u .

.
.
5774....

it; a

133.31%; . .

« !\c?.§rlé. .3
. 2.11 .r.

.

‘31. sutlli~
:3...‘ ,u.v1\s

a. v.5.
,

 

\ ‘
zlix;

I
.:
ad...

L )«fil... .3}.

.3.
x
1 3L.
....l:‘{ .
in: 4..) Ii is
.315?! étrhgikziaxf. ,
94 . . . ... flwtnxvn.2 .l. r
4.9... , .. .42....

. we???»
2... 3f » .2! .. .. .
‘ .3. I . ,y I

 

. Hz... a r. .. 2% 2.. \ti’ufl . .ififi‘ii‘.
‘ Am. i .13 V.) .. , . ‘ $1.}? f Li‘s.“ “in T ‘ . V

 

 

 

G 1 m! LIBRARIES

lxllglllllglllfiliill\wm u ll

 

 

 

 

 

 

 

 

 

 

3 00887 6504-

 

I":

LIBRARY
Michigan State
Unlvcrslty

 

 

 

This is to certify that the
thesis entitled

Continuous Processing of Polymers and
Composites With Microwave Radiation

presented by
Min Lin

has been accepted towards fulfillment
of the requirements for

_Mas_t.m:$__degree in Qhemical Engineering

 

./" f

/

Major professor

Datemr 1. 1993

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

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before dde due.

    

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ll

 

[ l

MSU Ie An Affirmative ActIorVEqueI Opportunlty InetItutIon
chfi-pfl

 

CONTINUOUS PROCESSING OF POLYNERS AND CONDOSITES
RITE NICROIAVE RADIATION

BY
Kin Lin

AN ABSTRACT OF A THESIS
Submitted to
MICHIGAN STATE UNIVERSITY
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Chemical Engineering

1993

Research Advisor
Dr. Martin C. Hawley

ABSTRACT

CONTINUOUS PROCESSING OF POLYMERS AND CONPOSITES
IITR NICROUAVE RADIATION

BY
Min Lin

Continuous microwave processing is a new approach for
microwave processing applications to polymers and
composites. This processing method has the potential to
process long and large profile parts which are highly
desired in industries. The continuous microwave processing
technique was applied to curing of a heat cable and
pultrusion of composites in this study.

A cylindrical tunable microwave cavity was modified to
perform the continuous processing. The curing of the heat
cable was studied at different cavity positions and cable
orientations. The microwave leakage was within safety
limits during processing even with conducting wires in the
processed cable. Microwave transparent dies were designed
to study the microwave pultrusion of glass/vinyl ester
composites. The possibility of microwave pultrusion of
composites was demonstrated. A process model for microwave
pultrusion was also developed to study the curing of the
composites. The model simulation agreed with experimental

results very well.

To my parents

and my wife, Dong

iii

ACKNOWLEDGEMENTS

I would like to thank Dr. Martin C. Hawley for his
support and guidance throughout the completion of this
thesis. I would also like to thank Dr. Jianghua Wei for his
valuable discussions and suggestions, thank Larry A. Fellows
for his proofreading of the thesis and valuable suggestions,
and thank Richard Delgado, Valerie Adegbite and Dhulipala
Ramakrishna for their help through the duration of this
study.

The most important thanks goes to my wife, Dong, for

her love, support and encouragement.

iv

TABLE OF CONTENTS

CHAPTER 1 INTRODUCTION . . . . . . . . . . .
1.1 Microwave Processing . . . . . . . .

1.2 Microwave Heating Mechanism . . . .

1.3 Microwave Distributions Inside a Cavity

1.4 Microwave Heating Advantages . . . .
1.5 Microwave Applicators . . . . . . .
1.6 Microwave Processing Techniques . .

1.7 Research Scope and Goals . . . . . .

CHAPTER 2 CONTINUOUS MICROWAVE PROCESSING OF HEAT

2.1 Introduction . . . . . . . . . . . .
2.2 Experiments . . . . . . . . . . . .
2.3 Results and Discussions . . . . . .
2.4 Conclusion . . . . . . . . . . . . .

CHAPTER 3 APPLICATION OF CONTINUOUS MICROWAVE
TO PULTRUSION . . . . . . . . . . .

3.1 Introduction . . . . . . . . . . . .
3.1.1 Pultrusion . . . . . . . . .
3.1.2 Microwave Processing . . . .

3.2 Experimental . . . . . . . . . . . .
3.2.1 Microwave pultrusion setup .
3.2.2 Material Preparation . . . .

3.2.3 Reaction Conditions . . . . .

PROCESSING

10

11

14

14

17

23

29

3O

3O

3O

33

35

35

39

42

3.2.4 Temperature Measurement . . . . . . .

3.2.5 Extent of Cure Measurement . . . . .

3.2.6 Fiber Fraction Measurement . . . . .

3.3 Results and Discussions . . . . . . . . . .

3.4 Conclusions . . . . . . . . . . . . . . . .
CHAPTER 4 PROCESS MODEL FOR THE MICROWAVE PULTRUSION
4.1 Introduction . . . . . . . . . . . . . . . .

4.2 Kinetics of resin polymerization . . . . . .

4.3 Composite Material Properties . . . . . . .
4.3.1 Literature Data . . . . . . . . . . .

4.3.2 Cp Measurement . . . . . . . . . . .

4.4 Microwave Power Absorption . . . . . . . . .

4.5 Simulation of Microwave Pultrusion Process .

4.6 Results and Discussions . . . . . . . . . .

4.7 Conclusions . . . . . . . . . . . . . . . .
CHAPTER 5 SUMMARY OF RESULTS . . . . . . . . . . . .
CHAPTER 6 FUTURE WORK . . . . . . . . . . . . . . . .

LI ST OF REFERENCES O O O O O O O O O O O O O O O O O

vi

43

45

45

48

55

56

56

57

62

62

64

67

69

72

79

80

83

86

Table

Table

Table

Table

Table

Table

Table
Table

Table

3.1

3.2

LIST OF TABLES

Heating Modes for Cable in Horizontal
Direction . . . . . . . . . . . . . . .

Curing Results at Different Pulling
Speeds O O O O O O O O O O O O O O O O O

Heating Modes for Cable in Vertical
Direction O O O O O O O O O O O O O O O

Heating Modes for the Cavity with Teflon

Die O O O O O O O O O O O O O O O O O O

Heating Modes for the Cavity with
Ceramic Die . . . . . . . . . . . . . .

Kinetics parameters for vinyl
ester/vinyl toluene polymerization . . .

Dielectric Properties of the Materials .
Heat Capacity Measurement . . . . . . .

Parameters for the Process Model . . . .

vfi

23

26

26

48

51

60

63

66

75

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure
Figure

Figure

Figure
Figure
Figure

Figure

LIST OF FIGURES

Resonant Mode Chart for 17.78 cm Diameter
Microwave Cavity . . . . . . . . . . . . . 5

Material Alignments with the Electric Field
Inside the Cavity . . . . . . . . . . . . 6

Microwave Cavity for Pultrusion
Application O O O O O O O O O O O O O O O 16

Microwave Circuit for Pultrusion
Application O O O O O O O O O O O O O O O 18

Sideview of Microwave Cavity with

Positions for Temperature Measurement on
Cable at Horizontal Direction . . . . . . 21
Sideview of Microwave Cavity with

Positions for Temperature Measurement on
Cable at Vertical Direction . . . . . . . 22

Heating Results for Cable at Horizontal
Direction O O O O O O O O O O O O O O O 24

Heating Results for Cable at Vertical
Axial Direction . . . . . . . . . . . . 28

A Sketch of Typical Pultrusion Machine . . 32

Sideview and Cross-Section of the Microwave
Pultrusion Die . . . . . . . . . . . . . . 37

Sideviews of Microwave Cavity with

Pultrusion Die . . . . . . . . . . . . . . 38
Prepregging Process . . . . . . . . . . . 40
DSC Measurement for Vinyl Ester Resin . . 41

Heat of Reaction Measurements Using DSC
for Vinyl Ester Resin . . . . . . . . . 42

Top view of the Microwave Cavity with
Positions for Temperature Measurement on

Iliii

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

4.3

4.6

Composites . . . . . . . . . . . . . . . . 44

Microscopic Image of the Microwave Pultruded
Glass/Vinyl Ester Composite Sample . . . . 47

Heating Tests at Different Modes for
Glass/Vinyl Ester Composite with
Microwave Input of 30 W . . . . . . . . 49

Heating Results at Mode 4 for
Glass/Vinyl Ester Composite with
Microwave Input of 60 W . . . . . . . . 51

Curing Results at Mode 4 for Glass/Vinyl
Ester Composite with Microwave Input of
100 w O O O O O O O O O O O O O O O O O 53

Curing Results for Glass/Vinyl Ester
Composite at Different Pultruding
Speeds . . . . . . . . . . . . . . . . . 54

Chemical Structures of the Resin . . . . 58

Heat Capacity Measurement Using DSC for
Microwave Pultruded Glass/Vinyl Ester
Composite Sample . . . . . . . . . . . . . 65

Modeling Configuration for Microwave
Pultrusion Process . . . . . . . . . . . 69

Modeling of the Microwave Pultrusion
Processing of Glass/Vinyl Ester
Composite . . . . . . . . . . . . . . . 72

Simulation of Curing and Temperature
Profiles of Glass/Vinyl Ester Composite
at the Pultruding Speed of 1 cm/min . . 73

Simulation of Curing and Temperature

Profiles of Glass/Vinyl Ester Composite

at the Pultruding Speed of 2 cm/min . . 74
Simulation of Curing Profiles of

Glass/Vinyl Ester Composite at Higher
Pultruding Speeds . . . . . . . . . . . 76

Simulation of Temperature Profiles of

ix

Glass/Vinyl Ester Composite at Higher
Pultruding Speeds . . . . . . . . . . . 77

Figure 6.1 A Sketch of Novel Microwave Pultrusion System
with Part of the Cavity Wall Served as
Pultrusion Die . . . . . . . . . . . . . . 85

CHAPTER 1

INTRODUCTION

1.1 Microwave Processing
Microwave heating is used as an alternative to
conventional thermal heating for the processing of

*“. Microwave heating process is achieved by

materials
placing a material in an electromagnetic field. This
heating process has been widely used in various applications
such as food heating, drying, material processing, and some
biomedical researchesumh The materials being processed
using microwave includes polymers, composites, and ceramic
materials. The general advantages of microwave heating
include: (1) fast, direct and outward heating, (2) better
control of material temperature profile and input power to
optimize the material processing, (3) selective heating
based on the lossy magnitude of constituents in a material,
and (4) improved mechanical properties for microwave
processed parts.
1.2 Microwave Heating Mechanism

The microwave heating takes the advantage of dielectric
properties of these materials. The frequency of a dipolar

oscillation in a material is in the microwave range. This

makes it possible for microwave energy to excite a dipole

1

2
group to its higher energy state. The heating mechanism can
described as follows. In microwave heating, the dipoles of
molecules in a material align themselves with the applied
electric field, and the relaxation of the dipoles causes a
friction among molecules. This friction will cause a
temperature rise of the material. As a result, microwave
energy is converted into thermal energy of the material.
Furthermore, the microwave heating is initiated from
molecule-dipole interactions. In the processing of
materials, this heating mechanism will lead to microwave
heating advantages over conventional thermal heating which
is achieved by conduction and convection.

The dielectric properties of a material are described
by the dielectric constant 6', and the dielectric loss
factor 6". The dielectric constant 6’, which represents the
electrical polarizability, is defined as the ratio between
the capacitance of a condenser filled with dielectrics and
the capacitance of the same condenser when empty. The
dielectric constant also reflects the ability of a material
to store electrical energy. The dielectric loss factor, 6”,
describes the molecular relaxation phenomenon which reflects
the ability of a material to dissipate electrical energy.
The conversion of microwave energy to thermal energy is
characterized by the material's dielectric loss factor. Ku
and Liepinl7 reported the results of heating materials with
different dielectric loss factors. The microwave power or

the time required to heat a material increases as its

dielectric loss decreases.
The microwave power absorption rate for materials can

be calculated using Poynting's Theorem“:

_ 1 , u, (1.1)
a ” = a ” + —3 (1.2)
of d 800
where w is the frequency of the electromagnetic waves in

rad/sec,

E¢W and EW are the effective dyadic loss factor

and relative dyadic loss factor of materials
respectively,

a is the dyadic conductivity in S/m, and
i and I?" are the electric field vector and its

conjugation in V/m.
The electric field can be calculated theoretically for
a empty cavity with boundary conditions of a given shape of
cavity. However, when a cavity is loaded with dielectric
lossy materials, the electric field inside the cavity will
be disturbed. The calculation of microwave absorption by

the material becomes very complicated.

4

1.3 Microwave Distributions Inside a Cavity

Microwave distributions inside a cavity are different
for different resonant conditions. Figure 1.1 shows the
microwave resonant modes at different cavity length at
different microwave frequency. These resonant modes can be
divided into two categories: Transverse Electric(TE) modes
.or Transverse Magnetic(TM) modes. In TE modes, the electric
fields are aligned perpendicular to the axis of the cavity.
In TM modes, their are aligned parallel to the axis of the
cavity. As a result, the set-up of the material inside a
cavity is very important for an effective microwave
processing. For an optimum processing result, processing
materials should be aligned along the electric field as much
as possible. Figure 1.2 shows the alignment at both
microwave resonant modes for the material inside a microwave
cavity. The alignment of the material will give the maximum
interactions between materials and microwave fields.
1.4 Microwave Heating Advantages

The difference in the energy dissipation into a
material between microwave heating and conventional thermal
heating lies in the different heating mechanisms. In
conventional thermal heating, a material is heated by
applying heat to the outer surface of the material, and the
heat is transferred to the inner sections by conduction.
This heating mechanism may result in temperature gradients
inside the material. For the processing of polymers and

polymer composites, the outer regions of the material will

vm

£26 0:323: .2255 .5 an: .5. :25 282 2.883. 3 2.55

2.9.2 .03ro So .4

ON . m. N. m
p - hi'bi - p P b b b . — b b b bi p b p o.—

 

 

/

Tv.~

 

 

 

 

 

(.(ouonbag luzuosor) 2H9 °}

£26 2.. use: sea grass 2.. at. 55.5.32 area: 3 2:5

 

 

 

 

All w :IIJ cans-.58 .m:
a _,..TI_ cm.
1 1 _ 1

 

 

 

 

 

 

 

 

 

 

 

7
be cured faster, forming a skin seal on the material. This
could cause problems during processing. Some polymer
reactions have side products or require solvents. The cured
outer skin on the material's surface can form a barrier and
trap the undesired evolving products. This leads to the
formation of voids inside the material. Voids decrease the
mechanical properties of polymers and composites. Also,
polymer cross-linking reactions are exothermic. The
exotherm can cause temperature excursions during processing,
which can lead to thermal degradation of the material.

Microwave heating, on the other hand, transfers energy
directly to the material's molecules. The inside of the
material can be heated selectively. The temperature
gradient is therefore outward. This "inside-out" heating
pattern may decrease the entrapment of the undesired
ingredients. Because the outer regions of a microwave
heated material are less reacted and less rigid than the
inside, the mechanical properties of the material may be
improved. Also, this heating mechanism may result in a
better consolidation of the part, or require a lower
processing pressure.

Another advantage of microwave heating is that it
allows better control of the temperature profiles. The
direct heating mechanism of the microwave heating has been
shown to eliminate the temperature excursions for exothermic
polymer reactions”. Jow demonstrated this by conduction

thermal and microwave heating experiments. The microwave

8
system can be turned off and on instantaneously to better
compensate for the reaction exotherm than the thermal
system. This provides an opportunity to process a material
at higher temperature without degrading it.

The polymers are sometimes not lossy enough to be
promising materials for microwave processing. However, the
addition of lossy materials can be added into the matrixes
to make microwave processing of these composites possible.
In a graphite composite, graphite fibers are good conductors
and very lossy. The microwave will heat the fibers by
inducing a current in them. The bindings between graphite
fibers and polymer resin will be enhanced due to the
concentrated reaction regions at fiber-resin interfaces.
The mechanical properties of the composite will thus be
improved. It has been shown20 that microwave coupling could
alter the interphase properties between fibers and matrix.
Compared with thermally cured composites, it was found that
not only the interfacial shear strength was different, but
the failure mode at the interface was also altered.

Microwave heating also affects the extent of reaction
for certain polymer reactions. It has been found that the
extent of reaction is higher for microwave processing than
thermal processing at a given temperatureuzh The result
could be due to the advantages of microwave heating in the
area of reaction kinetics. Faster reaction rates have been
reported by the directly applying microwave energy to the

reactive sites of polymers. Wei, DeLong, and Hawley used

9
thin film technique to study the reaction kinetics of epoxy
and found microwave processing to be faster”. Also,
enhanced reactions of polygamic acids in solutions have been
reported by Virginia Polytechnic Instituteuxfi Molecular
"hot spots" have also been reported where the reaction is
much faster than the observed reaction rate for conventional
thermal processing“.
1.5 Microwave Applicators

Several microwave applicators, such as waveguides,
commercial multimode microwave ovens, and single-mode
tunable resonant cavities, are commonly used in the
microwave processing of materials. Microwave curing of
epoxy has been studied using waveguidesn”, and multimode
microwave ovens””h However, the waveguides and multimode
microwave ovens are usually energy inefficient devices. In
a multimode microwave oven, the electromagnetic fields are
standing waves. In a waveguide, the electromagnetic fields
are travelling waves. Both microwave distributions inside
the microwave device can not be efficiently controlled for
the heating of materials.

In a single-mode tunable microwave cavity, however,
electromagnetic waves are resonant inside the cavity. The
microwave energy is well focused and distributed according
to the cavity boundaries and materials being processed. The
microwave energy being reflected and absorbed by the cavity
walls can be minimized by tuning the cavity and by selecting

a cavity wall material with lower electrical resistance.

10
Several microwave resonant modes can usually be found at
different cavity dimensions for a tunable cavity loaded with
processing materials. Jow used a single-mode resonant
microwave cavity to study the curing of epoxy. Efficient
microwave heating was found and materials were heated to
higher temperature in less time. In addition, the single-
mode resonant microwave cavity can be used to diagnose the
dielectric properties of a material while processing it”.
1.6 Microwave Processing Techniques

Microwave processing of polymer materials is
investigated extensively at Michigan State University.
Current research issues for microwave processing are: (1)
process scale-up; (2) processing of complex shapes of parts;
(3) automation of processing control; and (4) continuous
processing.

The scale-up of the microwave cavity was studied for
the batch processing of larger part of materials”. The
scale-up factors were found for 18" diameter cavities to
maintain the same heating patterns as 7" diameter cavities.
The effect of the orientation of graphite fibers in a
composite was studied to determine the processing
orientation of composites”. It was found that the heating
results were different for the composite for different fiber
orientations.

Processing of complex shape of composite materials is
also under study to extend tunable microwave processing to a

more extensive application. In order to achieve uniform

ll
microwave heating over the surface of a complex shape
material, a mode-switching technique has been recommended by
Larry“. The goal is to achieve a uniform heating profile
throughout the material by switching between microwave
heating modes. Each heating mode may yield a different
heating pattern for a sample, and the temperature gradients
across the sample may be minimized by alternating between
the different heating modes. A computer-based control
system has been developed that implements this technique to
automate and better control the processing of the simple and
complex shape materials”.
1.7 Research Scope and Goals

Microwave processing has been widely used in various
aspects of industrial and commercial applications. Most of
these applications are batch processing. In batch
processing, the size of a processed part is limited by the
size of the microwave batch applicator. Because materials
with large or continuous geometries are highly desired in
industrial and commercial applications, continuous microwave
processing of materials exhibits a great potential for
materials processing.

The idea of continuous microwave processing is to pass
materials through the microwave applicator. The material is
processed by the microwave environment inside the applicator
and exits the applicator when the processing is completed.
Many rubber materials, polar or non-polar, have been

processed continuously using microwave processing

12
systems“3K One of the key points in continuous microwave
processing is to control the microwave leakage from the
entry and exit ports of the applicator. The safety
threshold limit value (TLV) at 2.45 GHz is 10 mW/cm23fi.
The TLV presents the maximum radiation level to which
workers may be repeatedly exposed without adverse health
effects. For non-conductive materials, the microwave
leakage is easier to control during continuous processing.
Devices called double corrugated reactive chokes were
studied and used for reducing microwave leakage during the
continuous processing of polystyrene”. Conductive graphite
fiber with epoxy resin was also studied for continuous
microwave processing”. The microwave leakage control in
this case is far more difficult due to the tremendous
microwave leakage caused by the conductive graphite fibers.

The main goal of this research is to develop a low-
cost, high-speed microwave processing technique. The
research conducted in this study is to demonstrate the
feasibility of continuous microwave processing. Several
aspects of the continuous processing will be explored to
evaluate the microwave processing methods.

The possibility of processing a heat cable will be
tested. The heating and curing of the heat cable at
different heat modes and different sample positions will be
studied. Because the heat cable contains two electric
wires, the microwave leaking during processing will also be

studied. The continuous microwave application will also be

13
applied to pultrusion, a composite manufacturing process. A
microwave transparent die will be designed to perform the
microwave pultrusion. Heating and curing profiles of the
pultruding material will be studied at different processing
conditions. A process model, which includes reaction
kinetics and microwave power absorption models, will also be

developed to further study the microwave pultrusion process.

CHAPTER 2

CONTINUOUS MICROWAVE PROCESSING OF HEAT CABLE

2.1 Introduction

Continuous microwave processing of materials has been
studied for decades for polymer and composite materials.
Most microwave processing applications are with non-
conductive materials, such as rubbers. Conductive materials
such as graphite fibers have also been processed by
continuous microwave techniques. Microwave leakage during
the continuous microwave processing has been controlled
within safety limit.

In the study of graphite fiber composites, Wei modified
a batch tunable resonant microwave applicator for the
continuous processing”. As shown in Figure 2.1, two
rectangular slots were opened at opposite sides of the
cavity wall perpendicular to the coupling probe. A jacket
box was attached to each opening slot to prevent microwave
leakage. The entry and exit ports of the cavity were
specially designed to reduce the microwave leakage due to
the open space and conductive graphite fiber. A set of
doubly corrugated reactive chokes have been used for
reducing microwave leakage at the entry points during

continuous microwave processing”. In addition, Wei

14

15
suggested using conductive fins grounded to the cavity wall
openings. As a result, the majority of the induced current
in the conductive fibers is grounded by the fins. This
design allows a small portion of the microwave radiation
into the jacket box attached to the cavity wall. The dies
on the jackets were designed to fit prepregs of varying
thicknesses and widths. Finger stocks were used to further
reduce microwave leakage by making contact with the
composite material and grounding the induced current.

This project is a continuation of the continuous
microwave processing studies using the modified tunable
resonant microwave cavity described above. To extend the
application of continuous microwave processing, materials
containing conductive wires are under study for possible

processing.

16

seseo=nn< seam—5....— ..8 £50 356me EN 953..—

coca
mezaaoo

 

 

 

 

 

 

\

xon
Loxom_ A\\\\IIIIII/v

a2 \
@_Qm>OE -

 

 

17

2.2 Experiments

The continuous microwave processing system is similar
to that of batch processing systems. As shown in Figure
2.2, the system includes four basic units: a microwave
source circuit, a modified tunable cylindrical cavity, a
data acquisition unit, and a temperature measurement unit.
The low-power, swept frequency microwave source is a HP
83508 Sweep Oscillator with the HP 86235A RF Plug-In
(frequency range from 1.7 to 4.3 GHz). The high-power,
single frequency microwave source is an Opthos MPG-4M
microwave generator. The modified tunable cylindrical
microwave cavity with a 17.78 cm inner diameter is the same
as shown in Figure 2.1. The top plate of the cavity is
movable to adjust cavity length. By adjusting the top
plate, different microwave resonant modes can be found
inside the cavity. The microwave energy is conducted into
the cavity through the coupling probe. The data acquisition
unit is a data collection board in a personal computer
driven by a data processing software“. The temperature
measurement is performed with a Luxtron 755 multichannel

fluoroptic thermometer or an infrared thermometer.

18

:o:8=&< scum—5.5m sou 3:85 0:39.32 ad 953..—

 

 

 

 

 

.5

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5:5
o>okouom=

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  

a}
on
vaoq
5.58:9
l... 3.50m
253232

 

 

once moaned

 

 

 

I_ _
4 4
I
oncomo .om
“ :. o E
_ IIIIIII
a
soda—:25 _
mookm _
1..
.83on
3.8333:

  

SEEEC

 

19

Some microwave radiation is reflected back into the
microwave circuit from the cavity when the cavity is not
well tuned to a certain resonant mode. The power reflection
is monitored by an oscilloscope (Tektronix 2213) over the
swept range of microwave frequencies. Resonant modes can be
found at locations on the oscilloscope trace where the
reflected power is zero. Once the resonant modes at 2.45
GHz are found, the single frequency microwave source is used
for the processing of materials. The input power, P“ and
reflected power, P” are measured on-line during processing.
By tuning the cavity, the reflected power can be minimized,
and the microwave cavity is thus at a resonant condition.

The material in this study is a raw product of
Chromalox Division, Emerson Electric. It is a 0.15 cm thick
and 0.95 cm wide heat cable. The cable is made of two
copper wires emerged in a base compound, with the shape of
standard TV cable. The base compound is low density
polyethylene with 20% highly conductive carbon black and
zinc power. This mold compound is also mixed with a cross-
linking agent, di-cup peroxide(3% by weight). The
processing of the material is to cure the raw product with
the cross-linking agent, which initiates at 177%:.

An infrared thermometer (Omega 081100) was used to
measure the surface temperature on the center line of the
cable material during processing. The temperature
measurements were made through an opening at the center of

the top plate of the cavity. The cable material was passed

20
through the cavity by a pulling roller. The pulling speed
was controlled by a step motor which had a variable speed
control. The cable material was processed at optimum
resonant modes using a single frequency source at different
input powers. Microwave distributions inside the cavity for
samples at different cavity heights were also studied. The
temperatures were taken at different sample positions using
optical temperature probes attached directly onto the cable
surface. As shown in Figure 2.3, position 1 was at center
of the cavity, positions 2 and 3 were 15 mm off the center,
and position 4 was 30 mm off the center.

The possibility of processing cable materials through
the axial direction of the cavity was also tested. The
microwave distribution across this sample orientation was
measured by recording temperatures with optical temperature
probes attached to the surface of the cable. Figure 2.4
shows the cable positions from which temperatures were
taken. Positions 1 through 8 were along the axial direction
of the cavity. The positions were 2 cm apart with position
1 at 1 cm from the bottom plate of the cavity.

The processing speeds of the cable material for the
curing studies were 2.5 cm/s and 1.8 cm/s. The
corresponding residence times for the material in the cavity
were 7 and 10 seconds. The microwave input power was 70 W.
The curing of the material was studied using DSC to

determine the extent of cure.

21

=e:8..5 .8553: «a 03:0 so 2.2—.8382
2382—th 8.. 8338.— .EB 5250 033232 he Beta—gm ad «Eur.—

 

 

 

 

 

 

 

 

 

 

 

 

 
 

22

 

 

 

 

 
 

   

 

iiiliiiiiiiiiiiliiiii ii321331iii333iiiiiiiiilfllllllflllllllllllllllllllllllllll

BLQSV’EIZL

 

  
  
   
 

 

 

 

 

 

Figure 2.4 Sideview of Microwave Cavity with Positions for Temperature
Measurement on Cable at Vertical Direction

23
2.3 Results and Discussions

There were eight resonant modes available for the

horizontally loaded microwave cavity. The cavity dimensions

and mode types for each resonant mode are listed in Table
2.1.

Table 2.1. Heating Modes for Cable in Horizontal Direction

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Mode No. Cavity ModgType Max.
Length (mm) Temperature (°C)
I 1 63 . 0 TE," 150
2 72 . 0 mo“ 40
3 101. 5 TE,“ 36
4 108 . 5 TM,ll 42
5 130.0 TB", 112
6 145 . 5 mm 29
7 153 . 0 TE,“ 29 I
8 182 . 0 TE,” 28 I

 

 

From the preliminary heating results above, mode 1 and
5 show the most promising results for further processing
studies. Both resonant modes can heat the cable material
over 100°C at only 30 W input. The heating of the cable
material was studied at two cavity heights, 40 mm and 50 mm
from the bottom of the cavity. The heating results are
shown in Figure 2.5 using mode 1 and 5 at 30 W microwave
input power. At both heating modes, the heating for the

cable at height 50 mm was more effective than those at

24
height 40 mm. The microwave distribution patterns for both
heating modes were basically the same. The center of the
cable material was heated more efficiently than the rest of
the cable. This indicates that the microwave energy may
concentrate more in the center of the cavity. This finding
leaded to the subsequent study in which the cable material

was passed through the cavity in the axial direction.

 

 

 

300 - r T r
o rnodeI j
_ 0 mode 5 1
250 height 40mm
A ------- - height 50mm
9
:6 200 :— .............. ‘
3 ------- - ....................... -
*9: E’ ‘3‘
3
E 150 ‘0 ....................
o .....
’—

 

 

100

 

50

 

 

Pofifion

Figure 2.5 Heating Results for Cable at Horizontal Direction

25

The curing results of the material are shown in Table
2.2. The thermal curing result in the table is for the
conventional manufacturing processing condition. Compared
to the thermal curing, microwave processing is much more
efficient. At different microwave processing speeds, the
residence times for the cable material inside the microwave
cavity are different. A longer residence time will result
in a greater extent of curing for the same input power and
pulling speed. At a processing speed of 1.8 cm/s, the cable
material was 63% cured after being pulled through the cavity
for a microwave input power 70 W. At a faster processing
speed of 2.5 cm/s, the cable material was cured less even
though the material was passed through the cavity twice.
The reason is that the curing rate of the material is a
strong function of temperature. At a faster pulling speed,
the cable material is not heated to as high a temperature as
it is at a slower pulling speed. To maintain the rate of
curing at a faster processing speed, the microwave input

power must be increased.

26

Table 2.2 Curing Results at Different Pulling Speeds

 

 

 

 

 

 

 

 

 

 

Curing Type Residence Processing' Curing
Time(s) Speed(cm/s) Result
Thermal Curing 360 7.5 100%
Microwave Curing 10 1.8 65%
Microwave Curing 7 2.5 13%
Microwave Curing 14 2.5 45%
i—=__-_ .__._-.=.=..

The heating modes available for processing cable

material in vertical direction are shown in Table 2.3. The

disturbance of the material to the microwave field is

different for the processing material at different sample

positions inside the cavity.

As a result, the resonant

conditions are changed for the cable material in the

vertical direction compared with those at horizontal

positions.

Table 2.3 Heating Modes for Cable in Vertical Direction

 

 

 

 

 

 

 

 

 

 

Mode No. Cavity Mode Type Heating
Length (mm) Results (°C)
1 63.0 TB", 120
2 80.0 Tan, 45
3 111.0 TM", 42
4 129.5 TE", 67
5 153.0 TE”, 82
E—u—u—L :- _ =—

 

 

 

 

27

As can be seen in Table 2.3, modes 1 and 5 are most
effective for the processing system. The cable was heated
over 80°C at both heating modes with an input of 30 W. The
further heating results using these two modes are shown in
Figure 2.6. As the cavity length varies with different
resonant modes, the residence times at different resonant
modes for the vertical processing orientation are different.
By comparing with the heating results for the cable in the
horizontal direction, it can be seen that the microwave
heating for the cable in the vertical direction is less
efficient. This could attributed to the microwave energy
distribution changing dramatically with the cable running
through the axial direction of the cavity. The microwave
energy does not concentrate along the axial line of the
cavity for the vertical processing orientation.

In addition to the difficulties of designing the
microwave leakage control device for the vertical
processing, the heating of the cable material was less
efficient than that of the horizontal processing direction.
The processing of cable material along the axial direction

is not as promising as in the horizontal direction.

28

 

0 mode I
a mode 5

 

 

 

6‘
8,
8
3
*g 1004
0
O.
E
Q)
.—
50

 

 

Position

Figure 2.6 Heating Results for Cable at Vertical Axial Direction

One problem for the microwave curing of the cable
material is the surface roughness after processing. This
problem may be attributed to the fast heating rate of
microwave processing. The sudden temperature rise might
result in fast chemical reactions for the cable material.
Another possibility is burning of the polymer material
during intensive microwave processing. To address this
issue, an inert gas processing environment is needed. In
addition, a high pressure processing environment would help
to produce a better performance cable product.

The microwave heating is not only a function of the

electromagnetic resonant modes and the dielectric properties

29
of the material, but also a function of the location of the
material load in the cavity. The processing positions for
this study was confined by the openings on the cavity wall.
To find an optimum processing condition for this cable
material, the potential sample processing positions need to
be fully explored. It might be possible to achieve a better
processing result at a different microwave cavity position.
2.4 Conclusion

The potential of continuous microwave processing of
cable material was explored. With the modified tunable
microwave cavity, the feasibility of continuous microwave
processing of heat cable was demonstrated.

The processing of the cable material at two cavity
heights was studied. The processing at a height of 50 mm
yielded better heating results and temperature profiles.
The potential of processing the cable material along the
vertical cavity axis was also studied. The microwave
heating was less effective for the cable material along the

vertical direction.

CHAPTER 3

APPLICATION OF CONTINUOUS MICROWAVE PROCESSING TO PULTRUSION

3.1 Introduction
3.1.1 Pultrusion

Pultrusion is a continuous manufacturing method used to
produce fiber reinforced plastic profiles with a constant
cross-sectional area through the length of the product. In
pultrusion, the product's shape is determined by
continuously pulling the composite material through a die.
Pultruded composites consist of reinforcing materials, a
resin, and often a mat material and other ancillary
materials“. The reinforcing materials include mainly glass
fibers and graphite fibers. The resin materials are usually
thermosetting polymers which bind the fibers together.
During processing, the polymers are transformed from their
initial low viscosity liquid state to a rigid thermoset
polymer state as a result of a thermally induced
polymerization. The mat material is used to improve the
appearance of the composite surface.

As a result of efforts and achievements by processors,
material suppliers, universities, and research
organizations, pultrusion technology is evolving rapidly.

Pultrusion has become a key manufacturing method capable of

30

31
offering reliable and cost-effective products with
specifically defined properties. Pultruded composites have
been utilized over a broad spectrum of applications ranging
from civil constructions to aircraft and space vehicles.
Figure 3.1 shows the sketch of a typical pultrusion machine.
Fibers coming from roving are impregnated with resin
polymers. The preforming fixture brings fibers together and
form a consolidated shape. The polymers start
polymerization within the heated metal die and form a solid
profile. The cured composite is then pulled through the die
by the pulling mechanics at a constant rate. The product is
cut into desired lengths with the cut-off saw at the end of
the process.

One of the key steps in a pultrusion process is to
control the interactions among fiber, resin, and additives.
The conventional processing method is to heat the materials
inside the pultrusion die via thermal convection or
conduction. The pultrusion dies used for this processing
method are made of steels which are characterized by their
toughness and excellent heat conductivity. Both convection
and conduction heating convey heat through the surface of
the material. The heat is then conducted into the interior
of the material. This heating mechanism often causes
temperature excursions inside the processing materials

especially for large parts.

32

65:82 5.95:... .825. 5 68.5 < 3. 8&5

 

3(9 quhq-O

m3 omzmr . ..\nw.,\, fl
meats“. ozicoumma .

OZ_>OC mmaaau

.W.

hszNOKOquc #52

33
3.1.2 Microwave Processing

Compared to the thermal heating method, microwave
curing of composite materials is a faster and more direct
heating method that takes advantage of the dielectric
properties of the material being processed. The general
advantages of using microwave technology include: shorter
processing time, better control of temperature profiles
within the composite material, and improved mechanical
properties of the final part“. As a result, microwave
heating devices such as commercial microwave ovens“ and
waveguides“ are widely used in industry. Meanwhile, the
continuous microwave processing technique is highly desired
for processing large composite parts, such as long pipes or
panels. The critical problem of microwave leakage during
continuous processing can be controlled by using speCial
designed jackets attached to the entry and exit ports of the
microwave cavity“.

The applications of continuous microwave technique in
industries has been under investigation for decades. Most
of these applications use microwave energy to preheat or to
postcure materials after parts are made“. One pultrusion
application of processing materials continuously in a
waveguide has been patented“. The idea of using a
waveguide as part of a pultrusion die is excellent in terms
of the simplicity of the system. However, the dimension of
the waveguide have to be in accord with the dielectric

properties of the materials being processed in order to

34
create the suitable microwave field patterns”. This
requirement has greatly limited the wide application of the
technique in industry due to the large variety of materials
being processed.

In this study, a batch resonant microwave cavity was
modified to perform continuous pultrusion processing”.
Unlike a waveguide, a resonant microwave cavity can be tuned
to the optimum heating mode for different processing
materials. The microwave pultrusion process proposed here
is fundamentally different from the preheating and
postcuring applications described above. A microwave
transparent die was specially designed to fit into the
microwave cavity. The microwave energy will heat materials
directly through the die as they are pultruded through the

microwave cavity.

35
3.2 Experimental
3.2.1 Microwave pultrusion setup

The microwave circuit system for the microwave
pultrusion process is similar to that of a batch microwave
processing system described in chapter 2. The microwave
resonant modes were located using a swept frequency
oscillator. A single frequency (2.45 GHz) power source was
used for heating and curing tests. The input and reflected
power were measured on-line during processing. As is seen
in Figure 2.2, the microwave cavity was kept tuned by
adjusting the cavity length(Lc) and coupling probe depth(Lp)
so that the reflected power is minimized. A 17.78 cm inner
diameter tunable cylindrical batch microwave cavity was
modified for the microwave pultrusion processing.

Two dies were used in this study. A Teflon die was
used for heating studies, and a ceramic die was used for
pultrusion tests. Figure 3.2 shows the design of the die.
The opening shape of the die is an 0.294 cm high, 2.45 cm
wide rectangular. The top and bottom pieces of the die are
bonded together with nylon screws along both edges of the
die. The rectangular shape at one end of the die is to hold
the die in place during pultrusion. Figure 3.3 shows the
microwave cavity loaded with the pultrusion die. The die is
inserted inside the cavity through the openings on the
cavity wall.

The inner surface of the ceramic die was treated with a

mold release agent to prevent it from scratching as the

36
processing material is pulled through. The release agents
are mono-coat multi-release agents from Chem-Trend
Incorporated. The first layer of the agents, model number
E304, is used to seal the micro-holes in the die surface.
The second layer of the agents is applied to further smooth
the die surface. Both layers of release agents have to be
treated at processing temperature. The treatment will
improve the durability of the Mono-coat release film and
provide maximum number of releases. In the heating tests,
the temperatures were measured with optical temperature
probes attached directly to the surface of the prepreg. The
data were collected with a Luxtron temperature measurement
system. The system was connected to the computer and the
data were processed with data processing software.

In the pultrusion tests, the prepreg was pulled through
the cavity using a variable speed step motor which drove the
rollers. The input power for pultrusion tests was 100 W.
The prepreg was pultruded at different pulling speeds and
different input powers. The maximum microwave leakage

during processing was under the safety limit”.

 

 

 

 

 

38

ea saw—:3...— 5? 5:60 2:39.32 Co 96335 Rm 953..—

it

 

._ _.¢v mmcmm Ozamaou

 

 

 

zommbmhham

m:—

I »

.I.—..|L

 

man. H.535:

 

 

(1 I

 

 

Ommmmmm

 

 

H:— Zoumbrhaam

1, I

 

no“. ”.3990:

 

I e

 

 

 

39

3.2.2 Material Preparation

The material used in this study was a continuous
prepreg consisting of glass fiber/vinyl ester and vinyl
toluene resin. The resin contained 1 wt% of cross-linking
initiator Benzoyl Peroxide(BPO). The prepreg was prepared
separately in a fiber/resin prepregging machine. As shown
in Figure 3.4, the glass fiber tow is unrolled from the
roving and passed through the guiding wheels. The fiber tow
then passes through the resin pot which contains vinyl ester
polymers. After the fibers are impregnated with the
polymers, they are wound on the take-up drum. The drum
moves horizontally while rotating and taking up the fiber
tow. The fiber tow forms a sheet of prepreg on the drum.
By adjusting the drum rotating speed, various thickness of
prepreg can be obtained. The prepreg sheet is then cut into
widths of 2.45 cm plies. Several plies of the prepreg is
stacked together to form the composite prepreg tape with a
cross-section dimension of 2.45 cm by 0.3 cm for the

processing.

40

83E «530.595 En 8:9...—

EEO

@230 \\
ommEmO

mEn.

.09“.
O 05:03!"—

 

AmEQ uEDmocaw LMBV
.mcozaov

.5 NO
Encam— _.:||_~mmkv3 # I

L030 m _

 

i‘

41.

6.

58m nosed is; .8.— .sofiocsmeoz Own m2». Baum..—

.oo. ecsueeoneoh

 

 

 

 

DOWN “:5 DD Ow. v? HOLOCOQO
Wu 0?“ ON“ 00“ 00 on O?
,fillllell.ilrulL b LII L P > L .l-IL i .lhl. i if i -Ll . :Ilt ILfium.°.l
/ fio.o
fi
an...

 

fie.“

1m.«

 

I. A . I.'- ‘nl.’ie|I-I.O'QIL

 

manna nmILn<IN«
Can c«l

Hausa cam
"Lou-Leno

moo.omam> "snag

0mm

 

lo.~

.o com a» c«e\um “acessou

o.oo~ a» 2H:\uon "bonus:

as ooon.m« "seam

can an can: canon cease nae“) "museum

(O/M) nets 390a

42

3.2.3 Reaction Conditions

Polymer samples were taken for DSC measurements. The
results shown in Figure 3.5 give the information about the
polymer cross-linking reactions. The reaction starts at
about 80°C. The exothermic heat from the reaction is
studied by heating resin samples at different heating rates
for DSC measurements. From the DSC results in Figure 3.6,
the exothermic heat for the resin polymerization reaction is

found to be 272.5 J/g.

 

280 ' T ' ‘

270

260

250

Heat of Reaction (.J/g)

240

 

 

 

230 . 1 L l .
O 5 10 15

Heating Rote (OC/min)

 

Figure 3.6 Heat of Reaction Measurements Using DSC for Vinyl Ester Resin

43

3.2.4 Temperature Measurement

The temperature measurements for the heating experiment
were made by attaching optical temperature probes on the
surface of the prepreg. The temperatures on the positions
through the prepreg inside the cavity were recorded at
different heating modes. Figure 3.7 shows the top view of
the cavity with processing prerpeg. Positions 0 through 7
are along the pultrusion direction, while positions a, b,
and c are across the prepreg width. Position 0 is at the
entry port of the cavity and position 7 is at the exit port.
Since the cavity diameter is 17.78 cm, the positions along

the length of the prepreg are approximately 2.5 cm apart.

44

0:89:60 5 «cos—2.682
2582—809 .3. 33:8.— .23 5.50 o>aseso=z 2: Co 33> non. fin odour.—

oposm
wafiQSOO

 

 

-.zlmm+66.a

 

 
 

N. as 2. m m 2 c

.3?me 95.50.82:

45

3.2.5 Extent of Cure Measurement

A Differential Scanning Calorimetry(DSC) technique was
used to determine the extent of cure for the microwave
pultruded samples. The measurements were performed by
comparing the exothermic peaks of uncured and cured samples.
The extent of cure a is then calculated using the following
equation

= l- uwakavafl

(3.1)
nuukcwaflo

where [peak area] and [peak areaL,are the exothermic peak
integrations from the DSC results. They correspond to the
exothermic heat of the cured and uncured samples,
respectively.

The samples for DSC measurements were taken from the
positions on the prepreg where the temperatures were
measured: positions 0 through 7 and positions a, b and c.
3.2.6 Fiber Fraction Measurement

The microwave pultruded samples were also measured for
fiber volume fraction and void fraction. The measurements
were performed in a microscopic instrument with a data
processing software called Optical Numerical Volume Fraction
Analysis (ONVFA).

Samples were prepared through several steps of fine
grinding to make sure the measured surface is smooth. The
sample cross-section was analyzed through microscopic

images. The average fiber diameter was measured by marking

46
the fibers with electronic counting. The total fiber area
was then calculated by multiplying the number of fibers with
the average fiber diameter. The fiber volume fraction for
this analysis was found by comparing the total fiber area
with the cross-section area under analysis. After analyzing
fifteen cross-section areas at arbitrary spots of three

samples, the average fiber volume fraction was found.

47

 

Fiber diameter: 10 - 15 microns.

Figure 3.8 Microscopic Image of the Microwave Pultruded Glass/Vinyl Ester
Composite Sample

48
3.3 Results and Discussions
There were four resonant heating modes available in the
prepreg loaded tunable microwave cavity with Teflon die.
Table 3.1 lists the cavity length and coupling probe depth
parameters for each mode.

Table 3.1 Heating Modes for the Cavity with Teflon Die

 

 

 

 

 

Mode Mode Cavity Length Cavity Length Probe Depth
No. Type Lc(mm) Lc*(mm) Lp(mm)

1 1'30" 113 109.85 12.79

2' TB,” 134 130.51 9.74

3 mm 143 137.08 22.99

4 TE311 158 154.45 11.54

 

 

 

 

 

 

 

-- Lc is the theoretical cavity length for empty microwave
cavity.

-- Lc* is the cavity length with Teflon die and processing
material.

In order to find the optimum heating mode for the
prepreg system, further tests were conducted by heating the
prepreg at each heating mode with an input power of 30 W.
Temperatures were taken along the centerline of the prepreg
at a time interval of 5 seconds. As shown in Figure 3.9,
mode 4 is the heating mode with greatest heating rate for
the prepreg system and die geometry. The prepreg was heated
to 40°C in approximately 2 minutes at an input power of 30

We

49

 

 

 

 

45 l l l I l
0 model
40 h a mode 2
8 g 0 mode 3
\o/ A mode 4
tag 35 - -
5 i
i— )-
< ,
a 30 - _ .. , *
2 7:4." ""‘ ‘ ‘
Lu . - A .
'_ W
25 F _
20 1 l l l 1 l 1 l 1 l r
O 20 4O 60 80 100 120

TIME (s)

Figure 3.9 Heating Tests at Different Modes for Glass/Vinyl mer Composite
with Microwave Input of 30 W

50

More heating experiments were conducted to study the
absorption of microwave energy by the prepreg material
inside the cavity. The prepreg was tested at mode 4 and
temperatures were taken along the length of the prepreg at
the center and on the edges. Figure 3.10 shows the
temperature measurements at different positions on the
prepreg tape. Positions a and c refer to the edges of the
prepreg, while position b refers to the centerline of the
prepreg. The temperatures were taken along the length of
the prepreg from position 1 through 8 and the average of
these temperatures is presented in the graph. It can be
seen that the prepreg was heated to 80°C(the cross-linking
temperature) within 5 minutes. The temperature gradient
across the prepreg is small, with less than 10°C difference
between the center and edges of the prepreg.

When dielectric materials were loaded inside a
microwave cavity, the microwave resonant modes will be
disturbed. The extent of disturbance depends mainly on the
dielectric properties of the materials. Ceramic materials
are less transparent to the microwave than Teflon materials.
As a result, the resonant mode disturbance for the ceramic
die is greater than that for the Teflon die. By inspecting
Table 3.1 and 3.2, we can see that the cavity length Lc for
the cavity with the ceramic die loaded is shifted more than
for the cavity with Teflon die. The basic heating patterns

for the corresponding modes, however, remain unchanged.

51

  

Temperomrefi‘JQ
20 4o 60 so \00

‘4

Figure 3.10 Heating Results at Mode 4 for Glass/Vinyl Ester Composite with
Microwave Input of 60 W

Table 3.2 Heating nodes for tho Cavity with Ceramic Dio

 

 

 

 

 

 

Mode Mode Cavity Length Cavity Length Probe Depth
No. Type Lc(mm) Lc**(mm) Lp(mm)

1 T20“ 113 97.35 12.79

2 TEm 134 114.92 9.74

3 mm 143 120.64 22.99

4 TE,ll 158 146.76 11.54

 

 

 

 

 

--Lc is the theoretical cavity length for empty microwave
cavity.

--Lc** is the cavity length with ceramic die and processing
material.

 

52

The pultrusion tests were accomplished using the best
heating mode(mode 4) for the prepreg system at a certain
processing speed and 100 W input power. The ceramic die was
used because of its superior mechanical properties.
However, the input power had to be increased due to its
reduced microwave transparency as compared to that of the
Teflon die. As is seen in Figure 3.11, the extents of cure
were measured for pultruded prepreg samples cured at
different residence times. At short residence times, the
extent of cure is greater at the center than at the edges.
But as the residence time increases, the extent of cure is
consistent and complete across the width of the prepreg. To
increase the microwave processing speed, the input power has
to be increased. A higher microwave power source is
required for faster continuous microwave pultrusion

processing.

53

Q5

  
 

{fleck 0“ Cut e.

QB

Figure 3.11 Curing Results at Mode 4 for Glass/Vinyl Ester Composite with
Microwave Input of 100 W

To find the optimum operating conditions for the
microwave pultrusion processing, the relation between

processing rate and extent of cure was studied at an input

power of 100 W. The prepreg tape was processed at

pultruding rates of 1, 2, and 3 cm/min. The curing results
are shown in Figure 3.12. From the curing results, at
pultrusion speed of 1 cm/min, the composite prepreg was

fully cured at the beginning of the process. At 2 cm/min,

it was fully cured at the 2/3 of the cavity length(position

5). And at 3 cm/min, it was only cured 80% at the exit port

54

of the cavity(position 7). These results showed that the

optimum pultruding speed for this pultrusion system at an

input of 100 W is approximately 2.5 cm/min.

0.8

(16

0.4

extentofCLwe

O2

00

 

 

'1

0

'1

o

u—

H

E El

u El

4
t‘ 4
fl

0
0

k‘

0 3cnwfinm.
D ZCm/mm.
A 1cm/mm.

 

l A l A l 4 l

3 4 5 6 7

pomfion

Figure 3.12 Curing Results for Glass/Vinyl Fster Composite at Different

Pultruding Speeds

The curing rate of a composite material is determined

by the heating efficiency which, in turn, is dependent upon

the efficiency of microwave energy coupling into the

material.

The coupling of microwave energy with the

material will mainly depend on the dielectric properties of

the material.

Other factors include the efficiency of the

55

microwave cavity and the location of the pultrusion die
inside the cavity. other heating modes might be able to
heat the material as efficiently at different die locations.
Therefore, more detailed studies on die locations are
necessary for an optimized microwave pultrusion process.
3.4 Conclusions

It has been illustrated that continuous microwave
pultrusion is feasible in a modified tunable microwave
cavity with a microwave-transparent die fitted inside. The
absorption of the microwave energy is a strong function of
the resonant heating mode. The distribution of the
microwave energy is uniform across the width of the prepreg.
The prepreg is heated directly through the die while being
pultruded through the microwave cavity. The fast heating
rate shows the capability of coupling microwave energy into
the prepreg through the die. Also, the uniform curing of
the prepreg will be found the basis for producing high

performance microwave pultruded composites.

CHAPTER 4

PROCESS MODEL FOR THE HICRO'KVE PULTRUSION

4.1 Introduction

The feasibility of microwave pultrusion has been
explored in Chapter 3. To better understand the microwave
pultrusion process, a processing model is developed in this
chapter. The model focuses on the prediction of the curing
and temperature profiles at various pultrusion rates. The
limitation of pultrusion rate is also studied for the
optimum microwave pultrusion conditions.

Optimum material properties and reliable processing
conditions are necessary for the processing of high
performance thermoset composites. These data are usually
obtained through extensive testing of various designed
materials at different processing conditions. However, this
kind of testing is time consuming, and the tested results
are often limited to a certain extent. Sometimes the
experimental testing is limited by laboratory set-up and
instrumental conditions available. Also, the procedure of
finding the optimum operating conditions could be costly. A
fast and cost-effective way to achieve this objective is
through the numerical simulation of the processing. The

simulation usually uses a processing model which includes

56

57
chemical, physical and mechanical properties of the material
during processing. The model should include the information
of chemical changes and energy transfer of the material.

In this study, a kinetics model for free radical
polymerization is used for the vinyl ester's cross-linking
reaction. An empirical microwave absorption model is
developed to describe the microwave heating process. A
FORTRAN computer code was written to simulate the processing
at various operating conditions. The simulation can be used
to predict the curing and temperature profiles at various
processing conditions. It can also provide an optimum
processing condition based on the required properties of the
products.

4.2 Kinetics of resin polymerization

The resin used in this study is a methacrylic ester of
Diglysidyl Ether of Bisphenol-A (DGEBA) vinyl ester mixed
with 45 wt% vinyl toluene monomers. Benzoyl peroxide(BPO)
is added to the mixture at an amount of 1 wt% as the
initiator. The chemical structures of these chemicals are

shown below:

58

cgflggbmnfi:xg%_(:}EE(D}JJmfi:xn._(:}§2(D}qufi:xgbngfg

finflefla'
=CH2

CH3
vmwhwhun

<1?ng
0 O

BRO

Figure 4.1 Chemical Structures of the Resin

As can be seen in the chemical structure of vinyl
ester, the cross-linking of polymers achieves via the two
vinyl groups at the both ends of the structure. The
polymerization of the resin is through a free radical
mechanism. The reaction procedure is as follows”:
Initiation

Kc
I -———————> 2 R0

Propagation

R0 + M > Mf’

 

 

Mfi'+ M > up

 

M20 + M > M,-

143-+14 > M,-

 

Termination

 

Kw
M.‘ + "‘3 > Mr...

K.
H.-+M..- >M..+M..

 

The curing kinetics for the polymeric mixture
containing monovinyl and divinyl monomers can be described

by the following equations”:

 

 

l
%=K,(21:M)5[a’xxmx’l (4-1)
, CLsWZQD
= M (4.2)
[I] [I]. (14x)
sin—1 (4.3)
d

where X is the fraction conversion of the vinyl groups; t is
the reaction time; f is the initiator efficiency; kg is the
initiator decomposition rate constant; [I] and.[IZ]o are the
initiator concentration at reaction times t and zero,
respectively; 8 is the volume shrinkage factor; d, is the
density of the thermosetting polymer formed; and dm is the

density of the monomer mixture.

60

In this kinetics model, it is assumed that the
termination reaction is diffusion controlled at the
beginning of the polymerization. The model also assumes
that the vinyl groups of the monovinyl and divinyl monomers
have the same reactivity. The polymerization of methyl
methacrylate/ethylene glycol dimethacrylate system has been
reported using this model with a good correlation to
experimental results”.

The polymerization of DGEBA vinyl ester/vinyl toluene
system has also been studied using this model“. The system
was thermally cured at different processing temperatures.
The modelling prediction provided a good fit with the
experiment results. The following adjustable parameters of
the model in Table 4.1 were obtained through the fitting of
the experiment data with the prediction.

Table 4.1 Kinetics parameters for vinyl ester/vinyl toluene

 

polymerization
dm 1.04 g/cm?
dp 1.11 g/cm3
k. 8.02e7 exp(-8795.4/T) secJ
kc 2.13.25 exp(-6776/T) (L/mole-s) m
B. 2 .89
BM: 0. 58

* T is the temperature in %(

 

61

The curing rate equation using these parameters

 

becomes:
1
LIX = K.(2K.Ul)’ (1 - m1 + :10
ch B .1 (l-s-+23X)
(eprBxfi— - fin + I” 2
.fl f n»
where

 

K. = K L
P K,
K, = K“ + Kup
K r K“
W 1 1
EXPIBJV ' 7)]
fl .f
Kw = Bpr

Vp = [0,025 + 0.001 *(T -T,.)]
v, = V” + [0.00048*(T - To) - coon-(T - T,..)]*¢,

= X(l + e)

d” (1 +8X)

e - (d') -l
dp

(4.4)

(4.5)

(4.6)

(4.7)

(4.8)

(4.9)

(4.10)

(4.11)

(4.12)

62

The microwave pultrusion in this study used the
kinetics model and its parameters above except for the
activation energies in k,‘ and kc. The reason for this change
is that in microwave heating, the curing process may be a
little different due to the different energy dissipating
mechanism of the microwave heating. The activation energies
were obtained by minimizing the square of the error between
the experimental data points and the model prediction.

4.3 Composite Material Properties
4.3.1 Literature Data

From the optical fiber fraction analysis of the
pultruded composite in Chapter 3, the fiber fraction in the
composite is about 0.7. The composite density was obtained
from the reported densities of fibers and polymers. The
heat of the polymerization reaction was studied in Chapter 3
and should be approximately the same for the microwave
induced polymerization.

Other properties of the material include the dielectric
properties of the materials inside the microwave cavity:
composite material and pultrusion die material. Table 4.2
lists the dielectric constant and loss factor of these

materials.

63

Table 4.2 Dielectric Properties of the Materials

 

 

 

 

 

 

 

I —
Material Dielectric Constant Loss Factor
Resin 19.2 2x10'l
Macor Ceramic 5.67 42:10'2
'5‘ _— I —

 

 

 

 

 

As can be seen from the table, the dielectric loss
factor of polymers is much greater than those of glass
fibers and ceramic material. As discussed in Chapter 1, the
dielectric loss factor is a important criterion for
materials to absorb microwave energy. Therefore, the
microwave pultrusion processing of the composite material
will be modelled as if the die is transparent to microwave
and the polymers are heated selectively.

The gel point is an important parameter for the
processing of polymer composites, especially for the
microwave processing. The reason is that the dielectric
properties of polymers change dramatically after gel point.
The gel point is defined as the critical conversion in
thermosetting polymerization at which a three-dimensional
cross-linked network is formed. Walling” derived a
theoretical expression for the critical conversion in a
chain polymerization containing monovinyl and divinyl
monomers. The results showed a very low conversion (5%)
when the reaction mixture formed a gel. This is the reason
for the assumption of diffusion controlled termination

reaction in the kinetics model.

64

4.3.2 Cp Measurement

Heat capacity measurements were made using a TA
Instruments Differential Scanning Calorimeter mode 910.
Heat flow of an empty pan is first measured, then the heat
flow of sample material in the same pan is measured. Sample
material is heated to the desired temperature range to
measure the heat capacity.

The heat capacity is given by the equation

c = 60EAP

4.13
,, Hun ( )

Where Cp is the heat capacity in J /g°C; E is the cell
calibration coefficient at the temperature of interest; aP
is the difference in heat flow between the sample and blank
curves at the temperature of interest, in mW; H, is the
heating rate in °C/min; and m is sample mass in mg.

Values of E are calculated by measuring C, of a
standard Algh sample“. Using the calculated set of E
values, the heat capacity of the sample material is obtained
using equation 4.13. The DSC results of the sample material

are listed in the following table.

65

comm. can 4» oo.«> >a to>o
wmm omm on" cos on a

On“- IF 'P b b F P L L > 11.11

. u.v«.vsm o.«m.mm. u.vm.vm«
__ v :s . , -. - ‘
m.ou- o. . .mn
x: no.

r. . u.oo.v«~ .
m m o 38 mm. . ,h«.mm« oemmésmu oemmm cm
a In. mm .. 2. . .
H 1
a
C
«.1
“u v.ou1
O
M
m A
m

N.OI4

29:5 0:89—80 58mm— 354,335
coves—s.— o>a38o=z .2... Own «Em: «cos—8382 £3.39 .8: N... 8am“..—

.uo. ensueceaeeb

 

 

 

1‘11 'I. I‘ll" I‘l'l.‘bl.ul' t'

 

Tn!

1m:

Tu!

 

' I I'l:!“|‘lll

)

(Hm) "01$ 3'3H 380 I

66
Figure 4.2 shows the DSC results of the heat capacity
measurements. The calculated values are listed in Table
4.3. From these results, the average heat capacity of the
composite in the processing temperature range was found.
The value was used in the energy balance for the process

model.

Table 4.3 Heat Capacity Measurement

   
  
 
 
   

 

   

 

 

 

 

 

Temperature of Standard Coefficient Calculated Cp
Interest(°C) Cp for A1203 E of Samples
(cal/9°10 (cal/gm)
30 0.2090 1.378 0.241
126 0.2200 1.335 0.266
169 0.2362 1.344 0.267
214 0.2460 1.338 0.
E =3:

 

 

 

 

67

4.4 Microwave Power Absorption
A power absorption model is critical for the

development of a processing model. The temperature profiles
and curing rates are strong functions of the interaction of
microwave radiation with processing materials. In addition
to the dielectric properties of the material, the power
absorption is also controlled by the microwave distribution
pattern and electric field strength. Poynting's theorem can
be used to calculate the power dissipation into a material

as described in equation 4.1.

p, = lei-agar (4.14)
2

where P2 is the time average power dissipated in the
composite due to electric field of the incident transverse

electromagnetic (TEM) waves from top and bottom of the

material. if and E" are the electric field vector and its

conjugate respectively, and EK is the effective dyadic loss

factor of the material. Since the composite prepreg tape of
microwave pultrusion has very small edges (0.3 cm in
thickness), £5 can be approximated as the power absorbed by
the material.

For an empty microwave cavity, the electric field
inside the cavity can be calculated based on Maxwell's

electromagnetic equations and boundary conditions. For a

68
cavity loaded with small objects which only perturb the
resonant condition by a few percent, the electric field can
be calculated using a cavity perturbation theory”. For the
cavity loaded coaxially with a homogeneous, isotropic rod, a
mode-matching technique could be used. However, the
electric field for a cavity loaded with anisotropic material
becomes very complicated. The power absorption models
developed are only for specific type of materials and
processing situations, such as a five-parameter model for
the processing of a graphite composite plate”.

The microwave energy absorbed by polymers during
pultrusion processing is modelled by an empirical power
absorption model. The model describes the microwave
absorption ability according to the dielectric properties of

the material.

1

’Aell

 

P, = P,exp(- ) (4.15)
In equation 4.2, two adjustable parameters are used for

different processing conditions, such as different heating

modes and pultrusion positions on the cavity wall.

69

4.5 simulation of Microwave Pultrusion Process

The configuration of pultrusion processing of composite
prepreg tape is sketched in Figure 4.3. The prepreg tape is
enclosed in the ceramic pultrusion die. The microwave
energy dissipates into the prepreg material through the
microwave transparent die. The exothermic heat from the
resin polymerization reaction is conducted through the die

material into the environment.

L

 

' - r'23=h
pultrusuon due

composite tape
pultrusion die

 

 

L

— Z=0

 

 

 

>X

Figure 4.3 Modeling Configuration for Microwave Pultrusion Process

70

To simplify the computation process, following

assumptions are used in the simulation:

1. the microwave pultrusion is at steady state for a
defined input power and pultruding rate.

2. the temperature distribution is uniform along the
thickness and across the width of the prepreg
tape.

3. the heat transfer is only through the top and
bottom surface of the pultrusion die.

4. the die material and glass fibers in the prepreg
are transparent to microwave.

5. the temperature profile is uniform on the outer
surface of pultrusion die.

From an energy balance, a one-dimensional heat transfer

equation was derived as follows

dT - ¢1 dT
pCpua-t- = P. + pH + (72¢? (4.16)

with temperature boundary conditions

dT _
("—dz) lm - 0 (4.17)
dT _ _
“'01—sz - h.(T, T.) (4.18)

where p and Cl, are the density and specific heat of the
composite, k is the thermal conductivity of pultrusion die

material, H is the heat generation rate per unit rate by the

71
polymerization reaction, and R,is the heat generation rate
per unit volume by the absorbed microwave energy.

To solve for T(x) on the prepreg tape during microwave
processing, the reaction kinetics and microwave power
absorption models in chapter 4.2 and 4.3 are required. To
simulate the temperature and extent of cure profiles along
the length of the prepreg tape, implicit method of Lee's”
was used to numerically code the equations. The Lee's
method is based on the forward-difference approximation.
The finite-difference expressions for the partial
derivatives are:

E = Tim-1 ' TM 4 19
6x OX‘ ( ,

 

(4.20)

 

a kaT) - k[ (T1¢_§'1) (Ti’loj-T1,j)-(T1-%.j) (TigJ-T1'1,j)]
5E‘ 52 - 022

where i represents nodes in x direction, j represents nodes
in z direction.

With the models for P, and H, the temperature and
extent of cure profiles of the prepreg tape were computed.
The computer simulation was programmed in FORTRAN code and
presented in Appendix I. The microwave pultrusion
processing was studied at different input powers and
pultruding rates. A processing temperature limit was set in
the program to prevent the degradation of composite due to

temperature excursion.

72
4.6 Results and Discussions
The fitting of experiment data with the model
prediction is shown in Figure 4.4. The pultruding rates
under this study were 1 and 2 cm/min, and microwave input
power was 100 W. The pultrusion processing for this study
is at microwave resonant mode 4(Lc=146.76 mm) for the

composite system.

 

 

 

 

 

1.0 r
0.8 - _
8
5 0.6 - D -
“<3 .
1%
3 0‘4 _: ,- O 1 cm/min ‘
E f D a 2cm/mm p
0.2 _
0.0 :" 1 l l 1 1 1 L l 1 l l 1 I
o 1 2 :5 4 5 6 7

Pomhon

Figure 4.4 Modeling of the Microwave Pultrusion Processing of Glass/Vinyl Fster
Cbnuxshe

73
As can be seen from the graph, the computer simulation
curves agree with the experimental results quite well. The
predicted temperature profiles at each processing condition
were also shown in Figure 4.5 and 4.6. The simulation
predicted a temperature excursion occurring at beginning of
the process if the pultruding speed is too slow(at 1

cm/min).

 

 

 

 

 

 

 

 

 

200 —
or c or c:CI8
A150 " 1
9 <2
3 H—
33 100 -‘ g
3 - 0.43
E 1 33
,2
5O . 0 temperature _
C] extent of cure — 0.2
5 4
O 1 l l l l 1 0.0
O 1 2 3 4 5 6 7

Position

Figure 4.5 Simulation of Curing and Temperature Profiles of Glass/Vinyl Ester
Composite at the Pultruding Speed of l cm/min

74

200 . I v I r r 1 r

 

CD
CD
CD

1
.0
CI)

150* -

11
.0
CD

1 l
.0
4;

Temperature (°C)
E3
0
Extent of Cure

()1
O
0

temperature 2
C1 extent of cure ~

.0
N

 

 

 

Position

Figure 4.6 Simulation of Curing and Temperature Profiles of Glass/Vinyl Ester
Composite at the Pultruding Speed of 2 cm/min

75
The adjustable parameters, A and B, in the power
absorption model and activation energies, E, and 1:2,, in the
kinetics model were obtained by optimizing the fit of
predicted values to experiment data in Figure 4.4. A
general least square method was used to minimize the errors.

The optimization scheme is

S = 2 (a1 - Y1(aj)) “-201
Where i presents for the measuring points along the tape, ai
is the measured extent of cure at point i, and Yi(aj) is the
predicted value at point i using the processing model. The

parameters were obtained in Table 4.4.

Table 4.4 Parameters for the Process Model

 

A = 8

B = 0.5

E, = 7.963 KJ/mol
E, = 5.446 KJ/mol

 

Applying these parameters for the process model, curing
and temperature profiles at higher pultruding rates are
predicted. The simulation results are shown in Figure 4.7
and 4.8. The microwave pultrusion at pultruding rates of 6,
12 and 24 cm/min were studied. At pultrusion rates of 6 and
12 cm/min, the composite tape would be fully cured at the
exit port of the cavity(position 7). The corresponding

microwave input power would be 136 and 164 W, respectively.

76

 

     
 

 

 

 

1.0 I T j l _———“————3v2-
0.8 -
“2
5 0.6 -
B
E
3'3 0'4 h 6 cm/min
'“ D 12cmn/nfin 1
Cl 24 cm/min
0.2 ~ 1
.1
DO '1 1 1 1 1
0 1 2 3 4 5 6 7

Position

Figure 4.7 Simulation of Curing Profiles of Glass/Vinyl Ester Composite at
Higher Pultruding Speeds

77

 

300 r 1 r 1 r I

250

 

N
O
O

O 6cxn/nfin

Tenjperature(°C)
63
o

 

 

 

 

100 D 120m/mm ‘
A 24cm/mm
50 -
O l 1
0 1 2 3 4 5 6 7
Pomfion

Figure 4.8 Simulation of Temperature Profiles of Glass/Vinyl Ester Composite at
IfighérPuhnmfingEhnmds

The pultrusion speed limit for the microwave pultrusion
system was also studied. A processing temperature limit was
set at 275°C to prevent the pultruded composite from
degrading. At pultrusion rate of 24 cm/min, as shown in
Figure 4.7, the composite was only 80% cured at the exit of
the cavity. The reason is that for a defined microwave
cavity, fully cure of a composite material at a faster

pultruding rate requires higher microwave input power. But

78
higher microwave input causes a temperature rise in the
processing composite. The processing temperature would
exceed the temperature limit set in the simulation program
if a higher extent of cure is desirable at this pultruding
rate. To maintain the processing temperature under the
degradation temperature limit, the pultrusion rate for the
microwave processing is thus limited to a certain degree to
eliminate the degradation of the composite.

The temperature profile simulation in Figure 4.8 shows
a uniform processing temperature after composite material
reaches processing temperature. A uniform processing
temperature will provides a better control of cross-linking
reaction of the polymers. The cross-linking of polymers is
important for composite to achieve a high quality mechanical
properties.

The microwave processing simulation model provides a
good direction for further microwave pultrusion studies. It
also predicts the possible experimental results for a better
design of the microwave processing system. The model is
confined to the specific defined process. The simulation
results could be different for different processing

situations.

79
4.7 Conclusions

A one-dimensional processing model was developed for
microwave pultrusion processing of composites. An empirical
microwave power absorption model was proposed to describe
the interaction between microwave radiation and polymers in
composites. A free radical reaction mechanism was used to
model the polymerization of resin during microwave
processing. The activation energies for the microwave
curing are obtained by optimizing the fit of the model to
experimental data. The agreement between the model
simulation and the experimental data is quite good. The
optimized parameters are used to predict temperature and
extent of cure profiles during the microwave pultrusion.
The curing results provide the processing conditions for
microwave pultrusion at higher pultruding rates and higher
microwave input power.

The modelling results also provide a pultrusion rate
limit for the microwave pultrusion system by setting a
processing temperature limit in the simulation program. The
processing temperature limit prevents degradation of the
proceSsing composite material. The temperature profile
prediction shows a uniform processing temperature along the
length of the material during microwave pultrusion. The
uniform processing temperature is important for producing a

high quality composite.

CHAPTER 5

SUMMARY OF RESULTS

The feasibility of continuous microwave processing of
polymers and composites has been explored. The application
of continuous microwave processing has been studied by
processing heat cable with conductive wires. The microwave
pultrusion process has also been studied as an extensive
application of continuous microwave processing. A process
model has been developed to describe the microwave
pultrusion process. These studies have illustrated that
continuous microwave processing offers a potential
opportunity for various aspects of polymer and composite
processing.

In the study of the processing of heat cable, the main
results are summarized as follows:

* Microwave heating was very effective for the heat
cable. The cable material was heated over 100°C with
only 30 W of microwave input power.

* Microwave processing efficiency was a strong function
of the processing position in the cavity. It was found
that processing the cable at a cavity height of 50 mm
results in better heating than at height of 40 mm.

* The curing of the heat cable was much faster using

80

81
microwave processing compared to the thermal
processing.
The microwave processing of heat cable was less
effective for the vertical cable position than that for
the horizontal position.

In the study of the pultrusion application of

continuous microwave processing, the following results were

obtained.

*

The microwave processing of glass fiber/vinyl ester
composite was quite effective considering the
dielectric properties of the material.

Microwave distribution during processing was relatively
uniform across the width of composite prepreg. The
maximum temperature difference between the center and
edge of the prepreg was less than 10 %L

The microwave distributions for the cavity loaded with
Teflon and ceramic dies were basically the same.
Although the cavity lengths for the heating modes
shifted, the basic resonant modes were maintained.

The curing of the composite prepreg was uniform across
the width after being pulled through the pultrusion
die.

At pultrusion rate of 2 cm/min and 100 W microwave
input power, the composite was fully cured at the exit
port of the cavity.

In the simulation study of the microwave pultrusion

process, a process model containing kinetics and power

82
absorption models was developed. The following results were
obtained:
* The power absorption model worked well for the

simulation of the microwave pultrusion process.

* The modified kinetics model offers a good description
of the cross-linking reactions of the polymers.

* The results of the simulation model fit well with the
experimental data.

* The model described the temperature profiles for the
higher microwave processing rates, which provided
information to prevent thermal degradation.

* The model predicted a processing rate limit for the

microwave pultrusion system.

CHAPTER 6

FUTURE WORK

The continuous microwave application on the processing
of heat cable has been studied. The feasibility of
microwave pultrusion has been explored. The advantages of
microwave heating as an alternative to conventional heating
methods have been demonstrated in these applications. In
order to apply these processing techniques to practical
industrial applications, some fundamental and technical
issues need to be further investigated.

For the processing of heat cable, the following aspects
need further study:

* Processing of the cable material in an inert gas
environment to eliminate the burning of the cable
material during microwave processing.

* Design of a microwave cavity capable of maintaining
high processing pressures to improve the quality of the
processed materials.

For the microwave pultrusion study, some application
and fundamental areas need to be further explored.

* Microwave pultrusion with other composite systems to
extend the microwave pultrusion technique to a greater

variety of applications.

83

84

* Improvement of the existing microwave processing system
for high-speed microwave pultrusion by improving the
microwave cavity design, introducing high power
microwave source, and constructing new pulling
mechanics.

* Improvement of the microwave pultrusion die by
developing a more microwave transparent die to better
exploit the advantages of microwave heating.

* Use of the microwave cavity wall as the pultrusion die
itself eliminate the processing difficulties caused by
introducing the pultrusion die into the microwave
cavity. A preliminary approach to the idea of using
microwave cavity as the pultrusion die is shown in
Figure 6.1.

The pultrusion die between two tunable microwave
cavities is attachable. Different pultrusion dies could be
attached to the system for different shapes. Since the
pultrusion die is part of the microwave cavity, this
application would require the design of a suitable microwave

resonant cavity to provide a suitable heating mode.

85

  
 

0.3 sewnsuudsm me ve>uem Add: ”1230 on» no uuem new:
aounhm newness—E carotene“: H252 no couexm a «6 05.5:

 

 

 

2o com—353a

 

 

 

 

 

 

 

 

 

>:>mo o>m>>ono_E
63ch

LIST OF REFERENCES

10.

11.

12.

13.

LIST OF REFERENCES

W. 1. Lee and G. S. Springer, J. Composite Materials,
12. 387 (1984).

J. Jow, M. C. Hawley, M. Finzel and T. Kern, Polym.
Eng. Sci., 22. 1450 (1988).

J. Jow, M. C. Hawley, and M. Finzel, Rev. Sci. Instr.,
60(1), 96 (1989).

J. Jow, J. DeLong, and M. C. Hawley, SAMPE Quart.,
29(2), 46 (1989).

J. Jow, M. C. Hawley, M. Finzel, J. Asmussen, and R.
Fritz, AIChE Conf. Emerg. Tech. Mat., Aug. (1987).

J. Wei, J. Jow, J. DeLong, and M. C. Hawley, SAMPE
Journal, 21(1), 33 (1991).

J. Wei, J. Jow, J. DeLong, and M. C. Hawley, 2nd
Topical Conf. on Emerging Tech. in Mat., AIChE annual
meeting, San Francisco, CA, Nov. (1989).

M. C. Hawley and J. Wei, MRS Symp. Proc., 182, 413
(1990).

J. Wei, B. Thomas, and M. C. Hawley, 37th Intern.
Sampe, Anaheim, CA, March, (1992).

J. Wei, M. DeMeuse, and M. C. Hawley, ACS meeting, SF,
CA, April, (1992).

L. Fellows and M. C. Hawley, 37th Intern. SAMPE,
Anaheim, CA, March, (1992).

J. Wei, V. Adegbite, Y. Chang, and M. C. Hawley, 6th
annual ASM/ESD Advanced Comp. Conf./Exp., Session of
Emerg. Struct. Comp. Tech. I, Oct., (1990).

M. Lin and M. C. Hawley, 38th SAMPE Symposium and
Exhibition, Anaheim, CA, May, (1993).

86

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

87

M. Lin and M. C. Hawley, AIChE 1993 Summer National
Meeting, Seattle, WA, August, (1993).

R. Bruce, Microwave Processing of Material, Vol. 124 of
Materials Research Society Symposium Proceedings, Reno,
Nevada, April, (1988).

I. Chabinsky, "Applications of Microwave Energy Past,
Present, and Future 'Brave New World'", Microwave
Processing Materials, Vol. 124 of Materials Research
Society Symposium Proceedings, Reno, Nevada, April,
(1988).

C. C. Ru and R. Liepins, "Electrical Properties of
Polymers: Chemical Principles", Hanser Publishers, New
York, (1987).

R. F. Harrington, ”Time-Harmonic Electromagnetic
Fields", McGraw-Hill Book Company, Inc., (1987).

J. Jow, Dissertation for the Doctor of Philosophy,
Department of Chemical Engineering, Michigan State
University, 1988.

L. Drzal, Mat. Res. Soc. Symp. Proc. 121, (1988).

M. C. Hawley and J. A. Asmussen, Progress Report on
"Microwave Processing of Polyester/Glass Composites",
Presented to Michigan Materials and Processing
Institute, Novi, Michigan, April, 1991.

D. A. Lewis, J. C. Hedrick, J. E. Mcgrath, and T. C.
Ward, "The Accelerated Curing of Epoxy Resins Using
Microwave Radiation", Polymer Preprints, 21(2), 330
(1987).

J. Wei, J. D. DeLong, and M. C. Hawley, "Thermal and
Kinetic Study of Microwave Cured Epoxy Resins",
Proceedings of the American Society of Composites Fifth
Technical Conference, East Lansing, Michigan, June,
(1990).

D. A. Lewis, T. C. Ward, J. D. Summers, and J. E.
McGrath, "Cure Kinetics and Mechanical Behavior of
Electromagnetically Processed Polyimides", Polymer
Preprints, 22(1), 174 (1988).

J. C. Hedrick, D. A. Lewis, G. D. Lyle, T. C. Ward, and
J. E. McGrath, "Electromagnetic Processing of High
Performance Polymeric Matrix Resins for Composite
Structures", Proceedings for the American Society for
Composites, Fourth Technical Conference, Blacksburg,

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

88
Virginia, October, (1989).

D. A. Lewis, J. C. Hedrick, J. E. McGrath, and T. C.
Ward, "The Accelerated curing of Epoxy Resins Using
Microwave Radiation", Polymer Preprints, 21(2), 330

(1987).

A. Gourdenne, "Possible use of the microwave in polymer
science", International Conference on Reactive
Processing of Polymers

E. Karmazsin, P, Satre, and J. F. Rochas, "Use of
continuous and pulsed microwaves for quick
polymerization of epoxy resins: study of some
thermomechanical properties:, Thermochimica Acta, 22,
305 (1985).

L. K. Wilson and J. P. Salerno, "Microwave curing of
epoxy resins", Technical Report, AVRADCOM TR-78-46,
(1978).

N. S. Strand, ”Fast micrwoave curing of thermoset
parts", Modern Plastics, 51(10), 64 (1980).

J. Jow, Dissertation for the Doctor of Philosophy,
Chapter 3, Department of Chemical Engineering, Michigan
State University, 1988.

J. Wei, B. Thomas, and M. C. Hawley, 37th International
SAMPE Symposium and Exhibition, Anaheim CA, March,
(1992).

J. Wei, Dissertation for the Doctor of Philosophy,
Chapter 5, Department of Chemical Engineering, Michigan
State University, 1992.

L. A. Fellows and M. C. Hawley, "Microwave Processing
of LARC-TPI Polyimide Composites", ICCM/VIII, Honolulu,
Hawaii, July, (1991).

V. Adegbite and M. C. Hawley, AIChE 1993 Summer
National Meeting, Seattle, WA, August, (1993).

J. Microwave, Rubber Chem. Technol., 53(1), 294 (1991).
C. S. Wang, Rubber Chem. Technol., 51(1), 134 (1984).
Threshold Limit Values and Biological Exposure Indices

for 1988-1989, American Conference of Governmental
Industrial Hygienists, Cincinnati, Ohio, p103, (1988).

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

S4.

55.

89

A. L. VanKoughnett and J. G. Dunn, J. Microwave Power,
3(1), 101 (1973).

J. Wei, Dissertation for the Doctor of Philosophy,
Chapter 7, Department of Chemical Engineering, Michigan
State University, 1992.

J. Jow, Dissertation for the Doctor of Philosophy,
Chapter 2, Department of Chemical Engineering, Michigan
State University, 1988.

R. W. Meyer, "Handbook of Pultrusion Technology",
Chapman and Hall, New York, p.1, (1985).

J. Wei, et. al., SAMPE J., 21 (1), 33 (1991).

W. I. Lee and G. S. Springer, J. Comp. Mater., 12, 387
(1984).

W. I. Lee and G. S. Springer, J. Comp. Mater., 12, 357
(1984).

A. L. VanKoughnett and J. G. Dunn, J. Microwave Power,
a (1), 101 (1973).

S. H. Munson-McGee, "Opportunities for Innovation:
Polymer Composites", NIST GCR 90-577-1, August, (1990).

A. Cooper and J. M. Methven, UK Pat. 2,245,893 (to the
John Shaw Limited), Oct. 7, (1990).

S. R. Ghaffariyan and J. M. Methven, Mat. Res. Soc.
Symp. Proc., 122, 135 (1991).

J. Wei, et. al., 6th Annual ASM/ESD Advanced Composites
Conference/Exposition, Detroit, MI, Oct. (1990).

Threshold limit values and biological exposure indices
for 1988-1989, American Conference of Governmental
Industrial Hygienists, Cincinnati, Ohio, p.103 (1988).

G. Odian, "Principle of Polymerization", John Wiley &
Sons, Inc., (1991).

C. Chern and G. W. Poehlein, Polym.-Plast. Technol.
Eng., 22(586), 577 (1990).

R. Dhulipala, G. Krieg, and M. C. Hawley, ACS Meeting,
Denver, CO, Aug., (1992).

C. Walling, J. Am. Chem. Soc., g1, 441 (1945).

56.

57.

58.

59.

90

"Heat capacity of aluminum oxide table", J. Am. Chem.
Soc., 15, 522 (1953).

R. F. Harrington, "Time-Harmonic Electromagnetic
Fields", McGraw-Hill Book Company, Inc., (1987).

J. Wei, Dissertation for the Doctor of Philosophy,
Chapter 8, Department of Chemical Engineering, Michigan
State University, 1992.

B. Carnahan, H. A. Luther, and J. O. Wilkes, "Applied
Numerical Methods", Robert E. Krieger Publishing
Company, Inc., (1964).

"7'1111111111111111