml; E - First layer o f DS solution concentration 0.08 g/ml; and
F - Second layer of DS solution concentration 0.08 g/ml.
185
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
w
CM
CO
CO
CO
•H
00
a
4)
a
M
m
00
CO
■*-)
OS
t-H
1700
1650
1600
Wavenumbers
1550
1500
Figure 8.5. ATR FT-ER spectra in the O-H bending region for HP reacted to PVC
surface using direct solution exposure (dipping): A - HP solution concentration
0.0025 g/ml; B - HP solution concentration 0.005 g/ml; C - HP solution
concentration 0.01 g/ml; D - HP solution concentration 0.03 g/ml; E - HP solution
concentration 0.06 g/ml; and F - HP solution concentration 0.12 g/ml.
186
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
previously modified PVC surface. A new band detected at 1623 cm' 1 is attributed to
the O-H bending modes of HP-SCV^O*) groups. As the HP solution concentration
increases from 0.0025 (trace A) to 0.12 g/ml (trace F), intensity of the band at
1623 cm' 1 also increases. Furthermore, as the HP solution concentration increases,
another band develops at 1650 cm'1, which is attributed to the N-H deformation
modes. 16 HP contains SCV^sO*) and terminal CHO groups. Thus, both groups can
react with PEI: one reaction would involve HP ionic bonding through the formation
of ionic complexes between PEI-NH3 +(OIf) and HP-S0 3 '(H 3 0 +) groups, and the
other is covalent bonding through the formation of the -CH2 -NH- linkages between
PEI-NH2 and HP-CHO groups. Therefore, the presence of the N-H deformation
band detected at 1650 cm' 1 indicates that HP becomes covalently bonded to the PEI
modified PVC surface through the formation of the -CH2 -NH-linkages. On the other
hand, the band detected at 1673 cm' 1 is attributed to the N-H deformation modes of
unreacted PEI-NH3 +(OH') species. If the ionic complex was formed between
PEI-NH3+(OH') and HP-SOsXHjO*) groups, one would anticipate the presence of
the 1661 cm' 1 band attributed to the N-H deformation mode of -NH3 +S03'- ionic
complexes. The proposed reaction mechanisms responsible for HP immobilization
are depicted in Figure 8 .6 . It should be kept in mind that the data were obtained for
specimens for which every layer was deposited by direct solution exposure
(dipping).
187
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CH2° h
diazotation
CHO
RO
RO
1650 cm"l
(NH deformation)
I
^2
NH
^
(OH")
NH3 +
l+ ■—
I
CH2
NH
-L--| pE j
] DS
□ PEI
Z J DS
ZZU PEI
PVC
Figure 8 .6 . Proposed reaction mechanism of HP immobilization using direct solution
exposure (dipping).
188
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8.3.2 Quantitative Analysis
Although the objective of many surface studies is determining a chemical makeup
of surfaces, an ultimate challenge is to be able to quantify newly formed species. For
this reason, intensities of the bands responsible for the surface reactions were
measured and used to quantify each reacted layer. While the C=N stretching at
1560 cm' 1 and the N-H deformation at 1650 cm' 1 bands will be used as a measure of
the crosslinked and uncrosslinked PEI reactions, the O-H bending bands of
DS-S03* (H3 0*) and HP-S0 3 '(H 3 0 +) at 1642 cm' 1 and 1623 cm' 1 will be used to
determine the extent of DS and HP reactions on the PVC surface. Since quantitative
infrared analysis requires knowledge of extinction coefficients of these bands, 13
Table 8.2 lists extinction coefficients for 1650, 1560, 1642, and 1623 cm' 1 bands
obtained by measuring known concentrations of reactive species. Using these
values, Beer-Lambert law ((3 = e c, where e is the extinction coefficient, c is the
concentration, and P is the linear absorptivity) allows us to determine known
concentrations in a transition mode of detection.
However, to accomplish surface quantitative analysis, it is necessary to correct
ATR spectra for optical effects, 13' 14 and utilize proper quantitative algorithm. For
that reason, absorption index spectrum is refined by an iterative process that
minimizes the difference between the true and calculated reflectivity resulting from
optical effects. 13' 14 At the same time, the Kramers-Kronig relationship between
absorption (k) and refractive (n) indices is maintained. 13' 14 This iterative process is
used in conjunction with numerical double Kramers-Kronig transformation (KKT)
method to obtain ATR spectra suitable for quantitative analysis. Using corrected
189
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 8.2. Extinction coefficients of IR characteristic band for PEI, DS and
HP.
Species
Extinction Coefficient
S (cm2/g)
ERBand
crrr1
Crosslinked PEI
1560
621.12
Uncrosslinked PEI
1650
907.98
Dextran Sulfate
1642
623.54
Heparin
1623
1611.45
190
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ATR spectra, linear absorbtivity is obtained, thus allowing calculations of surface
concentrations from the Beer-Lambert law. Further details involved in the
applications and use of this algorithm which accounts for distortions of strong and
weak bands as well as quantitative ATR measurements can be found in the
literature. 13' 1 4 ,I8 ' 19
Using this methodology, Figure 8.7a was constructed and illustrates surface
concentration changes of crosslinked PEI on sulfonated PVC as a function of pH. It
appears that for the pH values ranging from 6 to 8 , the surface concentration of PEI
is 2.80x10*4 mg/m2. As the pH changes from 8 to 10, the PEI concentration increases
linearly from 2.80xl0'4 to 0.96xl0'3 mg/m2 to reach a steady state at
0.94x10° mg/m2 when pH > 10. These observations indicate that higher PEI
concentrations are obtained under basic conditions.
Figure 8.7b illustrates surface concentration changes of crosslinked PEI on the
sulfonated PVC surface plotted as a function of crotonaldehyde concentrations. As
the solution concentration of crotonaldehyde changes from 0.08x1 O' 2 to
0.48x1 O' 2 mol/1, crosslinked PEI surface concentration increases from 2.73x1 O'4 to
1.4 lxlO' 3 mg/m2. However, when higher concentration levels of crotonaldehyde
above 0.48x1 O' 2 mol/1 are employed, PEI crosslinks in the solution and precipitates
out, thus preventing further surface reactions with PEI. For this reason, crosslinked
PEI surface concentration decreases above 0.48x1 O' 2 mol/1 of crotonaldehyde.
Figure 8.7c illustrates surface concentrations of crosslinked and uncrosslinked PEI
on the sulfonated PVC surface plotted as a function of the PEI solution
concentration. Curve A represents surface concentration changes of the first
191
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0.13
o>
E,
0.10
LU
Q. ^
o CVJ<
. LU
O
0.08
•* -
c
o
o
©
O
as
0.05
0.03
k_
3
CO
0.00
5.00
6.40
7.80
9.20
10.60
12.00
pH
Figure 8.7a. Surface concentration changes of PEI on the PVC surface as a function
of pH obtained by direct solution exposure (dipping).
192
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 .2 0
03
E,
0.16
LJJ
QL ^
0.12
O •
. LU
O
c
o
o
<D
O
ffl
0.08
0.04
l.
o
CO
0.00
0.00
0.03
0.05
0.08
0.10
C rotonaldehyde conc. (mol/i)
0.13
(E-1)
Figure 8.7b. Surface concentration changes of PEI on the PVC surface as a function
of crosslinker content obtained by direct solution exposure (dipping).
193
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 .2 1
0.17
m
* ^
(*-> cs
o
1
. w
c
o
o
u
o
03
C4-H
bn
3
CO
0.13
0.08
0.04
0.00
0.00
0.08
0.04
0.11
0.15
0.19
PEI conc. ( g/m l )
Figure 8.7c. Surface concentration changes of PEI on the PVC surface as a function
of solution concentration obtained by direct solution exposure (dipping): A - First
layer; B - Second layer; and C - Third layer.
194
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
crosslinked PEI layer on sulfonated PVC surface, such as shown in Figure 8 .IB. As
the PEI solution concentration changes from 0.005 to 0.04 g/ml, the surface
concentration increases from 8.56xl0'4 to 1.47xl0' 3 mg/m2. In the second layer, the
surface concentration of PEI changes from 8.57X10"4 to 1.65xl0*3 mg/m2. This is
shown by curve B of Figure 8.7c. The first layer o f PEI is crosslinked to the
sulfonated PVC surface, but the second layer is crosslinked to the DS surface.
Therefore, higher PEI concentrations of the second layer are attributed to the fact
that DS provides higher content of the SC V ^O *) groups on the surface, thus
providing more opportunities to react with PEL When PEI solution concentration is
above 0.04 g/ml, surface concentrations of both layers remain unchanged. It should
be noted that the formation of the third PEI-DS layer was accomplished using
uncrosslinked PEI in order to provide higher NH2 concentrations accessible for the
bonding with HP, thus enhancing immobilization of HP molecules.
Surface concentration changes of PEI in the third layer was quantified using
extinction coefficient obtained for the N-H deformation band at 1650 cm'1. As
shown in Figure 8.7c, curve C, concentration levels range from 2.63x1 O'4 to
4.89x1 O'4 mg/m2, when PEI concentration in the solution changes from 0.005 to
0.04 g/ml. Compared to the crosslinked PEI layer, lower surface concentration was
obtained because uncrosslinked PEI layer is water soluble. With these results in
mind, let us consider how the pH of the solution will affect the surface reactions of
DS.
To form a charged surface with NH3 +(OIT) group for further reactions, the surface
should be treated with acidic solution. For that reason, acidic solution (pH < 6 ) was
195
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
employed. Figure 8 .8 a illustrates surface concentration changes of the first DS layer
on PEI plotted as a function o f pH. As seen, the DS surface concentration is similar
to the surface concentration of PEI, but when pH changes from 2 to 4, surface
concentration of DS increases from 7.51xl0'5 to 1.77x1 O'4 mg/m2. The pH changes
from 4 to
6
result in a linear increase, from 1.77x1 O'4 to 2.07xl0‘3 mg/m2. Like PEI
surface concentration, DS surface concentration levels off at 2.24x1 O'3 mg/m2, when
pH > 6 . These observations indicate that higher surface concentrations of DS can be
obtained under mild acidic conditions (4 > pH > 6 ) because the PEI surface exhibits
positive NH3 +(OH') groups under acidic conditions. On the other hand, when
pH < 4, lower surface concentrations o f DS are obtained because most of the
DS-S0 3‘(H3 0 +) associations lose their ionic strength to form DS-SO3H associations.
The effect o f DS solution concentration on surface reactions is plotted in Figure
. b. Curve A illustrates the surface concentrations o f the first layer and ranges from
8 8
1.24xl0‘3 to 6.20xl0 ' 3 mg/m2, when DS solution concentration ranged from 0.005 to
0.08 g/ml, respectively. On the other hand, the concentration of the second layer
increases from 1.28xl0'3 to 7.18xl0'3 mg/m2. This is shown in curve B, Figure 8 .8 b.
The second PEI layer exhibits higher surface concentration levels than the first PEI
layer. Higher surface concentration levels of the second DS layer result from the
increased surface concentration of the second PEI layer, which will provide more
opportunities to react with DS-SCVCHsO*) groups. On the other hand, surface
concentrations of both DS layers level off above 0.08 g/ml, indicating that PEI film
no longer supplies positive NH3 +(OH*) groups for further reactions with DS.
The primary objective of these studies was to make PVC surface thrombresistant
196
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 .3 0
bo
0.24
B
co
Q
CN
i
o
c
o
o
V
o
Ix
0.18
ffl
0.12
0.06
9
CO
0.00
0.00
1.60
3.20
4.80
6.40
8.00
pH
Figure 8 .8 a. Surface concentration changes of DS on the PVC surface as a function
of pH obtained by direct solution exposure (dipping).
197
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.00
00
a
0.80
CO
Q ^
O i
o. w'
G
O
o
<u
o
A
C
4-1
0.60
0.40
0.20
«-l
G
CO
0.00
0.00
0.04
0.07
0.11
0.14
0.18
DS concentration (g/ml)
Figure 8 .8 b. Surface concentration changes of DS on the PVC surface as a function
of solution concentration obtained by direct solution exposure (dipping): A - First
layer; and B - Second layer.
198
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
by attaching HP to PVC. Figure 8.9a illustrates surface concentration changes of HP
immobilized on multilayer modified PVC surface (Figure 8 . IB) plotted as a function
of pH. There are two reactive sites for HP that can potentially react with the
previously created PEI surface: HP-CHO and HP-S0 3 *(H30 +) entities. Multi-layered
coatings produced by direct exposure to respective solutions show that HP is
immobilized through the formation of the -CH2 -NH- linkages between HP-CHO and
PEI-NH2 groups. This was demonstrated by the formation of the 1650 cm' 1 band due
to N-H deformation modes of -CH2 -NH - linkages, and the reactions responsible for
the covalent bond formation are illustrated in Figure 8 .6 . When pH changes from 7
to 5, the surface concentration of HP immobilized on the PEI surface remains
constant. On the other hand, as the pH values decrease from 5 to 2, surface
concentration of HP increases from l.lOxlO'4 to 1.87xl0‘3 mg/m2. These
observations indicate that HP-CHO is activated when acidic conditions below pH 5
are utilized, resulting in immobilization on the outermost PEI layer.
The effect of the HP solution concentration on surface reactions is plotted in
Figure 8.9b. Curve A illustrates the surface concentration of HP immobilized on
crosslinked PEI surface, which ranges from 2.06xl0'3 to 3.10xl0 ' 3 mg/m2, for the
HP solution concentration changes ranging from 0.0025 to 0.12 g/ml. On the other
hand, surface concentration of HP immobilized on the uncrosslinked PEI surface
exhibits higher yield, and increases from 6.63xl0‘3 to 1.58xl0'2 mg/m2, compared to
its concentration level on crosslinked PEI surface. This is shown in Figure 8.9b,
curve B and indicates that, although surface concentrations of uncrosslinked PEI are
lower compared to crosslinked PEI (Figure 8.7), higher content of the NH2 groups
199
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Surface conc. of heparin (mg/m2)
0.20
0.16
0.12
0.08
0.04
0.00
1
2
3
4
5
6
7
8
pH
Figure 8.9a. Surface concentration changes of HP on the PVC surface as a function
of pH obtained by direct solution exposure (dipping).
200
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 .2 3
00
S
0.18
CQ
(X
X 7
<*-. w
O '
o
c
o
o
<u
o
CQ
«4«-i
-<
3
on
0.14
0.09
0.05
« A
0.00
0.00
0.03
0.06
0.08
0.11
0 .1 4
Heparin concentration (g/ml)
Figure 8.9b. Surface concentration changes of HP on the PVC surface as a function
of solution concentration obtained by direct solution exposure (dipping): A - HP
immobilized on the crosslinked PEI surface; and B - HP immobilized on the
uncrosslinked PEI surface.
201
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
exists on the uncrosslinked PEI surface, resulting in higher reaction yields o f HP.
8.4 Multi-layered Thin Films Obtained by Spin-coating
Although one perhaps would anticipate concentration level differences between
direct solution exposure (dipping) and spin-coating application methods of
individual layers reacted to PVC, it would be rather surprising to detect differences
in structural features of thin films. It appears that there are substantial changes in the
reactions of HP, when spin-coating is elected as an application method. In the series
of experiments, thin films were spin-coated using the same concentrations and
conditions as were utilized in the direct solution exposure experiments.
When spin-coating was used to apply PEI and DS, no significant spectral
differences were detected between both application methods, indicating that there
are no structural differences among the films. This situation changed, however,
when HP was spin-coated. Figure 8.10 illustrates ATR FT-IR spectra in the O-H
bending region for HP applied by spin-coating. Trace A shows the spectrum of the
surface HP spin-coated obtained using 0.005 g/ml HP solution. It appears that a new
band is detected at 1623 cm*1 which is attributed to the O-H bending modes of the
HP-SOsTHsO*) entities. As HP solution concentration was increased from
0.005 (trace A) to 0.10 g/ml (trace E), the band intensity also increased. In the
results of the dipping experiments (Figure 8.5), when HP solution concentration
increased, the bands at 1650 cm*1 and 1673 cm' 1 were detected. These bands were
attributed to the N-H deformation modes of the -CH2 -NH- linkages and unreacted
PEI-NH3 +(OH*) ionic complexes, respectively. However, for spin-coated HP, even
202
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
n
<M
CO
CO
CO
>>»
—
CO
a
V
C
1700
1650
1600
1550
1500
Wa v e n u m b e r s
Figure 8.10. ATR FT-IR spectra in the O-H bending region for HP reacted to the
PVC surface using spin-coating: A - HP solution concentration 0.005 g/ml; B - HP
solution concentration 0.01 g/ml; C - HP solution concentration 0.03 g/ml; D - HP
solution concentration 0.05 g/ml; and E - HP solution concentration 0.10 g/ml.
203
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
though the HP concentration in the solution was increased, the bands due to
formation of -CH2 -NH- linkages and unreacted PEI-NH3+ (OH-) ionic complexes
were not present. On the other hand, as shown in Figure 8.10, when HP is
spin-coated, a new band at 1661 cm*1 attributed to the N-H deformation modes of
PEI-NH3 +S03*-HP ionic linkages is present. Although one could anticipate a loss of
IR activity due to symmetry changes resulting from different application methods,
this is not the case. These observations indicate that there are significant differences
in bonding characteristics of the HP molecules when dipping and spinning are
employed.
The primary difference between dipping and spin-coating are shear rates and the
accessibility of HP molecules for reactions. Therefore, to identify how shear rates
may affect HP reactions, HP solution was spin-coated using various shear rates.
Figure 8.11 illustrates ATR FT-IR spectra in the O-H and N-H stretching region for
HP on the PVC surface. For reference purposes, trace A represents the spectrum of
HP immobilized using dipping. The band detected at 3571 cm*1 band is attributed to
the O-H stretching modes of HP-SO^CH^O - 14 As seen, for spin-coated films, the
3571 cm*1 band disappears when high shear rates are applied (traces D and E). On
the other hand, a broad band is detected around 2687 cm*1 and increases when shear
rates are changed from 23.8 (trace B) to 74.8 s*1 (trace E), indicating that the HP
immobilization occurs through the formation of ionic linkages between
PEI-NH3 +(OH*) and HP-S0 3*(H3 0 +) . 4 When high shear rates are employed, the band
at 3462 cm*1 develops (traces D and E) and is attributed to the N-H stretching modes
of newly formed -NH3 +S 03'- ionic linkages.
204
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
eo
co
CM
00
CM
CO
eo
co
Tl*
CM
CO
CM
3500
3000
2500
2000
Wa v e n u m b e r s
Figure 8.11. ATR FT-IR spectra in the O-H stretching region for HP reacted to the
PVC surface for various shear rates: A - HP deposited at 0 s' 1 shear rate; B - HP
deposited at 23.8 s' 1 shear rate; C - HP deposited at 47.6 s' 1 shear rate; D - HP
deposited at 60.2 s’ 1 shear rate; and E - HP deposited at 74.8 s' 1 shear rate.
205
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Although these observations clearly point toward HP immobilization through the
formation of ionic linkages between PEI-NH3 +(OH') and HP-SCV^C)*) groups
when spin-coating is applied, if this is the case, the N-H deformation region should
also confirm these findings. Figure 8.12 illustrates ATR FT-IR spectra in the N-H
deformation region for HP immobilized to the pre-treated PVC surface obtained by
spin-coating under various shear rate conditions. For reference purposes, trace A is
the spectrum of HP immobilized by dipping and exhibits the band detected at
1650 cm' 1 attributed to the N-H deformation modes resulting from the -CH2-NHformation. On the other hand, the 1661 cm*1 band appears when higher shear rates
are employed (traces E and F). This band is attributed to the N-H deformation modes
resulting from the formation o f PEI-NH3+S03‘- HP ionic linkages. When shear rates
change from 5.9 to 74.8 s'1, the 1661 cm' 1 band increases, while the 1650 cm' 1 band
significantly decreases. As shown in Figure 8.13, absorptivity of the 1650 cm' 1 band
decreases from 1.30x10° to 0.06x10° cm' 1 when the shear rate changes from
0 (dipping) to 74.8 s '1. These observations indicate that HP immobilization is the
shear rate dependent process that at higher shear rates, the formation of ionic
linkages occurs. Thus, using different shear rates, it is possible to produce covalently
or ionically bonded HP molecules to the PVC surface.
8.4.1 Orientation of HP Thin Films
Having identified that it is possible to obtain ionic or covalent bonding of HP
molecules by changing deposition shear rate, let us now consider orientation of HP
molecules. It is well-known that the band intensities in IR are also affected by
206
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CO
CV2
CO
to
CO
CO
CO
£
•H
03
G
u
-4<
->
G
OS
1700
1680
1660
1640
1620
1600
1580
Wa v e n u m b e r s
Figure 8.12. ATR FT-ER spectra in the O-H bending region for HP reacted to the
PVC surface for various shear rates: A - HP deposited at 0 s*1 shear rate; B - HP
deposited at 5.9 s' 1 shear rate; C - HP deposited at 23.8 s*1 shear rate; D - HP
deposited at 47.6 s' 1 shear rate; E - HP deposited at 60.2 s' 1 shear rate; and F - HP
deposited at 74.8 s' 1 shear rate.
207
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0.15
E
0. 12
> ^
0.09
Q .^
5 u
CO
.Q
0.06
o
CO
h.
CO
CD
0.03
0.00
0
16
32
48
64
80
S h ear rate (s '1)
Figure 8.13. Linear absorptivity of N-H deformation band for HP reacted to the PVC
surface for various shear rates.
208
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
orientation o f the electric vector responsible for a given vibration. Therefore, using
polarized light, it is possible to estimate a preferential orientation of the surface
species. Figure 8.14 illustrates ATR FT-IR spectra in the O-H bending region of the
HP immobilized on the pre-treated PVC surface obtained by spin-coating and
solution exposure (dipping). While traces A and B illustrate transverse magnetic
(TM) and transverse electric (TE) polarization spectra of specimens obtained by
direct solution exposure (dipping), traces C and D illustrate the same polarization
spectra of specimens produced by spinning. A comparison of the spectra of HP
immobilized by dipping in TM (trace A) and TE (trace B) modes indicates that the
enhanced intensity of the 1623 cm*1 band attributed to the O-H bending modes is
detected for TE polarization. In contrast, for spin-coated specimens, TM polarization
spectrum (trace C) exhibits higher intensity, and the band at 1650 cm*1 is not
detected. On the other hand, the 1661 cm*1 band appears. These observations
indicate that dipping produces O-H groups of HP-SCV^O*) which are
preferentially parallel to the surface, thus suggesting that the HP molecules are likely
perpendicular to the surface. However, for the spin-coated specimens, the O-H
groups of HP-S0 3 *(H30 +) are preferentially perpendicular, indicating that HP
molecules are also preferentially parallel.
Based on these observations, the following structural features of HP orientation
and bonding to the pre-treated PVC surface are proposed. Upon direct solution
exposure (dipping; 0 s*1 shear rate), HP molecules are preferentially perpendicular
resulting from the formation of the -CH2 -NH- linkages between PEI-NH2 and
HP-CHO groups. When spin-coating is employed, increased shear rates cause HP
209
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CO
CO
C
O
CV2
co
GO
a
c
os
V
*->
1700
1680
1660
1640
1620
1600
1580
W av e n u m b e rs
Figure 8.14. ATR FT-IR spectra in the O-H bending region for HP reacted to the
PVC surface: A - TM polarization-direct solution exposure (dipping); B - TE
polarization-direct solution exposure (dipping); C - TM polarization-spin-coating;
and D - TE polarization-spin-coating.
210
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
molecules to be oriented parallel to the surface, and for high shear rates, HP
molecules are attached to the surface through the formation of ionic complexes
between PEI-NH3 +(OH') and HP-SCVOHsO*) groups. The proposed reaction
mechanisms are depicted in Figure 8.15.
8.4.2 Quantitative Analysis
The results of quantitative analysis of spin-coated films on the PVC surface are
illustrated in Figure 8.16a where surface concentration of PEI is plotted as a
function of PEI solution concentration. For reference, curve A represents surface
concentration changes of PEI obtained by dipping. As seen, when PEI solution
concentration changes from 0.005 to 0.1 g/ml, the surface concentration changes
from 2.63xl0'4 to 4.78x1 O'4 mg/m2. For spin-coated PEI, the surface concentration
changes are shown in curve B. In this case, surface concentration levels are
significantly lower, ranging from 5.58xl0‘5 to 2.75X10"4 mg/m2.
The surface concentration changes of DS are shown in Figure 8.16b. Curve A
illustrates the surface concentration o f DS obtained by dipping, which ranges from
3.03x1c4 to 7.18x10'3 mg/m2, when solution concentration varies from 0.005 to
0.1 g/ml. For spin-coated specimens, DS surface concentration ranges from
4.52xl0'5 to 4.70x1 O'4 mg/m2 (curve B). Again, it is lower when compared to the
thin films obtained by direct solution exposure.
The effect of HP solution concentration on the HP surface concentration is shown
in Figure 8.16c. Curve A illustrates surface concentration changes of HP
immobilized on the PEI surface obtained by dipping and ranges from 6.90x1 O' 3 to
211
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A. HP Immobilization through Dipping
CHO
CHO
CHO
S 03 (H3 ° )
PEI - N H 2
fS o j(H 3o +)
_ (H 30 +)
L .s o J (h 3o +)
V s o 3 (H30 +)
s o 3 (H3o +)
(H30 +)
Randomly Oriented HP
PVC
B. HP Immobilization through Spin-coating
jh
3o + )
OHC-^r-T T ^ ^?°L ---^C H O
so3
(H30 +)
so3
PEI - N H 2
+ so3
(H3° ! ! _ j °3- <H3 ° +>
■ _
CHO
S 0 3 (H30 +)
nh3
so3
n h 3+
(H30 +)
SOf
CHO
^ CHO
ijfHa
PVC
Parallel Oriented HP
Figure 8.15. Proposed reaction mechanism of HP immobilization using spin-coating.
212
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0.70
00
0.56
s
• A
W
(4 -1
o
.
0.42
m
1
w
o w
a
o
o
<u
o
A
vm
f-H
3
CO
0.28
*B
0.14
0.00
0.00
0.02
0.04
0.07
0.09
0. 11
PEI concentration (g/ml)
Figure 8.16a. Surface concentration changes of PEI on the PVC surface as a
function of solution concentration obtained by spin-coating: A - PEI layer obtained
by direct solution exposure (dipping); and B - PEI layer obtained by spin-coating.
213
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 .0 0
00
0.8 0
£
'w '
GO
Q
^
o i
O. fwf l
c
oo
u
u
0 .6 0
0.4 0
0.20
<4-1
(l
3
CO
A B
0.00
0.00
0.02
0.04
0.07
0.09
0.11
DS concentration (g/ml)
Figure 8.16b. Surface concentration changes of DS on the PVC surface as a function
of solution concentration obtained by spin-coating: A - DS layer obtained by direct
solution exposure (dipping); and B - DS layer obtained by spin-coating.
214
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 .2 3
60
s
0.18
A
a4) /•—\
0.14
X 7
o
6
w
w
0.09
c
o
o
D
O
CO
«4M-l
s
W3
0.05
0.00
0.00
■A B
0.02
0.04
0.07
0.09
0.11
Heparin concentration (g/ml)
Figure 8.16c. Surface concentration changes of HP on the PVC surface as a function
of solution concentration obtained by spin-coating: A - HP layer obtained by direct
solution exposure (dipping); and B - HP layer obtained by spin-coating.
215
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.59xl0*2 mg/m2. For spin-coated specimens, the concentration levels are much
lower, ranging from 1.26X10"4 to 9.90X10*4 mg/m2 for the same HP solution
concentrations (curve B).
To compare relative thicknesses of respective thin films, Table 8.3 was generated
which illustrates estimated thickness fraction of each layer. As seen, film thickness
fractions of PEI and DS are similar for both conditions. On the other hand, direct
solution exposure exhibits 2.4 times higher HP film thickness.
8.5 Conclusions
In this study, multi-layered thin films of PEI and DS can be attached to the PVC
surface to immobilize the outer layer of HP. The procedure presented in this study
involves direct solution exposure (dipping) and spin-coating methods. Although
there are no significant differences between PEI and DS thin films deposited using
both methods, when direct solution exposure (dipping) is employed as an application
method for HP molecules, HP rings become preferentially perpendicular to the
surface, resulting in the formation of the -CH2 -NH- covalent linkages between
PEI - NH2 and HP-CHO groups. On the other hand, for spin-coated HP thin films,
HP molecules are preferentially parallel and are immobilized through the formation
of ionic linkages between PEI-NH3 +(OH’) and H P-SCV ^O*) groups. Quantitative
analysis indicates that surface concentration of each layer significantly depends on
pH of the solution. Concentration levels of HP immobilized to the modified PVC are
significantly higher for specimens produced by direct solution exposure.
216
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 8.3. A schematic representation of relative film thickness of multi-layered thin
films.
Film Thickness Fraction
Multi-layered Thin Films
Dipping
Spinning
Crosslinked PEI; first layer
0.08
0 .1 2
Crosslinked PEI; second layer
0 .1 0
0.13
Uncrosslinked PEI; third layer
0.03
0 .1 0
DS; first layer
0.13
0 .2 1
DS; second layer
0.15
0.23
HP
0.51
0 .2 1
217
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8.6 References
1. B. Casu, Adv. Carbohydr. Chem., 1985,43, 51.
2. J. Hoffinan, O. P. Larm, and E. Scholander, Carbohydr. Res., 1983, 117, 328.
3. J. S. Marcum and R. D. Rosenberg, Biochem. Biophys. Res. Common., 1985,
126, 365.
4. I. Bjork and U. Lindahl, Molecular and Cellular Biochemistry, 1982, 48, 161.
5. 0 . P. Larm, R. Larsson and, P. Olsson, Biomat., Med. Dev., Art. Org., 1983, 11,
161.
6
. 0 . P. Larm, L. A. Adolfsson, and K. P. Olson, US Patent 5,049,403,1991.
7. O. P. Larm, US Patent 4,613,665,1986.
8
. C. Golander and R. Larsson, US Patent 4,565,740,1986.
9. O. P. Larm, US Patent 4,810,784,1989.
10. D. J. Fink and R. M. Gendreau, Analytical Biochemistry, 1984, 139, 140.
11. R. M. Gendreau, R. I. Leininger, and S. Winters, Advan. Chem. Ser., 1982, 199,
371.
12. D. J. Fink and R. M. Gendreau, Analytical Biochemistry, 1984, 139, 140.
13. M. W. Urban, Attenuated Total Reflectance Spectroscopy o f Polymers-Theory
and Practice, American Chemical Society, Washington, DC, 1996.
14. M. W. Urban, Vibrational Spectroscopy o f Molecules and Macromolecules on
Surfaces, Wiley-Intersciences, New York, 1993.
15. L. J. Bellamy, The Infrared Spectra o f Complex Molecules, John Wiley & Sons,
New York, 1975.
16. Barcello and Bellanato, Spectrochim. Acta, 1956, 8 , 27.
218
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17. G. Socrates, Infrared Characteristic Group Frequencies, Wiley - Interscience,
New York, 1980.
18. J.B. Huang and M. W. Urban, Appl. Spectrosc., 1992,46(11), 1666.
19. J.B. Huang and M. W. Urban, Appl. Spectrosc., 1993,47(7), 973.
219
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CONCLUDING REMARKS
220
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Although there are various forms of energy available for conducting surface and
interfacial reactions, microwave energy generated plasmas appear to be an effective
source for reacting monomeric molecules to elastomeric surfaces without
destructing bulk polymer properties. These studies illustrated that ATR FT-IR
spectroscopy is a power tool to determine surface structures that develop on the
elastomeric surfaces as a result of plasma reactions.
For the microwave plasma reactions of imidazole discussed in Chapter 2, closed
and open flow microwave plasma reactors were developed and utilized to react
imidazole molecules to crosslinked PDMS surfaces. Using a closed reactor,
imidazole molecules are chemically bonded on PDMS surface through hydrogen
abstraction of the N-H bonds of imidazole. Their orientation, as determined by
polarized ATR FT-IR spectroscopy, appears to be preferentially parallel to PDMS
surface. The amount of imidazole reacted to PDMS surface increases at discharge
pressures not exceeding 53.3 Pa and discharge times not exceeding 20 sec. Extended
discharge times, however, are destructive to chemically attached imidazole
molecules to PDMS. In the open flow reactor, imidazole molecules are reacted to
PDMS surface by ring opening of the imidazole entity to form C=N surface species.
Both reactivity and the extent of the ring opening reactions can be controlled by the
plasma reaction parameters, discharge pressure, and time. The presence of silica
microdomains on PDMS surface inhibits imidazole reactions to PDMS.
While imidazole molecules were reacted to crosslinked PDMS surfaces through
hydrogen abstraction of N-H bonds, the issue of PDMS surface sites responsible for
imidazole bonding reactions remained open. The results and discussions presented
221
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
in Chapter 3 showed that the primary reactive sites o f PDMS surfaces are Si-CH2«
radicals, which are formed through the hydrogen abstraction of the Si-CH3 groups.
The Si-CH2« radicals react with the imidazole CHCHNCHN* radicals resulting from
hydrogen abstraction of the N-H bonds to form Si-CH2-Imidazole entities on PDMS
surfaces.
The issue of surface morphological changes resulting from microwave plasma
reactions of imidazole on PDMS surfaces was discussed in Chapter 4. In this study,
we combined ATR FT-IR analysis and AFM measurement to establish inhibition
mechanism of silica and morphological changes resulting from both closed and open
flow reactor conditions. These studies showed that silica containing PDMS is
composed of PDMS-rich and silica-rich domains. The aggregations of silica
particles in silica-rich domains is responsible for the inhibition of imidazole
reactions on PDMS surfaces. Under a closed reactor condition, a multi-layer of
imidazole rings is formed on PDMS surfaces. On the other hand, in an open flow
reactor condition, the imidazole ring is opened and grafted to form -(CH=CH)nspecies on PDMS surfaces.
Although these studies revealed that under Ar microwave plasma conditions,
reactions of imidazole with PDMS result in the formation of Si-CH2-imidazole
species, Chapter 5 discussed the issue of the discharge gas effect on microwave
plasma reactions. This study examined how imidazole reactions in the presence of
At, 0 2, and C 0 2 gases under microwave plasma conditions will affect surface
reactions on PDMS. When Ar microwave plasma reaction conditions are employed,
imidazole molecules react to the PDMS surface through hydrogen abstraction of the
222
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
N-H bonds to form Si-CH2-imidazoIe and Si-CH2-CH3 linkages. When 0 2
microwave plasma reactions are conducted for
20
sec discharge times or less,
Si-O-imidazole and Si-0-CH 3 species on the PDMS surface are formed. On the
other hand, above
20
sec discharge times, as a result of hydrogen abstraction, the
Si-0-CH 3 linkages are converted to Si-0-CH2-imidazole entities. The C 0 2
microwave plasma reactions in the presence of imidazole vapors result in the
formation of Si-0-imidazole-CH3 species on the PDMS surface, followed by
hydrogen abstraction, resulting in the formation o f Si-0-imidazole-CH2« radicals,
which react with subsequent imidazole molecules through the formation of CH2-N
linkages. Like previous studies, quantitative ATR FT-IR surface analysis showed
that the highest yields of imidazole reactions occur under
0 2
microwave plasma
conditions. All experiments utilized in this study allowed surface analysis at 1.3 pm
from the surface.
Due to desirable chemical and physical properties, PU elastomers and PVC are
often used in countless applications. However, to achieve necessary surface
properties, their surfaces may require modifications. Although we used the same
experimental setup of closed and open flow reactor conditions to obtain imidazole
entities on PU surfaces, we could not detect any newly created species resulting
from microwave plasma reactions of imidazole. Therefore, a new method for
reacting imidazole molecules on PU surfaces was demonstrated in Chapter 6 . To
facilitate
PU-imidazole
reactions,
PU
specimens
were
immersed
in
imidazole-containing methylene chloride solution and exposed to an Ar microwave
plasma. This approach, combined with microwave plasma reactions, appears to be
223
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
successful in creating imidazole species on PU surfaces. These experiments showed
that imidazole molecules can be bonded to polyurethane surfaces through C=N
opening of imidazole entities to create C-O-C linkage between imidazole and
polyurethane. Analysis of imidazole-urethane linkages was accomplished using
ATR FT-IR spectroscopy and revealed the same surface structures on polyurethane
surfaces, but different concentration levels of imidazole at various depths from the
surface can be created under closed and open flow reactor conditions.
Chapter 7 illustrated microwave plasma reaction of imidazole to PVC surfaces. In
this case, we focused on the surface reactivity and morphology of PVC. These
studies showed that the surface reactions on PVC depend heavily upon a prior
thermal history of the PVC substrate. It appears that the plasma reactions on
hot-pressed PVC not only results in the development of the CH2 linkages, but a
significant increase of crystallinity in the hot-pressed PVC inhibits the reactivity of
imidazole to the PVC surface. On the other hand, for a solvent-cast PVC with a
significantly lower surface crystalline phase content, imidazole reacts to the PVC
surface through a C=C bond opening. The amount of imidazole reacted to the PVC
surfaces changes as a function of depth. Using quantitative ATR FT-IR
spectroscopy, imidazole content can be quantified, and its concentrations are in the
I O'6 mole/cm2 range at about 0.8 - 1.2 pm for the PVC surface.
To
obtain
thrombresistant
surface,
multi-layered
thin
films
containing
polyethylene imine (PEI), dextran sulfate (DS), and heparin (HP) were deposited on
PVC surfaces. Direct solution exposure (dipping) and spin-coating are employed to
deposit multi-layers; and using ATR FT-IR spectroscopy, the effects of pH,
224
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
concentration of surface reacting species, and crosslinking conditions on the
formation o f multi-layered structures were investigated in Chapter 8 . Quantitative
ATR FT-IR analysis revealed that, when thin films are deposited by dipping, HP is
immobilized to form covalent -CH2 -NH- linkages between PEI-NH2 , and terminal
HP-CHO groups. However, when spin-coating is utilized as an application method,
HP takes parallel orientation, and ionic complexes between PEI-NH3 +(OH') and
HP-SCVO^O*) are formed. These studies illustrate the relationship between shear
rates of applied HP solution and structures that result from the application method.
Quantitative analysis indicates that surface concentration of individual layers is
significantly lower for spin-coated layers.
225
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Документ
Категория
Без категории
Просмотров
0
Размер файла
7 903 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа