close

Вход

Забыли?

вход по аккаунту

?

Investigation of microwave irradiation as an energy source in polymerization reactions

код для вставкиСкачать
INFORMATION TO U SERS
This manuscript has been reproduced from the microfilm m aster. UMI films the
text directly from the original or copy submitted.
Thus, som e thesis and
dissertation copies are in typewriter face, while others may be from any type of
com puter printer.
The quality of this reproduction is dependent upon th e quality o f th e copy
su b m itted .
Broken or indistinct print colored or poor quality illustrations and
photographs, print bleedthrough, substandard margins, and improper alignment
can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and
there are missing pages, these will be noted. Also, if unauthorized copyright
material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning
the original, beginning at the upper left-hand comer and continuing from left to
right in equal sections with small overlaps.
Photographs included in the original manuscript have been
reproduced
xerographically in this copy. Higher quality 6” x 9” black and white photographic
prints are available for any photographs or illustrations appearing in this copy for
an additional charge. Contact UMI directly to order.
Bell & Howell Information and Learning
300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA
UMI
800-521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
INVESTIGATION OF MICROWAVE IRRADIATION AS AN ENERGY SOURCE
IN POLYMERIZATION REACTIONS
Xiaomei Fang, Ph.D.
University of Connecticut, 1999
Variable frequency microwaves were investigated as an energy source to cure
unidirectional carbon fiber reinforced phenylethynyl-terminated polyimide composites,
and to synthesize the poly(e-caprolactam), poly(e-caprolactone) and copoIy(amideester) via ring opening polmerization.
The mechanism of the thermal and microwave cure reactions of a phenylethynylterminated imide model compound, 3,4’-bis[(4-phenylethynyl)phthalimido]diphenyl
ether (PEPA-3,4’-ODA) and a phenylethynyl-terminated imide oligomer PETI-5 (Mn ~
5000 g/mol) was studied by kinetics and solid-state 13C nuclear magnetic resonance
(NMR) spectroscopy. Both the model compound and PETI-5 exhibited much lower
activation energies and higher rate constants by the microwave cure process than by the
thermal cure process. Solid-state I3C-NMR studies revealed that the major cure reaction
in both the model compound and PETI-5 resin is an ethynyl to ethynyl addition
reaction, with a minor reaction to further form carbon-carbon single bonded structures .
Microwave energy was successfully applied to fabricate carbon fiber reinforced
phenylethynyl-terminated polyimide composites, PETI-5/IM7, with higher glass
transition temperatures (by 11° to 16°C) and enhanced mechanical properties at both
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Xiaom ei Fang — University of Connecticut, 1999
room temperature and 177°C and in one-half the time, compared to the standard thermal
process. Equivalent physical and mechanical properties were obtained from microwave
synthesized nylon-6 and poly(e-caprolactone) in reduced time relative to the
commercially produced thermal products. Anionic copolymerization o f e-caprolactam
with s-caprolactone via microwave irradiation produced poly(e-caprolactam-co-ecaprolactone) with higher yield, higher amide content, and higher Tg’s relative to the
thermally produced copolymer.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
INVESTIGATION OF MICROWAVE IRRADIATION AS AN ENERGY SOURCE
IN POLYMERIZATION REACTIONS
Xiaomei Fang
B.S. S h an g h ai Jiao Tong University, 1992
M.S. University of Connecticut, 1997
A Dissertation
Subm itted in Partial Fulfillment of the
R equirem ents for the D egree of
Doctor of Philosophy
at the
University of Connecticut
1999
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number. 9949658
Copyright 1999 by
Fang, Xiaomei
All rights reserved.
___
®
UMI
UMI Microform9949658
Copyright 2000 by Bell & Howell Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
Bell & Howell 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.
Copyright by
Xiaomei Fang
1999
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPROVAL PAGE
Doctor of Philosophy D issertation
IN VESTIG A TIO N O F MICROW AVE IR R A D IA TIO N A S AN E N E R G Y
SO U R C E IN POLYMERIZATION R E A C TIO N S
P resen ted by
Xiaomei Fang, B.S., M.S.
Major Advisor
DaniefA. S cola
A ssociate Advisor
Ja m e s P. Bell
c.
A ssociate Advisor
Anthony T. D iB enedetto
University of C onnecticut
1999
ii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGMENTS
I would like to express my sincere and deep gratitude to Prof. Daniel A. Scola,
my major advisor, for his guidance, support, understanding and encouragement through
the entire course o f this research. I would like to thank my associate advisors, Prof.
James P. Bell and Prof. Anthony T. DiBenedetto, for their professional advice and help.
Special thanks go to Prof. Samuel J. Huang for his direction and suggestions in the ring
opening polymerization section. Advice from Prof. Steven A. Boggs is also highly
appreciated.
I am very grateful to Dr. X.-Q. Xie for the assistance in the solid state NMR
studies. My cordial appreciation also goes to Dr. Ron Hucheon o f Microwave Properties
North (MPN), Ontario, Canada, for the dielectric properties measurement; Dr. Celene
DiFrancia, Dr. Hank Temme and Mr. Ron Jones o f Loctite Corporation for the
mechanical properties measurement; Dr. Zak Fathi, Dr. Billy Wei, Dr. Denise Tucker,
Mr. Keith Hicks and Dr. Richard Garard of Lambda Technologies, Inc. for the help in
equipment set-up and initial input in experimental design; Dr. Paul Hergenrother of
NASA Langley Research Center and Dr. Dan Reynolds o f Northrup-Grumman
Corporation for providing the polyimde resin and prepreg for this research. I am
grateful to the Yankee Ingenuity Grant of the State o f Connecticut, 96G059, for
financial support,
I am deeply thankful to the IMS faculty for their academic instruction, and to
IMS staff in the machine shop, electronic shop, IR lab, GPC lab, SEM lab, EIRC lab,
thermal lab, and rheology lab for their valuable assistance and patience. My work would
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
not have been done so well without their help and cooperation. Many thanks are
extended to my colleagues and friends in IMS for friendship and help in various ways,
especially to my present and former associates in Dr. Scola’s group, Eleonora, Chris
and Yumin, for all the cooperation, sharing and understanding.
My final appreciation goes to my brothers for being there for me all the time.
This dissertation is dedicated to my parents and grandma for their priceless love and
devotion.
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CONTENTS
Page
CHAPTER 1 - GENERAL INTRODUCTION
1.1. Microwave Irrad iation ---------------------------------------------------------------------1
1.1.1
Basic T h e o ry ------------------------------------------------------------------------ 1
1.1.2
Advantages over Thermal H eatin g --------------------------------------------- 3
1.1.3
Variable Frequency Microwave F u rn a ce ------------------------------------- 6
1.1.4
Temperature Control -------------------------------------------------------------- 8
1.2. Background of Materials to be Investigated--------------------------------------11
1.2.1
Advanced Phenylethynyl-terminated Polyimides ------------------------
11
1.2.2
Polymers from Ring Opening Polymerization ----------------------------- 13
1.3. Objectives of R esearch --------------------------------------------------------------------14
CHAPTER 2 - KINETIC STUDY OF THERMAL AND MICROWAVE CURE
REACTIONS OF A PHENYLETHYNYL-TERMINATED IMIDE MODEL
COMPOUND AND IMIDE OLIGOMER (PETI-5)
2.1. Introduction-------------------------------------------------------------------------------
15
2.2. Experim ental------------------------------------------------------------------------------- 17
2.2.1. Materials ----------------------------------------------------------------------------
17
2.2.2.
Dielectric Property Measurement --------------------------------------------- 17
2.2.3.
Thermal Cure Reaction Set-up -----------------------------------------------
19
2.2.4.
Microwave Cure Reaction Set-up --------------------------------------------
19
2.2.5.
Kinetic Study o f PEPA-3,4’-ODA by ER ----------------------------------- 20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.2.6. Kinetic Study o f PETI-5 by D S C -------------------------------------------- 20
2.3. Results and D iscu ssion ------------------------------------------------------------------ 21
2.3.1. Kinetic Study of Thermal Cure ----------------------------------------------- 21
2.3.2. Dielectric Properties of Model Compound and Imide Oligomer
(PETI-5) in the Microwave R eg io n ------------------------------------------ 30
2.3.3. Kinetic Study of Microwave Cure ------------------------------------------- 38
2.4. C onclusions---------------------------------------------------------------------------------- 45
CHAPTER 3 - INVESTIGATION OF MICROWAVE PROCESS TO
FABRICATE UNIDIRECTIONAL CARBON FIBER REINFORCED
PHENYLETHYNYL-TERMINATED POLYIMIDE, PETI-5/IM7
3.1. Introduction---------------------------------------------------------------------------------48
3.2. Experim ental------------------------------------------------------------------------------- 49
3.2.1. Materials --------------------------------------------------------------------------- 49
3.2.2. Thermal Process Conditions -------------------------------------------------- 50
3.2.3. Microwave Process Conditions ---------------------------------------------- 50
3.2.4. C haracterization----------------------------------------------------------------- 51
3.3. Results and D iscu ssion ------------------------------------------------------------------ 56
3.1.1. Process Conditions -------------------------------------------------------------- 56
3.1.2. Thermal and Physical Properties --------------------------------------------- 59
3.1.3. Mechanical Properties ---------------------------------------------------------- 67
3.1.4. Failure Surface Analysis ------------------------------------------------------- 69
3.4. C onclusions---------------------------------------------------------------------------------- 70
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 4 - INVESTIGATION OF RING OPENING POLYMERIZATION
VIA MICROWAVE IRRADIATION
4.1. Introduction-------------------------------------------------------------------------------- 79
4.1.1. s-Caprolactam Background --------------------------------------------------- 79
4.1.2.
e-Caprolactone B ackground-------------------------------------------------- 81
4.1.3.
Copolymerization B ackground-------------------------------------------------82
4.2. Experimental------------------------------------------------------------------------------- 85
4.2.1.
Starting Materials --------------------------------------------------------------- 85
4.2.2.
Dielectric Property Measurement ------------------------------------------- 86
4.2.3.
Microwave Equipment and Reaction Set-up ------------------------------ 86
4.2.4.
Microwave Synthetic Procedure -------------------------------------------- 87
4.2.5.
Thermal Synthesis o f Poly(e-caprolactam-co-e-caprolactone) -------
87
4.2.6.
Characterization ----------------------------------------------------------------
88
4.3. Results and D iscussion------------------------------------------------------------------ 89
4.3.1.
Dielectric Properties in the Microwave Region--------------------------- 91
4.3.2.
Ring Opening Polymerization of e-Caprolactam via Microwave
Irradiation ------------------------------------------------------------------------- 92
4.3.3.
Ring Opening Polymerization of e-Caprolactone via Microwave
Irradiation------------------------------------------------------------------------ 101
4.3.4.
Ring Opening Copolymerization of e-Caprolactam and
e-Caprolactone via Microwave Irradiation -------------------------------
104
4.4. Conclusions--------------------------------------------------------------------------------110
vii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 5 - A SOLID STATE 13C NMR STUDY OF THE CURE
REACTIONS OF 13C-LABELED PHENYLETHYNYL END-CAPPED
POLYIMIDES
5.1. Introduction--------------------------------------------------------------------------------112
5.2. Experim ental------------------------------------------------------------------------------ 115
5.2.1. General Inform ation-------------------------------------------------------------- 115
5.2.2. Synthesis o f l3C Labeled PEPA-3,4’-ODA ------------------------------
115
5.2.3. Synthesis o f 13C Labeled PETI-5 -------------------------------------------- 116
5.2.4. Characterization ----------------------------------------------------------------- 117
5.2.5. Solid State NM R Technique -------------------------------------------------- 118
5.3. Results and D iscu ssion ---------------------------------------------------------------118
5.3.1. PEPA-3,4’-ODA Model Compound Studies ----------------------------
119
5.3.2. PETI-5 Studies ------------------------------------------------------------------ 131
5.4. C onclusions------------------------------------------------------------------------------
142
REFERENCES ------------------------------------------------------------------------------------ 144
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
Table 1. Kinetic Analysis of Thermal Cure of PEPA-3,4’-ODA by IR ---------------- 28
Table 2. Kinetic Analysis of Thermal Cure of PETI-5 by DSC -------------------------- 34
Table 3. Kinetic Results of Microwave Cure of PEP A-3,4’-ODA ----------------------- 43
Table 4. Kinetic Results of Microwave Cure of PETI-5 ------------------------------------ 47
Table 5. Thermal and Microwave Cure Processes o f PETI-5/IM7 ----------------------- 55
Table 6. DMT A, TGA, TMA and Composition Results o f Microwave and
Thermally Cured PETI-5/IM7 Composites ---------------------------------------- 60
Table 7. Density, Composition and DMTA Results of Microwave Cured
PETI-5/IM7 Composites at Different Pressures ----------------------------------- 60
Table 8. Summary o f the Mechanical Results of Microwave and Thermally Cured
PETI-5/TM7 Composites ---------------------------------------------------------------- 71
Table 9. Summary o f the Mechanical Results of Microwave Cured PETI-5/IM7
Composites at Different Pressures ---------------------------------------------------- 72
Table 10. Summary o f Synthesis o f Poly(e-caprolactam) via Microwave Energy -- 100
Table 11. Synthesis o f Poly(e-caprolactone) via Microwave Energy ------------------
102
Table 12. Physical Properties of Poly(s-caprolactone) via Microwave Energy ----- 102
Table 13. Tensile Properties o f Poly(s-caprolactone) via Microwave Energy ------
103
Table 14. Temperature Effect on Microwave Copolymerization ----------------------- 107
Table 15. Catalyst Effect on Microwave Copolymerization -----------------------------
109
Table 16. Comparison of Microwave Energy and Thermal Energy in
Copolymerization Reactions ------------------------------------------------------- 110
Table 17. NMR Integration Results o f 13C-labeled PEPA-3,4-ODA
ix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
at a spinning rate o f 5 k H z ---------------------------------------------------------- 134
Table 18. NMR Integration Results o f l3C-labeled PEPA-3,4-ODA
at a spinning rate o f 10kHz --------------------------------------------------------- 134
Table 19. NMR Integration Results of I3C-labeIed PETI-5
at a spinning rate o f 8kHz ---------------------------------------------------------- 138
Table 20. NMR Integration Results of 13C-labeled PETI-5
at a spinning rate o f 1 0 k H z--------------------------------------------------------- 138
X
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
Figure 1. Heat-flow Profiles o f Thermal and Microwave Heating ------------------------- 4
Figure 2. Comparison o f Power Distribution of Microwaves ---------------------------------7
Figure 3. Variable Frequency Microwave Equipment ---------------------------------------- 9
Figure 4. Temperature Recordings o f Luxtron Fiber Optic Probe and Omega TC — 10
Figure 5. IR Set-up for Thermal Cure of Model Compound ------------------------------ 22
Figure 6. Set-up for Microwave Cure o f Model Compound and Oligomer ------------ 22
Figure 7. IR Spectra o f the Thermal Cure of PEPA-3,4’-ODA at 373°C --------------- 24
Figure 8. Comparison o f Extent o f Cure vs. Time at 373°C between Uncorrected
and Imide Carbonyl Corrected ------------------------------------------------------- 25
Figure 9. Extent of Cure vs. Time o f PEPA-3,4’-ODA at Various Temperature ----- 25
Figure 10. Plot of lnC^C vs. Time for Thermal Cure of PEPA-3,4’-ODA ------------ 27
Figure 11. Plot of ln£ vs. 1/T o f Thermal Cure of PEPA-3,4’-ODA -------------------- 27
Figure 12. DSC Thermograms of PETI-5 before and after C u re-------------------------- 28
Figure 13. Extent o f Cure vs. Cure Time o f PETI-5 from DSC Tg D ata---------------- 31
Figure 14. Relationship of Extent o f Cure and Tg of PETI-5 ----------------------------- 32
Figure 15. Plot of ln(l-x) vs. Time o f Thermal Cure o f PETI-5 (1st Order) ----------
32
Figure 16. Plot of (l-x)‘1/2 vs. Time o f Thermal Cure o f PETI-5 (1.5th Order) ------- 33
Figure 17. Plot o f InA: vs. 1/T o f Thermal Cure of PETI-5 --------------------------------- 33
Figure 18. Dielectric Properties o f PEPA-3,4’-ODA in Microwave Range ------------ 35
Figure 19. Dielectric Properties o f PETI-5 in Microwave Range ------------------------- 36
Figure 20. IR Spectra o f Microwave Cure of PEPA-3,4’-ODA at 320°C ---------------- 40
Figure 21. Kinetic Curve o f Microwave Cure of PEPA-3,4’-ODA via SiC Powder — 42
xi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 22. Plot o f ln(l-x) vs. Cure Time o f Microwave Cure o f PEPA-3,4’-ODA -- 42
Figure 23. Plot o f InA vs. 1/T o f Microwave Cure of Model Compound --------------- 43
Figure 24. DSC Thermograms o f Microwave Cure o f PETI-5 --------------------------- 44
Figure 25. Plot o f Extent o f Cure vs. Cure Time of PETI-5 via Microwave Cure — 46
Figure 26. Plot o f ln(l-x)'l/2 vs. Cure Time of PETI-5 via Microwave Cure ---------
46
Figure 27. Plot o f InA vs. 1/T o f Microwave Cure of PETI-5 ----------------------------- 47
Figure 28. Thermal and Microwave Cure C ycles----------------------------------------------- 53
Figure 29. Schematic o f Microwave Mold Design ------------------------------------------- 54
Figure 30. IR Spectrum o f Resin Extracted from Prepreg ---------------------------------- 58
Figure 31. TGA Curves o f PETI-5/IM7 Prepreg before and after Microwave
Curing at 250°C ------------------------------------------------------------------------- 58
Figure 32. TMA Thermograms o f Thermally Cured Composites ------------------------ 61
Figure 33. TMA Thermograms o f Microwave Cured Composites ----------------------- 62
Figure 34. DMTA Curves (1 Hz) o f Thermally Cured PETI-5/IM7 Composites —
63
Figure 35. DMTA Curves (1 Hz) o f Microwave Cured PETI-5/EVT7 Composites — 64
Figure 36. DMTA Curves (1 Hz) o f Microwave Cured PETI-5/1M7 Composites
at 250°C-0.5hr and 360°C-0.5hr under Different Pressures ------------------ 65
Figure 37. Comparison o f Processing Conditions and Properties between Microwave
and Thermally Cured PETI-5/IM7 Composites --------------------------- 73-74
Figure 38. ESEM Images o f Shear Failure Surfaces o f Microwave and Thermally
Cured PETI-5/IM7 Composites ------------------------------------------------ 75-78
Figure 39. Temperature-Time Profile of Microwave Process ------------------------------ 90
Figure 40. Dielectric Properties o f e-Caprolactam in Microwave Range -------------- 94
xii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 41. Dielectric Properties of e-Caprolactone in Microwave R a n g e -------------- 95
Figure 42. Dielectric Properties of Mixture o f e-Caprolactam and e-Caprolactone
in Microwave Range ------------------------------------------------------------------ 96
Figure 43. TGA Curves o f Microwave Synthesized- and Commercial Nylon-6 ----- 98
Figure 44. ER Spectra o f Microwave Synthesized Nylon-6 and PCL -------------------- 98
Figure 45. DSC Thermograms of Microwave Synthesized Poly(e-Caprolactam) —
99
Figure 46. DSC Thermogram o f Commercial Nylon-6 ------------------------------------- 99
Figure 47. Intrinsic Viscosity Extrapolation o f Nylon-6 in 85% Formic Acid ------
100
Figure 48. Stress-Strain Plot o f Microwave Synthesized Poly(e-caprolactone) -----
103
Figure 49. IR Spectra o f Microwave Synthesized PAE with Different
Compositions -------------------------------------------------------------------------- 107
Figure 50. lHNMR Spectra o f PAE Prepared at Different Catalyst Levels -----------
108
Figure 51. DMTA Curve o f Microwave Synthesized PAE
(160°C for Vz hr, catalyst 2%) -----------------------------------------------------
109
Figure 52. IR Spectra o f 13C-labeled and Unlabeled PEPA-3,4’-ODA
and PETI-5 ----------------------------------------------------------------------------- 123
Figure 53. IR Spectra o f 13C-labeled and Unlabeled PEPA-3,4’-ODA
and PETI-5 ----------------------------------------------------------------------------- 124
Figure 54. Solution NMR Spectra of l3C-labeled a) PEPA (in DMSO-c/^),
b) PEPA-3,4’-ODA (in NMP + DMSCW«s),
c) PETI-5 (in NMP + D M SO -^) ------------------------------------------------ 125
Figure 55. Results o f Ti Measurement through Inverse Recovery Experiments —
Figure 56. Solid State I3C NMR Spectra o f 13C-labeled PEPA-3,4’-ODA
XIII
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
127
before and after Curing (at a spinning rate of 5kHz) ------------------------- 128
Figure 57. Solid State I3C NMR Spectra of l3C-Iabeled PEPA-3,4’-ODA
a) before and b) after Curing (at a spinning rate o f 10kHz) ---------------- 132
Figure 58. Simulation o f the Solid State 13C NMR Spectrum o f Cured l3C-labeled
PEPA-3,4’-ODA for Quantitative Purpose ------------------------------------- 133
Figure 59. NMR Spectrum o f Hexaphenylbenzene in Solution
o f CH2 CI2 + CDCI3 ---------------------------------------------------------------------136
Figure 60. Solid State l3C NMR Spectra of l3C-labeled PETI-5 before and
after Curing (at a spinning rate o f 10kHz) ------------------------------------- 137
Figure 61. A Comparison o f Solid State l3C NMR Spectra o f 13C-Iabeled and
Unlabeled PEP A-3,4’-ODA after Curing (at a spin rate o f 10kHz) ------
140
Figure 62. A Comparison o f Solid State l3C NMR Spectra o f I3C-labeled and
Unlabeled PETI-5 after Curing (at a spin rate o f 10kHz) ------------------ 141
xiv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 1: GENERAL INTRODUCTION
1.1 MICROWAVE IRRADIATION
Microwave irradiation is a developing technology and its applications to polymers
and composites have attracted considerable attentions since mid-^SOs^1"61. W ith the
development o f microwave equipment and better understanding o f the interaction
between microwaves and materials, microwave processing theory and practical
applications in this area have been extended and evaluated in recent years[4‘5'7l
Microwaves were preliminarily used as an alternate heating source for materials
processing in the early 1960s, a decade after the first commercial microwave oven was
built by Raytheon. Marketing for home microwave ovens was booming and brought the
“microwave heating” concept into millions of families through food heating since mid1970s. However, up to the present, only a few microwave applications, such as food
preparation, materials drying and rubber vulcanization, are used in the processing
industries. Continuous efforts are being devoted to utilizing microwave energy to
process materials in a more effective and practical way, and exploring broad and
efficient microwave applications on an industrial level. The study described in this
thesis is directed toward investigating microwave irradiation as an energy source in
polymerization reaction, as it compares to conventional thermal processes.
1.1.1 Basic Theory
Microwave energy results from the oscillating electromagnetic waves in the
frequency range o f 0.3 to 300 GHz, (i.e. 1 m to 1 mm wavelength), located between
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2
radio frequency (RF) and infrared (IR) in the electromagnetic spectrum. For most
polymers, which belong to nonmagnetic materials, the interactions between microwave
energy and polymers are determined by the polymer dielectric properties.
For microwave interaction, the polymer dielectric properties at the macroscopic level
depend on their dipole moments at the molecular level. The rapidly changing electric
fields induced by microwave energy cause changes of electronic reorientation and
distortions o f induced or permanent dipole moments of the sample in the cavity. The
friction caused by these changes manifests itself in heat energy. This indicates that the
energy is stored in the sample and is usually characterized by the loss factor o f the
complex permittivity, s". Once the microwave frequency is close to the natural
reorientating frequency o f the dipoles o f molecules, the optimum coupling interaction
between microwaves and materials is reached and ideal microwave heating occurs. In
addition, a moderate value of the real permittivity, e \ which is characterized as energy
absorption, is also required to provide certain penetration of microwaves into samples.
If a conductor is involved in a microwave field, like carbon fiber in a reinforced
polymer composite, the electrons in the carbon fiber can move freely in the electric
field, which will give rise to an electric current instead o f reorientation o f dipole
moments as experienced in polymers. Therefore, the resistance of the carbon fiber will
result in heating o f the fiber, and hence the composite in the microwave environment, as
long as it is not a superconductor. Quantitative evaluations o f microwave power
absorption, Pa, and penetration depth, Dp, o f polymers or their composites can be
calculated by[81:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
where co is the angular frequency of the microwaves, So is the permittivity of free space
1 *7
I
(8.85x10' Fm' ), E is the root mean square internal electric field (depending on s', s"
and the field in the microwave cavity), fJo is the permeability o f free space (4tcx10'7 H
m '1), and fi' is the relative permeability o f the medium (equals to 1 for nonmagnetic
materials). In this case, Dp, also called skin depth, is defined as the distance from the
surface into the sample at which the electric field strength is reduced to 1/e (-37% ). It
has been reported that at the frequency o f a commercial microwave, 2.45 GHz, the
penetration depth of a unidirectional carbon fiber reinforced epoxy composite, Hercules
AS4/3501-6 is 0.28mm and 1.7mm for the electric field along and perpendicular to the
fiber direction, respectively11*. Microwave power absorption obviously depends on the
loss factor of the permittivity, s ”, as shown in equation (1). For a polymer composite
system, s" is contributed by*8*:
s ” = sD” +■ o /( cosq)
(3)
where Sd ” is the loss factor from dipolar reorientation polarization while
o /((osq)
is the
loss factor from resistive heating.
1.1.2 Advantages over Thermal Heating
Microwaves can penetrate into materials to heat the molecules inside materials,
while conventionally heat is transferred from external heating sources through thermal
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
convection and conduction. The penetration ability of microwaves is attributed from
their longer wavelength nature relative to the thermal heating. Most conventional
thermal heating energy is located in the infrared range. Therefore, this thermal energy
cannot penetrate materials but can be absorbed on the surface and then transferred into
inside materials by thermal conduction. Microwave energy penetrates into the bulk of
the material to heat molecules directly. This heat then dissipates into air from the
surface o f materials. This provides a reverse heat-flow profile (inside—^outside) relative
to the conventional thermal heat-flow (outside—>inside), as illustrated in Figure 1.
Thermal Heating
Microwave Heating
Figure 1. Heat-flow Profiles of Thermal and Microwave Heating
As indicated in equation (1), microwave absorption varies with the dielectric
properties and chemical crystal structures o f materials. Materials with stronger dipoles
and high mobility o f molecules provide more effective and efficient microwave
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
absorption. Microwaves as an electromagnetic radiation can be transmitted, absorbed
and reflected according to the laws o f optics. There are three typical materials defined
by their interactions with microwaves/91 that is, 1) Transparent - microwaves transmit
through some materials with negligible energy loss due to the very low dielectric loss o f
these materials. Poly(tetrafluoroethylene) (PTFE, TEFLON) and quartz belong to this
category. 2) Opaque - microwaves are almost totally reflected by the materials, no
absorption or penetration. Metal blocks (without sharp angles) belong to this category.
3) Absorptive — microwaves are partially or totally absorbed by the materials,
generating thermal energy inside as microwaves transmit through the materials. The
different absorption depends on the interaction between microwaves and dielectric
materials. M ost organic compounds and polymers with large dipole moments absorb
strongly. Dielectric materials containing a conductive or magnetic phase, like graphite
fiber (or carbon fiber) reinforced polymeric composites or polymer matrices blended
with carbon powder as an additive, behave as mixed microwave absorbers. In such
multiphase materials, the graphite fiber and carbon powder will absorb microwaves and
then transfer heat to the surrounding polymer matrix, since conductive phase like
graphite fiber or carbon powder is a much stronger absorber o f microwaves than the
polymer matrix. Such microwave absorption pattern was reported in the ceramic
composites earlier191.
Microwave processing o f polymeric materials, especially epoxy resins and epoxy
resin/composites^'2’6’10' 11], has been widely investigated. Most experimental results have
demonstrated that microwave energy provides improved reaction kinetics and
mechanical properties relative to the thermal process.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
The applications o f microwave energy in organic chemistry covering the field from
the first record application in 1969 to many references up to 1991 were collected by
Abramovich1121. In the last decade, considerable efforts have been devoted to
investigating the advantages o f microwave irradiation over the conventional thermal
processing in heating and synthesizing organic and polymer materials.113'171
In summary, compared to the conventional thermal heating, microwave energy
provides selective, instantaneous and volumetric heating, which can be easily controlled
by changing the applied electromagnetic fields. The reverse heat-flow pattern provides
the great potential for rapid processing of thick-sections and complex-shaped matrices
or composites. The internal heating is believed to produce an efficient reaction since the
functional groups of the starting materials, which have strong dipole moments, are the
primary source o f activation in the microwave electromagnetic field. Effective
microwave absorption directly by the materials will expedite the whole process in a
reduced time, and lead to energy saving as well. Energy consumption in microwave
cooking (chicken and beef) is only 24 - 35% that in the conventional oven1181.
1.1.3 Variable Frequency Microwave Furnace
Most commercial microwave ovens, like ones commonly used in the kitchen at
home, have a fixed frequency at 2.45 GHz. However, in the materials processing,
materials have shown non-uniform heating pattern caused by the non-uniform
distribution o f electromagnetic energy in the cavity, as defined in equation (1). More
recently, variable frequency microwave ovens have been developed to provide better
heating uniformity during microwave processing by controlling the sweep bandwidth
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7
and sweep rate119-221. This idea is based on the knowledge that the hot spot location
during microwave heating is related to the applied frequency. The hot spot will shift to
another location with the change o f applied frequency and thus a more uniform heating
pattern can be achieved in a given cavity by the frequency in a certain range swept at a
proper sweep rate. The power-distribution patterns o f microwaves with fixed frequency
and variable frequency are compared in Figure 2.
(a) Fixed Frequency
(b) Variable Frequency
Figure 2. Comparison of Power Distribution of Microwaves
These patterns were generated by placing carbon paper on unidirectional carbon
fiber reinforced polyimide prepreg (PETI-5/IM7). Two Teflon blocks were placed on
the top and at the bottom to maintain good physical contact between prepreg and carbon
paper. Once the prepreg absorbs microwaves to generate heat energy that is transferred
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
s
to the carbon paper, the color o f the carbon paper will turn from white at room
temperature to black at high temperature. The higher the temperature, the stronger the
microwave electric field, the darker the color o f carbon paper will be. As shown in
Figure 2, variable frequency microwaves provided a much more uniform power
distribution than fixed frequency microwaves by generating more uniform heating
pattern in the prepreg.
A variable frequency microwave furnace (VFMF) (model LT502Xb) used in this
study was designed and built by Lambda Technologies, Inc., Raleigh, NC, and is
illustrated in Figure 3. This model VFMF contains an incident center frequency from
2.4 to 7.0 GHz, a pressure system capable o f providing a force o f 3200 lbs and two
temperature monitoring and control systems. One by Luxtron fluoroptic temperature
probes is used at low temperatures (<250°C) and the other by grounded chromega/Al Ktype OMEGA thermocouples used at higher temperatures (>250°C). The size o f the
processing cavity is 15 inch x 12 inch x 12 inch (38.1 cm x 30.5 cm x 30.5 cm), and the
whole cavity, made o f stainless steel, is grounded.
1.1.4 Temperature Control
Microwave processing proceeded by controlling input power or reaction time instead
o f reaction temperature in earlier microwave studies, due to the difficulties in
determining accurate temperature o f samples. The Luxtron fluoroptic temperature probe
has been designed for accurately measuring temperature in microwave environment
without disturbing the electromagnetic fields inside cavity. However, this fiber optic
probe can be used only at low temperatures less than 250°C, and the safe temperature
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
for repeated use is even less than 200°C thus far. Therefore, in this work, grounded
Omega K type thermocouples (TC) have been used to replace Luxtron optic probes to
measure temperature during the microwave process studies.
L u x tr o n
4 -C h an n el
T e m p e r a tu r e
O m ega
P re s s u re
C o n tr o lle r
.^ 1
.
ttS T
M ic ro w a v e .
D ir e a c ti o n a l
C o u s le r
M a n u a lly
S o lid S ta te
V acuum
S y ste m
TV.rT
V a r ia b le
A tte n u a tio n
A m p l if i e r
2 .4 -7 .0 G H z
TW T
P o w e r S u p p ly
V o lt a g e C o n tr o lle d
M i c r o w a v e O s c illa to r
(H ig h )
C P U - D a ta
C o n tr o l a n d
I n te r f a c e B o a r d
A c q u is itio n
L o w V o l ta g e
P o w e r S u r p ly
Figure 3. Variable Frequency Microwave Equipment
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
260
240
220
Si 200
£
5
2 180
a)
a.
160
E
a>H
140
Luxtron
120
—- Omega
100
0
2
4
6
8
10
12
14
Tim e interval
Figure 4. Temperature Recordings of Luxtron Fiber Optic Probe and Omega TC
It has been reported that the presence of metallic temperature probe would cause
distortion of the electrical field and inaccurate temperature measurement as a result of
self-heating in the microwave environment.18'231 To minimize such negative effects, the
thermocouples were well-grounded during microwave processing, as suggested by
Lambda Technologies, Inc. In order to check the temperature accuracy determined by
the grounded thermocouples in the microwave environment, simultaneous temperature
measurement was undertaken with the thermocouples and fiber optic probes. Both
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11
thermocouple and fiber optic probe were inserted into two pre-drilled holes in a
soapstone block (a good absorber o f microwaves) and placed close to each other. Figure
4 shows very good agreement in the temperatures measured by the grounded
thermocouple with these measured by the fiber optic probe. Grounded thermocouples
were therefore selected to monitor reaction temperatures during microwave processing
studies.
1.2 BACKGROUND OF MATERIALS TO BE INVESTIGATED
1.2.1
Advanced Phenylethynyl-terminated Polyimides
Thermosetting polyimides have been used as resins with high performance and
matrices in advanced composites in the last three decades. [24'25] Those polyimides are
usually prepared in three steps: a) the reaction of diamines with dianhydrides in a polar
aprotic solvent to form low molecular weight poly(amic acid) precursor, where at least
one o f diamines or dianhydrides contain end-capped groups capable o f forming
crosslinks or chain extension reactions; b) the reaction of low molecular weight
poly(amic acid) precursor cyclodehydrated to form imide oligomer; c) the reaction of
the end-cappers undergoing addition to eventually form crosslinked polyimides. The
need for polymers to resist thermal and thermo-oxidative degradation at elevated
temperature has been the essential driving force for the development o f polyimides.
Aromatic polyimides are thermally stable, high performance polymers but are often
difficult to process by compression or injection moulding methods owing to its infusible
nature. The melting processibility can be altered by the incorporation o f more flexible
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
units, such as arylene ether segments or other flexible functional groups within the
polymer backbone. However, certain properties, such as the modulus and glass
transition temperature, are compromised by this way. A series of imide oligomers
terminated with reactive groups such as maleimide (BMI), nadimide (PMR-15),
benzocyclobutene, ethynyl (Thermid®) have been developed and investigated to
improve the processibility of polyimides. Some o f these have been commercially
available and widely used. More recently, phenylethynyl-terminated polyimides126*291
have been investigated with particular significance. The end-capped phenylethynyl
groups react at a high temperature (~371°C) via an ethynyl-ethynyl addition reaction
without the release o f volatiles and offer an excellent combination of high thermal and
chemical stability, good toughness, high mechanical properties and a relatively wide
processing window.
Fiber reinforced polyimide materials are high temperature, high performance
materials which provide high strength and light weight at elevated temperatures (250 400°C) and have potential application in airframes, propulsion systems, missiles and
land vehicles, such as automobiles and armored tanks. The High Speed Civil Transport
(HSCT) was designed to carry 250-300 passengers and to fly at a supersonic speeds (a
proposed speed o f Mach 2.4 according to the USA version)1301. This requires high
performance materials for service at 177°C for 60,000 hrs. LaRC PETI-5 is a
phenylethynyl-terminated imide oligomer with a molecular weight around 5,000 g/mole
that has been selected as the high temperature, high performance material for
applications on the High Speed Civil Transport (HSCT).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
1.2.2
Polymers from Ring Opening Polymerization
Poly(e-caprolactam) (Nylon-6) and poly(e-caprolactone) (PCL) are commercially
produced from the cyclic monomers by conventional thermal methods to induce ring
opening polymerization. Nylon-6 is one o f the most important polyamides and widely
used as a synthetic fiber and as an engineering resin. The ring opening polymerization
o f cyclic lactams can be initiated by bases, acids, and water131'32'. Water is the most
often used initiator for commercial polymerization. The lactam is hydrolyzed in the
presence o f water to form amino acid, which will react with lactam to produce high
molecular weight polyamide (nylon 6). Anionic initiation is also applied. Cationic
initiation is not useful because it yields low conversion and low molecular weight.
However, on a lab scale, the use of the water-initiated process is difficult as a result o f
high pressure system required to reach high molecular weight. To avoid this, a small
amount of co-amino acid can be introduced into reaction system instead o f water. Ring
opening polymerization of e-caprolactam thus can be initiated by amino acid, and
reaction conversion follows a step polymerization mechanism.
Poly(e-caprolactone) is one of the most popular biodegradable polymers, which has
good compatibility with other polymers, such as polystyrene polypropylene, poly(vinyl
chloride) and low density polyethylene.'33'341 Poly(e-caprolactone) can be prepared from
lactone by using organo-stannous compounds as a coordination polymerization
catalyst.135'361 A wide range of initiators such as organometal catalyst and alkanolamine
can also be used to conduct anionic and cationic polymerizations o f lactones.
Copoly(ester-amide)s (PAE) combine the good material properties of amide groups and
the biocompatibility and biodegradability o f ester groups.[37'401 The biodegradability of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
poly(e-caprolactone) ester structure has been shown to occur in copolymers o f ecaprolactone and e-caprolactam, where degradation involves the enzymatic hydrolysis
o f the ester groups in the backbone leaving the amide bonds unperturbed.1411
1.3 OBJECTIVES OF RESEARCH
This research has been directed to investigate:
(1) microwave irradiation as an energy source, as compared with thermal energy, in the
(a) kinetics o f the cure of a phenylethynyl-terminated imide model compound and
imide oligomer to understand the reaction mechanism. (Chapter 2)
(b) cure
of
carbon
fiber
reinforced
phenylethynyl-terminated
polyimide
composites, PETI-5/IM7, to identify whether microwave processing would
provide equivalent or enhanced physical and mechanical properties. (Chapter 3)
(c) ring opening polymerizations of e-caprolactam, e-caprolactone and mixture o f
these monomers to synthesize nylon-6, poly(e-caprolactone) and copoly(amideester). (Chapter 4)
(2) the general cure reactions of phenylethynyl-terminated polyimides by solid-state
l3C-NMR technique. (Chapter 5)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 2: KINETIC STUDIES OF THERMAL AND MICROWAVE CURE
REACTIONS OF A PHENYLETHYNYL-TERMINATED IMIDE MODEL
COMPOUND AND IMIDE OLIGOMER (PETI-5)
2.1 INTRODUCTION
Aromatic polyimides prepared from the phenylethynyl-terminated imide monomers
or oligomers have demonstrated advantages due to the favorable processability and
good materials properties,
[26-29,42-47]
.
and have provide high performance composites
[48]
with broad potential applications. The advantages o f these materials are that (1) no
volatiles are formed during the curing reaction, (2) the phenylethynyl group exhibits
better process control relative to the ethynyl group due to the larger processing
window,1421 and (3) improved thermal and thermo-oxidative stability over ethynylterminated imide oligomers. [26'27'42,47! Several studies on the thermal cure o f various
ethynyl
[48-53]
and phenylethynyl
[54-58]
-terminated monomers and oligomers in an attempt
to understand the chemistry of the process have been also reported in recent years. The
crosslinked structure o f phenylethynyl-terminated imide polymer was also proposed/ 1
However, the cure reaction mechanism is still poorly understood, although the triple
bonds o f terminated phenylethynyl groups are expected to react by chain extension,
crosslinking and branching. The expected steric effect o f the phenyl-ended functional
groups, relatively low concentration of phenylethynyl groups in the oligomers and
insolubility o f the cured products, add to the difficulties in elucidating the cure reaction
mechanism.
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
Furthermore, in order to identify whether microwaves provide enhanced cure
reactions compared to the conventional thermal cure, microwave kinetic studies on the
model compound and PETI-5 resin were carried out. The chemical reactions occurring
by microwave irradiation relative to the thermal process are always open to question.
Generally speaking, crosslinking o f a polymer causes an increase in the glass
transition temperature, Tg. At a given temperature, Tg increases with the extent o f cure
until it reaches an asymptotic value. A good estimate o f the extent o f cure or
crosslinking, x, can be deduced from the glass transition temperature by the use o f the
DiBenedetto equation159"621 which has been modified1611 to investigate the cure reaction
o f thermosetting polymers.
Objectives -
To understand reaction mechanism via microwave irradiation,
compared with thermal energy, the kinetic studies o f the cure o f a phenylethynylterminated imide model compound and imide oligomer were investigated. In this
chapter, kinetic analysis o f the thermal and microwave cure reactions o f an imide model
compound, 3,4'-bis[(4-phenylethynyl)phthalimido]diphenyl ether (PEPA-3,4'-ODA)
was determined by Infrared Spectroscopy (DR.) by following the absorbance of the triple
bond. Differential Scanning Calorimetry (DSC) was used in determining the kinetics of
the cure o f a phenylethynyl-terminated imide oligomer with an estimated molecular
weight o f 5000g/mole (PETI-5) by following the increase in the glass transition
temperature, Tg, as a function of cure. The relationship between Tg and relative extent
o f cure, x, was defined by the modified form23 o f the DiBenedetto equation.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17
2.2 EXPERIMENTAL
2.2.1 Materials
1) 3,4'-bis[(4-phenylethynyl)phthalimido]diphenylether (PEPA-3,4'-ODA)
The imide model compound (Scheme 1) used in this study was PEPA-3,4'-ODA
prepared from 4-phenylethynylphthalic anhydride (PEPA) and 3,4-oxydianiline (ODA)
in DMF solvent. 4-phenylethynylphthalic anhydride(PEPA) (21.37g, 0.086mole) and
3,4'-oxydianiline
(ODA)
(8.67g,
0.043mole)
were
dissolved
in
N,N’-
dimethylformamide (DMF) respectively, and the solution was stirred for 12 hours at
room temperature. An excess o f acetic anhydride was then added and the solution was
refluxed for an hour, cooled and the precipitated product was filtered and air dried. The
crude product was recrystallized from DMF, rinsed with methanol and dried under a
vacuum at 200 °C for 6 hours. The light yellow powder was obtained with 87% yield,
m.p. peak at 297 °C (sharp) (DSC) at a heat rate of 20 °C/min. ER (KBr), 2214 (C=C),
1774 and 1716 (imide C=0). Analysis calculated for
C 44H 24N 2O 5:
C, 80.0%, H, 3.6%,
N, 4.3%; found: C, 79.46%, H, 3.99%, N, 4.31%.
2) Phenylethynyl-TerminatedImide Oligomer (PETI-5)
The imide oligomer PETI-5 (Scheme 2) used in this study was obtained from NASA
Langley Research Center. This material had a theoretical molecular weight about 5000
g/mole, and a glass transition temperature o f 228.4 °C (DSC) at 20 °C/min.
2.2.2 Dielectric Property Measurement
Complex dielectric constants o f PEP A-3,4’-ODA and PETI-5 were determined by
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
o
PEPA
3,4'-ODA
DMF
RT
w 12 hr
Acetic Anhydride
C \
:n
o
A
1 1 hr
o-
Q
o
PEPA-3,4'-ODA
Scheme 1. Synthesis of Imide Model Compound, PEPA-3,4’-ODA
C=C
c=c
3 0
o
PETI-5
O
where: A r =
3,4'-ODA
1,3-Bis(3-APB)
Scheme 2. Chemical Structure of Imide Oligomer, PETI-5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
Microwave Properties North (MPN), using a system developed based on the cavity
perturbation technique, in the microwave frequency range from 0.4 GHz to 3 GHz and a
temperature range from room temperature to 385°C holding for 1 hour. The heating rate
was controlled at 2°C/min. The data were provided by Dr. Ron Hutcheon, Microwave
Properties North (MPN), Ontario, Canada.
2.2.3 Thermal Cure Reaction Set-up
The thermal cure o f the model compound was studied by using an on-situ set-up cell
in Nicolet SX-60 FT-IR. The KBr pellet of PEPA-3,4'-ODA was placed in the IR cell
connected with a temperature control as shown in Figure 5. The thermal cure and ER
determination (transmission) were conducted simultaneously. The thermal cure o f
PETI-5 at different time-temperatures was accomplished by DSC method at a heating
rate 20 °C/min under nitrogen atmosphere at a flow rate 20 cm3/min. Temperatures were
calibrated by indium and zinc standards.
The PETI-5 was weighed and sealed in
aluminum DSC pans. The sample was cured and cool controlled by the DSC system.
2.2.4 Microwave Cure Reaction Set-up
A variable frequency microwave furnace (VFMF) model LT 502 Xb (Lambda
Technologies, Inc.) was used to microwave cure PEPA-3,4’-ODA and PETI-5. Sample
was put in glass tube and buried in silicon carbide. Silicon carbide powder was used as
fugitive microwave absorber. The temperature was measured and controlled by a
grounded Omega K type thermocouple, which was calibrated by a Luxtron optic fiber
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
temperature probe beforehand. The grounded thermocouple was inserted directly into
the sample, as illustrated in Figure 6.
2.2.5. Kinetic Study of PEPA-3,4’-ODA by DR.
Infrared spectra were taken using Nicolet SX-60 and Nicolet Magna 560 FT-IR
systems. The KBr pellet was prepared samples was ground to powder. The absorbance
o f the triple bonds (2214 cm '1) of phenylethynyl groups as a function o f time, at specific
temperatures was measured. Based on the Beer-Lambert’s law, the extent o f cure, x,
was calculated by the ratio of absorbance o f triple bond during cure to the absorbance
before cure, normalized to imide carbonyl bonds at 1778 and 1716 cm '1, expressed as
follows:
X
=
^ ^ im td e C - O
1 —
)
f
£4^
(ACsC ^AimtdeC=o)l==0
where Ac^c and Aim,deC=o are absorbance o f the terminal ethynyl C=C at 2214 cm'1 and
the imide carbonyls C = 0 at 1778 and 1716 cm '1, t = 0 refers to initial absorbance at
time = 0, and t is absorbance at 2214 cm'1 when the sample temperature reaches the
melt state (~290°C).
2.2.6. Kinetic Study of PETI-5 by DSC
DSC measurement was performed by a Perkin-Elmer DSC, 7 series Analysis System
at a heating rate 20 °C/min under nitrogen at a flow rate 20 cm3/min. Temperatures were
calibrated by indium and zinc standards.
The original DiBenedetto equation as follows was used to calculate the reaction
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
extent, x, by determining Tg for a highly crosslinked network:
— — — = ----- —-----
(5)
where Tgo and Tg* refer to the glass transition temperature before curing and after fully
curing respectively, Tg is the glass transition temperature of the sample after microwave
cure at each temperature for a specific time, and X is the ratio o f isobaric heat capacity
o f fully cured material,
capacities, ACpoc and
A C p«,
ACpo.
to that of uncured material,
were determined by
D SC
ACpo.
The isobaric heat
through the glass transition
temperatures at a heating rate of 5°C/min under a nitrogen atmosphere.
In addition, even if the selected Tg00 does not correspond to the theoretically ideal
fully cured state, the extent of cure, x, is still valid for kinetic analysis according to the
modified DiBenedetto equation. In this case, Tg0o, A., and * are substituted by TgM, V and
x ’ respectively, where TgM represents Tg of a network, A.’ is the ratio of ACpm to ACpo
and x ’ refer to x/xM, a relative reaction extent. Since the above substitution only changes
the intercept and not the slope o f kinetic curves discussed below, it assures that it has no
effect on the rate constant, reaction order or activation energy.
2.3 RESULTS AND DISCUSSION
2.3.1 Kinetic Study of Thermal Cure
1) IR Analysis o f Thermal Cure o f PEPA-3,4’-ODA
The infrared (IR) spectra o f uncured and cured PEPA-3,4'-ODA are shown in Figure
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
Cartridge Heater
Thermocouple
KBr Crystal
Metal Holder
Figure 5. IR Set-up for Thermal Cure of Model Compound
•G ro u n d e d T C
S iC b a th
MW
S a m p le
Figure 6. Set-up for Microwave Cure of Model Compound and Oligomer
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
7.
The decrease in the absorbance at 2214 cm'1 with cure time indicates that the
phenylethynyl triple bonds have reacted during thermal cure.
Relative to the imide
carbonyl band, absorption o f the ethynyl group at 2214 cm '1 is weak. However,
magnification o f the region 2700 cm'1 to 2000 cm '1 allows for more accurate
measurements o f the absorption at 2214 cm'1. The changes o f integrated peak intensity
at wavenumber 2214 cm '1 (vC=C) with an increase in cure time were measured at four
different temperatures,
318°C,
336°C, 355°C
and
373°C.
For each infrared
measurement, the sample thickness, b, was kept constant. To determine whether the
absorptivity coefficient, e, followed Beer Lambert’s law A = sbc, where A =
absorbance, e = the absorptivity coefficient, b = sample thickness and c = concentration,
the absorbance at 2214 cm '1 was corrected by normalizing to the absorbance o f the
imide carbonyl group at 1778 and 1716 cm'1, assuming that the imide groups remained
constant during the cure process. The data (Figure 8) show that there is no obvious
change in absorptivity coefficient e. Therefore, other absorbances at 2214 cm'1 at the
various isotherms were used without normalization.
The results (Figure 9) show that the extent of reaction increased with time, and the
increased with temperatures, as expected. It is interesting to note that after 90 minutes
at 373°C, the extent o f cure is 90.4% indicating that about 9.6% o f ethynyl groups
remain after these cure conditions. This presence o f ethynyl group after 90 minutes at
373°C may be due to the restricted mobility or diffusion o f the phenylethynyl groups as
the material vitrifies to a crosslinked network.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
Absorbance
AT 373 C
15
30
Z2SO
2090
1030
.
Wave n u m b ers
Figure 7. IR Spectra of the Thermal Cure of PEPA-3,4’-ODA at 373°C
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
1.0
O
0.8
-
O
X
®"
O
©
0.6
"o
©
® 0.4 X
LU
©
8
0.2 - \
O
O
20
40
Uncorrected
Imide 0 = 0 Corrected
80
60
100
Time (min)
Figure 8. Comparison of Extent of Cure vs. Time at 373°C between Uncorrected
and Imide Carbonyl Corrected
1.0
•
0 .8
-i
®
3 0.6 H
a
s
X
LU
318 “C
336 °C
355 °C
373 °C
o.4 -
0.2
0.0
120
Time (min)
Figure 9. Extent of Cure vs. Time of PEPA-3,4’-ODA at Various Temperature
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
26
Plots of In ( O C ) vs. time for each temperature (Figure 10) show that the reaction
follows first order kinetics. Linear regression analysis using least squares method was
used to fit the data. The kinetic rate constants, k, derived from the slopes o f each
reaction isotherm, are listed in Table 1. Assuming that the Arrhenius relationship,
k=Ae
-E /R T
between temperature and rate constant holds, an apparent activation energy
Ea = 40.7 ± 2.7 kcal/mole was obtained from the plot of In k vs. 1/T (Figure 11).
2) DSC Analysis o f Thermal Cure o f PETI-5
Differential scanning calorimetry (DSC) has been used to determine the cure o f the
thermosetting phenylethynyl-terminated PETI-5 imide oligomer by integration o f the
exothermic peak due to the ethynyl-ethynyl cure reaction.[58* The glass transition
temperature, Tg, was also used to follow the cure in that study, but no kinetic
correlation was given. In this study, using the modified DiBenedetto equation (5), the
extent of the cure reaction o f PETI-5 was determined from Tg data obtained by the DSC
method. An illustration o f the T g’s before cure and after cure at 380°C for 60 minutes is
shown in Figure 12.
The extent o f cure, x, was calculated from the DiBenedetto equation as described in
the experimental section. The Tgo of uncured material is 228.6°C, and a fully cured
material is defined as the Tg« (269.9°C) after cure at 380°C for lhr. The isobaric heat
capacities ACP before cure and after cure at 380 °C for one hour were 0.26 ± 0.03 J/g °C
and 0.18 ± 0.03 J/g °C, respectively. For example, the Tg o f a sample cured at 360 °C
from 60 minutes is 264.1 °C. The extent of cure, x, calculated for this sample cured at
360°C for 60 min is 0.942. The kinetic curves for the extent of cure versus time for
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
-0.5
c
a
318 C
•
336 C
■ 355 C
-2.5
♦ 373 C
0
50
100
200
150
Time (min)
Figure 10. Plot of InC=C vs. Time for Thermal Cure of PEPA-3,4’-ODA
C
-1 0
-i--------
0.00154
0.00158
0.00162
000166
0.0017
1 /T
Figure 11. Plot of ln& vs. 1/T of Thermal Cure o f PEPA-3,4’-ODA
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
64 -
62 -
60 -
E. 58 -
56 X
54 -
52 -
50 -
100
200
300
400
500
Temperature (°C)
Figure 12. DSC Thermograms of PETI-5 before and after Cure
Table 1. Kinetic Analysis of Thermal Cure of PEPA-3,4’-ODA by IR
Temperature (°C)
Rate Constant (min'1)
Regression Coefficient
318
0.00157
0.991
336
0.00341
0.994
355
0.01229
0.997
373
0.02744
0.983
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
samples cured at the various temperatures 350 °C, 360 °C, 370 °C, 380 °C, and 390 °C
are shown in Figure 13.
As mentioned above, the fully cured state was defined as the Tg after thermal
treatment at 380°C for 1 hour.
Even if this assumption is not accurate, it has no
significant effect on the kinetic results. The significant effect in the study is that the
glass transition temperature, Tg, continues to increase as the cure temperature and time
increase even though the exothermic peak due to cure could not be detected. The
relationship between Tg and the relative extent of the cure reaction is displayed in
Figure 14. In a recent publication, Venditti and Gillham163' adapted the Couchman
equation1641, (which related the glass transition temperature to compositional variation
o f a polymer system), to model the Tg versus fractional conversion, x, relationship of
reactive thermosetting polymers. Based on a comparison o f measured ACpoo/ACpo values
derived from DSC measurements and predictive values, ACpoo/ACpo, calculated from the
modified Couchman and modified DiBenedetto equations, the authors stated that their
modified equation appears to be a better prediction tool than equation (5). In the present
work, the assumption inherent in equation (5), that is, ACpocl/T, was accepted and the
equation was used as a tool to calculate x required to follow the kinetics of the
crosslinking reaction. The fact that the overall reaction does not follow simple order
kinetics as shown below is most likely attributed to the complexity o f the reaction, and,
to a lesser extent, to the method used in determining x.
A plot o f the In (1-x) vs. time (Figure 15) for a first order reaction shows that
significant deviation from linearity occurs, particularly at the higher temperatures,
ruling out first order kinetics. A plot of In (l-x)'* vs time (Figure 16) gives a better fit
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
for a 1.5th reaction order than does a fit to second order kinetics. Kinetic plot o f In k vs.
1/T for 1.5th order is given in Figure 17. Assuming that the Arrhenius relationship k =
Ae'
p /n ’r
between rate constant and temperature holds, an activation energy o f 33.8 ± 2.0
kcal/mole with a statistical regression coefficient o f 0.990 was obtained. Hinkley158*
reported a reaction order o f 1.5 and an activation energy o f 33.2 ± 0.8 kcal/mole for
PETI-5 from heats o f reaction using the isoconversional plots as defined by Flynn and
Wall165* and Ozawa.166-671 However, this reaction order failed to describe the data over
the whole range o f conversion values for complete cure. This is strong evidence that
heats o f reaction data do not adequately define the reaction for complete cure. Johnston
et alt56] also reported similar effects in their investigations to follow the cure o f
phenylethynyl end-capped imide oligomers by infrared spectroscopy. The kinetic
analysis data for cure of PETI-5 are summarized in Table 2. The complicated 1.5th
reaction order implies that whole cure reaction includes phenylethynyl addition and
further intramolecular and bimolecular double bond addition reactions to form a more
highly crosslinked polymer.
2.3.2 Dielectric Properties of Model Compound and Imide Oligomer (PETI-5) in
the Microwave Region
As an alternative to conventional heating techniques, microwave irradiation provides
an effective, selective and fast synthetic method by heating the molecules directly
through the interaction between the microwave energy and molecular dipole moments
o f the starting materials. Therefore, dielectric properties o f these materials are directly
associated with the interactions between microwaves and the electric field in the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
molecules. The complex dielectric constant (e = e’-js") and loss tangent (tanS = s ’/s')
have been used as important parameters to evaluate both microwave power absorption
and penetration depth.[9'201 The dielectric properties o f PEPA-3,4’-ODA and PETI-5
were determined at six different frequencies over the microwave range, 397 MHz, 912
MHz, 1.429 GHz, 1.948 GHz, 2.466 GHz and 2.985 GHz, respectively (Figure 18-19).
0.8
-
£ 0.6
-
x
o
»*—
o
c
0
X
LLI
0.4 -
0.2
O
O
O
□
A
-
350
360
370
380
390
°C
°C
°C
°C
°C
0.0
10
20
30
40
50
60
70
Time (min)
Figure 13. Extent of Cure vs. Cure Time of PETI-5 from DSC Tg Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
32
1.0
0.8 H
x
I 0.6 H
o
O
&
0
A
□
O
I 0-4 0.2
&
-
0.0
o
-O r
230
240
250
350 °C
360 °C
370 °C
380 °C
390 °C
260
270
Tg(°C)
Figure 14. Relationship of Extent of Cure and Tg of PETI-5
1
0
-1
-2
? -3
c
-4
-5
♦
•
•
■
*
-6
-7
350
360
370
380
390
°C
°C
°C
°C
°C
-8
0
10
20
30
40
50
60
70
Time (min)
Figure 15. Plot of In(l-jc) vs. Time of Thermal Cure of PETI-5 (1st Order)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
33
9
8
o
o
350 °C
360 °C
370 °C
380 °C
390 °C
7
6
SI
a
*
5
+ 4
3
2
1
0
10
20
30
40
60
50
70
Time (min)
Figure 16. Plot of ( \ - x ) ' m
vs.
Time of Thermal Cure of PETI-5 (1.5th Order)
o
-0.5
-1
-
1.5
•2
-
2.5
-3 ----------------------------0 .0015
0.00152
—
0.00154
0.00156
0.00158
0.0016
-------------------
0.00162
Figure 17. Plot of ln& vs. 1/T of Thermal Cure of PETI-5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
34
Table 2. Kinetic Analysis of Thermal Cure of PETT-5 by DSC
Temperature (°C)
Rate Constant
350
0.0745
Regression
Coefficient
0.911
360
0.110
0.978
370
0.152
0.988
380
0.273
0.970
390
0.371
0.953
For microwave interaction, the polymer dielectric properties at the macroscopic level
depend on their dipole moment strength and mobility at the molecular level. The rapidly
changing electric fields induced by microwave energy cause changes of electronic
reorientation and distortions of induced or permanent dipole moments o f the molecules.
To minimize the deviation resulting from the powder compacting, PEPA-3,4’-ODA
sample was preheated to 305°C, just above the melting point (~297°C), and cooled
down to room temperature. And then the dielectric properties were determined from
room temperature to 385°C (Figure 18). Solid-state PEPA-3,4’-ODA gave a very low
dielectric loss tangent, which increased dramatically once reaching its liquid state
(above 297°C), as shown in Figure 18a. For example, the dielectric loss tangent (tan5)
was less than 0.007 (solid) below its melting point at 297°C and increased approaching
a value o f tanS at 0.105 (liquid) at 333°C. The dielectric constant ( s ’) increased from 1.5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35
0.16
0.14
- A
-
- v -
0.12
- 400
350
300
0.1 0
c?'
to
£
- 450
397 MHz
912 MHz
1.429 G Hz
1.948 G Hz
2.46 6 G Hz
2.98 5 G Hz
- 250
0.08
-
200
0.06
<o
h«w
0Q.)
£
<D
- 150
0.04
100
G
0.02
50
.q
0.00
p
PBi
0
0
8
(a)
-O
-D -G -O
-O -
r
397 MHz
912 MHz
1.429 GHz
1.948 GHz
2.466 GHz
2.985 GHz
&
450
400
GCCCO CO
350
300
c
rv
o
r\
250
200
150
c^CCgGCCCCS?
G
°Q
v
3
73
£
Q.
£0>
t-
100
o
50
P
0
4
Time (hr)
(b)
Figure 18. Dielectric Properties of PEPA-3,4’-ODA in Microwave Range
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
0.030
400
O O O O O u O O O
-o -
397 MHz
—
1.429 GHz
1.948 GHz
Z 466 GHz
2.985 GHz
—r-— 912 MHz
350
0.025
300
—'7-o
—o-
0.020
Da □
v
50 0.015
250
,
a r-. >
D-i;
200
0.010
150
O' ?
100
n
0.000
50
0
1
2
3
4
5
6
Time (hr)
(a)
Io o c o c o
dr"
400
350
o
o
300
o
o
o
o •*-
i
(^XDCXDQQOOCXX£>
397 MHz
- O 912 MHz
1.429 GHz
1.948 GHz
—■> 2.466 GHz
—O- 2.985 GHz
250
H
200
£
£
>
ha—
—C -
o
o
0
0)
3
150
100
50
3
Time (hr)
4
(b)
Figure 19. Dielectric Properties of PETI-5 in Microwave Range
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37
at room temperature to 2.7 at 330°C (Figure 18b). This strongly indicated that PEPA3,4’-ODA would absorb microwave energy effectively once the molecule reaches its
liquid state, i.e. the mobility of the ethynyl triple bonds in the starting materials is
sufficient to provide the required change o f dipole reorientation in the microwave
electromagnetic field for energy absorption. However, as temperature continued to
increase, the ethynyl-ethynyl addition reaction approached completion, and the resulting
high crosslink and large molecular structure limited the mobility o f the molecules. The
fact that higher-dipole triple bonds reacted into lower-dipole double bond also gave rise
to decreasing dielectric loss tangent. As shown in Fig 18a, the dielectric loss tangent
began to decrease with higher temperature as the temperature approached 333°C, and
dropped to zero after 373°C.
The dielectric properties of PETI-5 were also determined from room temperature to
385°C holding for 1 hour. No preheating was applied for this sample considering the
melting point (355°C) is too close to the curing temperature. Actually, the
phenylethynyl triple bonds of PETI-5 started to react at 350°C. As shown in Figure 19a,
PETI-5 showed a very low dielectric loss tangent in the solid state, which increased
dramatically as the temperature approached 350°C. The loss tangent increased from
0.001 (solid) at room temperature to 0.018 at 385°C. The dielectric constant { s ’)
increased from 1.6 at room temperature to 2.5 at 385°C (Figure 19b). The dielectric loss
tangent values of PETI-5 were much lower than that of PEPA-3,4’-ODA due to the low
concentration o f ethynyl groups, the higher molecular weight and the poorer mobility o f
the molecules as a result o f the high melt viscosity. During the isothermal period at
385°C, tan5 of PETI-5 decreased slightly with the cure time. This is a typical o f the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
38
mobility o f a small permanent dipole moment hindered with increasing molecular
structure. The chemical bond-type change from high-dipole triple bond to relatively
low-dipole double bond caused the decrease of dielectric loss tangent as well.
2.3.3 Kinetic Study of Microwave Cure
The microwave kinetic studies o f these materials presented some critical aspects
needed to be considered, namely that (1) the material under investigation absorbs
microwave energy to initiate a thermal reaction, and (2) precise time and temperature
control inside the reactor placed in the microwave cavity need to be maintained, and (3)
at a given temperature, reaction rates are faster by microwave irradiation than by
conventional thermal energy, and slightly inaccurate time and temperature measurement
would bring large deviation. The first condition can only be met if the materials reach
the liquid state. In thermal studies, this can be accomplished by heating the material to
its melting point, and then conducting the cure studies over the selected temperature
range. However, for microwave studies, since the solids are poor microwave absorbers,
a fugitive absorber has to be used to transfer the heat to the sample to attain melt, and to
satisfy the first condition. At this point, it is important to control the temperature of the
fugitive absorber to prevent overheat o f the sample. To accomplish this, a thermocouple
was inserted directly into the sample to provide precision o f temperature recording and
control, and silicon carbide (SiC) powder was used as the fugitive absorber.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
1) IR Analysis o f Microwave Cure o f PEPA-3,4'-ODA
The PEPA-3,4’-ODA sample was brought to its melting point temperature, 297°C,
and then kinetics o f cure was determined at 300°C, 310°C, 320°C and 330°C,
respectively. The infrared (IR) spectra of uncured and cured PEPA-3,4’-ODA are
shown in Figure 20. The decrease in the absorbance at 2214 cm'1 with cure time
indicates that the phenylethynyl triple bonds have reacted during microwave cure.
Relative to the imide carbonyl band, absorption of the ethynyl group at 2214 cm'1. The
changes o f integrated peak intensity at wavenumber 2214 cm'1(vC=C) with an increase
in cure time were measured at four different temperatures, 300, 310, 320 and 330°C. For
each sample, the integrated peak area at 2214 cm'1 was corrected by normalizing to the
integrated peak area o f the imide carbonyl group at 1778 and 1716 cm'1, by assuming
that the imide groups remained constant during the microwave cure process.
The results (in Figure 21) show that the extent o f reaction increased with time and
temperature, as expected. After 50 min at 330°C, the extent of reaction is 97%
indicating that only 3% of ethynyl groups remain after these cure conditions. Thermal
cure rate studies revealed that at 373°C after 90 min, the extent of cure was 90.4%
indicating that about 9.6% o f ethynyl group remained after these cure conditions. The
almost complete consumption of ethynyl groups after only 50 min at 330°C
demonstrates the efficiency and effectiveness o f the microwave process for this cure
reaction.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
40
jy
Before curing
MW lOmin
MW 20inin
MW 30min
MW 40min
2200
2000
1800
1600
1400
1200
1000
800
600
Wavenumbers (cm-1)
Figure 20. ER. Spectra of Microwave Cure of PEPA-3,4’-ODA at 320°C
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
Plots o f In(l-x) vs. time for each temperature (Figure 22) show that the reaction
follows first order kinetics, Linear regression analysis using the least squares method
was applied to fit the data. The kinetic rate constants, k, derived from the slopes o f each
reaction isotherm, are listed in Table 3. Assuming that the Arrhenius relationship k =
Ae'
p/pT
between temperature and rate constant holds, an apparent activation energy Ea =
27.6 ± 2.3 kcal/mol with a regression coefficient o f 0.987 was obtained from plot o f Ink
vs. 1/T (Figure 23).
2) DSC Analysis ofMicrowave Cure o f PETI-5
As we presented in the thermal cure reaction study o f PETI-5, the extent o f
microwave cure o f PETI-5 was also followed from Tg data obtained by DSC method.
DSC thermograms o f PETI-5 before and after microwave cure at 350°C through
different cure time are shown in Figure 24. Tg increased along with the increase of
microwave cure time as expected.
The extent o f cure, x, was calculated from the modified DiBenedetto equation as
described in the experimental section. Tgo of uncured material is 228.6°C, and a fully
cured material is defined as Tgaa (272.5) after microwave cure at 370°C for I hr. The
isobaric heat capacities ACP before cure and after microwave cure at 370°C for 1 hr
were 0.26 ± 0.03 J/g°C and 0.18 ± 0.03 J/g°C, respectively. The kinetic curves for the
extent of cure vs. time for samples cured at the various temperatures 350, 360, 370 and
380°C are shown in Figure 25. A plot of ln(l-x)'l/2 vs. time is given in Figure 26 by
following the 1.5th reaction order. Assuming that the Arrhenius relationship k = Ae’E RT
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
♦ 300C
0.2
• 310C
A320C
0.1
B 330C
0
0
10
20
30
40
50
60
Time (min)
Figure 21. Kinetic Curve of Microwave Cure of PEPA-3,4’-ODA via SiC Powder
-0.5
c
_i
-2.5
-3.5
0
♦
3000
A
320C
a
330C
10
20
30
40
50
60
T im e (m in)
Figure 22. Plot of ln(l-Jt) vs. Cure Time of Microwave Cure of PEPA-3,4’-ODA
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
-2.5
c -3.5 ■
-4.5 -
0.00164
0.00166
0.00168
0.0017
0.00172
0.00176
0.00174
1/T
Figure 23. Plot of \n k vs. 1/T of Microwave Cure of Model Compound
Table 3. Kinetic Results of Microwave Cure of PEPA-3,4’-ODA
Temperature (°C)
Rate Constant
Regression Coefficient
300
0.02504
0.932
310
0.03854
0.992
320
0.06313
0.990
330
0.08079
0.957
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Heat Flow (mW)
44
1 2 0
-i-
100
-
mw 350°C-60 min
80
mw 350°C-30 min
0
Q -j"-
mw 350°C-10 min
mw 350°C-5 min
40 H
Room temp
^
\
20 -J--------- ,------------------- .-------------------r
100
200
300
400
Tem perature(°C)
* mw - microwave cure
Figure 24. DSC Thermograms of Microwave Cure of PETI-5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
500
45
between temperature and rate constant holds, an apparent activation energy Ea = 17.1 ±
0.7 kcal/mol with a regression coefficient o f 0.997 was obtained from plot of ln& vs. l/T
(Figure 27). The kinetic analysis data for microwave cure o f PETI-5 are summarized in
Table 4.
2.4 CONCLUSIONS
Compared with thermal kinetic rate studies of model compound PEPA-3,4’-ODA
and PETI-5, microwave cure gave much higher rate constants, as shown in Tables 1 & 3
(results for PEPA-3,4’-ODA) and in Tables 2 & 4 (results for PETI-5). For PEPA-3,4’ODA, the rate constant of microwave cure at 320°C was 40 times that o f the thermal
cure at 3 18°C. Also the higher rate constants of microwave cure were observed at lower
cure temperature relative the thermal cure. For PETI-5, the rate constants of microwave
cure at 350°C and 360°C were twice that for the thermal cure reaction. As the cure
temperature increased, the rate constants tended to converge. At these temperatures, the
microwave irradiation induced cure reaction to a vitreous material. In the solid state,
microwave absorption decreased considerably and therefore the cure reaction became
much slower. Furthermore, microwave cure reactions displayed much lower activation
energies for both model compound and PETI-5. The activation energy o f the microwave
cure was 68% that o f the thermal cure for PEPA-3,4’-ODA, and 51% that of the thermal
cure for PETI-5. All these kinetic results indicated that, compared to the thermal
process, microwave process can provide more efficient and effective cure reactions
once the imide model compound and PETI-5 reach the liquid state.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
46
£3
o
o
c<D
LXU
!
10
20
30
n
O
□
A
40
350 C
360°C
370°C
380°C
50
60
Cure Time (min)
Figure 25. Plot of Extent of Cure vs. Cure Time of PETI-5 via Microwave Cure
8
7
-O
6
5
a
1?
4
I
3
&
__
is
I
°
2
I o
1
I
!
i
0
i
10
a
a
350°C
360°C
370°C
380°C
—r~
20
30
40
50
60
Time (min)
Figure 26. Plot of ln(l-x:)'1/2 vs. Cure Time of PETI-5 via Microwave Cure
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
-05
-0.7
-09
1.1
-1.3
■* -1.5
c
” -1.7
-
-1.9
-21
-23
-25
00015 00015 00015 0.0016 0.0016 0.0016 0.0016 00016 0.0016 0.0016
1/T(1/K)
Figure 27. Plot of ln& vs. 1/T of Microwave Cure of PETI-5
Table 4. Kinetic Results of Microwave Cure of PETI-5
Temperature (°C)
Rate Constant
Regression Coefficient
350
0.161
0.967
360
0.198
0.996
370
0.241
0.937
380
0.306
0.915
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 3: INVESTIGATION OF MICROWAVE ENERGY TO CURE
CARBON FIBER REINFORCED PHENYLETHYNYL-TERMINATED
POLYIMIDE COMPOSITES, PETI-5/IM7
3.1 INTRODUCTION
Fiber reinforced polyimide materials are high temperature, high performance
materials which provide high strength and light weight at elevated temperatures (250 400°C) and have potential application in airframes, propulsion systems, missiles and
land vehicles, such as automobiles and armored tanks. Considerable efforts have been
devoted to the application o f the microwave energy to cure polymer materials in the last
two decades.11'4’6’68'691 However, most work has been focused on epoxy resins and
epoxy/fiber composite system s/1'2’6’10"11’701 with only a few reports13’68"691 on the use of
microwave energy to cure polyimide resins. In this regard, the research presented in this
chapter was directed to the utilization of microwave as an energy source to cure
advanced polyimide/carbon fiber composites. For example, a high-performance
phenylethynyl-terminated polyimide PETI-5/IM7 composite material was selected for
structural applications in the High Speed Civil Transport (HCST). This material must
maintain its structural integrity at 177°C for 60,000 hours.1301 In addition to the strong
technical issues which this material must address, the ability to manufacture composites
more economically than can be accomplished by the conventional process, is a primary
factor in making the HSCT a commercially cost effective vehicle. Presently, cure o f
LaRC PETI-5 composite materials requires temperature up to 371°C and over a period
o f 3 hrs for high temperature curing.128'29’71'721 Processing o f thick-section parts by
48
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
conventional thermal techniques requires very low thermal ramp rates resulting in very
long processing times. A microwave cure process that would lower the cure temperature
and decrease process times would definitely contribute to reducing initial vehicle and
operating costs. For each specific application, whether the microwave energy can
indeed provide a better alternative to the conventional thermal process is still an issue
for debate. The need for reducing manufacture costs without sacrifice in thermal,
physical and mechanical integrity is always the essential driving force for the
development of novel processing methods with effectiveness and efficiency.
Objectives —The research in the composite area is directed toward investigating
microwave energy to cure PETI-5/IM7 polyimide/carbon-fiber prepreg into PETI5/IM7 composites with equivalent or better properties than the conventional thermal
cure process. Several cure cycles were investigated by the microwave and conventional
thermal processes. The thermal, physical and mechanical properties o f the composites
fabricated in this comparative study and observations of the failure surface of the shear
specimens are presented.
3.2 EXPERIM ENTAL
3.2.1 M aterials
Unidirectional carbon fiber reinforced phenylethynyl-terminated polyimide prepreg,
PETI-5/IM7, was provided by NASA Langley research center and Northrup-Grumman
Corporation. IM7 is one type o f continuous carbon fibers produced by Fiberite
Corporation, PETI-5 resin powder was provided by NASA Langley research center.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
3.2.2 Thermal Process Conditions
Thermal processing was performed using a programmable Tetrahedron press, model
MTP-14. A 10.2 cm x 15.2 cm (4 inch x 6 inch) 20-ply unidirectional lay-up was placed
in a stainless steel mold with a surface layer o f sprayed Teflon. Teflon glass cloth was
used above and below the ply lay-up. A full vacuum was applied through out the entire
cure period. Pressure was applied after the holding period at 250°C. For the thermal
processes, a pressure of 200 psi was used. Figure 28a is a schematic o f the thermal
processes.
3.2.3 Microwave Process Conditions
A variable frequency microwave furnace (VFMF) model LT502Xb (Lambda
Technologies, Inc.) was used to perform microwave processing studies o f PETI-5/IM7
composites. Due to its transparency to microwave energy, Teflon was selected as the
construction material for the molds required to fabricate composite samples. A 10.2 cm
x 10.2 cm (4 inch x 4 inch) 20-ply unidirectional lay-up was placed in the Teflon mold,
sandwiched between two quartz plates. Teflon glass cloth was placed between the
quartz plates and the laminate, as shown in Figure 29. Two quartz plates were used to
help maintain shape of the ply lay-up under the pressure. Two thermocouples were
inserted into both sides o f laminate samples, perpendicularly to the graphite fiber
direction o f the prepreg. A center frequency o f 4.69 GHz with a bandwidth o f 0.75 GHz
and a sweep rate of 0.5 second was used in the microwave processing. A full vacuum
was applied through out the entire cure period. Pressure was applied after the 250°C
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
holding. Various cure times, temperatures and process pressures were investigated and
evaluated in the microwave cure reactions as described below. Figure 28b is a
schematic o f the microwave process. All the thermal and microwave cure processes
investigated are listed in Table 5.
3.2.4 Characterization
TGA - Thermogravimetric analyses (TGA) of microwave and thermal cured
composites were performed by using a TA Instrument 2100 Thermal Analysis TGA
2950 at a heating rate of lO°C/min under nitrogen atmosphere.
DMTA - Dynamic mechanical thermal analysis (DMTA) was measured by a
Polymer Laboratories DMTA Mk II. The measurement was conducted by a shear mode
from room temperature up to 400°C at a heating rate of 2°C/min.
TMA - Thermomechanical analysis (TMA) was utilized to determine the glass
transition temperatures of microwave and thermally cured composites using a PerkinElmer TGA, 7 series analysis system at a heating rate of 2°C/min under helium
atmosphere.
Flexural Test - A 3-point bend test (ASTM D790), with a span-to-depth ratio of 16
to 1, was used to measure the flexural strengths and moduli of the composites by using
a INSTRON 4602 at a crosshead speed of 0.043 in/min (0.11 cm/min). The
measurement was conducted at both room temperature and 177°C. A nominal sample
size was around 1.9”x0.25”x0.1” (4.8cm x 0.64cm x 0.25cm). Five specimens were
used for each test.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
SBS Test - Short beam shear (SBS) test (ASTM D2344), with a span-to-depth ratio
o f 4 to 1, was performed by a INSTRON 4602 at a crosshead speed o f 0.05 in/min (0.13
cm/min). The interlaminar shear strength was measured at both room temperature and
177°C. A nominal sample size was 0.7”x 0.25”x 0.1”(1.8cm x 0.64cm x 0.25cm). Five
specimens were used for each test.
Density Test - Density o f microwave and thermal cured composites were measured
by using an OHAUS Density Determination kit at 23.5°C. Ethyl alcohol with a density
of 0.794g/ml was used as the test liquid to provide buoyancy.
Composition Test — Void, resin and fiber contents o f the composites were
determined by sulfuric acid - peroxide digestion method. The sulfuric acid (d = 1.84)
was heated to its fuming temperature, and then 30% H2 O2 was added dropwise to digest
the resin. Fiber remaining was washed in water and dried in the vacuum oven. The
weights o f the material before and after digestion were recorded. The densities of cured
PETI-5 resin (1.37 g/ml) and IM7 carbon fiber (1.77 g/ml)[711 were used to calculate the
void contents o f the composites.
ESEM Test - The environmental scanning electron micrographs o f the failure
surfaces o f short-beam shear specimens of both microwave and thermally cured
composites were taken using a Philips Instrument Model
ESEM 2020 with
magnifications o f 300 x and 1000 x.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
53
Temp
371 °C for lh r
Pressure
4 .5 ° C /m in
—3-hrs-----
Time
Vacuum
(a) Standard Thermal Cure Cycle
Temp
Pressure
360 °C for 30m in
250 °C for 30min
f
200 psi
-.■LXhrs....
Time
Vacuum
(b) Microwave Cure Cycle
Figure 28. Thermal and Microwave Cure Cycles
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
54
Pressure
W&/M
inless Steel
>TEFLON block
Seal rubber
Vacuumsystem
Grounded TC
TEFLON block
Kapton film
Quartz or Glass
Release cloth
Grounded TC
Prepreg sample
Release doth
Figure 29. Schematic of Microwave Mold Design
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5. Thermal and Microwave Cure Processes of PETI-5/IM7
Process No.
T h e rm a l
C u re
C y c le s
th(a)
250°C-0.5 hr + 360°C-0.5 hr / 200 psi
th(b)
250°C-0.5 hr + 371°C-0.5 hr / 200 psi
th(c)
250°C-1 hr + 371°C-1 h r / 200 psi
(NASA Standard Cure Cycle)
M ic r o w a v e
C u re
C y c le s
mw(l)
250°C-0.5 hr + 350°C-0.5 hr / 200 psi
mw(2)
250°C-0.5 hr + 360°C-0.5 hr / 200 psi
mw(3)
250°C-0.5 hr + 370°C-0.5 hr / 200 psi
mw(4)
250°C-0.5 hr + 360°C-0.5 hr /1 0 0 psi
mw(5)
250°C-0.5 hr + 360°C-0.5 hr / 50 psi
mw(6)
250°C-0.5 hr + 360°C-0.5 hr / 0 psi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
56
3.3 RESULTS AND DISCUSSION
3.3.1 Process Conditions
The PETI-5 resin was prepared by Imitec, Inc., from commercially available
monomers shown in Scheme 3, and applied to the IM7 carbon fiber by Fiberite
Corporation. The IR spectrum of resin extracted from prepreg (Figure 30) displayed
imide carbonyl absorbance at 1774 and 1720 cm*1, and acid carbonyl at 1684 cm'1 as
well. The PETI-5/IM7 prepreg at this point consists o f an amic acid and partially
imidized amic acid oligomers, which undergo two chemical curing steps to reach its
final cured state, (i) imidization at a lower cure temperature and (ii) crosslinking at a
higher cure temperature, as illustrated in Scheme 3. The prepreg showed no obvious
difference in coupling with microwaves within a range o f 2.4 to 7.0 GHz. A center
frequency o f 4.69 GHz with a bandwidth o f 0.75 GHz and a sweep rate o f 0.5 second
was used in this study. Based on the thermal processing method as developed at the
NASA Langley research center,[71*721 in which laminates were cured at 250°C for 1 hr
and 37l°C for 1 hr, (Figure 28a), excluding time to reach temperatures and cool down,
six microwave cure cycles were designed and investigated as shown in the experimental
section. The Figure 31 displays the TGA curves o f microwave cured PETI-5/IM7
prepreg at 250°C for 0.5 hr and 1 hr compared with the uncured prepreg. No obvious
weight loss was observed in the microwave cured prepreg at 250°C for 0.5 hr, which
gave similar TGA behavior as the microwave cured prepreg at 250°C for 1 hr,
indicating that complete imidization of amic acid in the PETI-5/IM7 prepreg was
already achieved in a half hour time period. Therefore, a half hour period was selected
as the cure time at 250°C in the microwave processes instead of 1 hr as used in the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
57
0.18
0.91
PEPA
BPTA
0.15
H2 N
- 0 r ° i 0 r o^ O r NH’ * ° 85 H!Nt ^ ° ^ O L n h ,
3,4'-ODA
1,3-Bis(3-APB)
30% Solids
NMP
Amide Acid Oligomer
O
©
O
r c
- < j n r o
where: Ar =
3,4’-ODA
APB
PETI-5
(Prepreg contains both amic acid and imide oligomer)
crosslin k
C= C
/
\ £. = C ✓
&
x
Reaction sequence for polymerization and crosslinking of PETI-5 resin
Scheme 3. Chemical Structures and Cure Reactions of PETI-5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
58
2.4
2 .0
t .2
0.0
0.2
4000
2000
3500
1500
Figure 30. ER Spectrum of Resin Extracted from Prepreg
100
\
95 -i
5. 90 “I
X
T
CD
§
85
80
Uncured Prepreg
MW 250°C-30min
MW 250°C-1hr
\
75
200
400
600
800
1000
Temperature (°C)
Figure 31. TGA Curves of PETT-5/IM7 Prepreg before and after Microwave
Curing at 250°C
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
59
thermal process. Furthermore, both dielectric measurement and kinetic studies showed
that the resin absorbs microwave effectively once it reaches its melt state. Therefore, the
temperatures investigated for the crosslinking reaction were 350°C, 360°C and 370°C,
respectively. These are around or above the resin melt temperature, which is around
355°C. In addition to the time-temperature processing conditions investigated, it was of
interest to determine if lower pressures (0 psi, 50 psi, 100 psi) at the optimum timetemperature conditions could be employed with the microwave processing. This effort
was devoted to identifying the optimum pressure for complete consolidation o f the
prepreg in the microwave process. In the thermal process studies, besides the NASA
standard cure cycle o f 250°C-lhr and 371°C-lhr, two other thermal cure processes, th(a)
250°C-0.5hr and 360°C-0.5hr and th(b) 250°C-0.5hr and 371°C-0.5hr, were investigated
in order to provide a direct comparison with the microwave cure processes.
3.3.2 Therm al and Physical Properties
Thermal properties o f microwave and thermally cured composites were determined
by TMA, DMTA and TGA methods. The results are summarized in Tables 6 and 7.
Glass transition temperatures o f both microwave and thermally cured composites were
determined by the onsets o f transition in TMA curves (Figures 32 & 33) and the
relaxation peaks of loss tan5 (at 1Hz and 10Hz) in DMTA spectra (Figures 34 & 35).
For the thermally cured composites, the glass transition temperatures increased from
199.5°C to 235.9°C according to TMA measurements (Figure 32) and from 244°C to
259°C according to DMTA measurements at 1 Hz (Figure 34) as the cure time and
temperature increased, from processes th(a) to th(c). For composite th(a), cured at
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
Table 6. DMTA, TGA, TMA and Composition Results o f Microwave and
Thermally Cured PETI-5/IM7 Composites
Process
Composite
No.
Tg at 1Hz Tg at 10Hz Tg at onset TGA wt. loss Resin voL
Void
(“Q-DM TA (°Q-DMTA (°Q-TMA before 450°C content content
th(a)
Thermal 250°C-l/2hr,
360°C-l/2hr
244
252
199.5
0.744%
41.7%
<0.5%
th(b)
Thermal 250°C-l/2hr,
371°C-l/2hr
244
253
218.2
0.335%
43.4%
<0.5%
259
270
235.9
0.219%
42.7%
<0.5%
274
286
249.4
0.147%
42.2%
<0.5%
273
281
251.7
0.081%
42.9%
<0.5%
270
281
251.8
0.099%
42.7%
<0.5%
th(c)
Thermal 250°C-lhr,
371°C-lhr
m w (l) Microwave 250°C-l/2hr,
350°C-l/2hr
mw(2) Microwave 250°C-l/2hr,
360°C-l/2hr
mw(3) Microwave 250°C-l/2hr,
370°C-l/2hr
Table 7. Density, Composition and DMTA Results of Microwave Cured
PETI-5/IM7 Composites at Different Pressures
Process No. - Pressure mw(2) - 200psi mw(4) - lOOpsi mw(5) - 50psi mw(6) - Opsi
Density (g/ml)
1.60
1.60
1.53
1.48
Resin vol. content (%)
42.9
42.4
44.5
47.8
Tg (°Q at 1Hz
273
275
274
273
Tg (°Q at 10Hz
281
280
281
281
Void Content (%)
<0.5
<0.5
4.4
7.3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
61
TtCRMAL PETI-5/IM7 C0M>0EITE5
2 .BO
2.58
2.56
EKpamlon
2.54
250C-0.5HR 360C-0.5W
2.50
250C-O.SW 37IC-0.5W)
2.4S
2.46 H
2S0C-ltfi 371C-IW
50.0
100.0
150.0
200.0
250.0
300.0
T w ce ritu re ('Cl
Figure 32. TMA Thermograms of Thermally Cured Composites
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
350.0
62
MIOOttVE PETI-5/IH7
COMPOSITES
2.40
2.36(«•) uotiuadia
2.36
2 .3 4 -
43 MM 370C-0.5hr
2.32
360C-0.5hr
2.30
2.26
41 MM 350C-0.3hr
2.26
50.C
100. C
150.0
200.0
Temperature
2 5 0 .0
3 0 0 .0
r*o
Figure 33. TMA Thermograms of Microwave Cured Composites
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
63
8.5
8.0
7.5
0.9
7.0
0.8
6.5
—0 ~ Thermal 250°C-0.5hr & 360°C-0.5hr
- o — Thermal 250°C-0.5hr & 371°C-0.5hr
—
Ther mal 250°C-1 hr & 371 °C-1 hr
0.6
Tan 5
0.7
O 6.0
O)
o
—I 5.5
0.5
5.0
0.4
4.5
0.3
4.0
0.2
3.5
0.1
3.0
50
100
150
200
250
300
350
400
0.0
450
Temperature (°C)
Figure 34. DMTA Curves (1 Hz) of Thermally Cured PETI-5/EM7 Composites
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
64
8.5
8.0
7.5
0.9
7.0
0.8
6.5
O)
o
_l 5.5
o - Microwave 250°C-0.5hr & 350°C-0.5hr
o - - Microwave 250°C-0.5hr & 360°C-0.5hr
Microwave 250°C-0.5hr & 370°C-0.5hr
0.6
Tan 8
0.7
O 6.0
0.5
5.0
0.4
4.5
0.3
4.0
0.2
3.5
0.1
3.0
50
100
150
200
250
300
350
400
0.0
450
T e m p e ra tu re (°C)
Figure 35. DMTA Curves (1 Hz) of Microwave Cured PETI-5/IM7 Composites
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
65
1.0
0.9
8
0.8
0.7
0 psi
50 psi
100 psi
200 psi
0.6
0.5
6
Tan 8
Log G* (P a)
7
0.4
0.3
5
0.2
0.1
tg K S E g g B ffisa a c q g fSg
50
100
150
0.0
200
250
300
350
400
450
T em p era tu re (°C)
Figure 36. DMTA Curves (1 Hz) of Microwave Cured PETI-5/IM7 Composites at
250°C-0.5hr and 360°C-0.5hr under Different Pressures
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
66
250°C-0.5hr & 360°C-0.5hr at 200 psi, the TMA curve showed that after the glass
transition temperature, another transition occurred commencing at about 315°C. This
can be attributed to melting o f uncured resin in the composite material. The presence of
uncured resin is confirmed much more clearly in the DMTA spectra. Both composite
materials from thermal processes th(a) and th(b) gave extra damping peaks in the high
temperature test range above 300°C (see Figure 34), which indicated that uncrosslinked
resin remains in the composite. The amount of uncured resin decreased from process
th(a) to process th(b) based on the decreasing damping magnitude. The glass transition
temperatures o f the microwave cured composites were relatively constant around 250°C
by the TMA method (Figure 33) and around 273°C by the DMTA method at 1 Hz
(Figure 35). The expansion coefficients ranged from 3.4 to 3.8 x 10‘5 mm/°C in the
glassy state before Tg and increased to a range of 1.4 to 1.7 x 1CT4 mm/°C in the
viscoelastic state after Tg. It is significant to note that the microwave cured composites
derived from processes m w(l) through mw(3) give higher glass transition temperatures
(by 11° to 16°C) than the thermally cured composites (see Table 6). The higher glass
transition temperatures observed in the microwave cured composites demonstrate that
the microwave technique offers a much more complete cure process relative to the
corresponding thermal process. This further suggests that a larger extent o f crosslinking
o f the phenylethynyl functional groups is achieved via microwave energy. Furthermore,
TGA data showed that the microwave cured composites lose less weight before 450°C
than the thermally cured composites. In the thermally cured composites, the TGA
weight loss decreased with an increase in cure time and temperature, from 0.74% in
process th(a) to 0.22% in process th(c); while microwave cured composites made by
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
67
processes mw(2) and mw(3) showed a weight loss o f less than 0.1%. The dynamic
mechanical thermal analysis (DMTA) curves of four composite systems fabricated at
four pressures (0, 50, 100, and 200 psi), as given in Figure 36, clearly show equivalent
Tg’s (273 - 275°C) based on the loss tan5. This is not unexpected since the cure
characteristics o f the polyimide resin are independent o f the pressure. However, other
properties such as density, void content, and mechanical properties, were influenced by
these pressure differences. Composites fabricated at 0 or 50 psi exhibited lower
densities and higher resin and void contents than those fabricated at 100 or 200 psi (in
Table 7), and composites fabricated at 100 and 200 psi exhibited equivalent densities,
resin and void contents. Both microwave and thermally cured composites gave resin
volume contents at 42-43% (Table 6), which were also similar to the data from the
conventional process1711.
3.3.3 Mechanical Properties
Two of the screening tests used extensively in evaluating composites are the flexural
and short beam shear tests. These tests are just two o f the many mechanical tests used
by NASA Langley Research Center in evaluating the PETI-5/IM7 composite systems.
Therefore, three-point flexural bending and short beam shear tests were measured at
both room temperature and at an elevated temperature (177°C). The results are
summarized in Tables 8 and 9. It is interesting to note that essentially no difference in
mechanical properties is exhibited among the thermally cured composites by the three
different processes. However, only thermal process th(c) yielded a composite with the
complete cure and the highest glass transition temperature, Tg, (Table 6). This is the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
68
conventional cure process selected by NASA.[71'72' For the composites fabricated at the
standard pressure o f 200 psi, flexural strengths of both the microwave and thermally
cured composites were similar at room temperature, although the microwave cured
composites demonstrated slightly higher values in flexural modulus than the thermally
cured composites (Table 8). However, at 177°C, higher flexural strengths and moduli
were observed in all o f the microwave cured samples relative to the thermally cured
composites. In short beam shear tests, microwave fabricated composites displayed
higher interlaminar shear strength at both room temperature and 177°C. At 177°C,
compared with the data from room temperature tests, flexural strength and modulus
retentions of 67-69% and 93-97%, respectively, were observed for microwave cured
composites, while the strength and modulus retentions of 62% and 87%, respectively,
were observed for the thermally standard cured composites. The interlaminar shear
strengths of the standard thermally cured composite at room temperature and 177°C
agree with the results in a NASA published report.[711 Relative to the short beam shear
strengths at room temperature, microwave cured composites demonstrated a strength
retention of 70-74% at 177°C while the thermally cured composites showed a strength
retention of 67%. A comparison o f microwave and thermal processes is illustrated in
Figure 37, where microwave process mw(2), 250°C - 0.5 hr and 360°C - 0.5 hr under
200 psi, was selected as the representative for microwave process and the conventional
standard process th(c), 250°C - 1 hr and 371°C - 1 hr under 200 psi, was chosen for
thermal process. The improved mechanical properties o f microwave cured composites
at high temperature demonstrated that the microwave irradiation is an effective method
to fabricate high quality PETI-5/IM7 composites with reduced cure time. The
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
69
mechanical tests results for microwave cured composites fabricated at different
pressures are listed in Table 9. Actually, composites fabricated at 100 psi and 50 psi
demonstrated equivalent flexural and shear strengths at room temperature and 177°C
relative to the results of the standard thermal cure process th(c), though their values
were lower than the results of the composites microwave cured at 200 psi. Also
considering the void content and density data (Table 7), these results indicated that the
microwave process mw(4) (250°C — 0.5 hr and 360°C - 0.5 hr / 100 psi) generated
PETI-5/IM7 composites with the equivalent physical and mechanical properties in one
half the time and lower pressure relative to the conventional thermal process th(c)
(250°C - 1 hr and 371°C - 1 hr / 200 psi).
3.3.4 Failure Surface Analysis
Inspection of the failure surface o f composites tested at room temperature showed
that more resin is adhering to the fiber surface of the microwave cured composite
(Figure 38(a)) than to the fiber surface o f the thermally cured composite (Figure 38(b)).
This suggests that adhesion o f the resin to the fiber surface for the microwave cured
composite was better than in the thermally cured composite. This is strong indication
that better wetting of the fiber by the resin occurs in the microwave cured system than
in the thermally cured system, most likely associated with excellent heat transfer from
each carbon fiber to the resin. Moreover, the resin adhering to the fiber surface itself
showed shear yielding topography indicating that the resin is tougher in the microwave
cured composite than in the thermally cured composites. The failure surfaces o f the
composites which were tested at 177°C showed the same trend as the room temperature
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
70
tests, but to a greater extent (Figure 38(c)-(d)). The failure surfaces o f the microwave
cured composites showed a considerable amount o f resin adhering to the fiber surface,
while the failure surface o f the thermally cured composites were essentially free o f
resin. Failure for these composites is mostly interfacial. For the microwave cured
composites, a mixed mode failure occurred consisting of both resin shearing and
interfacial failures. The higher Tg’s, higher flexural moduli and shear strengths at 177°C
o f the microwave cured composites relative to the thermally cured composites are
properties which can lead to the failure mode observed. Similar microwave enhanced
interfacial strength between the epoxy resin and glass fiber1731 or graphite fiber1741 has
been reported.
3.4 CONCLUSIONS
Six different microwave cure cycles and three thermal processes were investigated to
cure carbon fiber reinforced phenylethynyl-terminated polyimide composites, PETI5/1M7. Higher glass transition temperatures were observed in the microwave cured
composites. Thermally cured composites, fabricated from the same time-temperature
cure cycles as microwave processes, showed incomplete cure and much lower glass
transition temperatures. Compared to the standard thermally cured composites,
microwave cured composites exhibited higher flexural strength and modulus, and shear
strengths at 177°C. A microwave cure process to fabricate unidirectional PETI-5/IM7
polyimide/carbon fiber composites with superior thermal and mechanical properties
relative to the thermal process in one-half the time required for the thermal process, has
been demonstrated.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
Table 8. Summary of the Mechanical Results of Microwave and Thermally Cured
PETI-5/EM7 Composites
Process
Composite
at RT (kpsi)
a tR T (\^ » 0
at RT (kpsi)
a t 177°C(kpsi)
Thermal 250°C-l/2hr,
195.6
13X1
117.4
1X21
16.80
9.57
360°C-l/2hr
±6.7
±050
±7.4
±L25
±078
± 027
Thermal 250°C-I/2hr,
202.5
13.88
118.2
11.70
16.68
11.67
371°C-l/2hr
±1X2
±101
±4.3
±088
±038
±022
Thermal 250°C-lhr,
193.6
1X86
119.2
11.18
15.68
10.43
371°C-lhr
±8.0
±063
±8.1
±091
±046
±043
190.6
13.87
13X3
13.00
16.04
11.83
±4.6
±023
±50
±077
±054
±070
19X1
14.00
128.0
13.02
1732
1X61
±6.5
±026
±7.1
±042
±045
±045
19X1
1336
128.3
1X91
17.14
1X02
±4.6
±048
±3.5
±075
±066
±034
No.
th(a)
th(b)
th(c)
Flexural Stress Flexural Modulus Flexural Stress Flexural M xhilus Shear Strength Shear Strength
m w (l) M crow ave 250°C-l/2hr,
350°C-l/2hr
mw<2) M crowave 250°C-l/2hr,
360°C-L/2hr
mw(3) M crowave 250°C-l/2hr,
370°C-l/2hr
a t 177°C(kpsi) atl77°C (fcfcsi)
* RT - room temperature
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
72
Table 9. Summary of the Mechanical Results of Microwave Cured PETI-5/TM7
Composites at Different Pressures.
Process No. —
Flexural
Flexural
Flexural
Flexural
Shear
Shear
Process Pressure
Stress at
Modulus at
Stress at
Modulus at
Strength at
Strength at
RT (kpsi)
RT (Mpsi)
177®C (kpsi)
177°C (Mpsi)
RT (kpsi)
177°C (kpsi)
127.2
10.27
90.8
9.39
13.40
9.53
±20.9
±1.20
±8.4
±1.16
±1.26
±0.92
185.9
12.38
123.2
11.72
15.33
10.21
±7.2
±0.71
±5.3
±0.42
±1.29
±0.28
185.5
13.29
128.2
12.60
15.89
10.43
±9.3
±0.71
±6.9
±0.30
±0.58
±0.26
192.1
14.00
128.0
13.02
17.32
12.61
±6.5
±0.26
±7.1
±0.42
±0.45
±0.45
mw(6) - 0 psi
mw(5) - 50 psi
mw(4) -100 psi
mw(2) - 200 psi
* RT - room temperature
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
73
i □ Thermal Standard, th(c) El Mcrowave/200psi, mw(2) □ Mcrowave/10Opsi, mw(4) |
250
200
150
100
W
m
VV
rtV
W
'AA
V
fc&
i'?
m
was
■',Y'
Mass
5S&S?
i'V /j
50
vv.v
?
*
&
y
*
y
M
M
fta
SaAi
$$$
m
>/0A'/0
H
;A‘AV-
$$
Flexural
S trength a t
RT (kpsi)
i-SSS
5®
'S
88
SVSVK
S & V J!
0
Mk
Flexural
M odulus a t RT
(10A5psi)
F lexural
S tren g th a t
I77C (kpsi)
Flexural
M odulus at
177C (10A5psi)
S hear
S trength a t
RT (lOOpsi)
S h ear
S tre n g th a t
177C (lOOpsi)
* RT —room temperature
Figure 37. (a) Mechanical Properties
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
74
y
Z2
y
Biiii
IHP
®
fj
1
1 ■
y
1 1
T g (C )a t1 H z
m
III
W'
m 0 m
m
iii
lH
1
1
1
T g(C )at10H z
__
Density (10mg/ml)
(b) Physical Properties
400 r l
Highest Process
Temp (C)
(c) Processing Conditions
Figure 37. Comparison of Processing Conditions and Properties between
Microwave and Thermally Cured PETI-5/IM7 Composites
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
e 3»
<*o
Reproduced w ith permission
77
vrrc
figure
jg . <c) M i« ° " ave’ 3‘
78
x 300
x 1000
(d) Thermal, at 177°C
Figure 38. ESEM Images of Shear Failure Surfaces of Microwave and T h erm a lly
Cured PETI-5/IM7 Composites
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
C H A PT E R 4: INVESTIGATION O F RING OPENING POLYMERIZATION VIA
MICROWAVE IRRADIATION
4.1 INTRODUCTION
4.1.1 e-C aprolactam Background
Poly(e-caprolactam) (Nylon 6) prepared from e-caprolactam was first studied in the
late 1890s and expanded world-wide since World War II. Today, poly(s-caprolactam)
plays an important role in the field o f engineering thermoplastics due to its good
combination o f thermal and mechanical properties, broad processing range and
compatibility with additives and other materials. Also, poly(e-caprolactam) fiber, as a
widely used synthetic fiber, has a current global production capacity o f more than 3
million metric tons.[32i The two major commercial routes to manufacture poly(ecaprolactam) are hydrolytic polymerization and anionic polymerization. Through
hydrolytic polymerization, poly(e-caprolactam) is conventionally produced by heating
s-caprolactam at 250 - 270°C with a catalytic amount of water (5-10%) using a nylon
salt or aminocaproic acid as an initiator for 12-24 hrs.[31'
Three-step reactions have been involved in the ring opening polymerization of cyclic
e-caprolactam:I75'781 (l) hydrolysis of e-caprolactam to ©-aminocaproic acid; (2) ring
opening polymerization o f e-caprolactam through the amine nitrogen
o f ©-
aminocaproic acid or the amine end of the growing chain attacking on the carbonyl
carbon o f e-caprolactam by nucleophilic addition; (3) chain termination by the
condensation o f amine and acid end-groups, as illustrated in scheme 4.
79
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
80
Step 1
+
HoO
HOOC(CH2)5NH2
Step 2
+ H2N(CH2)5COOH
HOOCCCH^sNHCOCCH^sNHz
•NHCO(CH2)5NH2
Step 3
'/^CO O H + ''n^NH 2
Scheme 4. Ring Opening Polymerization of s-Caprolactam
Normally, there are 10 - 12 wt% monomers and 3% cyclic oligomers132’761 left in the
final step due to the reaction equilibrium. For industrial batch or continuous reactors,
water is added to initiate the reaction and removed later to increase molecular weight,
which is achieved in a high-pressure reactor. In this present investigation, coaminocaproic acid was selected as catalyst (and initiator) instead of water since the
microwave equipment in our lab can only be performed at atmosphere pressure.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
81
4.1.2 s-Caprolactone Background
Poly(e-caprolactone) (PCL) is an aliphatic polyester that has been synthesized from
anionic, cationic, and coordination polymerization from s-caprolactone.t35'36-79'83!
Poly(e-caprolactone) has been widely investigated as a potential biomaterial based on
its biocompatibility and biodegradability.[361 Poly(e-caprolactone) can not only be
degraded by microorganisms but also hydrolyzed under certain physiological
conditions. Two steps are involved in the degradation o f implanted poly(ecaprolactone).136' 84-851 Firstly, hydrolysis causes the formation o f low molecular weight
polymer chain, and then the body fluid or phagocyte carry away the hydrolyzed
polymer particles. It would take 2-4 years for the complete degradation o f poly(ecaprolactone), which is significantly slower than that o f poly(lactic acid) (PLA) or
poly(glycolic acid) (PGA).[86'87] This would lead to special interest in poly(ecaprolactone) for long-term, implantable drug-delivery systems or one year implanted
device.
The
mechanical
properties and
biodegradation
can
be balanced
by
copolymerizing or blending with other polymers to meet the various requirements. As a
biodegradable packaging
material,
poly(e-caprolactone)
can
copolymerize with
aliphatic amino molecules to improve mechanical properties and while retaining
biodegradability, which can lead to development o f the biodegradable poly(amideester)s like lactam-lactone copolymers.
Poly(e-caprolactone) is commercially produced from caprolactone at 110°C for over
12 hrs. Coordination polymerization, using stannous octoate (SnOct2 , tin(H)2ethylhexaoate) as a catalyst and aliphatic alcohol (like butanediol) as an initiator, is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
82
commonly applied
in the manufacture o f poly(e-caprolactone). Ring opening
polymerization o f e-caprolactone is illustrated in Scheme 5.
Oct
Sn
O ''
Oct
6+
OH
► init-0
initiation
+ HO-init
propagation
OH
initlO
Scheme S. Coordination Polymerization of s-Caprolactone
4.1.3 Copolymerization Background
Synthetic polyamides such as nylon 6 have favorable mechanical properties but are
not biodegradable. Modification o f nylon 6 with polyesters, other amino acids such as
a -L amino acids, renders these systems biodegradable1881. On the other hand, synthetic
poly(e-caprolactone)
(PCL)
is
biodegradable,
its
degradation
products
are
biocompatible and PCL is permeable to drugs. However, PCL lacks the mechanical
properties required for application as tough film for disposable packaging applications.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
83
However, polyester-amides composed o f esters and amides could potentially have better
physical and mechanical properties than polyester with retention of biodegradability
characteristics o f the polyesters189'931. Anionic copolymerization of e-caprolactam with
e-caprolactone has been investigated by Goodman et al[38’94'961, Gonsalves et al[391 and
Nakayama et al[971. This typical anionic copolymerization initiated by s-caprolactam is
illustrated in scheme 6.
Goodman et al[94’ showed that, in anionic copolymerization studies using N-sodium
caprolactam as a catalyst, at both high and low initial e-caprolactam/e-caprolactone feed
ratio (70/30 or 25/75), incorporation of e-caprolactam units into the copolymer
increased at higher catalyst concentrations (1 and 2%) and with reaction time. For a
catalyst level of 0.5%, at the same feed ratios, e-caprolactone is the major component in
copolymers. This implies that, under these conditions, e-caprolactone is the more
rapidly reacting monomer which polymerizes selectively at 0.5% sodium caprolactam
catalyst level. Goodman et al138’ further reported that higher temperatures and higher
catalyst concentrations and longer reaction times caused more equal incorporation o f
the two monomers, giving products with molecular weights in the order o f 104, and with
the lactam and lactone unit contents approximating the reactant feeds.
These anionic copolymers were favored to be crystallized over the whole range o f
compositions depending on the proportions of lactam and lactone unit and the thermal
history of the copolymers1951. Stiffness and extensibility varied with lactam content,
with the 25/75 and 10/90 lactam/lactone copolymer compositions, exhibiting the highest
brittleness. Microstructural features of the lactam/lactone copolymers have also been
investigated by Goodman et al[961.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
84
f>
N'M
+ BH
+ B*M+
NH
N'M
N—CO(CH2)5N'M+-|
O
O
N'M
Copolymer
r \
N— C0(C H 2)50 'M +
Scheme 6. Anionic Ring Opening Copolymerization
Gonsalves et al[391 investigated two methods to synthesize copoly(ester-amide)s. A
series of non-alternating copoly(ester-amide)s were synthesized by anionic ring opening
o f e-caprolactone and e-caprolactam using sodium caprolactam as a catalyst over the
temperature range 100 to 160°C. In the other method, 1,6-hexanediol was reacted with
excess adipoylchloride to give an acid chloride intermediate which was interfacially
reacted with 1,6-hexane diamine to yield alternating copoly(ester-amide)s. Both types
o f polymers were subject to hydrolysis in buffer solutions at a pH of 7.4 and were
degradable by the fungi, Fusarium moniliforme or Aspergillus niger.
Nakayama et al[9?i also reported the synthesis o f copoly(ester-amide)s by anionic
ring opening of e-caprolactone and e-caprolactam using sodium caprolactam as a
catalyst. These copolymers exhibited a single melting temperature between that of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
85
poly(e-caprolactone) and poly(e-caprolactam), characteristic of a random copolymer.
The ester rich copolymers were soluble in less polar solvents such as chloroform, while
the amide rich copolymers required formic acid for good solubility. All o f the
poly(ester-amide)s were hydrolyzed by Rhizopus arrhizus lipase and by hog liver
esterase.
Objectives - Microwave irradiation has been investigated to conducted ring opening
polymerizations o f e-caprolactam, e-caprolactone and mixture of these monomers to
synthesize nylon-6, poly(e-caprolactone) and copoly(amide-ester) in this chapter. For
energy conservation and environmental considerations, it would be o f interest to
investigate the application of microwave energy to synthesize organic compounds and
polymers. The physical and mechanical properties o f the microwave synthesized
polymers were measured to evaluate microwave effect in ring opening polymerization.
In addition, the microwave synthesis method from homopolymerization studies was
extended to anionic copolymerization studies o f e-caprolactone with e-caprolactam,
using lithium tri-tert-butoxyaluminohydride as a catalyst. The effects o f catalyst
concentration, reaction temperature and time on the copolymer composition and other
properties were investigated.
4.2 EXPERIMENTAL
4.2.1 Starting Materials
e-Caprolactone, e-caprolactam, 1,4-butanediol, o-aminocaproic acid and lithium tritert-butoxyaluminohydride from Aldrich Chemical; e-caprolactone 99% from Janssen
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
86
Chemical and stannous octoate from PFALTZ & BAUER, Inc.. e-Caprolactam was
recrystallized from acetone three times and dried in vacuum. e-Caprolactone (grade
99+%) and others were used without further purification.
4.2.2 Dielectric Property Measurement
Complex dielectric constants of e-caprolactam, e-caprolactone, and a mixture o f ecaprolactam and e-caprolactone (ratio 2:1) were determined by Microwave Properties
North (MPN), a system developed based on the cavity perturbation technique, in the
microwave frequency range from 0.4 GHz to 3 GHz. The data were provided by Dr.
Ron Hutcheon, Microwave Properties North (MPN), Ontario, Canada.
4.2.3 Microwave Equipment and Reaction Set-up
A variable frequency microwave furnace (VFMF) model LT 502Xb (Lambda
Technologies,
Inc.)
was
used to synthesize poly(e-caprolactam)
and
poly(e-
caprolactone) and poly(e-caprolactam-co-e-caprolactone). A TEFLON container was
selected as a synthetic reactor due to its transparency to microwave energy. A TEFLON
tube was connected to the TEFLON reactor to provide a nitrogen blanket during the
reactions. The polymerization temperature was measured and controlled by a grounded
Omega K type thermocouple, which was calibrated by a Luxtron optical fiber
temperature probe beforehand. A low microwave input power, 50 - 70 W, was applied
in these processes due to the high microwave absorption o f the starting materials. The
temperature control was achieved by changes in the applied electric field which was
programmed to maintain a set temperature by the pulse power on-off.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
87
4.2.4 Microwave Synthetic Procedure
Poly(e-caprolactam): e-Caprolactam with 10 mol% initiator o f co-aminocaproic
acid was placed in the microwave oven in the presence o f nitrogen at 250°C for several
hours. The starting materials were mixed and grounded thoroughly before the reaction.
The crude products were extracted by hot methanol for over 16 hrs and then dried in the
vacuum at 90°C overnight.
Poly(e-caprolactone): e-Caprolactone was mixed with small amount o f stannous
octoate as a catalyst, and approximate 1 mol% of 1,4-butanediol used as initiator. The
reactants were placed in the microwave oven for 2 hrs at different temperatures. The
crude products were dissolved in chloroform and precipitated from hexane. The
precipitate was dried in the vacuum overnight.
Poly(e-caprolactam-ctf-e-caprolactone): e-caprolactam and e-caprolactone with a
molar ratio of 2 to
1 were mixed with lithium tri-tert-butoxyaluminohydride,
LiAl[OC(CH 3 )3]3 H. (— 1% mol of total reactants). The mixture was heated up to the
certain temperature via microwave energy under the nitrogen blanket. The crude
product was dissolved in the mixture of hexafluoroisopropanol and chloroform and
precipitated from dimethyl ether, filtered and dried in vacuo at 60°C for 4 hours to give
a white product.
4.2.5 Thermal Synthesis of Poly(s-caprolactam-co-e-caprolactone)
The same reactant ratios and catalyst used in the microwave procedures were used
for the thermal synthesis. The reaction mixture was added to a 25 ml round bottom
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
88
flask, and placed in a preheated oil bath at 160°C for different time periods. Nitrogen
atmosphere was used for all the reactions. The purification o f thermal products was
performed in the same manner as with the microwave synthesized materials.
4.2.6 C haracterization
FT1R - Infrared spectra were taken directly on film by using a Nicolet Magna 560
FT-IR system. Nylon-6 film was cast at 100°C in vacuum from 90% formic acid, PCL
film was cast at vacuum from the solution of THF, and the copolymer film was cast at
60°C in vacuum from the solution of hexafluoroisopropanol and chloroform.
DSC - Differential Scanning Calorimetry (DSC) of nylon-6, poly(e-caprolactone)
and poly(s-caprolactam-co-e-caprolactone) was performed by a TA Instrument 2100
Thermal Analysis DSC 2920, at a heating rate of 20°C/min under nitrogen atmosphere
at a flow rate o f 20 cm3/min.
GPC - A Millipore Model 150-C Gel Permeation Chromatography (GPC) system
was applied to determine the molecular weight of PCL and copolymers. THF was used
as solvent for PCL, and NMP was used as the solvent for copolymers. The copolymers
were pre-dissolved in a small amount of hexafluoroisopropanol and mixed with NMP
(containing 0.05 M LiBr). The GPC data were calibrated by PMMA standard samples.
DMTA - Dynamic mechanical thermal analysis (DMTA) o f copolymer films was
measured by a Polymer Laboratories DMTA Mk II. Poly(caprolactam-co-caprolactone)
samples were melted at 140-150°C and pressed to -0.5 mm thick films. The
measurement was conducted by a single-bending mode from -80°C to 60°C at a heating
rate of 2°C/min.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
89
NMR —High resolution solution nuclear magnetic resonance (NMR) spectra were
acquired using a Bruker DMX500 NMR spectrometer. IHNMR spectra o f nylon-6 and
copolymers were collected from the solutions of deuterated trifluoroacetic acid (dTFA). IHNMR spectra o f PCL were collected from the solution of deuterated
chloroform frZ-CDCh). Chemical shifts were given with respect to tetramethylsilane
(TMS).
Solution Viscometry —Molecular weights of nylon-6 were determined by solution
viscometry. A Ubbelodhe viscometer, with a capillary diameter o f 0.25 - 0.30 mm, was
used to measure the viscosity o f the solutions at 25°C. Samples were dissolved in 85%
formic acid to prepare a solution with a concentration o f 0.005 g/ml.
Tensile Test —Tensile moduli and strengths of nylon-6 and PCL were measured by
Instron 1011 (ASTM 638), with a cosshead rate of 20 mm/min for PCL samples. The
dumbbell-like samples have a nominal size around 20 mm x 4.8 mm x 0.5 mm. Five
samples for each test.
TGA - Thermogravimetrical analyses (TGA) of nylon-6 products were performed
by using a TA Instrument 2100 Thermal Analysis TGA 2950 at a heating rate of
10°C/min under nitrogen atmosphere.
4.3 RESULTS AND DISCUSSION
Based on the molecular relaxation frequencies of e-caprolactam and e-caprolactone,
a variable microwave center frequency was chosen at 4.69 GHz with a frequency width
o f 1.0 GHz in this study. A low microwave forward power was generally selected to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
90
minimize any possible temperature deviation from the overheat o f the grounded Omega
thermocouple at high power range. Typical temperature-time profiles o f microwave
processes of poIy(e-caprolactam), poly(s-caprolactone) and poly(e-caproIactam-co-ecaprolactone) are shown in Figure 39. Temperature control was achieved by changing
the applied electric field resulting from the computer-controlled microwave forward
power on-off. The heating rate o f the microwave process depends on the forward
power, incident center frequency, the chemical structure and physical state o f materials.
c-cap rolarctanr
gfe-caprol; idtam + i:-cap ro lg c to n e
e-< iaproTacf o n e
Figure 39. Temperature-Time Profile of Microwave Process
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
91
4.3.1 Dielectric Properties in the Microwave Region
The polar nature of e-caprolactam and e-caprolactone results in materials with a high
dielectric constant. The microwave dielectric properties o f these materials, e ’ and e”, are
a measure of the ability o f the individual molecular dipoles to respectively move in and
absorb power from the alternating microwave electric field. The dielectric properties of
e-caprolactam, e-caprolactone, and a slurry mixture o f e-caprolactam and ecaprolactone (ratio 2:1) were determined at six different frequencies in the microwave
range, 397 MHz, 912 MHz, 1.429 GHz, 1.948 GHz, 2.466 GHz and 2.985 GHz,
respectively (Figure 40-42). For microwave interaction, the polymer dielectric
properties at the macroscopic level depend on their dipole moment at the molecular
level. The rapidly changing electric fields induced by microwave energy cause changes
o f electron reorientation and distortions of induced or permanent dipole moments o f the
molecules. The amide group and ester group in e-caprolactam and e-caprolactone act as
permanent dipoles in these molecules. For e-caprolactam with a melting point at ~7l°C,
its dielectric loss tangent values, tan5, were below 0.01 at room temperature and
increased dramatically above 75°C, approaching a value o f tan5 o f 0.46 at 100°C as
shown in Figure 40a. The relative dielectric constant increased from 2.8 at room
temperature to 14 at 100°C (Figure 40b). This strongly indicates that e-caprolactam
would absorb microwave energy effectively once the molecule reaches the liquid state,
i.e. the mobility o f the polar functional groups in the starting materials is sufficient to
provide the required change of dipole reorientation in the microwave electromagnetic
field for efficient energy absorption. Since s-caprolactone is liquid at room temperature,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
92
its loss tangent value was high initially (0.79), as expected, and maintained high values
through the entire temperature range, as shown in Figure 41a. The relative dielectric
constant maintained a high value from room temperature to 110°C (-42) (Figure 41b).
The dielectric properties o f a slurry mixture o f e-caprolactam and e-caprolactone
were also determined from room temperature to 160°C to study the microwave response
as the e-caprolactam melted, dissolved and formed a solution with e-caprolactone. This
is close to the simplified situation encountered in our microwave copolymerization
studies without the consideration of the effect caused by molecular chain growth and
viscosity increase during the catalytic polymerization reaction. The response o f the
mixture (in Figure 42) displayed the additive properties derived from individual ecaprolactam and e-caprolactone properties. The initial loss tangent values at the room
temperature were high due to the liquid-state e-caprolactone and increased to maximum
value once e-caprolactam reached its liquid state and a solution was formed. The
dielectric values o f both e-caprolactam and e-caprolactone in the liquid state were fitted
to a Debye relaxation model. e-Caprolactam gave a molecular thermal relaxation
frequency from -3 .6 GHz at 100°C increasing to -9 .8 GHz at 160°C, and ecaprolactone gave a molecular thermal relaxation frequency from -2.4 GHz at 30°C
increasing to -9.1 GHz at 100°C, both of which in the range of our microwave
irradiation studies.
4.3.2 Ring Opening Polymerization of e-CaproIactam via Microwave Irradiation
As shown in Figure 39, the heating rate increased sharply after 75°C. The dielectric
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
93
values (Figure 40) also increased significantly in the same temperature range. This
suggested that e-caprolactam would absorb microwave energy effectively once the
compound reaches the melting state. Based on the conventional thermal polymerization
o f e-caprolactam, the microwave polymerization was carried out at 250°C for lhr, 2hr
and 3 hr, and 280°C for 2hrs, respectively. No polymerization was observed at 220°C.
The color of the crude products changed from white to slightly light brown with the
increase in reaction temperature. The slightly dark color was probably due to
degradation at high reaction temperature. After extraction by hot methanol, the purified
sample showed no weight loss until 360°C, and provided the same thermostability as a
commercial nylon-6 sample, Capron 8202NL, from AlliedSignal Corporation, as
displayed in Figure 43. Also, no absorption at 870 cm*1, corresponding to caprolactam
monomer,[9S| was observed in IR spectra o f poly(s-caprolactam) (Figure 44). This
indicated that the high purification of nylon products was achieved. DSC thermograms
o f those nylon products synthesized at 250°C are shown in Figure 45, giving a glass
transition temperature around 53 -54°C and a melting point at 221 - 222°C, which are
close to the values o f the commercial product (Tg at 56.8°C and Tm at 227.4°C in
Figure 46).
The molecular weight o f the microwave synthesized poly(e-caprolactam) was
determined by solution viscosmetry.[32'991 The relative viscosity o f the dilute solution,
7jr, was measured by the ratio o f solution viscosity to solvent viscosity. The intrinsic
viscosity, [ tjJ, was calculated from a linear regression extrapolating to the infinitesimal
dilution of the inherent viscosity, rjinh = Inrj/c, and the reduced viscosity, (Tjr~l)/c, as a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
94
0.5
r
0 .4 -
—O —
—'v1—O —O -
3 9 7 MHz
9 1 2 M Hz
1. 4 29 G H z
1.948 G H z
2 .4 6 6 G H z
2 .9 8 5 G H z
TanS
0.3
0.2 H
0 .0
0 .0
Q
-
= 0 -0 0 0 ^
T"
40
20
60
80
1 00
120
1 40
160
180
T e m p e r a t u r e (°C)
(a)
16 -r-
I
14
JI
—O— 3 9 7
i
12
—o 10
M Hz
9 1 2 M Hz
1 .4 2 9 G H z
1 .948 G H z
2 .4 6 6 G H z
2.9 85 G H z
-
8
6
-
O
20
0 -0 0 0 0
40
60
80
1 00
120
140
160
180
T e m p e r a t u r e (°C)
(b)
Figure 40. Dielectric Properties of s-Caprolactam in Microwave Range
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
95
0.9
0.8
—c — 3 9 7 MHz
9 1 2 MHZ
—a - 1.429 G H z
V
1.948 G H z
2.466 G H z
O
-
0.7
<X
-s.
0.6
-
V.
0.5 0.4 -
x>
\
o
i
"O,
0.3 -
OO
~-V\
a.
0.2
a-
o0.1 H
o—
0.0
o_
-o-
T
20
40
60
80
10 0
120
T e m p e r a t u r e (°C)
(a)
50
O-
45 □
o~f? ~ O- —qI! "O-q
✓s — O —o -----
. o-
- a
40 -j
I
35 -j
30 -I
I
ts
25 -j
20
-
15
397 MHz
912 MHz
1.429 GHz
1.948 GHz
2.466 GHz
10
5
0
— i—
20
—
i—
40
60
80
100
120
T em p eratu re (°C)
(b)
Figure 41. Dielectric Properties o f e-Caprolactone in Microwave Range
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
96
0.6
0.5 -
oa
0.4 -J
•a
—
\
o ° ..
a-.._
0.3
0 .2
—O— 397 MHz
- a 912 MHz
—C*- 1.429 GHz
—■
<
7
— 1.948 G Hz
- O
2.466 GHz
—O - 2.985 G H z
-I
S
o - .
0.1
O — O -
0 .0
—
o
j—
20
40
60
80
100
120
140
160
180
T em perature (°C)
(a)
20
18 16 14
12
-u 10 I
8—q _ 3 9 7 MHz
6
—C3—
—V—
—o —
—o —
4
2H
0
91 2 MHz
1 .429 GHz
1 .948 G Hz
2 .4 6 6 GHz
2 .9 8 5 GHz
—i—
20
40
60
80
100
120
140
160
180
T em p erature (°C)
(b)
Figure 42. Dielectric Properties of Mixture of e-Caprolactam and s-Caprolactone
in Microwave Range
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
97
function o f solution concentration, c. An intrinsic viscosity plot of commercial nylon-6
by this method is presented in Figure 47, where c ' = c/co, and Co is the original solution
concentration before dilution.
The viscosity-average molecular weight of the nylon products can be obtained by the
Mark-Houwink equation:
In } - m -
(6)
where constants K and a are equal to 22.6 x 10'3 cm3/g and 0.82 in this solution
system.1’001 On this basis, the viscosity-average molecular weight of the commercial
nylon-6 was around 25.1 kg/mol, which is in the reasonable range according to the
company GPC data report (Mw at 34.0 kg/mol and Mn at 4.14 kg/mol). The crystallinity
o f the microwave synthesized poly(e-caprolactam) was around 31-33 %, calculated
from the endothermic melting peak of DSC thermograms based on the heat o f fusion o f
the crystalline region as 190.6 J/g. DSC data used for measuring crystallinity were
collected from the sample reheated above melting temperature after cooling down from
the melts. The commercial nylon-6 gave identical results. DSC data of microwave
synthesized and commercial nylon-6 cooling from melts were also collected at a cooling
rate o f 20°C/min. The exothermic peak for polymer crystallization during the cooling
exhibited a peak temperature at 179°C for microwave synthesized nylon-6 and 167°C
for the commercial nylon-6, and a crystallinity o f 36% for microwave synthesized
polymers and 31% for the commercial one. Results o f microwave synthesized nylon-6
are summarized in Table 10. Both molecular weights and yields increased with
increasing reaction time. It was noted that microwave synthesis at 250°C for 2hrs
provides products with similar physical properties and yield to that at 250°C for 3hrs.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
98
100-
>80
M
>60
s .
Microwave
Commercial
T
I
I
40-
-4 0
20-
100
300
Figure 43. TGA Curves o f Microwave Synthesized- and Com mercial Nylon-6
Poly(e-caprolactam)
Poly(e-caprolactone)
3500
3000
2500
2000
1500
1000
Wavenumbers fcrrwn
Figure 44. ER Spectra of Microwave Synthesized Nylon-6 and PCL
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
osc Heat Flow (mw)
99
MW 2 5 0 C -1 H R
-1 5 -
MW 2 5 0 C -2 H R
MW 2 5 0 C -3 H R
-2 5 -
-3 0
-100
-5 0
50
100
T em p e ra tu r e
150
I v e r la y
(*C)
^F !o O
TA I n s ? . 2 1 0 0
Figure 45. DSC Thermograms of Microwave Synthesized Poly(b-Caprolactam)
- 0 .3 -
o
2
O
- 0.0-
u_
m
f
1 .3 -
227.
1.0
100
-5 0
50
100
150
Te«peratur# (*CJ
200 PSC^.OB OuPon? 2100
Figure 46. DSC Thermogram of Commercial Nylon-6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
0.60
0.55
0.50
o
0.45
0.40
0.35
0 30
yc-
0.25
0.20
0.0
0.4
0.2
0.6
0.8
1 .0
c*
Figure 47. Intrinsic Viscosity Extrapolation of Nylon-6 in 85% Formic Acid
Table 10. Summary of Synthesis of Poly(e-caprolactam) via Microwave Energy
Reaction Condition
Yield (%) Tg(°C) Tm(°C)
(after
purified)
No reaction observed at 220°C
Color
Molecular
Weight
(kg/mol)
250°C-lhr, N2
79.1
53.6
221.5
white
19.3
250°C-2hr, N2
87.1
53.2
221.6
white
24.2
250°C-3hr, N2
88.5
53.1
221.3
26.8
280°C-2hr, N2
82.6
53.7
221.7
very slightly
brown
light brown
Commercial Product
(Capron 8202NL)
—
56.8
227.4
white
25.1
--
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15.5
101
Therefore, the process at 250°C for 2 hrs was selected as microwave process conditions
to synthesize poly(s-caprolactam).
4.3.3 Ring Opening Polymerization of 8-CaproIactone via Microwave Irradiation
Ring opening polymerization o f E-caprolactone was performed in the microwave
oven using a lower forward power system since the s-caprolactone monomer was at the
liquid state. The processing temperature-time profile is presented in Figure 39. The ecaprolactone was mixed with stannous octoate and placed in the microwave oven for 2
hrs by controlling at different temperatures o f 200°C, 180°C, 150°C and 120°C,
respectively. Two synthesis systems were investigated in this study, one with 1,4butanediol as initiator and the other without. For the second system without 1,4butanediol, trace amount o f water from the catalyst acted as initiator. Results are
summarized in Tables 11 and 12. No obvious reaction was observed in the microwave
process at 120°C for 2 hrs. Product yields and molecular weights decreased with the
increased temperature from 150°C to 200°C. Color of the crude products became darker
from white at 150°C to very light brown at 200°C. This suggests the degradation
probably occur during high temperature processing. The glass transition temperature of
-60 to -62°C and the melting temperature o f 57 to 60°C were similar to the
conventionally prepared poly(e-caprolactone). The high molecular weight, high yield
and lower polydispersity in the sample prepared at 150°C for 2 hrs compared favorably
with other products. The tensile properties of this microwave product without 1,4butanediol were also determined and are listed in Table 13. Compared with literature
data by conventional thermal process, the microwave synthesized product exhibited
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
102
Table 11. Synthesis of Poly(e-caprolactone) via Microwave Energy
Reaction Condition
Yield (%) (after purified)
Color
120°C-2hr w/ and
w/o diol
150°C-2hr, w/ diol
No obvious reaction
—
95.4
wh te
180°C-2hr, w I diol
95.1
Color
increase
200°C-2hr, w/ diol
93.9
very ligh brown
150°C-2hr, w/o diol
92.3
wh te
180°C-2hr, w/o diol
93.3
increase
200°C-2hr, w/o diol
90.1
very ligh t brown
1
Table 12. Physical Properties of Poly(e-caprolactone) via Microwave Energy
Sam ple
Tg (°C)
Tm (°C)
M w (kg/m ol)
M w /M n
150°C-2hr,
w /d ioi
-6 2 .5
5 7 .2
1 1 .9
1.7
180°C-2hr,
w /d iol
-6 0 .9
5 6 .7
9 .9
1.8
200°C-2hr,
w /d iol
-6 2 .5
5 6 .9
1 0.8
1.9
150°C-2hr,
w/o diol
-6 0 .4
5 8 .9
8 6 .3
2 .5
180°C-2hr,
w/o diol
-5 8 .7
6 0 .4
7 8 .7
2 .6
200°C-2hr,
w/o diol
-6 0 .4
5 9 .2
5 0 .7
2 .3
-62
57
44
—
Therm al PCL*
* Severian Dumitriu, Polymeric Biomaterials, Marcel Dekker, Inc. 1994
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
103
Table 13. Tensile Properties of Poly(e-caprolactone) via Microwave Energy
Sample
Tensile
Strength
(Mpa)
21.8 ± 2.5
Strain at
Break (%)
9.9 ± 1.5
Tensile
Modulus
(Mpa)
406 ± 27
7.0
400
16
80
Yield
strain (%)
Microwave PCL
Yield
Strength
(Mpa)
17.3 ±0.8
Thermal PCL*
—
514 ± 7 7
* Severian Dumitriu, Polymeric Biomaterials, Marcel Dekker, Inc. 1994
20
m
a.
2
(/}
0
100
200
300
400
500
600
Strain (%)
Figure 48. Stress-Strain Plot of Microwave Synthesized Poly(s-caprolactone)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
104
higher tensile strength, equivalent tensile modulus, and 6 times higher ultimate
elongation. A typical stress-strain curve is displayed in Figure 48.
The higher or equivalent physical properties generated for microwave-produced
poly(e-caprolactam) and poly(e-caprolactone) in a period o f only 2 hrs demonstrate that
the ring opening polymerization o f e-caprolactam and e-caprolactone can be performed
efficiently and effectively by pure microwave irradiation without any external thermal
heating. The much shorter processing time and the quality products produced, strongly
suggest that the microwave synthesis technique have great potential in this area.
4.3.4. Ring Opening Copolymerization of s-Caprolactam and e-CaproIactone via
Microwave Irradiation
1) Effect of Temperature and Reaction Time
The anionic copolymerization o f e-caprolactone and e-caprolactam via microwave
energy were investigated at different temperatures and reaction times, namely, ©
140°C-1 hr, ® 160°C-1 hr, (D 160°C-0.5 hr, © 180°C-0.5 hr. The ratio o f amide to ester
in the starting materials was maintained at 2 to 1, and the catalyst content was about 1%
mol. Infrared spectra o f copolymers are shown in Figure 49. Absorption at 1732 cm"1
due to ester carbonyl group, and that at 1639 and 1543 cm"1 due to amide carbonyl, are
noted. Solution ‘HNMR spectra o f the poly(amide-ester) copolymer (PAE) are
overlayed in Figure 50(b). The ratios o f amide unit to ester unit in PAE copolymers
were calculated from peak integration values o f methylene unit next to ester group, CKbCOO-. at 4.3 ppm and that next to amide group, -CKbCONH-. at 3.6 ppm. As
shown in Table 14, the yield and amide-to-ester ratio of PAE copolymers increased as
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
105
reaction temperature increased from I40°C to 160°C for I hr and reaction time
increased from 0.5 hr to 1 hr at 160°C. The yield from the 180°C / 0.5 hr run w as close
to that from the 160°C / 0.5 hr run, even though the ratio of amide to ester increased
slightly. The high reaction temperature at 180°C may cause degradation during
copolymerization, based on the relatively lower molecular weight for the run with 1%
catalyst.
2) Effect o f Catalyst Levels
Four different catalyst levels, 0.5 % mol, 1 % mol, 2 % mol and 3 % mol, were also
investigated. The processes were controlled at 160°C / 0.5 hr. These results are
summarized in Table 15. Obviously, the yield and amide-to-ester ratio o f PAE
copolymers increased with increasing catalyst level, especially for a catalyst level up to
3 %. At this level, the amide-to-ester ratio increased to 2 in agreement with the starting
lactone/lactam feed ratio, and yield was also improved to 78.2 %. For a catalyst content
o f 3 % and with a starting ratio of amide to ester of 1.0, the resultant PAE copolymer
exhibited an amide-to-ester ratio of 0.81, which is close to the starting ratio. The
molecular weight was slightly low at higher catalyst content, as might be expected. This
strongly suggests that the e-caprolactone reactivity is greater than the e-caprolactam
reactivity, in agreement with thermal studies o f Goodman et al[38!.
The microstructure of the microwave synthesized PAE copolymer appears to be
random, based on the fact that a single glass transition temperature, Tg, is observed for
each copolymer, as depicted in Figure 51. This random chain structure w as also
confirmed by comparing the Tg determined by the DMTA to the Tg evaluated from the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
106
Fox equation (7)[10l!,
where w refers to mass fraction, Tgi and Tg2 indicate glass transition temperature o f pure
polymer containing unit 1 or 2, respectively. Both the experimental and calculated
values were rather close as shown in Tables 14 and 15.
3) Comparison o f Thermal and Microwave Processes
In order to provide a direct comparison between microwave energy and thermal
energy in anionic copolymerization o f e-caprolactone and e-caprolactam, the same
reactant ratios and reaction conditions were used for the conventional thermal process
and microwave process. The results are summarized in Table 16. Both higher yields and
amide-to-ester ratios (Figure 50) were observed in microwave synthesized copolymers.
The higher yield generated by microwave copolymerization suggests that microwave
energy provides an effective synthetic method by heating the molecules directly through
the interaction between microwaves and molecular dipole moments o f the starting
materials. The higher amide-to-ester ratios in microwave synthesized PAE copolymers
further suggests that microwave energy has a greater affinity to amide units than for the
ester units during copolymerization. This also agrees with the dipole moment values o f
functional groups141 and chemical bonds11021. The amide group (-COONH-) has a dipole
moment of 3-4 D, while the ester group (-COOR-) has a dipole moment o f 1-2 D. The
N-H bond has a dipole moment o f 1.3 D. Dipole moments o f C -0 and C-N bonds are
0.7 D and 0.45 D, respectively. The chemical structures o f e-caprolactone and e-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MW, PAE, cat. 0.5%
MW, PAE, cat. 1%
MW, PAE, cat. 2%
3500
3000
2500
2000
1500
1000
Wavenumbers fcrn-ll
Figure 49. IR Spectra of Microwave Synthesized PAE with Different Compositions
Table 14. Temperature Effect on Microwave Copolymerization
MW PAE
MW PAE
MW PAE
MW PAE
140°C-l hr
160°C-0.5 hr
160°C-1 hr
180°C-0.5 hr
Starting Materials: EstenAmide
1:2
1:2
1:2
1:2
Yield (%)
57.8
61.9
67.4
61.5
1 :0.88
1 : 1.07
I : 1.27
1 : 1.18
Tm (°C), DSC
149.2
148.0
150.6
146.0
Tg (°C), (Taitf), DMTA
-25.0(1 Hz)
-14 (1Hz)
-6.0 (1H z)
-12.0 (1Hz)
Tg(°C), from Fox Equation
-20
-15
-10
-12
Molecular Weight, GPC, M w
22,921 (1.8)
21,984 (2.1)
22,141 (1.7)
17,191 (2.0)
Sample
Ester CHj at 4.3 ppm: Amide
at
3.6 ppm (TFA. NMR)
* MW - Microwave Process
* PAE - Poly(e-caprolactam-co-e-caprolactone)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
108
T H P A E , 1% cat.
J
u
T H P A E . 2 % cat.
T H P A E , 3 % cat.
T'””
r
(a)
M W P A E , 1% cat.
JU
M W P A E . 2% cat.
M W P A E . 3% cat.
i
r-r-r-f -s-.
oce
rnw-r
2
■■■■
6
- ■’■■»..............
5
(b)
Figure 50. !HNMR Spectra of PAE Prepared at Different Catalyst Levels
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
109
0.6
0 .5
0 .4
<0
a.
0 .3
LLJ
a>
o
0.2
1
0.0
-100
-80
-60
-40
0
-20
40
20
60
80
100
T e m p e r a t u r e (°C)
Figure 51. DMTA Curve of Microwave Synthesized PAE
(160°C for
Vi
hr, catalyst
2% )
Table 15. Catalyst Effect on Microwave Copolymerization
Sample
MW PAE, 0.5%
MWPAE, 1%
MW PAE, 2%
MW PAE, 3%
MW PAE, 3%
l60°C-0.5 hr
l60°C-0.5 hr
160°C-0.5 hr
160°C-0.5 hr
l60°C-0.5 hr
Starting Materials: Ester Amide
1:2
1:2
1:2
1:2
1:1
Yield (%)
45.7
61.9
70.1
78.2
66.7
Ester CH2 at 4.3 ppm Amide
1 :0.60
1: 1.08
1: 1.36
1:2.00
1:0.81
Tm(°C), DSC
128.8
151.5
154.8
145.4
142.3
Tg (°C), (TanS), DMTA
-27.0 (life)
-14.0 (1Hz)
-7.5 (life )
6.0 (life)
-22.0 ( life )
-30
-15
-8
4
-22
26,492(2.1)
21,984(2.1)
21,279(2.0)
16,197(1.5)
17,026(1.5)
CH2 at 3.6 ppm (TFA, NMR)
Tg
from Fox Equation
Molecular Weight, GPC, M\v
* MW - Microwave Process
* PAE - Poly(s-caprolactam-co-e-caprolactone)
* 0.5 %. 1 %. 2 %, 3 % - Catalyst level
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
110
caprolactam are very similar except for the amide and ester functional groups, and thus
higher values o f dipole moment for the amide group would provide better microwave
absorption in the liquid state and show greater reactivity during copolymerization. The
higher Tg’s o f microwave synthesized copolymers, relative to the thermally synthesized
copolymers, were a result o f the higher amide-to-ester ratios.
Table 16. Comparison of Microwave Energy and Thermal Energy in
Copolymerization Reactions
TH-PAE, 1%
TH-PAE. 2%
TH-PAE, 3%
M W PAE, 1%
MWPAE, 2%
M W PA E, 3 %
I60<1C-0.5hr
160°G0.5hr
1 6 (f& 0 ^ h r
16CrO0.5hr
160°C O ihr
I60°C-0.5 h r
1:2
1:2
1:2
1:2
12
1:2
Yield (°/Q
51.2
527
57.0
61.9
70.1
78.2
Ester CH, at 4 3 ppm: Amide
1:0.61
1 :1.19
1:129
1:1.08
1 : 126
1:200
-25.0 (life)
-18.5 (IFfe)
-14_5 (ltfe)
-14.0 (IFfe)
-7.5 (life)
6.0 (IFfe)
Sample
Starting Materials:
EsterAmide
CTfc ar 3.6 p p n (TEA., NMR)
Tg (“Q , (TanS), DMTA
Tg CQ, f a n Fax Equation
-29
-12
-9
-15
-8
4
M deatlar Wagfat, GPC, \h v
25.423(1.4)
19,862(l_5)
17.141(1.6)
21,984(21)
21,279(20)
16.197(13)
•
•
•
•
j
MW - Microwave Process
TH - Thermal Process
PAE - Poly(e-caprolactam-co-e-caprolactone)
1 %, 2 %, 3 % - Catalyst level
4.4 CONCLUSIONS
Ring opening polymerization of e-caprolactam and e-caprolactone using a variable
frequency microwave furnace has been investigated. Compared with the commercial
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Ill
products, equivalent physical and mechanical properties were obtained from the
microwave-produced nylon-6 and poly(e-caprolactone). The results revealed that
quality nylon-6 and poly(e-caprolactone) can be prepared by microwave irradiation
within 2 hours, versus more than 12 hours by the commercial thermal process. Ring
opening copolymerization o f e-caprolactam and e-caprolactone via microwave energy
showed that amide composition increased with the increased reaction temperature, time
and catalyst level, while molecular weights o f copolymers decreased with the increased
catalyst level. Glass transition temperatures of copolymers increased with increased
amide composition o f the copolymers, which are consistent with the T g’s calculated
from the Fox Equation based on the random microstructure of copolymers. Compared
with the corresponding thermal products, microwave synthesized copolymers gave
higher yield, higher amide composition, higher glass transition temperature and
equivalent molecular weight.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
C H A PT E R 5: A SO LID STATE 13C NM R STU D Y O F THE C U R E R EACTIO NS
O F 13C -L A B E L E D PHENYLETHYNYL EN D -C A PPE D POLYIMIDES
5.1 INTRODUCTION
Nuclear Magnetic Resonance (NMR) spectroscopy has been widely used to
characterize chemical structures. Many high-resolution techniques have been developed
to study polymers in solid state. Magic-angle spinning (MAS) and cross polarization
(CP) are most popular means applied in solid state I3C NMR measurement.1103' 1051 The
distribution of dipole-dipole interactions between many interacting nuclei and the
anisotropy o f chemical shift lead to line broadening in the solid-state NMR spectra. In
the liquid state, rapid and random molecular motion averages the dipole-dipole
interaction and chemical shift anisotropy to zero, giving sharp lines in NMR spectra.
For solid state, if a solid sample is rotated about an axis with a angle o f 54.7° with the
direction o f the external magnetic field, the dipolar broadening from anisotroptic parts
o f interactions can be minimize to zero. This technique is known as magic angle
spinning (MAS), and is used to achieve high-resolution solid-state NMR spectra. Cross
polarization (CP) is applied to increase the sensitivity in the NMR o f sparse nuclear
spins like 13C by taking advantage of the spin interaction with the abundant spins like
1H. The sensitivity is improved by the increase of the spin-up state population in the
sparse nuclei and the decrease of the relaxation time of the sparse nuclei. However, the
cross polarization technique will enhance the peak absorption for the sparse nuclei,
which will prevent accurate quantitative analysis. Therefore, the technique using single
pulse with gated proton decoupling during acquisition (1PDA)11061 was applied in this
112
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
113
study to provide quantitative high resolution solid state 13C NM R spectra.
The ethynyl group has become an important cure site for ethynyl end-capped
polyimide systems that have found applications in the electronics industry as low
dielectric coatings. The cure process of these end-capped polyimides has been widely
investigated. Studies by Swanson et al1531 using solid state I3C NMR spectroscopy have
brought some insight into the cure mechanism o f this resin system. In these studies,
trisubstituted benzenes derived by cyclotrimerization, thermal cyclization via DielAlder reactions or from biradical polyene products and condensed polycyclic aromatic
structures and alkenyl aromatics have been proposed. Other crosslinked structures o f
acetylene-terminated and nadic end-capped polyimides studied by this technique have
also been reported.t52,107' 1081
Pickard et al[491 have investigated the kinetics and mechanism of bulk thermal
polymerization o f 3-phenoxyphenyl acetylene. The soluble products generated in the
polymerization reaction were analyzed. The authors revealed that the soluble material
consists mostly o f (1) polyene having a trans-cisoidal configuration and (2) lower
molecular weight
species
such as
l,3,5-tris(3-phenoxyphenyl)benzene
and
3-
phenoxyphenyl acetylene dimer.
For the phenylethynyl terminated polyimides, although the triple bonds o f simple
terminated phenylethynyl imide molecules and more complex imide oligomers are
expected to react by an ethynyl to ethynyl addition chain reaction, or a simple ethynyl to
ethynyl chain extension reaction, a more complex ethynyl to ethynyl trimerization
reaction, ethynyl to vinyl and vinyl to vinyl crosslinking and branching reactions are
also very viable possibilities. The cure mechanism o f these systems is still poorly
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
114
understood. The expected steric and electronic effects o f the phenyl-ended ethynyl
group relative to hydrogen in ethynyl end-capped oligomers and the bulky nature o f the
oligomer repeat unit most likely contribute to the difficulties in elucidating the cure
reaction mechanism of these phenylethynyl end-capped resin systems. Therefore, the
simplest model o f the PETI-5 oligomer, PEPA-3,4’-ODA, was selected to develop an
understanding of the cure process in the expectation that this will contribute to our
understanding o f the PETI-5 imide oligomer cure process.
Objectives — Solid-state 13C-NMR is applied to investigate the general cure
reactions o f phenylethynyl-terminated polyimides in this chapter. Solid state l3C NMR
with a magic-angle spinning can provide high resolution spectra and is utilized to
determine the complicated chemical structure of solid polymers with great significance,
especially for the insoluble and infusible thermosetting polymers. As a continuous study
in the cure mechanism of phenylethynyl end-capped polyimides from Chapter 2, the
cure process of the 13C-seIectiveIy labeled model compound PEPA-3,4’-ODA (Scheme
7) and PETI-5 imide oligomer (Scheme 8) was investigated by solid state 13C NMR
techniques.
C —C
C -C
Scheme 7. Chemical S tructure of I3C Labeled PEPA-3,4’-ODA
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
115
O
O
0
6
Scheme 8. Chemical Structure of I3C Labeled PETI-5
5.2 EXPERIMENTAL
5.2.1 General Information
Unlabeled
4-phenylethynylphthalic
anhydride
(4-PEP A),
and
l,3-bis(3-
aminophenoxyl) benzene (APB) and LaRC PETI-5 resin powder were provided by
Imitec Inc.. Phenylacetylene-l,2-13C2 (>99% atom 13C) was purchased from Isotec Inc..
3,4’-oxydianiline (3,4’-ODA) was obtained from Mitsui Petrochemical Industries, Inc..
3,3’4,4’-biphenyltetracarboxylic dianhydride (BPDA) was purchased from Chriskev
Company, Inc.. 1,2-13C labeled 4-phenylethynylphthalicanhydride (13C-PEPA) was
prepared from l,2-l3C-phenylacetylene and 4-Bromophthalic anhydride (4-BrPAN) in
triethylamine (Et 3 N) in Dr. Scola’s group.
5.2.2 Synthesis of 13C labeled 3,4’-Bis[(4-phenylethynyl)phthalimido)diphenyl
Ether (13C-PEPA-3,4’-ODA)
The l3C labeled imide model compound (Scheme 7) used in this study was 13CPEPA-3,4’-ODA prepared from 13C labeled 4-phenylethynylphthalic anhydride (l3CPEPA) and 3,4’-oxydianiline (3,4’-ODA) in N,N’-dimethylformamide (DMF) solvent.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
116
13C-PEPA (0.45 g, 1.8 mmol) and 3,4’-oxydianiline (3,4’-ODA) (0.18 g, 0.9 mmol)
were each dissolved in DMF (2.0 ml) respectively. Those two solutions were then
mixed together and stirred for 12 hrs at room temperature. An excess o f acetic
anhydride was then added, and the solution was refluxed for an hour, cooled, and the
product which precipitated was filtered and air dried. The crude product was
recrystallized from DMF, rinsed with methanol, and dried in vacuo at 200°C for 6 hrs.
The light yellow powder (0.484 g) was obtained with a yield o f 80.5 %.
DSC, melting point peak at 297°C (sharp), at a heat rate o f 20°C/min under nitrogen.
ER. (KBr pellet), 2141 cm'1 ( 13C=l3C), 1778 and 1713cm'1 (imide C=0). Elemental
analysis, calculated for
C, 80.1%, H, 3.6%, N, 4.2%; found: C, 79.5%,
H, 3.8%, N, 4.2%.
5.2.3 Synthesis of I3C Labeled Phenylethynyl-terminated Imide Oligomer (13CPETI-5)
The I3C labeled imide oligomer (Scheme 8) used on this study was PETI-5 prepared
from
13C
labeled
4-phenylethynylphthalic
anhydride
( 13C-PEPA),
3,3’4,4’-
biphenyltetracarboxylic dianhydride (BPDA), 3,4’-oxydianiline (3,4’-ODA) and 1,3bis(3-aminophenoxy) benzene (APB) in N-methyl-2-pyrrolidinone (NMP) solvent by
the procedure of Bryant et al.[28'291 BPDA (2.672
g,
9.08 mmol) a mixture of 3,4’-ODA
(1.699 g, 8.50 mmol) and APB (0.439 g, 1.50 mmol) were each dissolved in NMP (7.0
ml), and then mixed and stirred for 1 hr at room temperature. l3C-PEPA (0.450 g, 1.80
mmol) was then added, the solution was diluted with NMP to approximately 30 % w/w
and stirred for another 24 hrs under the nitrogen. Toluene (20 ml) and small amount of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
117
NMP were then added, and the solution was heated to reflux using a Dean-Stark trap to
remove water as an azeotrope. Toluene (20 ml) and trace o f water were collected in the
first hour. The solution was refluxed for another 4 hrs and cooled. The solid which
precipitated on addition o f the solution to methanol (300 ml) was filtered and dried
under a vacuum at 200°C overnight to yield the yellow powder (4.65 g, 94.9 %).
DSC (at a heat rate o f 20°C/min under nitrogen), glass transition temperature (Tg) at
224°C and melting point peak at 351°C. IR(KBr pellet), 2137 cm '1 ( I3C s l3C), 1778 and
1716cm'1 (imide C=0). Elemental analysis, calculated for PETI-5: C, 74.4%, H, 3.2%,
O, 16.7%; found: C, 73.9%, H, 3.4%, O, 16.8%.
5.2.4 C haracterization
DSC - Differential Scanning Calorimetry (DSC) of imide model compounds,
PEPA-3,4!-ODA and 13C-PEPA-3,4’-ODA, imide oligomers, PETI-5 and l3C-PETI-5
was performed by a Perkin -Elmer DSC, 7 series Analysis System at a heating rate of
20°C/min under nitrogen atmosphere at a flow rate o f 20 cm3/min.
GPC - A Millipore Model 150-C Gel Permeation Chromatography (GPC) system
was applied to determine the molecular weight of l3C-labeled and unlabeled PETI-5
oligomer, NMP with 0.05 M LiBr was used as a solvent. The results were calibrated by
standard poly(methyl methacrylate) (PMMA).
IR - Infrared spectra of the model compound and PETI-5 (KBr pellet) were taken
by using Nicolet Magna 560 FT-ER. system.
Solution NM R - High resolution solution nuclear magnetic resonance (NMR)
spectra were acquired using a Bruker DMX500 NMR spectrometer. The solution NMR
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
118
samples o f 13C-PEPA-3,4’-ODA and 13C-PETI-5 were prepared from their NMP
solution with additional deuterated dimethyl sulfoxide (DMSO-cfe). PEPA-3,4’-ODA
and PETI-5 were dissolved in NMP solvent by heating to 200°C for 5 sec.
5.2.5 Solid-state NM R Technique
Solid-state l3C MAS NMR spectra were acquired by using single pulse gated proton
decoupling experiments (1PDA) at 75 MHz on a Chemagnetics CMX300 NMR
spectrometer with a commercial double-bearing 5 mm MAS probe. The magic angle
was set using the 79Br resonance o f KBr. Samples (~ 100 mg) were packed in a Zirconia
pencil rotor and spun at the magic angle at 5 kHz, 8 kHz or 10 kHz as desired. The 90°
pulse widths for lH and 13C were 5 ps. A relaxation time of 75 s ( >5 T l) was applied
for single pulse experiments to allow thermal equilibrium, and at least 16 scans were
utilized to achieve a proper signal-to-noise ratio. For more accurate data, some spectra
were collected after 128 scans. A 76 kHz proton decoupling field was used during the
acquisition o f the free induction decay. Chemical shifts were given with respect to
tetramethylsilane (TMS) by using an external sample o f solid glycine (176.03 ppm) for
13C as the secondary reference.
5.3 RESULTS AND DISCUSSION
Within the last five years, crosslinked structures o f the cured PETI polyimides have
been proposed based on the knowledge o f acetylene crosslinking chemistry and the
expected steric effects o f phenyl groups on the ethynyl structures. The principal views
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
119
on the crosslinked structures were focused on the polyene structures (Scheme 9a,
reaction (l))’27’ and cyclo-olefinic ring structures (Scheme 9b, reaction (6)) derived
from intra or intermolecular reactions of the polyene. All the possible ethynyl reactions
are summarized in Scheme 9, which includes the hexaphenyl substituted aromatic ring
structure (Scheme 9a, reaction (2)). Synthetic l3C-PEPA-3,4’-OD A and °C-PETI-5
gave ER spectra and DSC thermographs identical to the unlabeled model compound and
LaRC PETI-5 oligomer (Figures 52 & 53). The NASA LaRC PETI-5 and synthetic l3CPETI-5 gave similar molecular weights as determined by GPC based on the PMMA
calibration. A number-average molecular weight (Mn) of 8,380 for NASA LaRC PETI5 and o f 8,340 for synthesized l3C PETI-5 were obtained. A polydispersity o f 1.7 for
both samples was obtained. The GPC-Differential viscometry (GPC/DV) technique116’
gave a Mn ~ 8,050. The stoichiometry of PETI-5 was adjusted to give a theoretical Mn
o f 5,000, but the experimental data show that it is above 8,000.
In order to identify the chemical structures o f synthesized model compound and
PETI-5 resin, and to verify solid state NMR spectra, I3C solution NMR spectra o f the
starting materials, PEP A, PEPA-3,4’-ODA and PETI-5 were recorded and are shown in
Figure 54. The 13C-enriched ethynyl carbons gave strong peaks at 87-94 ppm, aromatic
carbons at 118-157 ppm, and imide carbonyl group at 167 ppm.
5.3.1 PEPA-3,4’-ODA Model Compound Studies
As an aid in the understanding of the crosslinking reaction and reaction products, a
model compound, PEPA-3,4’-ODA, was used. This compound helped to resolve the
signal overlaps generated by the PETI-5 oligomer during the cure process. Selectively
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
120
13
13
c=c
\ l 3
/
13 J f
c= c
\W
13^
Polyene Structure
Aromatic Ring Structure
Scheme 10. (a) Polyene and Aromatic Ring Structures from Phenylethynyl Group
Reactions
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
121
C) c ^ n >
>
v
_
r~ (
/
^
^
^
c
=
/
^
V
- J
O
Intramolecular
ene-ene
^ c'
o
c^ c= c f '
^
o
r
^
r
+
c=
A
□ >
.
>
— c:
x
0
- ©
Intermolecular
ethynyl-polyene
\ r_ r ^
-c —
.t**
'C'""
c> ^ 0 >
i c" 7 \
6 r > ' S
u
- ^
^
«
-
'X
\
0
^
J
„/
X
b
.'
>
Intermolecular
polyerte-polyene
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
>
122
(6)
*~**c
W
9 i—
P
\
d
"
c v c lo o le fin
f u s e d r in g
p o ly e n e
M u lti- fu s e d rin g stru c tu re
"C = C
o
r
(7)
or
c= c"^
V _ ;/ /
/ ..
\
c=c
\
d
Intermolecular polyene-polyene
d
f
i
'
Crosslinked unsaturated-saturated structure
(b) Further Reactions from Polyene Structure
Scheme 9. Proposed Cure Reactions of Phenylethynyl End-capped Group
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
123
C
C-PEPA-3,4’-OD A
Unlabeled PEPA-3,*T-ODA
3500
3000
2500
2000
Wavenumbers (cm-1)
1500
500
1000
C-PETI-5
Unlabeled PETI-5
3500
3000
2500
Wavenumbers
2000
1500
100Q
(cm
-\\
Figure 52. IR Spectra of 13C-Iabeled and Unlabeled PEPA-3,4’-ODA and PETI-5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
124
100
PEPA-3,4'-ODA
13C PEPA-3.4-ODA
80 |
E,
60 -
I
40 -
Li­
en
X
20 -
-20
100
150
200
250
300
350
400
450
Temperature (°C)
150
LaRC PETI-5
13C PETI-5
140 130 r
E 120
-
'tto 110 -
<v
X
100
-
9 0 ------80
100
150
200
250
300
350
400
450
500
Temperature (°C)
Figure 53. IR Spectra of u C-labeIed and Unlabeied PEPA-3,4’-ODA and PETI-5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
125
r*\
cn
ri
13C-PETI-5
00
00
uidd
0C
b) 13C-PEPA-3.4’-ODA
JIJJ.
a) 13C-PEPA
JL
ppm
160
ISO
—
10I—
0
Figure 54. Solution NMR Spectra of 13C-labeled a) PEPA (in DMSO-</«), b) PEPA3,4’-ODA (in NMP + DMSO-</«s), c) PETI-5 (in NMP + DMSO-r/6)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
126
I3C-enriched material was used to probe the triple bond signal decreases and follow new
bond formation during the curing process. The model compound was cured
cumulatively in the solid state according to the following schedule: 318°C - 80 min,
336°C - 80 min, 355°C - 30 min, 373°C - 30 min, 373°C - 60 min, and 380°C - 1 hr.
Stacked spectra of PEPA-3,4’-ODA, shown in Figure 56, were acquired by using gated
13
*
•
C single
pulse experiments as the curing temperature was increased
from room
temperature up to 380°C. The chemical shift assignment was based on the chemical
environment of each individual l3C resonance. These values compared favorably with
the published 13C chemical assignment of similar compounds,[4*5’ 121 and were confirmed
by our solution NMR studies (Figure 54).
To ensure that the acquisition pulse delay o f 75 s is sufficient (i.e. longer than 5 Ti)
for accurate I3C peak integration, inversion recovery spin-lattice relaxation time (Tur)
measurements were performed, and Tur values were calculated by using curve-fitting
analysis o f variable recovery time, The results showed that a 75 s acquisition delay time
is essential for both cured model compound (Ti = 9.5 s for PEPA-3,4’-ODA cured at
380°C, as shown in Figure 55) and cured imide oligomer (Ti = 10.3 s for PETI-5 cured
at 400°C).
As shown in Figure 56, in the uncured state, the 13C labeled triple bond signals were
observed in a range of 85 to 92 ppm, whereas almost no other signals appeared except
the spinning side bands o f the l3C-enriched triple bond carbons. After curing at 318°C
for 80 min, two groups o f new peaks emerged at 125 - 145 ppm and 50-65 ppm,
attributed to the newly-formed l3C=13C double bonds and saturated $ 13C-13C \ single
bonds, respectively. As the temperature increased, we observed that triple-bond signals
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
127
_Tt =9.5 s
150
xoo
ppm
50
Figure 55. Results of Ti M easurem ent through Inverse Recovery Experim ents
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
128
+ 373°C - 60 min
+ 336°C - 80 min
318°C - 80 min
Room Temp
rp n r i i | r i i i j t -i i i
1*75
150
12 5
j
tt
lOO
p p m
ri | i i i i | i i
75
50
Figure 56. Solid State 13C NMR Spectra of I3C-IabeIed PEPA-3,4’-ODA before and
after Curing (at a spinning rate of 5kHz)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
129
gradually decreased whereas signals arising from the double bonds and single bonds
increased. After 373°C for 1 hr no signal due to C=C was observed. This was confirmed
by measuring the polymer after cure at 380°C under a high spinning rate o f 10 kHz and
after 128 repetitive scans (Figure 57).
Quantitative analyses o f the NMR data acquired at a spinning rate o f 5 kHz and 16
scans for l3C-PEPA-3,4’-ODA were performed by calculating the relative percentage of
I3C peak integration as shown in Table 17. This was done by calculating the percentage
o f each bond type by integration o f the assigned absorption for each. The overlapped
spinning side bands o f the 13C-enriched ethynyl groups were subtracted for this
calculation. Spectral simulations were carried out to deconvolute the overlapping peaks
and ensure accurate peak integration. An example o f computer simulated spectrum and
its deconvulated peak regions is shown in Figure 58. As the curing reaction proceeded
from 318°C to 373°C, the peak integration values for the triple-bond signals decreased
from 91.5% to 3.1% and finally disappeared, while the integration values for
unsaturated double bonds increased from 8.5% to 87.2% and then decreased to 81.9%
after continued cure at 380°C for 2 hrs, and the aliphatic single bonds increased from
zero to 10.8 %-15.5 %. The carbon percentage for carbonyl signal remained relatively
constant during the entire curing period as expected. The percentage differences o f peak
integration for the double bonds and single bonds suggested that the major reaction o f
C s C is the formation of new
C= C double bonds, and a small fraction o f the
13C=l3C bonds reacted further to form 13C-I3C single bonds. O f special significance is
the appearance o f 1.8% aliphatic 13C-13C single bonds after cure at 318°C for 80 min
which increases to a concentration of 12.6% after 336°C for 80 min. Clearly, this shows
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
130
that the double bond 13C=13C formed from I3C=13C continues to react intermolecularly
or intramolecularly to form 13C-I3C single bonds. The possible structures that can form
single bonds are shown in reactions (3) to (7) in
Scheme 9(b). Analysis o f these
structures in terms o f the concentration of carbon-carbon single bonds and carboncarbon double bonds and on the complexity o f reaction clearly suggests that molecules
with structural features shown in reactions (3), (4) and (7) are more likely than
molecules with structural features shown in reactions (5) and (6). If the structures in
reactions (3) and (4) represented 100% of the carbon-carbon single bonded reaction
products, the percentage of double bonds and single bonds would be 66.5% and 33.5%,
respectively. Since only about 10.0% single carbon-carbon bonds are present the
molecules with these structural features must represent a small fraction of 13C present.
Therefore, the percentage difference of carbon peak integration for the double bonds
and single bonds (Table 17) suggests that the major reaction o f 13C=I3C is the formation
o f new 13C=I3C double bonds with only a small fraction of 13C=l3C reacting further to
form 13C -I3C single bonds. The formation of new aromatic structures resulting from a
trimerization reaction (reaction (2)) to form hexasubstituted benzene can not be
discounted nor verified from the l3C NMR data. The broad l3C absorption over the
range from 110 to 150 ppm prevents differentiation between aromatic double bonds and
olefinic double bonds. For example, I3C NMR of hexaphenylsubstituted benzene
(Figure 59) shows a range o f absorption from 125.4 to 140.9 ppm which fall within the
range for non-aromatic double bonds.
Finally, the results in Table 18 summarize the single pulse 13C NMR spectra
acquired at a spinning rate o f 10 kHz, which suppresses fully the spinning side bands
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
131
and are therefore quantitatively more reliable. Theoretically, based on the chemical
structure o f uncured I3C labeled PEPA-3,4’-ODA, the calculated l3C percentages o f
C=C, C=C, C = 0 are 90.8, 8.3% and 0.8% respectively, which are consistent with the
values derived from integration o f these peaks (Table 18). For both the 5 kHz and 10
kHz spinning rates, the integration error may increase as the peaks become broader and
overlapped due to increases in the high molecular weights. The long chain chemical
structures cause long relaxation times, resulting in weak signal absorption.
5.3.2 PETI-5 Studies
Solid state NMR experiments were also applied to study the curing process o f the
imide oligomer PETI-5 using an approach similar to that described for the PEPA-3,4’ODA study. The following cure processes for PETI-5 were carried out in this study:
350°C - 15 min, 360°C - 20 min, 370°C - 30 min, 370°C -1 hr, 380°C - 1 hr, 390°C - 2
hrs, 400°C - 3 hrs. Three separate samples were used for the first three schedules. One
sample was used for the study commencing at 370°C - 1 hr to a final cumulative cure at
400°C - 3 hrs. The relatively high molecular weight for uncured PETI-5 caused greater
difficulty in NMR integration due to the severe line broadening and overlapping o f
NMR peaks. Analyses o f these spectra were again based on the calculation o f the peak
integration percentages o f each component of interest. The results obtained at spinning
rates of 8 kHz and 10 kHz are summarized in Tables 19 & 20. The stacked 13C NMR
spectra o f PETI-5 oligomer at a spinning rate o f 10 kHz before cure and after cure at
400°C are shown in Figure 60. The uncured material gave signals for the C=C triple
bond at 85-92 ppm, the C=C double bond at 108-154 ppm and the C = 0 carbonyl
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
132
c=c
After Curing (b)
c=o
C-C
c=c
c=o
175
150
125
Room Temp
100
75
SO
Figure 57. Solid State 13C NM R Spectra of 13C-labeIed PEPA-3,4’-ODA
a) before and b) after Curing (at a spinning rate of 10kHz)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
133
Simulated Curve
Peak Simulation
Original Spectrum
T— I— I— |— t— t— i— I— |— I— r
IT S
IS O
13S
T
100
TS
SO
Figure 58. Simulation of the Solid State I3C NMR Spectrum of Cured 13C-IabeIed
PEPA-3,4’-ODA for Quantitative Purpose
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
134
Table 17. NMR Integration Results of I3C-labeled PEPA-3,4-ODA
at a spinning rate of 5kHz
Sample
C Carbon percentage from integration
C=C
C=C
C=0
Room Temp
91.5%
8.5%
—
318°C-80min
56.2%
42.0%
+ 336°C-80min
11.6%
72.3%
3.5%
12.6%
+ 355°C-30min
10.5%
74.1%
2.5%
12.9%
+ 373°C-30min
3.1%
83.7%
2.4%
10.8%
—
C-C
—
1.8%
+ 373°C-60min
—
87.2%
1.9%
10.9%
+ 380°C-
- -
81.9%
2.6%
15.5%
Table 18. NMR Integration Results of 13C-Iabeled PEPA-3,4-ODA
at a spinning rate of 10kHz
13C Carbon percentage from integration
Sample
C=C
C=0
90.9%
8.3%
0.8%
—
Theory: 90.8%
8.3%
0.9%
—
- -
88.2%
3.8%
C=C
Room Temp
380°C-120min
Actual:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
C-C
8.0%
135
around 165 ppm. For I3C labeled PETI-5 with a number-average weight o f 8,340
(before curing), the calculated 13C percentages o f C=C, C=C and C = 0 are 43.7%,
48.7% and 7.6%, respectively, which are very close to the NMR integration values as
shown in Table 20.
The observed cure mechanism o f PETI-5 was similar to the model compound as the
cure process proceeds over the temperature range. As given in Table 19, the
phenylethynyl triple bond decreased with the increase of cure extent and was
completely consumed after cure at 370°C for 1 hr. The ethynyl group reacted to produce
unsaturated double bonds as predicted. A weak signal for carbon-carbon single bonded
product is shown in Figure 60 and confirmed in Figure 62. However, the low
concentration o f 13C triple bonds initially prevents integration of this weak signal
quantitatively.
In order to identify the nature o f the double bond formed from the phenylethynyl
triple bond, a subtraction o f solid state 13C NMR spectra of the cured samples from
unlabeled and labeled starting material is given in Figure 61 and 62 for PEPA-3,4’ODA and PETI-5, respectively. For the model compound, NMR peak integration gives
90.8% o f carbon percentage in 13C-labled triple bond due to the 99% l3C-enriched
labeled carbons, whereas only 9.1% of carbon percentage in unlabeled triple bond.
Therefore, the double bond appearing in the NMR spectrum from unlabeled sample
should be mainly attributed to the aromatic benzenes in the starting compound, while a
majority o f double bond in the cured 13C-enriched sample should be derived from
phenylethynyl triple bond. It is o f interest to note that the main double bond peak o f
cured unlabeled model compound appears at 126 ppm, and that of the cured labeled one
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ppm
180
160
HO
ISO
100
—
I—
ao
—
i-----.-----1
-----.-----1
----->----- 1
60
>10
?0
Figure 59. NMR Spectrum of Hexaphenylbenzene in Solution of CH 2 CI2 + C D Cb
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
137
After curing
Room temp
175
150
125
100
pm
75
50
Figure 60. Solid State I3C NMR Spectra of 13C-IabeIed PETI-5
before and after Curing (at a spinning rate of 10kHz)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
138
Table 19. NMR Integration Results of I3C-IabeIed PETI-5
at a spinning rate of 8kHz
Sample
l3C Carbon percentage from integration
C=C
C=C
c=o
Room Temp
44.7%
50.5%
4.8%
350°C-I5min
42.2%
52.3%
5.5%
360°C-20min
22.5%
65.0%
12.5%
370°C-30min
3.1%
85.4%
11.5%
370°C-60min
—
86.0%
14.0%
+ 380°C -lhr
-
85.9%
14.1%
+ 390°C-2hr
--
85.9%
14.1%
+ 400cC-3hr
--
89.3%
10.7%
Table 20. NMR Integration Results of 13C-labeled PETI-5
at a spinning rate of 10kHz
Sample
1 C Carbon percentage from integration
C=C
Room Temp
400°C-3hr
C=C
C=0
Actual
42.2%
51.0%
6.8%
Theory-
43.7%
48.7%
7.6%
85.1%
14.9%
—
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
139
appeared at 136 ppm, as shown in Figure 61. This indicates that the unsaturated double
bonds formed directly from the phenylethynyl triple bonds are essentially different from
the double bonds in the aromatic ring structures o f the starting material, and therefore
could be attributed to a trimerization reaction to form hexasubstituted benzene.
A similar trend was also observed in PETI-5 samples, given in Figure 62. The signal
peak at 136 ppm increased to a shoulder in the NMR spectrum of the cured labeled
sample. The subtraction spectrum for I3C-PETI-5 is not as clearly defined as for I3CPEPA-3,4’-ODA because o f the low concentration o f I3C=13C triple bonds in this high
molecular weight l3C-PETI-5, even though the spectra were taken after 700 repetitive
scans. The subtraction spectrum was scaled up to give better observation. Carboncarbon single-bond formation observed by the weak broad absorption in the region 4570 ppm. The increase o f peak intensity at 136 ppm is much weaker. However, this is an
indication that the trimerization reaction to The increase o f peak intensity at 136 ppm is
much weaker, form substituted aromatic ring structures occurs in the PETI-5 system.
The weak and broad spectrum of PETI-5 after subtraction prevented a quantitative or
even semi-quantitative determination.
From studies in Chapter 2, the kinetics of the thermal cure o f PEPA-3,4’-ODA was
investigated by following the disappearance of the ethynyl group over the temperature
range from 318 to 373°C. This reaction followed first order kinetics as determined by
infrared spectroscopy. However, the solid state I3C NMR results demonstrated that not
only do the ethynyl triple bonds react to form carbon-carbon double bonds but also
some o f the newly-formed double bonds further react to form carbon-carbon 13C -l3C
single at the low temperatures of 3 18°C and 336°C. This suggests that the earlier kinetic
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
140
a ) 1JC-labeIed PEPA-3,4’-ODA after Curing
b) Unlabeled PEPA -3.4-O D A after Curing
Subtraction Unlabeled from Labeled, a) - b)
Ii i
ppst
r rp iT T T T T T
160
i" I I i i ■ i | i r r - ! - r
140
120
TTT
T
T'
100
Figure 61. A Comparison of Solid State !3C NMR Spectra of I3C-IabeIed and
Unlabeled PEPA-3,4’-ODA after Curing (at a spin rate of 10kHz)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
141
a) 13C-Labeled PETI-5 after Curing
b) Unlabeled PETI-5 after Curing
Subtraction Unlabeled from Labeled, a) - b)
T
T
T
■»150
1”
T
T
“T*
T
125
T
T
T
“1
100
T
T
T
T
T
T
T
Figure 62. A Comparison of Solid State 13C NMR Spectra of 13C-IabeIed and
Unlabeled PETI-5 after Curing (at a spin rate of 10kHz)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
142
results by the ER technique do not reflect the entire cure reaction mechanism but only
the first step reaction from ethynyl triple bond to double bond. The first order kinetic
data from IR study were based on starting materials, and the concentration o f the triple
bonds. Therefore, if we consider the whole cure reaction o f PEPA-3,4’-ODA including
the single bond formation, the reaction order should be > 1. The kinetics of the thermal
cure o f PETI-5, based on the increase of glass transition temperature or the decrease o f
the exothermic peak o f cure reaction^58* by DSC, gave 1.5th order kinetics for the
overall reaction. Those DSC kinetic data were derived from the final chemical
structures o f the materials after curing, and therefore, reflected the whole cure reaction
mechanism. The complicated 1.5th order kinetics suggests the more than one reaction
steps are involved in the cure reaction, even though solid state I3C NMR data cannot
provide sufficient evidence to differentiate between ene and aromatic structures or
differentiate between ring or chain 13C-13C single bonded structures in cured PETI-5
resin, which could help in the interpretation of the kinetic studies. This limitation in
following the full cure reaction through the disappearance o f the triple bond and in
determining structural features by solid state 13C NMR cure studies becomes clear from
this study.
5.4 CONCLUSIONS
Solid-state I3C nuclear magnetic resonance (NMR) was performed to determine the
crosslinked structures o f phenylethynyl end-capped polyimides, as a continuous
investigation to Chapter 2. For PEPA-3,4’-ODA, the NMR results revealed that the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
143
major reaction was the ethynyl to ethynyl reaction to form double bonds, and a minor
reaction of double bond to double bond or ethynyl to conjugated double bond to further
form single bonds was also observed. In the PETI-5 study, equivocal evidence for
formation of the single bonded structures was also observed as a result o f the cure
reactions. The major curing reaction for the phenylethynyl end-capped PETI-5 oligomer
is ethynyl to ethynyl to produce chain extension or polyene structures. The
hexasubstituted aromatic ring structure could also be formed from PEPA-3,4’-ODA
model compound or PETI-5. However, the broad absorption peaks in the region o f 110
-1 5 5 ppm prevented their identity and integration for quantitative evaluation.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
REFERENCES
1) Lee, I.L.; Springer, G.S., J. Compos. Mater., 18, 387(1984).
2) Agrawal, R.K.; Drzal, L.T., J. Adhesion, 29,63(1989).
3) Lewis, D.A.; Herdrick, J.C.; Ward, T.C.; McGrath, J.E., Polym. Prep., 28,
330(1987).
4) Chen, M.; Siochi, E.J.; Ward, T.C.; McGrath, J.E., Polym. Eng. Sci., 33,
1092(1993).
5) Lewis, D.A., Mater. Res. Soc. Symp. Proc., 269, 21(1992).
6) Wei, J.; DeLong, J.D.; DeMeuse, M. and Hawley, M.C., Polym. Eng. Sci., 33,
1132(1993).
7) Clark, D.E. and Sutton, W.H., Anna. Rev. Mater. Sci., 26, 299(1996).
8) Committee on Microwave Processing of Materials et al., in Microwave Processing
o f Materials, National Academy Press, Washington, D.C., 1994.
9) Sutton, W.H., Ceram. Bull., 68, 376 (1989).
10)Thuillier, F.M. and Jullien, H., Makromol. Chem., Macromol. Symp., 25, 63(1989).
11)Mijovic, J., Fishbain, A. and Wijava, J., Macromolecules, 25, 986(1992).
12) Abramovich, R.A., Org. Prep. Proc. Int., 23(6), 683(1991).
13) Cablewski, T.; Faux, A.F.; Strauss, C.R., J. Org. Chem., 59, 3408(1994).
14)Majetich, G.; Hicks, R., J. Microwave Power Electromagn. Energy, 30, 27(1995).
15) Jullien, H.; Valot, H., Polymer, 26, 505(1985).
16) Lewis, D.A.; Nunes, T.L.; Simonyi, E., Polym. Mater. Sci. Eng., 72, 40(1995).
17)Chia, H.L.; Jacob, J.; Boey, F.Y.C., J. Polym. Sci. Part A, 34, 2087(1996).
144
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
145
18) Platts, J., in Microwave Ovens, Peter Pereginus Ltd., London, 1991, page 81.
19) Bible, D.W.; Lauf, R.J., US Patent 5 321 222, 1994.
20) Wei, J.B.; Shaidaker, T.; Hawley, M.C., Trends in Polym. Sci., 4, 18(1996)
21)Fathi, Z.; Tucker, D.A.; Lewis, W.A.; Wei, J.B., Mater. Res. Soc. Symp. Proc., 430,
21(1996).
22)DeMeuse, M.T.; Parodi, F.; Gerbelli, R.; Johnson, A.C., 39th Intl. SAMPE Symp.,
13(1994).
23) Janney, M.A.; Kimrey, H.D.; Schmidt, A.S; Kigans, J.O., J. Am. Ceram. Soc., 74,
1675(1991).
24) Wilson, D.; Stenzenberger, H.D.; Hergenrother, P.M., in Polyimides, Blackie &
Son Ltd, London, 1990.
25)Scola, D.A., Polym. Eng. & Sci., 31, 6(1991).
26)Bryant, R.G; Jensen, B.J; Hergenrother, P.M., Polym. Prep., 33, 910(1992).
27)Hergenrother, P.M.; Smith Jr, J.G., Polymer, 35, 4857(1994).
28) Bryant, R.G; Jensen, B.J; Hergenrother, P.M., J. Appl. Polym. Sci., 59, 1249(1996).
29) Hergenrother, P.M.; Bryant, R.G.; Jensen, B.J.; Havens, S.J., J. Polym. Sci. Part A:
Polym. Chem., 32, 3061(1994).
30) Hergenrother, P.M. Trends in Polym. Sci. 4, 104(1996).
31) Odian, G., Chap 7 in Principles o f Polymerization, John Wiley & Sons, New York,
1991.
32)Kohan, M.I., in Nylon Plastics Handbook, Hanser/Gardner Publications, New York,
1995.
33)Brode, G.L.; Koleske, J.V., J. Macromol. Sci. Chem., A6, 1109(1972).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
146
34)Koleske, J.V.; Lundberg, RJD., J. Polym. Sci., PartA-2, 1, 795(1969).
35)Dumitriu, S., in Polymeric Biomaterials, Marcel Dekker, Inc., New York, 1994.
36) Pitt, C.G., in Biodegradable Polymers as Drag Delivery Systems, Chasin, M.;
Langer, R.; Eds, Marcel Dekker, Inc., New York, 1990, pp 71-120.
37) Gonsalves, K.E.; Mungara, P.M., Trends in Polym. Sci., 4, 25(1996).
38) Goodman, I.; Vachon, R.N., Ear. Polym. J., 20, 529(1984).
39) Gonsalves, K.E.; Chen, X.; Cameron, J.A., Macromolecules, 25, 3309(1992).
40) Ellis, T.S.; J. Polym. Sci., Part B: Polym. Phys., 31, 1109(1993).
41) Chen, X.; Gonsalves, K.E.; Cameron, J.A. J. Appl. Polym. Sci., 50, 1999(1993).
42) Paul, C.W.; Schulze, R. A.; Fenelli, S.P., Advances in Polyimide Science &
Technology (Eds. C. Feger, M.M. Khojasteh, and M.S. Htoo.), Technomic
Publishers, Lancaster, PA., 1991, pp. 220-244.
43) Jensen, B.J.; Hergenrother, P.M.; Nwokogu, G., Polymer, 34 , 630(1993).
44) Meyer, G.W.; Jayaraman, S.; McGrath, I.E., ACS Polym. Prepr., 34, 540(1993).
45) Bryant, R.G.; Jensen, B.J.; Hergenrother, P.M., ACS Polym. Prepr., 34, 566(1993).
46) Smith, Jr., J.G.; Hergenrother, P.M., ACS Polym. Prepr., 35, 353(1994).
47) Jayaraman, S.; Srinivasan, R.; McGrath, J.E.; J. Polym. Sci. Part A: Polym.
Chem., 33, 1551(1995).
48)Kovar, R.F.; Ehlers, G.F.L.; Arnold, F.E., J. Polym. Sci, Polymer Chem. Ed. 15,
1081(1977).
49) Pickard, J. M.; Jones, E.G.; Goldfarb, I.J., J. Macromolecules 12, 895(1979).
50)Ratto, J.J.; Dynes, P.J.; Hamermesh, C.L., J. Polym. Chem. Ed., 18, 1035(1980).
51) Douglas, W.E.; Overend, A.S.; Eur. Polym. J., 29,1513(1993).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
147
52) Sefcik, M.D. ; Stejskal, E.O.; McKay, R.A.; Schaefer, J., Macromolecules, 12,
423(1979).
53) Swanson, S.A.; Fleming, W.W.; Hofer, D.C., Macromolecules, 25, 582(1992).
54) Sastri, S.B.; Keller, T.M.; Jones, K.M.; Armistead, J.P., Macromolecules, 26,
6171(1993).
55) Jones, K.M.; Keller, T.M., ACS Polym. Mater. Sci. Eng. 68,97(1993).
56) Johnston, J.A.; Li, F.M.; Harris, F.W.; Takekoshi, T., Polymer, 35, 4865(1994).
57)Takekoshi, T.; Terry, J.M.; Polymer, 35, 4874(1994).
58)Hinkley, J.A., J. Advanced Materials, 27, 55(1996).
59)DiBenedetto, A.T., J. Polym. Sci., Part B: Polym. Phys., 25, 1949(1987).
60) Hale, A.; Macosko, C.W.; Bair, H.E., Macromolecules, 24, 2610(1991).
61)Pascault, J.P.; Williams, R.J.J., J. Polym. Sci., Part B: Polym. Phys., 28, 85(1990).
62) Nielsen, L.E., J. Macromol. Sci., Revs. Macromol. Chem., C3, 69(1969).
63) Venditti, R.A.; Gillham, J.K., J. Appl. Polym. Sci., 64, 3(1997).
64)Couchman, P R., Macromolecides, 20, 1712(1987).
65) Flynn, J.H.; Wall, L.A., J. Res. NBS, _70A(6), 487(1966).
66) Ozawa, T., Bull. Chem. Soc. Jpn, 38, 1881(1965).
67) Ozawa, T.; J. Therm. Anal. 2,301(1970).
68) Lewis, D.A., 39th Intl. SAMPE Symp., 1(1994).
69) Ward, T.C.; Chen, M., Polym. Mater. Sci. Eng., 66, 335(1992).
70) Jordan, C.; Galy, J.; Pascault, J.; More, C.; Delmotte, M.; Jullien, H., Polym. Eng.
Sci., 35, 233(1995).
71)Hou, T.H.; Jensen, B.J.; Hergenrother, P.M., J. Compos. Mater., 30, 109(1996).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
148
72) Connell, J.W.; Smith J.G.; Hergenrother, P.M.; Rommel, M., 28th International
SAMPE Technical Conference, 14(1996).
73) Yue, C.Y.; Looi, H.C., Composites, 26,767(1995).
74) Wei, J.; Hawley, M.C., 39thIntl. SAMPE Symp., p 45(1994).
75) Reimschuessel, H.K., Chap. 7 in Ring-Opening Polymerization, Frisch, K.C.;
Reegen, S.L.; Eds., Marcel Dekker, Inc., New York, 1969.
76) Reimschuessel, H.K., J. Polym. Sci. Macromol. Rev., 12, 65(1972).
77) Hermans, P.H.; Heikens, D.; van Velden, P.F., J. Polym. Sci., 30, 81(1958).
78) Hermans, P.H.; Heikens, D.; van Velden, P.F., J. Polym. Sci., 44, 437(1960).
79) Young, R.H.; Matzner, M,; Pilato, L. A., Chap. 11 in Ring-Opening Polymerization,
Saegusa, T.; Goethals, E.; Eds., American Chemical Society, Washington, D.C.,
1977.
80)Lundberg, R.D.; Cox, E.F., Chap. 6 in Ring-Opening Polymerization, Frisch, K.C.;
Reegen, S.L.; Eds., Marcel Dekker, Inc., New York, 1969.
81) Bigdelli, E.; Lenz, R . W Macromolecules, 11, 493(1978).
82) Slomkowski, S.; Penczek, S., Macromolecules, 13, 229(1980).
83) Slomkowski, S.; Penczek, S., Macromolecules, 9, 367(1976).
84) Pitts, C.G.; Chasalow, F.I.; Hibionada, Y.M.; Kilmas, D.M.; Schindler, A., J. Appl.
Polym. Sci., 26, 3779(1981).
85) Pitts, C.G.; Schindler, A., in Biodegradation o f Polymers in Controlled Drug
Delivery, Vol. 1, Bruck, S.D.; Eds, CRC Press, Boca Raton, 1983, pp53-80.
86) Lewis, D.H., Chap. 1 in Biodegradable Polymers as Drug Delivery Systems, Chasin,
M.; Langer, R.; Eds., Marcel Dekker, Inc., New York, 1990.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
149
87) Vert, M.; Li, S.M.; Garreau, H., J. Controlled Rel., 16, 15(1991).
88)Gebelein, C.G.; Carraher, C.E., Eds., Biosactive Polymeric Systems: An Overview;
Pelnum Press: New York, 1985.
89) Field, R.D.; Rodriguez, F.; Finn, R.K., J. Appl. Polym. Sci., 18, 3571(1974).
90)Mathisen, T.; Lewis, M.; Albertason, A.C., J. Appl. Polym. Sci., 42, 2365(1991).
91)Helder, J.; Diijkstra, D.J.; Feijen, J., J. Biomed. Mater. Res., 24, 1005(1990).
92) Barrows, T.H.; Johnson, J.D.; Gilbson, S.J.; Grussing, D.M., in Polymers in
Medicine II, Chiellini E., et al, Eds., Plenum Press: New York, 1986, p85.
9 3 )Horton, V.L.; Blegen, P.E.; Barrows, T.H., Quarfoth, G.L.; Gibson, S.J.; Johnson,
J.D.; McQuinn, R.L., In Progress in Biomedical Polymers; Gebelein, C.G., Dunn,
R.L., Eds; Plenum Press: New York, 1990, p263.
94) Goodman, I.; Valavanidis, A., Eur. Polym. J., 20, 241(1984).
95) Goodman, I.; Vachon, R.N., Eur. Polym. J., 20, 539(1984).
96) Goodman, I., Eur. Polym. J., 20, 549(1984).
97)Nakayama, A.; Higashi, T.; Yalcayama, A.; Iyoda, J.; Ukita, M.; Hayashi, K.;
Yamamoto, N., Chem. Express, 8, 181(1993).
98)Haslam, J.; Willis, H.A; Squirrell, D.C.M., in Identification and Analysis o f
Plastics, Butterworth& Co., Ltd., London, 1972. pp314-315.
99) Sperling, L.H., in Introduction to Physical Polymer Science, John Wiley & Sons,
New York, 1992. pp 100-108.
100)
Mattiussi, A.; Gechele, G.B.; Francesconi, R., J. Polym. Sci., A2. 7, 411(1969).
101)
Fox, T.G., Bull. Am. Phys. Soc., 1, 123(1956).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
150
102)
Ku, C.C.; Liepins, R., in Electrical Properties o f Polymers, Chemical
Principles, Hanser Publisher: New York, 1987, p41.
103)
Stejskal, E.O.; Memory, J.D., Chap. II and IV in High Resolution NMR in the
Solid State: Fundamental o f CP/MAS, Oxford University Press, New York, 1994.
104)
McBrierty, V.J.; Packer, K.J., Chap. 4 in Nuclear Magnetic Resonance in Solid
Polymers, Cambridge University Press, New York, 1993.
105)
Wehrli, F.W.; Wirthlin, T., in Interpretation o f Carbon-13 NMR Spectra,
Heyden & Son Ltd., London, 1978.
106)
Albert, K.; Schmid, J.; Pfleiderer, B.; Bayer, E., in Chemically Modified
Surfaces, Elsevier Science Publishers, 1992, ppl05-117.
107)
Baugher, A.H.; Espe, M.P.; Goetz, J.M.; Schaefer, J.; Pater, R.H.,
Macromolecules, 30, 6295(1997).
108)
Meador, M.A.B.; Johnston, J.C.; Cavano, P . Macromolecules, 30, 515(1997).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Документ
Категория
Без категории
Просмотров
0
Размер файла
5 661 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа