close

Вход

Забыли?

вход по аккаунту

?

Study of Dow Cyclotene surface amination using a downstream microwave plasma

код для вставкиСкачать
STU DY OF DOW CYCLOTENE SURFACE A M IN A TIO N USIN G
A DOW NSTREAM M ICROW AV E PLA SM A
by
Lijiang W ang
A Dissertation Presented in Partial Fulfillm ent
o f the Requirements for the Degree
Doctor o f Philosophy
ARIZONA STATE U NIVERSITY
Decem ber 2005
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 3194983
INFORMATION TO USERS
The quality of this reproduction is dependent upon the quality of the copy
submitted. Broken or indistinct print, colored or poor quality illustrations and
photographs, print bleed-through, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send 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.
®
UMI
UMI Microform 3194983
Copyright 2006 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
STUDY OF D O W CY CLO TEN E SURFACE A M IN A TIO N USIN G
A D O W N STREA M M ICROW AVE PLA SM A
by
Lijiang Wang
has been approved
N ovem ber 2005
APPROVED:
, Chair
//
/ Supervisory Committee
ACCEPTED:
Department Chair
d A l* - ' / r
Dean, Division o f G raduate Studies
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A BSTRA CT
Plasm a surface m odification can im prove biocom patibility and biofunctionality
between the physiological environm ent and the biom aterial surface w ithout changing the
m aterial’s bulk properties. This dissertation is therefore focused on m odifying Dow
Cyclotene surfaces through plasm a treatm ent to improve biocom patibility, and on
determining and correlating changes in physical and chem ical states o f the Cyclotene
surface to plasm a process parameters. The prim ary goals o f this research are to explore
the capabilities and lim itations o f utilizing a dry plasm a one-step technique for am inating
D ow Cyclotene as a function o f four processing param eters (power, pressure, tem perature
and time), and to develop fundamentally-based models describing the surface am ination
behavior to provide insight about the reaction m echanism s o f the plasm a surface
m odification. The effects o f the four processing param eters on the extent o f nitrogen
incorporation into polym er surface were studied. X-ray Photoelectron Spectroscopy
(XPS), Fourier Transform Infra-red Spectroscopy-Attenuated Total R eflection (FTIRATR), and A tom ic Force M icroscopy (AFM) were used to characterize the surface
chem istry and topography structures. A TR and XPS results showed that nitrogencontaining functional groups were introduced onto the polym er surface through am m onia
plasm a treatm ent. The N /C ratio on the surface reached a m axim um o f 0.24 under the
high level intensity plasm a conditions accompanied w ith argon plasm a pretreatments.
Covalent coupling o f oxidized dextran to aminated Cyclotene surface was subsequently
realized, as indicated by the outcomes o f cell adhesion and spreading studies in which
dextran-coated am inated Cyclotene surfaces exhibited significantly reduced cell adhesion
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and spreading com pared w ith untreated Cyclotene surfaces. Cell adhesion reduction
correlated w ith am m onia plasm a treatm ent conditions and N /C ratio. The chem ical
stability o f plasm a m odified surfaces is poor, and surface restructuring typically leads to
an effective decrease over tim e o f the treatm ent effects. Therefore an aging study was
perform ed in three different storage environm ents; the study indicated that the surface
m odification effect, including both N /C ratio and amino selectivity (NH 2 /N), degraded
w ith storage tim e and was dependent on the storage media. Finally, a plasm a chem istry
m odel in the gas phase and an am ination m odel on a surface w ere proposed, and the tw o
descriptive m odels were evaluated using experim ental data.
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
To m y husband, H ua
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGEMENTS
The w ork w ith this dissertation has been extensive and trying, yet exciting and
fun. W ithout help, support, and encouragem ent from the follow ing people, this w ork
w ould never have been finished.
First o f all, I w ould like to thank m y advisor, Dr. G regory Raupp for his
invaluable guidance and comments during the whole w ork w ith this dissertation.
Also I w ould like to thank my dissertation com mittee, Dr. Stephen K rause, Dr.
M ichael Sierks, Dr. Jiping He and Dr. Stephen M assia for review ing this docum ent and
their technical feedbacks.
N ext I am very grateful to Kee-keun Lee for his assistance w ith Cyclotene sam ple
preparations, to G holam Ehteshami for all cell adhesion assays and PBS solutions, to Tim
Karcher, Shawn W haley for their help w ith material characterization and their tolerance
to m y questions, to Jay Schwartz for his assistance w ith A FM analysis. I also w ould like
to acknowledge financial support from D ARPA Bio-Info-M icro program , Grant #
M D A 972-00-1-0027.
Last but not least, I owe a debt o f gratitude to m y parents for their unconditional
support all the w ay through the entire trying process. And more im portantly, to m y
husband Hua: w ithout your love, support and encouragement, I w ould never have done
this!
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
Page
LIST OF T A B L E S ................................................................................................................................x
LIST OF F IG U R E S ............................................................................................................................ xi
CH A PTER 1
IN T R O D U C T IO N ................................................................................................. 1
CH A PTER 2
LITERATURE R E V IE W ..................................................................................14
2.1
Introduction.................................................................................................................. 14
2.2
Overview o f Process Chem istries and Associated Process Param eters
for Different P olym ers...............................................................................................16
2.3
D iscussion o f Process V ariable Effects on Surface M o d ificatio n .................. 18
2.3.1
Power effects................................................................................................. 18
2.3.2
Gas pressure effects........................................................................................ 21
2.3.3
Treatment time effects.................................................................................... 23
2.3.4
Temperature effects.........................................................................................25
2.4
Characterization o f M odified Polym er Surfaces................................................ 26
2.5
Possible Reactions in Plasm a Surface M odification......................................... 27
2.5.1
Dissociation of ammonia in the plasma gas phase.......................................28
2.5.2
Reaction on the polymer surface during plasma treatment.......................... 31
2.6
Im m obilization R eaction S trateg y ......................................................................... 41
2.7
Aging Effect S tu d y ....................................................................................................42
CH A PTER 3
EXPERIM ENTAL M ETH O D S.......................................................................48
3.1
O verview ......................................................................................................................48
3.2
Plasm a S ystem ........................................................................................................... 49
vii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Page
3.3
Sam ple P reparation................................................................................................... 52
3.4
C haracterization Techniques................................................................................... 53
3.4.1
XPS.................................................................................................................. 54
3.4.2
FTIR-ATR.......................................................................................................54
3.4.3
A F M ................................................................................................................ 55
3.5
M icrow ave Plasm a T reatm en t................................................................................56
3.6
Experim ental Design for Plasm a T reatm ent........................................................57
3.7
3.6.1
Response character:.........................................................................................58
3.6.2
Factorial design.............................................................................................. 59
Plasm a Surface M odification Effects on Cell A dhesion and Spreading
3.7.1
3.8
CH APTER 4
60
Dextran coating method for aminated Cyclotene film ................................60
A ging Effect S tu d y ...................................................................................................62
3.8.1
Storage media and experimental.................................................................... 63
3.8.2
Surface characterization................................................................................. 63
EX PERIM EN TA L RESULTS AND D IS C U S S IO N .................................65
4.1
O verview ..................................................................................................................... 65
4.2
D oE (D esign o f Experiments) R esults.................................................................. 6 6
4.3
D etailed Investigation o f the Effects o f Processing Param eters on Nincorporation on Cyclotene Surfaces..................................................................... 6 8
4.4
Surface Characterization and Bio-test R esults.................................................... 74
4.4.1
XPS characterization results...........................................................................75
4.4.2
FTIR-ATR characterization results................................................................80
4.4.3
AFM results..................................................................................................... 82
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Page
4.4.4
4.5
Results of cell adhesion and spreading studies after covalent coupling
of oxidized dextran to aminated Cyclotene surfaces....................................84
A ging Effect Study R esults....................................................................................... 8 8
4.5.1
XPS analysis of nitrogen and oxygen incorporation................................... 8 8
4.5.2
XPS spectra analysis: aging in PBS and air..................................................92
4.5.3
Discussion........................................................................................................95
CH APTER 5
TW O PROPOSED M ODELS FO R A M M O N IA PLASM A
R E A C T IO N ..........................................................................................................99
5.1
O verview ....................................................................................................................... 99
5.2
Plasm a Chemistry M odel in the A fterglow R eg io n ......................................... 100
5.2.1
Kinetic analysis of ammonia decomposition.............................................. 100
5.2.2
Plasma chemistry model in ammonia discharges.......................................103
5.2.3
Model validity............................................................................................. I l l
5.3
CH A PTER
Surface Am ination M odel and R eaction M echanism E x p lo ratio n
6
115
5.3.1
Reactions of reactive species with polymer surfaces................................ 116
5.3.2
Amination model...........................................................................................118
5.3.3
Model evaluation...........................................................................................123
5.3.4
Hypothetical reaction mechanisms.............................................................. 128
5.3.5
Surface reaction mechanisms on Cyclotene................................................129
CONCLUSIONS AND RECO M M END A TIO NS FOR FUTURE
W O R K ................................................................................................................ 136
A PPEND IX
A
PLASM A SYSTEM O PERATING P R O C E D U R E ................................ 145
B
BOND DISSOCIATION ENERGIES OF SELECTED SPEC IES
ix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
151
LIST OF TABLES
Table
Page
2-1
Processing param eters o f surface m odification in the literature................................16
2-2
Effect o f plasm a power on the surface m odification results o f p o ly m ers
2-3
Com parison o f ion density for different system s...........................................................21
2-4
Effect o f gas pressure on the surface m odification results o f various p o ly m ers. 22
2-5
Effect o f treatm ent time on the surface m odification results o f p o ly m e rs............. 24
3-1
Factors and le v e ls.................................................................................................................58
3-2
Summary o f design...............................................................................................................59
3-3
M atrix design....................................................................................
4-1
Result o f matrix design for am m onia plasm a treatm ent w ithout Ar
pretreatm ent......................................................................................................................... 6 6
4-2
Result o f m atrix design o f am monia plasm a treatm ent w ith Ar pretreatm ent
4-3
Com ponent XPS peaks and possible N -containing attributions................................ 79
4-4
FTIR-A TR peaks and their assig n m en ts.........................................................................81
5-1
Units for corresponding variables or constant............................................................111
5-2
Parameters employed in the chem istry m odel and the extracted m e a n s
5-3
Summary o f the values o f all constants for this kinetic m odel................................. 112
5-4
Definitions o f the symbols used in the am ination m o d e l.......................................... 123
5-5
Estim ated rate constants o f the am ination m odel.........................................................125
x
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
59
67
Ill
LIST OF FIGURES
Figure
Page
1-1
Cyclotene m onom er.............................................................................................................. 2
1-2
Cyclotene polym erization m echanism .............................................................................. 3
1-3
Typical Cyclotene w ater uptake response c u rv e ............................................................5
1-4
Schem atic o f plasm a surface m odification w ithin a plasm a re a c to r......................... 9
1-5
Schem atic o f the reaction m echanism s o f plasm a surface m odifications
2-1
Schem atic o f the down stream reactor configuration.................................................. 15
2-2
Radical shift reaction during am m onia plasm a trea tm e n t......................................... 33
2-3
Two possible reaction paths o f PFA by A r or N 2 p lasm a.......................................... 33
2-4
Possible direct oxidation o f PFA by O 2 plasm a trea tm e n t........................................ 34
2-5
Reaction m echanism s o f PC during exposure to UV light: a) direct
photodegradation mechanism; b) production o f m acroradicals (R 3 )
through the induced photodegradation mechanism; c) abstraction o f
m ethyl groups..................................................................................................................... 37
2-6
Proposed plasm a-induced crosslinking between siloxane chains in silicon
rubber................................................................................................................................... 39
2-7
M ethyl group elim ination reaction in PP by N 2 plasm a treatm en t.......................... 39
2-8
Radical induced double bond formation during plasm a treatm ent o f
polyolefines........................................................................................................................ 40
2-9
Reaction sequence for PP surface m odification using N 2 p la sm a ............................40
2-10
Possible overall reaction in the plasm a-treated polyim ide f il m ............................... 41
2-11
Reaction scheme for the covalent im m obilization o f periodate-oxidized
dextran onto plasm a-aminated surfaces....................................................................... 41
3-1
A ctual picture o f the microwave plasm a system em ployed in this study............... 50
xi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
Figure
Page
3-2
Schematic o f the experimental setup for the plasm a sy stem ...................................51
4-1
N /C ratio o f the plasm a-treated Cyclotene surface vs cham ber pressure as a
function o f pow er w ith and w/o A r pretreatm ent....................................................... 69
4-2
N /C ratio o f the plasm a-treated Cyclotene surface vs substrate tem perature
as a function o f power w ith and w/o A r p retreatm en t..............................................69
4-3
N /C ratio o f the plasm a-treated Cyclotene surface vs treatm ent tim e as a
function o f pow er w ith and w/o A r pretreatm ent....................................................... 70
4-4
O/C ratio o f the plasm a-treated Cyclotene surface vs cham ber pressure as a
function o f pow er with and w/o A r pretreatm ent....................................................... 72
4-5
O/C ratio o f the plasm a-treated Cyclotene surface vs substrate tem perature
as a function o f power w ith and w/o A r p retreatm en t..............................................72
4-6
O/C ratio o f the plasm a-treated Cyclotene surface vs treatm ent tim e as a
function o f pow er with and w/o A r pretreatm ent....................................................... 73
4-7
N Is spectra o f the Cyclotene before and after N H 3 plasm a treatm ents................76
4-8
C Is spectra o f the Cyclotene before and after N H 3 plasm a treatm ents................ 77
4-9
O l s spectra o f the Cyclotene before and after N H 3 plasm a treatm ents................ 78
4-10
FTIR-A TR spectra for am m onia (w ith Ar plasm a pretreated) plasm a-treated
and untreated Cyclotene. Treatm ent conditions are: Ar (150 W ,0.4 Torr,
180 s, 60 °C) an d N H 3(250 W, 0.6 Torr, 180 s, 175 °C)..........................................81
4-11
A FM images for am monia plasm a-treated Cyclotene films: a) control;
b) am monia plasm a (250W, 0.6Torr, 100 °C, 300 s); c) argon plasm a
( 150W, 0.4 Torr, 60 °C, 180 s) followed by am m onia plasm a (250W,
0.6Torr, 100 °C, 300 s); d) argon plasm a ( 150W, 0.4 Torr, 60 °C, 180 s)
followed by am monia plasm a (250W , 0.6Torr, 200 °C, 300 s).............................. 83
4-12
M icroscopic images o f adherent and spread cells on the substrates (scale bar
= 50 pm): (a) untreated Cyclotene surface; (b) N H 3 plasm a treatm ent under
low level operating condition; (c) N H 3 plasm a w ith A r plasm a pretreatm ent
under low level operating condition; (d) N H 3 plasm a treatm ent under high
level operating condition; (e) N H 3 plasm a w ith A r plasm a pretreatm ent
under high level operating condition............................................................................ 8 6
xii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Page
Figure
4-13
Cell adhesion value and N /C ratio under different operational co n d itio n s
4-14
N/C ratio as a function o f aging time in 3 different m e d ia......................................90
4-15
O/C ratio as a function o f aging tim e in 3 different m e d ia ......................................91
4-16
O Is spectra o f N H 3 plasm a-treated Cyclotene (w ith A r plasm a pretreatm ent):
(a) fresh sample im m ediately after plasm a treatm ent; (b) after 3 m onths o f
aging in P B S........................................................................................................................93
4-17
O Is spectra o f N H 3 plasm a-treated Cyclotene (with A r plasm a pretreatm ent):
(a) fresh sample im m ediately after plasm a treatm ent; (b) after 3 m onths o f
aging in air...........................................................................................................................93
4-18
N Is spectra o f N H 3 plasm a-treated Cyclotene (w ith A r plasm a pretreatm ent):
(a) fresh sample im m ediately after plasm a treatment; (b) after 3 m onths o f
aging in P B S....................................................................................................................... 94
4-19
N Is spectra o f N H 3 plasm a-treated Cyclotene (with A r plasm a pretreatm ent):
(a) fresh sample im m ediately after plasm a treatment; (b) after 3 m onths o f
aging in air...........................................................................................................................95
5-1
M odel-predicted N H 2 density as a function o f (a) pow er; (b) pressure
5-2
Reaction pathways for amino grafting: (a) addition o f active nitrogen;
(b) N H insertion; (c) N H 2 attachment; (d) conversion o f nitrogen groups;
(e) rem oval o f nitrogen groups..................................................................................... 118
5-3
Com parison o f the m odel-predicted and experimental density o f am ino and
non-amino groups as a function o f (a) power; (b) pressure, for N H 3 plasm a
treatm ent alone conditions.............................................................................................126
5-4
Com parison o f the m odel-predicted and experimental density o f am ino and
non-amino groups as a function o f (a) power; (b) pressure, for N H 3 plasm a
treatm ent w ith A r plasm a pretreatm ent conditions.................................................. 127
5-5
Possible reaction pathways for am monia plasm a treatm ent on Cyclotene
surfaces: (a) hydrogen abstraction from pendent CH 3 groups and further
oxidation process; (b) hydrogen abstraction from cyclobutane ring and
ring opening process; (c) direct nitrogen addition to C = C .....................................133
xiii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
87
115
CHAPTER 1
INTRODUCTION
At an estim ated US$80.3 billion in 2005, the U.S. m edical device m arket looks
set to break the US$100 billion barrier in the next five years . 1 N ew m edical products,
m aterials and surgical procedures keep im proving current health-care procedures. M any
o f these innovations involve polym eric devices that m ust m eet certain clinical and cost
requirem ents. The use o f synthetic materials in biom edical application has increased
dram atically during the past few decades. A lthough m ost synthetic biom aterials possess
the physical properties that m eet or even exceed those o f natural tissue, upon
im plantation they often result in a num ber o f adverse physiological reactions such as
throm bosis formation, inflammation and infection. Therefore, one o f the m ost urgent
requirem ents is the need for biocom patibility betw een the physiological environm ent and
the biom aterial surface. Plasm a surface m odification can im prove biocom patibility and
biofunctionality without changing a m aterial’s bulk properties. The key practical
application o f this study is to understand how plasm a treatm ent can be used to provide
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2
biom edical devices w ith stable hydrophilic, cell culture suitable and protein repelling
surfaces.
Selecting a m aterial w ith appropriate bulk and surface properties is critical to the
success o f the im plantation o f the biom edical device. D ow Cyclotene 4026 A dvanced
Electronic resins are I-line/G-line sensitive photopolym ers that have been developed for
use as dielectrics in thin film microelectronics applications .2,3,4,5 R esearchers at A rizona
State U niversity 6 have recently verified that this material is a good candidate for use in
the biom edical field. This polym er is derived from B-staged bisbenzocyclobutene (Dow
Cyclotene) chem istry and has excellent physical, chemical and electrical properties w hich
include: low dielectric constant, good processability (including photosensitivity to enable
simplified patterning), low moisture absorption (< 0 . 2 w t %) and good adhesion
properties. D ow Cyclotene 4026 is the Diels-Alder reaction product o f Cyclotene
monomer. The m onom er structure and the polym erization m echanism are show n in
Figure 1-1 and Figure 1-2, respectively.
CH
CH
S i—O — Si
CH
CH
Figure 1-1 Cyclotene monomer.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3
CH3
—
i
St
CH.
.
i
BCB M o n o m e r
D ie ls-A lder rea ction
■Si— O — Si
d im e r
B C B po ly m e r
Figure 1-2 Cyclotene polym erization m echanism.
D uring design/manufacturing o f im plantable devices such as neural interfaces, the
general issues we need to consider are: packaging, biocom patibility, regulation (Food &
Drug Adm inistration), reliability, electrom agnetic interference (EMI) issues, system
traceability- 1 0 0 %, size, weight, volume, m oisture resistance, mechanical shock,
vibration, therm al shock, battery life, corrosion, sterility, supplier developm ent and
reliability. In search o f the ‘ideal’ polym eric material for such applications, w ork relating
7 8
to the reliability and/or the processability o f Cyclotene has been reported. ’ These reports
have shown that Cyclotene is a high perform ance and reliable material. Cyclotene has
num erous processing advantages, such as high degree o f planarization, good
com patibility w ith copper (no migration), low curing tem perature (200-300 °C), no
volatiles evolved during cure, can be rapidly cured (in m inutes), negligible shrinkage
during cure, resistant to most processing chemicals, and low m oisture uptake (< 0 .2 wt%).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
The last property is critical for medical im plant applications. As is w ell know n, m oisture
ingression w ith subsequent failure o f electronic circuitry is a prim ary issue in polym er
packaging applications. Cyclotene was developed in the electronics industry to elim inate
this failure mode. Typical Cyclotene w ater absorption/desorption behavior 9 is show n in
Figure 1-3.
In m edical device manufacturing, a wide range o f materials is used for packaging,
including polym ers. For polymers, the prim ary issue is the m oisture ingression w ith
subsequent failure o f electronic circuitry. M EM s (M icro Electro M echanical Systems)
and m icroelectronics are very sensitive to water. Ingress o f only 104 pL w ater to bond
pads in packages can cause galvanic corrosion. Our candidate Cyclotene m aterial has a
m uch lower m oisture uptake (<0.2 wt %) com pared to the com m only used polyim ide (4-6
w t %).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
0,14
81% m
0.10
4-*
JC
OR
a
6% RH
200
400
«0#
800
WOO 12©0 1400
Time (s)
Figure 1-3 Typical Cyclotene w ater uptake response curve.
To achieve higher speeds and lower pow er in m icroelectronic packages, there is a
need for low dielectric constant interlevel insulators. In addition to low perm ittivity, these
insulators m ust also have excellent adhesion, low moisture absorption, and high glass
transition temperatures. Polyim ides suffer from the shortcom ings o f high dielectric
constants, high w ater absorption levels and inherent anisotropy. Cyclotene has
advantages over polyim ide in dielectric constant and m oisture absorption, but we still
need to realize that its final electrical properties, glass transition tem perature, and therm al
stability are process-dependent.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
B iocom patibility is the ability o f a m aterial to perform w ith an appropriate host
response in a specific application. On a local level, nonspecific protein adsorption occurs
quickly after im plantation. Surface characteristics, e.g., roughness, porosity, electrical
charge, chem ical nature o f the surface, surface energy (o f the solid), texture, subsurface
grain structure and w ettability im pact the type and am ount o f proteins that adsorb. A fter
proteins adsorb to a surface, they becom e the solid interface in contact w ith the
physiological environm ent, so protein adsorption is a critical factor for the success o f
im planted devices. Adverse physiological reactions such as throm bosis formation,
platelet adhesion and aggregation can occur locally. O n a m ore global level, the
im planted biom aterial interface can cause inflam m ation and infection as w ell as toxicity.
Implants not only have to be biosafe and biostable in term s o f cytotoxicity and
degradation, they also have to satisfy ‘biological’ system solutions, i.e., they should
m im ic the biological tissue near the im plant w ith regard to sm ooth edges, a density close
to that o f the tissue, and high flexibility to prevent m echanic nerve traum atisation often
know n as m echanical or structural biocompatibility.
For active im planted microdevices, a significant opportunity is to use the design
and fabrication tools o f microsystem technology and tailored m aterials’ surface properties
to design and fabricate implantable devices that satisfy biological system solutions.
Substrates w ith electronics, on one hand, and package and encapsulation, on the other
hand, should no longer be considered as two parts o f a system but as a synergism.
Flexible m icrom achined substrates with hybrid assem blies o f electronics and w ell-know n
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7
encapsulation material Cyclotene m ay in this w ay lead to production o f robust flexible
m icroimplants.
Biom aterials that come in contact with blood or protein require special surface
treatm ents to enhance biocom patibility. Am ine functional groups, w hich are attached by
am m onia plasm a treatm ent, act as hooks for anticoagulants, such as heparin, and thereby
decreasing throm bogenicity . 10 Synthetic polym eric im plant m aterials can be surface
activated using plasm a techniques to enable covalent im m obilization o f cell-binding
peptides derived from the extra cellular m atrix protein: fibronectin and lam inin. The
resulting grafted peptides can prom ote com plete coverage o f a surface with a m onolayer
o f intact, healthy endothelial cells to form a natural blood com patibility surface. The
endothelialized bioim plants will have im proved biocom patibility and reduce antigenicity
and throm bogenicity . 11
The work described in this dissertation is aimed at the follow ing technological
goal: em ploy plasm a surface m odification to inhibit undesirable cell adhesion to
im planted device and to enable im m obilization o f bioactive m olecules. Identification o f
how plasm a treatm ent can be used to provide biom edical devices w ith stable hydrophilic
surfaces that are suitable for cell culture and w ith protein repelling surfaces is the key
fundam ental objective o f this study. Since dextran has protein rejection properties and is
multivalent, a covalent im m obilization o f oxidized dextran on plasm a m odified
Cyclotene surface followed by a cell adhesion and spreading study w ill be conducted as
the test vehicle for functionalization o f the surface. If amino functional groups are
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
8
successfully introduced by am m onia plasm a treatm ent, the dextran will be bound onto
polym er surface and little cell/protein adsorption w ill be observed.
A plasm a is a partially ionized gas containing electrons, ions, and various neutral
species at m any different levels o f excitement. Plasm a is often referred to as a "fourth
state o f matter". Cold gas plasm a is an energetic process utilizing electrical energy to
transform very small am ount o f innocuous gases, such as air, nitrogen, oxygen, and argon
into very chem ically reactive and aggressive species. The energetic gas particles in the
plasm a interact w ith solid surfaces placed in the plasm a environm ent, thereby causing
m odification o f the molecular structure to create desired surface properties. This process
is schematically shown in Figure 1-4.
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
P rocess G as In
Vacuum Reaction Chamber
Glow discharge
(Short wave & long wave UV)
Excited G as S p ecies
Plastic Substrate
RF
®ource
P rocess G as Out
Ground Electrode
Figure 1-4 Schematic o f plasm a surface m odification w ithin a plasm a reactor.
The gas is energized using techniques such as radio-frequency energy,
m icrowaves, alternating current or direct current. The energetic species in a gas plasm a
include ions, electrons, radicals, metastables, and photons in the short-wave ultraviolet
(UY) range. Surfaces in contact w ith gas plasm as are bom barded by these energetic
species and their energy is transferred from the plasm a to the solid. Energy is dissipated
w ithin the solid by a variety o f chemical and physical processes as schem atically
illustrated in Figure 1-5, 12 to result in the surface modification. These processes include:
a) crosslinking (create a polym er network by m aking links in polym er chains); b) etching
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
(remove atomic or molecular layers on the polym er); c) deposition (deposit atom ic or
m olecular layers on polymer); and d) functionalization (replace specific chem ical
groups).
M
M
-P-P-P-P-P-P-PM: -
Energy
-P:
Glow Discharge
Inert G a ses
Monomers
Organics
Reactive G a ses
Polymer chain
UV
-P-P-P-P-P P-P-
Termination of
Free Radicals
y
A) Crosslinking
B) Etching (Degradation)
C) Deposition (Grafting)
-M
P-P
-p-p-p-p
-p -p-p-p-
p .p .p .p .
-p-p-p
P
M-M
p .p .
M
-p -p -p -p -p
D) Functionalization
MM-M-
M M
ill
I I
p-p-
-p -p -p -p -p
M
I
p-p-
p-p-p-
Figure 1-5 Schematic o f the reaction mechanisms o f plasm a surface m odifications.
The unique surface m odification that can be achieved using the plasm a process
results from the effects o f the photons and active species in the plasma. These reactive
particles react w ith surfaces in depths from
1 -1 0 0
nm w ithout changing the bulk
properties o f the biomaterial. The chem ical and physical characteristics o f a plasm a can
be directly affected by a wide variety o f processing param eters such as gas types,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11
absorbed pow er, operating pressure, treatm ent tim e and substrate tem perature; and also
can be varied by the physical system param eters, such as reactor design, electrode
location, gas inlets and vacuum. This wide range o f param eters offers greater control over
the plasm a process than that offered by therm al or other high-energy radiation processes,
so that the subsequent surface chemistry obtained by plasm a m odification can be
effectively tailored.
The next chapter is a review o f the body o f literature on plasm a surface
m odification o f various polymers. M ost papers described the study o f the effect o f
experimental param eters on the characteristics o f surface m odification o f polym ers. V ery
few have studied surface am ination o f low dielectric polym ers, and none o f them has
reported inform ation on surface am ination o f Cyclotene. A lthough substantial data exist
on the effects o f process parameters on the surface m odification o f polym ers, little has
been reported on controlling reaction mechanisms, correlations betw een the extent o f
am ination and the biocom patibility o f m odified polymers. M ost papers assume that
tem perature w ill not have a measurable effect on the final am ination result, and in fact no
systematic investigation o f this potentially important param eter has been performed.
A m ajor challenge for this w ork is apparent process reproducibility, because the
therm al and chem ical stability o f surface modified films is poor, in that exposure to air
leads to surface restructuring and post-treatment chem ical changes. A spontaneous post­
plasm a oxidation can occur due to surface radicals created during the plasm a process.
This post-plasm a oxidation also can degrade the amino selectivity (NH 2 /N) on the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
polym er surface by incorporation o f various oxygen functionalities. This research focuses
on characterizing the surface am ination o f D ow Cyclotene 4026 through experim entation,
m odeling and identifying the m odification effect on biocom patibility im provem ent o f the
polym er, as well as exploring the reaction m echanism on Cyclotene surface. The
experim entation includes characterizing the im pact o f the process param eters: absorbed
power, cham ber pressure, substrate tem perature, and treatm ent time. A nalytical
techniques (XPS, FTIR-ATR and A FM ) and biotest (cell adhesion and spread study) are
also applied to study the reaction m echanism and to identify the final effect o f surface
m odification.
Following the literature review, the subsequent chapter describes the experim ental
m ethods employed in this study. The experim ental results including im m ediate p o st­
processing and the tim e evolution o f the surface properties for plasm a-treated Cyclotene
stored in various environm ents are then described. Finally, two kinetic m odels, (one
plasm a kinetic model in the gas phase and one am ination model on a surface) w ere
proposed, and the two models were evaluated. These models provide a fram ew ork for
understanding possible important reaction pathways for the surface am ination o f
Cyclotene.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
References
1
M edical D evice M arket Outlook, March, 2005.
2
Im, J.-H.; Ii, E. O. S.; Theodore Stokich, J.; Strandjord, A.; Hetzner, J.; Curphy, J.;
Karas, C.; M eyers, G.; Hawn, D.; Chakrabarti, A.; Froelicher, S. J Electron Packag
2 0 0 0 , 1 2 2 , 28.
3
M eyers, G. F.; Dineen, M. T.; Schaffer, E. O.; Stokich, T.; Im, J.-H.
M acrom odecular Symposia 2001,167, 213.
4
Elenius, P.; Janssen, R.; Strandjord, A. J. G. in Proceedings o f Semicon W est, 1997.
5
Clearfield, H. M .; W ijeyesekera, S.; Logan, E. A. in Proceedings o f the 7th.
International Conference on M ultichip M odules, Denver, CO, 1998.
6
Ehteshami, G.; Singh, A.; Coryell, G.; M assia, S.; He, J.; Raupp, G. J B iom ater Sci,
Polym er Ed 2003, 14, 1105.
7
Garrou, P. E.; Heistand, R. H.; Dibbs, M. G.; M ainal, T. A.; M ohler, C. E.; Stokich,
T. M.; Townsend, P. H.; Adema, G. M.; Berry, M. J.; Turlik, I. IEEE Transactions
On Components, Hybrids, And M anufacturing Technology 1993, 16, 46.
8
Chong, C. K.; Peng, C. C.; Ngoh, L. G.; Teo, M., http://ww w .ellipsizm icrofab.com /icep .pdf.
9
Scheck, D. W LP Seminar, Ultratech Stepper, D ow Chem ical Com pany 2001.
10 Yuan, S.; Szakalas-Gratzl, G.; N.P. Ziats; D.W. Joacobsen; K ottke-M archant, K.;
M archant, R. E. J Biom ed M ater Res 1993, 27, 811.
11
Kiaei, D.; Hoffman, A. S.; Ratner, B. D.; Horbett, T. A.; Raynolds, L. O. J Appl
Polym Sci: Appl Polym Symp 1998, 42, 269.
12 Loh, I.-H., http://ww w .astp.com /PDFs/PS_biom ed.pdf.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 2
LITERATURE REVIEW
2.1 Introduction
Polymers have become w idely accepted for in-vivo and in-vitro medical
applications. M any o f these m aterials have properties that lend themselves well to the
m anufacture o f medical appliances or devices; moreover, polym ers are easily m olded or
form ed into com plex shapes and bulk physical properties that m ay be selected from a
w ide range o f param eters such as rigidity and tem perature stability. U nfortunately,
fabrication procedures that require bonding are difficult to achieve, and biological
interface reactions w ithin the body or in the laboratory can lim it their in-vivo and in-vitro
performance.
Plasm a modification o f polym er substrates is a proven technology 1 that has been
used to attach bio-m olecules to substrate surfaces .2,3 This w ork is focused on surface
am ination using downstream am m onia plasma.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15
A dow nstream plasm a system (schem atically shown in Figure 2-1) can be used to
create a m ixture o f photons, electrons, ions, radicals and atoms that have the potential to
react with the substrate surfaces. For dow nstream plasm as, the concentrations o f
electrons and ions are generally too low to contribute to surface reaction. D ow nstream
plasm a is therefore a m ilder plasm a process relative to a direct plasm a, and therefore
m inim izes or avoids undesirable surface damage. For any gas com position, three
sim ultaneous processes alter the outer m olecular layers o f the polym er: ablation,
crosslinking and activation. The relative contribution o f each depends on the chem ical
nature o f the gas plasm a, the properties o f polym ers and the processing param eters.
r
d - 24 cm
Pumping
G as Feed
System
1* Quartz Tube
OPC?S Cavity
X
Sam ple Holder
Coaxial Slug Tuner (CST)
’ U
4” Stainless Steel Cham ber
MWG
Figure 2-1 Schematic o f the down stream reactor configuration.
This literature review summarizes topics that form the foundation o f this work.
These topics include an overview o f process chem istries and associated process
param eters em ployed for surface modification o f various polym ers, process variable
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
effects on surface m odification, characterizations o f m odified polym er surfaces, possible
reactions in plasm a surface modification, and aging effect studies.
2.2 Overview of Process Chemistries and Associated
Process Parameters for Different Polymers
A m m onia or nitrogen plasm as are used to modify the chem ical structure and
reactivity o f polym er surfaces in order to incorporate the desired functional groups, and
to improve surface adhesion, interfacial shear strength, and cell adhesion. Literature
ranges o f process variables for plasm a treatm ent o f different polym ers are listed in Table
2-1. The processing param eters on w hich researchers tend to focus their efforts are:
absorbed power, cham ber pressure, feed gas flow rate and treatm ent time. Few reported
values o f temperature.
Table 2-1 Processing param eters o f surface m odification in the literature
Power
(W)
Pressure
(Torr)
polyethylene(U HPE)
aramid fibers
(Aromatic Polyamide)
polyvinylidene or
polypropylene fluoride
(PVDF)
RF 30-100
0.5
Gas
flowrate
(seem)
nh3
RF100
0.25
NH3/ 20
RF5-60
0.15-0.5
[7]
carbon fibers
RF 50
0.18
[8,9,10]
polytetrafluro-ethylene
(PTFE)
[8]MW200~500
[9,10] d.c. 2.3-15
[8]N/A
[9,10]
0.03-0.08
RF 20
0.3
nh3
RF 1200
0.1
NH3 / 44
RF 300
0.038 (5 Pa)
Ref.
Polymer
[4]
[5]
[6]
[11]
[12]
[13]
polytetrafluro-ethylene
(PTFE)
polytetrafluroethylene
(PTFE)
polyester (PES)
n h 3/
0.5-10
NH3 /35
n h 3/
30-200
Temp
(°C)
Time
(Min)
100
1/6-25
N/A
5
N/A
30
room
temp.
1-20
N/A
1-5
N/A
1-4
room
temp,
n h 3/ n 2
low temp.
small ratio (< 60 °C).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0.08-32
3
17
Ref.
Polymer
Power
(W)
Pressure
(Torr)
Gas
flowrate
(seem)
Temp
(°C)
Time
(Min)
[14,15]
polystyrene
RF 60
0.5
nh
3
room
temp.
0.5
3.7
nh3
room
temp.
1
2x10'4
nh3
30
30
0.008-0.04
h2
N/A
0 -3 0
[16]
[17]
[18]
silicone rubber (dimethyl
and methylvinyl siloxane RF 50
copolymers)
RF 430
silicone
polyethylene
terephthalate (PET),
MW 300
polypropylene (PP),
polyethylene (PE)
RF 100
[23] MW 60W
0.76
0.23
-10% n 2
+90% Ar
n2
nh3
Ar
NH3,N 2 10
N 2 / 20
[24] RF20-23
1.5xl0'4~0.(
N 2; Ar
polypropylene (PP)
CD 800
0.38-7.6
nh3
[26]
polyimide (PI)
MW 45-225
0.075-0.25
[27]
polyimide (PI)
fluorinated ethylene
[28 29]
’
ProPylene (FEP)
polytetrafluoroethylene
(PTFE)
polypropylene (PP) +
polystyrene (PS)
(A 12/PS) /PEFE
polycarbonate (PC)
polydimethylsiloxane
(PDMS)
Poly(D,L-lactide) film
Poly ethyl en eterephtal ate
(PET)
Poly(ethylene-2,6naphthalate) (PEN)
[19]
perfluoroalkoxyvinyl
ether (PFA)
r?n911
polyethylene (PE)
polypropylene (PP)
[22]
polypropylene (PP)
[23,24]
polypropylene (PP)
[25]
[30]
[31]
[32]
[33]
[34]
[35]
MW 600
0.08-4.6
MW 1200
0.08-4.6
RF 100
N/A
0.3-3
N/A
N/A
N/A
N/A
N/A
3
1
10
1-20
5
N/A
1-15
N/A
150 mA at 20 kHz 0.1
N/A
0 2 and N 2 room
additive
temp
N/A
Ar, N 2
RF 20-120
0.2 -0.5
NH3/1~6
N/A
1 /6 -2
RF 185
0.38
NH3/55
N/A
2
MW 200
0.06
Ar/ 85
N/A
6
RF10 -4 0 0
0.3 - 0 .7
Ar, N2
N/A
0.05-5
RF 30-90
0.15-0.61
nh3
N/A
0.17-2
RF 10-150
N/A
nh3
N/A
0.02-2
RF 60-600
0.05-0.15
n
2
N/A
0.01-0.05
RF: radio frequency source; MW: microwave source; CD: corona discharge; N/A: not available.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0.02-0.23
3
18
2.3 Discussion of Process Variable Effects on Surface
Modification
As the table illustrates, principal adjustable independent processing param eters
are power, pressure, tem perature, treatm ent time, feed gas com position and flowrate.
Some researchers have investigated single param eter effects on the surface m odification
performance, but none have perform ed a detailed study on the com bined effects o f all the
processing parameters. It is expected that the total effects o f the plasm a treatm ent are due
to the synergy o f some o f these factors .36,37 Chevallier et al . 11 have shown that the surface
chem istry can be m odulated through appropriate selection o f the plasm a treatm ent
parameters. It should be possible to minimize undesirable reactions, such as chain
scission, through the appropriate selection and tuning o f experim ental parameters.
2.3.1 Power effects
Table 2-2 summarizes the observable effects o f plasm a pow er on the surface
m odification o f various polymers. From this table, we can see the particular surface
m odification result, w hich is indicated by extent o f nitrogen incorporation, interfacial
shear strength and contact angles, etc., increase with an increase in plasm a pow er for
m ost cases. A n exception is the study 4 which showed that the interfacial shear strength
(IFSS) values increased w ith power and reached a m axim um at a pow er level o f 30 W
and further increase in the power input resulted in a decrease o f the IFSS values. A
possible explanation for this observation is that at higher pow er levels, m ore energy is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
available to break the am monia m olecules into ions or sm aller free radicals, w hich m akes
it difficult to obtain higher prim ary am ine group concentrations.
The effect o f pow er can be com plex and can affect more than ion density. O verall,
the ion density at the selected operating conditions is a com plex function o f input pow er
and total pressure. Pringle et al .9 reported that at low er pressure the ion density is low er
than that m easured at higher operating pressure, and that the variation w ith pow er is
alm ost linear. The same effect was also observed by H opw ood et a l .38 H owever, B osw ell
and Porteous 39 found that the ion density w as not linear w ith pow er in their system. Table
2-3 com pares the plasm a characteristics for different plasm a systems operated under
various conditions.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
Table 2-2 Effect of plasma power on the surface modification results of polymers
Power
(W)
Pressure
(Torr)
Gas
flowrate
(seem)
Time
(Min)
Surface
modification
effect
100
5
Interfacial shear
strength
increased
48-109% .
N/A
30
Fluorescence
intensity
7.5%~5.5%.
N/A
[8] N/C ratio
[8] 2
0.08-0.14
[9, 10] 1.2 [9] N/C ratio
0.26-0.35.
N/A
3
N incorporation
increases with
power.
nh3
N/A
1
Interface
toughness T/Tjo
3.6-3.8
Ar
N/A
5
Interface
toughness T/Tjo
4.9-6.9.
0 2 and
2
additive
25-85
N/A
Contact angle
60°~20°.
Ref.
Polymer
[4]
polyethylene
(UHPE) fibers
RF 30-100
0.5
nh3
[6]
polyvinylidene
or
polypropylene
fluoride
(PVDF)
R F5- 60
0.15
NH3/ 0.5
[8]MW
200-500
[9, 10] d.c.
2.3-15
[8]N/A
[9,10]0.03
[8]NH3/
30-200
[9,10]
N/A
polytetrafluro[8,9,10] ethylene
(PTFE).
[19]
perfluoroalkoxyvinyl
ether (PFA)
MW
600-1200
-10% n 2
+
0.08-4.6
90% Ar
2
Temp
(°C)
n
polyethylene
(PE),
Polypropylene
(PP)
RF 100-300
[26]
polyimide (PI)
MW/DC
45-225
[32]
polydimethylsiloxane
(PDMS)
R F10- 400
0.3 - 0 .7
Ar, N 2
N/A
0 .0 5 - 5
Adhesion in the
interface
increases with
power.
[33]
Poly(D,Llactide) film
RF 30-90
0.3
nh3
N/A
2
Contact angle
(H20 ) 44°~35°.
[34]
Polyethyleneterephtalate
(PET)
RF 10-150
N/A
nh3
N/A
0.02-2
N/C ratio
0.12-0.4.
[35]
Poly(ethylene2,6-naphthalate) RF 60-600
(PEN)
0.05-0.15
n
2
N/A
0.01-0.05
Nitrogen %
increases with
power.
[20,21]
0.075-0.25
n
RF: radio frequency source; MW: microwave source; CD: corona discharge; N/A: not available
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
Table 2-3 Com parison o f ion density for different systems
Pressure
(mbar)
Ion density
(cm-3)
Plasma
potential
(V)
R ef
System details
Input power
(W)
[40]
ECR 2.45 GHz
400
0.02
4 x 1010
—
[41]
Induct coupled
200
0.01
1.9 x 10u
18.1
[38]
Induct coupled
300
0.001
1 x 10"
30
[39]
Induct coupled
50
0.02
2 x 109
12-15
[42]
Cap coupled
100
0.1
1.5 x 109
—
[40]
Helicon
1500
0.02
2.5 x 1011
—
[43]
Helical resonator 11-21 MHz
280
0.002
1.5 x 1011
—
[44]
Helical resonator 13.6 MHz
—
0.07
4.9 x 109
—
[44]
Helical resonator 13.6 MHz
500
0.002
2.5 x 10'°
—
[9]
Helical resonator 16.0 MHz
15
0.1
1.5 x 1010
15
[9]
Helical resonator 16.0 MHz
15
0.04
1.3 x 1010
24
The higher ion density plasm a produced is due to the presence o f a 100 G
electrom agnet surrounding the lower part o f the resonator shield .43
2.3.2 Gas pressure effects
Table 2-4 summarizes the effects o f plasm a cham ber total pressure on the surface
properties. Overall, m ost references reported an optimal surface m odification
perform ance over the range o f pressures studied.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
Table 2-4 Effect of gas pressure on the surface modification results of various
polymers
Ref.
Polymer
Power
(W)
Pressure
(Torr)
Gas
flow rate
(seem)
Temp
(°C)
Time
(Min)
Surface modification
effect
[9,10]
polytetrafluroethylene
(PTFE).
d.c. 15
0.03-0.08
N/A
N/A
1.2
N/C ratios 0.23-0.26.
[19]
perfluoroalkoxyvinyl
ether (PFA)
MW 1200 0.08-4.6
2
N/A
3
N incorporation
increases with
pressure.
[24]
polypropylene
(PP)
RF20-23
1.5x1 O'40.08
N 2; Ar
N/A
1-15
Sticking coefficient
o f Mg decreases with
pressure.
[26]
polyimide (PI)
MW 200
0.075-0.25
0 2, (N2
additive)
25-85
N/A
Peel strength
decreases with
pressure.
[32]
polydimethylsiloxane
(PDMS)
R
lvlFin~
1U
400
0.3 - 0 .7
Ar, N 2
N/A
0 .0 5 - 5
Adhesion in the
interface increases
with pressure.
[35]
Poly (ethyleneRF
2,6-naphthalate)
60-600
(PEN)
0.05-0.15
n
N/A
0.01-0.05
Nitrogen increases
with pressure.
n
2
RF: radio frequency source; MW: microwave source; CD: corona discharge; N/A: not available.
Gas pressure range is source type dependent w ith 0.008 ~ 4.6 Torr for m icrow ave
plasm a sources, and 2 x l0 ‘4 ~ 3.7 Torr for radio frequency plasm a sources. The variation
in the optimal pressure m ost likely arises due to different processing param eters and
reactor geometries. None o f the authors give a detailed explanation about the effect o f
pressure on the observed surface modification. However, the general interpretation for
this phenom enon is due to the ‘glow discharge lim itation’ associated w ith m ean free path
variation w ith pressure. W hen the chamber pressure is low, the m ean free path is large
and collisional excitation events are limiting. As pressure increases from low values one
therefore observes an increase in excitation events, ion density and hence observable
surface m odification effects. However, as the pressure further increases, the m ean free
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
path begins to decrease, and the num ber o f recom bination events (especially three-body
events improbable at low pressures) that consum e reactive species start to outw eigh the
generation events, w hich leads to low er concentration o f reactive species. In addition, the
residence tim e inside the delivery tube is pressure dependent. H igher pressure leads to
longer residence tim e and subsequently m ore reactive species are consum ed through
recom bination processes in the dow nstream regions o f the plasm a process tool. Overall
one therefore expects to observe a m axim um in treatment effect if the pressure region
studied is sufficiently wide.
2.3.3 Treatment time effects
A third factor that controls m odified surface characteristics is the treatm ent tim e
o f plasmas. As shown in Table 2-5, m any researchers have investigated the treatm ent
tim e effect on the m odified surface properties. M ost reports show that there is an
optim um m odified surface characteristic at a given time point.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
Table 2-5 Effect of treatment time on the surface modification results of polymers
Ref.
Polymer
[4]
polyethylene(U
HPE) fibers
0.5
uas
^ aS
flowrate
(seem)
NH3
„
100
room
temp.
Time
(Min)
Surface
modification effect
1/6-25
Interfacial shear
strength (IFSS)
reached maximum
at 5 min.
1-20
Transverse tensile
strength reached
maximum at 5 min
and surface energy
reached maximum
at 10 min.
NFT 735
[8]MW
polytetrafluro200
ethylene (PTFE). [9,10]
d.c. 15
[8]N/A
[9,10]0.03
[8] NH3/
30 ~ 200
[9,10]
N/A
N/A
polytetrafluroethylene (PTFE) ^
0.3
NH3
N/A
1-4
Concentration o f
amines increases
with time.
0.1
NH3 / 44
room
temp.
0.08-32
Average pull
strength reached
maximum at 1 min.
carbon fibers
[8,9,
10]
[ 11 ]
[18]
* * 30
Pressure
(Torr)
0.18
[7]
[ 12]
Power
(W)
ethylene (PTFE)
polyethylene
terephthalate
(PET),
polypropylene
RF 50
zu
RF 1200
[8] 1-5
[9, 10]
1- 1.6
MW 300
0.008-0.04
H,
N/A
0-30
Bond strength
reached maximum
at 0.3-5 min (PE);
30 min (PP); 0.01-5
min (PET).
perfluoroalkoxyvinyl
ether (PFA)
MW 600
0.08-4.6
-10% N 2
+ 90%
N/A
Ar
0.3-3
N incorporation
increases with time.
polypropylene
Rp m
0 76
N H 3 ,N 2
,
0 -20
Concentration o f
amine groups
reached maximum
at 2 sec.
N/A
0-60
Nitrogen %: 0-18.
1-15
Sticking coefficient
o f Mg reached
maximum at 20-30
sec.
(PP),
polyethylene
(PE)
[19]
[22]
[23]
[24]
[8] N/C ratios
0.06-0.14
[9] N/C ratios
0.26-0.30;
10
polypropylene
(PP)
polypropylene
(PP)
MW 24
0.23
N 2 / 20
RF
20-23
1.5x10''
0.08
N,; Ar
N/A
[25]
polypropylene
CD gQ()
q.38-7.6
NHj
N/A
0.02-0.23
N/C ratios
0.01-0.14; Surface
adhesion increases
with time.
[28,29]
fluorinated
RF 20-
0 .2 -0 .5
NH3/1~6
N/A
1/6 - 2
N/A
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
Ref.
Polymer
Power
(W)
Pressure
(Torr)
Gas
flowrate
(seem)
Temp
(°C)
Time
(Min)
Surface
modification effect
N/A
0.05-5
Adhesion in the
interface increases
with time.
ethylene
120
propylene (FEP),
polytetrafluoroet
hylene (PTFE)
[32]
polydimethylsiloxane
(PDMS)
R F10400
0 .3 - 0 .7
Ar, N 2
[33]
Poly(D,Llactide) film
RF
30-90
0.3
nh
3
N/A
0.17-2
Contact angle
(H20 ) 60.5°-40.5°.
[34]
Polyethyleneterephtalate
(PET)
RF 150
N/A
nh3
N/A
0.02-2
N/C ratio 0.08-0.4.
RF: radio frequency source; MW: microwave source; CD: corona discharge; NA: not available
The effect o f surface m odification increase with treatm ent tim e can be explained
by the hypothesis that polar m olecules generated in the topm ost layer o f the polym ers
increase with tim e during the plasm a process. For longer plasm a treatm ent time,
Chevallier et al . 11 observed a partial conversion o f the nitrogen-containing species (other
than amines) to amino groups. M oreover, other undesirable side reactions, such as the
form ation o f alkenes, occur, and these m ay track the formation kinetics profile w ith that
o f the amino group formation. Chain scission may also occur.
2.3.4 Temperature effects
A thorough literature search failed to uncover any papers that reported the effect
o f tem perature on the surface chem istry o f the film in plasm a surface modification.
However, com mon sense and chemical intuition suggest that the higher the tem perature,
the faster a given chem ical reaction will proceed. This relationship betw een the rate a
reaction proceeds and its temperature is quantitatively expressed by the Arrhenius
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
26
Equation. The activation energy is the energy barrier for the reaction in question. The
Arrhenius equation is shown as follows:
k=A*exp(-Ea/R*T)
(2-1)
where k is the rate coefficient, A is a constant, Ea is the activation energy, R is the
universal gas constant, and T is the tem perature (in degrees K elvin). R has the value o f
8.314 x 10‘3 kJ m o l^K '1.
However, in plasm a chemical reactors, the existence o f the various ionized
particles and surfaces imposes several differences from classical therm al hom ogeneous
reaction systems. From the kinetic point o f view, the presence o f highly excited species
requires the use o f state-specific data, while A rrhenius-type extrapolations can’t describe
the variation o f the rate constant o f these non-therm al electron-m olecule reactions as a
function o f the energy o f the system.
2.4 Characterization of Modified Polymer Surfaces
For the plasm a m odified polym er surfaces listed in Table 2-1, many researchers
have reported the use o f XPS (X-ray photoelectron spectroscopy) and SEM (scanning
electron microscopy) to characterize the surface properties after plasm a treatm ents. Other
characterization techniques that have been em ployed to analyze the m odified polym er
surfaces include FTIR-A TR (attenuated total reflection Fourier transform infrared
spectroscopy), AFM (atomic force microscopy) and HREEES (high -reso lu tio n electron
energy-loss spectroscopy). Gerenser et al .35 and A ndre et a l .25 also reported static SIMS
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
(secondary ion m ass spectrometry) analysis on m odified polym er surfaces. The
researchers 25,35 tried to understand better the exact nature o f those high m olecular
fragments w hich m ay be produced by the nitrogen or am m onia (bond breaking, ruptures
in the polym eric chain). Shahidzadeh et al .45 developed a new application o f CIA
(capillary electrophoresis ion analysis) which perm itted them to identify the acidic low
m olecular w eight fragm ents formed at the plasm a-treated polypropylene surface.
2.5 Possible Reactions in Plasma Surface Modification
Plasm a treatm ent operates through the interaction o f energetic particles and
photons with the polym er surface. The action o f plasm a results in activation (radical
formation), degradation (chain scission), unsaturation, and crosslinking, w hich depends
on the plasm a processing conditions and on the polym er properties. A ctivation can favor
chem ical bonding betw een the treated polym er and the adherent, an effect w hich is more
readily achieved by active gas plasm a such as N H 3 , N 2 , and O 2 , w hich can graft reactive
groups. W hen treated w ith inert gas plasm a, such as Ar, the “w eak boundary layer”
(W BL) at the surface can be crosslinked which enhances the cohesion strength o f the
surface.
Plasm a m odification to produce functionalized groups on the polym er surface
typically consists o f two steps: 1) The plasm a gas forms surface radical sites; 2) The
radicals on the surface undergo grafting reactions w ith adsorbed or/and im pinging
reactive species.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
The m icrow ave plasm a dow nstream treatm ent doesn’t involve the participation o f
ions and electrons in superficial m odification process. The role o f UV irradiation has
been found negligible in the case o f polypropylene m icrow ave plasm a dow nstream
treatm ent .46 However, with PTFE, it has been shown that U V irradiation can m odify the
surface when using long exposure tim e ,47 i.e., tim es that are m uch larger than those listed
in Table 2-1. U nder such conditions, it is possible to realize a superficial physicochem ical
m odification o f the polym er w ithout degrading its bulk properties.
2.5.1 Dissociation of ammonia in the plasma gas phase
Plasm a m ay be generated by a wide range variety o f techniques (d.c., r.f. -diode,
ECR, etc.). As described above, the processing param eters including pow er, operating
pressure, substrate temperature, and treatm ent tim e are know n to have a m easurable
effect on the plasm a characteristics. Hence, a surface exposed to one plasm a m ay possess
totally different properties to that o f another treated in a different system. A n additional
difficulty arises from the fact that m any plasm a-treated surfaces adsorb m oisture or react
w ith oxygen in an uncontrolled w ay upon exposure to the atmosphere. This “aging” issue
is discussed in section 2.7 below.
In m any cases the specific plasm a conditions (i.e., electron and ion density and
energy, radicals’ concentrations) are unknown, which m akes it difficult to directly
quantitatively compare results from different research groups. The next several
paragraphs summarize the few studies in w hich careful plasm a characterization to
determ ine the active species present was undertaken.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
S. D. Pringle et al. 9 conducted plasm a diagnostics and in situ XPS analysis while
running a helical resonator plasm a at 2.3 ~ 15 W and at 0.04 - 0 . 1 mbar. Radical
detection was attem pted using threshold ionization spectroscopy. The threshold
ionization technique is based on selective ionization o f neutral species using electron
im pact ionization w here the electron energy is w ell defined. For radical detection the
spectrom eter was tuned in residual gas analysis (RGA) m ode at the required mass and a
large (>30 Y) potential was applied to the extraction electrode to repel any incident
positive ions. The residual gas analysis showed that the neutral com position was
predom inantly N H 3 w ith trace amounts o f O 2 and N 2 . A n R G A spectrum taken from
am m onia plasm a shows that N , N H and N H 2 mass peaks, respectively w hich are form ed
in the spectrom eter ion source by electron im pact dissociation o f N H 3 .
A positive ion spectrum and RGA showed that an ion at m ass 18 was form ed via a
proton transfer reaction or ion-molecule collision
or
48
o f the form
N H / + N H 3 => N H 4++ N H 2 •
(2-2)
NH, + H + => N H 4+
(2-3)
The ions observed at larger masses were produced through clustering reactions o f
N H 4+ w ith N H 3 in the ambient atmosphere in the cell.
M ass 35: N H / + NH, o
NH, + N H /
(2-4)
M ass 52: N H / N H , + NH, <=> N H / +NH,NII,
(2-5)
M ass 69: N H / N H , N H , + NH,
(2 -6 )
N H /(N H ,),
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
U sually the growth o f ion clusters continues until a critical size is reached. A t this
point the rates o f dissociation and association o f the clusters becom e equal .
W ojcik and Bederski 49 investigated the ion-m olecule reactions o f am m onia using
an electron beam ion source, and only observed ion clusters o f m ass 35. In their study the
am m onia pressure range studied was 0.027 ~ 0.27 mbar. G losik et al . 50 observed cluster
ions o f the form N H 4+(NH 3 )n, where n = 1, 2, 3, 4 (max. m ass = 80 amu). Their reaction
system was based upon a flowing after-glow in w hich N H 3 atoms w ere ionized via
charge-exchange reaction w ith He+ ions produced in a rem ote hollow cathode d.c.
discharge. The ionized am m onia was then allow ed to undergo clustering reactions. The
pressure regim e em ployed was ~9 mbar, w hich is significantly higher than m ost present
studies.
The mechanisms responsible for these clustering reactions are com plex and often
involve a third party, such as O 2 and N 248. The reaction forming the cluster at mass 35 has
27
been reported to have a reasonably high rate constant o f the order o f 3 ,4x 10"
6
cm m ol'
2
s '1, and that O 2 is a suitable third party.
For purposes o f the present work, it is important to note that studies o f the plasm a
decom position o f am monia 5 1,52,53 suggest that -N Ff radical is the m ost likely Ncontaining reactive species present in the plasm a. It is logical to conclude that these
radicals could be the source o f surface functional groups such as -N H 2 on the polym er
surface.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
2.5.2 Reaction on the polymer surface during plasma treatment
The generally accepted theories o f plasm a surface m odification 54 are: 1) the
mechanical interlocking theory, which is frequently called the “anchor effect” ; 2 ) the
electronic or electrostatic theory; 3) the diffusion theory; 4) theories based on surface
energy, wetting and adsorption; 5) chemical bonding; and 6 ) the w eak boundary layer.
However, from the reaction mechanism viewpoint, surface reactions during plasm a
treatm ent can be categorized as: induced by free radicals to form new chem ical bonding;
induced by w eak boundary layer to form crosslinking structures; and induced by photon
to degrade the surface. The property changes o f a polym er surface upon plasm a treatm ent
can attribute to the m echanism s presented above w hich can occur at the surface or
interfacial region either in isolation, or m ost likely, in com bination to produce the final
bonds on the surface. The subsequent section below is the literature review corresponding
to these three categories.
o
p
i n
j q
M any researchers ' ’ ’ have reported that am m onia plasm a treatm ents are
effective in m aking PTFE surfaces hydrophilic. They all agreed that since the C-F bond is
considerably m ore stable than C-C bonds, the predom inance o f chain scission over
fluorine abstraction during plasm a treatm ent w ould be anticipated. Thus, the facile
incorporation o f nitrogen to the PTFE surface via flourine abstraction w ith am m onia
plasm a exposure was surprising. K aplan et al . 12 investigated w hether plasm a
polym erization would provide a modified adherable layer to PTFE. The rapid plateauing
in the am m onia activation study supported the hypothesis that an equilibrium is rapidly
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
32
achieved in w hich backbone chain scission elim inates ‘activated or m odified’ chain
segments almost as soon as they are formed.
For perfluorinated polymers, G engenbach et al.
suggested a fluorine abstraction
m echanism. By observing that the am ount o f fluorine abstracted w as considerable larger
than the am ount o f nitrogen incorporated, the authors assum ed that the first step in the
am m onia plasm a m odification is the hem olytic cleavage o f C-F bonds. Thus, the
difference between fluorine abstraction and nitrogen incorporation suggested that a
substantial fraction o f the carbon-centered radicals created by the plasm a exposure didn’t
react w ith an am monia molecule, or a nitrogen-containing fragm ent thereof, arriving
from the plasm a phase. Some o f the rem aining radicals reacted w ith atm ospheric oxygen.
The m ajority o f the radicals presum ed to have been created by C-F bond scission during
the plasm a treatm ent underwent reactions other than addition o f an extraneous species.
Possible reactions were the com bination o f radicals to form crosslinks, and the surface
recom bination o f radicals which form double bonds .55 A nother assum ption is that a
substantial fraction o f C‘ radicals can undergo other reactions before oxygen addition
occurs, and the form ation o f additional crosslinks by the com bination o f neighboring
radicals would occur relatively rapidly given the high mobility o f perfluorocarbon
polym er segments. Also another reaction sequence that m ay proceed sim ultaneously to
reduce the F/C ration is the decom position o f m etastable peroxides into alkoxy radicals,
w ith the subsequent liberation o f small, volatile fragments rich in fluorine, such as by a
radical-shift reaction:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
33
O
•
'
XX0
'ma/wwa/w'CF2— C F — CF2 — CF2
>
I
>www w w .Q p — q
+
\
CF3
C h 2- O r 2
c f :3
Figure 2-2 Radical shift reaction during am m onia plasm a treatment.
Sprang et a l . 19 also studied the fluoropolym ers reaction m echanism during Ar, N 2
and 0 2 plasm a treatment. For perfluoroalkoxyvinyl ether (PFA), they proposed the
following reactions in the case o f A r or N 2 plasm a treatment.
i)
c 3f 7
I
0
o'
1
.
CF2 — CF — CF2 —
----------->
I
— CF2— C F — CF2—
0
1
— C2F5— CF2
+
— C2F5— CF2
O
II
----------->
— C2F5— C — F
+
F
ii)
C 3 F7
0
1
o'
I
- C F 2 — C F — CF2 —
> — CF2 — C F — CF2 —
o'
c f — c f 2— —
C3F7
+
c f 2—
o
I
— c f 2—
+
II
>
—
c f 2— c — f
Figure 2-3 Two possible reaction paths o f PFA by A r or N 2 plasma.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
34
In the case o f O 2 plasm a treatm ent, they proposed the direct oxidation pathw ays
as follows:
a)
CF 2— CF2—
--------> _ _ C F — q f 2—
+
b)
0
1
— CF — C F ,— +
O
-> — CF — CF,
c)
0
1
— CF — CF2—
O
II
------- > — C — F
+
CF2—
Figure 2-4 Possible direct oxidation o f PFA by O 2 plasm a treatment.
A sim ilar reaction path is reported for PTFE exposed to ionizing radiation . 56
N itrogen plasm a treatm ent yields an acid fluoride peak (also observed with argon
plasm a). Therefore, it was assumed that there w ere also additional ways to generate the
acid fluoride end group. Possibly, the nitrogen plasm a produced nitrogen-containing
groups at the surface o f the PFA, w hich subsequently reacted w ith atm ospheric oxygen.
A similar reaction is observed for plasm a treatm ent o f polyethylene.
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35
However, Bhat and U padhyay
reported that no crosslinking on am m onia
plasm a-treated PP surface was observed and explained that am m onia contributed m ainly
to incorporating groups such as -N H 2 , -N H , and -N O 2 .
For argon plasm a treatm ent o f a polym er fdm , m any researchers 21’58’59 reported
that the argon plasm a treatments lead to surface oxidation and thus to an increase in the
w ettability o f that surface. The surface oxidation can be explained by potential sources:
1) The argon plasm a introduces radicals at the polym er surface, w hich rapidly react w ith
oxygen or nitrogen when the plasm a-treated surfaces are exposed to air57; 2) A lthough all
possible precautions are taken, an oxygen source like a small air leak or adsorbed w ater
on the reactor walls or the oxygen in the polym er itself m ight be present during the
plasm a treatments. In this case, not only radical formation, but also surface oxidation will
occur during the plasm a treatment.
V allon and Drevillon
T1
proposed three possible reaction m echanism s for argon
plasm a treatm ent on PC (polycarbonate) surface (shown in Figure 2-5). The first
m echanism is due to the excitation o f the polym er by “short-w avelength” photons (254
nm), the absorption band o f PC being centered at 265 nm. A fter carbonate bond breaking,
three types o f reactions occur sim ultaneously. First, two successive photo-Fries
rearrangem ents occur: the phenylcarbonate units rearrange into phenylsalicylate unit (Li),
that in turn rearrange into dihydroxybenzophenone groups (L 2 ). Concurrently, the
radicals originating from the carbonate breaking can recom bine after decarbonylation or
decarboxylation, leading to chain fracturing (second reaction) or to generating various
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
photoproducts L 3 (third reaction). The second m echanism , w hich is first described by
Rivaton et al .60 for the photodegradation o f PC, refers to the effects induced by defects or
impurities o f the material. By absorption o f “long-w avelength” photons (365 nm), R'
radicals can be produced. These radicals react with PC by rem oving a hydrogen atom
from a methyl group, and this new radical isom erizes into a tertiary m acroradical R 3 to
increase its stability. R 3 may oxidize in the presence o f oxygen. The third possible
mechanism is the abstraction o f methylgroups w hich m ay produce radicals and lead to
crosslinking, as w as already observed for PP (polypropylene) in a nitrogen plasm a
23
and
in an argon plasm a31. The existence o f this mechanism is supported by the fact that the
plasm a contains photons o f higher energy than the short-w avelength photons (254 nm)
used by Rivaton et al.60. These higher energy photons can directly break C-C and C-H
bonds without the presence o f impurities.
l.
a
r
r
A
hv
photo-F ries
chain b r e a k in g ^
OH
L,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37
b
A bsorbing impurities
hv
--------- ^
R
CH*
CH,
I
O -C -O -
II
O
CH,
>
+RH
ch3
o
isomerisation
o
R3
c
CH,
ch3
o
ch3
o
Figure 2-5 R eaction m echanisms o f PC during exposure to UV light: a) direct
photodegradation mechanism; b) production o f macroradicals (R 3) through the induced
photodegradation mechanism; c) abstraction o f m ethyl groups.
U rban and Stewart 61 combined infrared spectroscopy (ATR-mode) and dynam ic
m echanical analysis to study the effects o f single argon, carbon dioxide and am m onia
plasm a treatm ent (50 W, 0.48 Torr, 10 min) on industrial-grade silicone rubber. D uring
the am m onia plasm a treatm ent for silicone rubber, the authors observed higher glass
transition temperature and storage module, w hich were interpreted by increased
crosslinking o f the “weak boundary layer” (W BL) at the surface. A lso in accordance w ith
the present study, U rban and Stewart 61 concluded that a single carbon dioxide plasm a
treatm ent causes chain cleavage. However, no indications for crosslinking o f argon
plasm a-treated silicon rubber were reported, w hich is in contrast to Everaert et al . 16 who
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
38
dem onstrated that repeated treatm ents o f silicone rubber w ith argon and am m onia
plasm as lead to enhanced crosslinking in the surface layers o f elastom er. They speculated
about the pathw ays by w hich silicone rubber is m odified during and after repeated
plasm a treatm ent, w hich are shown in Figure 2-6.
(a) Crosslinking in medical grade silicone rubber
Si— O — Si— O — Si— O
CH 3
ch3
\ CH2/
Si— O —
S i—
CH 3
ch3
O — Si— O
ch3
(b) Crosslinking due to radical recom bination
ch3
Si— O
ch3
Si— O
9 h3
Si— O
CH 3
ch3
ch3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
(c) Crosslinking due to (poly) condensation
CH'
CH'
CH 3
I
Si— O
-
—
-
1-
/ X
\
?
'
/
/
\ k
/
\? y
-
- .... Si - o ..... 6h!
C H 3
Si—O
i
---
/ p H ''
: i
'
'4 -
f - °
Si - 0
H -'
I
'>
•
■
' O
\ I
;
C H 2
;
'
f O
ch3
Figure 2-6 Proposed plasm a-induced crosslinking between siloxane chains in silicon
rubber.
For nitrogen plasm a, Foncin-Epaillard et al .23 suggested an exom ethylene double
bond m echanism, w hich caused a slight degradation o f PP. This behavior is not
consistent w ith that obtained by O 2 or CO 2 plasm a treatments. The loss o f m ethyl groups
m ay occur by the following reaction:
CH3
CH2
1
CH
CH2
-CH
_ UM3>
— CH2 — CH — CH2—
Figure 2-7 M ethyl group elim ination reaction in PP by N 2 plasm a treatm ent.
After elim ination o f CH3, the m onom er unit having this radical is sim ilar to a PE
(polyethylene) unit, w hich could lead to branching under irradiation .62,63,64 Indeed,
exom ethylene double bonds have already been observed in A r or O 2 plasm a treatm ents .65
A n alternative crosslinking mechanism during plasm a treatm ent o f polyolefm es was
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
40
proposed by Clark et al ,66 where double bonds form ed through H -atom free radical
abstraction (Figure 2-8) seem to initiate the reaction.
CH,
-CH2 — CH — CH2— .
CH,
- H
— CH, — CH — CH,
ch2
H
II
ch2
c —
c h 2-
Figure 2-8 Radical induced double bond form ation during plasm a treatm ent o f
polyolefines.
Similarly, Bhat and Upadhyay 22 gave the following reaction sequence for PP
surface m odification using N 2 plasma.
CH,
VW W W W W i
CH,
3 min
C H — CH- AW vW /vW VS
>
CH— X
•*« *« * C H — CH2yWWWWvWN
C H — CH-
vW A V A W A N
N2 plasma
CH— X
I
WAWMW/i C H — CH2A W M W W .
CH,— X
CH,— X
.
10
mir^ AWvWAWt Q __
/M W iW M
Q
____
N2 plasma
WVWVWVvW.
0
M W AW M
I
X — CH2
Figure 2-9 Reaction sequence for PP surface m odification using N 2 plasma.
Inagaki et a l 27 investigated nitrogen and argon plasm a treatm ent on polyim ide
surface and concluded that, in a chemical sense, plasm a treatm ent is a radical-substitution
reaction o f C-H bonds in polymers. They also proposed a possible reaction process that
could form the carboxyl and the secondary amide groups in the plasm a-treated polyim ide
film, w hich is a com bination o f the cleavage o f im ide groups and the subsequent reaction
w ith w ater as shown in the figure below.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
- n/ cy
§\
v
0
^
vo_
/ ^
S
0
o
Figure 2-10 Possible overall reaction in the plasm a-treated polyim ide film.
2.6 Immobilization Reaction Strategy
Several research groups 67,68’69 have reported a reductive am ination m ethod for
im mobilizing dextran to am inated plasm a surfaces and observed reduced cell adhesion on
dextran-coated substrates. The schematic depiction o f the form ation o f the covalent
interfacial amine linkage is shown in Figure 2-11. Dextran is activated by oxidation o f
the glucose subunits w ith sodium metaperiodate to convert glucose subunits to cyclic
hem iacetal structures. The polym er surface is activated by am m onia plasm a surface
modification. O xidized dextran is then reacted w ith surface-bound am ines on the plasm atreated polym er surface.
O o
p
H
>
~
0
'
■CD
.H c S
°
^
oh'
OH
“ ° V
_
1
HY
0
+ H 20
-
-„,o
0
- ° V
HOO
H
~“C-/- ‘V o
-
]l
^ 0
-
''OH
nh2
nh2
V x x T t + H20
CD
=
d
e
x
t
r
a
n
b
a
c
k
b
o
n
e
C D —° \
C D —o
f
NH
h
2
o
NaCNBH,
[
NH
^
N
° \ _
0^ C
1|
N
Figure 2-11 Reaction scheme for the covalent im m obilization o f periodate-oxidized
dextran onto plasm a-am inated surfaces.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
D
42
Researchers70,71 further reported that am m onia and oxygen plasm a-treated
polyethyleneterephthalate (PET), polytetrafluoroethylene (PTFE) and fluorinated
ethylene-propylene copolymer (PEP) surfaces can im prove the biocom patibility o f
hum an endothelial cells.
2.7 Aging Effect Study
In order to promote applications o f polym ers in biom edical devices, m any
researchers have investigated the reproducibility and stability o f the plasm a m odified
*70*70Q **7/1 *7C
surfaces. ’ ’ ’ ’ It has been established that tw o principal m echanism s occur: chain
relaxation processes and further chemical reactions w ith the environm ent74. A m m onia
plasm a-treated polym ers invariably contain oxygen, even after careful leak testing and
pum ping dow n o f the reactor. Oxygen incorporation increases w ith tim e as the samples
w ere stored in air after treatments. The fast, initial oxygen uptake can be attributed to the
atm ospheric oxygen reaction at carbon-centered radicals created in the surface and subco
n£nn
yo
surface layers by the plasm a exposure. Previous studies ’ ’ ’ have shown that radicals
created by plasm a exposure can react with atmospheric oxygen, following venting o f the
plasm a chamber. The addition o f O 2 to carbon-centered radicals is a very fast process,
and oxygen can diffuse rapidly through the top few nanom eters o f the polym er surface.
G erenser et al.57 have reported substantial post-treatm ent oxygen incorporation into
polyethylene after only 30 s exposure to the atmosphere.
Researchers79,80,81 have reported that the slow, extended oxygen uptake is likely
due to secondary reactions analogous to those involved in the oxidative degradation o f
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
polyolefin, w hich proceeds by several radical reactions. G engenbach et al.
9&
have show n
that the gradual decay o f m etastable peroxides can m aintain oxidative reaction cycles
over extended periods o f time, thus producing a continuing, long-term oxygen uptake. A
corresponding study o f the long-term oxidative reactions in a fluorocarbon polym er after
plasm a treatm ent showed that the reaction pathways and cycles involved are com plex.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
References
1
Kaplan, S. L.; Rose, P. W. Int J Adhes Adhes 1991, 11, 109.
2
Sipehia, R.; Chawla, A. S. Biom ater M ed Dev A rtif Organs 1982, 10, 229..
3
M utlu, M .; M utlu, S.; Rosenberg, M. F.; Kane, J.; Jones, M. N.; Vadgama, P. J M ater
Chem 1991, 1,447.
4
Li, Z.-F.; N etravali, A. N .; Sachse, W. J M ater Sci 1992, 27, 4625.
5
Brown, J. R.; M athys, Z. J M ater Sci 1997, 32, 2599.
6
Muller, M.; Oehr, C. 14th Int. Symp. on Plasm a Chem ., Prague, Aug. 2-6, 1999; p
1791.
7
Chang, T. C. J Ind Technol 1998, 15.
8
Badey, J. P.; Urbaczewski-Espuche, E.; Jugnet, Y.; Sage, D.; Due, T. M .; Chabert, B.
Polym 1 9 9 4 ,3 5 ,2 4 7 2 .
9
Pringle, S. D.; Joss, V. S.; Jones, C. Surf Interface A nal 1996, 24, 821.
10 Pu, F. R.; W illiams, R. L.; M arkkula, T. K.; Hunt, J. A. Biom ater 2002, 23, 2411.
11
Chevallier, P.; Castonguay, M.; Turgeon, S.; Dubrulle, N .; M antovani, D.; M cbreen,
P. H.; W ittmann, J.-C.; Laroche, G. J Phys Chem B 2001, 105, 12490.
12 Kaplan, S. L.; Lopata, E. S.; Smith, J. Surf Interface A nal 1993, 20, 331.
13
Janca, J.; Stahel, P. Czech J Phys 2000, 50, 449.
14 L'argentiere, P. C.; Taher, H. A. J Chem Technol Biotechnol 2000, 75, 25.
15
Larocca, P.; Barker, S. Cell Line 1996, 6 , 1.
16
Everaert, E. P.; Chatelier, R. C.; V an Der Mei, H. C.; Busscher, H. J. Plasm as Polym
1997, 2 ,4 1 .
17
Husein, I. F.; Chan, C.; Chu, P. K. J M ater Sci Lett 2000, 19, 1883.
18
Petasch, W.; Rauchle, E.; W alker, M.; Eisner, P. Surf Coat Technol 1995,74-75,
682.
19
Sprang, N .; Theirich, D.; Engemann, J. Surf Coat Technol 1998, 98, 865.
20
Li, V. C.; Wu, H. C. U.S. Patent No. 5,788,760, Issued on Aug. 4, 1998
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
45
21
Terlingen, J. G. A.; Brenneisen, L. M .; Super, H. T. J.; Pijpers, A. P.; H offm an, A.
S.; Feijen, J. J Biom ater Sci, Polym Ed 1993, 4, 165.
22 Bhat, N. V.; Upadhyay, D. J. J Appl Polym Sci 2002,
86,
925.
23
Poncin-Epaillard, F.; Chevet, B.; Brosse, J. C. J A dhes Sci Technol 1994, 8 , 455.
24
Collaud, M .; N owak, S.; Uttel, O. M. K. F.; Schlapbach, L. J A dhes Sci Technol
1994, 8 , 435.
25 Andre, V.; Arefi, F.; Amouroux, J.; Puydt, Y. D.; Bertrand, P.; Lorang, G.; Delam ar,
M. Thin Solid Films 1989, 181, 451.
26 Egitto, F. D.; M atienzo, L. J.; Blackwell, K. J.; K noll, A. R. J A dhes Sci Technol
1994, 8 ,4 1 1 .
27 Inagaki, N .; Tasaka, S.; Hibi, K. J A dhes Sci Technol 1994, 8 , 395.
28
Gengenbach, T. R.; Xie, X.; Chatelier, R. C.; Griesser, H. J. J A dhes Sci Technol
1994, 8 , 304.
29 Xie, X.; Gengenbach, T. R.; Griesser, H. J. J Adhes Sci Technol 1992, 6 , 1411.
30 M ason, M .; Vercruysse, K. P.; Kirker, K. R.; Frisch, R.; M arecak, D. M.; Prestw ich,
G. D.; Pitt, W. G. Biom ater 2000, 2 1 ,3 1 .
31
Vallon, S.; Drevillon, B.; Poncin-Epaillard, F.; K lem bergsapieha, J.; M artinu, L. J
Vac Sci Technol A 1996, 14, 3194.
32
Owen, M. J.; Smith, P. J. Polym Surf M odifi: Relev to A dhes K L M ittal (Ed) 1995,
3.
33
Yang, J.; Bei, J.; Wang, S. Polym A dv Technol 2002, 13, 220.
34
Creatore, M.; Favia, P.; Tenuto, G.; Valentini, A.; D'agostino, R. Plasm as Polym
2000, 5 ,2 0 1 .
35
Gerenser, L. J.; Grace, J. M.; Apai, G.; Thom pson, P. M. S urf Interface A nal 2000,
29, 12.
36 Veprek, S. Plasm a Chem Plasm a Process 1989, 9, 29.
37 Puchkarev, V.; Gundersen, M. Appl Phys Lett 1997, 71, 3364.
38 Hopwood, J.; Guarnieri, C. R.; W hitehair, S. J.; Cuomo, J. J. J Vac Sci Technol A
1993, 11, 152.
39 Bosw ell, R. W.; Porteous, R. K. Appl Phys Lett 1987, 50, 1130.
40
Teperm eister, I.; Ibbotson, D. E.; Lee, J. T. C.; Sawin, H. H. J Vac Sci Technol, B
1994, 12, 2322.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
46
41
M ahoney, L. J.; W endt, A. E.; B arrios, E.; R ichards, C. J.; Shohet, J. L. J A ppl Phys
1994, 76, 2041.
42
Cox, T. I.; Deshmukh, Y. G. I.; Hope, D. A. O.; Hydes, A. J.; Braithw aite, N. S. J.;
Benjamin, N. M. P. J Phys D: A ppl Phys 1987, 20, 820.
43
Cheng, C. C.; Guinn, K. V.; Donnelly, V. M .; Herm an, I. P. J V ac Sci Technol A
1994, 12, 2630.
44
Ibbotson, D. E.; Chang, C. P. Dry Etch Technology 1991, 1593, 130.
45
Shahidzadeh, N .; Arefi-Khonsari, F.; Chehimi, M. M.; A m ouroux, J. S urf Sci 1996,
352, 8 8 8 .
46 Norm and, F.; M arec, J.; Leprince, P.; Granier A. M ater Sci Eng A 1991, 139, 103.
47
Egitto, F. D.; M atienzo, L. J. Polym D egrad Stabil 1990, 30, 293.
48
Knewstubb, P. F. M ass Spectrom etry & Ion-m olecule Reactions; Cam bridge
U niversity Press, 1969.
49
W ojcik, L.; Bederski, K. Int J M ass Spectrom Ion Processes 1993, 127, 11.
50
Glosik, J.; Zakouimageil, P.; H anzal, V.; Skalsky, Y. Int J M ass Spectrom Ion
Processes 1995, 149/150, 187.
51
Allred, R. E.; M errill, E. W.; Roylance, D. K.; Ishida, H.; Kumar, G. Ed.; Plenum
Press, N ew York, 1985, pp 333.
52
Captelli, M.; M olinari, E. in Topics in Current Chemistry; Springer-V erlag, N ew
York, 1980, pp 59.
53
D'agostino, R.; Cramarossa, F.; Debenedictis, S.; Ferraro, G. Plasm a C hem Plasm a
Process 1981, 1, 19.
54
Buchwalter, L. P. J Adhes Sci Technol 1990, 5, 697.
55
Youxian, D.; Griesser, H. J.; Mau, A. W.-H.; Liesegang, R. S. Polym 1991, 32, 1126.
56 Fisher, W. K.; Corelli, J. C. J Polym Sci, Part A: Polym Chem 1981, 19, 2465.
57
Gerenser, L. J. J Adhes Sci Technol 1987, 1, 303.
58
Yasuda, H.; M arsh, H. C.; Brandt, E. S.; Reilley, C. N. J Polym Sci, Part A: Polym
Chem 1977, 15, 991.
59
Grant, J. L.; Dunn, D. S.; M cclure, D. J. J Vac Sci Technol A 1988, 6 , 2213.
60
Rivaton, A.; Lemaire, J. Polym er D egradation and Stability 1988, 23, 51.
61
Urban, M. W.; Stewart, M. T. J Appl Polym Sci 1990, 39, 265.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
62 Dunn, T.; W illiam s, E.; W illiams, J. Radiat Phys C hem 19 8 2 ,1 9 , 287.
63
Zhu, Q.; Horil, F.; Kitamaru, R.; Yamaoka, H. J Polym Sci, Part A: Polym Chem
1990, 28, 2741.
64
Seguchi, T.; H ayakaw a, N.; Tamura, N.; H ayashi, N .; K atsum ura, Y.; Tabata, Y.
Radiat Phys Chem 1989, 33, 119.
65
Occhiello, E.; Garbassi, F.; M orra, M. Surf Sci 1989, 211/212, 218.
66
Clark, D. T.; Dilks, A.; Schuttleworth, D. in The application o f Plasm a to the
Synthesis and Surface M odification o f Polymers; John W iley, N ew York, 1986.
67 M eng, W.; H yun, J.-Y.; Song, D.-I.; Kang, I.-K. J A ppl Polym Sci 2003, 90, 1959.
68
M assia, S. P.; Stark, J. J Biom ed M ater Res 2001, 56, 390.
69
Dai, L.; Stjohn, H. A. W.; Bi, J.; Zientek, P.; Chatelier, R. C.; Griesser, H. J. Surf
Interface A nal 2000, 29, 46.
70 Ram ires, P. A.; M irenghi, L.; Romano, A. R.; Palum bo, F.; N icolardi, G. J Biom ed
M ater Res 2000, 51, 535.
71
Griesser, H. J.; Chatelier, R. C.; Gengenbach, T. R.; V asic, Z. R.; Johnson, G.;
Steele, J. G. Polym Int 1992, 27, 109.
72
Occhiello, E.; M orra, M.; Morini, G.; Garbassi, F.; H um phrey, P. J Appl Polym Sci
1991,42, 551.
73
Johannes, G.; Terlingen, A.; Hermina, F.; Gerritsen, C.; H offm an, A. S.; Feijen, J. J
A ppl Polym Sci 1995, 57, 969.
74 X ie, X.; Gengenbach, T. R.; Griesser, H. J. in Contact Angle, W ettability and
Adhesion; Good, R. J. Ed.; Elsevier, Amsterdam, 1993, pp 509.
75
Chatelier, R. C.; Xie, X.; Gengenbach, T. R.; Griesser, H. J. Langm uir 1995, 11,
2576.
76
N akayam a, Y.; Takahagi, T.; Soeda, F.; Hatada, K.; N agaoka, S.;Suzuki, J.; Ishitani,
A. J Polym Sci, Part A: Polym Chem 1988, 26, 559.
77
Chappell, P. J. C.; Brown, J. R.; George, G. A.; W illis, H. A. S u rf Interface Anal
1991, 17, 143.
78
Everhart, D. S.; Reilley, C. N. Anal Chem 1981, 53, 665.
79 Lazar, M.; Rychly, J. Adv Polym Sci 1992, 102, 189.
80
Gugumus, F. Polym Deg Stab 1990, 27, 19.
81
Lacoste, J.; Carlsson, D. J.; Falicki, S.; Wiles, D. M.Polym D eg Stab 1991, 34, 309.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 3
EXPERIMENTAL
METHODS
3.1 Overview
Although the effects o f am m onia plasm a surface m odification have been
investigated by many researchers, no studies have yet to report on dow nstream am m onia
plasm a (with or without argon plasm a pretreatm ents) surface m odification o f Cyclotene,
nor have researchers explored the synergetic effects o f all processing variables on the
am ination degree and controlling reaction mechanisms. The prim ary experim ental goals
for this research were to identify the preferred operating conditions to graft am ine groups
onto Cyclotene surfaces, and to provide a firm foundation for theoretical m odels to be
subsequently proposed. In this chapter, efforts were also made to correlate the operational
param eters to cell adhesion performance on the plasm a-treated Cyclotene surfaces.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
This chapter describes the apparatus, m ethods and experim ental design for the
investigation o f the im pact o f im portant operational param eters (power, pressure, tim e
and temperature) on the degree o f surface am ination and for the study o f aging behavior
o f plasm a-treated surfaces in three different storage media. B oth D esign o f Experim ent
(DoE) and one-factor-at-a-tim e (OFA) experiments were applied for the experim ental
designs. Several com plem entary surface characterization techniques (XPS, A TR and
A FM ) and biocom patibility tests that were used to quantify the am ination effects are also
described.
3.2 Plasma System
The dow nstream plasm a reactor, shown in Figure 3-1, consists o f a 2.54 GHz
plasm a source (A STEX AX2000) 250 W m icrowave pow er generator, 4 ” stainless steel
cham ber and the pum ping system (a Pfeiffer/Balzers turbo pum p TP180H backed by an
Edw ards E1M -18 tw o-stage mechanical pump). Typical base pressures are less than
7x10-7 Torr. Samples are m echanically mounted on an alum inum block. The argon and
am m onia gases are introduced into the reactor through individual m ass flow controllers
(PFD-301).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
Figure 3-1 A ctual picture o f the microwave plasm a system em ployed in this study.
A schem atic diagram o f the down stream reactor apparatus is shown in Figure
3-2. The plasm a source is attached to a port on a 4” stainless steel cham ber by a 1” quartz
tube. The distance d between the sample and the center o f the discharge is 24 cm. In this
way, all o f the samples are downstream treated. The gas flows are controlled by using
m ass flow controllers over the
1 -2 0 0
seem range.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
microwave
generator
glow discharge
cham ber
p ressu re gau g e
rough
p ressu re gau g e
(Tc)
( Tc)
plasm a reactor
'glow
Wv
throttle valve
roughing valve
plasm a source
/*)
ionization gauge
foreline
pressu re gauge
temp
controller
high vacuum
valve
turbo pump
mechanical
pump
It
Ml
W v
foreline valve
exhaust
N2
N
H
3
0
2
A
t
Figure 3-2 Schematic o f the experim ental setup for the plasm a system.
The m ajor advantage o f this type o f system is the elim ination o f highly energetic
excited state ions and the associated ion-induced reactions, including potential sputtering
or other dam age on the exposed surfaces to be treated. In the flowing afterglow, the
substrate is exposed only to relatively long lifetime neutral species (at least m icrosecond
lifetim es) and photons, and so only these species are expected to react w ith polym er
surfaces. Undesirable polym er degradation reactions can be particularly troublesom e
w hen the substrate is directly in the plasm a because o f ion or/and potentially electron
bom bardm ent. Therefore, whenever possible, the substrate is positioned in the flowing
afterglow o f the discharge in order to realize benefit from the functionalization by lower
energy neutral species while minimizing ion-induced degradation and surface
dam age . 1,2’3’4 M oreover, it is more straightforward to distinguish and interpret the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
observed results since the num ber o f potentially active species and reaction pathw ays is
much lower.
3.3 Sample Preparation
Because organic contaminants may adhere to the substrate surface, the sam ple
(Si/Si 0 2 substrate) to be coated w ith Cyclotene is first cleaned in a Plasm aLab pP80
reactive ion etcher for 5 m inutes at 50 watts using 50 seem O 2 at 100 m T orr total
pressure. Im m ediately after cleaning, the sample is placed into a Specialty Coating
Systems model P-7608D program m able spin coater, and D ow A P3000™ adhesion
prom oter is dispensed onto the sample surface using a clean, glass dropper bottle. Enough
A P3000™ was applied to cover the entire surface (typically about 2-3 m L for a onefourth slice o f a 4” wafer). The sample is spun at 800 rpm for 30 seconds to spread the
adhesion prom oter, followed by a linear ramp to 2000 rpm over 10 seconds. The sample
is then spin-dried at 2000 rpm for 30 seconds.
After adhesion prom oter application, the prew arm ed Cyclotene (D ow 4026-46) is
dispensed onto the center o f the sample surface; for the one fourth sam ple size m entioned
above, a typical dispense volum e is 3-5 m L o f Cyclotene. The spinner bow l cover is put
in place, and the spin program used is as follows: linear ram p to 800 rpm over
seconds; 800 rpm “spread” for
10
seconds; linear ramp to
2000
rpm over
10
10
seconds;
2000 rpm “spin” for 30 seconds; linear ramp to 0 rpm over 10 seconds. A typical postsoft-bake thickness o f 13 pm is achieved using the
2000
rpm spin speed.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
53
After spin-coating the Cyclotene, residual m esitylene is driven from the film by
soft baking in an N 2 -purged convection oven at 70-80°C for 20 minutes. Then the sam ple
is transferred to a Therm co M B-80 M inibrute furnace, w here it is purged w ith N 2 at room
tem perature for one hour to provide an inert atm osphere in the furnace; this purge is
necessary to prevent oxidation o f the film during curing. A fter the one-hour purge, the
sample is cured in the inert atmosphere by rapidly raising the tem perature to either 210°C
for 40 m inutes (partial cure for 1st Cyclotene layer) or 250°C for 60 m inutes (full cure for
2nd Cyclotene layer). After the required cure time, the heater is turned o ff and the sam ple
was allowed to cool for several hours to room tem perature while still in the inert
atmosphere. Finally the whole w afer is cleaved into small square pieces (3x3 cm ) and
ready for plasm a treatment.
3.4 Characterization Techniques
Due to the complexity o f the possible surface reactions w ith plasm a treatm ents, a
reasonably com prehensive understanding o f the surface chem istry requires the use o f
several com plementary surface-sensitive characterization techniques. In this study, the
surface o f the plasm a aminated Cyclotene films was characterized w ith a com bination o f
the follow ing surface-sensitive spectroscopies: XPS (X-ray Photoelectron Spectroscopy),
FTIR -A TR (Attenuated Total Reflection Fourier Transform Infrared Spectroscopy) and
A FM (A tom ic Force M icroscopy). Each o f these techniques is described in the
subsections that follow.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
54
3.4.1 XPS
XPS analyses w ere perform ed using a K ratos X SA M 800 system . The instrum ent
includes an ion-pum ped sample analysis cham ber and a turbopum ped sample
introduction chamber. The base pressure was 2x10-1 0 Torr and the pressure w as 10“ 9 Torr
while acquiring data. A m onochrom atic Mg Ka x-ray source (240 W ) w as used and the
take-off angle w as 90° unless otherwise stated. H igh-resolution spectra were accum ulated
at 0.05 eV intervals and after a Shirley-type background subtraction, the com ponent
peaks were separated using an in-house, nonlinear least m ean squares program. This
program uses peak w idths and atom ic sensitivity factors previously determ ined on know n
pure materials. The elemental com position o f the surface was determ ined based on a first
principles approach; atomic ratios were calculated from integral peak intensities using a
non-linear Shirley-type background and published values for photo-ionization cross
section.
3.4.2 FTIR-ATR
For investigations o f biological polymers, the functional group specificity o f
infra-red spectroscopy profits substantially if coupled to a surface sensitive technique.
A TR (attenuated total reflection) represents such a surface sensitive technique. In this
study, the infrared spectra were m easured with a Fourier Transform Infrared (FTIR)
spectrom eter (Bruker, IFS
6 6 V/S)
in the ATR mode to obtain surface sensitive IR-
spectra. Typically 1024 scans w ith a resolution o f 4 cm ' 1 were taken and averaged to get
one spectrum. The samples were brought into contact w ith one side o f the internal
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
55
reflection elem ent (IRE) germanium (Ge) using an adjustable spring to ensure
approxim ately the same pressures for all samples.
Unfortunately, quantitative A TR m easurem ents are difficult, because a non-ideal
contact between the IRE and the sample is very com mon. The signal intensity therefore
depends on m any factors, which can not be controlled quantitatively, such as the sam ple
roughness, contact area, the applied pressure, etc.
W ith the intention o f investigating the m odification depth o f the surface, w e need
to consider the penetration depth. The penetration depth dp depends on the radiation
w ave-length A, the refractive index o f the IRE np and the sample ns, and the angle o f
incidence o f the beam a.5
j
A
P
(
■
2
2 V /2
27m p (sm a - n sp)
where nsp=njnp. For Cyclotene, « s= l .543, and for Ge, np=4.0 at 1000cm '1, then
the penetration depth dp ranges from 0.17 pm at 4000 cm ’1 to 1.12 pm at 600 c m '1.
3.4.3 AFM
The surface o f a polym er used in a m edical device is often the interface betw een
the body and the device. By controlling the surface properties o f the polym er, the m edical
device designer can enhance or inhibit various reactions o f the body to the device. The
interaction betw een the polymer and its environm ent depends in large part on surface
com position and structure. Plasm a treatm ent can produce the desired surface
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
56
characteristics and topographies. Since surface com position and topography play such an
im portant role in the perform ance o f the polym er, precise surface characterization can be
a large part in the rapid deploym ent o f new m aterials or understanding o f problem s and
behaviors in existing materials. AFM , w ith its high-resolution surface m apping
capabilities, can be a key com ponent o f that characterization.
A N anoScope E M ultiM ode AFM (Veeco Digital Instrum ents, Santa Barbara,
CA) and standard AFM cantilevers (Veeco D igital Instruments, Santa Barbara, CA) w ith
pyram idal Si 3N 4 tips were used to image surface structures o f Cyclotene samples. The
spring constants o f the cantilevers em ployed for im aging w ere 0.12 N ew tons per m eter as
reported by the manufacturer. The samples were im aged in air at room tem perature. A
2m m by 2m m area was scanned at a rate o f 4 Hz. The integral and proportional gain
values were set to 2.0 and 3.0, respectively.
3.5 Microwave Plasma Treatment
M icrowave plasm a treatm ent took place at 2.45 GHz in the system described
previously. Gases used for plasm a treatm ent w ere o f com m ercial variety [NH 3 purity:
class 4.5, M atheson gases; Ar purity: class 5, LIQUID A IR corp.]. For am m onia plasm a
treatm ent, the plasm a was m aintained at a pow er from 50 to 250 W, total pressure from
0.2 to 0.6 Torr, an am monia flowrate from 50 to 110 seem, a treatm ent tim e from 60 to
300 s, and a tem perature from room temperature to 200 °C. For argon plasm a
pretreatm ent, the experiment was conducted at a power o f 150 W, total pressure o f 0.4
Torr, argon flowrate o f 35 seem, treatm ent time o f 180 s, and a tem perature o f 60 °C.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
57
After the one-step or tw o-step (Ar plasm a pretreatm ent plus am m onia plasm a treatm ent)
surface treatment, the samples were im m ediately transferred to the “sm art desiccator”
w ith nitrogen gas flowing inside so that the possibility o f oxidation and w ater uptake can
be minimized. Then the samples were ready to be characterized using X PS, FTIR -A TR
and AFM.
3.6 Experimental Design for Plasma Treatment
In this study, we experim entally investigated the efficiency o f surface
modification o f Cyclotene using am m onia plasm a w ith or w ithout argon plasm a
pretreatm ents. A “one variable at a tim e” approach for experim entation w ould require a
large number o f experiments and yet may not provide a com plete picture. In contrast,
design o f experim ent (DoE) is a systematic, m athem atically sound approach that uses
few er experiments, requires rigorous but easy-to-use analytical techniques, and yields
m ore inform ation such as second or higher order interactions. Thus, a tw o-level 2)~v'
fractional factorial (screening) experim ent was conducted to identify significant factors
and proper factor settings in the process o f Cyclotene surface m odification. The factors
that m ay have effects on the am ination o f the surface were proposed as follows:
m icrowave power, cham ber pressure, treatm ent time, and sample tem perature. The
response character was the efficiency o f am ination o f Cyclotene surface w hich was
m easured by XPS analysis. The factors and levels are shown in the Table 3-1 below. This
design is w ell suited for phenom enological modeling o f m easured responses as a function
o f the four factors. Outside o f this range, the limitations were either due to plasm a source
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
58
pow er delivery range, plasm a stability, m ass flow controller range and accuracy, or due
to degradation o f polym ers.
Table 3-1 Factors and levels
Low level
(A)
(B)
(C)
(D)
power (W)
pressure (Torr)
treatment time (s)
temperature (°C)
(-1 )
50
0.2
60
Room temp.
Center point
(0 )
150
0.4
180
60
High level
(1 )
250
0.6
300
100
3.6.1 Response character:
The efficiency o f am ination o f Cyclotene was evaluated by XPS analysis. W e
took the N /C ratio as determined by XPS as the principal response m easurem ent, w hich
quantitatively represented the extent o f am ination o f Cyclotene.
C onsidering the limitations o f the surface am ination experiments, a 24~' fractional
factorial design w ith single-replicate, two blocks and
2
center points w ithin each block
w as applied. The design generator was D =A BC; the block generator was CD=AB. The
sum m ary o f this design and the matrix design are shown in Table 3-2 and Table 3-3,
respectively.
To m inim ize unintentional aging effects, every effort was m ade to keep the same
intervals betw een treatm ent and XPS analysis. Due to the large am ount o f time required
on X PS analysis, runs were performed on 2 different test days. Both the XPS analysis and
XPS operators w ere blocked in the design.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
59
The results o f the experim ent in the next chapter will indicate w hich m ain effects
and two-factor interactions are significant. The linear m odel determ ined by D esignExpert software will be provided as well.
3.6.2 Factorial design
Table 3-2 Sum m ary o f design
Experiment design
Block description
Number o f runs
2 ]'v' design
Blocki:
XPS analysis (test day 1)
4+2
Block2.
XPS analysis (test day 2)
4+2
Total run
12
2 blocks
Single replicate
2 center points in each block
Table 3-3 M atrix design
Standard order
Block
A
B
c
1
Block 1
1
1
-1
2
Block 1
0
0
0
0
3
Block 1
1
-1
1
-1
4
Block 1
-1
-1
-1
-1
Run order
d
=a b c
-1
5
Block 1
0
0
0
6
Block 1
-1
1
1
-1
1
7
Block 2
1
1
1
8
Block 2
0
0
0
9
Block 2
-1
-1
1
10
Block 2
-1
1
-1
1
11
Block 2
0
0
0
0
12
Block 2
1
-1
-1
1
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
srr
N/C ratio
60
3.7 Plasma Surface Modification Effects on Cell
Adhesion and Spreading
To dem onstrate the am monia surface m odification effects on high-density
im m obilization o f biologically active molecules, the covalent coupling o f oxidized
dextran to surface m odified Cyclotene was conducted. Cell viability assays were then
utilized to determ ine surface m odification effects on cell adhesion and spreading.
3.7.1 Dextran coating method for aminated Cyclotene film
3.7.1.1
Periodate oxidation o f dextran
Dextran was oxidized to produce aldehyde groups through standard periodate
m ethods w ith m inor m odification .6,7 Dextran (MW 40 kDa, Sigma), lg , was dissolved in
30 m L deionized water. Sodium periodate (NalCfi, QAM) was prepared for im m ediate
use. The NalCfi solution was then added to the solution o f dextran to make a 50% m olar
ratio o f NalCfi to dextran (moles o f glucose monomer). The reaction mixture was stirred
at 4°C overnight and protected from light by covering the reaction flask w ith alum inum
foil. The solution was then purified by precipitation o f un-reacted periodate and iodate
products using an equimolar aqueous solution o f BaCL. The purified oxidized dextran
solution was then lyophilized and stored (if not immediately used) at 4°C in a 50-m L
conical centrifuge tube protected from light. The product was analyzed by Fourier
transform infrared (FTIR) spectroscopy. The results showed a peak at 1700 cm '1,
indicating the characteristic aldehyde groups w ithin the dextran chain.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
61
3.7.1.2
Covalent coupling o f oxidized dextran to surface anim ated base m aterials
Im m ediately following plasm a surface am ination procedures, the am inated
Cyclotene film s were placed in 6 -well m ultiw ell dishes. O xidized dextran w as dissolved
in 0.2 M sodium phosphate buffer, PH, at a final solution concentration o f 0.02 g/m l. This
solution was added to the wells containing surface am inated Cyclotene substrates. The
substrates w ere allowed to incubate at room tem perature for 16 hours on a rocker
platform and protected from light. Follow ing incubation, the reaction m ixture was
decanted from the culture wells, and replaced by fresh 0.1 M solution o f sodium
borohydride (NaBH4) to reduce Schiff bases form ed and to quench any free unreacted
aldehyde groups present on the oxidized dextran chain. The substrates w ere allow ed to
incubate for 2 hours on the rocker platform. The N aBH 4 solution w as then decanted and
the substrates were rinsed gently several tim es w ith deionized w ater to rem ove unbound
dextran.
3.7.1.3
Cell adhesion and spreading studies
Cell lines
3T3 fibroblasts (ATCC #CRL-6476), utilized for adhesion studies, were
m aintained in D ulbecco’s M odified Eagle M edium (DMEM) w ith 10% Phosphate buffed
saline (PBS). All cell culture m edia and reagents were obtained from Life Technologies,
Inc.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
62
Cell adhesion assay
Cell adhesion assay was perform ed according to M assia et a l .6,7 to assess the cell
adhesion behavior o f the dextran coated surfaces. Six-well culture plates were initially
coated with a 0.5% pH EM A (in 95% ethanol) solution to reduce cell attachm ent to w ell
surfaces. Following thorough air drying o f pH EM A -coated culture plates under the sterile
hood, cleaned and sterile material samples were placed in each well. A pproxim ately 2 ml
o f 3T3 cell suspension in m edia (15,000 cells/ml) was added to each well o f the culture
dish. Control sample wells were uncoated w ith/w ithout m aterial sam ples served as
references. The culture plates were then incubated at 37 °C, 5% CO 2 for 24 hr. Follow ing
incubation, samples w ere fixed (3.8% form aldehyde in PBS, 5 m in) and stained (0.1%
aqueous toluidine blue, 5 min). Stained cells were then exam ined using phase contrast or
stereom icroscopy (Leica) at 100x m agnification. Cell m orphology was assessed
qualitatively on all material samples and com pared to adherent and spread cells on tissue
culture plastic. Three random 100* fields were selected for each substrate for analysis.
3.8 Aging Effect Study
In order to investigate the surface m odification caused by the gas plasm a
treatm ents and changes in the surfaces properties due to aging in sim ulated biological
environm ents, the treated and untreated polym ers were stored in three different
environm ents: air, phosphate-buffered saline (PBS) and a “sm art desiccator” (w ith
nitrogen flowing inside to minimize the oxygen content) for various periods up to 3
months.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
63
3.8.1 Storage media and experimental
The storage m edia used for this study include: PBS solution, “smart desiccator”
and air. PBS w as purchased from GIBCO (PBS, com prised o f the following: 136.9
m m ol/1 NaCl, 2.68 mmol/1 KC1, 8.09 mmol/1 Na2HP04, 1.47 mmol/1 KH2PO4, 0.9 mmol/1
CaCl 2 , and 0.49 mmol/1 M gCh). The “smart desiccator” was purchased from Terra
Universal Co. This desiccator is a cost effective apparatus w ith autom ated nitrogen
control system that can automate clean, dry storage to elim inate m oisture-related
degradation and m inim ize oxygen incorporation, and is ideal for biological and
pharm aceutical samples and other sensitive materials. The equipm ent and procedure for
plasm a surface m odification are the same as described before.
3.8.2 Surface characterization
After different periods o f storage, m aterials w ere characterized using XPS, FTIRA TR and dextran im m obilization test to determine the changes o f surface chem istry and
functionality induced by different storage media.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
64
References
1
Poncin-Epaillard, F.; W ang, W.; A usserre, D.; Scharzenbach, W.; D erouard, J.;
Sadeghi, N. Eur Phys J 1988, AP 4, 181.
2
Golub, M. A.; W ydeven, T.; Cormia, R. D. Polym 1989, 30, 1571.
3
Foerch, R.; M cintyre, N. S.; Sodhi, R. N. S.; Hunter, D. H. J Appl Polym Sci 1990,
40,1903.
4
Foerch, R.; Johnson, D. S urf Interface Anal 1991, 17, 847.
5
Griffiths, P. R.; Haseth, J. A. D. Fourier Transform Infrared Spectroscopy; W iley
and Sons: N ew York, 1986.
6
M assia, S. P.; Stark, J.; Letbetter, D. S. Biom ater 2000, 21, 2253.
7
Dai, L.; Stjohn, H. A. W.; Bi, J.; Zientek, P.; Chatelier, R. C.; Griesser, H. J. Surf
Interface Anal 2000, 29, 46.
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
CHAPTER 4
EXPERIMENTAL RESULTS
AND DISCUSSION
4.1 Overview
N itrogen-containing groups have been incorporated into Cyclotene surfaces by
am m onia plasm a w ith or without argon plasm a pretreatm ents, w hich was confirm ed by
XPS and A TR spectra. By DoE and actual experimental investigation, the preferred
operating w indow was determined. The highest degree o f surface am ination was obtained
w ith an argon plasm a pretreatm ent at 150 W, 0.4 Torr, 180 s, 60 °C, and a subsequent
am m onia plasm a treatm ent at 250 W, 0.6 Torr, 240 s, 175 °C. In this section, DoE results,
XPS and A TR spectra and N /C ratios as a function o f processing variables are presented
first, then results o f A FM analysis and cell adhesion and spreading study are shown.
Finally the discussion o f those results and a possible reaction m echanism are given.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.2 DoE (Design of Experiments) Results
For am m onia plasm a treatm ent alone (w ithout A r pretreatm ent), the principal
responses for the DoE m atrix design are show n in Table 4-1.
Table 4-1 R esult o f m atrix design for am m onia plasm a treatm ent w ith ou t A r
pretreatm ent
Standard order
B
c
D=ABC
Response:
N/C ratio
1
1
-1
-1
0.057
0
0
0
0
0.040
-1
1
-1
0.052
-1
-1
-1
-1
0.026
0
0
Run order
Block
4
1
Block 1
10
2
Block 1
6
3
Block 1
1
1
4
Block 1
9
5
Block 1
0
7
6
Block 1
-1
1
1
-1
0.035
8
7
Block 2
1
1
1
1
0.206
11
8
Block 2
0
0
0
5
9
Block 2
-1
-1
1
1
0.041
3
10
Block 2
-1
1
-1
1
0.018
12
11
Block 2
0
0
0
0
0.050
2
12
Block 2
1
-1
-1
1
0.086
A
0.043
0.036
The dependence o f the N /C ratio on the processing param eters (from Design
Expert software output) is given by
N /C ratio = 0.065 + 0.035 * power + 0.014 * pressure + 0.01 8 * time + 0.023 * temp.
+0.017 * power * pressure + 0.0140 * power * time + 0.023 * power * temp.
w here pow er is in W, pressure is in Torr, tim e is in s, and tem perature is in °C.
The result implies that we can identify the following significant factors according
to the order o f significance coefficients: absorbed power, interaction betw een pow er and
tem perature, temperature, the interaction betw een pow er and pressure, cham ber pressure,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
67
treatm ent tim e, and interaction betw een pow er and time. The reduced linear m odel
indicates that the extent o f am ination increases w ith increasing absorbed plasm a pow er,
increasing cham ber pressure, increasing treatm ent tim e, and increasing sample
tem perature, respectively.
For A r pretreatm ent, the process conditions were: pow er 150 W, pressure 0.4
Torr, tim e 180 s and tem perature 60 °C. The factors and levels for am m onia plasm a
treatm ent are the same as those in Table 3-1. The design m atrix is show n in Table 4-2.
Table 4-2 R esult o f matrix design o f am m onia plasm a treatm ent w ith A r
pretreatm ent
Standard order
Run order
Block
A
B
c
D=ABC
Response:
N/C ratio
4
1
Block 1
1
1
-1
-1
0.170
10
2
3
Block 1
0
0
0
0
0.052
6
Block 1
1
-1
1
-1
0.071
1
4
Block 1
-1
-1
-1
-1
9
Block 1
0
0
0
7
5
6
Block 1
-1
1
1
-1
0.056
8
7
Block 2
1
1
1
1
0.237
11
8
Block 2
0
0
0
0.047
0.058
0.049
5
9
Block 2
-1
-1
1
1
0.055
3
10
Block 2
-1
1
-1
1
0.050
12
11
Block 2
0
0
0
0
0.050
2
12
Block 2
1
-1
-1
1
0.099
The dependence o f the N/C ratio on the processing param eters (from Design
Expert software output) is given as follows:
N /C ratio = 0.098 + 0.046 * power + 0.030 * pressure + 0.006 * time + 0.012 * temp.
+0.032 * power * pressure + 0.003 * power * time + 0.012 * power * temp.
where pow er is in W, pressure is in Torr, time is in s, and tem perature is in °C.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
68
The qualitative tendency o f the N /C ratio w ith A r pretreatm ent is the same as that
w ithout Ar pretreatm ent. However, the quantitative N /C ratio is higher w ith Ar
pretreatm ent than that corresponding to the ratio w ithout A r pretreatm ent. M oreover, the
order o f relative im portance o f the independent param eter effects on N /C is different. For
am ination w ith A r pretreatm ent, the order is as follows: absorbed pow er, the interaction
between pow er and pressure, cham ber pressure, tem perature, interaction betw een pow er
and tem perature, treatm ent time, interaction betw een pow er and tim e.
4.3 Detailed Investigation of the Effects of Processing
Parameters on N-incorporation on Cyclotene
Surfaces
Further investigation o f the different param eter effects on N /C ratio o f the plasm atreated Cyclotene surface has been carried out. A ccording to the results o f the DoE as
described above, we observed that there are three significant interaction effects for both
am m onia treatm ent alone, and w ith Ar pretreatm ent cases. These three interaction effects
are: power*pressure, pow er ^temperature and pow er dim e. Based on this finding, we
focused an additional investigation o f these three interaction effects by perform ing
corresponding one-factor-at-a-time experiments. The N /C ratios versus cham ber pressure,
substrate temperature, and treatment time as a function o f pow er w ith and w ithout (w/o)
A r pretreatm ent are displayed in Figure 4-1-F ig u re 4-3, respectively.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
69
0 .2 4 -
—
0 .2 2 -
- • - N H 3,150W
0 .2 0 -
- ■ - Ar+NH3,50W
NH3, 50W
■*— NH3,250W
Ar+NH3,150W
0.18
Ar+NH 250W
0.16
o
0.14
O
-Z_
0.12
0.10
_
0 .0 8 0 .0 6 0 .0 4 0.2
0.4
0.3
0.6
0.5
Chamber pressure (torr)
Figure 4-1 N/C ratio o f the plasm a-treated Cyclotene surface vs cham ber pressure as a
function o f pow er w ith and w/o Ar pretreatm ent.
0.40
0 .3 5 0 .3 0 0 .2 5 -
O
03
0 .2 0 -
O
z
0 .1 5 0.10
-
0 .0 5 0.00
20
40
60
80
100
120
140
160
180
Substrate temperature (°C)
Figure 4-2 N/C ratio o f the plasm a-treated Cyclotene surface vs substrate tem perature as
a function o f power w ith and w/o A r pretreatment.
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
70
0.36 -i
-I
—■— NH ROW
•
NH3, 150W
NH3, 250W
a
Ar+NH3, 50W
*
0 .2 4 -
Ar+NH3, 150W
- A-A r+N H 3, 250W
0.2 2 -
Z
—A
_A-
■A-
3
4
.
0 .1 4 0.1 2 0. 1 0 0 .0 8 0.06 0 .0 4 -
•
a
0.02
2
5
Treatment time (min)
Figure 4-3 N /C ratio o f the plasm a-treated Cyclotene surface vs treatm ent tim e as a
function o f pow er w ith and w/o A r pretreatment.
Figure 4 -l~ F ig u re 4-3 show that for N /C ratios increase w ith increasing power,
operating pressure and substrate temperature as expected based on the DoE outcomes.
The figures further reveal the nature o f the interaction w ith each param eter pair. Figure
4-1 and Figure 4-2 show that the dependence on pressure and the dependence on
tem perature becom e stronger at higher plasm a power. Figure 4-3 shows that nitrogen
incorporation onto the polym er surface mainly occurs w ithin the first 4 m inutes for
m edium and low pow er and then essentially saturates; at high pow er, nitrogen
incorporation occurs w ithin first 3 minutes and then a slight decrease in the N /C ratio
occurs.
The tim e-dependent behavior may be tentatively explained by a com petition
betw een the nitrogen incorporation process and a polym er surface degradation process.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
A t higher pow er one would expect a relatively greater contribution o f the dam age
reaction pathways.
It is also apparent that under same am m onia plasm a processing conditions, the
extent o f nitrogen incorporation onto the polym er surface w ith A r plasm a pretreatm ent is
consistently substantially higher than that w ithout pretreatm ent. This behavior m ay be
attributed to a “CA SIN G ” (crosslinking via activated species o f inert gases) effect.
N um erous researchers 1’2,3,4,5’6 have reported that A r plasm a can cause highly branched
and crosslinked structures near the polym er surface through creation o f surface free
radicals that undergo subsequent rearrangem ent and recom bination reactions. It is
therefore reasonable to postulate that argon gas plays a role in creating free radicals that
could then react w ith ammonia to produce an am inated surface m ore easily.
O/C ratios can provide some insight inform ation on understanding the reaction
m echanism s o f the plasm a surface m odification process. The O/C ratios versus cham ber
pressure, substrate temperature and treatm ent tim e as a function o f power w ith and
w ithout Ar pretreatm ent are shown in Figure 4-4 ~ Figure 4-6, respectively.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
72
0 .6-1
—
NH3, 50W
—
NH3, 150W
-
0 .5 -
NH3, 250W
-
a
--■--Ar+NH3, 50W
-••--A r+ N H 3, 150W
-
-a -
- Ar+NH , 250W
0 .4 o
0 .2
-
0.2
0.3
0.6
0.5
0.4
Chamber pressure (torr)
Figure 4-4 O/C ratio o f the plasm a-treated Cyclotene surface vs cham ber pressure as a
function o f pow er w ith and w /o A r pretreatment.
0.6
- ■ - N H 3, 50W
- • - N H 3, 150W
a
0 .5 -
-
NH3, 250W
--■--Ar+NH3, 50W
Ar+NH3, 150W
- -
0 .4 -
a
- -
Ar+NH,, 250W
o
0.2
-
20
40
60
80
100
120
140
160
180
Substrate temperature (°C)
Figure 4-5 O/C ratio o f the plasm a-treated Cyclotene surface vs substrate tem perature as
a function o f pow er w ith and w/o A r pretreatment.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
73
0 .6 -,
NH3, 50W
- • - N H 3, 150W
a
0 .5 -
NH3,250W
■ Ar+NH3, 50W
♦
-
a
Ar+NH3, 150W
Ar+NH3, 250W
-
0 .4 -
o
2
o
—
d
0 .2
•
-
1
2
3
4
5
Treatment time (min)
Figure 4-6 O/C ratio o f the plasm a-treated Cyclotene surface vs treatm ent tim e as a
function o f pow er with and w/o A r pretreatment.
Figure 4-4 ~ Figure 4-6 show that oxygen incorporation increases w ith
increasing pow er, substrate temperature and treatm ent tim e, but that it is not strongly
correlated w ith the operating pressure (independent o f w hether or not the A r pretreatm ent
was em ployed). The XPS spectra o f am m onia plasm a-treated Cyclotene (both w ith and
w ithout A r pretreatm ent) contain more oxygen than that o f untreated Cyclotene samples.
W e note that careful leak testing and thorough virtual contam ination source evaluation o f
the cham ber was performed and expect that the oxygen-containing active species in the
plasm a is very low. Samples were, however, transferred through air before the XPS
m easurem ents were made. Surface-bound radicals created by plasm a exposure can,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
74
following venting o f the plasm a reactor, react w ith atm ospheric oxygen. The addition o f
oxygen to carbon-centered radicals is a very fast process, and oxygen can diffuse rapidly
through the few nanom eters o f the material probed by XPS. G erenser et a l .7 reported that
substantial post-treatm ent oxygen incorporation into polyethylene w as observed after 30
seconds o f exposure to the atmosphere.
In addition, we see the degree o f oxygen incorporation for the case w ith Ar
pretreatm ent is higher than that without A r pretreatm ent. This behavior is m ost likely also
related to the CASING effect. Argon gas can create a greater concentration o f free
radicals w hich can react w ith both am m onia and im purity gases (including O 2 ) in the
vacuum system during both plasm a treatm ent and post-oxidation periods, therefore
leading to greater oxygen introduction.
4.4 Surface Characterization and Bio-test Results
Surface atomic concentrations (or relative atomic ratios) as a function o f the
experim ental param eters were displayed in Figure 4-1 ~ Figure 4-6. These results
indicated that the surface modification occurred largely w ithin the first 4 m inutes o f
treatm ent, and the degree o f surface m odification increased with the absorbed power,
operating pressure and substrate temperature. In this study, FTIR-ATR, XPS and A FM
w ere used to obtain the nature o f the surface chem ical species and surface topography.
The typical spectra for surface before and after treatm ents are presented in the subsequent
subsections. M icroscopic images o f adherent and spread cells on the substrates are then
show n in the subsection that follows.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
75
4.4.1 XPS characterization results
The Cyclotene surface was exposed to am m onia plasm a w ith or w ithout argon
pretreatm ent for up to 5 minutes, and the typical resultant N Is, C Is, and O l s XPS
spectra are presented in Figure 4-7 through Figure 4-9. The operational param eters for
this treatment are: am m onia plasm a at pow er 250 W, pressure 0.6 Torr, tim e 300 s and
temperature 175 °C, w ith argon plasm a pretreatm ent at pow er 150 W , pressure 0.4 Torr,
time 180 s and tem perature 60 °C.
From the XPS spectra, we observed that the m ajor portion or the peak shifts were
realized in the first few m inutes o f treatm ent, w hich m eant the effects o f the plasm a
reached a relatively steady level after 4 m inutes’ treatment. The C Is spectra were
characterized by the rapid disappearance o f the
7t*
<—n shake-up and a slight asym m etry
on the high-energy side o f the peak, as new com ponents were introduced.
In deconvoluting the XPS spectra, we assumed that apparent peak shifts are
caused by the introduction o f new peaks due to the effect o f bond breaking or form ation.
Therefore the peak fitting program was instructed to find one or m ore new peaks,
growing in intensity but not shifting in position by increasing plasm a treatm ent time.
A lthough the O Is requires only one additional peak, the C Is envelope requires tw o. The
new C 1s peaks, at higher binding energies, are also due to new structures form ed by
plasm a treatment. As m any as four peaks may be present in the N Is spectra. The
com ponent XPS peaks and possible nitrogen-containing attributions are shown in Table
4-3.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
76
a )N Is
N,
130001250012000
after treatment
11500—
11000 -
nj,
10500
•'I'
10000
S
_c
9500 9000before treatment
850080007500-
m
ti r i M » . >
i m
7000-
“ i—
390
395
405
400
410
Binding energy (eV)
b) N 1s after plasm a treatment
N 1; 401.9 eV
N2: 4 0 0 . 8 eV
N3: 3 9 9 . 9 eV
N4: 3 9 9 . 2 eV
N2
N3
>.
N4
-4—•
c
404
400
402
Binding Energy (eV)
398
Figure 4-7 N Is spectra o f the Cyclotene before and after N H 3 plasm a treatm ents.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
77
a) C Is before plasma treatment
Center: 284.6 eV;
CD
c/J
0)
c.
"c
288
284
286
282
280
Binding Energy (eV)
b) C 1s after plasm a treatm ent
C 1: 2 8 8 .4 eV
C2: 2 8 7 .2 eV
C 3: 2 8 6 .4 eV
C4: 2 8 5 .3 eV
C3
CD
-I—*
<f)
c
(1)
c
C4
290
288
286
284
Binding Energy (eV)
Figure 4-8 C Is spectra o f the Cyclotene before and after N H 3 plasm a treatm ents.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
78
a) 0 1s before plasma treatment
C enter: 5 3 2 . 0 eV;
CO
%
4>—
'
CO
c
0)
c
536
532
534
528
530
Binding Energy (eV)
b) O 1 s after plasm a treatm ent
0 1 : 534.7 eV
0 2 : 5 3 3 . 9 eV
0 3 : 5 3 2 . 8 eV
0 4 : 5 3 2 . 0 eV
02
4>.
—
*
c
40—
'
c
03
04
538
536
534
532
530
Binding Energy (eV)
F ig u re 4-9 O l s spectra o f the Cyclotene before and after N H 3 plasm a treatments.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
79
Table 4-3 Com ponent XPS peaks and possible N -containing attributions
Envelop
Component
Binding energy
Possible N-containing attributions
c„
c,
288.4
0 = C -N -C = 0 , -NHCHO, - O N 0 2, -C = N
c2
287.2
—N -C = 0 , —C=N
C3
286.4
c- n h
C4
285.3
—
Ni
401.9
C=N, C=N, 0 = C -N -C = 0
n
2
400.8
-NCO, N -C = 0
n
3
399.9
0 = C -N
n
4
399.2
N -C
Oi
534.7
C -0 -N 0 2
o2
533.9
C -0
03
532.8
C = 0 -N
o4
532.0
0=C
N ls
Ois
2, C-NHOH, C-C=N
It is difficult to interpret the plasm a-induced surface chem istry from the C 1s
spectrum alone due to the possibility o f num erous overlapping peaks. For instance, some
possible species that can be introduced by am m onia plasm a treatm ent are: am ine (C -N )
at -2 8 5 .4 eV, im ine (C=N) at ~ 287.0 eV, am ide (0 = C -N ) at ~ 287.8, and im ide ( 0 = C N -C = 0 ) at -2 8 8 .0 eV. The standard binding energies are approxim ate and can shift ± 0.3 eV, depending on the chemical environm ent o f the species. Also, the individual
peaks for these C -N species overlap w ith their C - 0 counterparts. N onetheless, by
considering additional factors, we still can derive some hints about the new functional
group formation. For instance, an imide is not as likely to be form ed as an am ide, because
it w ould require an additional carbonyl group to be available to bond to a nitrogen atom
that has already attached to a carbonyl group.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
80
From the N Is spectra, the N Is peak for N H 3 plasm a-treated Cyclotene is
relatively w ide and symmetric for all plasm a conditions. The w idth o f the peak is ~ 3.5
eV, which suggests that nitrogen is present in a variety o f groups, possibly including
amine (-399.1 eV), imine (-400.3 eV), and amide (-3 9 9 .9 eV).
4.4.2 FTIR-ATR characterization results
A com parison o f FTIR-A TR spectra for wave num ber from 4000 - 600 cm ' 1
before and after am monia treatm ent (250 W, 0.6 Torr, 180 s, 175 °C), w ith A r plasm a
pretreatm ent (150 W, 0.4 Torr, 180 s, 60 °C), (shown in Figure 4-10) reveals that some
peaks appear upon treatm ent and some disappear. These various infrared absorption
peaks are sum m arized in Table 4-4, along w ith their possible assignm ents to chem ical
species. As discussed below, N -containing species (NFIn) were grafted to the Cyclotene
structure as it underw ent treatment. The absence o f specific Si-N peaks indicated that the
new N -containing species were grafted to C only.
A m m onia plasm a treatm ent results in creation o f a new wide strong peak 3100 3500 cm ' 1 and a weaker peak around 1620-1590 cm '1. These new additions indicate that
N H n groups have been grafted on to the Cyclotene surface. In addition, the existing
hydrocarbon peaks at 3007, 2953, 2898, 2837 cm ' 1 decrease in intensity, w hich indicates
that CH 3 and CH 2 groups have been rem oved from the polym er surface by plasm a
treatment. Decreases in the intensities o f peaks at 1500, 1256 and 1120 cm "1 indicate
corresponding decreases in C=C, Si-C and S i-0 groups, respectively, on the plasm atreated polym er surface.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
81
K
K
CO
0.02
O
o
co
in
n.
o
o
0)
CD
Untreated
O
C
CO
-Q
L_
CD
o
CD
hLOI
CO
<
Treated
4000
3500
3000
2500
1500
2000
1000
_1
Wavenumber (cm )
Figure 4-10 FTIR -A TR spectra for am monia (with A r plasm a pretreated) plasm a-treated
and untreated Cyclotene. Treatment conditions are: Ar (150 W, 0.4 Torr, 180 s, 60 °C)
and N H 3 (250 W, 0.6 Torr, 180 s, 175 °C).
Table 4-4 FTIR-ATR peaks and their assignm ents
Wavenumber (cm'1)
Gained or increased
— Assignments
Lost or decreased
3100-3500
v (NH„), v (OH),
3007, 2953,2900,
v (CH„),
2837
v (cyclobutene CH)
1625-1587
aromatic p (C=C)
8 (CH) + v (CN)
1620-1590
1498
Aromatic p (C=C), polymer backbone
1445
5 (CHn)
1406
v (C-H) in RCH=CH2
1256
v (Si-C) in Si(CH3)2,
1120
v (Si-O) in Si-O-Si
833, 790,
v (C-H) in RjR2 C=CR3, aromatic finger print
700
5 (C-H) in R 1CH=CHR2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
82
4.4.3 AFM results
Typical AFM im ages for plasm a-untreated and treated Cyclotene surfaces are
presented in Figure 4-11. The differences betw een treated and untreated sam ples are
dramatic. Surface m odification by am m onia plasm a w ith or w ithout argon pretreatm ent
produces fine, bulge-like ridges and additional structure superim posed on the gentle
undulations native to untreated C yclotene film. However, for the am m onia (w ith argon)
plasm a-treated surfaces, the bulge-like ridges are obviously larger than that for am m onia
alone plasm a-treated ones. In addition, the polym er surface topographic condition
changes w ith operational parameters.
The root-m ean-square (rms) roughness o f the surface for the control (untreated)
sample is about 2.5 nm (shown in a), w hich is consistent with a planarization effect on
the Si/SiC >2 substrate by Cyclotene. U nder the same operational condition (250W ,
0.6Torr, 100 °C, 300 s), the rms o f the surface that was treated by am m onia plasm a
w ithout Ar plasm a pretreatm ent is 8.5 nm (shown in b), and the rm s for the surface
treated with argon plasm a (150W, 0.4 Torr, 60 °C, 180 s) followed by am m onia plasm a
treatm ent is about 25 nm (shown in c), or a factor o f ~3 greater roughness than w ith
am m onia treatm ent alone.
U nder higher operational tem perature conditions (250W, 0.6Torr, 200 °C, 300 s),
rm s is greater than 80 nm (shown in d). For this case, roughness is very uneven and the
uniform ity is poor. We conclude that 200 °C is too high. By com bining A FM im age
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
83
observation and plasm a treatm ent experimental exploration, a conclusion can be drawn
that 175 °C is the tem perature upper limit for am m onia plasm a treatm ents.
i
1.0 nm
3.0 nm
0.5 nm
1.5 nm
0.0 nm
0 .0 nm
i
5.00 nm
5.00 nm
b) Ammonia plasma trea ted Cyclotene
a) Untreated Cyclotene
*.—
9.0 nm
■ |—| 9.0 nm
jirS t € I I
4.5 nm
4.5 nm
I
P
H
f e w
0.0 nm
5.00 um
c) Ammonia plasma treatment
(with Ar plasma pretreatment)
0
0 .0 nm
5.00 um
d) Ammonia plasma treatment (with
Ar plasma pretreatment) at 200°C
Figure 4-11 AFM im ages for am monia plasm a-treated Cyclotene film s: a) control; b)
am m onia plasm a (250W , 0.6Torr, 100 °C, 300 s); c) argon plasm a ( 150W, 0.4 Torr, 60
°C, 180 s) followed by am m onia plasm a (250W, 0.6Torr, 100 °C, 300 s); d) argon plasm a
( 150W, 0.4 Torr, 60 °C, 180 s) followed by am m onia plasm a (250W , 0.6Torr, 200 °C,
300 s).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.4.4 Results of cell adhesion and spreading studies after covalent
coupling of oxidized dextran to aminated Cyclotene surfaces
The Cell A dhesion value is determ ined on all plasm a-treated Cyclotene surfaces
and expressed as a percentage o f the control adhesion (fraction o f the surface covered by
cells) on an untreated Cyclotene surface. Representative optical m icroscopic fields o f
dextran-coated Cyclotene surfaces w ith or w ithout plasm a treatm ents under different
operating conditions are shown in Figure 4-12. The results clearly show that adhesion
values on plasm a-am inated Cyclotene surfaces decrease dram atically com pared w ith the
control (untreated) sample. In addition, for low level conditions (50 W, 0.2 Torr, 60 s,
room tem perature), the adhesion value (2 1 .2 ± 4 .6 % ) is higher than that (4 .6 ± 1.1%) for
high level conditions (250 W, 0.6 Torr, 180 s, 175 °C). For am m onia plasm a alone treated
sample, the adhesion (4.6 ± 1.1% at high level) value is higher than that (1.0 ± 0.3% at
high level) for am m onia with argon plasm a pretreated samples.
In order to understand the relationship am ong operational param eters, am ounts o f
nitrogen incorporation and cell adhesion values m ore clearly, a colum n plot is displayed
in Figure 4-13. It has been concluded that the extent o f surface am ination increases w ith
increasing absorbed power, cham ber pressure, substrate temperature, and treatm ent time
based on the experimental design and the previous experimental investigations. Figure
4-13 illustrates how the adhesion value changes w ith operational condition and
nitrogen/carbon ratio (N/C). The adhesion value decreases w ith increasing nitrogen
incorporation. U nder the high level experimental condition (highest intensity plasm a)
w ith argon plasm a pretreatment, the cell adhesion value reaches the m inim um
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
85
(1.0 ± 0.3%). This clear correlation suggests that the surface covalent im m obilization o f
dextran has been achieved to reduce 3T3 cell adhesion and spreading. These results are
consistent w ith w hat other researchers 8,9,10 have observed on other dextran-coated
materials. Furtherm ore, the extent o f reduction o f cell adhesion and spreading depends on
the plasm a operating conditions. In this study, under the operating conditions that
m axim ize N /C ratio (am m onia plasm a w ith pow er 250 W, pressure 0.6 Torr, tim e 180 s,
tem perature 175 °C; w ith argon plasm a pretreatm ent: pow er 150 W, pressure 0.4 Torr,
tim e 180 s, tem perature 60 °C), the best cell rejection result w as obtained.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(d)
(e)
Figure 4-12 M icroscopic images o f adherent and spread cells on the substrates (scale bar
= 50 pm): (a) untreated Cyclotene surface; (b) N H 3 plasm a treatm ent under low level
operating condition; (c) N H 3 plasm a w ith Ar plasm a pretreatm ent under low level
operating condition; (d) N H 3 plasm a treatment under high level operating condition; (e)
N H 3 plasm a w ith A r plasm a pretreatm ent under high level operating condition.
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
87
100
T3
<
-
N/C ratio
cell adhesion value
80-
0.8
60-
0.6
40-
0.4
20-
0.2
~
0)
o
0.0
C o n tro l
nh3
N H 3+ A r
NH3
N H 3+ A r
lo w le v e l
lo w le v e l
h ig h le v e l
h ig h le v e l
Operational Conditions
Figure 4-13 Cell adhesion value and N /C ratio under different operational conditions.
We compared this result w ith w hat M assia et al . 11 obtained using a tw o-step dryw et am ination technique. In the tw o-step dry-w et surface am ination technique, tw o steps
are needed: an oxygen dry plasm a treatm ent, w hich is used to carboxylate the surface,
follow ed by a poly-lysine w et step, w hich is used to am inate the surface. The conclusion
is that an equivalent or better cell adhesion and spreading result can be obtained via onestep dry plasm a m ethod compared w ith that via two-step dry-wet am ination method.
However, for the present dry plasm a one-step technique, the procedure is m uch simpler.
In addition, the dry am monia plasm a one-step technique does not em ploy an oxygen
plasm a that can deleteriously oxidatively dam age the polym er surfaces.
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
88
4.5 Aging Effect Study Results
Under selected optimal condition, the ammonia plasm a treatm ents produced
highly modified surfaces w ith high N /C ratio. However, these N /C ratios changed w ith
tim e during storage in different environments. Changes in N /C ratio, O/C ratio and
chemical state as a function o f tim e as m easured with XPS spectra are described in the
following subsections.
4.5.1 XPS analysis of nitrogen and oxygen incorporation
Figure 4-14 shows the N /C ratio, as determined w ith XPS, as a function o f
storage time after plasm a treatm ents in three different environm ents. N o visible change
was observed after the untreated Cyclotene sample was stored in PBS or air for up to 3
months, w hich confirm s the inert nature o f Cyclotene. H owever, once plasm a-treated
Cyclotene sam ples were stored in their aging environments, changes in the N /C ratios
w ith aging tim e becam e appreciable immediately. For all three environm ents, decreases
in N /C ratios occurred at a relatively high rate in the first few days o f storage and then
became more gradual; after 20 days’ storage, there was relatively little change. The
reaction kinetics varies w ith different aging media. Over the long term , overall changes
were m ost dram atic for samples stored in air or PBS; the samples stored in the dessicator
were relatively m ore stable.
This instability behavior has been ascribed to surface reorientation : 12 unfavorable
interfacial energetics associated with the presence o f polar groups at the interface w ith air
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
89
provides a driving force that displaces m odified polym er chains from the surface into
deeper regions o f the polymer, with concom itant em ergence o f hydrophobic m aterial. For
all three environm ents, decreases in N /C ratios occurred at a relatively high rate in the
first few days o f storage and then became m ore gradual; after
20
d ay s’ storage, there was
relatively little change. Decreases in the N /C ratios on aging in different environm ents
could result from evaporation o f low m olecular w eight m aterial or from surface
restructuring . 13 During surface restructuring, the nitrogen-containing groups are m oved
aw ay from the surface into the polym er so that the photoelectron escape probability is
reduced and thus the N Is signal is attenuated.
Alternatively, Gerenser 7 suggested that imines produced by nitrogen plasm a
hydrolyze w hen absorbing atmospheric w ater vapor, thus the N /C ratios decrease.
A ccording to m ost literature reports , 14,15’16,17 it is m ore reasonable to ascribe the tim e
dependence o f N /C ratios to the surface restructuring o f m odified polym ers. H ow ever, the
observation that am m onia plasm a-treated samples experienced the highest degree o f
aging in the PBS solution while experienced the low est degree o f aging in the “sm art
desiccator” , is consistent with Gerenser 7 hydrolysis m echanism. In the “smart
desiccator”, the w ater vapor content was reduced to the low est degree, while in PBS
solution im ine groups can easily becom e hydrolyzed.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
90
0 .2 7 0 .2 6 0 .2 5 0 .2 4 0 .2 3 -
Desiccator
0 .2 2 -
o
0.21
'A -
-
O
0.20-
2
0 .1 9 0.18
0.170.160.150.14
0
20
40
60
80
100
Storage time (days)
Figure 4-14 N /C ratio as a function o f aging tim e in 3 different media.
The XPS spectra o f plasm a-treated Cyclotene specim ens all contained a signal
assignable to oxygen even before aging. The oxygen content increased w ith tim e w hen
samples were stored in different aging media, as shown by O/C ratio vs time during
storage in the three aging environm ents in Figure 4-15 below.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
91
0.620.600.580.56
0.54
0.52-1
o
’co 0.50 H
O 0.48
O
0.46-1
0.44
— Air
^ — Desiccator
0.420.400.3820
40
60
80
100
Storage time (days)
Figure 4-15 O/C ratio as a function o f aging time in 3 different media.
In a fashion similar to the N /C decreases, the oxygen uptake underw ent three
distinct stages: an initial fast process on the first day o f storage; a m ore extended and
gradual process that continued for about 3 weeks; and a final relative stability o f the O/C
ratio.
The initial fast oxygen uptake can be explained by the addition o f in-diffusing
atm ospheric oxygen to carbon-centered radicals on the surface and sub-surface layers.
The addition o f oxygen to the carbon-centered radicals is a very fast process. O xygen can
diffuse quickly through the few nanom eters o f the polym er surface. The extended oxygen
uptake can be attributed to the secondary reactions related to those involved in the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
92
oxidative degradation o f polym er chains, w hich occurs through a series o f com plex
radical reactions . 18,19,20 The gradual decay o f m etastable peroxides can m aintain oxidative
reaction cycles over extended period o f time, thereby producing a long-term oxygen
uptake. A study o f the long-term oxidative reaction o f a hydrocarbon plasm a polym er
dem onstrated that these reaction pathways and cycles are com plicated.
21
4.5.2 XPS spectra analysis: aging in PBS and air
In order to gain further insight into the chem ical reactions associated w ith aging
o f the samples, deconvoluted high resolution O ls and N ls spectra are presented below
(Figure 4-16 ~ Figure 4-19). The O ls spectra for aged m aterials provided im portant
inform ation regarding the chemical changes induced by aging m edia. As indicated in
Figure 4-16, all O Is spectra associated with am m onia plasm a-treated Cyclotene can be
fitted into three peaks which represent - C = 0 - , - C - 0 - and 0 - C = 0 - type groups,
respectively. Significant changes in the ratios o f integrated peak areas were observed
upon aging. The - C - 0 - / - C = 0 - ratio increased from 1:1.05 to 1:0.55 after aging in PBS
for 3 m onths, w hich implied that - C = N - instead o f - C - N - groups w ere being replaced
by - C = 0 groups. In contrast, for aging in air, the - C - 0 - / - C = 0 - ratio decreased from
1:1.05 to 1:1.75 after 3 months o f storage, as shown in Figure 4-17. For aging in the
“ sm art desiccator”, a similar result w ith that for aging in air w as obtained.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
93
C-O, Si-0
A g ed in P B S
C - 0 , Si-O
F resh
3
3
<0
(d
&
C=0
(A
c<
D
c
(A
C
Q
)
C
542
540
538
536
534
532
530
528
526
524
542
522
540
538
B inding e n e rg y (eV)
536
534
532
530
528
526
524
Binding e n e rg y (eV)
(a)
(b)
Figure 4-16 O Is spectra o f NH 3 plasm a-treated Cyclotene (w ith A r plasm a
pretreatment): (a) fresh sample im m ediately after plasm a treatm ent; (b) after 3 m onths o f
aging in PBS.
C-O , S i-0
F resh
•C = 0
A ged in air
3
3
C=0
<d
I
(A
C
0)
C-O, S i- 0 ^
c
542
540
538
536
534
532
530
Binding en e rg y (eV)
(a)
528
526
524
522
542
540
538
536
534
532
530
528
526
524
Binding E n erg y (eV)
(b)
Figure 4-17 O Is spectra o fN H 3 plasm a-treated Cyclotene (with A r plasm a
pretreatm ent): (a) fresh sample im m ediately after plasm a treatm ent; (b) after 3 m onths o f
aging in air.
The N 1s spectra associated w ith plasm a-treated Cyclotene aged in different
m edia also provided valuable inform ation concerning the nature o f post-chem ical
reactions on polym er surfaces and sub-surfaces. As indicated in Figure 4-18, the N Is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
94
spectra are curve-fitted into three peaks w ith binding energies o f approxim ately 398.8,
400.2 and 401.7 eV, representing - C - N - , - C = N - and -C = N —, respectively. For aging in
PBS, a decrease o f the - C - N - /- C = N - ratio from 1.0:0.9 to 1.0:1 .6 was observed after 3
months o f storage, w hich im plied that - C -N H 2- groups w ere being replaced by more
polar -C = O N H - groups. A slight reduction at peak 401.7 eV was also observed
following prolonged storage in PBS. For aging in air, as show n in Figure 4-19, the - C N -/- C = N - ratio was increased from 1.0:0.9 to 1.0:0.7, w hich suggested that imine
groups rather than amine groups were converted during aging process. For aging in the
“sm art desiccator”, a similar change in - C - N - /- C = N - ratio w as observed, im plying a
sim ilar reaction m echanism to that occurred in air.
.,
F re sh
j
,
1
,
1
1
1
1
/- \
A g ed in P B S
V " C=N
-
/
D
-
yc-N
TO
C=N
&
(/)
c
0)
A
/
c
\
\
/
1
404
402
400
398
Binding e n e rg y (eV )
(a)
396
.
/\
1
_
\
1
1
.
1
-
394
Binding e n e rg y (eV)
(b)
Figure 4-18 N Is spectra o f NH 3 plasm a-treated Cyclotene (with A r plasm a
pretreatm ent): (a) fresh sample immediately after plasm a treatm ent; (b) after 3 m onths o f
aging in PBS.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
95
1
Fresh
'
1
•
1
- Aged in air
C-N
'
1
'
/ \
-
C'
C=N
C=N
*
ca>
V
/
/ W
\
\
_
CO
’
c
/A v
A
\
.............................................. .........
404
400
402
398
Binding energy (eV)
(a)
396
394
402
400
■
398
Binding energy (eV)
(b)
Figure 4-19 N Is spectra o f N H 3 plasm a-treated Cyclotene (w ith A r plasm a
pretreatm ent): (a) fresh sample im m ediately after plasm a treatm ent; (b) after 3 m onths o f
aging in air.
4.5.3 Discussion
Based on the evidence presented in this chapter, am m onia plasm a-treated (w ith
argon plasm a pretreatm ent) Cyclotene surfaces contain amine and other nitrogencontaining groups. However, these plasm a-treated surfaces are unstable and undergo
aging processes during storage. To understand the aging process in different practical
storage media, it is necessary to investigate the chemical transform ations and their
kinetics. The principal processes occurring during aging are chem ical reactions w ith the
storage medium, and relaxation o f side groups and chain segments in the surface region.
In the case o f storage in an aqueous medium, chemical reaction results in the
substitution o f N by O. The stability o f N 2 plasm a-treated surfaces aging in air or m oist
air w as studied by several researchers who concluded that a dom inant process was the
conversion o f amines to am ides .23 In this study, such a transform ation was indicated by
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
96
changes in elem ental ratios measured by XPS. Furtherm ore, the reaction occurred m ore
rapidly in PBS than in air and in the desiccator. The detailed reaction steps are as yet
unknown, but m ay be attributable to the substitution o f am ines w ith either am ide or
hydroxyl groups.
It was reported that reconditioning air-stored samples in PBS can partially restore
the surface m odification effect by plasm a treatm ents, w hich is also attributed to the chain
or group relaxation24. The surface w ettability o f air-stored PTFE was partially reinstated
by a 10 m inutes’ reconditioning in PBS, w hich suggested that pre-aging o f plasm atreated surfaces in a aqueous environm ent for a short period m ay stabilize the surface
properties o f polym ers for an application in vivo.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
97
References
1
Everaert, E. P.; Chatelier, R. C.; V an Der Mei, H. C.;Busscher, H. J. Plasm as Polym
1997, 2 ,4 1 .
2
Sprang, N .; Theirich, D.; Engemann, J. Surf Coat Technol 1998, 98, 865.
3
Collaud, M .; N owak, S.; Uttel, O. M. K. F.; Schlapbach, L. J A dhes Sci Technol
1994, 8 , 435.
4
Vallon, S.; D revillon, B.; Poncin-Epaillard, F.; K lem bergsapieha, J.; M artinu, L. J
Vac Sci Technol A 1996, 14, 3194.
5
M ason, M .; Vercruysse, K. P.; Kirker, K. R.; Frisch, R.; M arecak, D. M .; Prestw ich,
G. D.; Pitt, W. G. Biom ater 2000, 2 1 ,3 1 .
6
Inagaki, N.; Tasaka, S.; Hibi, K. J Adhes Sci Technol 1994, 8 , 395.
7
Gerenser, L. J. J A dhes Sci Technol 1987, 1, 303.
8
M assia, S. P.; Stark, J.; Letbetter, D. S. Biom ater 2000, 21, 2253.
9
M assia, S. P.; Stark, J. J Biomed M ater Res 2001, 56, 390.
10 M archant, R. E.; Yuan, S.; Szakalas-Gratzl, G. J Biom ater Sci, Polym Ed 1994,
549.
11 M assia S.; Holeko M.; Ehteshami G. J Biom ed M ater Res 2004,
68
6
,
A, 177.
12 Xie, X.; Gengenbach, T. R.; Griesser, H. J. J Adhes Sci Technol 1992, 6 , 1411.
13 G engenbach, T. R.; Xie, X.; Chatelier, R. C.; Griesser, H. J. J A dhes Sci Technol
1994, 8 , 304.
14 Occhiello, E.; M orra, M.; Cinquina, P.; Garbassi, F. Polym 1992, 33, 3007.
15 Griesser, H. J.; Da, Y.; Ehighes, A. E.; Gengenbach, T. R.; M au, A. W. H. Langm uir
1991,7, 2484.
16 Youxian, D.; Griesser, H. J.; Mau, A. W.-H.; Liesegang, R. S. Polym 1991, 32, 1126.
17 G arbassi, F.; M orra, M.; Occhiello, E.; Barino, L.; Scordam aglia, R. S urf Interface
Anal 1989, 14, 585.
18 Lacoste, J.; Carlsson, D. J.; Falicki, S.; Wiles, D. M. Polym D eg Stab 1991, 34, 309.
19 Lazar, M .; Rychly, J. Adv Polym Sci 1992, 102, 189.
20
Gugumus, F. Polym Deg Stab 1990, 27, 19.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
98
21
Gengenbach, T. R.; Vasic, Z. R.; Chatelier, R. C.; Griesse, H. J. J Polym Sci, Part A:
Polym Chem 1994, 32, 1399.
22
Chatelier, R. C.; Xie, X.; Gengenbach, T. R.; Griesser, H. J. Langm uir 1995, 11,
2576.
23
Xie, X.; Gengenbach, T. R.; Griesser, H. J. in Contact Angle, W ettability and
Adhesion; R. J. Good, Ed.; Elsevier, Am sterdam , 1993, pp 509..
24
W ilson, D. J.; W illiams, R. L.; Pond, R. C. Surf Interface A nal 2001, 31, 397.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 5
TWO PROPOSED MODELS
FOR AMMONIA PLASMA
REACTION
5.1 Overview
This chapter focuses on the form ulation o f two num erical m odels designed to
provide a descriptive understanding o f the dominant reaction pathw ays in remote
m icrowave am m onia plasm a treatm ent o f Cyclotene surfaces. A plasm a kinetic m odel in
the gas phase o f the downstream microwave plasm a system is used to describe the
reactive species densities in the gas phase, and a theoretical surface am ination m odel is
used to describe the reaction processes o f plasm a surface m odification by these reactive
species.
Plasm a functionalization provides a way to introduce am ino groups to a polym er
surface and has been studied by many researchers. However, the m echanism s are not well
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
100
understood. One o f the reasons for this deficit is the lack o f experim ental (or theoretical)
gas phase com positions close to the substrate surface during the process. V arious
nitrogen-containing species are candidates for direct amino or unspecific nitrogen
grafting. The im portance o f the selectivity o f am ino group form ation is obvious,
especially for basic research and biom edical device applications.
5.2 Plasma Chemistry Model in the Afterglow Region
5.2.1 Kinetic analysis of ammonia decomposition
To understand the chemistry o f plasm as, it is im portant to pair experim ents w ith
theoretical considerations. Accurate num erical kinetic m odels or even sem i-quantitative
m odels are indispensable to better understand the com plex kinetics and processing effects
observed in em pirical research on nitrogen-containing plasmas. Favia et al. have
proposed and discussed a competitive and synergetic m echanism for am m onia
dissociation . 1 Gerenser has reviewed the present know ledge o f the chem istry o f N H X
radicals , 2 which, however, has a focus on therm al reactions and therefore high
tem perature conditions. N H 3 decom position behavior under different plasm a conditions
has been directly investigated , 3’4’5’6,7 but a definitive decom position m echanism and the
different roles played by various intermediate species such as N H 2 , N H , N 2 H 2 , etc. has
not been drawn.
The aim o f this work is to gain m ore insight into the relative im portance o f the
various reaction pathways for the plasm a-chem ical dissociation o f N H 3 in a m icrow ave
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
101
2.45G H z discharge, operated in the 0 .2 -0 .6 Torr pressure range, w ith pow er input
50-250 W and am m onia flowrate 3 0 -1 1 0 seem. The model for the low -pressure N H 3
plasm a system yields expected radical concentrations, w hich can help elucidate the
polym er surface reaction mechanisms.
Am m onia dissociation can occur either through direct electron im pact
dissociation o f ground state m olecules, or through bim olecular dissociation involving
vibrationally excited m olecular or through the m ore likely jo in t vibrational m echanism as
discussed by Capitelli et al .8 This m echanism involves dissociation occurring through
processes typical o f the pure vibrational m echanism (PVM ) and through the direct
electron mechanism (DEM ) from each vibrational level.
Based on the published rate constants and spectroscopic observations o f other
researchers , 9’10,11 the likely initial m olecular level elem entary steps for electron im pact
dissociation o f N H 3 are given by:
e + NH2^ - > e + NH2 + H
(5-1)
e + N H ,—^ e
(5-2)
+ NH + 2H
w here e represents an electron in the plasm a glow discharge. Follow ing the prim ary
decom position in m icrowave discharges, there are a variety o f subsequent radical transfer
or recom bination reactions that have been proposed as follows:
h
+ n h 2—
>n
h
+h 2
R e p r o d u c e d with p e r m i s s io n of t h e cop y rig h t o w n e r. F u r th e r r e p ro d u c tio n prohibited w ith o u t p e rm is s io n .
(5-3)
102
N H 2 + N H 2 -------------- >NH + N H
+
H2
N H + N H --------------- >N2 + H 2
h
+ n h 3—
>n h 2 + h 2
h
+n h 2+m
—
h
+ n h 2 + h 2—
H +H +M
---------------
(5-4)
(5-5)
(5-6)
>n h 3 + m
(5-7)
>n h 3 + h
(5-8)
>H2 + M
(5-9)
The proposed initial steps for the N H 3 decom position closely parallel those
applied for CH 4 and other hydrocarbon decom position, w here zero-order kinetics could
be explained by com bining the non-equilibrium dissociation R H - ~ ^ R + H, w hich
involves the rupture o f a C-H bond, w ith the therm al process RH + H —>R + H 2 . The
energy involved for the rupture o f the first C-H and N -H bonds and for H 2 dissociation
does not differ significantly (104 kcal mole "1 for CH 4 and H 2 and 110 kcal m ole "1 for
N H 3 , respectively.). For hydrocarbons, it is know n that the therm al process occurs at very
high rate even under therm al conditions, which w ill lead to very low value o f H atom
concentrations. For ammonia, the value o f the rate constant reported by D ove et al.
12
(k ~
109 cm 3 mole "1 sec ' 1 at 800 K) w ould lead to H concentration o f -1 0 % o f stable products.
As it is known, for vibro-rotationally excited m olecules the rate constant can be several
orders o f m agnitude higher than that evaluated under therm al conditions. Therefore it is
reasonable to assume that processes described by Eq. (5-6) proceed at higher rates
com pared to those described by Eq. (5-1) and (5-2), so that the H concentration is very
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
103
low w ith respect to parent m olecules (in this case am m onia) and reactive products o f the
plasma.
5.2.2 Plasma chemistry model in ammonia discharges
The num ber o f interdependent chemical and electrochem ical reactions required to
describe a plasm a discharge is generally too large to be handled intuitively. The term
“understanding” o f a plasm a process m ay therefore m ean that im portant and relevant
characteristics properties o f a plasm a discharge can be sim ulated by a relatively sim ple
num erical model. Based on the other researchers’ kinetic analyses o f the decom position
o f N H 3 , and an analogous kinetic study on etching o f SiC>2 in a CF 4 plasm a ,9,11’13’14’15 a
num erical plasm a kinetic model was developed in this section.
D uring developm ent o f the am m onia plasm a chem istry m odel, the follow ing
principal assum ptions are made:
■
W ell m ixed and well-confined (and defined) plasm a volum e;
■
Steady state operation;
■
Ideal gas behavior;
■
Electron im pact dissociation pathways for am m onia are lum ped into a single
effective reaction w ith a lum ped rate param eter;
■
The dissociation o f N H Xradicals to generate additional H atoms is ignored;
R e p r o d u c e d with p e r m is s io n o f t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
104
■ N H 2 is the dominant nitrogen-containing reactive species (i.e., the parallel
dissociation pathway to produce N H radicals as represented by Eq. (5-2) and a
potential secondary sequential electron im pact dissociation pathw ay N H 2 are
both negligible);
■
Wall (reactor surface) recom bination reactions are negligible.
In am m onia plasmas, the sequential generation o f N H 2 radicals and H atom s in
N H 3 discharges by electron im pact dissociation can be represented as
e + N H 3 — *B -> N H 2 + H + e
(5-10)
The species mass balance for N H 3 molecules, N H 2 radicals and H atom s can be
expressed as
d n N H 3 / d t ~ ^ N H 3 ~ K o n e n N H 3 ~ YlN H i / T 0
(5 -1 1 )
d n NH2 f d t ~ K o n en NH3 ~ n NH2 / 'TNH2
(5-12)
dnHj dt = klQnenNH^ —nH/ th
( 5 - 13)
w here km and ne are the (energy dependent) electron im pact reaction rate constant for
reactions represented by Eq. (5-10) and the electron density, respectively; nNH3, nNH2, and
hh are the density o f am monia molecules, N H 2 radicals and H atoms, respectively; q m 3 is
the m olecular flow rate o f am m onia feed gas divided by the reactor volum e (V); t 0 is the
residence time o f feed gas in the reactor; and
and rH are the decay tim e o f N H 2
radicals and H atoms in the reactor, respectively.
R e p r o d u c e d with p e r m i s s io n of t h e cop y rig h t o w n e r. F u r th e r r e p ro d u c tio n prohibited w ith o u t p e rm is s io n .
105
Assuming steady-state conditions, d n NH3 / d t = d n NH^ / d t = d n H / d t = 0 , the
following simple expressions for the am m onia, N H 2 and hydrogen atom concentrations
can be obtained:
I
(5-14)
*tOWe +
^ ; \ '7 / 3 T N H 2 K ( ) n e
n NH2 ~
I
(5 _15)
Kon e +
T0
H
,
1
W
(5-16)
+ -
In the absence o f a rigorous plasm a physics model to provide electron density and
energy, we introduce a new “plasm a efficiency” param eter c to relate kwne to the
absorbed microwave pow er P w,
Kone =
(5-17)
w here typical units for £ and Pw w ould be s"1W ' 1 and W, respectively. Then Eq. (5-15)
and (5-16) can be rew ritten to
n NH2 =
n H
(lN H - i T N H 2 y/
~ Qn h S
h
V
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
(5-18)
(5-19)
106
(5-20)
where
w ith y = <ffr0, w hich denotes the dissociation efficiency o f feed gas m olecules per input
power. The param eter i// corresponds to the dissociation degree: (n0 - nNH^) / n 0, w here no
is the value o f n m 3 at i V =
0
.
The density o f am m onia m olecules in the plasm a at steady state can be obtained
from Eq. (5-14) and (5-17) and the ideal gas relation to give
nNH3
9 nh3
P
(5-21)
w h ere p is the feed gas pressure in the chamber, ks is B oltzm ann’s constant and T is the
absolute gas temperature.
In a study o f the density o f CHF 3 in electron resonance plasm a by Takahashi et
ah , 16 the inverse density o f CHF 3 was shown to be proportional to the inverse pressure o f
the chamber. Similarly, we assume that £ can be em pirically expressed in a form
dependent on the inverse o f pressure as
P
R e p r o d u c e d with p e r m i s s io n of t h e cop y rig h t o w n e r. F u r th e r r e p ro d u c tio n p rohibited w ith o u t p e r m is s io n .
(5-22)
where a ; and
are positive em pirical constants. Therefore, the dissociation efficiency y
can be w ritten as
r = ‘h 0 = — I----- (atp + a2)
B
(5 23)
Q NH?,
Gas phase reactions include recom bination o f chem ically active species and b i­
radical reactions as shown in Eq. (5-7)~(5-9). I f the dissociation o f N H Xradicals to
generate additional H atoms is ignored, the approxim ate lifetim e o f the different radicals i
can be expressed as
1
7i
1
~
1
~i , F + ~
1
i,D
+ r i ,R
<5' 24)
where r,,/r, r ; D, r , « denote the flow -related decay time, the diffusion-related decay time,
and the bulk (plasm a volume) reaction-related decay time, respectively.
The flow -related decay time xf = to, and is thus given by
n„
TF= qm
p
kBTqNH3
C5-25)
where n0 is the density o f N H 3 at Pw= 0.
The expression for the diffusion-related decay time xd was obtained from
literature 17 and is given by
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
108
(5-26)
The first term in the right side o f Eq. (5-26) is related to the diffusion o f H atom s to the
surface, where Ao is the geometrical diffusion length determ ined by cham ber physical
design, and Du is the diffusion coefficient o f H in the gas phase. The second term is
related to the surface reaction w here lo is the volum e-to-surface ratio o f the reactor, au is
the reactive sticking coefficient o f H atoms to the surface, and vth is the therm al speed o f
H atoms.
If au « 1 (i.e., reactive recom bination at reactor w alls is negligible as assum ed
above), rD can be approxim ated by
According to the literature 18 ao in oxygen plasm as is given by
2
1 + v,6A / 2 D
(5-28)
o
w here X is the distance intercept o f the extrapolation o f the concentration profile beyond
the wall. Assum ing an analogous Eq. (5-28) holds for H atoms in the am m onia plasm a,
we have
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
109
l 0A
D~ n
(5-29)
H
Since gas diffusivity is inversely proportional to the pressure, Eq. (5-26) can be
written
r
rsj
D
~
P
c^ H
(5' 3°)
where C# is an constant, independent o f the pressure, and is given by
r
- DhP
-
, ,
.
(5-31)
If Eq. (5-7) is the dom inant bulk reaction process and N H 2 is the dominant
chem ical species among the possible N H Xspecies, then the bulk reaction-related decay
time can be expressed as
1
T
R ~ k Rn
NH2
( 5 ' 3 2 )
19 20
where kR is the rate constant for Eq. (5-7) which can be obtained from literature. ,
For H atoms, introducing Eq. (5-25), (5-30), and (5-32) into Eq. (5-24), the
approxim ate decay tim e th can be obtained as
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
110
p
+
& R n NH2
P
(5-33)
Similarly, for N H 2 radicals, the approxim ate decay tim e can be expressed as
P
k B T <lNH3 + C Nh 2
+ kRnHp
(5-34)
Introducing Eq. (5-33) and (5-34) back to Eq. (5-19) and (5-18), respectively, nn
and hmh2 can be written as
QnhJPP
kpTq NIIi + CH + kRnNIh p
(5-35)
and
(Inh^ P
k'NH2 + kRnHp
Eq. (5-35) and Eq. (5-36) are the m anipulated species balances for
(5-36)
and n m 2
w hich can be solved simultaneously for these two unknowns. Direct solution o f the hnh3
species balance, Eq. (5-21), provides the final unknow n am m onia gas num ber density.
For the model calculations all units for corresponding variables or constants are
listed for ease o f reference in Table 5-1.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
Ill
Table 5-1 Units for corresponding variables or constant
Variable
knh2 or yin
Pw
P
Q NH 3
T
U nit
cm ' 3
W
Torr
cm ' 3 s ' 1
K
Constant
ai
a2
Ch or C m 2
kR
kB
U nit
s^W
s‘1W '1Torr
Torrs ' 1
cm 3 s ’ 1
Torrcm 3 K ''
1
5.2.3 Model validity
Parameters for this m odel include: aj, a2, y/, Ch, C m 2 and kR. All param eters and
how they are obtained are listed in Table 5-2 below.
Table 5-2 Param eters em ployed in the chem istry m odel and the extracted means
alt a2
Constants will be estimated from data given in [16,21]. Plot reverse density as a function o f
reverse pressure, and a, a2 can be calculated from the intercept and slope o f given line. The unit
o f at is [s'1W'1], the unit o f a2 is [s^W^Torr].
QnH3/V
(V= chamber volume)
cm'3s'1
Can be obtained from Eq. (5-35)
mH
Slope o f plot p/nH~ p
c/ n h
niff can be derived
mTorr'cm3
[9,21]
C„ can be derived
mTorrcm
[9,21]
,C
Intercept o f plot p/nH~ p 2 =
QnHiV
In Eq.(5-33), if
kRnHp
kRcan be estimated to be
- « 1,
( kBTqNHj + Cm 2)
cH(kBTqNH^ +CNH^)
cm s
a straight line can be obtained by
plotting p/nH~ p 2.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
112
Final estimated param eter values are show n in Table 5-3.
Table 5-3 Summary of the values of all constants for this kinetic model
Parameters
Value
Unit
Source
Note
a,
3 .8 x l0 '3
s 'V 1
[16,21]
...
a2
2 .9 x l0 '5
s'V 'T o r r
[16,21]
...
<JNH3
1.86x l 0 15
c m 'V
...
Q/V
V
CH
6 .5 x l0 '2
-
...
Eq. (5-13)
0.58
Torrs"1
[9, 21, 24]
—
CnH2
kR
8.4
Torrs'1
[9,21, 24]
...
6 .6x l 0‘n
[19]
...
kB
1.035xl0'19
3
1
cm s
TorrcnfK'1
To estimate constant aj and a?, we need to plot inverse density o f N H 2 versus
inverse pressure o f N H 2 using experimental data. However, due to the lack o f the
experim ental data for N H 2 densities at different pressures, we estim ated ai and a 2 using
the data from a CHF 3 electron cyclotron resonance plasm a . 16 The prerequisite for the
reasonableness o f this estim ation is that the bond energies for C-H, C-F and N -H aren’t
rem arkably different (C-H 413 kJ/mol; C-F 488 kJ/m ol; N -H 391 kJ/mol).
By introducing Eq. (5-22) into Eq. (5-21), we can obtain the intercept o f line 1/n ~
1/p is (ctiPJqNH3 ), and the slope o f the line is (a2 PJqNH3 +kBT). The calculated gradient
o f this line is 2.44x1 O' 10 cm3Torr and the calculated intercept o f this line is 2.8x1 O' 14 cm3.
Thus ai and a2 can be determined (Table 5-3).
For the extraction o f mn and Ch, the detailed process is as follows: Firstly, we
need to correlate kRnNH2 in Eq. (5-35) to nu or p. Based on the study o f C F 4 discharges in
the m icrow ave plasmas, researchers 15,22 reported that bulk phase reaction rate is directly
proportional to p if the decay rate o f active species by gas phase reaction always
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
113
increases w ith increasing pressure. If w e assum e that this behavior is follow ed by H
atom s in N H 3 plasm a as well, Eq. (5-35) can be expressed as
<Inh , V P
kBTqm +CH+mHp 2
n"
<” 7>
w here mp is an empirical constant. N ow we can plot p /n n ~ p using given experim ental
data ,23 and then calculate the slope and the intercept o f the plot. A s listed in Table 5-2,
YYl
the slope o f the plot = — — where q m 3 and 1// are already know n, so the mu can be
k y j 1 'q ;)y'!_j
“ I- t
' 11
derived; the intercept o f the plot = --------- 5---------, w here all other param eters are know n,
QnHi V
and thus Ch can then be derived. Similarly, m^m and C m 2 can be extracted follow ing the
sam e procedure. From the given experimental data , p /n n ~ p can be sim ulated as a
straight line, the gradient o f the line is about 2.1x10'
12
3
1
cm Torr' and the intercept o f the
line is about 5 .5 x l0 ' 15 cm 3 Torr. Unfortunately, w e don’t have data for the determ ination
1
o f C m 2 (8.4 T o r r s ') for the system used by Tserepi et al,
91
so instead w e used a value
C m 2 w hich was estimated from a RF CFH 3 plasm a system .24,15
For the recom bination kR, we used the literature value (6.6x10'
11
3
1
cm s ' ) given by
G ordiets et al .20 If we follow the procedure described in Table 5-2, the estim ated value o f
kR is about 1 .2 x l 0 ' 12 cm 3 s’1, so we can see that the two values obtained from different
channels are quite comparable.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
114
To visually com prehend how the m odel-predicted densities o f active species
correlate the operational variables (pressure, power), the density o f N H 2 as a function o f
pow er and as a function o f pressure are displayed in Figure 5-1 (a) and (b), respectively.
Figure 5-1 (a) reveals that the reactive N H 2 radical density increases linearly w ith
increasing pow er at low absorbed pow er (up to about 100 W), as is typically observed for
low power glow discharge systems, and then exhibits a sub-linear dependence on power.
The sub-linear behavior is at least partially caused by am m onia source gas depletion at
the higher power.
0.6 Torr
0.4 Torr
CO
£o
o
*
0.2 Torr
X
c
0.9 -
0
50
150
100
200
250
300
Power (W)
(a)
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
115
250 W
150 W
E
o
CM
o
*
I
z
CM
0 .9 -
0.8
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Pressure (Torr)
(b)
Figure 5-1 M odel-predicted N H 2 density as a function o f (a) power; (b) pressure.
Figure 5-1 (b) shows a sub-linear increase in N H 2 radical density w ith increasing
pressure. A m axim um in radical density w ith further increases in pressure, as is typically
observed for m icrowave glow discharge systems, is not predicted by the present m odel. It
is anticipated that for higher levels o f pressure than explored in this study, greater levels
o f recom bination and de-excitation w ould be experienced at higher pressures, and w ould
lead to the observation o f an N H 2 radical num ber density maximum.
5.3 Surface Amination Model and Reaction Mechanism
Exploration
M odeling o f plasm a processes provides a way to test w hether a hypothetical
reaction pathw ay m ay explain experimental observations, and to evaluate the relative
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
116
importance o f individual reaction channels. In this chapter, a prelim inary m odel is
developed for the surface am ination o f Cyclotene. Due to the absence o f experim ental
information on the gas phase com position, it is not possible to develop a definitive model.
However, the relatively simple model could serve as a starting point for m ore rigorous
models, or could be used in an engineering process developm ent scenario.
The model developed in the following section describes am ino and nitrogen
grafting processes on a Cyclotene surface, and is used to describe the surface density o f
amino and non-am ino nitrogen groups as a function o f densities o f reactive species in the
gas phase.
5.3.1 Reactions of reactive species with polymer surfaces
A plasm a in am m onia produces a large number o f radicals, electrons, ions, and
neutral species w ith different excitation levels. Some o f these species possess sufficient
energy to initiate a chem ical reaction w ith a polymer, w hich then leads to m odification or
functionalization o f the polym er surface. Since molecular and dissociated (radical)
am monia and atomic nitrogen have m any energetic excitations, energetic details are
ignored and each o f the potential reactive species is assum ed to possess sufficient energy
in the surface functionalization context. Radical reactions either produce stable products
or a new radical w ith more or less reactivity. The former reactions are called com bination
or attachment, the latter reactions are called addition, substitution, abstraction, or
fragm entation as distinguished by the pathways illustrated as follows:
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
117
Combination: A +B' —>A - B
Attachment:
R-A+B'->R-A-B
Addition:
R = A +B '
Substitution:
R - A + B ' —» R - B + A
Abstraction:
R —A + B ' —t R ' + B —A
B- R- A
F ragm entation: R - A -R '+ B
R~ + B - A - R '
In general, two possible reaction m echanism s m ay result in the generation o f
prim ary amino groups on surfaces: a direct grafting o f prim ary am ino groups, and a
conversion o f non-am ino nitrogen groups to amino groups. The first m echanism contains
three principal steps: addition o f active nitrogen, N H insertion, and N H 2 attachm ent; the
second m echanism contains two principal steps: conversion and rem oval. These different
reaction steps are illustrated in Figure 5-2.
N
R— C—c —R
R— C = C —R
rearrangement
(a)
NHt +
R — H -------------►
^
HN— H - - R
2
H N— R
(b)
H
R— C— R +
Ho
I
H
H
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
118
H
I
•
NHo + R— C— R --------^ R— C— R + NH3
2
I
I
H
3
H
NH2
•
•
1
H
H
.
NH2 + R— C— R ------- a- R— C— R
2
I
I
(c)
R
.
,
H
H
I
1
I
2 H + R — N— R --------»■ H + R — N— R + R
------- ^ R— N— H + 2 R
H
H
I
4 H + R — C = N --------^ 2 H + R — C = N H ------- ^
I
R— C— NH,
I
H
H
H
H
H
3 H + R — C = N —R -------s- h ' + R — C— N—R
H
R— C— N—H + R
A
A
(d)
NHo
•
,
H +
1
R— C— R ------- 3-
I
H
•
R— C— R + NHo
I
H
(e)
Figure 5-2 R eaction pathways for amino grafting: (a) addition o f active nitrogen; (b) N H
insertion; (c) N H 2 attachment; (d) conversion o f nitrogen groups; (e) rem oval o f nitrogen
groups.
5.3.2 Amination model
The plasm a am ination model distinguishes only four gas phase species, [H],
[NH 2 ], [NH 3 ], and active nitrogen atoms [Nact]. On the surface side, the m odeled
functionalities include those that could be identified by XPS: prim ary amino groups and
non-am ino nitrogen groups. In this kinetic model, amino and other nitrogen group
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
119
grafting process on a surface will be described. For a given treatm ent duration and given
concentrations o f the four phase gas species, the surface density o f am ino and non-am ino
nitrogen groups can be estimated.
A ssumptions used in the developm ent o f this am ination m odel are listed as
follows:
■
Physical adsorption is negligible;
■
Chemical adsorption and absorption are negligible;
■
The surface reaction probability o f a reactive gaseous species is assum ed to be
proportional to its gas phase concentration and the surface density o f the
reaction partner at the surface (Eley-Rideal m echanism );
■
Surface diffusion is negligible;
■
Catalytic effects o f the metallic reactor walls are negligible;
The am ination model predictions will be fitted to experim entally-obtained surface
com position data (XPS) by adjusting the rate constants o f the model. Thus, it is possible
to derive suggestions for probable grafting mechanisms. A m ore stringent test o f the
model w ould require explicit gas phase data on various species. The reaction rate o f a
gaseous species is assum ed to be proportional to its concentration in the gas phase and to
the surface density o f its reaction partner, i.e., we assume that these elem entary reaction
steps occur through an Eley-Rideal pathway.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
120
For simplicity, the prim ary gaseous reactive species considered in the plasm a
m odel were H, N H 2 , and N H 3 . N ote that the am ination reaction pathw ays sum m arized in
Figure 5-2 include steps that involve NH radical and reactive N atoms, and hence the
plasm a model outcom es will need to be adjusted to account for these additional species,
or their contributions need to be ignored. In the present m odel we choose to neglect the
contribution o f N H insertion reaction steps (i.e., the reaction show n in Figure 5-2 (b))
since we anticipate that the [NH] radical concentration will be m uch less than that for
N H 2 . Active nitrogen is however considered in a non-rigorous, ad hoc fashion, w hich, as
described below, places the overall mass conservation statem ent in im balance by -1 5 % .
On the surface, the modeled functionalities include those that could be
distinguished by XPS: primary amino groups, A, and non-amino nitrogen groups, B. In
addition, active polym er sites, R, are included in the model. For these gaseous and solid
reaction partners, the model describes four basic heterogeneous processes described
below.
A ]/B: D irect grafting o f amino and non-am ino groups by active nitrogen. This
type o f step is shown in Figure 5-2 (a). The surface reaction rate for this step is
proportional to the free bonding capacity o f the surface Q and to the concentration o f
active nitrogen in the gas phase. The rate constants are aj and /? for amino and non-amino
grafting, respectively. For amino groups, this direct grafting is expected to play a m inor
role.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
121
A 2 '. A ttachm ent o f N H 2 to open surface radical sites. This type o f step is
illustrated by the second two reactions in Figure 5-2 (c). Surface reaction rate is
proportional to the density o f surface radicals, R, and to the concentration N H 2 in the gas
phase, and the rate constant is a.2 . N ote that the attachm ent reaction destroys (consum es) a
surface radical site.
A l B : Fragm entation-affected selective etching o f amino and non-am ino groups.
Chemical etching o f am ino groups is shown in Figure 5-2 (e); an analogous reaction for
etching o f non-amino groups is not shown. Rates o f these reactions are proportional to
the respective nitrogen group densities on the surface and the concentration o f atom ic
hydrogen in the gas phase. The rate constants are a and j5 for am ino and non-am ino
etching, respectively. N ote that these etching reactions create a surface radical site.
R : Creation o f surface radicals by hydrogen attack. This step is a prerequisite to
N H 2 radical attachment, and is illustrated by the first reaction in Figure 5-2 (c). The
reaction rate is proportional to Q and the density o f atomic hydrogen [H], and the rate
constant is p.
The model is form ulated as system o f three simultaneous differential equations
describing the rates o f form ation o f surface amino groups A, non-am ino groups B, and
active radical polym er sites R, respectively, as follows:
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
122
= a \nN„Q + a n NHR { f ) - a n HA{t)
d%
t = ^
, Q
-
M
O
d R / = pnHQ + anHA(t) + finHB(t) - a 2n„HiR(t)
7 dt
w here nt are the gas phase num ber densities o f the respective species i as before, and the
rate param eters are as defined above. All symbol definitions are sum m arized for
convenience in Table 5-4. Given initial conditions, these three m odel equations contain
four unknowns A, B. R and Q. A surface site balance for Q com pletes the model:
Q = Q-A(t)-B(t)-R(t)
A
w here Q is the m axim um capacity for bonding nitrogen or tolerating active sites (a fixed
number).
For model solution and surface am ination simulation, [H], [NH 2 ], and [NH 3 ] are
provided by the am m onia plasm a model. Active nitrogen atom concentration [Nact] can
25
be estim ated to 15% o f [H] as found in an am m onia plasm a dissociation system. .
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
123
Table 5-4 Definitions of the symbols used in the amination model
ai
reaction rate parameter for grafting am ino groups by active nitrogen
a2
reaction rate parameter for grafting am ino groups b y N H 2 attachment
a
rem oval reaction rate parameter for amino groups by H
P
reaction rate parameter for grafting non-am ino groups by active nitrogen
rem oval reaction rate parameter for non-am ino groups by H
P
surface radical creation rate param eter by H
A
the density o f am ino groups on the surface
B
the density o f non-am ino groups on the surface
R
the density o f active polym er surface sites
n Nact
the concentration o f active N in the gas phase
nH
the concentration o f H in the gas phase
K nH 2
the concentration o f N H 2 in the gas phase
K nH 3
the concentration o f N H 3 in the gas phase
5.3.3 Model evaluation
By inputting the estim ated densities o f transient species in the gas phase in term s
o f the plasm a kinetic chem istry model and adapting the rate constants o f the model, the
m odel predictions can be fitted to experim ental surface com position data. A stringent test
o f the model w ould require explicit gas phase data on various species.
A
In the surface am ination model, Q couldn’t be determ ined absolutely due to the
sem i-quantitative character o f the XPS m easurem ents and the unknow n surface radical
A
densities. In this case, Q is assum ed to be 100% for this model (in effect w e have
norm alized the surface concentration); correspondingly, units for all surface densities o f
am ino groups, non-amino groups and active polym er sites are norm alized as well. The
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
124
concentration o f [NH 3 ], [NH 2 ] and [H] are provided by the am m onia plasm a chem istry
model for given process conditions. A ccording to the results by M eyer-Plath,
•j/rin a
m icrowave plasm a system, the removal rate o f amino groups a by atom ic hydrogen was
constrained to be
1 /1 0
o f that o f non-am ino groups /? ; we apply the sam e constraint
here. The rem aining rate constants w ere fitted to the experim ental XPS data. First, plot
experimental data A ~ t and B ~ t, respectively, under proper pressure and pow er
conditions, and then calculate the slopes ( dA/dt and dB/dt) o f the above two plots at
different treatm ent time. Finally, inserting these data and the densities o f N actj H, and N H
resulting from the plasm a chemistry m odel under corresponding p and Pw, and the
reaction rate param eters can be extracted. The procedures to extract these rate param eters
for am m onia plasm a alone treatm ent and for am m onia w ith argon plasm a pretreatm ent
accom panied are similar except that the initial values for R are different. For am m onia
plasm a treatm ent alone, the initial value for R(t=0) is zero, w hile for the case w ith argon
plasm a pretreatm ent, the initial value for R(t=0) is an unknown. Therefore R(t=0) was
taken as an adjustable parameter. The values o f these rate constants resulting from this fit
have no absolute meaning, because the concentrations o f required species were estim ated
from the previous plasm a chemistry model, and XPS surface densities are semiquantitative only. However, their relative values still provide useful inform ation on
reconstructing the general behavior exhibited by the data. The estim ated rate constants
from this fit are listed in Table 5-5. These rate param eters fitted from different feeding
gases (with A rgon or without Ar plasm a pretreatment) shift on the order o f only 15-20%.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
125
This relative quantitative consistency o f the model extracted param eters provides a good
measure o f confidence in the overall m odel framework.
Table 5-5 Estim ated rate constants o f the am ination model.
Value
Rate constant
Unit
With Ar
Without Ar
a,
9.2xl0'lb
8 .6 x l 0 'lb
a2
2.4xl0 ' 15
2 .8 x l 0 ' 15
cm 3s-l
cm3s-i
a
2 .0 x l 0 ' 16
1.7xl0 ' 16
cm 3s-1
P
5.8xl0 ' 15
5.2xl0 ' 15
cm3s' 1
1
2 .0 x l 0 ' 15
1.7xl0 ' 15
cm3s' 1
P
3.9xl0 ' 15
4.6xl0 ' 15
cm s
3 -1
W ith these fitted rate param eters, the concentrations o f am ino and non-am ino
groups can be described under any given set o f operational param eters (Pw, p , T and t). In
addition, the model allowed the first test o f the suggested reaction m echanism s o f the
grafting o f amino and non-amino nitrogen groups. Results are consistent w ith the
suggestion that the N H 2 attachm ent mechanism is im portant during the surface am ination
process.
Plots o f the surface density o f amino groups and non-am ino groups as a function
o f pow er (Pfv) or pressure (p) under am m onia plasm a treatm ent alone and argon plasm a
pretreatm ent accompanied conditions are given in Figure 5-3 and Figure 5-4,
respectively. Reasonable agreement is found between the model predicted amino or non­
amino densities (solid lines) and the experimental data (points) for both N H 3 plasm a
treatm ent alone and A r plasm a pretreatm ent accom panied conditions.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
126
0.14
0.13
Amino
Non-amino
0.12 H
0.6 Torr
0.6 Torr
0.11
0.10
>, 0.09
-M
c
Q>
<1)
0.08-
0.2 Torr
■o 0.07-I
% 0-06 -|
d
<
/}
0.05 H
0.040.030 .0 2 0.01
-
0.00
T
-
50
0
150
100
200
250
300
Power (W)
(a)
0 .1 4 0 .1 3 -
a
•
Amino
Non-amino
250 W
0.12
0.11
250
-
150 W..,
>, °-10_
<D
0 .0 9 -
73 0.0 8 -|
0
0.0 7 -|
W 0 .0 6 -
50 W
0 .0 5 -
50 W
0 .0 4 0.03
0.02
0.1
—I—
—I—
i
—I—
—I—
I
0.2
0.3
0.4
0.5
0.6
0.7
Pressure (Torr)
(b)
Figure 5-3 Comparison o f the m odel-predicted and experimental density o f am ino and
non-am ino groups as a function o f (a) power; (b) pressure, for N H 3 plasm a treatm ent
alone conditions.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
127
0.6 Torr
A Amino
• Non-amino
0 .140.13 -
0.6 Torr
0.4 Torr
0.1 0 -
0.2 Tort
CO
g 0 09 '
8 008:
0.4 Torr
0.0 7-
w 0.0 6-
0.2 Torr
0 .050.040.03-
0 .0 2 0
50
100
200
150
250
300
Power (W)
(a)
0.15
0 .1 4 0.13-
*
•
Amino
Non-amino
250 W
0.12-
250W > ^i5Q w
0. 1 0 ^
0.09 -
150 W
(U
73 0.08 -
50 W
ro 0.07 -
t:
rn 0.060.05-
50 W
0.040.03-
0.0 2 -
0.1
0.2
0.4
0.3
0.5
0.6
0.7
Pressure (Torr)
(b)
Figure 5-4 Com parison o f the m odel-predicted and experim ental density o f amino and
non-am ino groups as a function o f (a) power; (b) pressure, for N H 3 plasm a treatm ent w ith
A r plasm a pretreatm ent conditions.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
128
5.3.4 Hypothetical reaction mechanisms
M aking progress in the understanding o f plasm a surface m odification not only
requires more precise gas phase and surface diagnostic techniques, but also new ideas
that simulate future research w ith w ell-designed experim ents. By considering all
experim ental data and the two proposed kinetic m odels in this chapter, some possible
principal surface m odification m echanism s are suggested in this section.
5.3.4.1
N H 2 attachm ent
N H 2 attachm ent is a radical attachm ent reaction that incorporates prim ary am ino
groups on the polym er surface at open reaction surface sites. To be efficient, N H 2
attachm ent requires atomic hydrogen, partially for the creation o f open reaction sites at
the polym er, partially to shift the gas phase reaction balance tow ards N H 2 production.
5.3.4.2
NH insertion
N H insertion is based on a radical addition reaction and is potentially a process
for the production o f prim ary amino groups. H owever, this effect has not been reported in
the literature. For the amination model developed in this section, this process is not
included due to the expected low n m value and its absence in the plasm a chem istry
m odel. However, compared to a tw o-step m echanism like N H 2 attachment, N H insertion
is a direct process, which may contribute to the fast and selective am ination process
observed in microwave plasmas.
R e p r o d u c e d with p e r m is s io n o f t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n p rohib ited w ith o u t p e r m is s io n .
129
5.3.4.3 Fragm entation-affected selective etching
In fragm entation-affected selective etching, the conversion o f nitrogen groups to
prim ary amino groups by hydrogen radical reactions is less probable than their rem oval.
The probability o f a successful conversion decreases w ith the num ber o f hydrogen
addition reactions required for a given type o f nitrogen functionalities, and fragm entation
o f the polym er chain may occur. This pathw ay requires a higher persistence o f prim ary
am ino groups, com pared to other nitrogen functionalities, partially due to the nitrogen
loss during the conversion, and partially due to the higher polarity o f the N -C o f prim ary
com pared to secondary and tertiary amino groups.
5.3.5 Surface reaction mechanisms on Cyclotene
The am ination model along with the XPS and FTIR -A TR data provide insightful
understanding about the Cyclotene am ination process. As we know , the possible
functional groups formed on the Cyclotene surfaces by am m onia plasm a treatm ent
include: prim ary amine, amide, imine, imide and nitrile groups. As concluded by Inagaki
et al ,27 in a chem ical sense, plasm a treatment is a radical-substitution reaction o f C-H
bonds in polymers. Hydrogen abstraction by the collision o f electrons, ion, or radicals
lead to carbon radicals in the polym er chains, and then the carbon radicals react with
simple radicals such as oxygen and nitrogen in the plasm a to yield oxygen and nitrogen
functionalities in the polym er chains. In a remote microwave plasm a the ion-induced and
electron-induced pathways should be negligible, and hence only hydrogen abstraction is
the initiating reaction pathway. From the viewpoint o f hydrogen abstraction, aromatic
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
130
polym ers are not favorable substrates for plasm a treatm ent because hydrogen abstraction
from aromatic carbons is not as easy as that from aliphatic carbons. Thus, for Cyclotene
polym er, hydrogen m ay be preferentially abstracted from the pendent CH 3 groups instead
o f phenyl carbons.
Three possible reaction pathways for this am ination process are proposed in this
section, and are sum m arized in the figure below (Figure 5-5). The first pathw ay, shown
in Figure 5-5 (a), occurs through hydrogen abstraction from pendant CH 3 groups, w ith
subsequent N H 2 radical attachment. The resulting amino group m ay undergo plasm a
radical attack to produce non-am ino nitrogen-containing groups, or in the presence o f
oxygen and/or w ater vapor (in the reactor cham ber or upon post-processing exposure to
air) m ay experience subsequent oxidation.
The second pathway, shown in Figure 5-5 (b), is initiated by hydrogen
abstraction from cyclobutane rings. Subsequent N H 2 radical attachm ent produces amino
groups on the cyclobutane ring; alternatively, successive oxidative attack at the
cyclobutane radical site can lead to ring opening. The amine itself can be attacked by
hydrogen radicals to produce =NH, or by oxidizing species to produce oxygen containing
groups.
The third alternative pathway, shown in Figure 5-5 (c), is direct addition o f
nitrogen-containing groups on to the polym er surface, w ith the possible subsequent
form ation o f imino groups and oxygen-containing groups by H radical attack and oxygen
incorporation upon exposure to oxidizing species.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
131
CH;
CH;
CH;
CH;
N H 3 p la s m a
NH;
CH
CH;
• S i— O — S iCH;
CH;
-2 H
NH2
V
c h
3
NH
CH3
-S i—
CH3
O — S iCH3
(a)
R e p r o d u c e d with p e r m i s s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e rm is s io n .
132
R e p r o d u c e d with p e r m i s s io n of t h e cop y rig h t o w n e r. F u r th e r re p ro d u c tio n p rohibited w ithout p e rm is s io n .
133
c h
• S i— 0 — S iCH3
N H 3 p la s m a
n h
—
S i—
3
CH:
o — s iCH;
2
n h
- S i — O — Si-
2
R
n h
2 c h 3
c h
3
-2 H
H
OH
NH
ch3
CH;
• S i— 0 — S
NH
CH3
CH3
OH
(c)
Figure 5-5 Possible reaction pathways for am m onia plasm a treatm ent on Cyclotene
surfaces: (a) hydrogen abstraction from pendent CH 3 groups and further oxidation
process; (b) hydrogen abstraction from cyclobutane ring and ring opening process; (c)
direct nitrogen addition to C=C.
In term s o f the amination model, the hydrogen abstraction process could lead to
subsequent N H direct insertion and N H 2 attachm ent steps. D uring N H 2 attachm ent
processes, atomic hydrogen and other radicals (denoted by -R) attack the polym er and
create the open reaction sites on the surface so that the N H 2 groups can be attached to
these open sites to form primary amine groups. In a sim ilar way, the secondary hydrogens
can be abstracted from such a group to form im ine and nitrile groups.
For the further reaction o f amine groups on contact w ith active oxygen species,
the carbonyl, carboxyl and the secondary amide groups could form through the
com bination o f the cleavage o f imide groups and the subsequent reaction w ith water.
R e p r o d u c e d with p e r m i s s io n o f t h e cop y rig h t o w n e r. F u r th e r re p ro d u c tio n p rohib ited w itho ut p e r m is s io n .
134
References
1
Favia, P.; Stendardo, M. V.; D 'agostino, R. Plasmas Polym 1996, 1, 96.
2
Alfassi, Z. B. N -C entered Radicals; John W iley & Sons: Chichester, England, 1998.
3
Hanes, M. H.; Bair, E. J. J Chem Phys 1963, 33, 672.
4
Carbaugh, D. C.; M unno, E. J.; M archello, J. M. J Chem Phys 1967, 47, 5211.
5
Cramarossa, F.; Colaprico, V.; D'agostino, R.; M olinari, E. 3rd Int. Symp. on Plasm a
Chemistry, Limoges, 1977.
6
Seideman, T. J Chem Phys 1995, 103, 10556.
7
Soucy, G.; Jurew icz, J. W.; Boulos, M. I. Plasm a Chem Plasm a Process B 1995, 15,
693.
8
Captelli, M.; M olinari, E. in Topics in Current Cheistry; Springer-V erlag, N ew York,
1980, pp 59.
9
D'agostino, R.; Benedictis, S. D.; Cramarossa, F. Plasm a Chem Plasm a Process
1981, 1, 19.
10 Capezzuto, P.; Cramarossa, F.; D'agostino, R.; M olinari, B. P lasm a Phys 1977, 17,
205.
11 Nicholas, J. E.; Spiers, A. I.; M artin, N. A. Plasm a Chem Plasm a Process 1986, 6 ,
39.
12 Dove, J. E.; N ip, W. S. J Chem 1974,52, 1171.
13
Kim, M. T. J Electrochem Soc 2 0 00,147, 1204.
14 Kim, M. T. Appl Surf Sci 2003, 211, 285.
15 Kim, M. T. J Electrochem Soc 2002, 149, G218.
16 Takahashi, K.; Hori, M.; Goto, T. Jpn J Appl Phys 1994, 33, 4745.
17
Chantry, P. J. J Appl Phys 1987, 62, 1141.
18 Booth, J. P.; Cunge, G.; Chabert, P.; Sadeghi, N. J Appl Phys 1999, 85, 3097.
19 Levchenko, A.; Alexeev, G. K INEL-Chem Kinet M odell Program , V ersion 4.2,
1994.
20
Gordiets, B.; Ferreira, C. M .; Pinheiro, M. J.; Ricard, A. Plasm a Sources Sci Technol
1 9 9 8 ,7 ,3 6 3 .
R e p r o d u c e d with p e r m i s s io n of t h e cop y rig h t o w n e r. F u r th e r r e p ro d u c tio n prohibited w ith o u t p e rm is s io n .
135
21
Hopwood, J.; C.R.Guarnieri; W hiehair, S. J.;
1993, 11, 152.
Cuomo, J. J. J Vac Sci Technol A
22
M iyata, K., Arai, H.; Hori, M.; Goto, T. J Appl Phys 1997, 82, 4777.
23
Tserepi, A.; Schwarzenbach, W.; Derouard, J.; Sadeghi, N. J Vac Sci Technol A
1997, 15,3120.
24
Haverlag, M.; Stoffels, E.; Stoffels, W. W.; Kroesen, G. M. W.; Hoog, F. J. D. J Vac
Sci Technol A 1994, 12, 3102.
25
Brocklehurst, B.; Jennings, K. R. Prog R eact
K inet 1967, 4, 1.
26
M eyer-Plath, A. A. Greifswald, Univ., Diss.,
2002.
27
Inagaki, N .; Tasaka, S.; Hibi, K. J Adhes Sci
Technol 1994, 8 ,395.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
CHAPTER 6
CONCLUSIONS A N D
RECOMMENDATIONS
FOR FUTURE WORK
This work investigated the surface m odification o f dielectric polym er D ow
Cyclotene 4026 (BCB). This polym er has excellent electrical insulating and barrier
properties, plus extremely low moisture uptake and good therm al stability properties,
w hich makes it a good candidate for both im plantable biom edical devices and
m icroelectronic packaging and interconnect applications. The am ination o f Cyclotene
using a one-step dry plasm a technique is a relatively sim pler and in principle more
readily controlled process that does not employ w et chemicals. In tw o-step dry-w et
surface amination, two steps are needed: first, an oxygen dry plasm a is used to
carboxylate the surface; second, a poly-lysine w et step is used to aminate the surface. In
the dry am m onia plasm a one-step technique there are no concerns about potential for
reactive oxygen damage during the first step. The more im portant technological outcome
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
137
o f this dissertation is the dem onstration that one can realize equivalent or better
animation extent and better surface quality w ith less damage using the one-step am m onia
plasm a treatm ent com pared w ith those using the tw o-step dry-w et technique.
The overall goal o f this study is to explore the capabilities and lim itations o f the
am ination process utilizing a dry plasm a one-step technique; identify the preferred
operating conditions through D oE and experim ental investigation to achieve the best
surface adhesion and biocom patibility; determ ine and correlate changes in physical and
chem ical states o f the Cyclotene surface due to plasm a treatm ents; and finally develop
fundam entally-based models w hich can describe the surface am ination behavior as a
function o f four processing param eters (absorbed power, operating pressure, substrate
tem perature and treatm ent tim e) to provide insight into reaction m echanism s o f the
plasm a surface modification.
The effects o f the four processing param eters on the extent o f nitrogen
incorporation into the polym er surface were studied. X-ray Photoelectron Spectroscopy
(XPS), Fourier Infra-Red Spectroscopy-Attenuated Total Reflection (FTIR-A TR), and
A tom ic Force M icroscopy (AFM ) were used to characterize the surface chem istry and
topography structures. A study about aging effect in different storage m edia was
conducted. A plasm a kinetic chem istry model in the gas phase and an am ination model
on a surface were developed and evaluated. Finally the possible reaction pathw ays on the
Cyclotene surface were proposed based on the am ination model and surface
characterization results.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
138
The experim ental investigation leads to the follow ing conclusions: XPS and
FTIR-A TR characterization results indicated that nitrogen-containing functional groups
were introduced onto Cyclotene surfaces by am m onia w ith/w ithout argon plasm a
pretreatm ent, and that the functional groups exist in the tw o m ost probable form s (am ine
and amide groups). The N /C ratio on the surface reached a m axim um o f 0.24 under high
level plasm a condition w ith argon plasm a pretreatm ent. Results o f design o f experim ents
showed that the extent o f surface am ination increases w ith increasing absorbed plasm a
power, increasing cham ber pressure, increasing treatm ent tim e, and increasing sample
tem perature for both am m onia alone treatm ents and am m onia w ith argon plasm a
pretreatm ents. There are also some second-order interactions during the am ination
process for both cases. The preferred operating condition is: argon plasm a pretreatm ent
(absorbed pow er 150 W, chamber pressure 0.4 Torr, substrate tem perature 60 °C,
treatm ent tim e 180 s), followed by am m onia plasm a treatm ent (absorbed pow er 250 W,
cham ber pressure 0.6 Torr, substrate tem perature 175 °C, treatm ent tim e 240 s). U nder
identical am m onia plasm a processing conditions, the extent o f nitrogen incorporation into
the polym er w ith Ar plasm a pretreatm ent is higher than that w ith am m onia plasm a alone.
This im provem ent is m ost likely due to the CASING (crosslinking via activated species
o f inert gases) effect, w hich can cause highly branched and crosslinked structures near
the polym er surfaces w hich serve as reactive sites for subsequent amination.
The effect on cell attachment and spreading w hen Cyclotene surfaces are treated
using am m onia/argon plasm as is dramatic: the dextran-coated plasm a am inated
Cyclotene surface significantly reduced 3T3 cell adhesion and spreading; moreover, the
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
139
reduction extent o f cell attachm ent and spread for am m onia w ith argon plasm a
pretreatm ent is much higher than that w ith am m onia plasm a alone treatm ent, and results
correlated w ith extent o f surface am ination. This result suggests that the surface covalent
im m obilization o f dextran has been achieved to reduce 3T3 cell adhesion and spread for
im planted biom edical device applications. It m ay ultim ately be possible to design the
chem ical com position o f a surface for optim um cell adhesion and spread reduction.
One o f the difficult challenges in this w ork is the process reproducibility, since
the therm al and chemical stability o f surface m odified films is poor in the presence o f
oxidizing atmospheres, and surface restructuring typically leads to an apparent decrease
o f the treatm ent effects over time. Therefore, the study o f evolution o f the post-treatm ent
surface com position and chem ical bonding state was investigated in detail. M oreover, in
the context o f biom edical devices, it is im portant to investigate the surface m odification
caused by the gas plasm a treatments and changes in the surface properties over storage
tim e in simulated biological environments. Three different m edia w ere chosen to store
both treated and untreated polym ers for various periods up to 3 months: air, phosphatebuffered saline (PBS), “sm art desiccator” (w ith nitrogen gas flowing inside to m inim ize
the degree o f humidity and oxidation). The aging processes for both untreated and plasm a
treated Cyclotene surfaces were identified using XPS and FTIR-ATR. The
characterization results showed that the am ination effects degraded w ith storage tim e for
all these three cases, but that the reaction kinetics varies with different aging media. The
reaction occurred more rapidly in PBS than in air and in desiccators. D ecreases in the
N /C ratios on aging in different environm ents could result from evaporation o f low
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
140
molecular w eight m aterial or from surface restructuring . 1 The observation th at am m onia
plasm a treated samples experienced the highest degree o f aging in the PBS solution, and
experienced the low est degree o f aging in the sm art desiccator, can be explained by the
hydrolysis m echanism suggested by Gerenser. In the hydrolysis m echanism , im ines
produced by nitrogen plasm a hydrolyze when absorbing atm ospheric w ater vapor, thus
the N /C ratios decrease. In the “smart desiccator”, the w ater vapor content w as reduced to
the lowest degree, w hile in PBS solutions, im ine groups can becom e readily hydrolyzed.
We speculate that the principal m echanism s occurring during aging are chem ical
reactions w ith the storage medium and relaxation o f side groups and chain segm ents in
the surface area . 3 Each o f these mechanisms proceeds w ith characteristic kinetics. In the
case o f storage in an aqueous medium, the chem ical reaction is the substitution o f N by
O. The detailed reaction is as yet unknown, but m ay be attributable to the substitution o f
am ines w ith either am ide or hydroxyl groups. It has been reported that reconditioning airstored samples in PBS can partially restore the surface m odification effect by plasm a
treatm ents , 4 which suggested that pre-aging o f plasm a-treated surfaces in an aqueous
environm ent m ay stabilize the surface properties o f polym ers for an application in vivo.
To provide a prelim inary understanding o f the relationship betw een processing
param eters (power, pressure, temperature and time) and final surface chem ical
com position (N/C ratio, N H 2/N), a plasm a kinetic chem istry m odel and a surface
am ination model were developed to describe the surface am ination process using the
operational variables as inputs and the surface chem ical com position as output. The
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
141
plasm a chem istry model in the gas phase describes the concentrations o f m ajor species in
the plasm a, w hich are the input param eters for the am ination m odel. Based on the
proposed m odels and the surface characterization results, the possible surface reaction
pathways were suggested. Surface initiation reactions that produce active sites on the
polym er surface that w ere hypothesized include: a) hydrogen abstraction from pendent
CH 3 groups, and further oxidation process on contact w ith oxygen and m oisture; b)
hydrogen abstraction from cyclobutane rings and further oxidation by active oxygen
species, which finally leads to the ring opening process and form ation o f carbonyl,
carboxyl and the secondary amide groups; and c) direct nitrogen addition to unsaturated
carbons and subsequent reactions to form imine groups and oxygen-containing groups.
Difficulties and challenges occurring during this w ork stim ulated several
suggestions for subsequent research. First, to better identify prim ary amino and other
functional groups, some derivatization reactions can be applied. M eanw hile, the plasm a
functionalization depth o f Cyclotene should be investigated in m ore detail using angleresolved XPS technique, since understanding the depth o f different derivatization
reactions is very im portant to label prim ary amino and other functional groups.
To better understand volum e and surface reactions o f transient plasm a species, an
extended gas phase diagnostics study is desirable. Optical in-situ m easurem ent technique,
e.g., Optical Em ission Spectroscopy (OES), is an ideal tool to investigate densities o f
reactive species in plasm a gas phase. Such a study could provide useful inform ation for
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
142
evaluating and im proving the proposed plasm a kinetic chem istry m odel and am ination
model.
The plasm a chem istry and am ination m odels developed in this w ork were
dem onstrated to be good descriptive models o f the observed Cyclotene am ination
behavior. The predictive nature o f the m odel should be explored through sim ulations over
a w ide operational w indow to identify optimal process conditions. Once identified, these
optim al conditions could be experim entally tested.
The dependence o f aging effects on tem perature should be investigated and an
apparent activation energy (or energies) would be extracted. These values w ould likely
provide valuable clues for reaction m echanism s in different aging environm ents.
To enhance the applications o f plasm a m odified polym ers, actions should be
taken to reduce post-plasm a surface relaxation processes. As w e know, surface radicals
are the driving force for post-oxidation o f polym ers. I f we could find a way to quench the
plasm a treated surfaces efficiently, it surely w ould have a great influence on m any
applications o f plasm a treated surfaces. The m ethyl radical is considered to be a
prom ising candidate to saturate open reaction sites without changing surface
functionalities. Keudell et al . 5 has suggested a m ethod for producing m ethyl radicals in
w hich methyl and atomic hydrogen radical beam s were used as source beam to grow
m ethyl groups on the surface o f hydrogenated carbon film. The grow th synergism was
explained as: first, the incoming atomic hydrogens create dangling bonds via the
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
abstraction o f surface bonded hydrogen, and then these dangling bonds serve as
adsorption sites for incoming m ethyl radicals.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
144
References
1
Gengenbach, T. R.; Xie, X.; Chatelier, R. C.; Griesser, H. J. J Adhes Sci Technol
1994, 8 ,304.
2
Gerenser, L. J. J Adhes Sci Technol 1987, 1, 303.
3
Chatelier, R. C.; Xie, X.; Gengenbach, T. R.; Griesser, H. J. Langm uir 1995, 11,
2576.
4
W ilson, D. J.; Williams, R. L.; Pond, R. C. S urf Interface A nal 2001, 31, 397.
5
Keudell, A. V.; Schwarz-Selinger, T.; M eier, M .; Jacob, W. A ppl Phys Letters 2000,
76, 676.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
APPENDIX A
PLASMA SYSTEM OPERATING PROCEDURE
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
146
1
Safety:
The m icrow ave plasm a system can be operated in a safe m anner if appropriate
procedures and safety precautions are followed. For all o f these chem icals, use Safe
Cylinder H andling Procedures.
A m m onia is a kind o f corrosive and harm ful gas under pressure and is held in a
gas cabinet. Its PH A has been approved by the Safety M anager. Also, the M SDS sheets
for all o f these chem icals can be found on the sh elf o f the lab.
2
Emergency shutdown procedure:
Shut o ff all electrical pow er to the microwave pow er supply and to the vacuum
pum p and all the m ajor switches o f the gas cylinders once any one o f the potential
hazards occurs.
2.1 Operating procedure:
2.1.1 Pre-run checklist
•
Check a ll the valves in the gas lines perpendicular to the gas lines.
•
Check rough valve is closed and fo re lin e valve is open.
•
Check that the gas in let toggle is closed.
•
Check p o w e r on; m ech an ical p u m p on; H i Vac on.
•
Turn on chiller com pressor and circulator. Also m ake sure that the w ater level is
w ithin one inch o f the top. If the level is low, fill w ith DI water.
•
Check that the chiller w ater temperature set point is 20 °C.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
147
2.1.2 Pump startup (not typically performed):
•
M ake certain all o f the pum ping valves: (HV), (F) and (R), are closed.
•
Start the roughing pum ping by plugging it in.
•
Open the fore line (F) valve.
•
W hen the pressure on the Granville-Phillips fore line pressure readout is less than
20 m Torr turn on the turbo pump on the turbo pum p pressure controller.
•
The system is now ready for use.
2.1.3 Pumps down the chamber:
•
Close the nitrogen needle valve located in the w hite flow controller box.
•
Tighten the reactor cham ber door.
•
Close fore line (F) valve and ensure that the high vacuum (FIV) valve is closed.
N EV ER OPEN THE H IGH VACU U M V ALVE W H ILE THE R EA C TO R
CH AM BER IS AT ATM OSPHERIC PRESSURE.
•
Open the roughing (R) valve. THE FORELINE V A LV E AND R O U G H IN G
V ALVE SHOULD N EV ER BE OPEN AT THE SAM E TIME.
•
Let the cham ber pum p dow n to ~30 mTorr on the cham ber pressure readout.
•
Close the roughing valve.
•
Open the high vacuum valve.
•
Open the fore line valve.
•
Switch o ff the nitrogen purge line on the mass flow control panel.
•
The cham ber should pum p down to 7x 10'7 Torr w ithin 30 m inutes to one hour
2.1.4 Bringing reactor chamber up to atmospheric pressure:
•
Close the high vacuum valve, labeled HV. The fore line valve, labeled F, should
be open and the roughing valve, labeled R, should be closed.
•
U nscrew the black knob on the quartz w indow door on top o f the reactor
chamber.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
148
•
Open nitrogen purge needle valve located on the purge line in the w hite flow
controller box.
•
Turn on the nitrogen purge flow by turning on the purge switch on the gas flow
control panel.
•
Open the nitrogen valve leading into the plasm a chamber.
•
As soon as the pressure is -7 4 0 Torr as indicated on the MKS Type 146 pressure
readout, the chamber door can be opened.
2.1.5 Replacing a sample:
•
Remove the sample holder from the reactor cham ber carefully by pulling straight
up. Place new sample on holder by placing the corners o f the sam ple underneath
the four washers. N ext, gently tighten the screws.
•
Place the sample holder assem bly into the reactor cham ber by gently pushing it
onto the connector wires.
•
Close the reactor cham ber door, but DO N OT tighten the knob.
2.1.6 Start ammonia plasma treatment
•
Check GCF (gas correction factor) and setpoint on the 247 C M FC pow er supply
readout.
•
Turn on N 2 flow to N upro valve.
•
Turn on gas flow at cylinder and at black valve in gas cabinet
•
Turn on flow at 247 C.
•
Open am m onia valve leading to chamber.
•
Start temperature controller.
•
Set pressure by adjusting throttle valve.
•
Strike plasma, run for desired period o f time and shut o ff m icrow ave pow er
supply when done.
2.1.7 Stop an ammonia plasma treatment run:
R e p r o d u c e d with p e r m i s s io n of t h e co p y rig h t o w n e r. F u r th e r r e p ro d u c tio n p rohibited w ith o u t p e r m is s io n .
149
•
Open throttle valve
•
Shut o ff gas flow at cylinder and at black valve in gas cabinet.
•
W ait for the am m onia flow to reach 0 seem.
•
Turn o ff flow at 247 C.
•
Turn o ff N 2 flow to Nupro valve.
•
W ait 5~ 10 sec.
•
Close am m onia valve leading to cham ber
2.1.8 Purge the system
•
Close high vacuum valve (HV).
•
Close fore line valve (F).
•
Open roughing valve (R)
•
Run N 2 through the rough pum p at a pressure o f ~2 Torr for 10 min.
•
Turn o ff N 2 flow.
•
Close roughing valve (R)
•
W hen pressure < 20 m Torr on Granville-Phillips fore line pressure readout, open
fore line valve (F)
•
O pen high vacuum valve (HV).
•
R un A rgon through turbo at a flowrate o f 40 seem for 30 min.
2.1.9 Daily shutdown
•
Put a clean silicon sample on the sample holder.
•
Put the sample holder in the reactor chamber.
•
Follow “pumping down the cham ber” .
•
Turn o ff the chiller compressor and circulator.
•
T urn o ff all mass flow controller panel switches and close all gas lines leading
into the plasm a chamber.
•
Turn o ff the microwave power supply.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
150
2.1.10 Pump shutdown (not typically performed):
Bring the cham ber to atm ospheric pressure.
•
Turn o ff the turbo pum p on the turbo pum p controller.
•
Switch the knob on the side o f the turbo pump. This will vent the turbo in order to
slow it down.
•
Slowly (over one minute) open the knob until it is out o f the housing.
•
The turbo will eventually stop spinning.
•
U nplug the roughing pum p by turning the plug to the left and then pull.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
APPENDIX B
BOND DISSOCIATION ENERGIES OF SELECTED
SPECIES
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
152
Dissociation bond energies*
Bond type
D issociation energy (kJ/m ol)
H -H
436
0=0
498
C -H
413
N -H
391
C -C
345
C -F
488
N=N
945
ON
890
N =N
614
C=N
456
C = 0 (amide)
748
C=C
610
C -N
305
C -0
357
N -C O
360
N —0
230
C -S i
347
S i- 0
460
H -F
564
H -0
464
C6H 5-vinyl
451
C = 0 (aldehyde)
740
C = 0 (ketone)
744
*The bond energies listed here were taken from the follow in g sources:
Sanderson, R. T., Polar C ovalence, A cadem ic Press, N e w York, 1983.
Sanderson, R. T., Chem ical Bonds and B ond Energy, A cadem ic Press, N e w York, 1976.
R e p r o d u c e d with p e r m is s io n of t h e cop y rig h t o w n e r. F u r th e r r e p r o d u c tio n prohibited w ith o u t p e r m is s io n .
Документ
Категория
Без категории
Просмотров
0
Размер файла
3 317 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа