".55..
7 Jr ‘

tin.

           

   

Wm.» .
‘ 5“an
u... 4...:

‘ A . é
3;. .....

            

 

 

 

 

 

   

                     

barb:

I-

A-

 

,_

5" ~.

 

’yo'.

    

ha}.
.Cifl .

 

 

‘ . Hump , .
Jufiarra 3;

.uflqyu‘ .

            

.
uzmnsfifl J;
.. flmfiéuwn
3mg ,

 

 

1 Lytolwu.”
H .. H..‘o-.‘nll

3 r1, film»?

                                     

 

‘ :Fvlgmulubfi

 

  

:TIL Ir; .
0..)

A 3...
‘ 1511.3
{Furry

.n; is?!
. 2.32
L .

    

      

.. : . a}: v {I .

. ‘ . n , .. y . . . , ‘ ‘ .5... 3.. I. 1 bahfli‘rfpfl. 4H”;
. . . , 4 . . . . . w . . y. . .. . . y . . . 9...! r In“. 1X»nu..:lc."

. ... . , ‘ , . < . 1. 1141'. {n}.

n: &

  

:, .. Iv -1...“ Lat

    

 

.n b 11 . .I.‘
v.,:,|..lpl.:lr.

 

        
 

             

  

       
  

 

: t I} J A :1!
. ...f.r..é..4.k .AII
.. . . . ‘ H 1.43 f..:.:.?.:..
‘ t v | I . I . ‘l of".
. u ‘ I! ‘ .
, y . . . . ‘ | . ‘ . .
\ V L Q } I -- o -
.o o u 4 fi 1 V § I
. . .. - ‘ ‘
.- u ‘ Y I . ‘ .
L ‘. v u . I t n
r . .
C -. .32 flat:
‘ , . 1 ‘I On” ; f . 5! out I? .llttdzi...r.fldz anti. . 161:}:
. . . ¢ 1 1. J .
.

 
  

I. I .Ian .-f v! I.
I ‘ . 3 .. é .lvk.,|4:brr.oolh3..1 3..
I ‘ A I T ‘ .

luv uni. ."u.

 

3.1.. cf”... 3.1:...w.,.m.fi.fi.#.3fl..Ania». $51.11. . . . .aruinr. .Tafi .5214 i ii 5:, ii .

 

 

THESE;

l
(9'13 0

Jllllllllllll'lilll'filllllllilllllllllllllllflllllllllllll ”

3 1293 01834 5995

 

LIBRARY

Michigan State
Unlvaralty

 

 

This is to certify that the

thesis entitled

THERMO-OXIDATION AND HYGROTHERMAL EFFECTS
ON CYANATE ESTER POLYMER MATERIALS

presented by

Zitao Liu

has been accepted towards fulfillment
of the requirements for

_Mast.e.n__ degree in .mma-Ls—Science

  

Major professor

0-7639 MS U is an Afirmative Action/Equal Opportunity Institution

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1M C/CIHCIDItoDmpGS-p.“

 

THERMO-OXIDATION AND HYGROTHERMAL EFFECTS
ON CYANATE ESTER POLYMER MATERIALS

By

Zitao Liu
A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Materials Science and Mechanics

1 999

ABSTRACT

THERMO-OXIDATION AND HYGROTHERMAL EFFECT ON
CYAN ATE ESTER POLYMER MATERIALS

By
Zitao Liu

The aim of this research is to study thermo—oxidation and hygrothermal effects on
cyanate based polymer materials. Cyanater ester (CE) and siloxane-modified cyanate
ester (SMCE) and their graphite composites were investigated. Gravimetric experiment,
differential scanning calorimetry (DSC), Fourier-transform infrared spectros00py (FTIR),
three-point bend tests were employed to study the mechanical and physical mechanisms
of thermo—oxidation on cyanate based polymer materials. Water absorption experiments
and three-point bend tests were used to characterize the water diffusion and hygrothermal
degradation of cyanate based poylmer materials.

Thenno-oxidation experiments showed that SMCE gains weight when aged in
the air at 100°C but shows virtually no weight gain when aged in argon at 100°C. CE
resins show no weight gain when aged in the air or argon at 100°C.

Differential scanning calorimetry (DSC) tests show that glass transition
temperature of SMCE decreases about 35°C afier aged in the air at 100°C for 3.4x10° sec
(~ 40 days). Three-point bend tests indicated that the flexure strength of SMCE
decreased by 20% after aging in the air for 7.63 x10° sec (~ 3 months). FITR results
provide information on the bonding character and chemical structure of CE & SMCE

materials in this research. Results suggest SMCE degrades at 100°C in the air due to the

thermo-oxidation process. The degree of thermo-oxidation of the polymer resin depends
strongly on the supply of oxygen from the environment and the diffusion of oxygen in
SMCE resin.

Hygrothermal tests show water diffusion in cyanate ester based polymer and its
composites is non-Fickian type. FTIR spectroscopy shows that water molecules diffusing
into the polymer structure interact chemically with the cyanate ester polymer network.
The interaction causes some water molecules to be trapped in the polymer network, thus
the water absorption curves deviate from F ickian behavior. Calculation of activation
energy supports the contention that water diffusion occurs mainly through the polymer
matrix in the composite materials.

Three-point bend tests show that the flexure strength of CE and SMCE decreases
with increasing aging temperature. The flexure strength of CE based composites
decreases significantly only at high temperature. Results also show the surface
degradation of SMCE and CE based composite materials and delamination of SMCE

based composite materials after aging in 95°C distilled water for more than 2000 hours.

TO MY PARENTS

ACKNOWLEDGEMENTS

I would like to express sincere thanks to the following people. First, I would like
to thank my thesis advisor Dr. James P. Lucas. His guidance, help, fiiendship and
suggestions during my graduate work and research are greatly appreciated. It has been a
great privilege to work with him.

My special thanks go to Dr. Andre Y. Lee and Dr. Melissa J. Crimp for being in
my graduate committee. Thank you also to thank my colleagues Guo F u and Alan
Gibson for their contributions, guidance, and friendship.

I also acknowledge the Composite Materials and Structure Center for the use of
much of the equipment necessary for conducting my research.

Finally, I thank my parents and my sister and brother-in-law for their support and

encouragement.

TABLE OF CONTENTS

LIST OF TABLES ................................................................................ viii
LIST OF FIGURES ................................................................................ ix
LIST OF ABBREVIATIONS AND SYMBOLS ............................................. xi
CHAPTER I
INTRODUCTION ..................................................................................... 1
CHAPTER [1
LITERATURE REVIEW ............................................................................ 5
2.1 Cyanate Ester Based Polymer Materials ........................................... 5
2.2 Thermo-oxidation of Polymer Materials ........................................... 7
2.2.1 Thermo-oxidation Mechanism ............................................. 8
2.2.2 Chemical Change of Polymers in the Oxidation Process .............. l 1

2.2.3 Effect of Chemical and Physical Structure on Oxidation .............12
2.3 Hygrothemal Effect on Polymer and Composite Materials 13

2.3.1 Water Absorption Kinetics .................................................. 13
2.3.2 Plasticization and TS Variation ........................................... 16
2.3.3 Effect on Mechanical Properties ........................................... 18
2.3.4 Hygrothermal Degradation of Polymer and Composite Materials...20
2.3.5 Hygrothemal Effect on Interphase Phase ................................ 22
CHAPTER III
THERMO-OXIDATION OF CYANATE BASED POLYMER ......................... 24
3.1 Experimental ......................................................................... 24
3.1.1 Materials ................................................................... 24
3.1.2 Specimen Preparation ..................................................... 26
3.1.3 Gravimetric Experiments .................................................. 26
3.1.4 Glass Transition Temperature Measurement ........................... 26

vi

3.1.5 Three-point Bend Test ..................................................... 27

3.1.6 FTIR Test ...................................................................... 28
3.2 Results and Discussion .............................................................. 28
3.2.1 Assessment of Thermo-oxidation of CE & SMCE by Gravimetric
Experiments ................................................................. 28
3.2.2 Degradation of Physical and Chemical Properties of CE & SMCE
Materials ...................................................................... 36
3.2.3 Thermo-oxidation Mechanism of SMCE ................................. 41
3.3 Summary/Conclusions ................................................................. 49
CHAPTER IV
HYGROTHERMAL EFFECT ON CYANATE ESTER BASED MATERIALS....50
4.1 Experimental ........................................................................ 50
4.1.1 Materials .................................................................... 50
4.1.2 Hygrothermal Exposure ................................................... 52
4.1.3 Three-point Bend Test ..................................................... 53
4.2 Water Absorption Results and Discussion ....................................... 54
4.2.1 Water-absorption Kinetics of Cyanate Based Polymer Materials....54
4.2.2 Diffusivity in Cyanate Ester Based Polymer and Composites ....... 58
4.2.3 Activation Energy of Water Diffusion in Cyanate Based Polymer
Materials ...................................................................... 65
4.3 Hygrothermal Degradation of Cyanate Based Polymer Materials ......... 68
4.4 Three-point Bend Tests on Hygrothermally-exposed Materials ............ 72
4.5 Summary/Conclusions ................................................................ 77
CHAPTER V
CONCLUSIONS AND RECOMMENDATIONS ............................................ 78
5.1 Conclusions ............................................................................ 78
5.2 Recommendations ................................................................... 79
BIBLIOGRAPHY ................................................................................. 80

vii

Table 3.1

Table 3.2

Table 3.3

Table 3.4

Table 4.1
Table 4.2
Table 4.1
Table 4.2
Table 4.3

Table 4.4

LIST OF TABLES

Nominal physical properties of CE at room temperature (p. 25)
Nominal physical properties of SMCE at room temperature (p. 25)

Glass transition temperature of CE aged in different environment (argon
and air) (p. 36)

Glass transition temperature of SMCE aged in different environment
(argon and air) (p. 36)

Nominal physical properties of CE composite (p. 51)

Nominal physical properties of SMCE composite (p. 51)

Diffusivity of cyanate ester resin (p. 64)

Diffusivity of cyanate ester composite (p. 64)

Diffusivity of siloxane-modifred cyanate resin (p. 64)

Diffusivity of siloxane-modified cyanate composite (p. 64)

viii

Figure 2.1

Figure 3.1

Figure 3.2

Figure 3.3

Figure 3.4

Figure 3.5

Figure 3.6

Figure 4.1

Figure 4.2

Figure 4.3

LIST OF FIGURES

Formation of cyanate ester network from dicyanate monomer (p. 6)

Weight change versus exposure time profiles is shown for cyanate ester
and siloxane modified cyanate ester resin materials aged in the air and in
the Argon at 100 °C. (p. 29)

Weight change of SMCE aged in the air then aged in the argon at 100°C
(experiment A) (p. 31)

Weight change of SMCE aged in the argon first and then aged in the air
and aged in the argon again at 100°C (experiment B) (p. 32)

Weight change of siloxane-modified cyanate ester aged in the air at 100°C

(p. 35)

Three-point bend test of cyanate based polymer material (aged and
nonaged) (p. 40)

FTIR spectra (1660 cm" — 2380 cm") of SMCE aged in the air. (a)
Change in the spectra with aging time shows the increase of the 1726 cm'

peak (p. 42) (b) Change in the spectra with aging time shows the decrease
of 911 cm' and 966 cm'l peaks (p. 43)

I

FTIR spectra of cyanate ester resin aged in the water at 95°C for 3000hrs.
The spectrum shows the increasing of 3360-3340 cm”1 (N -H group) and
1745 cm" (C=O group) (p. 56)

The reaction of cyanate molecule with water forming carbamate (p. 57)

Water absorption curves of siloxane-modified cyanate at different
environmental temperatures with square root of time (p. 59)

ix

Figure 4.4

Figure 4.5

Figure 4.6

Figure 4.7

Figure 4.8

Figure 4.9

Figure 4.10

Figure 4.11

Figure 4.12

Figure 4.13

Water absorption curves of siloxane-modified cyanate fiber composite at
different environmental temperatures with square root of time (p. 60)

Water absorption curves of cyanate ester at different environmental
temperatures with square root of time (p. 61)

Water absorption curves of cyanate ester fiber composite ester at different
environmental temperatures with square root of time (p. 62)

The activation energy of water diffusion for cyanate ester based polymer
and composite materials (p. 66)

The activation energy of water diffiision for siloxane-modified cyanate
resin based polymer and composite materials (p. 67)

Surface morphology of aged siloxane-modified cyanate @95°C for 3000b
(a) as — received (b) sample aged in water (p. 70)

Cross-section of aged siloxane-modified cyanate composite @95°C for
3000b (a) as-received (b) sample aged in water (p. 71)

Surface morphology of aged cyanate ester composite@95°C for 3000hrs
(a) as - received (b) sample aged in water (p. 72)
The results of the three-point bending test of the cyanate ester and

siloxane-modified cyanate (p. 74)

The results of the three-point bending test of the cyanate ester composites
and siloxane-modified cyanate ester composites. (p. 75)

BMI

C.F.lPEEK

C.F.lPSF

CE

De

DFT

DMTA

DSC

ESEM

FTIR

Gr/Ep

LIST OF ABBREVIATIONS AND SYMBOLS

Bismaleimide

Carbon fiber/polyetheretherketone composites

Carbon fiber/polysulfone composites

Cyanate ester

Diffusion coefficients of composites

Delamination fracture toughness

Dynamic mechanical thermal analysis

Diffusion coefficients of resin

Differential scanning calorimetry

Potential energy

Environmental scanning electron microscopy

Fourier-transform infrared spectroscopy

Graphite/Epoxy composite

Moisture concentration

Initial moisture content

xi

Mr

PEEK

PMMA

PPS

SEM

SMCE

UV

w%

Wo

Equilibrium moisture content

Moisture content at time t

Poly-ether-ether-ketone

Poly(methyl methacrylate)

Polyphenylene sulfide

Activation energy

Scanning electron microscopy

Siloxane-modified cyanate ester

Glass transition temperature

Ultraviolet

Weight percentage change

Original weight of samples

Weight of aged samples

xii

Chapter 1

INTRODUCTION

Polymer composite materials [1-7], especially epoxy resin matrix composites,
have been used for decades in the electronics, aerospace, sporting goods and automotive
industries due to their high strength, excellent dielectric properties, high dimensional
stability and low specific weight. One of the major concerns of composite materials [4-9]
in industry and commercial applications has been their long-term durability under
environmental conditions such as moisture, temperature, oxygen and radiation. These
effects, particularly the effects of moisture absorption and oxidation on polymer and

polymer composite materials, have been extensively studied since the early 19703 [4—12].

Most investigations [4-12] have shown that epoxy based composite materials
absorb significant amount of water which often induces the degradation and dimension
instability of polymer materials. Such changes tend to limit their application in high-
technology industries which usually demands high environmental stability standards of
materials. Intensive research [13-14] has been performed in effort to provide new
thermosetting resins with a reduced tendency of moisture absorption without penalizing
the processibility. Low moisture absorbing materials were achieved by the development

of cyanate ester resin based materials [15-17].

Cyanate ester resin matrix composites reinforced by glass or aramid fibers

established their presence in the application for multi-layer electric circuit boards as early

as the late 19708 [15]. The use of cyanate ester resins was pursued for circuit board
applications for the following reasons: (1) The glass transition temperature of cyanate
ester exceeds those of epoxy resins and match molten solder temperature (220°C-270°C);
(2) low dielectric loss properties; (3) excellent (epoxy like) processibility; (4) low
moisture absorption; (5) excellent adhesion to metals.

The significant reduction in water absorption of cyanate ester (CE) allows for
much improved dimensional stability than epoxy based polymer materials. Improved
dimensional stability makes CE very promising for applications in aerospace industry.
Recently, successful efforts in toughening the cyanate ester resin by thermoplastic
modifier such as siloxane-modified cyanate ester (SMCE) has led to a serious
consideration of cyanate ester resin matrix composites for use in primary structure
applications [14-1 6].

There have been several studies of thermo-oxidation effect on the durability of
thermoplastic-toughened cyanate ester resin (Fiberite 954-2) and its composites by
Paravatareddy et al. [17,18]. The samples were aged for periods of up to 9 months at
150°C and in one of three different gas environments: nitrogen, low pressure air and
atmospheric ambient air. Results have shown the glass transition temperature of cyanate
ester was affected substantially by the aging time and the environments, and the bending
strength of the materials decreases by 30-40% over 6—month period at 150°C. But the
thermo-oxidation mechanism of cyanate ester materials is still not clear. There has been
no report on the thermo-oxidation effect on siloxane-modified cyanate ester materials.

There have been many studies on the hygrothermal effect on the cyanate ester

based polymers materials [19-21]. Results show that the water absorption decreases the

glass transition temperature and mechanical strength of cyanate based polymer materials.
But the mechanism of water diffusion in cyanate based material is still not clear and some
conflicting results have been observed. Cinquin et al. [19] studied water absorption of
cyanate ester materials and found the humidity absorption of cyanate ester follows
Langmuir's law. Blair et a1. [20] studied the moisture absorption of cyanate ester resin
laminate. Their results showed that the moisture absorption curves of these materials
follow Fick's second law of diffusion. Lee et a1. [21] studied the water absorption of
graphite fiber-reinforced cyanate ester resin composites. Their results showed these
materials absorbed a remarkably small amount of moisture when exposure to high
humidity and the moisture absorption curves follow Fick's law in the early stage of water
absorption. They also observed that the degree of moisture absorption underwent a
sudden increase to a new equilibrium level and the water absorption became non-Fickian
after prolonged exposure to high humidity.

This research plans to investigate the thermo-oxidation effects and hygrothermal
effects on CE and SMCE and their composites using gravimetric, analytical, and
spectrographic techniques. The weight change of CE and SMCE has been measured
subsequent to exposure in air and water environments at different temperatures. FTIR
was used to monitor the chemical structure change during aging (thermo-oxidation)
process. Glass transition temperature (Tg) of these materials was measured to study the
extent of degradation of cyanate ester networks. Three-point bend tests were used to
study environmental effects on mechanical strength of these materials. Further research
will assess environmental effects on the polymer and composite interface. When

completed, the efforts will clarify the mechanism of thermo-oxidation on cyanate ester

based materials and hygrothermal environment on cyanate ester based polymer and

composite materials.

Chapter II

LITERATURE REVIEW

2.1 Cyanate Ester Based Polymer Materials

Cyanate ester resin, also known as cyanate ester, cyanic esters, or trizine resins,
features the polymerizable functional group —O~CEN on an aromatic backbone.

Like epoxies, these thermosetting resins are derived from bisphenols or polyphenols, and
are available as monomers. All commercial dicyanate monomers fit the model

compound structure illustrated below.

CH3

NEC—O—@—CIIT—@—O—CEN

CH3

When heated, cyanide functionality undergoes cyclomerization to form
symmetrically substituted trizine structures. This ring-forming addition polymerization
results in a therrnoset network of oxygen-linked trizine rings and bisphenol esters. Cured
cyanate resins are classified as polycyanurates. The cyclotrimerization curing process is

illustrated in Fig. 2.1 [22]

(T33
NEc—o—©—<l:—-@—o—CEN

CH3

Dicyanate monomer

C\ 1 heat é/N

1 \Tmm
<3

CH3— —CH3
C
m
N
Prepolymer resin

Fig. 2.1 Formation of cyanate ester network from dicyanate monomers

The cyanate ester network has many interesting properties [23], such as high glass
transition temperature (250-300°C), high toughness, low moisture absorption, good
thermal stability, low dielectric constants, and excellent adhesion to metals. This
combination of properties is unique in 250°C Tg resins and commends its use in high
performance, technological applications, i.e., aerospace and electronic industries.

Polysiloxanes [24] as shown below are molecules with long Si-O-Si chains, side
hydrocarbon groups and end functional groups. The chemical structure is shown as

following:

CHzR CH2R

X Si—O —Si X
I n
CHZR CH2R

X: -CN R: H n:10—1000

Owing to the long Si-O-Si chains, polysiloxanes have a number of unique
properties including low temperature flexibility, hydrophobicity, and ultraviolet and
radiation resistance. It has been reported [24] siloxane-modified cyanate ester has better
properties, such as high toughness, low water absorption and oxygen plasma stability,

than the cyanate ester resin.

2.2 Thermo-Oxidation of Polymer Materials
Although all polymers degrade at high temperatures in a non-oxygen
environment, degradation of polymers is much faster in the presence of oxygen at

elevated temperatures. This phenomenon is often referred to as thermo-oxidation of

polymer materials [25-28]. The initiation of thermo-oxidation of polymer materials at
elevated temperature is mainly due to the free radicals generated by thermolysis of
polymer molecules [25-27]. The free radicals generated react rapidly with oxygen
supplied by the environment. Therrno-oxidation reaction of polymer materials usually
causes the degradation of polymers, i.e., lower tensile strength, lower glass transition
temperature and lower dimensional stability. For high-temperature engineering polymer
materials, thermo-oxidation degradation of these materials has been major concern for

their practical application in high temperature environments.

2.2.1 Thermo-Oxidation Mechanism

In polymers, the oxidation propagates through a free-radial chain mechanism.
The difference of thermo-oxidation from other chain reactions is that, besides the usual
steps, i.e., initiation, propagation and termination, two additional important steps must be
considered: (1) conversion of the formed hydrocarbon radials to peroxy radicals and (2)
degenerate chain branching. A scheme including the most important steps in polymer
oxidation is shown as follows [28]:
(I) Initiation

PH -> P' + H' PH is a carbon-hydrogen bond , P' and H' are free radicals

(11) Radical conversion
P' + 02 —) P02' P02' is peroxide radicals
(III) Chain propagation
P02'+ PH —) POOH + P' POOH is peroxide hydrogen bond

(IV) Degenerate chain branching

POOH —> a R (IV.])
POOH+ POOH—> BP' (IV.2)

a P' and BP' are degenerated radicals

(V) Termination

P“ + P' —~> inactive product (V.l)
P' + P02° —> inactive product (V.2)
P02' + POz' —-) inactive product (V3)

(1) Initiation

The free radicals may be initialized by several external stimuli, such as
1ight(photo-oxidation) and heat (thermo-oxidation), and will react most readily with
molecular oxygen. Initiation by direct reaction of molecular oxygen with the polymer
material and in step I has been observed during thermo-oxidation of polymers [28-30].
This reaction is endothermic and very slow at low temperatures. The probability of this
reaction occurring is higher when the polymer contains reactive hydrogen. Researchers
report [29,30] that tertiary-hydrogen and secondary-hydrogen adjunct electron-rich
groups are very active.

The degenerate chain branching reaction (IV) (the decomposition of the
hydroperoxides into radicals) may take over the role of initiation in the later phase of
oxidation reaction. The chain branching produces a radical or radicals from a non-radical

intermediate in the process. When peroxides are present in the original polymers, e.g.,

those formed during processing or storage, reaction (IV. 1) may dominate the initiation
reaction.

Audouin et a1. [32] studied the mechanistic schemes of radical oxidation of
hydrocarbon polymers in which initiation is only due to unimolecular or bimolecular
hydroperoxide decomposition. The results of their kinetic analysis have been compared
with data relative to the thermo-oxidation of polypropylene in solid state (60-160°C).
Their results show the unimolecular initiation scheme is in good agreement with
experimental results. The main characteristics of the unimolecular initiation scheme are:
(1) the quasi-independence of the kinetic behavior with initial conditions, and (2) the
initiation period depends only on the rate constant of unimolecular hydroperoxide
decomposition.

(11) Radical conversion

The conversion of hydrocarbon radicals to peroxy radicals is very important
because the majority of oxygen is absorbed by the polymer in this step. As soon as a
radical is generated in an initiation reaction, it reacts very easily with the oxygen.
Ground state oxygen is unusual in that it exists in the triplet state, i.e., it is a diradical.
The reaction of alkyl radical with oxygen is essentially a radical coupling reaction that is
very rapid in nature. The rate of this reaction, however, depends on the concentration of
the oxygen inside the polymer, i.e., on its pressure outside and its ease of diffusion.
Thus, with low oxygen pressure and/or with high sample thickness, the oxidation may

become diffusion controlled.

10

(III) Chain propagation
Chain propagation in polymer oxidation consists of the hydrogen abstraction
reaction of the peroxide radicals. When the rate constant of this reaction increases, the
radical generated in step I and II will react faster with polymer molecules, generate more
free radicals, and increase the rate of oxidation. The reactivity of hydrogen toward free
radicals increase in the following order: primary<secondary<tertiary, in accordance with
the decrease in bond dissociation energies [33]. Because of the high reactivity of the
tertiary hydrogens, intermolecular propagation could be a major reaction path in the
oxidation of some polymers such as polypylene and polystylene.
(IV) Degenerate chain branching
The decomposition of hydroperoxides to radicals is the most important step in
polymer oxidation because of the formation of oxidation products [33]. The thermal
decomposition of hydroperoxides can proceed unimolecularly, bimolecularly, or with the
participation of RH groups, breaking down the polymer chains, forming new chemical
groups such as carbonyl and carboxyl, and generating volatile products such as alcohols,
acids, aldehydes.
(V) Termination
Unimolecular or bimolecular termination of free radicals are most common type
of termination in polymer oxidation. Unimolecular termination is the reaction of alkyl
radicals with molecular oxygen. The unimolecular termination usually occurs when solid
polymer is oxidized in oxygen rich environment. Bimolecular termination is the reaction
of two alkyl radicals forming an inactive particle. This process is very important at the

low oxygen pressures.

11

2.2.2 Chemical Change of Polymers in the Oxidation Process

Fourier-transform infrared spectroscopy (F TIR) has been used to assess the
chemical changes that occurs during thermo-oxidation process of hydrocarbon polymers
[34-35]. FTIR results show that aldehydes (1735 cm") and ketones (1720 cm") are the
main species present during the early stages of degradation, although carboxylic acids
(1710 cm") are the species dominating the later stages. The presence of carboxylic acids
in the later stages indicates the chain scission of polymer molecules during the thermo-
oxidation process. The chain scission of polymer molecules during thermo-oxidation
usually generates small molecules and decrease mechanical properties of the polymers.

The thermo-oxidation of hydrocarbon polymers was investigated by Celina et al.
[34] using FTIR spectroscopy. Polymer degradation was studied by measuring the
spectroscopic changes that occurred during thermo-oxidation of polymers under air at
temperatures ranging from 150 to 250°C. Their results show the formation of carbonyl
groups and related oxidative reactions, weight loss and volatilization, as well as

formation of conjugation and specific polymer reactions.

2.2.3 Effect of Chemical and Physical Structure on Oxidation

The chemical composition and structure of the molecules determine the thermo-
oxidation stability of polymers. The rate of diffusion of oxygen into the polymer affects
the rate of thermo-oxidation reaction. The rate of oxygen diffusion is much larger in the
amorphous regions than in crystalline domains [3 6] because the crystalline core is
inaccessible to molecular oxygen. This implies that the rate of oxidation is greater in the

amorphous regions than in the crystalline regions. The mobility of radicals is also much

12

larger in amorphous regions than the crystalline regions. The diffusion rate of oxygen is
also affected by the morphological degradation and the existence of oxygen containing
groups built into the polymer.

The chemical structure of a polymer strongly affects its ability to resist oxidative
degradation [37]. The rate of thermo—oxidation degradation may increase with increased
chain branching of polymer structures, because polymers with branch chains have more
tertiary hydrogen atoms than pure linear materials and the tertiary hydrogen atoms are
more vulnerable to radical attack than are primary or secondary hydrogen atoms. The
presence of double bonds in a polymer backbone also affects the mechanism of oxidative
degradation because the radicals can react easily with double bonds. Oxidation products
may also affect the rate of oxidation of the parent polymers. Many of the first formed

oxidation products are more easily oxidized than the parent polymer.

2.3 Hygrothermal Effect of Polymer and Polymer Composite Materials

Polymer and composite materials absorb some amount of water slowly from its
surroundings and this process is controlled by the diffusion of water in the polymer
materials and environmental temperatures. The absorbed water often causes
hygrothermal effects such as the degradation, plasticization, swelling, and lowered T3 of

the polymer materials.

2.3.1 Water Absorption Kinetics

Moisture absorption process in polymer and polymer composites can be described

by F ick's law of diffusion which can be expressed as

13

2
°FAt4-=D°ax]:4 (2.1)

 

where M represents the concentration of the diffusing medium and x is a dimensional
coordinate. By applying the appropriate boundary conditions and integrating, an

expression predicting moisture absorption with time can be expressed as

______..M' M" =1—_8—22_1 xeXp{ Dani-:21) n I} (2-2)

Where Mt, Mo, and M00 are the moisture content at time t, the initial moisture content, and
the equilibrium moisture content, respectively, D is the coefficient of diffusion and h is

the sample thickness. The coefficient of diffusion, D, is given by:

2 2
h _
D = ”[_] [M] (2,3)
4M.» t/EZ—tE
Diffusion coefficient depends on the environmental temperatures. The

dependence of diffusion constant on temperatures follows the following expression:

D(T) = D0 exp(— fig—T) (2.4)

where Do is the diffusion constant. Q is the activation energy. T is the temperature.

The diffusion constant also depends on the physical properties of specimens. Rao
et a1. [3 8] studied the moisture diffusion characteristics of unidirectional composite
specimens. They found that the diffusion coefficient of the composites is affected by the
cut-edge surface area and by lay-up sequences of composite materials, which was
attributed to tortuosity caused by fiber orientations.

Water absorption of some polymer materials has been observed to deviate from

the Fick's law after prolong exposure to water. Chateauminois et al.[39] investigated the

hygrothermal degradation of glass fiber/epoxy unidirectional composites. They showed
water absorption can be described by Fick’s law after short immersion times when water
sorption occurs mainly by diffusion through the epoxy network. They noticed water
absorption curves slowly deviates positively from Fickian behavior after a long time
aging. The deviation was attributed to the occurrence of debonding at the interphase
region during hygrothermal aging. They concluded that the hygrothermal degradation of
the interface occurs essentially in the non-Fickian absorption steps when the epoxy
matrix is close to saturation.

The Langmuir model is often used to describe non-Fickian water absorption. The
Langmuir model assumes the absorbed water molecules exist in two phases, one free, as
in the F ickian model, and the other trapped due to the interactions between the molecules
of water and matrix. Coefficients were introduced into this model to account for the
possible interactions between the water and polymer molecules. Cai et a1. [40] studied
the effects of non-Fickian water diffusion in fiber-reinforced polymeric composites. The
departure from classical diffusion was attributed to the time-dependent response of
polymers.

Langmuir-type two-step or one step model could be used to fit the experimental
data. The Langmuir-type two-step model [41] can be represented by equation 2.5 and

2.6.

15

62n an 6N

—— = — + — 2.5
6x2 5: at ( )
6N
—— = yn — ,Bn (2.6)
6:

Here, n is the number density of mobile molecules; N is the number density of
bound molecules; 7 is probability per unit time that during which a bound molecule of

water become mobile; B is probability per unit time that a mobile molecule of water

become bound; D is the coefficient of diffusion.

2.3.2 Plasticization and Tg Variation

Moisture absorption in most polymer and fiber-reinforced polymer composites is
associated with a decline in the glass transition temperature (Tg) attributed to
plasticization of the matrix. A number of material factors could influence the magnitude
of the Tg drop including water content, absorption temperature, exposure history,
composite moisture diffusivity, resin matrix chemistry and initial matrix properties. The
influence of each of these factors must be keenly investigated to provide insight into the
details of moisture absorption. Such critical insight is required to design polymer and
composite materials systems with improved resistance to hygrothermal degradation.

Previous researches [42-46] have shown that the T8 of epoxy resins and Gr/Ep
composites decreases as the water content increases. Kelly et al. [42] derived an
expression to calculate the Tg of a plasticized system based on the free volume concept of

polymeric materials.

T : apvagp +ad(l 411,)ng

8

 

2.8
apvp+ad(l-vp) ( )

Here, a is the expansion coefficient, v is the volume fraction, and subscript p and d are
for polymer and diluent, respectively.

Zhou et al. [44] found Tg depends not only on water content, but also on its
exposure time and history. The longer the exposure time and higher exposure
temperature, the less the drop in glass transition temperature. Cinquin et al. [19] studied
the water absorption of composite materials based on different resin families:
bismaleirnide(BMI), cyanate and epoxy. Their data shows the glass transition
temperature of cyanate ester based composites decreases more than BMI and epoxy-
based composites. Ma et al. [45] investigated the hygrothermal behavior of carbon fiber-
reinforced ploy(ether-ether-ketone) (PEEK) composites. They found the hygrothermal
effect on the glass transition temperature of PEEK composites is negligible.

The glass transition temperature is reversible in most case which the material
showing a recovery to the original Tg upon drying. Rice et al. [46] and VanLandingham
et al. [47] evaluated the hygrothermal effects on the polyimides composite. They
observed the glass transition temperature T8 was lowered with moisture absorption but
was recovered when the sample was re-dried. Chateauminois et al. [3 9] investigated the
plasticization of a glass/epoxy unidirectional composite by dynamic mechanical thermal
analysis (DMTA). They found the plasticization effect on the therrno-mechanical
property of the composite was temperature independent and reversible after re-drying of
the aged material. In some polymer materials, however, moisture can cause irreversible
degradation of polymer structure with the Tg change becomes irreversible. Xiao et a1.
[48] studied the hygrothermal aging effects on epoxy resin by FTIR, gravimetric tests and

DMTA. The results show that water immersion at 50, 70, and 90°C lead to the

17

introduction of carbonyl groups in the resin and chain scission of the cross-linked
structure. These irreversible structural changes induce an irreversible decrease of the
glass transition temperature of epoxy resin.

A glass transition temperature increase was observed in some cases when the
moisture diffusing into the polymer caused crosslinks of polymer molecules. Morgan et
al. [49] studied the hygrothermal effects on BMI-carbon fiber composites using Fourier
transform infrared spectroscopy (FTIR) and differential scanning calorimetry (DSC).
Their result revealed that these resins are not fully cured and can continue to react as a
result of dehydration-induced ether crosslinks. This further crosslinking causes Tg

increases and mechanical property decreases.

2.3.3 Effect of Moisture Absorption on Mechanical Properties

Mechanical properties of fiber reinforced composite materials depend on those of
their components, the fibers and the matrix plus the quality of the fiber matrix interface.
A mechanism that results in the degradation of the matrix, fibers or their interface will
result in a loss of characteristics of any structure made from composite materials.

Hygrothermal aging often cause the plasticization of polymer matrix with
depression of Tg and swelling of specimen. These effects usually decrease the tensile
properties of polymer and polymer composites. Mohd et al. [50] studied hygrothermal
aging of nylon 66 and its carbon fiber-reinforced composites. They found hygrothermal
aging reduced the tensile property of both unreinforced and reinforced nylon 66. This
behavior was explained by moisture-induced plasticization and interfacial degradation.
Haque et a1. [51] studied the effects of moisture on tensile properties of Kevlar 49/epoxy,

graphite/epoxy, single fiber and Kevlar 49-graphite/epoxy hybrid composites. Their

18

results showed variations in tensile strength, elastic modulus and Poisson's ratio with
changes in the hygrothermal conditions. Their results indicated that moisture degrades
the tensile strength of hybrid and single fiber composites.

Uschitsky et al. [52] studied the hygrothermal effect on epoxy filled with silica
and alumina nitride particles. They showed that moisture diffusion leads to a decrease in
the compound's strength and to a substantial increase in the material's plasticity. Bouadi
et al. [53] studied the effects of moisture on Young's modulus, Poisson's ratio and
material damping of the epoxy matrix. Their results showed that the effects of moisture
were negligible at room temperature. As temperature increased, the moisture induced
significant changes in Young's modulus and Poisson's ratio.

Hygrothermal effects also led in a temperature dependent decrease in fatigue
properties of composite materials. Ma et al. [54] studied the hygrothermal effects on the
fatigue behavior of the carbon/PEEK laminated composites. Their results indicated that
the fatigue lives of immersed specimens were shorter than those without hygrothermal
aging. Vauthieret et al. [55] studied the effects of hygrothermal aging on the fatigue
properties of a unidirectional glass/epoxy composite. They found the endurance
properties of the composite declined after water aging. Komai et al. [56] investigated the
effects of water absorption on fatigue strength of carbon-epoxy composites. They found
the fatigue lives of composite specimens decreased in water. They also found water
absorption lowered the yield strength of the epoxy resin, changing the shape of the stress-

strain curves of composites.

19

2.3.4 Hygrothermal Degradation of Polymer and Composite Materials

Degradation of composite materials may be caused by environmental exposure to
humidity at increased temperature. The absorbed water could change the state of residual
stress of composite materials and causes rnicrocracks and delamination of composites.

Microcracking is a very important form of damage in composites. Microcracks
could form pathways for the accelerated ingress of moisture and chemicals into the
materials. Long-time exposure to hygrothermal environments could cause microcracks
of polymer composite materials. Shirrell et a1. [57] studied the exposure of epoxy
composites to several different hydrothermal environments and examined the
hygrothemally induced microcarcks by scanning electron microscopy (SEM). They
observed severe rnicrocracks at 82°C; the severity and frequency of these rnicrocracks
increased with relative humidity. Obst et al. [5 8] studied the evolution of rnicrocracks
under static and fatigue loading and the effect of rnicrocracks on the mechanical
performance. They found the presence of rnicrocracks reduces the measured bearing
strength as well as the interlaminar shear strength of the composite. Their results also
showed a significant increase in the apparent diffusion coefficient in microcracked
specimens.

Delamination is a very important failure mechanism of composite because it
reduces the stiffness, strength and fatigue life of composite materials. Long-time
exposure to hygrothermal environments could cause the delamination of polymer
composite materials. Several studies [6,7,59] show that moisture absorption of thermoset
polymer composite degraded the delamination fracture toughness (DF T) of composite

laminates. Lucas et al. [6] studied the water absorption on Model I (out-of-plane

20

loading) delamination of resin-matrix composites. Their results show significant
reduction delamination fracture toughness occurred with moisture absorption under
severe hygrothemal conditions. Other studies [5 9], however, reported the interlaminar
fracture toughness of thermoplastic polymer composite increases as these materials
absorb moisture. Hoa et al. [60] investigated the hygrothermal effect on Mode 11
interlaminar fracture toughness of a carbon/polyphenylene sulfide (PPS) unidirectional
laminate. It was found that the interlaminar fracture toughness of the carbon/PPS
laminate increases with both moisture content and temperature. Hooper et al. [61]
studied the water absorption on Model I and Mode II delamination of graphite/epoxy
composites. Their results revealed that moisture absorption tends to toughen the
composite material. The difference in fracture toughness between thermoset and
thermoplastic composite could be attribute to the fact that thermoset polymer composites
absorbed much more moisture than thermoplastic polymer composite. The extra
moisture in the thermoset composite degrades the matrix and matrix-fiber interface,
losing the gain of fracture toughness from the plasticization of the matrix, that usually
increases the fracture toughness.

Hydrolysis of some polymer network has been observed after a long time
exposure to water. Mohd et al. [62] studied the injection molded short glass fiber
reinforced poly(butylene terephthalate) subjected to hygrothermal aging. Examination
of fracture surfaces using a scanning electron microscope (SEM) revealed the hydrolysis
of the polymer matrix. They observed that the hydrolysis resulted in the formation of

microvoids and degradation at the fiber-matrix interface.

21

2.3.5 Hygrothermal Effect on Interphase Phase

The fiber-matrix interface is a three-dimensional region around the fiber where
properties are different from fiber and matrix. The performance of the composite is
largely controlled by the strength of the bond across this interface. Water absorption has
been shown to lead to a reduction in the mechanical properties of polymer composites
[63-65]. This has been attributed, in part, to the hygrothermal degradation of fiber-matrix
interface bond. Several researchers [63 -66] observed that the hygrothermal conditions
could cause an interfacial debonding in glass fiber/epoxy resin composites. Kaelble et
al. [63] observed that the interlaminar shear strength could be reduced by 30-50% after
immersion for 200 hours in 100°C water. They suggest that the degradation process is
irreversible and is directly related to cumulative moisture degradation of fiber-matrix
interfacial bond. Morii et al. [64] studied the weight changes in the interphase region of
hygrotherrnally aged fiber-reinforced plastics. Their results showed that the weight gain
in the interphase was caused by water penetration into the interphase region. They
suggested that improvement in the water resistance of the binder at the interphase is
important to enhance the water resistance of glass-fiber-reinforced plastics. Meyer et al.
[65] investigated the long-term durability of the interfacial bond in carbon fiber
polyetheretherketone (CF/PEEK) and carbon fiber/polysulfone (C.F./PSF) composites
after exposure to hygrothermal environments. Their results indicated that interfacial
bond strength in C.F.IPSF and C.F./PEEK composites is significantly influenced by long-
term exposure to water, especially at high temperatures. Their results indicated that
water saturation reduces interfacial bond strength in C.F.lPSF and C.F./PEEK composites

as a function of both time and temperatures. Sohn et al. [66] investigated hygrothermal

22

stability of graphite fiber/epoxy interphase. They found that the interfacial shear strength

decreased gradually with water immersion time.

23

Chapter III
THERMO-OXIDATION OF CYANATE BASED POLYMER MATERIALS

3.1 Experimental
3.1.1 Materials

Fiberite 996HM and 954-3HM resins were used in this study. All the materials
were received from Fiberite, Inc. F iberite 996HM resin is a 177°C curing siloxane-
modified cyanate resin with a -128°C to +121°C service temperature. F iberite 954-3HM

is 177°C curing cyanate resin with a - 128°C to +121°C service temperature.

The chemical structures of these materials are shown as follows
Polysiloxane:
X|(|3H2R CH2R
Si—O—Sli X
n
(I3H2R CH2R

X: -CN R: H n:10-1000

Cyanate ester monomer:

CH3
N=C—O—<: :C>—— (III—<: :>—0—C=— N
CH3

24

The nominal physical properties of 954—3 HM cyanate ester resin and 996HM

siloxane-modified cyanate ester resin were listed in Table 3.1 and Table 3.2 [23]

Table 3.1 Nominal physical properties of CE at room temperature

 

 

 

 

 

Tensile Strength 56.5 MPa
Tensile Modulus 2.76 GPa
Flexure Strength 119.27 MPa
Flexure Modulus 2.96 GPa
Density 1.19 g/cc

 

 

 

 

Table 3.2 Nominal physical properties of SMCE at room temperature

 

 

 

 

 

Tensile Strength 51.0 MPa
Tensile Modulus 2.96 MPa
Flexure Strength 75.8 MPa
Flexure Modulus 3.24 GPa
Density 1.146 g/cc

 

 

 

 

3 .1 .2 Specimen Preparation
All the samples were polished into 1 mm thick thin plate. Specimens of the

dimension of 25 mm x 25 mm were cut form the plate using diamond saw. The weight of

25

the specimen ranged from 2 to 2.5 grams. The edges of the specimens were subsequently
ground using 600 grit abrasive paper to achieve consistently smooth edge surfaces. All
the samples were marked with a vibration pen and cleaned with methanol using
ultrasonic cleaner. The samples were handled very carefirlly during the experiments to

avoid any surface contamination.

3.1 .3 Gravimetric Experiments

Samples were tested in an environmental test chamber consisting of gaseous
aging environments of either ambient air or argon at 100°C. Gravimetric experiments
were used to monitor weight of specimens during aging. Specimens were exposed to the
aging environments for up to 3000 h. Specimens were taken out of the oven periodically
during the aging experiment. Specimens were weighed using the analytical electronic
balance with 0.01mg resolution. The experimental error of the measurement is i 0.02%.

The weight percentage change is calculated as follows

Wt —W0

 

w%=

x 100% (3.1)
wo

Where w% is the weight percentage change, w is the weight of aged samples, and wo is

the original weight of samples.

3.1.4 Glass Transition Temperature Measurement

A TA Instruments 930 differential scanning calorimetry (DSC) system was used
to determine Tg variations at different thermo-oxidation stages. Specimens of 3 mm x 3

mm x 1 mm were cut from the thermal aged samples at different time periods. The

26

samples were sealed in small aluminum pans to prevent any weight loss during the
experiment, and the weight of each sample pan was measured before and after each test.
The samples were put into testing chamber protected by pure nitrogen gas. The
temperatures range from 25°C to 350°C, and the temperature increase rate is 5°C/min.
The T8 values were determined by the intersection of two tangential lines of the DSC

curve before and after the glass transition regime.
3.1.5 Three-point Bend Test

Three-point bend flexure testing was performed to study the effects of therrno-
oxidation on mechanical properties of both CE and SMCE. The standard procedure
(ASTM D790-90) was used as guide for specimen preparation and testing. A
representative sample of five specimens was tested for each aging condition, i.e., as-
received samples and samples aged for 1 and 3 months at 100°C in the air. Tests were
conducted using Instron® mechanical test machine, with a 10 KN load cell. The nominal
specimen dimensions used in this study were about 25 mm long, 3-5 mm wide and 1 mm
thick. A cross-head speed of 2 mm/min was used in the test. For every experiment, 5

specimens were tested and the average data points were taken.

The flexure strength was calculated by formula

 

3PL
S = 3.2
2bd2 ( )
where S = Flexure strength (MPa)
P = Brealdoad (KN)
L = Outer (support) span (mm)
b = Specimen width (mm)
d = Specimen thickness (mm)

27

3.1.6 FTIR Testing

Infrared spectroscopy was used to determine the change of chemical structure of
cyanate based polymer materials associated with bonding characteristics at various
thermo-oxdation stages. A Nicolet 700 Series F TIR was used in this study. The cyanate
polymer sample was polished to 0.025 mm thick thin fihn. The thin film was secured to
a holder designed for F TIR test. The films with the holder were put into environmental
chamber with dry air at 100°C for different periods of time up to 1200 h. The holder was
taken out of the oven during different period of time. The sample was cool to room

temperature in the air and transferred into the FTIR testing chamber.

3.2 Results and Discussion

3.2.1 Assessment of Thermo-oxidation of CE & SMCE by Gravimetric Experiments

To study the therrno-oxidation process of cyanate based polymer materials,
cyanate ester and siloxane-modified cyanate were aged in an environmental test chamber
consisting of either ambient air or argon at 100°C. Gravimetric experiments were used to
monitored weight change of the specimens during thermal aging. The weight change of
CE and SMCE are shown in Fig. 3.1. The graph shows that weight reduction of SMCE
specimens was observed initially and followed with weight gain after 9h of exposure.

The initial weight lose is due to the desorption of the moisture absorbed on the specimen

surface.

28

 

+siloxane-nndifiedcyanateagedintheair -I-cyanateesl:eragedintheair
+siloxane-rmdifiedcyanateagedinfl1eargon +cyanateeswagedintheargon

0.5
04 -L__2_H.LHL_L -W.W-Wfl /.

-I___.L.wmm_. ////7

L-flz

 

COO
NOD

A
l

 

 

 

 

 

l

 

weight change %
O
O

55.6
N

 

 

 

 

 

 

0 5 10 15 20 25
. . 1 2
squarerootofasmgm (h ’ )
Fig. 3.1 Weight change versus exposure time profiles are shown for siloxane-

modified cyanate and cyanate ester resin materials aged in the air and argon at 100 °C.

29

The graph also shows the weight of the samples kept increasing with increase aging time
up to 400 h. The surface color of SMCE specimens turns slowly from yellow to brown
when aged in the air. Examination of specimen cross-section shows that the discoloration
tended to start at the surface and gradually progress inward to the specimen interior. In
contrast, no weight gain or color change was observed of SMCE when aged in the argon
in excess of 6h. At the same time, only weight reduction was noticed in aging of CE both
in the air and in the argon at 100°C. The weight lose is due to the evaporation of moisture

absorbed on the sample surface.

To further confirm the therrno-oxidation process, experiments were conducted to
study the weight gain of SMCE in different therrno-oxidation processes. In the first
experiment (experiment A), samples were heated in the air at 100°C for 850 hours ( step
1) and then heated in the argon for 400 hours (step 2). The results of experiment A are
shown in Fig. 3.2. The graph shows SMCE specimens gain weight when they were aged
in the air for 850 hours. The graph also shows SMCE didn't gain any weight when aged
in the argon after they were aged in the air for 850 hours. In another experiment
(experiment B), samples were heated in the argon for 200 hours ( step 1) then heated in
the air for 1400 hours ( step 2) and heated in the argon again ( step 3) for 400 hours. The
results of experiment B are shown in Fig. 3.3. The graph shows SMCE specimens didn't
gain any weight when they were aged in the argon. The graph also shows that the
samples start and keep gaining weight when they were aged in the air after aged in the
argon for 200 hours. Both experiments show that the thermal degradation process ceased

when the supply of oxygen in the aging environment was halted.

3O

A< 32:22.on
0.2: a new... a: 5 Baa 5.: ea 2: 5 Sue 855. he emcee Sue? 3 .5

A35 2:: ”fine no .8.— 98.5w

 

om

P

9 m

 

 

 

 

 

 

 

 

 

 

 

 

 

 

:m 05 5 wow“ fl 93m

 

cows 05 E coma N mam

 

 

1|
II

 

 

 

 

 

 

md-

fio-

o.o

_.o

Nd

md
v.0

% afiuaqo 11181:) AA

md
ed
5.0
wd

31

Am “conning—cg 062: «a Ewan Swan
2: 5 Baa as ea 2: 5 .6? :2: as ace .3»: a: 5 3? 82m .e ewes: £363 on 5...

A25 2:: «fine we “8.. 22:5

9 2 9 mm Om mN ON 2 2 m o 2 m

 

 

 

F

 

coma
05 5 come ~ @on

 

 

 

\\

 

 

\

 

com;
05 5 come m mam

he 05 5 come N mam

 

 

 

 

 

 

 

 

 

od

Nd

Yo

0.0

wd

o.—

% afiueqo 111819,“

32

Therefore, the role played by the supply of oxygen in the therrno-oxidation
process of SMCE is key to understand the therrno-oxidation process. These results show
SMCE gains weight only when oxygen is present in the environment. Therefore, the
weight gain should be caused by the thermo-oxidation of SMCE by oxygen in the air.
The results also show that pure CE didn't gain any weight when they were aged in the air
or in the argon, which suggests that no thermo-oxidation occurred in the pure CE resin at
this temperature. Siloxane-modified cyanate ester (SMCE) is cyanate ester (CE)
modified with polysiloxane molecules, as shown in section 2.1. The distinguishing
difference of their thermal aging behavior in the air suggests that the oxidation of SMCE
could only be introduced by thermo-oxidation of the polysiloxanes, which was used to

modify the cyanate ester resin.

For a more detailed assessment of thermo-oxidation process of SMCE in the air,
we extracted data of the weight gain of SMCE aged in air from experiment A and
experiment B and plotted them in Fig. 3.4. The graph shows that the weight gain curves
in the therrno-oxidation of SMCE are nearly parallel. Both curves are linear with the
square root of oxidation time.

Equation 3.3 to follow is the empirical relation of the average weight gain related

to the aging time derived from the Fig. 3.4 is
w = wo + kt"5 ,k==0.032 g-h'°'5 (3.3)

where w is the weight gain, we is the original weight, k is a empirical constant ( average

slope of the curves).

33

Theoretically, therrno-oxidation of polymers usually involves two steps: (1) the
diffusion of oxygen into the polymer and (2) the reaction of oxygen with polymer
network groups. The rate of therrno-oxidation process is usually controlled by one of
these two steps. For the diffusion controlled thermo-oxidation process, the weight
change was found to be proportional to the square root of time, t °‘5 [36]. It is
distinguished from the chemical reaction controlled thermo-oxidation process, in which
weight gain was found to be linearly related to time, t.

Equation 3.3 we derived from the experiments is found to be proportional to t°'5,
which suggests that thermo-oxidation of SMCE is controlled by the oxygen diffusion rate
in the polymer network. Nam et a1. [3 6] reported similar results. They observed the
weight change of bismaleimide/carbon fiber composite during the therrno-oxidation
process was controlled by oxygen diffusion, and the reaction rate was found proportional
to t 05.

Thermo-oxidation of SMCE involves two basic mechanisms: (a) the diffusion of
oxygen into the SMCE and (2) the chemical reaction associate with free radicals. Since
the rate of therrno-oxidation reaction of SMCE polymer molecules with oxygen is much
faster compared to the rate of diffusion of oxygen in siloxane-modified cyanate ester at

100 °C below its glass transition temperature (169 °C), the process controlling thermo-

oxidation in SMCE is the oxygen diffusion in the polymer network.

34

 

 

9 experiment A
Linear (experiment A)

I experiment B
_ _ ._ Linear (experiment B)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.8
2: I
o_7 Hafiz—0.9891 R 0i
/
aged in the air /

0.6 L
\o /
O
m 6.5 .1
g) /
m /
.C r J.
3 °" [J u
.5 k=0.031 / k=0.032
.— 0.3 -— l
(D i
a / I’

0.2 / L

0.1 /

/o i
0 . . . . .
0 10 20 30 40 50 60
square root of time ( hm)
Fig. 3.4 Weight change of siloxane-modified cyanate ester aged in the air at
100°C

35

Therrno-oxidation was not observed in the pure cyanate ester at 100 °C. Thermo-
oxidation degradation of cyanate ester has been observed at temperature above 150 °C in
the air [17,18]. This suggests that the initial temperature required for thermo-oxidation of
cynanate ester is higher than for siloxane-modified cyanate ester. It was reported [24]
that SMCE has lower degree of crystallinility than the cyanate ester. Researchers [29,3 0]
have shown that the initiation temperature of thermal-oxidation increases with decreasing
crystallinility of the polymers, which makes the structure more vulnerable to be attacked

by oxygen.

Some authors [34,37,68] observed weight loss of the polymer sample during the
thermo-oxidation of polyolefm polymer at temperatures above or close the glass
transition temperature (usually above 200 °C). The weight loss was contributed to
volatilization of small molecules generated by the oxidative reactions. In our
experiments, the aging temperature (100°C) is much lower than the glass transition
temperature (170°C) of siloxane-modified cyanate ester. At this temperature, the
molecules generated by the thermal-oxidation reaction are more likely to stay in the

polymer, increasing the weight of experiment samples

3.2.2 Degradation of Physical and Chemical Properties of CE & SMCE Materials

During thermo-oxidation process, polymer function groups react with oxygen
molecules diffused into the polymer network. At the same time, the molecular chain
scissions and rearrangement occurs, causing reduction of the molecular weight and

changing the molecular weight distribution of polymers. The glass transition temperature

36

(Tg) of the polymer also decreases because of chain fragmnetation and lower molecular

species generated during the aging process.

Several authors[l 7,18,37]observed the decrease of the glass transition temperature
during the thermo-oxidation of polymer. Hinsken et al. [3 7] studied the therrno-
oxidation of polypropylene and found glass transition temperature decreases 25%. They
also observed chain scission of the macromolecules. Gedde et al. [69] studied the
therrno-oxidation of polyolefin in the hot air. They observed the decrease of T8 and
degradation of the polymer by thermo-oxidation after a long time exposure to the hot air.
The effect of thermo-oxidation on the durability of cyanate ester resin (F iberite 954-2)
has been studied by Paravatareddy et al. [17,18]. Their results show that the glass
transition temperature of CE decrease substantially (lo-30%) when aged in the air for 3

months at 150°C.

To evaluate the thermal degradation of SMCE and CE, we measured the glass
transition temperature (by DSC) of both materials aged in the argon or air at 100°C and
140°C. The results are shown in Table 3.3 and Table 3.4. Table 3.3 shows that the glass
transition temperature of SMCE decreases almost 20% (35 °C) when aged in the air at
100 °C for 40 days. On the contrary, Table 3.3 shows no change of Tg of SMCE when
aged in the argon even at higher temperature (140°C). In comparison, cyanate ester resin
samples show no decrease of Tg in both environments and temperatures, therefore, they

show no sign of degradation.

37

Table 3.3 Glass transition temperature of CE aged in different environments (argon and

 

 

 

 

air)
Materials (CE) Tg ( °C )
As received 240
Heated in Argon for 3.4x10rsec (~ 4 240
days) at 140°C
Heated in the Air for 3.4x 16° sec ( ~ 40 240
days) at 100°C

 

 

 

 

Table 3.4 Glass transition temperature of SMCE aged in different environments (argon

 

 

 

 

and air)
Materials (SMCE) Tg ( °C )
As received 195
Heated in Argon for .4x105 sec (~ 4 195
days) at 140°C
Heated in the Air for 3.4x10° sec ( ~ 40 160
days) at 100°C

 

 

 

 

38

The decrease of glass transition temperature of SMCE suggests that the thermal
degradation of SMCE takes place during the therrno-oxidation process. The difference in
thermo-oxidation behavior of CE and SMCE is related to the chemical structure
difference. The chemical structure difference of SMCE and CE (shown in section 2.1)
suggests that the thermal degradation of SMCE is due to the thermo-oxidation of

polysiloxane molecules, which may involve chain scissions of polysiloxane molecules.

To study the degradation of mechanical strength of cyanate ester based polymer
materials during the thermo-oxidation, three-point bend tests were used to characterize
effect of oxidation on the flexure strength of SMCE and CE before and after the thermal
aging. The results were plotted in Fig. 3.5. Results from the three-point bend test of the
CE and SMCE show a clear decrease of the flexure strength of SMCE. After being aged
in the air at 100°C for 3 month, the flexure strength of SMCE decreased almost 20%. In
comparison, the flexure strength of CE remains almost the same after aged in the air at

100°C for 3 months.

The decrease of the flexure strength of SMCE during the thermo-oxidation
process may be due to the chain scissions of polysiloxane molecules. The chain scission
breaks the long Si-O-Si bonds of polysiloxanes chains into smaller molecules. The
flexure strength of the polymer will decrease when the polymer chains are degraded in

the aging process.

39

 

 

F lexure Strength (MPa)

 

 

CE arr-received CE aged SMCE as-reoeived SMCE aged

 

Fig. 3.5 Three-point bend test of cyanate based polymer material (aged and as-received).

Samples aged in the air at 100°C for 3 months

40

3.2.3 Therrno-oxidation Mechanism of SMCE

Infrared spectroscopy (FTIR) was used to determine the mechanism of therrno-
oxidation and the change of chemical structure of SMCE. The sample was aged in the air
at 100°C for more than 1200 hours. From the experimental results, shown in Fig. 3.6 a
and 3.6 b, it was evident that the intensity of the peak at wavelength 1720-1730 cm'1
increases with thermo-oxidation time. The decrease of the intensity of the peaks at 911

cm'1 and 966 cm“1 with increase aging time was also observed.

The peak at wavelength 1720-1730 cm'1 is the characteristic absorption of
carbonyl groups ( -O=C-O-). The appearance and increase of this peak indicates a
formation of carbonyl groups during the thermo-oxidation. The peak at wavelength 911
cm“1 and 966 cm'1 is the characteristic absorption of the CH bonds of the siloxane
-SiOCH2R group. The decrease of its intensity indicates a degradation of these groups in
the therrno-oxidation process.

Several authors [34,3 5] observed similar functional groups formed in the thermo-
oxidation process. Celina et al. [34] used FTIR spectroscopy to study polymer oxidation
by measuring the spectroscopic change that occur during thermo-oxidation degradation of
polymers exposed to air at temperatures ranging from 150°C to 250°C. Their results
show that the formation of carbonyl groups during the thermo-oxidation reactions.
Yoshitaka et al. [3 5] studied the thermo-oxidation degradation of poly(methyl
methacrylate) (PMMA). They found that the initial period of the oxidation is

characterized by a decrease in the intensity of absorption band of C-H groups.

41

 

40=C4}-
1776
1620 1110 1660

1980 1000

aiio 2060
navenunaea

2220

2500

 

 

380

’1'? 0'2 9‘1 2'1 8' ‘1' 0"
EONUQHDSQU

Fig. 3.6 (a) FTIR spectra (1660 cm" — 2380 cm") of SMCE aged in the air.
Change in the spectra with aging time shows the increase of the 1726 cm'1 peak

42

015

07s

1005

 

85

 

 

l

‘1'? 0’2 9’! 2'1 8' h' 0'
EONUBHDSQU

Fig. 3.6 (b) FTIR spectra (765 crn'l — 1035 cm") of SMCE aged in the air.
Change in the spectra with aging time shows the decrease of 911 cm’1 and 966 cm’I
peaks.

43

HHVENUMBER

From the experimental results we observed, a probable mechanism illustrating the

experimental finding could be represented as following:
(1) Thermal initiation

Initiation by direct reaction of the polymer with molecular oxygen in step I has
been observed during therrno-oxidation of most polymer materials [32-34]. This
reaction is endothermic and very slow at low temperature. The probability of this
reaction occurring is higher when the polymer contains reactive hydrogen or/and at
elevated temperature. Researchers have reported [3 2-33] that tertiary hydrogen or
secondary hydrogen close to electron rich group such as Si-O is very active. We
observed the thermo-oxidation of SMCE happens at elevated temperature (100 °C). We
also found that the absorption of C-H bonds of the siloxane -OSiCH2R group decreases
with aging time. These results suggest the initiation reaction of thermo-oxidation of

SMCE is the breaking of C-H bonds of the siloxane -OSiCH2R group as following:

CH2R CH2R RHC- -H CHzR
V\Si—O—Si/\/—‘* \Asi—o—SiA/

CHzR CH2R CH2R CH2R

44

(II) Radical conversion

This step usually involves the conversion of hydrocarbon radicals to peroxy

radicals. This step is very important because the majority of oxygen molecules is
absorbed by the polymer in this step. As soon as the MO is generated on side group in
the initiation reaction, it reacts very easily with the oxygen, and the radical is quite stable
below 300°C. During the thermo-oxidation of SMCE, we observed that the therrno-
oxidation and degradation happens only in the oxygen surrounded environment at 100°C.

The weight gain of SMCE can be explained by the reaction of these radicals with the
oxygen diffused into the polymer. It also suggests the amount of oxygen absorbed is
controlled by the rate of oxygen diffirsion as observed in section 3.3.1. The whole

scheme follows.

RHC' CH2R RHCOO ° CH2R

\ASi—o—Si/\/ $1» V\Si-—o—Si/\/

 

 

 

CH2R CH2R CH2R CH2R

45

(III) Chain propagation

The radical generated in steps I and II kept react with siloxane molecules,
generating more free radical on the side group RHCOz'. The rate of oxidation increases

as the kinetic chain length becomes larger.

The radicals formed in this step also form more stable cyclic peroxide ring that
has been observed in most thermo-oxidation of engineering polymers [29,30,70].
Researchers also observed that the cyclic peroxide ring in polysiloxane copolymer was
formed by the intramolecular exchange reactions involving the Si-O bonds. During the

thermo-oxidation of SMCE, a RHCOz' radical can form a peroxide cyclic ring radical

with a Si-O bond, which is chemically more stable than the RHCOz'. The whole scheme

 

follows.
RHOOO CHzR RHC— O \ (IZH2R
0
V\si—O— st/\/ _* V\Si—o/— Sl/\/
CH2R CH2R CH2R CH2R

(IV) Degenerate chain branching

The decomposition of hydroperoxides to radical is a very important step in
polymer oxidation because the formation of oxidation products is caused by the

degradation of these radicals. The thermal decomposition of hydroperoxides can break

46

down the polymer chain, forming new chemical groups such as carbonyl and carboxyl,

generating volatile products such as alcohols and acids.

As shown in section 3.2.2, the thermo-oxidation of SMCE accompanies decreases
in the glass transition temperatures and flexure strength. It also suggested the
degradation of SMCE by chain breaking of Si-O-Si chain of polysiloxane group during
the thermo-oxidation process. F TIR results indicate the formation of carbonyl groups
during the thermo-oxidation process. The gravimetric experiment only shows weight
gain, which means no volatile products were generated in this process. These
experimental results suggest that the decomposition of hydroperoxides during thermo-
oxidation of this polymer involves breaking of Si-O-Si chain, forming carboxyl group

with rearrangement of chemical structure and forming non-volatile products as following.

(V) Termination

Under most conditions [32-33], unimolecular or bimolecular termination of free
radicals in polymer oxidation occurs almost exclusively. The reaction of alkyl radicals

with molecular oxygen is very fast. The unimolecaur termination usually occurs when

47

solid polymer is oxidized. This mechanism provides an explanation for rapid
termination of thermo-oxidation of SMCE when the sample is removed from the oven

and stored at room temperature.

48

3.3 Summary

The thermo-oxidation effect on cyanate based polymer materials can be
summarized as follows:

- Gravimetric experiments show that SMCE gains weight when aged in the air at
100°C but shows no weight gain when aged in the argon at 100°C. Cyanate ester resins
show no sign of weight gain when aged in the air or argon at 100°C.

0 DSC tests show that glass transition temperature of aged SMCE decreases about
35°C after being aged in the air at 100°C for 40 days. Three-point tests show that the
flexure strength of SMCE decreases about 20% after aged in the air for 3 months.

0 F ITR shows that thermo-oxidation process is going through the oxidation of the
polysiloxane side group and breaking the main chain at the same time.

0 Results suggest degradation of SMCE occurs at 100°C in the air through thermo-
oxidation process. The process is controlled by the supply and diffusion of oxygen in

SMCE.

49

Chapter IV
HYGROTHERMAL EFFECT ON CYANATE BASED POLYMER MATERIALS

4.1 Experimental

4.1.1 Materials

The materials under study were received from F iberite Inc.: 996HM, 954-3 HM,
M4OJ/996, UM55/954-3. Fiberite 996HM is a 177°C curing siloxane-modified cyanate
resin with a -128°C to +121°C service temperature. Fiberite 954-3 HM is 177°C curing
cyanate resin with -128 °C to +121 °C service temperature. M40J/996 is unidirectional
carbon fiber (1M8) reinforced 996HM composite plate. UM55/954-3 is unidirectional
carbon fiber reinforced 954—3 HM composite plate.

The chemical structures of these materials are shown as follows

Polysiloxane:

CH2R CH2R

X Si—O — Si X

| I n

CHZR CH2R

X: -CN R: H n:10-1000

50

Cyanate ester monomer:

CH3

NEC—O—@—C|lZ—@>—O—CEN

CH3

The nominal physical properties of 954—3 HM cyanate resin and 996HM siloxane-
modified cyanate resin have been listed in Table 3.1 and Table 3.2. The typical physical

properties of cyanate resin composite and siloxane-modified cyanate resin composite

were listed in Table 4.1 and Table 4.2 [23]

Table 4.1 Nominal physical properties of CE composite provided by manufacturer

 

 

 

0° Tensile Strength 2 GPa
Modulus 290 GPa
90° Tensile Strength 27 MPa
Modulus 5.6 GPa
0° Compression Strength 407 MPa
Modulus 276 GPa

 

 

 

 

51

Table 4.2 Nominal physical properties of SMCE composite provided by manufacturer

 

 

 

0° Tensile Strength 2 GPa
Modulus 317 GPa

90° Tensile Strength 42 MPa

0° Compression Strength 890 MPa
Modulus 342 GPa

 

 

 

 

4.1.2 Hygrothermal Exposure

All hygrothemally-exposed samples were ground down to 1mm thickness. Specimens
having the dimension of 25 mm x 25 mm were cut from the laminate with a diamond
saw. The edges of the specimens were subsequently ground using 600 grit abrasive
paper to maintain consistently smooth edge surfaces. The ratio of edge surface area to
face surface area was 0.04 for all hygrothemally-aged specimens. Therefore, edge
sorption effects were assumed to be negligible. The specimens were conditioned by
heating at 100°C for 48 hours in the air to remove sorbed moisture on the surfaces and to
eliminate residual stress caused by sample fabrications. Specimens were then placed in
distilled water chambers at constant temperatures of 35°C, 65°C, 80°C and 95°C. The
temperature variation is 4:2 °C. Specimens were weighed periodically. Once the
specimens were taken out of the environmental chambers, the surface water was absorbed

using a clean dry filter paper. Then, the samples were left for 2 min at the ambient

52

temperature and humidity conditions before weighing. Then specimens were weighed
by using an analytical balance with 0.01 mg resolution. The experimental error of the
test is :t 0.02%. The percentage moisture absorbed in the resin materials is calculated
using equation 3.1

Optical microscopy, FTIR and environmental scanning electron microscopy
(ESEM) were used to investigate surface modification associated with cracking and mass

loss.
4.1.3 Three-point Bend Test

Three-point bend flexure tests were performed to study the effects of thermo-
oxidation on mechanical properties of both CE and SMCE materials. In performing the
flexure tests, ASTM D790-90 was used as a guide. A representative sample of six
specimens was prepared and tested for each data point that relates to different aging
temperatures. Tests were conducted using Instron® mechanical test machine with a 10
KN load cell. The nominal specimen dimensions used in this study were about 25 mm
long, 3-5 mm wide and 1 mm thick. A cross-head speed of 2mm/min was used in this
study. For every datum point, six specimens were tested and the average of these data

points was plotted. The flexure strength was calculated by equation 3.2.

53

4.2 Water Absorption Results and Discussion

4.2.1 Water-absorption Kinetics of Cyanate Based Polymer Materials

Hygrothermal aging experiments of CE polymer and composite materials shows
that the water-absorption curves follow F ick's law in the early stage, then the absorption
curves deviate from the F ickian behavior, and become non-Fickian in the late stage of

experiments.

Non-Fickian water-absorption of cyanate ester materials has been observed by
some researchers [19,34]. They suggested that non-Fickian water-absorption is caused
by chemical interaction or reaction of water molecules with cyanate ester polymer
network. The chemical interaction traps extra water into the polymer network, causing
samples keep gaining weight even after the saturation of water of polymer network. The

chemically absorbed water was accounted for the non-Fickan behavior of cyanate ester

polymers.

The chemical interaction also leads to hydrolysis of cyanate ester at high
temperature. Researchers reported [21,71] that cyanate ester laminates stored in humid
condition at high temperature encountered hydrolysis and blistering phenomena. They
found that the cyanate ester blisters much easier than epoxies, although cyanate ester
networks absorb less water than do epoxies. They suggested [67] that blistering is caused

by the hydrolysis of the polymer network that produces gaseous products.

54

To study the water molecules interaction and reaction with the cyanate ester
network, the FTIR spectrum of cyanate ester as-received and cyanate ester aged in the
95°C water for 3000b have been taken. The result is plotted in Fig. 4.1. The comparison
of FTIR spectrum of hygrothemally-aged and as-received samples shows that some new
absorption peaks appear (three peaks in the range of 3360-3340 cm’1 and a peak at 1745
cm") in the spectrum of cyanate ester aged in the water. The peak of the range 3360-
3340 cm'l is related to N-H groups and the 1745 cm'1 peak corresponds to C=O groups.

These new chemical groups have been shown to be the product of water molecules

reacting with -CEN groups of cyanate ester as shown in Fig. 4.2 [71].

The PT IR results suggest that the interaction which cause the non-Fickian water

absorption of cyanate ester materials is chemical interaction of water molecules with the

unlinked —CEN group of cyanate ester polymer network. The chemical interaction is
most likely to be the hydrogen bonding of water molecules with the unlinked —CEN

groups, given the fact that most organic compounds with unlinked -C‘='N groups tend to

form hydrogen bonds with waters.

55

ABSORBANCE

0007

 

U

a a (g
E a.

DJ
0
g g t -_ g
S
93 ' 6’
0
9..
(D
'1
a.
g a

O

5.

2

 

 

Fig. 4.1 FTIR spectra of cyanate ester resin aged in the water at 95°C for

3000hrs. The spectrum shows the increasing of 3360-3340 cm"1 (N-H group) and 1745

cm'1 (C=O group)

56

 

CH3
N=c—o+ ow:- _.
CH3

H201 H+

OH

I CH3
we too—11
CH3

1 Rearrangement

C”)

CH3
H2N— |C — O—O—Jf :H:O—O—C —NH2

Fig. 4.2 The reaction of cyanate molecule with water forming N-H groups and

C=O groups

57

4.2.2 Diffusivity in CB Based Polymer and Composites

In the Fig. 4.3-4.6, the weight change profiles are plotted for siloxane-modified
cyanate, siloxane-modified cyanate carbon fiber composite, cyanate ester and cyanate
ester based composite at the different temperatures. Symbols represent the experimental
data of water absorption. Each data point was based on the average data of three
samples. Because the ratio of thickness/width of the thin plate specimen was smaller
than 4%, calculations of diffusion parameters were determined using a one-dimensional
approach without incurring significant error.

As shown in Fig. 4.3, for siloxane—modified cyanate polymer, the water
absorption curves at 35, 65,80 and 95°C was Fickian in the early stage of absorption, then
became non-Fickian in the late stage of water absorption. The 95°C water absorption
curve shows a slight weight reduction after being immersed in water for 1000 hours. As
shown in Fig. 4.4, for siloxane-modified cyanate composite materials, the water
absorption curves at 35 , 65,80 and 95°C is Fickian in the early stage of absorption, then
becomes non-Fickian. After being immersed in water for 1000 hours, a significant
weight gain of samples aged at 95°C was observed.

As seen in Fig. 4.5, the water absorption curve of cyanate ester at 35, 65, 80 and
95°C is Fickian in the early stage of absorption, then becomes non-Fickian. At the late
stage of the absorption curve, the curves of samples aged at 80°C and 95°C diverge from
other samples. After being immersed in water for 1600 hours, the weight gain of
samples immersed at 80°C and 95°C water is even more. At the same time, some blisters

were observed inside the specimen.

58

 

l * +35dageeC +65dsgeeC

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

L -)(—mcbgeeC +%cbgeeC
25
2
o\° .
.g 1.5 /'_ ,1 '1
o:
.0—0
.c
.9 1-
a)
3
(15 .
0 I T r l l
0 1o 20 a) 40 a) 60

sqaemotdurrem‘”)

Fig. 4.3 Water absorption curves of SMCE at different environmental temperatures with

square root of time

59

 

+35 degree C +65 degree C
-)(-80 dggree C +95 degree C

I. /
WW /
, /

 

 

 

 

 

.0
0)
I

weight gain %

 

 

 

 

 

0.4 . .-.__. M
o 2 . M "=’ ‘—
‘1‘"
y
o 4' 4 f , . .
o 10 20 30 4o 50 so

square root of time (hm)

Fig. 4.4 Water absorption curves of siloxane-modified cyanate fiber composite at

different environmental temperatures with square root of time.

60

 

I +35degreec +65degreec

l

] -)(—80 deger +95 degreeC

 

 

 

 

 

 

 

2.5 l
2.0
o\°
.E 1.5 ~——v——
(U
U)
1“
'5:
'0.) 1 0 X¥ I
3 ’19"
_—H
05 t” __ 7 -

 

 

 

 

 

0.0 4 . . . .
o 10 20 30 4o 50
square root of time (hm)

Fig. 4.5 Water absorption curves of cyanate ester at different environmental

temperatures with square root of time.

61

 

+35 degree C +65 degree C
-)(-80 degree C +95 degree C

 

 

 

 

.3
A

 

—\
lo
1
I

 

1
l
l
l

 

 

l
t
l

.0
o:

 

 

weight gain %

 

 

.0
.3

 

.0
N

 

 

 

.0
o

0 1 O 20 30 4O 50

square root of time (hm)

Fig. 4.6 Water absorption curves of cyanate ester fiber composite ester at

different environmental temperatures with square root of time

62

 

60

As shown in Fig. 4.6, the water absorption curves of cyanate ester composite
materials at 35 , 65 and 80 °C are Fickian in the early stage of absorption, then become
non-Fickian. The curve of specimen aged at 95°C diverges form this behavior in the late
stage. A slight reduction in weight of the specimen has been observed after being

immersed in water for 1000 hours.

The comparison of water absorption profiles of CE and SMCE (Figs. 4.3 and 4.5)
shows that SMCE absorbed more moisture than the CE. This is contrary to the claim by
the manufacturer (Fiberite”)[23] that SMCE has lower moisture absorption than the CE.
It is contended the greater of moisture absorbed by SMCE in our experiment is due to the
thermo-oxidation degradation of SMCE when these specimens were conditioned in the
air at 100°C before water absorption tests were conducted(shown in 4.2). Thermo-
oxidation of the SMCE which involves chain scission of macromolecules and surface

degradation made the material more susceptible to moisture absorption.

Assuming the temperature and moisture distribution in the material are uniform
and following Fick's model in the early stage of diffusion, diffusivity of water in these

materials can be calculated by equation 4.1

2 2
h _
D=fl[_] [M] (4,1)
4M. J2 47.
All M, values used the saturation water absorption data before the water absorption

curves deviates from the Fickian behavior.

The diffusivity calculated is list in Table 4.3-4.6.

63

Table 4.3 Diffusivity of CE resin

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Temperature (°C) 35 65 80 95
D(mmzs‘l) 6.1x10'7 7.9 1110'7 9.2 x10'7 1.2 x10'6
Table 4.4 Diffusivity of CE composite
Temperature (°C) 35 65 80 95
D(mmzs") 8.3 x10'8 1.6 x10’7 2.2 x10'7 2.5 x10'7
Table 4.5 Diffusivity of SMCE resin ester
Temperature (°C) 35 65 80 95
D(mmzs") 2.88 x10'7 6.84 1110'7 1.32 1110*6 2.64 x10'6
Table 4.6 Diffusivity of SMCE composites
Temperature (°C) 35 65 80 95
D(mmzs'l) 3.2 x10’8 7.5 x10'8 2.2 x10'7 5.1 1110‘7

 

 

 

 

 

64

 

 

 

 

4.2.3 Activation Energy of Water Diffusion in Cyanate Based Polymer Materials
The Arrhenius plots of the logarithms of the diffusion coefficient versus 1/T for
the neat resin and fiber composite are shown in Fig. 4.7 and Fig. 4.8. Solid circles
represent the data of the neat resin and solid squares represent data of the composite.
The activation energy of diffusing was calculated by equation 4.2 by the slope of

the Arrhenius plots.

ln(D) = ln(Do) — R—QT- (4.2)

where Do is the diffusion constant, Q is the activation energy.

Results show that the activation energy for the water diffusion in CE and CE
composites is about 10 kJ and the activation energy for the water diffusion in SMCE and
SMCE composites is about 24 kJ. The larger the activation energy, the more difficult for
water to diffuse in polymer materials. Researchers reported [17-19] that polysiloxane
molecules are hydrophobic. When the hydrophobic polysiloxane group is connected to
the cyanate ester polymer network, the whole polymer network becomes more
hydrophobic. Therefore, water diffusion is more difficult in the siloxane-modified
cyanate ester than in a pure cyanate ester polymer network

The slopes of the Arrhenius plots of both SMCE resin and SMCE composite (Fig.
4.7) and CE resin and composite (Fig. 4.8) are almost same, indicating that water
diffusion in both resin and composite occurs by the same diffusion process. The

diffusion occurs through the bulk and/or the fiber-matrix interfacial region.

65

 

 

 

 

 

 

 

 

 

333

 

 

 

 

 

 

 

 

O cyanate ester . l
I cyanate ester composnte ,
A 98199932291392 8%?99 on equatIon4-3 l
-1.8
-2
-2.2 «—
M
-24 a
’u?
\
N a-
E 2.6
v ‘2.8 r . 1 1 1 1
a 0.0326 0.0027 0.0028 0.0029 0.003 0.0031 0.0032 0.0
a) -3
2 h _ .I _. !
_3.2 l“ —— —----g_
-3.4 L—
-3.6
-3.8

Fig. 4.7 The Arrhenius plots of the logarithms of the diffusion coefficient versus

1rr (16‘)

NT for cyanate ester and fiber composite.

66

 

 

 

 

 

 

 

 

 

[ O siloxane-modified cyanate
1 I siloxane-modified cyaoomnate uggosites
] A calculated data_ based on eq
-2
-2.2 ]~
0
_2 4 \

-2:6 \

 

 

0.0326 QMZB 0.0029 0.003 00031 00032 00333
-3.2 9 .

-3.4 ____._22__

log D (mmzls)
. 1'0
0’ m
. ‘4/

 

 

 

 

 

 

 

 

 

1fl'(k'1)

Fig. 4.8 The Arrhenius plots of the logarithms of the diffusion coefficient versus

UT of siloxane-modified cyanate ester and fiber composite

67

 

 

The transverse difiusion coefficient of the composite can be estimated from the diffusion
coefficient of the resin, if all the fibers are parallel to the surface through which the

moisture passes. Shen et al. [72] derived the expression:

V
D. = 11(1— 2,/—’ J (4.3)
71'

Where, Dc and Dr are the diffusion coefficients of composite and resin respectively. VfiS

the volume fraction of carbon fibers.

The diffusion coefficients of both CE and SMCE composites were calculated
from the diffusion coefficients of resins listed in Tables 4.3 and 4.5 using equation 4.3.
The calculated data was plotted in the Figs. 4.7 and 4.8 to compare with the measured
data of the diffusion coefficient of both CE and SMCE composites. The solid triangle in
the Figs. 4.7 and 4.8 with dash lines represent the calculated data. Both Fig. s show that
the agreements between measured and calculated values are quite close. The same
agreement has been observed by several authors [15,73]. The agreement has been
explained as the water diffusing only through the polymer matrix without any diffusing
through the fiber-matrix interface in the composite. In some polymer and composite
systems, researchers found large discrepancies [15,74]. The large discrepancies reported
in the literature were explained by the degradation of fiber-matrix interface and the high
diffusivity along poorly bonded fiber-matrix interfaces in the composite. The agreement
of our data indicates that water diffusion in both CE and SMCE resin and composite
occurs by the same diffusion process. The water diffusion in CB and SMCE composites

is mainly through the matrix.

68

 

4.3 Hygrothermal Degradation of Cyanate Based Polymer Materials

The weight loss of SMCE at the 95°C shown in Fig. 4.3 can be attributed to the
surface degradation of specimen. Fig. 4.9 is an ESEM photograph which compares the
surface morphology of a sample aged in water for 3000b at 95°C with the surface of a
non-aged sample. The micrography shows that some of the surface area of aged
specimen was damaged by hot water. Some surface materials have been etched off the
sample surface. The weight loss of cyanate ester composite materials at 95°C shown in
Fig. 4.6 could be attributed to the observed surface degradation.

Fig. 4.10 shows the surface morphology of a CB composite specimen aged in
water for 3000b at 95°C, compared with the surface of non-aged sample. The
photograph shows that some surface area of the composite aged at 95°C was damaged.
The damage occurs mainly at the resin rich region, and some micro-cracks were also
observed on the damaged surface. Similar surface degradation [5] has been observed in
hygrothermally aging of Graphite/Epoxy composite at 90°C.

The weight gain of SMCE composites at 95°C can be attributed to the
delamination of composite laminate. Fig. 4.11 is an ESEM photograph of cross-section
of a SMCE composite sample aged at 95°C compared with a non-aged sample. The
cross-section photograph shows that the delamination happens inside the specimen. The
delamination forms some large cracks inside of the specimen. These cracks could trap a

large amount of water and cause a large increase of the sample weight as observed in Fig.

4.4.

69

 

   

(b)

Fig. 4.9 Surface morphology of aged siloxane-modified cyanate ester @95°C for
3000hrs (a) as-received, and (b) sample aged in water

70

 

(b)
Fig. 4.10 Surface morphology of aged cyanate ester composite@95°C for 3000hrs
(a) as-received, and (b) sample aged in water (The basket-weave imprint shown in both
Figs. is due to the breather cloth used during processing of panels)

71

 

 

 

Fig. 4.11 Cross-section of aged siloxane-modified cyanate composite @95°C for

3000h (a) as-received, and (b) sample aged in water

72

4.4 Three-point Bend Test on Hygrothermally-Exposed Materials

To determine the hygrothermal effects on the strength of cyanate based polymer
materials, three-point bend test was used to determine the flexure strength of these
materials after exposed to water. Fig. 4.12 shows the results of three-point bend tests of
hygrothermally-aged CE and SMCE. The graph shows a decrease of the flexure strength
of SMCE with increasing aging temperatures. Afier aged in the air at 95°C for 3 months,
the flexure strength of SMCE decreased almost 70%. The graph also shows a decrease
in the flexure strength of CE with increasing aging temperatures. The flexure strength of
cyanate ester decreased 20% after aged in the air at 100°C for 3 months.

During hygrothermal aging of SMCE and CE, especially in the high temperatures,
hydrothermal degradation of cyanate ester and polysiloxane chains occurred. The flexure
strength of these materials decreases when their chains break. The significant decrease of
the flexure strength of SMCE at 80°C and 95°C could be attributed to the surface
degradation of SMCE observed in Fig. 4.10 and thermo-oxidation of SMCE discussed in

chapter 3.

73

 

l D siloxane-modified cyanate ester
I cyanate ester

 

 

 

1 80
160
140
120
1 00
80
60
40
20

as-received

 

 

 

 

35°C

 

 

65°C

 

 

 

 

 

80°C

 

 

95°C

 

 

Flexure strength ( MPa)

 

 

 

 

 

 

 

 

     

 

 

 

Fig. 4.12 The results of three-point bend tests of the cyanate ester and siloxane-

modified cyanate ester for 3000 hrs

74

Fig. 4.13 shows the results of the three-point bend test of the CE composites and
SMCE composites after aged in the water at 100°C for 3 months. The graph shows no
significant change of the flexure strength of the cyanate based composite materials when
the aging temperature is lower than 95°C. The flexure strength of the CE composites and
. SMCE composites decrease significantly only at the high aging temperature (95°C). The
flexure strength of SMCE composite decreased by almost 30% and the flexure strength of
CE composite decreased by almost 20%.

The flexure strength of composite materials is mainly controlled by the strength of
fibers and fiber-matrix interface [74]. As long as no environmental degradation of the
fiber-matrix interface and fibers occurs, the flexure strength of composite materials
should not affect significantly by the variance of environmental temperatures even there
could be a degradation of polymer matrix. Fig. 4.13 shows no significant change of the
flexure strength of the cyanate based composite materials when the aging temperature is
lower than 95°C. When the temperature is above 95°C, thermal degradation of siloxane-
modified ester and hydrolysis of cyanate ester polymer takes place. These effects could
cause a significant decrease of the strength of the fiber-matrix interface of these
composite materials. The degraded interface can cause the decrease of the flexure

strength of SMCE and CE composites.

75

 

El siloxane-modified cyanater ester composite
I cyanate ester composite

 

 

 

 

.A-t
QM

 

as—reoeived 35°C

 

 

 

6 5°C 80°C

4...;
N¥O§

 

 

 

 

 

95°C

 

 

 

 

 

 

 

 

 

 

 

 

.09
0500

 

 

F lexure strength ( GPa)

.0
A

 

.0
N

 

 

 

 

 

 

 

 

   

0

 

Fig. 4.13 The results of three-point bend test of the CE composites and SMCE

composites for 3000 hours

76

4.5 Summary/Conclusions

o Gravimetric measurement shows that the water diffusion curves of cyanate ester

based polymer is non-Fickan type.

0 Hygrothermal degradation of CE based polymer and composite materials have
been observed at high temperature (95°C) after immersing in water for more than 2000
hours. Surface degradation of SMCE resin and CE composites was observed. Hydrolysis
of cyanate ester polymer network with water molecules was also observed. Delamination

of SMCE composites was also observed.

0 The activation energy shows that water diffusion occurs mainly through the

polymer matrix in the composite materials.

0 The three-point bend test shows that the strength of CE based polymer materials
decreases with increase aging temperature, and the strength of CE based composite

materials is affected significantly by moisture at high temperatures.

77

Chapter V

CONCLUSIONS AND RECOMMENDATIONS

5.1 Conclusions
The major results of this dissertation are concluded as following:

o The gravimetric experiment shows that siloxane-modified cyanate ester
gain weight when aged in the air at 100°C. Cyanate ester resin gains no weight even
after a long period of time.

o DSC testing shows that the glass transition temperature of aged siloxane-
modified cyanate decreases about 20°C after aged in the air at 100°C for 40 days. The
three-point bend tests show that the flexure strength of siloxane-modified cyanate
decreases 20% after being aged in the air for 3 months. These results indicate that the
polymer network degrades through thermo-oxidation process and that this process is
controlled by oxygen diffusion in the polymer.

0 FITR results show that the therrno-oxidation process starts with the
oxidation of C-H bonds of the polysiloxane side groups. The reaction propagates with
the breaking of the Si-O-Si chain and the formation of carbonic compounds.

0 Gravimetric measurement shows that water diffusion in cyanate ester
based polymer is non-Fickan type. The activation energy shows that water diffusion is
only through the polymer matrix in the composite materials.

0 Degradation of polymers and composites occurs at high temperature

(95°C) after being immersed in water for more than 2000 hours. Three-point bend tests

78

 

show the strength of polymers decreases with increases in aging temperatures, and the
strength of composites are affected significantly by moisture at high temperatures.
5.2 Recommendations
Many methods, such as F TIR, DSC, ESEM and three-point bend test, have been

used to characterize the environmental effects on cyanate based materials. Also, more
experiments are needed for a more complete understanding of environmental effects on
cyanate based materials.

0 Study therrno-oxidation effect on the mechanical behavior of cyanate
composite, especially the effect on matrix-fiber interface.

0 Study thermo-oxidation of cyanate based polymer and composite materials *
in different temperature (from 80°C to 170°C).

0 Hygrothermal effect on the delamination (Model I and Model H ) of

composites.

79

 

10.
11.
12.
13.

14.

15.

16.

BIBLIOGRAPHY

. D.Hull, An Introduction to Composite Materials, Cambridge University, London,

U.K., 1981.

Springer, G.S. (ED.), Environmental Effects on Cogmposite Mgterials, v 1, Technical
Publ. Co., Lancaster, PA, 1981.

Springer, G.S. (ED.), Environmental Effects on Composite Ma_terials, v 2, Technical
Pub]. Co., Lancaster, PA, 1984.

Springer, G.S. (ED.), Environmental Effects on Composite Materials, v 3, Technical
Pub]. Co., Lancaster, PA, 1988.

Zhou, J. and Lucas, J.P., Journal of Thermoplastic Composite Materials, v 9, n 4, Oct
1996, p. 316-328.

Lucas, J .P. and Zhou, J., JOM, December 1993, p. 37-40.

Lucas JP. and Odegard, B.C., A_dvances in Thermopla_stics Composite. STP1044,
Newaz 6., ed., 1989, p. 231.

Wang, CS. and Chang,G.C., The effect of moisture absorption on composite
laminates, Composites Structures, v 2, 1983, p. 260.

Joshi, O.K., The effect of moisture on the shear properties of carbon fiber composites,
Composites Structures, v14, 1983, p. 196.

Wolff, E.G., SAMPE JOURNAL, v 29(3), 1984, p. 11.
DeNeve, D. and Shanahan, J.L., Polymer, v 34, 1993, p. 5099.
Adamson, M.J., Journal of Materials Science, v 15, 1980, p. 1736.

Grayson, MA and Wolf, C.J., Proceedings of 5‘° International Conference on
Composite Materials, San Diego, 1985, p. 1463.

Shimp, D.A., Christenson, J .R. and Ising, S.L, Proceedings of 34’” International
SAMPE Symposium, 1989, p. 222-233.

McConnell, V.P., Advance Composite, v 7(3), 1992, p. 28.
Speak, S.C., Sitt, H. and Fuse, R.H., Proceedings of 36'” International SAMPE
Symposium, 1991, p. 336.

80

17.

18.

19.

20.
21.

22.

23.

24.

25.

26.
27.

28.

29.

30.
31.
32.

33.

Parvatareddy, H., Wang, 1.2. and Ward, T.C., Composites Science and Technology, v
53, 1995, p. 399.

Parvatareddy, H., Wilson, RH. and Dillard, D.A., Composites Science and
Technology, v 56, 1996, p. 1129.

Cinquin, J. and Abjean, P., Proceedings of 38” International SAMPE symposium,
1993,p.1539.

Blair, C. and Zakrzewski, 1., Design of optical Instruments, SPIE 1690, 1992, p. 300.

Lee, F.W. and Lefebvre, P., Proceedings of 3 4’” International SAMPE symposium,
l990,p.162.

David A S., Cyana_tes, Guide to Engineering Pla_stics Families: Thermosetting Resins.
Elsevier Scientific Publishing Company, Amsterdam, 1992.

F iberite data sheet of cyanate ester and cyanate siloxane resin from HEXCEL,
October 15, 1995.

Arnold, C., Mackenzie, P., Malhotra, V., Pearson, D., Chow, N., Hearn, M. and
Robinson, G., Society for the Advancement of Material and Process Engineering
(USA), 1992, p. 128-136.

Zolotova, NY. and Denisov, Y.T, Journal of Polymer Science, A1(9), 1971, p. 3311.
Iring, M. and Kelen, T., Polymer degradation and Stability, v 1, 1979, p. 287.

Holrnstrom, A. and Sorvik, E.M., Journal of Applied Polymer Science, v 18, 1974, p.
761-779.

Kim, J.H. and Kwei, T.K., Chemical Engineering Communications, v 116, 1992, p.
105-116.

Schnabel, W. Polymer Degradation: Principle and practical applications, Elsevier
Science Ltd Oxford Engl ,1981.

Kelen, T., Polymer Degradation. Elsevier Science Ltd Oxford Engl,1982.

 

Gugumus, F., Polymer Degradation and Stability, v 52, n 2, 1996, p. 159-170.

Audouin, L., Gueguen, V., Tcharkhtchi, A. and Verdu, J ., Journal of Polymer
Science, Part A: Polymer Chemistry, v 33, n 6, p. 921-927.

Chien, J.C. and Boss, C.R., Journal of Polymer Science, A1(5), 1967, p. 3091.

34. Celina, M., Ottesen, D.K., Gillen, K.T. and Clough, R.L., Polymer Degradation and

Stability, v 58, n 1-2,1997, p. 15-31.

81

 

35. Yoshitaka, T. and Seikichi, T., Journal of Applied Polymer Science, v 42, 1991, p.
2811-2817.

36. Nam, JD. and Seferis, J.C., SAMPE Quarterly, v 24, n 1, Oct 1992, p. 10-18.

37. Hinsken, H., Moss, 8., Pauquet, J .R. and Zweifel, H., Polymer Degradation and
Stability, v 34, n 1-3,1991, p. 279-293.

38. Rao, R.M., Shylaja, H.V. and Suresh R., Journal of Reinforced Plastics and
Composites, v 14, n 5, May 1995, p. 513-522.

39. Chateauminois, A., Vincent, L., Chabert, B. and Soulier, J .P., Polymer, v 35, n 22,
1994, p. 4766-4774.

40. Cai, L.W. and Weitsman, Y., Journal of Composite Materials, v 28, 11 2,1994, p. 130-
154.

41. Lee, MC. and Nikolaos A.P., Journal of Applied Polymer Science, v 47,1993,
p.1349-1359.

42. Kelly RN. and Bueche, F ., Journal of Polymer Science, L(1961), p. 549.
43. Mijovic, J. and Weinstein, S.A., Polymer Communication, v 26,1985, p. 237.

44. Zhou, J. and Lucas, J .P., Journal of Thermoplastic Composite Materials, v 9, n 4, Oct
1996, p. 316-328.

45. Ma, C.M., Huang, Y.H., Chang, M. and Sheu, M.F., Conference Proceedings In
Search of Excellence 49th Annual Technical Conference -ANT EC '91, v 37, May 5-9,
1991, p. 2092-2096.

46. Rice, BF. and Lee, C.W., Proceedings of the 1997 29th International SAMPE
Technical Conference, v 29, 1997, p. 675-685.

47. VanLandingham, M.R., Eduljee, RF and Gillespie, J .W., Proceedings of the 1995
Symposium on High Temperature and Environmental Eflects on Polymeric
Composites, v 1302, Nov 13 1995, p. 50-63.

48. Xiao, G.Z. and Shanahan, M.E., Journal of Applied Polymer Science, v 69 n 2,1998,
p. 363-369.

49. Morgan, R.J., Shin, E.E., Lincoln, J.E., Choi, J. and Lee, A., Proceedings of the 28’”
International SAMPE Technical Conference, v 28, Nov 4-7 1996, p. 213-224.

50. Mohd Ishak, Z.A. and Berry, J .P., Journal of Applied Polymer Science, v 51, n 13,
1994, p. 2145-2155.

51. Haque, A., Mahmood, S., Walker, L. and Jeelani, S., Journal of Reinforced Plastics
and Composites, v 10, n 2, Mar 1991, p. 132-145.

82

 

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

65.

Uschitsky, M. and Suhir, E., Proceedings of the 1997 ASME International
Mechanical Engineering Congress and Exposition, v 21, Nov 16-21, 1997, p. 141-
1 66.

Bouadi, H. and Sun, C.T., Journal of Materials Science, v 25, n 1B, 1990, p. 499-505.

Ma, C.M., Lee, C.L, Chang, M.J. and Tai, N.H., Polymer Composites, v 13, n 6, Dec
1992, p. 448-453.

Vauthier, E., Chateauminois, A. and Bailliez, T., Polymers and Polymer Composites,
v 4, n 5, 1996, p. 343-351.

Komai, K., Minoshima, K. and Shiroshita, 8., Proceedings of the Joint Japan- USA
Seminar on Advanced Materials for Severe Service Applications, A143, 11 1-2, 1990,
p. 155-166.

Shirrell, C.D., Advance Composite Materials. —Environmental Eflects, Vinson J .R.
ed., ASTM STP 658, p. 21.

Obst, A.W., VanLandingham, M.R., Eduljee, R.F., Gillespie, J .W., Griesheim, GE.
and Tosi, K.F., Proceedings of 28'” International SAMPE Technical Conference, v
28, Nov 4-7,1996, p. 34.

Wang, Q. and Springer, G.S., Journal of Composite Materials, v 23, 1989, p 434-447.

Hoa, S.V., Lin, S. and Chen, J .R., Journal of Reinforced Plastics and Composites, v
11, n1, Jan 1992, p. 3-31.

Steven J .H and Rarnaswamy, 8., Composite Materials: F atigpe and Fracture, v 4,
ASTM STP 1156, 1993, p. 318.

Mohd Ishak, Z.A. and Lim, N.C., Polymer Engineering and Science, v 34, n 22, Nov
1994, p. 1645-1655.

Kaelble, D.H, Dynes, P.J. and Crane, L.W., Journal of Adhesion, n5, 1975, p. 211.

. Morii, T., Ikuta, N., Kiyosumi, K. and Hamada, H., Proceedings of the 1996 6”

International Conference on Composite Interfaces, ICCI-6, May 5-8 1996, v 57, n 8,
p. 985-990.

Meyer, M.R., Latour, R.A. and Shutte, H.D., Journal of Thermoplastic Composite
Materials, v 7, n 3, July 1994, p. 180-190.

66. Sohn, D.W. and Sung, N., Proceedings of the ACS Division of Polymeric Materials:

67.

Science and Engineering, Apr 1990, v 62, 1990, p 43-47.

Iring, M., Szesztay, M., Stirling, A. and Tudoes, F., Journal of Macromolecular
Science - Pure and Applied Chemistry, A29, n 10, 1992, p. 865-884.

83

 

 

68. Song, J ., Fischer, CH. and Schnabel, W., Polymer Degradation and Stability, v 36, n
3, 1992, p. 261-266.

69. Gedde, U.W., Viebke, J ., Leij strom, H. and Ifwarson, M., Polymer Engineering and
Science, v 34, n 24, Dec 1994, p. 1773-1787.

70. Abdel-bary, EM. and Dessouki, A.M., Journal of Applied Polymer Science: Applied
Polymer Symposium, v 55, 1994, p. 37-46.

71. Leo J .K., Idelette, H., Christopher W.M. and David A.S., Journal of Applied Polymer
Science, v 64, 1997, p. 107.

72. Shen, CH and Springer, G.S., Journal of Composite Materials, v 10, 1996, p. 1.
73. Zhang,Q. and Morgan, R.J., Journal of Composite Materials, v 27, 1993, p. 1465.

74. Thomason, J .L., Composites, v 26, 1995, p. 467.

84

nrcurcnu STATE UNIV. LIBRARIES
Illll1111111111111111HIllllllllllllllllllllHI
31293018345995