synthesis and characterization of boron-containing
Post on 27-Nov-2021
4 Views
Preview:
TRANSCRIPT
W&M ScholarWorks W&M ScholarWorks
Dissertations, Theses, and Masters Projects Theses, Dissertations, & Master Projects
1996
Synthesis and characterization of boron-containing polymeric Synthesis and characterization of boron-containing polymeric
materials for neutron shielding applications materials for neutron shielding applications
Michael B. Glasgow College of William & Mary - Arts & Sciences
Follow this and additional works at: https://scholarworks.wm.edu/etd
Part of the Nuclear Commons, Polymer Chemistry Commons, and the Polymer Science Commons
Recommended Citation Recommended Citation Glasgow, Michael B., "Synthesis and characterization of boron-containing polymeric materials for neutron shielding applications" (1996). Dissertations, Theses, and Masters Projects. Paper 1539623878. https://dx.doi.org/doi:10.21220/s2-pj26-2v30
This Dissertation is brought to you for free and open access by the Theses, Dissertations, & Master Projects at W&M ScholarWorks. It has been accepted for inclusion in Dissertations, Theses, and Masters Projects by an authorized administrator of W&M ScholarWorks. For more information, please contact scholarworks@wm.edu.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter tace, while others may be
from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in reduced
form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6" x 9" black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly to
order.
UMI A Bell & Howell Information Company
300 North Zeeb Road, Ann Arbor MI 48106-1346 USA 3I3n61-47oo soots2I-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
SYNTHESIS AND CHARACTERIZATION OF BORON-CONTAINING
POLYMERIC MATERIALS FOR NEUTRON SHIELDING APPLICATIONS
A Dissertation
Presented to
The Faculty of the Department of Applied Science
The College of William and Mary in Virginia
In Partial Fulfillment
of the Requirements for the Degree of
Doctor of Philosophy
by
Michael B. Glasgow
1996
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 9701088
UMI Microform 9701088 Copyright 1996, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized copying under Title 17, United States Code.
UMI 300 North Zeeb Road Ann Arbor, MI 48103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPROVAL SHEET
This dissertation is submitted in partial fulfillment of
the requirements for the degree of
Doctor of Philosophy
Michael Bruce Glasgow •
Approved, May, 1996
Committee Chairman/Thesis Advisor
D~~ Dr. Denn~s Manos
Dr. Robert L. Vold
Dr. She1la A. Th1beault NASA-Langley Research Center Hampton, Virginia
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
ACKNOWLEDGMENTS
LIST OF TABLES
LIST OF FIGURES
ABSTRACT
CHAPTER 1. INTRODUCTION
1.1 Galactic Cosmic Radiation
1.2 Neutron Capture
1.3 Boron-Containing Materials
1.3.1 carborane-Polymers
1.3.2 Filled Epoxy Resins
1.4 Applications
1.4.1 Aerospace Applications
1.4.2 Nuclear-Powered Ships
1.4.3 Accelerator Facilities
CHAPTER 2. EXPERIMENTAL STUDIES
2.1 Boron-Epoxy Materials
2.1.1 Fabrication and Processing
2.1.2 Thermal Characterization
2.1.3 Mechanical Tests
2.2 Carborane Polyamides
2.2.1 Synthesis and Characterization
2.2.2 Structure Confirmation
2.2.3 Molecular Weight Determinations
2.2.4 Thermal Characterization
iii
Page
v
vi
viii
xii
3
10
13
19
22
25
27
28
30
35
39
44
48
50
52
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.3 Boron-loaded Polyimide Films 59
2.3.1 Synthesis 60
2.3.2 Film Characterization 63
2.4 Carborane Polyimides 67
2.5 Neutron Exposure 72
2.6 Surface Characterization Methods
2.6.1 X-ray Photoelectron Spectroscopy 76
2.6.2 static Secondary Ion Mass Spectrometry 77
CHAPTER 3 • RESULTS AND DISCUSSION
3.1 Boron-Epoxy Materials 79
3.2 carborane Polyamides 144
3.3 PMDA-ODA Films 159
CHAPTER 4. APPLICATIONS AND FUTURE WORK
4.1 Introduction 162
4.2 Aerospace 163
4.3 Nuclear Power 165
4.4 Accelerator Facilities 166
CHAPTER 5. SUMMARY AND CONCLUSIONS 167
LITERATURE CITED 172
APPENDICES 1-4 183
VITA 245
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGMENTS
This work was supported by the NASA Space Exploration Initiative (SEI)
Program, the Virginia Space Grant Consortium (VSGC), and the Department of
Applied Science at the College of William & Mary.
I wish to express my deepest appreciation to Dr. Orwoll, under whose
professional guidance this investigation was undertaken, and to thank him for all the
patience, advice, helpful criticism, and grammar lessons he bestowed upon me
throughout my graduate career. I thank Dr. Thibeault and Dr. Ed Long for their
navigational skill in sailing the NASA-Langley research facility. I am very grateful
for the thoughtful contributions and encouragement of Dr. Kiefer, Dr. Manos, and
Dr. Void.
I would like to thank all of my friends who helped me to complete this work,
especially Dr. Donnie Sandusky, Dr. Dave Hood and Captain Phil Smith (and their
wives) for their encouragement and close friendship. My sincerest thanks are
extended to Sheeba, Dave E., and the others in The Orwoll Group who made life
memorable in the laboratory.
Most of all, I would like to express my heart-felt gratitude to my wife
Bonnie, my love and companion for life, for her emotional, physical and fmancial
support of my graduate pursuits. I owe you much more than I can ever repay.
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table
1.
2.
LIST OF TABLES
Elemental analyses of boron-loaded PMDA-ODA
poly(pyromellitimide) films.
Compositions of boron-epoxy formulations.
3. Theoretical and actual densities of boron
epoxy materials.
4. Glass transition temperatures of boron-epoxy
materials.
5. Initial DMA frequencies and glass transition
temperatures of boron-epoxy materials.
6. Mechanical properties of boron-epoxy materials
from DMA measurements.
7. Thermal stability of boron-epoxy materials.
8. Coefficient of thermal expansion values from
TMA measurements.
9. CTE comparison between TMA and interferometer
measurements.
10. Flexure strength of boron-epoxy materials.
11. Compressive properties of boron-epoxy
materials.
12. Reaction efficiency of boron-epoxy materials
in LEO orbit.
13. 1H-NMR peak identifications of aliphatic
carborane-polyamides.
vi
Page
63
80
82
90
92
95
95
100
105
115
122
143
147
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14.
15.
Narrow molecular weight standards for GPC.
Solution viscosities of carboJ.:ane-polyamides
in DMAc.
vii
149
150
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure
1.
2.
LIST OF FIGURES
Average energies of space radiations.
Cascade reactions of GCR particles.
3. Attenuation of 605 MeV amu·• iron beams in
various weight fractions of boron-containing
polyetherimide.
4.
5.
6.
7.
8.
Boron (n,a) nuclear reaction.
Structures of carborane isomers.
Exopolyhedral ring structure in o-carboranes.
Repeat unit of TGMDA epoxy resin crosslinked
with DDS.
Dose equivalents from GCR received on air
carrier flights.
9. Schematic drawing of neutron shielding
Page
5
6
9
12
15
17
21
24
materials in nuclear submarines. 26
10. Ultrasonic C-scan of boron-epoxy plaque. 32
11. Mechanical configuration of DMA sample arm. 37
12a. Schematic diagram of laser-interferometric
dilatometer system. 39
12b. Schematic diagram of Fizeau-type
13.
14.
interferometer.
Schematic diagram of dc{dt crosshead
displacement instrument.
ASTM tensile test dogbane with composite tabs.
viii
40
41
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15. Text fixture for flexure samples. 44
16. Functionalization reactions of m-carborane. 46
17. Diamines used in polyamide synthesis. 49
18. Synthesis of m-carborane polyamides. 51
19. 1H-NMR spectrum of MC-MDEA polyamide. 52
20. 13C-NMR spectrum of MC-MDEA polyamide. 53
21. GPC chromatogram of MC-1,12 DDA. 55
22. GPC molecular weight calculations for MC-1,12
DDA. 56
23. GPC universal calibration curve. 57
24. Synthesis of PMDA-ODA polyimide. 62
25. FT-IR spectrum of PMDA-ODA polyamic acid. 65
26. FT-IR spectrum of PMDA-ODA polyimide. 66
27. Reaction scheme for "dry" and "wet" m-
carborane-diamine synthesis. 68
28. Friedel-Crafts m-carborane "protected" amine
synthesis. 71
29a. Schematic drawing of Pu/Be source container. 73
29b. Neutron exposure rod with sample mount. 75
30. Principal neutron reactions in Pu/Be source. 73
31. Counting reaction of beta-emitting indium
isotope.
32. Schematic illustration of the PHI Model 560
ESCA/SAM.
33. Optical surface photograph of boron-epoxy
material (200X].
ix
76
78
81
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
34. Optical micrograph of 10%B-Epoxy [lOOOX]. 83
35. DSC cure isotherm at 177°C of 934 epoxy resin. 87
36. DSC thermogram of pure epoxy. 89
37. DMA of 20%B-Epoxy and 20%B-mix materials. 94
38. Plot of char yield vs. percent boron. 96
39. TMA plot of 10%B-Epoxy runs. 98
40. TMA plot of 20%B-Epoxy runs. 99
41. Schematic diagram of interferometer-
dilatometer sample holder. 102
42. Graph of CTE for 10%B-Epoxy from dilatometer
instrument. 103
43. Graph of CTE for 20%B-Epoxy from dilatometer
instrument. 104
44. Tensile properties of boron-epoxy materials. 108
45. Optical micrograph of 20%B4C-Epoxy [200X]. 109
46. Tensile strength of boron-loaded epoxy
materials. 110
47. Tensile strain of boron-loaded epoxy
materials. 112
48. Breaking stress points of flexure specimens. 114
49. Flexure strength of amorphous boron-epoxy
50.
51.
materials.
Schematic diagram of stresses in 4-point
flexure test.
Optical micrographs of boron clusters.
52. Compressive strength of boron-epoxy
X
116
117
119
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
53.
54.
materials.
Compressive modulus of boron-epoxy materials.
Plot of initial activity of indium foil, pure
epoxy-shielded indium foil.
55. Graph of initial activity of boron-epoxy
56.
57.
58.
59.
60.
61.
62.
shielded indium foils.
Neutron absorption of boron-epoxy materials.
Normalized graph of neutrons absorbed vs.
weiqht percent boron.
Photoelectric effect in XPS.
TGA curves of linear aliphatic carborane
polyamides.
TGA curves of MC-MDEA and MC-MDMA in air.
TGA curve of MC-ODA in air.
TGA curve of MC-ODA in argon.
63. Neutron absorption of boron-loaded PMDA-ODA
films.
xi
123
124
128
129
130
132
135
153
155
156
157
161
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
The development of boron-containing polymeric materials for neutron shielding applications was undertaken. Three types of materials were characterized for physical and thermal properties: boron powder-filled epoxy composites, carborane polyamides having boron chemically bonded into the polymer, and boron-loaded polyimide thin films. Addition of amorphous submicron boron powder did not affect significantly the thermal performance of the epoxy. The 17% boron loading produced a 26% increase in compressive failure strength and a 68% increase in the compressive modulus. 0.125 inch thick specimens containing 17% boron absorbed 92% of incident neutrons from a 5-curie PU/Be source compared with <1% for the neat epoxy. Dispersion of the boron in the epoxy was improved with the addition of larger size crystalline boron powders. Carborane polyamides containing up to 35% boron were thermally stable up to 400°C in air. The polymers had hydrogen/boron ratios from 2.0 to 3.8 and were soluble in several organic solvents. Polymer solutions were processed into clear, colored thin films. Boron-filled polyamic acid solutions of a PMDA-ODA polyimide containing up to 10% boron were processed into thin films. Neutron absorption of the opaque films measured in a 5-Curie PufBe neutron source was linear with boron concentration and film thickness. The fraction of neutrons absorbed varied linearly with boron concentration and film thickness. The applicability of boron-containing materials to the aerospace, nuclear power and accelerator industries was investigated.
xii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
SYNTHESIS AND CHARACTERIZATION OF BORON-CONTAINING
POLYMERIC MATERIALS FOR NEUTRON SHIELDING APPLICATIONS
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER ONE: INTRODUCTION
Radiation shielding materials are vital to the
development of future modes of transportation, nuclear
energy sources, and nuclear safety models. In applications
ranging from the exploration of planets to the containment
of nuclear waste and the safe operation of nuclear reactor
facilities, shielding materials will play a crucial role in
the technology of tomorrow.
It is the nature of humans to reach beyond their means
to attain that which was thought to be unattainable. The
national space program, under the management and direction
of the National Aeronautics and Space Administration, has
captured the imaginations of scientists and engineers for
three decades with the lunar landing in 1969. An entire
generation of Americans have wondered to themselves, "What
else is out there?"
A manned mission to Mars has been one of the goals of
the space exploration effort. The development of new
2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
strong, lightweight materials is a large factor in the
calculation of costs of manned missions to other planets.
In addition to being lighter than conventional metal
structural components of spacecraft, polymers and composite
materials have a greater strength per unit weight and more
durability without deformation under extreme loads.
However, the deep space environment beyond the van Allen
radiation belts that surround and protect Earth from
galactic cosmic radiation (GCR) is hazardous to most known
materials. Radiation shielding materials that combine the
outstanding mechanical and thermal properties of polymeric
composites with protection from the deleterious effects of
cosmic radiation would effectively serve a dual purpose:
structure and shielding.
In order to fully understand the requirements of
shielding materials in spacecraft, it is necessary to know
the composition and energies associated with galactic cosmic
radiation.
1.1 GALACTIC COSMIC RADIATION
There are two principal sources of radiation entering
the Earth's atmosphere at any given time: solar cosmic
radiation (SCR) and galactic cosmic radiation (GCR). The
sun emits low-energy (<1 GeV amu-1) cosmic rays which
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
contribute to the flux of background radiation. The 11-
year cycle of solar activity modulates the amount of
radiation observed. In addition, during periods of high
solar flare activity, the sun emits large numbers of charged
particles, which affect the total flux of radiation measured
on Earth. When sun spot activity is high, the SCR flux is
decreased because of the sun's increased magnetic field,
which deflects low-energy particles.
Highly energetic GCR cosmic rays (1-105 GeV amu-1) come
from a variety of sources, including supernova explosions
and stars in our galaxy (intragalactic GCR) and from beyond
our galaxy (extragalactic GCR). Intragalactic cosmic
radiation, the principal source of GCR for Earth orbit,
consists of highly penetrating charged particles, creating
the potential for long-term exposure risk at high altitudes
and at the Earth's magnetic poles. Figure 1 summarizes flux
density vs. particle energy curves of the common components
of cosmic radiation.• Although the flux of GCR is very low,
the average energy per particle (102 to 104 MeV) is the
highest of the background radiation types affecting Earth.
Therefore, development of shielding materials for effects of
GCR is important for applications that involve long-term
travel at high altitudes on Earth and in Earth orbit.
Perhaps the strongest reason for radiation shielding in
space is the interaction of GCR with atomic nuclei in
spacecraft hull materials. GCR is composed almost
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1Q15 Solar wind protons
Auroral electrons
(outer zone)
Solar stonn protons
IQ-2 100 Panicle energy. MeV
Trapped protons (inner zone)
Fiqure 1 Average energies of space radiations. 1
completely of protons, electrons, positrons and helium
nuclei, but a small percentage (<2%) consists of larger
nuclei and elements with atomic numbers up to Z=90. These
energetic, highly charged heavy particles cause the most
damage to materials and tissue. GCR particles impacting on
materials in space cause cascade reactions which result in a
flux of sub-atomic product particles, including protons and
neutrons. Figure 2 shows the development of cascades caused
by incident proton and alpha particle (helium nuclei)
radiation. 2 Secondary protons, being positively charged,
are stopped effectively by Coulombic interactions with
matter. The secondary neutrons are uncharged and have a long
mean free path. Neutrons may penetrate far into the inside
5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
p
,--, \ \
-~
Proton initiated cascade
Protons
Neutrons a
Alpha initiated cascade
Fiqure 2 Cascade reactions caused by GCR particles. 2
compartment of a spacecraft. Neutrons are slowed to thermal
energies by elastic collisions with atomic nuclei. At
thermal energies, neutrons are absorbed by the nuclei of
target elements at a probability proportional to their
respective high thermal neutron cross sections.
which absorbs thermal neutrons forms a compound
A nucleus
nucleus/excited state atom which breaks apart into a variety
of decay products, including protons and neutrons. These
product particles can react with more nuclei to begin a
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
cascade series of reactions, as depicted in Figure 2. This
series of "chain" reactions is the reason why secondary
shieldinq materials are needed, even thouqh the flux of
heavy, enerqetic primary ions is such a small component of
qalactic cosmic radiation.
Structural materials, electronic devices, and human
tissue can be damaqed by the effects of the secondary
particles qenerated by the exposure of primary shieldinq
materials to GCR. These effects can be manifested in loss
of plasticity and mechanical properties for polymer-based
composites, sinqle event upsets (SEU) of electronic devices,
and possible cancer risk caused by cell damaqe in tissues.
Theoretical study of SEUs and tissue damaqe in mouse cells
is a topic of much research interest. The ramifications of
tissue studies could possibly affect the future of manned
space missions. As important as these issues are, the scope
of this work will be dedicated to the effects of secondary
neutrons on polymeric materials.
The effects of qalactic cosmic radiation on shieldinq
materials have not been demonstrated yet in manned
spacefliqht experiments. The Lonq Duration Exposure
Facility (LDEF) was recently retrieved from low Earth orbit
with the Space Shuttle, and shorter term experiments have
been conducted on board the Shuttle. But to qain a true
measure of the effects of GCR on polymeric structural
7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
materials requires a mission to deep space.
Researchers at NASA-Langley Research Center and
colleagues at research universities have developed computer
methods and transport codes that estimate the flux of
charged heavy particles and their impact effects on atoms of
materials. 3 In her recently published dissertation, Kim
utilized a heavy ion/nucleon transport code for space
radiations (HZETRN) to estimate effects of heavy ions (in
particular Fe+~) on a variety of shielding materials. 4
Kim's study concluded that shielding materials for GCR
should have a high shielding content (polymers) and that
boron-containing polymeric materials showed promise in their
effectiveness to absorb secondary neutrons from cascade
reactions. The exposure dose rates of various shielding
materials were calculated. Kim's plot of a polyetherimide
containing different percentages of boron is reproduced in
Figure 3. 5
Boron-containing polymeric materials can serve at least
two purposes simultaneously in future spacecraft missions:
to provide secondary neutron shielding to prevent the
deterioration of materials and devices inside the spacecraft
and to serve as structural support members to avoid
parasitic weight and volume in spacecraft design. Another
possible benefit would be reduction of costs of duplication
in electronic circuits and "radiation-hardened" components.
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
~ 0.010 E u
'-..... ~ 0.008 u
....... I...
g_ 0.006
Q)
u ~ 0.004 :::J
u_
0.002
• • • • • Polyetherimide/20 ~ 8 G&eee Polyetherimide/15 ~ 8 G&ea£J Polyetherimide/1 0 ~ 8 ~ Polyetherimide/ 5 ~ 8 1 1 1 1 1 Polyetherimide
0. 000 +--,r-T""-r-r--r--r-r--,-......,..-r-1~-r-r--,-..,......,... ........... 3 13 23
Atomic number of projectile fragment
Figure 3 Attenuation of 605 MeV amu-1 iron beams in various weight fractions of boroncontaining polyetherimide (18 gfcm2). 5
The approach to developing polymeric materials with high
boron content was essentially two-fold. The first method
was to fabricate boron-filled epoxy laminates to establish
the baseline mechanical and thermal properties of a boron
powder-filled material. The second method was to synthesize
novel boron-containing polymers with varying hydrogen
content for evaluation as possible matrix resins for fiber-
based composites.
The following section describes the thermal neutron
9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
capture process for boron-10, an isotope of boron with a
high neutron capture probability.
1. 2 NEUTRON CAPTURE
Neutron capture processes are reactions between nuclei
and neutrons. The probability of neutron capture is related
to the neutron capture cross section, a, expressed in barns
or 10·24 cm2 • It is a measure of the "largeness" of the
target nucleus as it appears to the incident neutron.
Neutron capture cross section data are tabulated for nearly
all isotopes of the elements in the periodic table. Cross
sections for most isotopes range from 0.002 to 1.0 barns.
some nuclei, such as 113cd, have cross sections over 104
barns. 6 The data for target nuclei are usually reported for
thermal neutron capture cross section -- capture of thermal
neutrons. Thermal neutrons have energies of less than 0.1
eV at standard ambient temperature. Because they are slow
moving, thermal neutrons are the most easily captured
species of the neutron energy spectrum. Most of the thermal
neutron capture data has been accumulated from neutron
activation studies and nuclear research efforts by
scientists in the United States and the Soviet Union. 7
In a nuclear reactor, neutrons are produced from
epithermal energies (below thermal energies) to ultrafast
energies of about 10n eV. Fast neutrons (1-10 MeV) are
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
generally not harmful to organic tissue or electronic
components since the cross section for fast neutron capture
is extremely low for most elements. In addition, since
neutrons do not encounter coulombic charge interactions with
matter, the mean free path of fast neutrons is extremely
long, and they pass through most organic material readily.
However, fast neutrons are slowed to thermal energies by
elastic collisions with atomic nuclei. These collisions are
more effective with decreasing nuclear size. Hydrogen
containing materials, including most organic polymers, are
most effective in thermalizing neutrons. Effective neutron
shielding materials can be made of hydrogen-containing
materials that have 1) elements with a large neutron capture
cross section for a range of neutron energies and 2) a high
concentration of neutron absorbing nuclei. In addition,
increasing the thickness of shielding materials provides a
greater volume in which to slow neutrons to thermal
energies.
In this work, neutron capture experiments focused on
boron-10, an isotope of boron (19.8% natural abundance) with
an unusually large thermal neutron capture cross section,
about 3838 barns. 8 The neutron capture nuclear reaction is
summarized in Figure 4. 9 Boron-10 absorbs a neutron,
forming a compound nucleus excited state 118 atom, which
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10S + 1n ----> [ 11B] 6% ----> 7Li + 4He + 2. 79 MeV
94% ----> 7Li• + 4He + 2. 31 MeV
----> 7Li + ~ + 2.31 MeV
Figure 4. Boron(n,a) nuclear reaction.
instantaneously breaks apart into an energetic lithium atom
and alpha particle. The capture products, lithium and an
alpha particle, are not radioactive and are easily stopped.
The energy of these atomic particles, about 2.4 MeV, is
dispersed over a radius of only about 10! in tissue. 10
Boron-containing polymeric materials are excellent
candidates for neutron shields. Polymers, composed of
mostly carbon and hydrogen, can slow intermediate-to-high
energy neutrons to thermal energies. Boron can stop thermal
neutrons with a relatively high efficiency because of its
high neutron capture cross section. Ideally, thick
polymeric materials with high boron content should provide
good protection from neutrons of a range of energies. In
addition, the higher the boron content, the greater the
neutron attenuation. For cost-sensitive applications where
volume and weight savings are critical, boron-containing
polymeric materials can provide a combination of good
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
performance with a low cost/volume ratio.
1.3 BORON-CONTAINING MATERIALS
1. 3.1 CARBORANE POLYMERS
Introduction
Most of the early work on incorporating icosahedral
carborane moieties into polymers can be attributed to large
scale industrial laboratory efforts in the 1960's. Reasons
for the research and development of carborane polymers
include improving thermal stability, resistance to oxidative
degradation, and resistance to chemical attack. The
rationale for these characteristics lies in the crystalline
lattice of carborane which, in addition to its inherent
structural stability, acts as an energy sink, absorbing
thermal energy from adjacent bonds in the polymer. Several
promising carborane-polymers were discovered, most notably
the poly(carborane siloxanes) developed by Green and
coworkers. 11"13
High-temperature, stable, carborane-based polymers,
such as the "PNP" and "POP" polymers which have phosphorus
nitrogen-phosphorus and phosphorus-oxygen-phosphorus groups,
respectively, between carborane groups in the polymer
backbone had poor processibility, were insoluble in common
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
organic solvents, and were prohibitively difficult to
synthesize. Polymers with sidechain carborane qroups were
easier to synthesize, but their thermal and mechanical
properties suffered. These unwanted characteristics,
coupled with the advent of high-temperature aromatic
polyimides such as Kapton® and the LaRcTm polymer series,
resulted in declining interest in carboranes as high
temperature, high performance polymers.
Although some of the research remains in proprietary
databases, the published literature of polycarboranes
("carbaboranes" in the Russian literature) is extensive. A
review of the chemistry of carboranes, in particular
icosahedral carboranes, with emphasis on the synthetic
routes, thermal and mechanical properties, and neutron
absorbing capabilities follows. carborane polymers are
typically classified into two groups: Type I -- polymers
with carbon-carbon carborane linkages and Type II -
carboranes with inorganic linkages. This discussion will be
restricted to the organically linked carborane polymers.
carboranes -- General Remarks
Carboranes, named for the two primary constituent
elements, have been isolated in many different forms and
formulas. The three most common classifications of
carboranes are the nido carboranes, the arachne carboranes,
and the closo carboranes. By far the most synthetically
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
important carboranes of the three types are the icosahedral
carboranes of the close family. The discussion of synthetic
polymer chemistry is henceforth limited to the icosahedral
carboranes.
Xcosahedral carboranes
Icosahedral carboranes are clusters of boron, carbon,
and hydrogen which have the general formula, c;B1oH12 • There
are three isomers of these dicarbacloso-dodecaboranes,
ortho-, meta-, and para-, in which the carbon atoms occupy
respectively the (1,2), (1,7), and (1,12) positions in the
dodecahedron lattice (Figure 5) • 14 Their preparation from
decaborane and acetylene was reported by Grimes. 15
F iqure 5 • Structures of 1, 2-, 1, 7-, and 1, 12 -<;B1oH12 • 14
Isomerization of Q-carborane to m-carborane occurs via a
high temperature rearrangement at about 650°C. 16 Heating of
carborane to 700°C produces a 75%/25% mixture of the meta
and para isomers, which are separable on columns loaded with
basic alumina using a petroleum ether mobile phase. Because
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of the expense involved in isomerization and purification of
R-carborane, the isomer is no longer produced commercially.
For this reason, discussion of the icosahedral carboranes
will be limited to that of m-carborane and its extensive
derivative chemistry.
Although the chemistry of the ortho isomer is best
known, _extensive derivative chemistry has been reported for
all three isomers, including reactions to form diol, diacid
chloride, diazide, dichlorosulfonate, dihydroxymethyl,
dianhydride and diamine functional groups. The difunctional
carborane monomers can be homo- or co-polymerized to form a
variety of polymers including polyformals17 , polyesters1B-20,
polyethers21 , polyamidesn, polysulfonesn, polyacrylatesu,
polyurethanes25 , and polyimides26•27 • Meta- and para-carborane
isomers are of greater versatility for synthesizing
thermally stable polymers owinq to decreased steric
interactions of the bulky carborane polyhedron. In
addition, Papetti and coworkers28 have reported that o
carboranes can underqo cyclization to yield exopolyhedral
ring structures (Figure 6}, especially when the Q-carboranyl
qroup is attached to silicon in siloxane monomers.
Poly(carborane siloxanes) containinq m-carborane groups in
the polymer backbone have found the most utility, and their
thermal and mechanical properties are best characterized.~
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 6. Exopolyhedral ring structure in o-carboranes. 28
Today the polymer chemistry of the icosahedral
carboranes is limited to reactions involving m-carborane
moieties for their synthetic utility, ability to generate
long-chain linear polymers, and lower cost when compared to
the para isomer.
H-carborane-containinq polymers
Linear polymers containing m-carborane groups possess
outstanding thermal stability and more synthetic facility
than those with Q-carborane units in the polymer backbone.
Carborane polyamides and polyimides in particular are stable
to over 250°C in air and are processible into thin films.
Polyimides having phenylene rings display markedly higher
thermal stability than their polyamide counterparts. The
polymerization of carborane polyimides has been reportedly
achieved starting from m-carborane-1,7-diamine~ and
dianhydrides.
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Polyaai4aa
The first m-carborane-containinq polyamides were
synthesized by Korshak and coworkers in 1969. 31 The reaction
of m-carborane-1,7-diacid chloride with various diamines was
studied extensively in the following years.n
The objective of this early work was to develop polymers
with unusual thermo-oxidative stability and solubility in a
variety of common organic solvents. Reaction conditions
were studied, including the influence of solvent on the
molecular weightn and morphologiM of the polymers.
Preparation of polymers in amide solvents yielded low
molecular weight polymers, owing to the high activity and
poor selectivity of the carborane acid chloride.
Transparent m-carborane polyamide films were cast, and the
increased vulnerability of the amide bonds to cleavage in
the presence of water was noted.~
The present study examines the synthesis and
characterization of several carborane-containing polyamides
not previously reported in the literature. Solubility in
various solvents was investigated, as well as determination
of the thermo-oxidative stability and properties of films
cast from tetrahydrofuran.
An additional goal of this work was to prepare
polyamides with a range of hydrogen/boron ratios. Since
neutron attenuating polymers contain large amounts of
hydrogen to slow fast neutrons to thermal energies,
18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
investigation of the hydrogen content in polymer with
naturally high boron content is a reasonable task. Thin
films cast from concentrated solution were evaluated for
creasability, brittleness, color, and transparency.
Evaluation of the polyamides prepared as described in the
Experimental section as neutron-shielding materials would be
straightforward.
1.3.2 FILLED EPOXY RESINS
The polymer chemistry of epoxies has been known for
over thirty·years. The reader is referred to several
excellent reviews of epoxy polymers and their history,
preparation, and rich derivative chemistry.~39
Epoxies are thermosetting polymers which have high
resistance to chemicals and solvents, exhibit tack and
adhesion to a variety of different materials, and possess
good electrical and impact properties. The primary uses of
epoxies are as surface coatings and as structures in weight
bearing applications. coating applications include
automotive and aerospace primers, pipe and floor finishes,
and applications where heavy wear resistance is required.
Structural applications are found in the construction and
electrical industries, as well as in composite materials
manufacturing. About 300 million pounds per year of epoxy
polymers are produced in the United States.~ Their
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
relatively low cost and the breadth of the processing
science of epoxies explains why epoxy resins are one of the
most widely used polymeric materials today.
In this study, an epoxy resin based upon the
tetraglycidyl ether of methylenedianiline (TGMDA) premixed
with a diaminodiphenylsulfone (DDS) crosslinking (or curing)
agent will be evaluated for its basic structural and thermal
properties when combined with boron powder. The repeat unit
of the epoxy is shown within the dashed lines in Figure 7.
In addition to possibly improving mechanical
properties, loading the epoxy with boron powder should
produce a composite material with outstanding neutron
shielding properties. Elemental boron powder contains
almost 20% 1~, which has a high thermal neutron capture
cross section, as discussed in Section 1.2 above. By
combining the structural reinforcement and neutron absorbing
features of boron powder with a thermally stable, chemically
resistant epoxy matrix resin, load-bearing structural
materials can be fabricated with a lower density than
conventional metal building materials such as aluminum or
steel. Several applications depend on weight and volume
conservation in order to limit the cost of manufacture and
functional use, including spacecraft and nuclear-powered
ships. This work will investigate the attractive properties
of boron-loaded epoxy composites and the possible use of
the materials in several key applications.
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
, I
I
v I
1- -.c-::
I
1- -I c-:
I .....
·...;
I :--::::: .... -
Fiqure 7 Repeat unit of TGMDA epoxy resin crosslinked with DDS.
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.4 APPLICATIONS OF NEUTRON-SHIELDING MATERIALS
1.4.1 AEROSPACE
Spacecraft
The benefits of a lightweight, strong neutron shielding
material in spacecraft applications are many. As discussed
above, the outer hull of a spacecraft is exposed to ionizing
radiation from galactic cosmic rays, which results in a
cascade of neutrons, protons, and other sub-atomic particles
within the hull materials. Exposure of electrical
components and humans aboard manned space missions to
neutrons results in damage to circuitry and tissues. Single
event upsets and permanent damage to even radiation-hardened
components can jeopardize missions. In addition, protection
of human occupants of space vehicles from secondary
particles can lower the incidence of cancer and other health
problems. 41
The materials used to build spacecraft must serve more
than one purpose (neutron shielding) to be cost effective.
Because of shrinking NASA space budgets, developing
materials which can function as structural members as well
as neutron shields is critical. One overriding criterion in
the design of materials for spacecraft construction is that
they be lightweight, yet strong. In this study, low
density, boron-containing polymeric materials with good
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
mechanical properties and neutron absorption are
investigated.
Biqb-altituda aircraft
Airplanes which cruise at high altitudes have a small,
y~t significant exposure risk to ionizing radiation from
GCR. One recent Federal Aviation Administration study
estimated dose equivalents for nonstop one-way flights of
long duration.~ The results are found in Figure 8. The
long, transoceanic flights from New York to Tokyo and from
Athens to New York had the highest dose equivalents (99, 93
microsieverts respectively). These intercontinental flights
typically fly at altitudes from 30,000 to 38,000 feet. The
exposure risk of continuous service on these high-altitude
flights is a factor to consider for aircrews and pregnant
women, and the FAA study provides recommendations to
minimize exposure to the effects of GCR.
Secondary shielding may be needed in the future for
high-altitude aircraft. The NASP (National Aero-space
Plane) and HSR (High Speed Research) aircraft were designed
to travel at ultra-high altitudes. HSR aircraft may fly at
altitudes up to 70,000 feet with more than double the
exposure dose to ionizing radiation. Neutrons produced from
cascade reactions of aircraft fuselage materials with GCR
can be effectively shielded with materials with a high boron
content. This study will examine the effectiveness of
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
liltlt lUlU! •t-tll flit•t m Uact •nrc' I! pall altitdt, Air (att il tile, alact licnainuu • lilliaiauru•
oritia • autiaatioa Uauua Uul ... n• (ailliml l•illirul I 2 3 I ! I
111utoa lwi • uma (usi 21 eJ ... l.l(Utl t.1 1111 auttle (WI • rartlu• (rDll 11 1.1 u 1.1 1.2 liui (WI • fUJI (n&l u u 1.1 1.1 1.1
It Lilli (nLI • tdu lULl l! 1.1 1.1 1.1(1.211 1.1 11111 1upa ln&l • It Llaia (ULI II u 1.2 !.( 1.1 au Jau n ISJII • uui (lUI 3! 1.1 1.! u 2.1 Ill DritUI (Uti • IU <nio (UTI " 1.1 l.l l.l 1.1 lUlU (lUI • liutlfllil (urI u 1.1 1.! (.1 u
Ill tort lUll • Ill JIU rt (IIIII 11 1.1 u UIUI l.! 1.11 .&afdU IWI • IIAaohla IDLI 1S !.1 1.1 n 1.1 cticato (DIDI • ••• tort 1m1 n u 1.1 1.1 1.1 IIAaohh IDLI • Los .<tln (WI (f !.1 u n 1.2 lu.iattaa, D.c. IWI • Lll .<tlu (WI l! (.1 !.1 11 u totJI Jl (UUI • La& <du (WI n ... 1.1 " 1.1 U& .<till (WI • totJO Jl (UUI (f 11.1 11.1 n .., Ill tor11Jill • QICifl (OUI 11 1.1 !.l u !.1
liuupolh IWI • 111 tart (Jill n 1.1 1.1 II !.1 Loadoa II (lml • Dlllll/lt lort• (IRI 11 1.1 11.1 !l !.1 Dallu/ll lort• IIIII - Uadoa II (IWI n 1.! 1.1 u !.3 Suttle (SUI - .&lcbnge (uti l! u 3.1 Zl !.1 Luka ro ILrnl • 111 tort (ml l! u 1.1 u u cticago IDUI - su rrucuco IUOI 11 1.1 1.1 25 u Suttle (SUI • ru•iattll, D.C. (lUI l1 1.1 (.( 1! u Lo .. oa II (IGLLI • Ill tort (Jill n 1.1 1.1 u '·' 111 tort IJUI • Suttle (SUI " (.1 !.3 II u sea rrucuco ural - ctmgo IDUI u l.l 1.1 lt 1.1 tOtJa Jl (UUI • Ill tort (Jill u ll.l 12.1 " u Loalloa II (IGLLI • Los .&agtlu IWI " u.s 11.1 .. 1.1 Qtcata (OIOI • Laadaa II (IGtLI n 1.3 1.1 !I u 111 tart (Jill • totya U (U.&al II IU U.l " lt.ll 1.1 Loadaa II (IGLLI • Qicato (DIOI II 1.1 1.1 n l.l llhu G& (I.Qtl • 111 tart IJUI u 1.1 1.1 ll 1.1 (till
• th •!act ~nn of a llltU ••tia nu th urcnft lmu th •!acts khn talea(( u4 ud nu it ruche th •!acts d ter ludllf. 11 couidlr Ill Uoct ku& accuahtad Ia II comntin auuc 11 a repruutatiu rart year.
' rar ucl lhtlt 11 tstiutd Clt dan -..inlut far air tiu uiat HI fhpt plu, tlliat hta accnat c•utu 11 alutade ud t•auttetlt llmau frn tw•U to ta~clilon.
' lllhumru 11 !II blact h1u • (ISO 1 ticroumrtc it 111 flitUI I (!Itt 1 bloct hm it 111 llitltl.
Fiqure 8 Dose equivalents from galactic cosmic radiation received on air carrier flights. 42
boron-loaded epoxy samples and boron-containing polyimide
films in absorbing incident thermal neutrons. (The cost
considerations that apply to spacecraft also pertain to
commercially-produced aircraft with shielding materials.)
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.4.2 NUCLEAR-POWERED SHIPS
Polymeric materials with excellent thermal, mechanical
and chemical resistance are good candidates for service on
nuclear-powered surface ships and submarines. Volume is a
crucial quantity to be considered in ship design, especially
in submarines. Low-density, high performance polymers can
serve as replacement structural members in some noncritical
applications ships. Although weight savings is not as
important as volume savings, polymeric composites may
eventually find use in marine environments.
Nuclear-powered ships are designed with multiple safety
mechanisms in order to prevent exposure of the crew to
radiation. In particular, nuclear submarines have several
layers of neutron-absorbing materials which reduce the
background radiation level to well below surface background
levels. 43
Although information on the precise volume of shielding
materials currently in use is classified, it is estimated
that a thickness of several inches of each type of shield
exists in state-of-the-art shielding design on nuclear
submarines. These layers may include but are not limited to
polyethylene, borated polyethylene, hafnium-plating, and
lead. A typical arrangement of these layers is shown in
Figure 9. Polyethylene has a high hydrogen content, and
serves to slow fast neutrons to thermal energies. The boron
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
borated PE
reactor core PE
(relative dimensions estimated)
Piqure 9 Schematic of neutron shielding materials in nuclear submarines.
in borated polyethylene absorbs the thermal neutrons,
forming nonradioactive products which are trapped in the
matrix. Lead is typically used for protection from gamma
rays, which are produced in large, energetic amounts in
nuclear reactors. Finally, hafnium coating provides
strength and material integrity while absorbing slow
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
neutrons.
Polymeric materials with higher boron loadings than
borated polyethylene can absorb neutrons more effectively
with less volume. In this work, boron-containing epoxy
materials will be shown to possess outstanding neutron
absorption in a small material thickness.
1.4.3 ACCELERATOR FACILITIES
Electron accelerators and nuclear reactors used in
boron-neutron cancer therapy (BNCT) produce a small number
of neutrons from collisions of accelerated particles with
the inner walls of the devices. Neutrons, having a long
mean free path because of their neutral charge, can escape
the matter of the accelerators and cause a small but
measurable flux in the outside environment. Although the
flux of neutrons that emerge from the primary shielding
materials is low, the need for some shielding from the
neutrons exists.
Materials developed in this work can be applied to
shielding of neutrons from accelerators. Polymeric
materials with a high boron content may be used to reduce
the level of neutron radiation to below background levels.
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER TWO: EXPERIMENTAL STUDIES
2.1 BORON-EPOXY COMPOSITES
Achieving a high boron concentration in polymeric
materials is not trivial. There are at least two
approaches: chemically bondinq boron-containing moieties
into the polymer chain, and physically mixing a boron
containing compound into a polymeric matrix. The former is
an ideal method to achieve a homogeneous polymer system with
well-defined properties, albeit with stoichiometrically
determined limits on the weiqht percentage of boron. The
second option consists of physically mixing a hiqh-boron
material, either fibers (such as boron or boron nitride) or
powders (such as elemental boron) into a polymer matrix and
crosslinking or otherwise solidifying the mixture. one
question that arises with physically mixing two or more
different materials together is whether or not there is qood
consolidation of material with optimum polymer properties.
This experimental section describes the fabrication and
synthesis of some boron-containing polymeric materials by
each method above. Characterization of each new material by
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
composition, heat resistance, and mechanical properties was
undertaken.
Neutron absorptive properties were investigated in an
effort to determine if the materials could be effective
neutron shields.
nT~I~S
Crystalline 250-micron boron powder, 99.9+%, and two
grades of amorphous submicron boron powder, nominally 90%
and 99% purity, were purchased from Johnson Matthey catalog
co. (Ward Hill, MA). Crystalline, 20-44 micron boron
carbide, 99.999% pure, was purchased from Noah Technologies,
Inc. (San Antonio, TX). All powders were vacuum dried at
ll5°C for at least 24 hours before use. The purity of the
90%-labelled amorphous boron was determined from inductively
coupled plasma (ICP) measurements to be 87.0% boron. The
principal impurity was magnesium. The ICP method could not
be used for the crystalline powders because of their
insolubility in concentrated acid solutions necessary to the
method.
A tetrafunctional epoxy resin, EA 934, was purchased
from ICI/Fiberite, Inc.(Tempe, AZ), stored under cold
conditions, and used as received. The resin contained a
diaminosulfone curing agent in a pre-mixed form, therefore
necessitating storage in a freezer to prolong the shelf life
of the resin. Elemental analysis conducted at NASA-Langley
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Research Center revealed that only carbon, hydrogen,
nitrogen, oxygen, and sulfur were present.
2.1.1 FABRICATION AND PROCESSING
Bot-aelt processinq
Boron-containing epoxy resin materials were processed
in an 8.125-inch square welded steel mold. Part thicknesses
were chosen from the range of 0.375 to 0.700 inches in order
1) to minimize processing time and 2) to maximize the number
of mechanical test specimens that could be cut from each
plaque. Thicker plaques could be made; however, the
processing time required would preclude fabrication of void
free parts, owing to the reaction rate of the epoxy resin at
mixing temperatures. Boron powder was added to measured
amounts of epoxy in nominal percentages from 5 to 20% by
weight. Density calculations were used to determine the
volume of each part, and all parts were between 0.375 and
0.700 inches thick.
A typical procedure for processing a powder-filled
epoxy composite material is as follows. First, the mold was
release-coated with Freekote 44N (The Dexter Corporation,
Seabrook, NH), a hydrocarbon-based release agent spray, and
heat-treated according to the manufacturer's specifications.
The epoxy resin was then placed in the mold and heated to
57°C under vacuum until all entrapped air was eliminated,
typically 0.5-1.0 hours. Boron powders were thoroughly
30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
mixed with the resin in the mold while at 57°C, a
temperature at which the resin is fluid enough that thorough
mixing was possible. The mixture was placed in a vacuum
oven and deaerated at 57° to 62°C under vacuum until there
was no visual evidence of entrapped air. The boron-epoxy
mixture was then placed in an air convection oven under
moderate air flow and the temperature was increased to 121°C
at 2°Cfmin. Following a one-hour hold, the temperature was
increased to 177°C at 2°C/min and held for two hours. This
cured material was cooled overnight to ambient temperature
and removed from the oven. Both the neat epoxy and the
boron-containing epoxy materials appeared to be well
consolidated.
Density measurements were made by weighing machined
specimens of each material in and out of water and carefully
recording the volume changes. Ultrasonic c-scan analyses
were conducted at NASA-LaRC using a computer-controlled
Panametrics, Inc. (Waltham, MA) ultrasonic scanning
instrument with software from Sonix, Inc. (Springfield, VA).
The transducer frequency was 10 MHz and c-scans indicated
that the parts were essentially void-free (See Figure 10).
Actual boron concentrations in the cured materials were
determined by a Thermo Jarrell Ash (Thermo Instruments
systems, Inc., Franklin, MA) Atomscan 25 inductively coupled
plasma (ICP) elemental analyzer. In this spectroscopic
method, acidic solutions are prepared of each sample for
31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
+4.375~--~--~----r---~--~----~--~----r---~---.
+2.625
+9.875
-9.875
-2.625
Figure 10 illtrasonic C-scan of boron-epoxy plaque.
injection into the ICP instrument. Concentrations of
crystalline powders in the epoxy could not be explicitly
determined owing to the insolubility of the crystalline
powder in strong concentrated acid solutions necessary for
analysis. The actual concentrations were estimated to be
close to the theoretical weight percent of crystalline boron
powder combined with the actual weight percent amorphous
boron powder. This approximation produced negligibly small
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
errors in the data.
Powder processinq
934 Epoxy resin was ground to a fine powder using a
mortar and pestle in a walk-in freezer. The freezer
provided a low-humidity, cold (-l8°C) environment in which
to grind and mix the brittle epoxy. The powdered epoxy was
sifted through a 0.0625 inch mesh and weighed. Amorphous,
submicron boron powder was added to powdered epoxy in
nominal weight percentages of 10 and 20%. The boron-epoxy
mixture was ground in the mortar in order to achieve a
smaller average particle size, and transferred to a mold
prepared as above.
The mold was placed in a vacuum oven and deaerated at
approximately -l8°C for twenty hours under vacuum. After
eliminating air from the material, the vacuum oven
containing the loaded mold was removed from the freezer.
The sample was warmed to ambient temperature while still
under vacuum in the oven. The material was heated to 65-
750C and allowed to settle. When bubbling of the material
ceased, the mold was transferred from the vacuum oven to an
air convection oven and heated under atmospheric pressure
according to the cure program above, i.e., 2°C/min. to 12l°C,
hold one hour, heat 2°C/min. to 177°C, hold for two hours,
cool to ambient temperature and remove from the mold. The
neat epoxy and boron-containing epoxy materials appeared to
33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
be well-consolidated and of similar thickness as plaques
described above. Ultrasonic analysis indicated that the
parts were mostly void free, with pin-hole voids only on the
surface of the material.
Solution processing
Five percent solutions of epoxy resin were made by
slowly dissolving the epoxy in a suitable solvent (such as
acetone) in a 1-L volumetric flask under nitrogen. The
yellow-gold solutions were allowed to stand overnight.
Boron powder was weighed into 300 mL beakers in 0.63, 1.25,
and 2.5 gram amounts and 200 mL of epoxyfacetone solution
was added carefully to each beaker. Mixing was accomplished
with a mechanical stirrer equipped with a 4-blade steel
rotor. Most of the solvent was evaporated in air at room
temperature to a minimum volume, then the mixtures with
little residual solvent were transferred to weighed,
release-coated aluminum pans. Following additional solvent
evaporation, the materials were placed in a vacuum oven at
ambient temperature and deaerated under vacuum. The samples
were heated to 65-75°C under vacuum. After deaeration at
65-75°C, the materials were heated according to the cure
program described above and removed from the convection oven
after cooling to ambient temperature.
The solution-processed boron-epoxy materials were
generally poorly consolidated and large voids were observed,
34
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
probably owing to entrapped solvent in the cross-linked
polymeric material. The rate of solvent release from the
material was crucial to achieving a well-consolidated, void
free cured epoxy panel.
Acetone and acetonitrile are solvents for the epoxy.
Acetonitrile has a lower polarizability (6.4 x 10"24 cm3 vs.
4.4 x 10·24 cm3) than acetone1 , and it has less ability to
hydrogen-bond to the epoxy resin during cure, owing to the
presence of a much weaker hydrogen-bonding substituent. In
addition, acetonitrile is less volatile than acetone, as it
evaporates more slowly under vacuum. As a result,
evaporation of acetonitrile from the solvent-laden boron
epoxy mixture is less likely to cause bubbling of the cured
composite part. For these reasons, acetonitrile proved to
be the solvent of choice for the solution processing method.
Small, cured epoxy samples have low void content when
processed by the solution method with acetonitrile as
solvent. However, voids could not be sufficiently minimized
to yield composite materials with excellent consolidation.
Thoroughly drying the powders, degassing the solvent, and
completely deaerating the epoxy resin dramatically improve
the quality of the composite.
2.1.2. THERMAL CHARACTERIZATION
Differential acanninq calorimetry (DSC)
DSC scans were conducted using a Perkin-Elmer DSC 7
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
differential scanning calorimeter in an effort to determine
the glass transition temperature {T1 ) and thermal
performance curves of the cured boron-loaded epoxy
materials. A true glass transition as strictly defined by
Florr may not be observed in crosslinked epoxy-based
polymers. The term "glass transition temperature11 will be
understood in this section to indicate the transition known
as the softening point of the boron-filled epoxy materials.
The samples were heated twice from 50° to 350°C at 20°C/min.
Dynamical mechanical analysis (DMA)
Dynamic mechanical analyses were conducted3 using a
Dupont 982 Dynamical Mechanical Analyzer connected to a
Dupont 1090 Thermal Analyzer system at NASA-Langley Research
center. The DMA specimens were dried in vacuum at 80°C, and
weighed to constant weight. The specimen dimensions were
measured carefully with digital calipers and the initial
transmitted frequency of each mounted specimen in the
analyzer was noted. The specimens were heated in air from
ambient to 320°C at 5°C/min. The oscillation frequency of
the DMA was recorded as 18 Hz, with an oscillation amplitude
of 0.2 mm. The mechanical configuration of the DMA test
module is shown in Fiqure 11.
The glass transition temperatures of the boron-loaded
samples were calculated from the tangent line drawn between
the two slopes of the frequency vs. temperature curve.
36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MECHANICAL CONFIGURATION I DMAl
ELECTROMECHANICAL TRANSDUCER
SAMPLE CLAMPS
FJgUre 11 Mechanical configuration of DMA sample arm.3
Thermogravimetric analysis (TGA)
Thermogravimetric analyses of the boron-loaded epoxy
materials were conducted to study the chemical stability at
high temperatures. The TGA curves were determined by heating
in air from ambient to 650°C at 2.5°C/min using either a
Shimadzu TGA-50 or a Dupont 9900 thermogravimetric analyzer.
Heat resistance of these materials was determined by
measuring the temperature at which weight loss was s, 10,
and sot.
The amount of char yield of each boron sample was
calculated theoretically for complete combustion. The
37
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
amount of char, presumably boron oxide, ~03 , remaining might
yield information about how these boron-epoxy materials
degrade in a harsh thermal environment. As expected, higher
boron loadings produced incrementally larger char yields.
calculations to determine whether the char was pure ~03 were
inconclusive. The amount of sample remaining after TGA
analysis was insufficient for elemental analysis.
Thermoaechanical analysis (TMA)
Thermomechanical analyses were conducted to determine
the T1
and coefficient of thermal expansion (CTE) of each
sample. Samples measuring 36 mm2 in area and 3 mm thick
were tested using a Shimadzu T.MA-80 thermomechanical
analyzer in penetration mode. Three samples of each
material were heated from ambient to 350°C at 5°C/min. in air
and the expansion coefficients were determined in the range
from 8 0°C to 13 0°C.
Dilatoaeter measurements
A high-precision, Fizeau-type laser-interferometric
dilatometer system4 (see Figure 12-A) was used to determine
the CTE of two boron-loaded epoxy materials, 10% and 20%
boron by nominal weight percent. Figure 12-B shows the
sample configuration in the holder used in measuring each
material's coefficient of thermal expansion, with a suitable
reference material present to generate the necessary
38
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LASER BEAM
LAS[R OPTICAL TRAIN
Fagure 12-A. Schematic diagram of the laserinterferometric dilatometer system. 4
interference pattern. The reference materials used for
boron-epoxy CTE determination were quartz rods, with well-
defined thermal behavior. The data were collected and
plotted as thermal strain vs. temperature. From these
plots, the coefficients of thermal expansion were plotted
vs. temperature in °F.
2.1.3. MECHANICAL PROPERTY CHARACTERIZATION
compression tests
Specimens for compression tests were machined from the
panels of cured polymer at NASA-Langley Research Center.
The dimensions of the compression test specimens were 1.0" x
0.5 11 x 0.25 11 • ASTM Method 695 "Compression Testing of
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Plastics" was rigorously followed in all tests. 5 The
IMACINARY TRIANCUI.AR
SUPPORT FOR TO~FlAT SP£Cit.'(N ::..,.. I REF. RODS
', I PARTIAL •
PEDESTAL MIRROR
~
SPECIMEN
DATA ACQUISITION
PEOESTAL
SUPPORT lAS£
FigUre 12-B Schematic diagram of Fizeau type interferometer. 4
specimens were instrumented with biaxial strain gauges
measuring load (or stress) and strain. The instrumented
samples were tested on a Baldwin-Tate-Emery 120-kip (120,000
lb. full scale) universal test plant connected to a SATEC
systems controller equipped with an Analogic ANDS 5400
series data acquisition system to record the data from the
strain gages. Each boron-loaded epoxy specimen and pure
epoxy was tested. Poisson's ratio was calculated for each
specimen as well as ultimate compressive strength and
modulus. In addition to stress and strain, a calibrated
displacement transducer (from either Trans-Tek, Inc.,
40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Ellington, CT or Hewlett-Packard, Rockville, MD) measured
the change in length of sample with respect to time (dcfdt)
to supplement the strain gage data (See Figure 13). The
FJ.gUre 13
1 OICILI.• I ATQR I l ,.__~
I I ........ --~-_ .. ____ _.. __ -.~
Schematic diagram of del dt crosshead displacement instrument. 6
fractured pieces of each specimen were retrieved whenever
possible for fracture analysis.
Tension tests
Tension dogbane specimens were machined from cured
boron-epoxy panels according to ASTM Method 6930 "Tensile
Testing of Plastics."7 Each dogbane was instrumented with
two biaxial strain gauges mounted on each flat side at the
center of the gage length. The gage length of all specimens
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
was no less than four inches. The specimens were tested
using a 10,000 lb. Instron test mechanical test plant
equipped with an Analogic ANDS 5400 series data acquisition
system to record data from the strain gauges. A few tension
dogbanes fractured outside the gage length and were omitted
from the data. The dogbanes were modified with two tabs on
each end to prevent slippage when mounted in the test
fixtures (See Figure 14). The fractured pieces of each
Top view
........... - /
/ / ~- ' I strain gauges
Stde vtew tabs
Figure 14 ASTM tensile test dogbone with composite tabs.
specimen were retrieved for failure analysis.
Flexure tests
Flexural test specimens were machined from boron-epoxy
plaques to 6" x 0.5" x 0.25" according to ASTM Method 0790-
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71 "Flexure Testing of Plastics. "8 The specimens were not
instrumented with strain gages because of the sample
geometry; however, the strain was measured with a dc/dt
probe. An MTS Systems model #661.23A-Ol SO-kip (50,000 lb.
full scale) mechanical test plant was used with a 1000 lb.
load cell mounted in series in order to achieve a 100 lb.
full sqale load at the lowest setting. An MTS systems model
458.20 microconsole controller connected the MTS SO-kip test
plant to an Analogic ANDS 5400 series data acquisition
system. Careful calibration and electronics checks ensured
accurate data. A four-pronged cell mount was utilized to
center the breaking stress of each specimen (See Figure 15).
Load was plotted vs. percent boron loading and normalized
against the neat resin. The fractured pieces of each
specimen were collected and analyzed for possible fracture
mechanism.
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
center of breaking stress
FJgUre 15 Test fixture geometry for flexure samples.
2.2 BORON-CONTAINING POLYAMIDES
MATERIALS
M-carborane (99+% pure) was purchased from DEXSIL,
Corp. (Hamden, CT), and used without further purification.
1,8-Diaminooctane, 1,10-diaminodecane, and 1,12-
diaminododecane were purchased from Aldrich Chemical Co.
(Milwaukee, WI), and vacuum fractionally distilled prior to
use. 4,4 1 -0xydianiline, 4,4'-methylenebis(dimethyl
aniline), 4,4'-methylenebis(diethylaniline), and m- and R-
44
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
phenylenediamine were purchased from Aldrich and
recrystallized from ethanol prior to use. Dimethylacetamide
(DMAc) and tetrahydrofuran (THF) were either HPLC grade or
were distilled from calcium hydride and stored under
nitrogen with molecular sieves prior to use. N-Butyllithium
was purchased in a Sure-SealS bottle from Aldrich as a 2.5 M
solution in hexanes. Pyridine was purchased from Fisher
Scientific, Inc. (Pittsburgh, PA) and used as received.
Phosphorus pentachloride, 95%, and triethylamine, 99+%, were
purchased from Aldrich and used as received.
MONOMER PREPARATION
H-carborane was functionalized to the dicarboxylic acid
dichloride for polymerization with aliphatic and aromatic
diamines. A typical procedure with modifications9•10 is as
follows.
Preparation of m-carborane-1,7-dicarboxylic acid9 (Fig. 16)
10.0 grams (0.0694 mol) of m-carborane dissolved in 100
mL dry ether were charged into a 250-mL 3-neck round bottom
(RB) flask equipped with a magnetic stirrer, nitrogen
inletjoutlet, and an addition funnel capped with a rubber
stopper. A large rubber stopper was inserted into one neck
to facilitate addition of gaseous reagents. The flask was
cooled to -10 to -5°C with an icejacetone bath, and the
solution was stirred for o.s hours. 56 mL (0.1387 mol) of
45
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
l.n-BuLi ether
2. C02 0-10°C
3.~
m-carborane (MC)
(each dot= BH)
PC15
ether reflux
MC-1,7-dicarboxylic acid MC-1,7-diacid chloride
FigUre 16 Functionalization of m-carborane.
n-butyllithium in hexanes was added rapidly (1 mLfsecond)
with stirring. The mixture was stirred for 1 hour, then
gaseous carbon dioxide was bubbled rapidly into the solution
for 0.5 hour. The mixture was acidified with dilute (10%)
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
hydrochloric acid, extracted with ether and separated with a
separatory funnel. The extracts were dried over magnesium
sulfate and the solvent evaporated to yield a yellow liquid,
which was recrystallized from toluene to yield a white
powder, m-carborane-dicarboxylic acid, m.p. 202-204°C.
Yield: 75% of theoretical. Calculated: C 20.69%, H 5.20%,
B 46.55%, o 27.56%, found: c 20.85%, H 4.63%, B 43.47%, o
31.05%. IR: o-H, 3150-2950 cm·1 (b, s), c-H 2900-2850 cm·1
(b,s), B-H 2632 cm·1 (m,s), C=O 1726 cm·1 (b,s), c-o 1418,
1277 cm·1 (m, s) •
Preparation of m-carborane-1,7-diacid chlori4e9 (Fig. 16)
7.5 grams (0.0322 mol) of m-carborane-1,7-dicarboxylic
acid were charged into a 500-mL 3-neck RB flask equipped
with a magnetic stirrer, water condenser, and nitrogen
inlet/outlet. 300 mL of ether were used to aid the
transfer. To the stirring solution were added 14.4 grams
(0.0645 mol) of phosphorus pentachloride (95% pure, 2%
excess). The mixture was refluxed for 5 hours, over which
time most of the solvent evaporated. The black oily residue
was vacuum fractionally distilled at <1.0 Torr (b.p. 103-
1050C) twice to afford a white crystalline solid, mcarborane-1,7-diacid chloride, m.p. 26-28°C. Yield: 8.50
grams, 98% of theoretical.
Purification of the diacid chloride of m-carborane is
difficult and tedious. It involves distilling the crude
47
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
reaction product twice via fractional distillation with a
packed column under <1.0 Torr pressure vacuum. Since the
diacid chloride is extremely reactive, storage under inert
gas does not prevent the compound from reacting with
advantageous water to form the acid. Reverse reaction of
the acid chloride may not occur to both functional groups.
However, if one group is converted to the carboxylic acid,
functionalization of the monomer to the diamine is limited
to low yields.
2.2.1 POLYAMIDE PREPARATION
All polymerizations were carried out under dry
conditions in nitrogen in the presence of a HCl-acceptor
such as triethylamine or pyridine. Use of pyridine as
acceptor provided the simplest isolation of the resulting
polymers, owing to pyridine hydrochloride's high solubility
in water, the precipitating medium. A variety of aliphatic
and aromatic diamines (See Fiqure 17) were purified and
reacted with m-carborane-1,7-diacid chloride. The
polyamides obtained from the polymerizations had a range of
H/B mole ratios from 2.0 to 3.8. The weight percentage of
boron in the polyamides was from 21 to 35%. A typical
polymerization procedure with modifications11 is as follows
(See Figure 18).
48
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
~N-t CH2rNH2 1,8-diaminooctane (1,8 DAO) 8
H2N-f CH2l._NH2 1,12-diaminododecane (1,12-DDA) 712
H2N (_) 0 (-) NH2
4,4' -oxydianiline
(ODA)
CH3CH2 CH2CH3
4,4'-methylenebis(diethylaniline)
(MDEA)
H3C CH3
4,4'-methylenebis(dimethylaniline)
(MDMA)
Fagure 17 Diamines used in polyamide synthesis.
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Polyaerisation of a-carborane-1,7-diacid chloride and 4,4'
ozydianiline11
To 1. 4880 qrams (7. 43 x 10-3 mol) of 4, 4 •-oxydianiline
stirring in 80 mL of THF in a 100-mL 3-neck RB flask were
added 3 mL of pyridine. 2.000 qrams (7.43 x 10-3 mol) of m
carborane-1,7-diacid chloride were added slowly to the
solution. Stirring was increased to compensate for the
increase in viscosity of the mixture. The temperature was
kept at 20-25°C with a water bath. The reaction was allowed
to stir for 8 hours, followed by precipitation into 1500 mL
deionized water. The beige-white solids were filtered by
aspiration and washed with 300 mL deionized water to remove
residual salts. The polymer was collected, dried at 60°C
overnight under vacuum, and weighed. Yield: 2.95 qrams, 85%
of theoretical.
2.2.2 STRUCTURE CONFIRMATION
The polyamides synthesized above were characterized
structurally by solution 1H and 13C NMR with deuterated
dimethyl sulfoxide using a General Electric QE-300 NMR (300
MHZ for 1H), and by FT-IR spectroscopy using a Nicolet DX-20
FT-IR spectrometer (potassium bromide pellet). Excellent
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 0
8 :~c,8 Cl" 'c~ 'CI + diamine
m-carborane-1, 7 -diacid chloride
HCl scavenger
0 0 II II
C :~c"C-N-R-N '~ .
X Carborane-polyamide
FJgUre 18 Synthesis of m-carborane polyamides.
quality KBr pellets were made from a pellet press assembly
(Spectra-Tech, Inc., Stamford, CT) at 15,000 lb. of force in
a hydraulic press. Typical NMR traces are found in Figures
19 and 20. FT-IR spectra for the polyamides are found in
Appendix 1.
51
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.2.3 MOLECULAR WEIGHT CHARACTERIZATION
The weight average molecular weight, M_, was found by
gel permeation chromatography (GPC) using a Viscotek Model
222 HPLC pump connected to a Viscotek Model 200 viscometer
and differential refractometer detector system. Two GPC
columns were connected in series: column A was a styragel
divinylbenzene crosslinked polystyrene stationary phase
6 4 2 0 PPM
FigUre 19 1H-NMR spectrum of MC-MDEA polyamide.
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I 200
I ,.II l ~-~ l -I I
150 I
100
I . , __ .;. . i
1 . ~~ .. '
F1g11re 20 13C-NMR spectrum of MC-MDEA polyamide.
(pore size 500 A, mean particle size 10~m), manufactured by
American Standards (Mentor, OH), designed to separate
molecular weights from 1,000 to ao,ooo gfmole. Column B,
also a Styragel column (linear pore size, mean particle size
10~m), was designed to separate molecular weights from 1,000
53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to 1,000,000 qjmole. The carrier solvent was THF with a flow
rate of 1.0 mL/min. The columns were calibrated with a
series of narrow molecular weiqht distribution (MWD)
polystyrene standards ranqinq from 4,075 to 184,200 qjmol.
The calibration generated a Mark-Houwink plot of molecular
weiqht vs. the constants K and a. A typical GPC trace,
molecular weiqht calculations, and calibration curve are
found in Fiqures 21-23.
Specific viscosity determinations were conducted at
25°C in an Ubbelohde viscometer (Thomas Scientific,
Philadelphia, PA, No.1, #263) with DMAc as the solvent.
DMAc was chosen as the solvent because its flow time of
approx. 115 seconds was larqe enouqh to quantitate errors to
less than 0.5%. Care was taken to measure the viscosities
as quickly after dissolution as possible after temperature
equilibration, owinq to the decomposition of the polymer in
DMAc over time. 12
54
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
i ; N .... 0 ~ (') a 3 ~
I 0 ....., s::
U1 U1
(') I -.. -N tj tj
~
UISCOTEK CORP. FILENAME: pa2
l.I!U!t
... + Sl ... )(
!) .000
:t
..I a: z ~ M
UCAL 4.04 ENDED: 11/28/95 17:52 RUN ID: Carborane-1,12dtaMtne
CONCENTRATION CHROrtATOGRAH
MC-1,12 DDA
""(\ 4,075 MW Std
VI ·--~-:!::.-.._,..__.-J\.., . ..._
~---/""--~-~ ....... :'\ .. , __
- 1 . 00101. e I 121. 0 t I I I I \\ 1'~ 14 • 0 16
•0
I -f----11-t-. 8.\ t, 18 .e 20 .e ~
RET UOL Rescale HP-plot Label Print Ouerla4 Bnoue Erase
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
r ;z ~ 0 ~ 8 if ~ ~ 9.
OQ :r ... U1 ~ 0\
c: -~ g. ::s rn
Ql ""1
~ (') I --N
0 0 ~
IJCAL 4.04 UNICAL SUMMARY REPORT ================================================================================ oa2 ENDED: 11/28/95 17:52: Carborane-1,12diamine =============================================~=========================~========
PARAMETERS CONCENTRATION (mQ/mL) INJECTION VOLUME (mL) DPT SENSITIVITY (mv/Pa) INLET PRESSURE (KPa) FLOW RATE (ml/min) VISCOMETER OFFSET (ml) ACQ. START TIME (min) ACQ. STOP TIME (min) DATA INTERVAL (sec) SIGMA (mL) TAU (~)
TAU (V)
THRESHOLD
METHOD: UCAL-BROAD CAL FILE mike1
MARK-HOUWINK CONSTANTS ALPHA .399 LOG I< -2.265
MOLECULAR WEIGHT VALUES 2.710 Mn ( avQ) = 2.405E 3
.100 Mw (avg) = 9.020E 3 1.225 Mz ( avQ) = 2.272E q
22.03LJ 11p = 1. 456E 4 1.000 Mv (avQ) = 6.021E 3 -.093 Mn(no RI)= 2.715E :~
3.000 30.000 POLYDISPERSITY RATIOS
3.2'10 ~lw/Mn = 3.750 • ~~09 Mz/Mn = 9.444 .231 SKEWNESS OF DISTRIBUTION
,... ... '"Y
• .:.~1.) SKEW(n) = 5.737 .020 SKEW(w) = 1.297
INTEGRATED DETECTOR SIGNALS BASELINE X y
CONC (mv-mL) = 28.29 L. VISC 18~~ --20.89 VISC ( mv-mL) = 31.1~~ R. VISC 286 -20.89
IV (dl/Q) .171 L. CONC 176 -.. 2::;~,. 83 VISCOTEK MODELH 200 F:. CDNC 301 --2~i 5. 83
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
f ;:z N ~
0 "1:1 n c: e. < ~ ~
~ a iiJ l=f', 0 ::s
Ul (') c: ~ ~ !D
UISCOTEJ( CORP. FILENAME: "'ike!
7.88 -r-
UCAL 4.84
~ + S)
6.88 + .... ,
~' 5.88-+ .-1 )(
UNIUERSAL CALIBRATION PLOT
,,~
-.... ~ ..... * ~, ... 4.88-+
~ ~
* :t
Ul c ~
3.88-+
A8: 3.656E1
A1:-3.735E8 2.88 +
A2= 9.942E-2
1. 8812'. 8 131.8
·'it ...... ~ '· ... ~-...... * ........... ____ __
-----............ '*----. --...... ____ ____
14'.0 15',0 16T.8 17'.0
RET.UOL (M\)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.2.4 THERMAL CHARACTERIZATION
All polyamide powders were characterized thermally by
DSC and TGA. To eliminate advantageous solvent, the powders
were annealed at 150°C under vacuum for eight hours prior to
analysis. The DSC scans were run over the temperature range
50-350°C at 10°C/min in a nitrogen atmosphere. Glass
transition temperatures, melting temperatures, and onset of
degradation were recorded. The TGA analyses were conducted
in the temperature range from 100 to 650°C in air. The
temperatures of five, ten, and fifty percent weight loss
were recorded, as well as the temperature at which the
polymer degradation began. Two polymers, MC-ODA-1 and MC
ODA-2, were analyzed a second time in argon because of a
large net weight gain in air.
Characterization of polymer fi1ms
Thin films of each polyamide in THF solution were made
using a doctor blade. Films were doctored onto clean, dust
free glass plates and placed in a dry-box under nitrogen at
a flow rate of 50 mL/min. After drying to a firm
consistency (5 days}, the films were removed from the plates
and evaluated for physical properties such as color,
clarity, brittleness, and creasability.
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.3 BORON-CONTAINING POLYIMIDE FILMS
Kaptons is a hiqh-performance polyimide with excellent
chemical and thermal properties. owinq to its qood
mechanical properties, Kapton• can be fabricated into
sheets, tape, and thin films. Kapton•'s low dielectric
constant allows it to remain the polymeric binder of choice
in the microelectronics industry. This combination of
properties makes Kapton• very desirable for spacecraft
applications. For these reasons, investiqation of the
effects of mixinq boron powder into the polymer is
important. The imidized polymerization product of 4,4'
oxydianiline and 1,2,4,5-benzenetetracarboxylic acid
dianhydride was used as the laboratory counterpart to
commercial Kapton•.
nT~IllS
4,4 1 -oxydianiline was purchased from Aldrich Chemical
co., recrystallized from ethanol, and vacuum dried prior to
use. 1,2,4,5-benzenetetracarboxylic acid dianhydride was
purchased from Aldrich, recrystallized from acetic
anhydride, and vacuum sublimed immediately prior to use.
Dimethylacetamide (DMAc) was purchased from Fisher
Scientific, Inc., stirred with calcium hydride for 24 hours,
and distilled before use.
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.3.1 SYNTHESIS
The polyamic acid precursor of the polyimide was
synthesized from the room-temperature condensation of 4,4'
oxydianiline (ODA) with 1,2,4,5-benzenetetracarboxylic acid
dianhydride or pyromellitic dianhydride (PMDA) •13 The
reaction is summarized in Figure 24. The polymerization was
carried out at ambient temperature with stirring in DMAc in
a nitrogen atmosphere. A stoichiometric quantity of PMDA
was added slowly to a stirring solution of ODA in a 500-mL
round bottom flask cooled with a water bath. After addition
of the monomers, the solution was stirred at RT for two
hours, then stored in a freezer until film casting. The
polyamic acid solution in DMAc was approximately 10 to 15%
solids.
Thin films of polyamic acid in DMAc were spread onto
clean glass plates with a doctor blade set to the desired
thickness. The films were placed in a dry box under
nitrogen (50 mL/min.) to evaporate most of the solvent.
Thermal imidization was accomplished by heating the films to
100°C for 1 hour under forced nitrogen gas, 200°C for 1 hour
in air, and 300°C for 1 hour in air. The polyimide films
were removed from the glass plates under running water,
dried under vacuum, and stored in a desiccator. PMDA-ODA
polyimide films processed by this method were clear, strong,
and creasable, but were of inferior quality compared to
commercially available Kapton® film. synthesized films from
60
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0.002 to 0.006 inches thick had less strength and stiffness
and were generally less homogeneous than commercial 0.003
inch thick film. One possible reason could be that the
PMDA-ODA polyimide cast from polyamic acid had a lower
molecular weight than commercial Kapton• film.
PMDA-ODA films containing amorphous, submicron boron
powder were prepared by mixing boron powder with the
polyamic acid solutions in concentrations ranging from 5 to
15% by weight. First, the boron powder was mixed with
enough DMAc (about 25mL) to produce a slurry. The boron
powder/DMAc slurry was added, with stirring, to known
quantities of polyamic acid solution and agitated. The
mixtures were spread onto clean glass plates as above and
thermally imidized according to the temperature program
described above. The boron powder particles were
homogeneously spread throughout the cured polyimide films.
Interestingly, x-ray photoelectron spectroscopic (XPS)
analysis of the films revealed no boron on either glass or
air surfaces of the film. This characteristic will be
discussed in more detail in the Results section of this
work. Boron-containing PMDA-ODA polyimide films were strong
and creasable, yet were opaque owing to the brownish-black
boron powder. Elemental analyses of 5% and 10% boron-loaded
films are listed in Table 1.
61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 0
ODA
H 0 0
I II II
N2 DMAc
RT
0
PMDA
N-C--'
HO-C
II
C-OH
~-,-o-0 0 0 H
poly(amic) acid
thermal I imidization
0~ ~0
~c~c~ o\~/-\ j 0
~ c~ 0 ~0
PMDA-ODA polyimide
Fagure 24 Synthesis of PMDA-ODA polyimide.
62
0
X
X
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 1. Elemental analyses of boron-loaded PMDA-ODA polypyromellitimide films.
~~ BQBQN lQ~ BQRQN Calc. Found Calc. Found
Carbon 65.8 64.4 62.8 60.1
Hydrogen 2.50 2.62 2.30 2.75
Nitrogen 6.98 7.06 6.66 6.44
Boron 4.50 3.86 9.00 7.56
Oxygen* 19.9 22.1 19.0 23.2
(* = by difference)
2.3.2 FILM CHARACTERIZATION
Fully cured neat PMDA-ODA polyimide films were
characterized by FT-IR spectroscopy using a Nicolet 20DXB
FT-IR spectrometer. Boron-containing polyimide films were
analyzed by attenuated total reflectance (ATR) FT-IR using a
50mm x 10mm x 3mm KRS-5 thallium bromide dense crystal
(International Crystal Labs, Garfield, NJ) as background.
The angle of the crystal with respect to the infrared beam
was 45°. The presence of the imide absorbances at 1786,
17 3 o , 13 8 o , and 7 3 o cm·1 and the absence of the broad amic
acid -OH peak indicated that the polyamic acid was fully
63
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
imidized to the polyimide. (See Figures 25 and 26) The
absence of water in the spectra was further proof that the
polymer had fully reacted.
64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I ~~---
_,1=--~--<i~
< -c
---~
~ \.'--z
? (
"-------, -__...5'
z\
0 0 0 .,....
0 0 Ill .,....
0 o-o ..... N E ~ Ill
.8 E :J
0 c: 0 Q) Ill > N~
0 0 0 (')
0 0 0
0 0 .,....
0 Q)
0 'If
<fl.t-'-IIIC:IDE·---IIIC:UQI J
Figure 25 Ff-IR spectrum of PMDA-ODA polyamic acid.
65
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 CX)
-- I --- -~~-1 ~-- i
----~-~~ i
~=----- -g I ------ .3 ..-
----:-~-, -=-...;-
-= ----~---~~ _)l 0 -----::-~- 0
.c.:~.::--·---- ------ -·I ~ ·---~
·--..,_ ------- l
--- ----- -=·-~ _11 --
.:::::-.... < .--
0 co
--"'? -- --
i I I
0 0 11'1 (')
iO ·--- r~-- ,' g
0 0 0 ~ ~ N
Fagure 26 Fr-IR spectrum of PMDA-ODA polyimide.
66
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.4 CARBORANE-POLYIMIDES
Synthesis of m-carborane-1,7-diamine was attempted via
three pathways: direct "dry" method12 , direct 11wet" method15 ,
and a "protected amine" synthesis (See Fiqures 27, 28). The
direct methods utilized the curtius rearrangement of a
diazide to the diamine. The protected amine synthesis
involved a Friedel-Crafts acylation mechanism followed by
hydrolysis and neutralization to the free diamine. The
following procedures describe the synthetic methods.
MATERIALS
Benzanilide and sodium azide were purchased from
Aldrich Chemical Co. and used without further purification.
Potassium hydroxide and hydrochloric acid were purchased
from Fisher Scientific, Inc. and used as received. Toluene
and acetone were distilled from calcium hydride and stored
with molecular sieves under nitrogen.
Preparation of activated sodium azide16
Activated sodium azide was prepared by moistening 20.0
grams of sodium azide with 1 mL 85% hydrazine hydrate,
mixing thoroughly, and letting stand for 12 hours or
overnight. The solid was dissolved in a minimum amount of
hot deionized water (typically about 40 mL) and precipitated
into 1000 mL of cold acetone. The reconstituted white solids
67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
aq. acetone
<10°C
"Wet" Method
N2
0 0 II II
N,......c ..... c~c .... c, 3 ~ N3
MC-diazide
(isolate)
+ 2 NaN3
sodium azide
"Dry" Method
toluene
80-90°C
2 hours
toluene
80-90°C
O=C=N,:~c..,.....N=C=O c~
MC-diisocyanate
cone. HCl 1 hour
80°C
H2N :~c/NH2 'c~
MC-diamine
Figure 27 Reaction schemes for "dry" and "wet" m-carborane(MC)-diamine synthesis.
were filtered, washed with 10 mL cold acetone and dried in
air for 1 hour. The azide was active for about 24 hours.
68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Preparation of m-carborane-1, 7-diaaine via "dry" metho412
(Fig. 27)
1.50 grams (0.00186 mol) of m-carborane-1,7-diacid
chloride were mixed with 45 mL dry toluene and 8 grams
(0.123 mol) of activated sodium azide and refluxed for 4
hours under nitrogen blanket in a 100-mL 3-neck RB flask.
After cooling to room temperature, the mixture was filtered
and washed with toluene. Evaporation of the solvent fraction
to <5 mL and addition of 20 mL concentrated hydrochloric
acid produced a black, oily residue. The mixture was heated
to 65°C for 2 hours, then poured over ice/deionized water
with stirring. After neutralization with potassium
hydroxide, three ether extractions (500 mL total) were
collected and dried over magnesium sulfate. Evaporation of
solvent left a dark brown residue with light brown solids.
Preparation of m-carborane-1, 7-diaaine via "wet" metho415
(Fig. 27)
To a stirring, nitrogen-filled 250 ml 3-neck RB flask
containing 4.06 grams (0.015 mol) of m-carborane-1,7-diacid
chloride in 75 ml dry acetone were added 8 mL of a 3.78 M.
solution (0.030 mol) of sodium azide in water. After slow
addition of the sodium azide solution, the mixture was kept
below 15°C for 2 hours with vigorous stirring. The mixture
was then poured into 1000 mL of deionized water. The volume
of the aqueous filtrate was reduced to 300 mL with heating,
69
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
cooled in an ice bath, and clear crystals were collected.
Preparation of •-car~rane-1,7-diamine via Frie4el-crafts
aetho4 (Fiq. 28)
To a mixture of 2.516 qrams (0.0189 mol) of aluminum
chloride in 100 mL carbon tetrachloride in a 250-mL 3-neck
RB flask were added 2.52 qrams (0.094 mol) of m-carborane-
1,7-diacid chloride slowly with stirring. 3.685 qrams
(0.0188 mol) of benzanilide were slowly added and
transferred into the flask with an additional 20 mL of
carbon tetrachloride. The mixture was refluxed for 2 hours
at 80°C and cooled. An ice/water solution was added and the
dark brown carbon tetrachloride layer separated. 80 mL of
95% ethanol was added to the flask and the mixture refluxed
for 2 hours with 110 mL of a 4.06 M. potassium hydroxide,
leavinq a brown residue. Solids were collected from three
fractions of the residue by filtration and stored for later
analysis.
70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
+
N2
reflux
1 hour
I II o-HO-o 2 N-C \. J
0 H 0 0 _ H 0-o o- II 1-o II 11-o-l II \. J C-N \ ~-C,C~C \ J N-C \._ J
I WJH:z()
F'"agure 28 Friedel-Crafts m-carborane "protected" amine synthesis.
71
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2. 5 NEUTRON EXPOSURE
Xrradiation of aatarials
Two types of amorphous, submicron boron-containing
materials were evaluated as candidate materials for neutron
absorption: thick epoxy plaques and thin PMDA-ODA polyimide
films.
thick.
The epoxy specimens measured 1" square and 0.125 11
The polyimide films were from 0.001 to 0.003 11 thick
and were cut into 1" squares. The boron weight percent
loading in the epoxy materials was from 0% (neat epoxy) to
17.43%. The polyimide films contained 5, 10, and 15%
nominal weight percentages of boron powder evenly dispersed
in the polymer. All materials were well-consolidated and
had no visible voids or occlusions, owing to careful
processing described above.
Neutron exposure methods were designed and exposures
were conducted at William & Mary. The neutron source for
the exposures was a 5-Curie alpha particle plutonium/
beryllium source contained in a paraffin block inside a five
gallon metal drum (See Figure 29-A). The paraffin block
served two purposes: to fix the source within the container
and, more importantly, to slow down the resultant neutrons
from the beryllium reaction. The nuclear reactions present
in the source are summarized in Figure 30.
A nylon exposure rod was fitted with a bevelled collar
a few inches from one end and a screw-down sample holder at
72
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
hole for sample rod Pu/Be neutron source
I I f-- 0 • ..----/! •:!/ /
!, ;:/ I :
I ' :· 1.~ :I I
: : f I, I
/,, :,
Lf ~~ I ~II/ 1 1!/1/
Figure 29-A Schematic drawing of Pu/Be source container.
(X (X
239Pu --> 235U --> 235U decay series to 207Pb (stable)
n ~e + 4He -> 13C. -> 12C (stable)
Fagure 30 Principal neutron reactions in Pu/Be source.
73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the other (See Figure 29-B). The top end of the rod and the
metal drum were marked so that the sample orientation within
the source did not change between exposures. Minimization
of orientation effects on neutron irradiation was crucial to
accurate data collection. The bevelled collar facilitated
placement of the shielded samples in the source, thereby
forming a tight seal with the drum.
Samples of boron-containing materials were sandwiched
around 0.003" thick, square indium foils. The boron-indium
boron "sandwich" was then placed into the sample holder and
held in place with two nylon screws. The nylon exposure rod
was then placed in the source for twenty-hour exposures.
This time was necessary to completely saturate the indium
foils with slow neutrons, ensuring the maximum exposure and
neutron absorption. The boron-loaded materials shielded the
indium foils from interaction with neutrons to form 116In, a
beta-emitting radioisotope with a 54-minute half-life (See
Figure 31). After irradiation, the rod was removed, quickly
disassembled, and the indium foil placed inside a thick lead
enclosure for counting. A Model 580 Geiger-Muller tube
counter (The Nucleus, Oak Ridge, TN) was used for tabulation
of counts of the radioactive foils. Beginning at
approximately ten minutes after removal from the source, the
number of counts was collected at five minute intervals up
to two half-lives, about 110 minutes. The logarithm of the
number of counts was plotted vs. exposure time and
74
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
indium foil~
:: ~~~ boron-loaded materials~-""'"'
F"JgUre 29-B Neutron exposure rod with sample mount.
75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
~ 115In > + n ---- 116In * ---- > 116Sn (stable)
till= 54 min
F"JgUre 31 Counting reaction of beta-emitting indium isotope.
extrapolated to time zero, removal from the source (See
Appendix 2) •
After counting, the nylon rod and indium foils were
tested with a hand-held Geiger counter to ensure that no
latent radioactivity remained. Five hours after counting
two half-lives of indium, no radiation was detected.
2.6 SURFACE CHARACTERIZATION METHODS
2.6.1 X-RAY PHOTOELECTRON SPECTROSCOPY
x-ray photoelectron spectroscopy (XPS) and static
secondary ion mass spectrometry (SSIMS) were conducted17 on
samples of boron-filled epoxy composites and boron
containing PMDA-ODA polyimide films which were exposed to
atomic oxygen in low Earth orbit (LEO). The samples were
placed in the ram direction of Flight STS-51 of the Space
76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Shuttle for forty hours exposure to the LEO environment,
whose principal component is atomic oxygen. The exposed
samples and the control samples were analyzed at the
Brisbane Surface Analysis Facility (BSAF) at the University
of Queensland in Brisbane, Queensland, Australia. XPS was
performed with a Perkin-Elmer (Eden Prairie, MN) PHI Model
560 ESCA/SAM instrument (See Figure 32) using soft x-rays to
probe the surface composition. Surface penetration of the
Ka x-rays of magnesium (1250 eV) produced emitted electrons
from elements within 100 nm of the surface according to the
photoelectron effect, and the electrons were detected and
counted by a PHI Multiple-technique Analytical Computer
System (MACS). The XPS data were plotted as intensity
(arbitrary units) vs. energy of the recoil electrons.
2.6.2 STATIC SECONDARY ION MASS SPECTROSCOPY (SSIMS)
The same instrument was utilized with an optional SSIMS
module to generate 5.0 nanoampere argon ions for surface
characterization. The SSIMS instrument sputtered argon ions
onto each boron-containing material's surface and evaluated
the secondary molecular ions generated from surface species
77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The PHI Model 560 ESCA/SAM
fltvu• ti"Hll
llltrll• 1rna
till IllS
••a
~UPIIICS IISPUT
r.UYSIIUitl IIIIGICV• CG81101.S
lllf IIIIJtl
IIICI' YlliiCf
liT IIUII car~t•
Fagure 32 Schematic illustration of the PID Model 560 ESCA/SAM.
with a Perkin-Elmer PHI SIMS II detector. carbon, nitrogen,
oxygen, boron, sulfur were detected from the polymeric
materials, and the major contaminant species detected were
silicon and fluorine. SSIMS data were plotted as intensity
(arbitrary units) vs. mass number. The XPS and SSIMS data
are found in Appendix 3.
78
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER THREE: RESULTS AND DISCUSSION
3.1 BORON-LOADED EPOXY COMPOSITES
Materials processing
Amorphous submicron boron, crystalline 250~ boron, and
crystalline 20-44~ boron carbide powders were mixed with an
aerospace-qualified epoxy resin according to the methods
described in the experimental section above. The weight
percentages of amorphous boron in the parts ranged from
3.95% to 17.43% as determined from inductively coupled
plasma (ICP) measurements. Panels containing 10% and 20%
nominal weight percent boron were made by mixing amorphous
and crystalline boron powders (50% amorphous, 50%
crystalline) together into the epoxy resin. Finally, boron
carbide (B4C) was mixed with the epoxy at 20% nominal
loading. All composite parts were essentially void-free and
of excellent quality. Table 2 lists the nominal weight
composition of boron and designation of each boron
containing epoxy material. The ICP method could not be used
to determine the amount of crystalline powders present in
the epoxy owing to their insolubility in concentrated acid
79
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
solutions necessary for sample preparation. The weight
Table 2. Compositions of boron-epoxy formulations.
DESIGNATION MATERIAL COMPOSITION
O%B-Epoxy pure epoxy resin
S%B-Epoxy epoxy + 5% amorphous boron powder
10%B-Epoxy epoxy + 10% amorphous boron powder
15%B-Epoxy epoxy + 15% amorphous boron powder
20%B-Epoxy epoxy + 20% amorphous boron powder
10%B-mix epoxy + 10% boron (amorph. + cryst.)
20%B-mix epoxy + 20% boron (amorph. + cryst.)
2 O%B4C-Epoxy epoxy + 20% crystalline boron carbide
compositions of crystalline powders were assumed to be close
to the nominal percentages.
The reactive epoxy resins were processed by hot-melt,
powder mixing, and solution mixing methods. Of the three
mixing variations, the hot-melt method produced the most
well-consolidated molded plaques with the fewest voids.
Ultrasonic c-scan analysis revealed only a few pinhole
surface voids in the final materials. The voids were not
observable to the eye, and optical micrographs showed no
voids over the bulk of the material (See Figure 33). The
small black pits in the micrographs are due to pull-out
voids created by the specimen polishing method, and are not
80
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Piqure 33 Optical surface photograph of boron-epoxy material. [200X]
evident of entrapped air or volatiles in the material.
Density measurements were in good agreement, in general,
with the calculated densities. The densities of the epoxy
resin and boron powders used in the calculations were: 1.30
gfcm3 (934 Epoxy), 2.35 gfcm3 (amorphous submicron boron),
2.37 gfcm3 (crystalline boron), and 2.52 gfcm3 (boron
carbide). Table 3 lists the calculated densities and
measured specific gravities of all boron-epoxy materials.
81
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 3. Theoretical and actual densities (gfcm3 )
of boron-epoxy materials.
KATEBIAL TBEOBETICAL DENSITY ACTUAL QENSITY
O%B-Epoxy ]..300 ]..30
5%B-Epoxy 1.330 ]..33
10%B-Epoxy 1.361 1.36
15%B-Epoxy 1.393 1.39
20%B-Epoxy 1.427 1.42
10%B-mix 1.361 1.39
20%B-mix 1.428 1.40
20%B4C-Epoxy 1.439 1.42
One difficulty with the hot-melt processing method was
the persistence of boron clusters within the continuous
boron powder-loaded epoxy matrix. Optical micrographs taken
at soox magnification show relatively large clusters of pure
boron powder (See Figure 34) with the rest of the boron
dispersed evenly in the epoxy. Powder processing and
solution processing were attempts to mix the boron powder
more homogeneously into the epoxy. The effects of the
clustering of the boron were reflected in mechanical test
data (discussed in a later section of this document).
The powder processing method seemed to mix the boron
powder more intimately with the crushed epoxy powder. The
small particle size of the amorphous boron (0.6 ~m) aided
82
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fiqure 34 Optical micrograph of 10%B-Epoxy (lOOOX].
the grinding of the epoxy to a fine powder. The cold
processing environment was produced by a cyclical
refrigeration unit, which resulted in periodic levels of
high humidity during mixing. Consolidation during the cure
program was hindered by large amounts of entrapped air
and/or water vapor. The materials processed in this fashion
required much longer deaeration times, and melt flow was
further slowed by the cure reaction of the epoxy. The
83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
composite parts produced were voidy and brittle. Estimated
void content of materials mixed by this method was 2-10% by
weight.
The possible causes of boron clusters in the epoxy were
not explicitly determined in this study. Static electricity
effects, in conjunction with surface tension of the boron
powder were thought to produce the clustering of boron in
the epoxy. One recent study found similar behavior for
carbon black-filled polypyrrole composites.• In an effort
to minimize these effects, the epoxy was dissolved in a
suitable solvent (acetonitrile), mixed with boron powder,
and poured into shallow molds for curing. Most of the
solvent was removed first by air flow, then under vacuum.
Some solvent remained in the material, and this contributed
to the extremely high void content of the finished
composites. The cured composite parts processed by this
method showed somewhat less clustering of the boron.
However, the large voids in the parts prevented the
consideration of solution processing as a way of achieving
finely dispersed boron in the epoxy. Further attempts to
produce void-free cured materials were unsuccessful.
THERMAL PROPERTIES
The thermal history and processing of polymeric
materials have a profound effect on their properties. Panels
84
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of boron-containing epoxy and neat epoxy resin were cured
for two hours, including the time specified by the
manufacturer required for post-cure. Thermal measurement
methods such as DSC, DMA, and TMA are sensitive to the
processing methods of thermosetting polymers such as
epoxies. The glass transition temperature (T1), percent
crystallinity, ultimate strength and modulus are several
parameters that are affected by cure proqram. osc, DMA,
TMA, and mechanical testing were the thermal
characterization methods chosen to investigate these
properties.
Glass transition temperatures of crosslinkable
polymers, including epoxies, are changing constantly,
theoretically owing to the presence of unreacted epoxide
groups in the "cured" resin. (We will again use the term
"glass transition temperature" instead of "second order
transition" to mean the softening point of the crosslinked
polymer system. 2) As epoxy resins are cured, the viscosity
increases dramatically until vitrification. At this point,
the curing agent molecules are not sufficiently mobile to
reach unreacted epoxide groups. A principle in polymer
synthesis is that optimum reaction conditions occur with
equal accessibility of the functional qroups. When reactive
qroups are isolated due to gelling of the matrix, the
molecular weight and thermal properties are adversely
affected. The concentration of unreacted species in epoxies
85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
is dependent on the processing program. The T1 's of epoxies
are also affected profoundly by the cure rate, hold time and
temperature of crosslinking. Some thermosetting polymers
such as epoxies and cyanate esters exhibit much higher T1 's
than their cure temperatures. In general, the T1 is higher
for polymer systems crosslinked at higher temperatures and
for longer hold times at final temperature. Since the epoxy
materials developed in this study were crosslinked by the
same cure program, the effects of cure temperature and hold
time were virtually eliminated.
Differential Scanning Calorimetry
Differential scanning calorimetry (DSC) measures the
heat released or absorbed by the sample relative to an empty
sample pan. Heat is added or removed to maintain the sample
and reference pans at the same temperature.
Crystallization, glass transitions, and melting phenomena
are recorded as exotherms, changes in slopes, and endotherms
in the DSC thermogram. T1 's were measured for boron-epoxy
materials, and some qualifying remarks regarding these
measurements follow.
Analysis of the neat epoxy isotherm at 177°C revealed
that a high degree of crosslinking (or cure) was achieved
after only 25 minutes (see Figure 35). The cure reaction
was almost complete after so minutes, as indicated by the
asymptotic levelling of the heat vs. time curve. Addition
86
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
If) Cl Cl
(T')
r:: c:5 0
0 N
> 0 :z c: Cl
0 E
:E ~
CTlo me ~(T) >. L • X :JIO 0 u 0... c J UJ
u :l.t: U1 tTl~ 0 -- CTl
0 Cl Cl - ..... :a - c: "'0 - (I Ql Ill L > Ql 0... :l L..- E 0 :l - c c ULL.Ul =:I
l'iqure 35 resin.
m m 10
I
10
m 10
N
CTi 10
0
CTi 10
m N 10
10
N 10
'¢ N
N N 10 10
0
f-ci 0
0 ,_IIi
r--
0
-ci If)
0
f-ui N
0
ci 0
N 10
osc cure isotherm at 177°C of 934 epoxy
87
'iii Cl .... :l c
.5 (J E
1-
c i Q
d ~
i;j X ;:
u
0
~ ~
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of boron to the epoxy did not significantly affect the time
required to reach full cure.
Glass transition temperatures were measured for
multiple samples of each specimen of boron-containing epoxy
material in an effort to determine the effects of the
addition of boron powder to the epoxy. DSC curves of the
boron-epoxy materials were determined at 20°C/min. in
nitrogen in the range of 50-350°C. A typical DSC plot is
shown in Figure 36. The relatively fast heating rate was
selected to accent the transitions in the brittle, composite
materials. Some authors~ have noted that heating polymer
samples at a high temperature rate may drive the T1
higher
than the "true" T1 • However, since some of the cured epoxy
samples did not show measurable transitions at slower
heating rates (2 to 5°C/min) , the faster rate was selected
in order to observe the glass transition. Table 4 lists the
glass transition temperatures determined by osc of the
88
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
0)
ID
~r-~--~~~------------------------------------------------~--------.... Curve 1• DSC '9 Ftla inf'o. naatapm<1 Tua Nov 22 18a 57a 18 1994 1:; Sa.pla Wa!ghta .4. 800 •g • 934 Epoxy rae1n w Cft
c (/]
0
rt ::r CD
~ ~ 0 HI
't:J s:: t1 CD
CD 't:J 0
~ .
~ • Q .....
I.&.
~ :!
lilA:
----Xl 161.800 •c X2 224. BOO •c
49.8 -l Peak 194.083 •c Area 32.449 Ill
49.8 -l 4H 7.054 J/g
~·1 Height 0.181 •"
49.2.,
49.0
48.8 !
48.5 J '
48.4 j' 48.2 '
T--·r-,---·r-r 100. 0 150.0
.M:8 ~ TJMEh a..a •In RATIUo aa..a Cl'•&n
I I
I ·---r---r·----.·-··-·r-,---r--_j
200. 0 250. 0 300. 0
T-.paratura c•c>
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4. Glass transition temperatures for boron-epoxy materials.
KATEBIAL ~~
O%B-Epoxy 194°C
5%B-Epoxy 215°C
10%B-Epoxy 238°C
15%B-Epoxy 227°C
20%B-Epoxy 237°C
10%B-mix 243°C
20%B-mix 238°C
20%B4C-Epoxy 244°C
neat resin and various boron-loaded epoxy materials studied.
The T1's increase from 194°C for the neat resin to 244°C of
the 20%B4C-Epoxy. This trend reflects the hindering of
movement of the epoxy polymer network with addition of
filler. Heating the boron-epoxy materials to higher
temperatures is required to observe softening. No
crystalline melt endotherms nor crystallization upon cooling
was observed, as is typical of amorphous crosslinked
networks.
Dynamic Mechanical Analysis
Dynamic mechanical analyses (DMA) were performed on
specimens of boron-epoxy panels containing O% (neat resin),
90
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10% boron, 10% mixed powders, 20% boron, 20% mixed powders,
and 20% boron carbide (B4C). DMA utilizes a pair of
mechanical arms to exert a cyclical load on the sample at a
given frequency. Investigation of the glass transition is a
frequency-dependent measurement and hence does not depend on
heat transfer to the degree that DSC does.
Sample specifications were used in this DMA study for
comparison to NASA High Speed Research (HSR) materials. The
HSR program goal is to develop new composite materials for
high-speed high altitude aircraft. Several aerospace
companies submitted test specifications for the HSR
polymeric materials development, and the DMA specimens used
in this study were machined and tested according to HSR
criteria.
The glass transition temperatures of the materials were
determined by drawing lines tangent to the linear portions
of the frequency vs. temperature curve and extrapolating the
point of their intersection to the temperature axis. The
initial transmitted frequency was recorded for the boron
epoxy specimens in order to compare the data with duplicate
samples exposed to high energy electrons. Table 5 lists
initial frequency and T1
data from the DMA measurements for
the boron-loaded epoxy specimens.
91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5. Initial DMA frequencies and T,s for boron-epoxy materials.
MATEBIAL INITIAL VD' Hz ~~
O%B-Epoxy 16.35 191
10%B-Epoxy 18.59 218
20%B-Epoxy 19.43 228
10%B-mix 18.64 229
20%B-mix 19.87 229
2 O%B4C-Epoxy 21.19 235
The initial frequency (V0 ) increases with increasing boron
concentration in the epoxy. The neat epoxy has a V0 of
16.35 Hz, while the 20% B4C material has a V 0 of 21.19 Hz.
There is a trend of increasing frequency in the boron-epoxy
materials relating to the amount of crystalline powder
present:
20% B4C > (20% B-mix, 20% B) > (10% B-mix, 10% B) > 0% B.
This implies that with increasing crystallinity and boron
loading in the epoxy, the more dimensional stability the
composite material retains. The shape of the DMA curve
changes slightly for the crystalline boron containing
92
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
materials. Yielding before the glass transition occurs in
materials with crystalline boron-containing powders. Figure
37 shows the difference in shapes of DMA curves of 20%B
Epoxy and 20%B-mix materials. The slope of each linear
portion is not as clearly defined as for the amorphous boron
materials. The glass transition temperature varies from
190°C of the neat epoxy to 235°C of the 20% B4C-filled epoxy,
an increase of about 45°C. Addition of boron-containing
powders does not adversely affect the glass transition
temperature of the epoxy resin.
Tensile storage modulus, loss modulus, and tan delta
data for the specimens are found in Table 6. The storage
modulus, e', is calculated from the slope of the stress
strain curve plotted by the DMA instrument. Boron-epoxy
materials show dramatic increases in stiffness (modulus)
relative to the neat epoxy.
Thermoqravimetric Analysis
High-temperature stability of the boron-epoxy
composites was measured with thermogravimetric analysis
(TGA). Weight loss was plotted vs. temperature in the range
of 100-650°C at a heating rate of 2.5°C/min in air. The
temperatures at which weight loss was 5, 10, and 50% of
original weight are found in Table 7. The 5% weight loss
temperatures decrease with increasing boron concentration in
the materials while the 50% weight loss temperatures
93
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
~ 0 X Q. e w
I I
m CD
I ~I ~ 0 0 N
0> 0
c: 10 C')
c: ·-co 0 0 ...... C')
c: 0 en 0 ca 0 I 10
c: •t: N Q)
0 ..... u L.- ca 0 -0 :E oCb
N5 m ~
iii ~
~
Cl)
0 0 OCl.
0 IOE a. ..-CI)
N w 1-
'+-0 0
0
<t: 2 0
0 10
0
0 0 0 «) ex) ci .,....
ZH 'k>uanb<U.:f
Fiqure 37 DMA overlay of 20%B-Epoxy and 20%B-mix materials.
increase. The 10% weight loss numbers are ambiguous,
probably owing to the high weight loss rate
in that portion of the curves. The amount of char remaining
94
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 6. Mechanical properties of boron-epoxy materials from DMA measurements.
MAT~IAI.! ~mi~ILE STQRAGE ~ENSILE l.!OSS TAN MODULUS. GPa MODULUS. GPa DELTA
0%8-Epoxy 3.25 0.03 0.009
10%8-Epoxy 4.35 0.06 0.014
20%8-Epoxy 5.00 0.07 0.013
10%8-mix 4.28 0.10 0.023
20%B-mix 4.84 0.20 0.040
20%B4C-Epoxy 4.40 0.07 0.016
after heating to 650°C (see Figure 38) also increased with
weight percent boron filler, as expected. As the organic
epoxy degrades, the boron powder forms boron oxides, most
probably Bz03 • The amount of ash was not sufficient for
Table 7. Thermal stability of boron-epoxy materials from TGA.
MATERIAL TEMP at 5% Wt. Loss
(oC)
O%B-Epoxy 313
10%B-Epoxy 309
20%B-Epoxy 303
10%B-mix 292
20%8-mix 287
TEMP at 10% Wt. Loss
(oC)
338
345
341
339
341
352
95
TEMP at sot Wt. Loss
(oC)
471
488
498
498
502
499
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 c N 0 ~
0 m ..... c LO G) ~
(.) ~ c G) 0
D.. '-0
• CD UJ 0 +J
> ~c Q)
"'C ~ - Q) G) a. ·-> ~ ca LO
.c 0 ..... 0 .c
0 c. ca 0 LO 0 LO 0 LO 0 ~
('f) N N ~ ~
(!) 0/o 'PI8!A Jelf8
Fiqure 38 Graph of char yield vs. % boron.
elemental analysis. The 50% weight loss temperature
increases to about 500°C at the highest boron loading.
Clearly, addition of boron to the epoxy does not adversely
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
affect the thermal stability of the composites.
Theraomechanical Analysis
In an effort to determine coefficients of thermal
expansion (CTE), thermomechanical analyses (TMA) were
conducted on boron-loaded materials in penetration mode.
Uncertainty in the TMA test arises from surface preparation
of the specimens. The smoother the sample surface, the more
accurate the test because the penetration needle distributes
the load uniformly on the surface. TMA on 5%B-, 10%B-,
15%B-, 20%B-, and neat epoxy were heated from ambient to
250°C at a rate of 5°C/min under constant load. The
expansivity (CTE) of the materials was measured over a large
temperature range to minimize the effects of noise in the
TMA signal. It should be noted that the variability in the
test for three samples of each concentration of boron was
high. Figures 39 and 40 show the variability between
individual runs of 10% and 20% B-epoxy. Table 8 lists the
values of the CTE from TMA for
97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 0 0 0 ,...
~ Cl'.) ~ __... ro j ·~ -----$-( --------Q) _____ ... -----ta 0
I 0 ' 0
~ ' ' '
~ ' 0 ' N
'
~ ', ,.... u
0 0: 0.. e
'. u
~ \ 1-<
I ' co 0 0
~ \ 0
0 .. 0 ,...._.. ~ 0
~ 0
~ ~ 0
~ ~ 0 0 0 ~ 0 0 0 0 N 0 00 10 - - 0 0 M ,... ,... ,...
Piqura 3t TMA graphs of l0%B-Epoxy runs.
98
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 0 0 0 !'1
rn ~
ro ·~ ~ Q)
~ 0 0
~ 0 0 N
>-. ,_. ~ u 0
....... c.
0-c E II
~ !-<
I co 0 0
~ 0 0
0 N ~ \ 0
~ :
0
~ 0 0
~ § 0 0 0 0 0 0 0 0 N 0 00 I() - - 0 0 M M !'1 M
Piqure 40 TMA plot of 20%B-Epoxy runs.
99
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 8. CTE values from TMA measurements.
KAT ERIAL CTE. ppmrc
OtB-Epoxy 70.6 ± 1.5
StB-Epoxy 69.8 ± 2.3
lOtB-Epoxy 65.3 ± 0.4
15tB-Epoxy 66.1 ± 0.3
20tB-Epoxy 65.4 ± 1.1
each material. Addition of boron to the epoxy lowers the
CTE by approx. 5°C. From the rule of mixtures, one would
expect the CTE to decrease, owing to boron's lower CTE
(approx. 6 ppmrc).
Dilatoaeter measurements
A Fizeau-type interferometer-dilatometer was used to
measure coefficients of thermal expansion in order to
supplement the TMA measurements above and to determine
coefficients for the lOtB-mix and 20%B4C-Epoxy materials.
To measure CTE accurately, a wide temperature range is
necessary. In addition, several factors influenced the CTE
measurements from the interferometer: sample size,
reference material, sample surface preparation, and shimming
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
effects. First, sample size differences between the TMA and
the interferometer could have had an effect on the CTE
measured. The sample size for the TMA measurement was 6 mm2
x 3 mm thickness, small compared to the size of the
interferometer specimen (77 mm x 25.4 mm x 6.4 mm
thickness). The larger the sample, the less effect sample
inhomogeneities have on the measurement. Second, the
reference rods play a large role in the interferometer
determination of the CTE. The interference pattern has too
many bands to count when the CTE difference between the
reference material and the sample is large. For
experimental materials such as the boron-filled epoxy, this
factor limits the range over which the CTE may be
determined. Sample surface preparation is important because
of the mechanics of the test. If the ends of the specimens
(Figure 41) are too rough, the accuracy of the test can be
compromised. Finally, shimming a sample in order for its
interference pattern to be counted against the reference may
change the CTE value if the shim's CTE is markedly different
from the sample. The quartz glass reference rods used in
this experiment have a CTE of about 0.1 ppm/°C, or about two
orders of magnitude less than the sample CTE. The
temperature range over which the interference pattern was
able to be counted was from 25°F to 125°F. Outside this
range, the number of"interference lines was too great to be
counted. Care was taken to machine the sample ends to
101
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
smoothness to minimize roughness effects. Owing to
differences in CTE between the reference and the sample, the
20%B4C-Epoxy specimen had to be shimmed with a polymeric
material whose CTE was in the same range as the sample. The
effects of shimming the sample were not corrected for in the
test, and the measurement was not as accurate as the 10%B-
mix epoxy specimen.
Interferometric measurements were attempted for the
full temperature range of the instrument, -150°F to 150°F.
Data were not recorded over the entire temperature range
owing to the fact that a suitable reference material at all
temperatures was not available. Plots of CTE vs.
temperature in degrees~ are found in Figures 42 and 43 for
TOP VIEW
SPECIMEN ONE OF TWO PARALLEl REF. RODS
Piqure 41 Schematic diagram of interferometerdilatometer sample. 5
102
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 0 ......
0 co
Fiqure 42 Graph of CTE data for 10%B-Epoxy from dilatometer measurement.
103
-u. L-
CD .... ::I -ro .... CD a. E ~
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
u.. 0 1.0
"' ..-.9 u.. 0 0 1.0 @)
~ ~ .a d) m E ... ·-d) 0 c.d>
ic% 1- d)
.S :g ltl
CD -0 J: ..
I 0
"'
0
"'
Fiqure 43 Graph of CTE data for 20%B-Epoxy from dilatometer measurement.
104
0 0 ...-
0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the 10tB-mix and 20tB4C-Epoxy materials.
Table 9 lists the expansion coefficients determined
Table 9. CTE values from TMA, dilatometer experiments.
10tB-mix
TMA CTE. ppm/C
65.3 ± 0.4
Dilatometer CTE. ppm/C **
43.2 ± 0.8
65.4 ± 1.1 44.1 ± 0.9
(** maximum probable error)
from the interferometer measurements and from the TMA runs
for comparison. Coefficients of thermal expansion
determined from this method were consistently lower than the
average CTE determined from TMA measurements. However, the
interferometer measurements were probably more accurate,
considering the larger temperature scale and lower
instrument noise of the test.
Addition of boron powder to the epoxy resin lowers the
CTE of the neat epoxy by up to 5 ~m/m/°C. A decrease in the
CTE from the neat resin was expected, although the magnitude
of the change is larger than anticipated, considering the
relatively small weight percent of boron.
The average values of the CTE of these boron-loaded
epoxy materials are useful in determining the applicability
105
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of such materials in structural components and hybrid
structures. In particular, applications in which panels of
boron-filled epoxy resin are bonded to steel, concrete,
wood, or other plastic materials would require precise
knowledge of the expansion coefficients for structure
design.
MECHANICAL PROPERTIES
To evaluate the possible structural or reinforcement
properties of the boron-epoxy materials, three mechanical
tests were performed according to their respective ASTM
Methods:~ tension, compression, and flexure tests.
Structural applications of polymeric materials may
require excellent behavior under both tensile and
compressive loads. However, some commercial composite
materials lack this versatility. The optimum polymeric
composite for a particular application is dependent
ultimately on the requirements for reinforcement. Asphalt,
a composite of tar and stone, can withstand extremely heavy
loads in compression, but its tensile strength is low.
Aromatic polymer fibers have outstanding strength in the
axial direction under tensile loads, but are brittle in
compression. Many composites are made with woven fiber
fabric reinforcement to minimize the anisotropy of
106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
unidirectional tapes. Powder-filled composites typically
have less strength when pulled in tension than when loaded
in compression.
This section describes the mechanical properties of the
boron-filled epoxy composites processed above, and relates
these data to the structural requirements of some neutron
shielding applications.
Tensile Tests
Specimens of boron-filled epoxy were prepared according
to the procedure discussed in Section 2. The boron-epoxy
specimens as machined had a smooth, hard surface. Glass
fiber-epoxy composite tabs (Section 2, Figure 14) were
adhesively bonded to the ends of each dogbone-shaped
specimen to prevent slippage when mounting the specimens in
the tester. All tensile tests involved at least five
specimens of each material and concentration. The materials
tested were: neat epoxy (0% boron), 5%, 10%, 15%, and 20%
amorphous boron-epoxy, 10% boron(mix)-epoxy, and 20% boron
carbide-epoxy. Each specimen failed within the gage length,
which indicates a true tensile test result. Figure 44 lists
the tensile properties determined for each material.
The ultimate tensile strength (UTS) varies from 2.97
Ksi for the 10% B-mix epoxy to 8.36 Ksi for the neat resin.
A plot of UTS vs. percent boron (Figure 46) for the
amorphous boron-filled epoxy materials shows the dramatic
107
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MATER_Il4T. SIBENGTH MODULUS STRAIN POISS. (Ksi) (Msi) (%) RATIO
0%8-Epoxy 8.36 ± 0.63 0.58 ± 0.00 1.67 ± 0.12 0.37 5%8-Epoxy 8.25 ± 0.71 0.64 ± 0.01 1.09 ± 0.12 0.36
10%8-Epoxy 4.35 ± 0.55 0.72 ± 0.01 0.62 ± 0.09 0.36 15%8-Epoxy 4.11 ± 1.36 0.82 ± 0.00 0.51 ± 0.18 0.35 20%8-Epoxy 5.84 ± 0.25 0.94 ± o.oo 0.65 ± 0.02 0.34 10%8-mix 2.97 ± 0.43 0.70 ± 0.01 0.41 ± 0.06 0.37 20%8C-Epoxy 3.43 ± 0.15 0.78 ± 0.01 0.44 ± 0.02 0.36
Piqure 44 Tensile data for boron-epoxy materials.
decrease in strength in the composite materials. The
strength decreases from 8.36 ksi of the neat resin to 4.11
ksi of the 15% 8-epoxy, then rises to 5.84 ksi for the 20%
B-epoxy material. The 10% mixed amorphous/crystalline
boron-filled epoxy and the 20% 84C-epoxy had low tensile
strengths of 2.97 and 3.43 ksi respectively. These results
were somewhat unexpected because the optical micrographs of
these materials indicated less clustering of the amorphous
boron powder and more homogeneous mixing of the crystalline
particles (see Figure 45). A more uniform dispersion of the
boron particles throughout the epoxy matrix should result in
greater strength properties for the composite.
Powder-filled composites typically display low tensile
strength properties owing to the fact that the filler
particles act as stress concentration points in the
108
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
material. Clusters of filler particles generally have zero
strength, so the stress is greater at these non-loadbearing
points. The greater the number of stress points in a
material, the less strength a material has and the shorter
time to failure in tension.
These boron-epoxy materials have clusters of boron
particles embedded in the epoxy matrix, as well as a uniform
dispersion of boron throughout the epoxy. The clusters of
Fiqure 45 Optical micrograph of 20%B4CEpoxy. [200X]
109
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
boron are essentially stress concentration points and serve
to lower the strength of the material in tension. Various
~ . ~
10
~ CP\
~ -~
t/?. 0 :i 0
~ ; z ..
~ - ~ i
I 0 I ! = a
~ ! ~ 1.1 . ln
0 ·1 r · ~
::1 ~
I 0 I
I I t:>
i = ]ln = ....c
1i fo-e; C'l)
Figure 46 Tensile strength of boron-epoxy materials.
110
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
processinq methods (See Section 2) were attempted to
minimize the clusterinq. However, each of these methods,
powder processinq and solution processinq, resulted in a
hiqher void content of the composite material than hot-melt
processinq. The effect of boron clusters on the strenqth of
boron-epoxy materials will be discussed further in the
flexural testinq section.
Addition of fillers, whether in fiber or particulate
form, to a polymeric matrix usually increases the modulus of
the material. The stiffness, or modulus, of the boron
containinq materials increases from 0.58 Msi for the neat
epoxy to 0.94 Msi for the 20% B-epoxy, an increase of 62%.
While not unexpected, the dramatic increase in modulus
indicates that the boron powder-filled epoxy composites are
more riqid than unfilled epoxy. Small (sub-micron) boron
particles may occupy sites in the crosslinked epoxy network
and restrict the polymer chains from translational motion
under stress. The effect of filler content is also evident
in the tensile strain data. The decrease in tensile strain
is shown qraphically in Fiqure 47. The more hiqhly filled
the epoxy material, the less strain to failure under tensile
stress. Poisson's ratio was calculated from the slopes of
the axial and transverse strain curves. Values of Poisson's
ratio (0.34 to 0.37) were typical of polymeric materials.
While the tensile strenqth decreases rapidly for boron
filled epoxy composites, the increases in tensile modulus
111
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
~ -~
~ D
~ JH ~ • :Q
~ I 0 I ~
~ I 0 I s ; .. z -~ f ~ I 0 I
0 J = ~ • ll'l
~ 0
. ~ I .... c N' c
~·
~ ]f
Fiqure 47 Tensile strain for amorphous boron-epoxy materials.
and strain to failure indicate that boron-epoxy materials
112
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
may possibly be useful in structural applications in which
tensile load is present.
Plezure testa
Specimens of boron-epoxy materials were subjected to
flexural stresses in a four-point flexure test fixture. (See
Section 2, Figure 15.) Flexure testing of polymers usually
involves notching the polymer sample and breaking it under
three-point bend stress in order to concentrate the stress
on the center line of the specimen. Because a three-point
test fixture was unavailable for the mechanical test
instrument used in the tests, a four-point fixture was
substituted.
As pictured in Figure 15 of Section 2, the center of
breaking stress in a four-point fixture is focused between
the two inner rollers. However, the specimens subjected to
this test actually broke into three pieces of roughly the
same length. Figure 48 shows the locations of the fracture
sites of the test specimens. ASTM Method 0790-71 supports
the use of a four-point bend test, provided that the radii
of the top rollers are within the range 3.2 mm to 1.5 times
the specimen depth. 6 The four-point flexure tests conducted
as above conformed to the ASTM criteria. The data for the
flexure tests are listed in Table 10. Five specimens of
each of five materials (neat resin, 5% B-epoxy, 10% B-epoxy,
15% B-epoxy, and 20% B-epoxy) were tested. The graph of
113
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Piqure 48 Breaking stress points in flexure specimens.
flexure strength vs. weight percent boron are shown in
Figure 49. The trend that was evident in the tensile data
(see discussion above) is repeated for the flexure data. A
sharp decrease in the flexural strength
occurs between the neat resin and the 10% B-epoxy material.
However, the flexural strength increases for the 15% and 20%
B-epoxy materials. The tensile strength graphs show a
pattern similar to that of the flexural strength graphs.
114
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 10 Flexure data for boron-epoxy materials.
MA~EBIAL STRENGD! (Ksi)
O%B-Epoxy 0.834 ± 0.12
5%B-Epoxy 0.602 ± 0.16
10%B-Epoxy 0.456 ± 0.03
15%B-Epoxy 0.472 ± 0.05
20%B-Epoxy 0.591 ± 0.06
115
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
tn -'~-ca 0·... .cs .....,ca ~:2 ~~ ...... >< cno ~c. ::s'Y >< c Cl) 0 - ... LLo
m
m 0
tO """ <0 l() . . . 0 0 0 0
!S>t '4lDU9JlS X9l.:f
0 N
c 0 '-0 co
0 _. ~c
Q)
~ Q)
a..
l()
0
Fiqure 49 Flexural strength of amorphous boron-epoxy materials.
116
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Here a decrease in flexural strength with increased boron
content is closely related to the likely cause for the sharp
drop in the ultimate tensile strength data. To better
understand this phenomenon, a comparison of the two
mechanical tests needs to be made.
The ASTM standard tensile test described above involves
pulling apart polymeric specimens by generating tensile
stress along the axial direction of the tensile dogbane.
The flexure test also described above can be thought of as a
combination of tensile and compression tests (See Figure
50). Along the underside of the test specimen mounted in
the four-point bend test fixture, the primary force the
FORCE
ll ll [
~ compression ~
J tension .....
Figure 50 Diagram of stresses in 4-point flexure test.
sample experiences is tension. This is created by the
bending of the specimen and a "stretching" of the sample
117
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
plane opposite the load application point. At the same
time, the top side of the specimen experiences compression,
owinq to deformation of the sample from the same downward
force. Thus, the sample is beinq both pulled apart and
pushed toqether, on opposite edqes. It is not surprisinq
that the two mechanical test modes, tensile and flexure,
should yield similar data for the same materials.
It is postulated here that the reason for the similar
behavior in the two is physical. At low boron
concentrations in the epoxy (5-10% by weiqht), larqe
clusters of pure boron powder appear infrequently in optical
microqraphs (See Fiqures 51 a-d). Larqe areas of almost
pure epoxy are visible. Reqardless of the maqnification,
boron clusters measurinq rouqhly SO~m x ao~m are irreqularly
118
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5\B-Epoxy (SOOX] A 10\B-Epoxy [500%] B
15\B-Epoxy [SOOX] c 20\B-Epoxy [SOOX] D
Figure 51 Optical micrographs of boron clusters.
119
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
found over large areas of the microscope specimen. At 15 to
20 boron weight percent, the clusters of boron are still
present (perhaps in slightly greater numbers), but the
clusters are smaller relative to the ones found in 5% and
10% boron-epoxy materials. In addition, a larger fraction
of the boron powder appears to be more well-mixed with the
epoxy than in materials that have a lower boron content.
One possible reason for this interesting phenomenon
could be that at higher boron concentrations, large clusters
of boron come into contact with each other in the mixing
process. The shear forces present in mixing causes clusters
of boron to break apart and disperse into the material.
Some of the clusters essentially annihilate each other, and
the boron powder is more effectively distributed throughout
the epoxy matrix. Not all the clusters are broken apart by
the vigorous, if rudimentary, hand-mixing of the boron into
the epoxy. However, more mixing occurs during the
deaeration stage of processing, further assisting the
breaking up of the clusters, therefore homogenizing the
mixture to a greater extent than at low boron
concentrations. At boron concentrations lower than 10-15%
by weight, the clusters of boron are too far apart to
destructively interfere with each other. Further
experimental support for this theory is contained in the
error bars surrounding the tensile and flexure data. As the
concentration of boron increases in the materials, the test
120
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
data have smaller uncertainties associated with the
measurements. Reproducibility of mechanical test
measurements is an indication of a more uniform composition.
Hence the boron-epoxy materials with higher boron
concentration are more homogenous and better mixed over the
large areas of the mechanical test specimens.
Compression teats
Compressive strength and modulus are key parameters for
structural applications. In addition to bearing the weight
of an external load, a structural member must support its
own weight. Polymeric materials such as boron-filled epoxy
composites are shown in this section to perform very well
under compressive loads.
Specimens of boron-epoxy materials were machined to
ASTM test specifications, instrumented with strain gauges,
and loaded in compression. The fracture mode was recorded,
along with ultimate compressive strength, modulus, and
compressive strain. In addition, Poisson's ratio was
calculated from the data for each specimen. Some of the
specimens fractured axially, while others failed along the
horizontal midline. The test run was stopped immediately
after fracture was detected.
Ultimate compressive strength, modulus and compressive
strain for each boron-epoxy material are listed in Table 11.
121
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 11. Compressive properties of boron-epoxy materials.
MAT~IAL STMHGnf MODULUS POISSON'S (Ksi) (Msi) RATIO
O%B-Epoxy 25.6 ± 0.05 1..57 ± 0.00 0.40 5%B-Epoxy 25.9 ± 0.07 1.73 ± 0.01 0.39
10%B-Epoxy 27.2 ± 0.44 1.94 ± 0.01 0.39 15%B-Epoxy 29.4 ± 1.03 2.28 ± 0.03 0.38 20%B-Epoxy 31.7 ± 0.90 2.63 ± 0.04 0.38 10%B-mix 27.7 ± 0.74 0.80 ± 0.10 0.38 20%B4C-Epoxy 28.4 ± 0.37 0.82 ± 0.19 0.37
Addition of boron powder improved the compressive
properties of the neat resin. The ultimate compressive
strength increased from 25.6 to 31.7 ksi, a 24% chanqe,
for the amorphous boron-loaded materials (see Fiqure 52).
The compressive modulus increased from 1.57 to 2.63 Msi, a
chanqe of almost 68% over the neat resin (see Fiqure 53).
Compressive strains were not meaninqful because the
maqnitude of the strain data exceeded the strain qaqe's
upper limit (3.75% strain). Poisson's ratio was not
siqnificantly altered by the presence of boron powder in the
matrix.
The increases in compressive strenqth and modulus are
typical for powder-filled polymeric composites. Elemental
amorphous and crystalline boron powders and boron carbide
122
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
till ... 1 • "' N
so 1»0 rtS Rl'tS 1111 t-'·RI l»ln •-qn m ..,. . <
Rl
m rt 11 Rl .... ::s
N \Q w rt
::r 0 H)
tr 0 11 0 ::s I
Rl 'tS 0
~
COMPRESSIVE FAILURE STRENGTH OF BORON-LOADED EPOXY
40 ,..
35 1-
Compressive strength, ksi
30 1-
- Maximum
o Average
- Minimum
a
0
-2
0
25 ~------------~~------------~----------------~----------------~ 0 5 10 15 20
Boron loading, weight %
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
S'lll Pl ...
~1 ..... PI COMPRESSIVE MODULUS t-'UI mw .
OF BORON-LOADED EPOXY 0 0 .g ,.., (D m
3 r ~-I I - Maximum < (D
2.J I 0 Average s
0 Minimum I 0
0. -...... ~ N .... ~ r::
m Compressive
0 modulus, Msi 1 Q HI
tr 0
g) 2 I 0 I
(D
'tJI 0 I ;;
X "<
1.5 0 5 10 15 20
Boron loading, weight %
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
are hard, high-melting ceramic materials. Mixing ceramic
powders with polymers improves the compressive properties by
reinforcement of the polymeric matrix. In a mechanical test
or in a load-bearing application, the matrix component of a
composite encounters an outside stress. The force pulls on
the polymer chains, thereby stretching the bonds in the
polymer backbone. Assuming good adhesion to the reinforcing
medium (fibers or particulates), the matrix transfers the
load to the load-bearing reinforcing material. The purpose
of the reinforcing material is to bear the in-plane loads.
With boron-filled epoxy composites, the boron particles are
the load-bearing materials, and the epoxy resin is the load
transfer medium. carbon or boron fibers have a similar
effect on fiber-reinforced composites (FRP); they act as
load-bearing members in the axial direction. FRP's are
typically anisotropic, with much less strength in the
transverse direction. However, the properties of powder
filled composites are isotropic since the powder is randomly
mixed with the polymer. The addition of boron and boron
carbide to epoxy resin in this study reaffirms the utility
of powder-reinforced composite materials.
NEUTRON ABSORPTION MEASUREMENTS
Thin indium foils surrounded by one-inch square
125
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
specimens of boron-filled epoxy resin were exposed to
thermal neutrons in a s-curie plutonium/beryllium (Pu/Be)
source in the Department of Physics at William & Mary. In
addition, indium foils were exposed without any shielding to
establish the maximum number of counts possible from the
neutrons present in the source. Followinq a 20-hour
saturation exposure in the PufBe source, the radioactivity
induced in the indium was counted with a Geiger counter {See
Appendix 2 --Counting Data). The number of counts obtained
by the counter is proportional to the number of
disintegrations of indium-116 atoms decaying to atoms of the
stable tin-116 isotope. Plots of the natural logarithm of
the counts vs. time are linear for radioactive decay.
This experiment is very similar to neutron activation
analysis (NAA), in which samples of known atomic composition
are irradiated with thermal neutrons in a nuclear reactor.
The flux of thermal neutrons in the reactor is typically
between 105 and 1014 neutrons per second. In an NAA
exposure, the flux of neutrons and their energies are very
precisely known in order to calculate recoil products and
energies. However, in our experiments, the neutron flux is
not precisely known. Calculation of neutron attenuation
factors is the principal information gained from a neutron
activation experiment. Because the flux of neutrons was not
precisely known for our PufBe source, neutron attentuation
factors were not expressly calculated from the data.
126
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Graphs of ln(counts) vs. time were plotted for two
half-lives of the radioactive indium foils, in order to
present quantitative comparisons in the data. Figure 54
shows the counting data for unshielded indium foil and
indium foil surrounded by neat epoxy resin. The number of
counts for the indium foil alone (4,134) is less than the
number of counts for indium shielded by neat resin (4,792).
However, the Pu/Be source, like all neutron sources, emits
neutrons in a range of energies, from epithermal to fast
neutrons. The one-eighth inch thick epoxy polymer slabs
slow down a portion of the more energetic neutrons to
thermal enerqies, resulting in a higher thermal neutron flux
exposure of the indium foils than when the foils alone are
exposed. The greater thermal neutron flux results in a
proportionally larger number of counts for the O% shielded
foil vs. the unshielded foil. The data for the foils
surrounded by 5%-20% nominally boron-loaded epoxy are
therefore normalized to the data from the foils surrounded
by neat resin, not to the unshielded indium foils.
A qraph of the initial activity of the shielded foils
is found in Figure 55. The data show increased shielding
imparted by increasinq boron content in epoxy resin for
samples of the same thickness. Similar graphs of PMDA-ODA
polyimide in Section 3.3 confirm that
127
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
~ 0 c. w ~ ::s a. .. E ::s ·-"'C c -.... 0
~ ·-> ·-..... (J
<C -ns ·-..... ·-c -Fiqure 54
~ E w :::J C1)
' ~ ~ '-:J a.
<OVNCOCOC.OV cococo t-t-t-
(SlUnO:J)Ul
N t-t-
0 N "'r"'""
0 0 "'r"'""
0-. coc
E o-<Oaf
E Ol-v
0 N
0
Plot of initial activity of indium foils.
the amount of neutron absorption in boron-containing
shielding materials varies linearly with both boron
concentration and thickness of absorber.
Figure 56 shows a plot of the counting data for the 0%,
128
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 N
tn -·-0 LL z l()
~ -"'C c: CD 0 "'C '-- 0 CD OJ ·- o ..... .c ~ c: en Q)
u '-.,._ Q)
0 a..
~ ·- l()
> ·-~ (J <( -ns ·-~ 0 ·-c 0 L() 0 l() 0 l() 0 L() . - 0') co co "'- "'- co co L()
(SlUn08) Ul
Piqura 55 Plot of initial activity of shielded indium foils.
5%, 10%, 15%, and 20% nominal boron-containing epoxy
shielded foils. As expected, the lines fit to the
129
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
regression equations in Appendix 2 are very nearly parallel.
(Since the decay of indium-116 is the same regardless of the
shielding material surrounding it during exposure.)
~ 0
tn -as c. w ·-...
(I) Q) '-
~ ::J • 0.;
... as ~
~ 0 c. w 'I-0 c: 0 ·-... c. ... 0 tn .c
<C c 0 ... ... 0 ::s m (I)
0
z
c 0 '-0 m ~ . 0 • l{)
c c 0 0 '- '-0 0 co co ~ 0 ~ 0 0
+ lO ~ ~ ~
. i !
' • ilil l; I I! fi / f ! i I !j
1! I !f f f I !. ! f I ; f j f I .
! I ! ! I
i I i I It
/ I/ r ' I ! _,' j .
l tl
/1 I i I !
,' : ~ ! I if I i.: I
J / !! ,' :! I
! i : f i I
! ! : i I I •• •
0 0 0
c 0 '-0 co
:$?. 0 0 C\1
0 C\1 ~
0 0 ~
o_ ex>c
E o-COG.>
E ol-~
0 C\1
0 0 ~
Piqure 56 Plot of ln(counts) vs. percent boron for boron-epoxy shielding materials.
130
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A few observations can be made regarding the plots.
First, the natural logarithm plot shows an order of
maqnitude difference between the shielding of the neat resin
and the shielding of the 17.43% boron-loaded epoxy. This is
a remarkable factor of "neutron attenuation" considering
that it is the result of only 0.125 inch of boron-epoxy
shielding surrounding the indium foil during neutron
exposure. Thicker neutron shielding boron-epoxy materials,
logically, result in greater neutron absorption. Second,
the scatter in the data increases as the number of counts
approaches the background level. In each exposure, the
background was less than 17 counts/min., less than one-sixth
of the number of counts at two half-lives of the 17.43%
shielded foil. Finally, as the weight percentage of boron
in the shielding materials increases, the fraction of
neutrons absorbed increases almost exponentially toward
unity. Figure 57 is a normalized graph of fraction of
neutrons absorbed vs. weight percent boron in the boron
loaded epoxy materials. At 3.95% boron loading, the epoxy
shielding material absorbs 75% of the incident neutrons.
The maximum absorption was for the 17.43% boron-epoxy, which
absorbed over 92% of the incident neutrons. For cost
sensitive applications, the nominal 5%B-Epoxy material may
prove to have the best neutron absorption for the raw
materials cost.
Enriched boron-10 is commercially available at over 95%
131
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
'1-0 c .2 ~ ....,o c. a. ~..w o-c (I)~
.QCU <C~ cs o~..o ....,m ::s CD z
•
•
•
CX)
0
I
\0 . 0
o:t' N . . 0 0
paqJOSqe SUOJlnau JO UO!lOEJ::J
0
0 N
~ c e 0 m -oc
~~
0
CD c. -.c tn
~
Piqure 57 Normalized graph of neutrons absorbed vs. weight percent boron.
~32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
enrichment. Since the elemental boron powder used in these
experiments has the natural abundance of boron-10 (approx.
20%), using enriched boron-10 should greatly increase the
neutron absorption of the boron-epoxy materials.
SURFACE CHARACTERIZATION
Boron-epoxy materials were exposed to the low Earth
orbit (LEO) aboard the Space Shuttle flight STS-51 in 1993.
The chief component of the atmosphere in LEO is atomic
oxygen. The effects of atomic oxygen on polymeric materials
have been reported in the scientific literature,' and
exposure to atomic oxygen is one of the several criteria for
polymeric materials use in space.
Many types of atomic oxygen-resistant materials have
been developed in the past twenty years, including polymers
doped with tin, silicon, and more recently, phosphorus.
This section discusses the results from surface
characterization of boron-loaded epoxy composite materials.
Surface spectroscopy of pure PMDA-ODA polyimide films is
included for completeness.
z-ray Photoelectron Spectroscopy
XPS was performed on two sets of 0.75-inch boron-epoxy
discs, those exposed to LEO and the unexposed control qroup.
Energetic K alpha x-rays of magnesium were used to expel
133
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
inner shell electrons from the atomic species on the surface
of the discs accordinq to the photoelectric effect (See
Fiqure 58). The electrons, having a characteristic energy
signature, were detected, counted, and identified by the
MACS system software. The elements found in all boron-epoxy
samples were carbon, nitroqen, oxygen, sulfur, silicon, and
fluorine. The PMDA-ODA polypyromellitimide samples
contained only carbon, nitrogen and oxygen. Silicon and
fluorine were considered contaminants because elemental
analysis indicated that the 934 Epoxy resin consisted of
only carbon, nitrogen, oxygen and sulfur (hydroqen is not
detected in XPS). Contaminants are a major concern when
evaluating the surface of specimens exposed in the LEO
environment. Several experiments on board the Lonq Duration
Exposure Facility (LDEF) were shown to be contaminated by
silicon from silicone adhesives. 8•9 In spite of the cleaning
procedures employed by persons handling the specimens pre
flight, some contamination consisting of silicon and
fluorine was present in the boron-epoxy samples.
A typical XPS spectrum plots intensity (arbitrary
units) vs. the electron binding energy from highest to
lowest energy. The binding energy's siqn is reversed in
these spectra to show the energies properly (the software
proqram displays the peaks in reverse order of ascending
binding energy).
134
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
photoelectron
photon. hv
atom
Fiqure 58 Photoelectric effect in XPS.
PMDA-ODA polypyromellitimide films
Figure A-1 of Appendix 3 shows the XPS survey spectrum
of the elements on the surface of pure PMDA-ODA polyimide
film. Peaks represented are oxygen, nitrogen, and carbon
from left to right. The control sample is shown in bold
offset. The LEO-exposed plot has some differences from the
control. The carbon 1s peak is smaller in intensity while
the oxygen 1s is larger than the corresponding control
peaks. The reason for this is that as atomic oxygen strikes
the surface of the specimen, it forms a variety of compounds
of carbon and oxygen, including carbonyls, ethers, and
carbonates. In addition, carbon oxides are formed which
135
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
leave the surface. These two factors account for the
difference in the relative intensity of the carbon and
oxygen peaks. As more atomic oxygen reacts to form carbon
oxides, the intensity of the carbon ls peak diminishes while
the oxygen ls peak gains intensity. Atomic oxygen also
forms various oxides with nitrogen on the polymer surface.
As the polymer is eroded by atomic oxygen, more nitrogen is
exposed. This is the reason for the increased intensity of
the nitrogen ls peak in the spectra.
Closer examination of the individual peaks reveals the
chemical changes that occur when atomic oxygen reacts with
carbon and nitrogen present in the polyimide. Figure A-2
shows the carbon ls peaks of the control and LEO-exposed
samples, centered at a binding energy of about 285 ev, in
greater detail. The peak of the exposed sample (bold line)
broadens, indicating that bonding of the carbon becomes more
complex than in the control. The additional peak in the
exposed sample at higher binding energy indicates that the
carbon is bound to more electronegative atoms, such as
oxygen. The more oxygens a carbon atom is bonded to the
more electropositive it becomes. The greater positive
charge on carbon causes it to bind its electrons more
tightly. Thus, the ls electrons that are expelled by the x
rays via the photoelectric effect have more energy. The
intensity of the carbon exposed peak is almost 3.5 times
less than the control, indicating a loss of carbon from the
136
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
surface of the polymer.
The nitrogen 1s peak of the polypyromellitimide (Figure
A-3) centered at a binding enerqy of about 400 eV shows
similar changes to carbon. The peak broadens, indicating a
qreater variety of chemical bonding states, and the peak
shifts slightly toward high binding enerqy, indicative of a
qreater number of bonds to the more electronegative oxygen
atoms. The small intensity of the control and exposed peaks
produces the low siqnal to noise ratio in the qraph.
In contrast, the LEO-exposed oxygen 1s peak (Figure A-
4) centered at a binding enerqy of about 532 ev changes only
slightly from the control. The peak broadens slightly, but
does not shift position relative to the carbon and nitrogen
shifts above. Even though the exposed peak is greater in
intensity than the control, the peaks are centered at a
binding enerqy of about 533 ev.
Boron-Epoxy materials
Comparison of the neat epoxy, 10% B-epoxy, and 20% B
epoxy XPS spectra yields several interesting phenomena. The
carbon, nitrogen, oxygen, sulfur and boron peaks are roughly
centered at 285, 400, 532, 169 and 187 eV respectively.
However, the LEO-exposed materials show differences in peak
width and peak shifts from the control samples that deserve
explanation. This section will describe in detail these
changes of the electron binding enerqy peaks, as well as
137
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
differences in relative intensities of the spectra. The
spectra of the silicon and fluorine contaminants are
included at the end of Appendix 3 (see Figures A-19 to A-
24) •
The carbon 1s peaks of the three materials are very
similar (Figures A-5 to A-7). Upon exposure to atomic
oxygen, all the spectra show a broadening toward higher
binding enerqy which results from the carbon on the surface
reacting with oxygen. The exposed carbon, which is more
electropositive, binds its electrons more tightly which
results in a higher 1s electron binding enerqy. As
discussed above for the PMDA-ODA polyimide films, the
specimens had less carbon on the surface after exposure
owing to loss of carbon oxides in the atomic oxygen flux.
The spectra of the nitrogen 1s peaks (Fiqures A-8 to A-
10) of the boron-epoxy materials indicate similarities also.
There is an average of 1.7 times the amount of nitrogen on
the surface for the exposed samples than for the controls.
The explanation for this is the formation of nitrogen-oxygen
bonds on the material's surface coupled with the loss of
other surface elements such as carbon and hydrogen. As in
the case of the carbon 1s spectra, the nitrogen 1s exposed
peaks are shifted to higher electron binding energy owing to
the nitrogen-oxygen bonds, and nitrogen becomes more
electropositive in the exposed samples.
Analysis of the oxygen 1s spectra of the epoxy
138
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
materials indicates more oxygen on the surface after atomic
oxygen exposure, as expected (Figures A-ll to A-13).
Formation of carbon-oxygen, nitrogen-oxygen, and perhaps
sulfur-oxygen bonds is the primary reason for this
phenomena. However, the peaks are shifted to lower electron
binding energies as the boron content in the epoxy
increases. When oxygen bonds to surface elements, it
becomes less negative, and hence holds its electrons less
tightly, resulting in lower binding energy in XPS spectra.
The sulfur 2p spectra are almost identical for the
three boron-epoxy materials (Figures A-14 to A-16). One
trend evident in the data suggests that as the boron
concentration in the epoxy increases, the amount of sulfur
on the surface after exposure decreases. This is not
unreasonable, given that more boron is present near the
surface at higher loading.
The XPS spectra of the boron ls electrons (Figures A-
17, A-18) is the best evidence that there is very little
boron on the surface of unexposed boron-epoxy materials.
Only after exposure to atomic oxygen does the boron ls peak
show. One possible explanation for this fact could be that
in forming the powder-filled composites, the powder
gravitates to the bottom edge of the mold. However, the
epoxy is probably too viscous to allow this process to
occur. In addition, boron-loaded PMDA-ODA polymellitimides
did not have any boron on either the air or glass sides of
139
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the film. A more probable reason for this interesting
property of these epoxy materials is related to wetting of
the boron powder. At sufficiently high boron
concentrations, some boron powder is not completely wetted
by the epoxy resin. Below this critical "solubility" of the
boron in the epoxy, the resin can fully wet the powder,
thereby coating the powder particles with at least a thin
layer of polymer. This may explain the absence of boron on
the surface of these materials.
Static Secondary Ion Mass Spectrometry
Characterization of the surface of the samples by SSIMS
provides evidence of atomic oxygen's dramatic effects. In
addition, mass spectrometry can determine the elemental
surface composition of polymeric materials. Mass
spectrometry results serve in this study to complement the
results of the x-ray photoelectron spectra discussed above.
SSIMS was conducted on pure PMDA-ODA polyimide films
and boron-loaded epoxy resin samples at the BSAF in
Australia. Some of the boron-epoxy materials were flown on
Flight STS-51 of the Space Shuttle to expose them to atomic
oxygen in the LEO. Bombardment of each specimen with argon
ions produced secondary molecular ions of the surface
elements. The mass spectra were analyzed with a MACS data
collection system, and plotted as counts vs. mass units.
SSIMS spectra are divided into three graphs per specimen:
140
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
low range (0-50 mass units), medium range (51-100 mass
units), and high range molecular weight ions (101-150 mass
units). These plots are found in Appendix 3. LEO-exposed
sample spectra appear in bold type, while control samples
appear in fine type.
PMDA-ODA polypyromellitimide
The SSIMS mass spectrum for the pure PMDA-ODA polymer
is found in Figures A-26 through A-28 of Appendix 3.
Principal peaks are N-H (m=15), sodium (m=23), various
organic groups (m=27,29,39,41,43,55,57,67, and 69), two
large peaks at m=74 and its dimer at m=148. These two peaks
are characteristic of common silicone contamination of the
sample. The mass spectra show substantial changes between
the control and LEO-exposed samples. After exposure to
atomic oxygen, many of the peaks decrease substantially.
Atomic oxygen reacts with elements on the surface to cause a
loss of organic matter and to form oxides with a variety of
elements including carbon, nitrogen, silicon, sulfur, and
phosphorus. The PMDA-ODA polyimide in this study is mostly
carbon and hydrogen, so weight loss by the effects of atomic
oxygen is large. This is reflected in the SSIMS spectra,
which shows a large reduction in almost all of the peaks.
The peak at approximately 130 mass units of the LEO-exposed
sample (heavy line) is probably some form of contamination
which occurred while in flight in the low Earth orbit. It
141
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
is seen in several of the LEO-exposed samples, including the
epoxy materials.
Boron-epoxy materials
SSIMS mass spectra for neat epoxy, 10%B-Epoxy, and
20%B-Epoxy materials are found in Figures A-28 to A-41 of
Appendix J. Several important features of these spectra are
compared and contrasted to the pure PMDA-ODA spectrum.
First, similar hydrocarbon peaks appear at m=27,29,39,41,43,
55,57,67 and 69. Like the pure polyimide, these peaks lose
intensity in the mass spectrum owing to atomic oxygen
effects, however not as much material is lost. The reaction
efficiency for boron-epoxy materials is much less than PMDA
ODA films in space, and the amount of weight loss due to
atomic Olcygen is less. Table 11 shows mass loss and
reaction efficiency for PMDA-ODA and boron-epoxy materials
determined from data obtained from LEO exposure on the Space
Shuttle.
Next, the peaks corresponding to silicone contamination
present in the polyimide films do not show up in the boron
epoxy spectra. Finally, the elemental composition of the
boron-epoxy materials is not surprisingly different from the
polyimide, and this is reflected in the different peak
heights and positions.
To investigate the surface phenomenon of the absence of
boron in the boron-epoxy materials revealed by x-ray
142
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 11. Reaction efficiency of boron-epoxy materials with atomic oxygen in LEO orbit.
MATERIAL
PUre Epoxy
10%B-Epoxy
20%B-Epoxy
~~ (mg)
0.89
0.74
0.53
REACTION EFFICIENCY (cm3/atom x 1024
)
4.40
3.59
2.15
photoelectron spectroscopy, the samples containing 20%
nominal percent boron were scraped with a razor to expose
the surface physically. Spectra of original, unexposed
boron-epoxy and scraped boron-epoxy materials were compared.
Further comparisons are made in this section between
materials with different boron concentrations before and
after exposure to LEO.
In the low mass unit range, the presence of boron in
the 10% and 20% boron-containing epoxy materials is
confirmed by the large peaks at 10 and 11 mass units. In
addition, substantial peaks are recorded at mass 23
indicating the presence of sodium. The boron powder
contains some impurities, including sodium. Therefore, the
height of the sodium mass peak remains constant or increases
due to greater exposure to boron powder in the boron-epoxy
samples. The intensity of the sodium peak (also present as
an impurity) in the pure epoxy spectrum decreases owing to
143
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the effects of atomic oxygen.
The medium and high mass unit ranges (51-150 mass
units) display characteristic mass losses due to atomic
oxygen effects. The one interesting peak is at mass 130,
which is probably present due to contamination of the
samples in space.
3.2 CARBORANE POLYAMIDES
POLYMERS
Six m-carborane-containing polyamides were synthesized
from the condensation polymerization reaction of m
carborane-1,7-diacid chloride with various diamines.
Figures 17 and 18 of Chapter 2 show the repeat units of the
polymers, including the five polyamides not found previously
in the polymer literature: m-carborane-1,7-dicarbonyl-1,8-
diaminooctane [MC-1,8-DDA], m-carborane-1,7-dicarbonyl-1,10-
diaminodecane [MC-1,10-DDA], m-carborane-1,7-dicarbonyl-
1,12-diaminododecane [MC-1,12-DDA], m-carborane-1,7-
dicarbonyl-4,4'-methylenebis-(diethylaniline) [MC-MDEA], and
m-carborane-1,7-dicarbonyl-4,4'-methylenebis(diethylaniline)
[MC-MDMA]. The polyamide synthesized from 4,4 1 -oxydianiline
was reported previously by Korshak and coworkers. 10
The polycondensations were conducted in tetrahydrofuran
(THF) at approximately 5-10°C in a nitrogen atmosphere.
Isolation of the polymers was achieved by dilution of the
144
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
reaction mixture in water, followed by filtration, multiple
washes with deionized water, and vacuum dryinq. Care was
taken to minimize entrapped salts by repeatinq the
dissolution, precipitation, and filtration steps at least
twice. Typical yields were 80-100% of theoretical.
Polymers obtained by this method {Reference 10 with
modifications) were fine powders from dark or liqht brown to
brownish-yellow in color. The aromatic diamines yielded the
more colorful polymer powders owinq to electron transfer
from the more electropositive aromatic diamine to the
electron deficient carborane moiety.
STRUCTURAL CHARACTERIZATION
Nuclear maqnetic resonance and Fourier transform
infrared spectroscopy were used to characterize the
polyamides structurally. Proton and carbon-13 NMR spectra
were obtained to confirm the connectivity of the carborane
polyamides. The FT-IR spectra are found in Appendix 1, and
FT-IR and NMR spectra are discussed in the text of this
section.
Infrared Spectroscopy
FT-IR spectra for each of the six carborane polyamides
confirm the presence of the carborane moiety and the amide
bond. Stronq absorbances present in each of the polymers
145
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
are found at 2600 em-• (B-H stretch) and at 1430 em-• (amide
bond vibrations). The B-H stretch confirms the bonding of
the carborane unit into the polymer, and the amide
absorbances confirm the presence of the amide bond. In
addition, the absence of the broad OH absorbance in the
3300-3000 cm-1 range in the spectra indicates that the
polymers were completely reacted and dry when the spectra
were taken. The background interferences at 22 oo em-• are
characteristic of carbon dioxide. If carbon dioxide and
water vapor are not purged from the IR sample chamber
completely, they can cause distortions in the spectra.
Variability in the age of the background spectrum also
influences the presence of carbon dioxide and water
absorbances in the FT-IR spectra. However, the differences
in background age do not affect the positions of the
absorbances, and since the peaks were not quantitatively
analyzed, no error is introduced into peak identification.
NMR Spectroscopy
Carbon-13 and proton NMR was used to characterize the
polyamides. Peak identification in the aromatic polymers
was complicated by the large number of resonances and lack
of spectra in the literature with which to confirm
absolutely the specific peaks of each substituent.
Identification of the aliphatic polyamide (MC-1,8 DDA, MC-
1,10 DDA, and MC-1,12 DCA) peaks was fairly straightforward,
146
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and the peak listing appears in Table 13. NMR spectra
combined with FT-IR spectra are consistent with the
polyamide connectivity and structure. Boron-11 NMR has
proven to be the tool of choice for verifying the structure
of carborane polyamides. Given the limited utility of
proton and carbon-13 NMR for carborane polymers, future
studies will focus on boron-11 NMR as a diagnostic for
structure elucidation.
Table 13. Peaks identified in NMR spectra of aliphatic polyamides.
fOl!YMEB PEAK CENTER IOENTIFICA~ION
(ppm)
MC-1,8 OOA 7.3 N-H [1H] 3.4 B-H [5H] 1.3 C-H CH2 [10H]
MC-1,10 ODA 7.3 N-H [1H] 3.1 B-H [5H] 1.3 C-H CH2 [10H]
MC-1,12 ODA 7.3 N-H [1H] 2.9 B-H [5H] 1.4 C-H CH2 [10H]
MOLECULAR WEIGHT CHARACTERIZATION
The molecular weight (MW) and molecular weight
distribution (MWD) of polymers influence the thermal,
147
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
mechanical, and physical properties profoundly. With
increasing mol~cular weight, a polymer generally becomes
more thermally stable, forms tough, creasable thin films,
and has better mechanical properties. The utility of a
synthetic polymer depends greatly on its molecular weight
characteristics. For these reasons, higher average
molecular weight is desired in most applications.
Investigations of the molecular weight of polymers make
use of two types of methods: primary and secondary
measurements. Primary measurements of polymer molecular
weight include light scattering, osmometry, and
ultracentrifugation. Secondary methods include gel
permeation chromatoqraphy (GPC) and intrinsic viscosity.
Secondary methods must have an internal standard or include
a calibration curve in order to measure accurately the
molecular weight. In this study, GPC and specific viscosity
measurements were made on carborane polyamides to determine
their molecular weight in order to characterize their
physical properties. A calibration curve of narrow
molecular weight polystyrene standards (Table 14) allowed
calculation of the molecular weight of the polymers
directly.
148
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 14. Narrow molecular weight standards for GPC runs.
NARROW STANDARDS (gfmol)
4,075 12,860 24,150 32,660 45,730 95,800
184,200
The molecular weights of the polyamides from GPC were
lower than expected. All polymers except two, MC-1,10 DDA
and MC-1,12 DCA, had peaks in the GPC that eluted at or
after the lowest polystyrene standard of the calibration
curve (4,075 gfmol). The molecular weights of the two
polymers were MN = 20,500 gfmol for MC-1,10 DDA and ~ =
11,000 gfmol for MC-1,12 DDA. Although these molecular
weights are higher than the other four polyamides, they are
not close to the molecular weights of carborane polyamides
in the literature. 11
The most probably reason for the low MW of the series
of polymers is impurity of m-carborane diacid chloride
monomer. The diacid chloride was meticulously vacuum
fraction distilled at low pressures at least two times,
149
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
however the monomer's reactivity is very high and it reacts
with advantageous water in air at a high rate. Future
preparations could include a direct distillation into the
reaction flask in order to prevent the reverse reaction of
the functional groups to the carboxylic acid from occurring.
In an attempt to achieve higher molecular weight of the
polymers, the reaction solvent was changed to
dimethylacetamide. The polymerization in THF resulted in a
viscosity increase, yet some solids remained during the
reaction. Both monomers are completely soluble in OMAc and
the change of solvent was designed to promote a higher
degree of conversion because of the increased solubility
(and accessibility) of the monomers. Reaction in DMAc was
unsuccessful in producing high molecular weight polymers, as
indicated by the solution viscosity measurements (see Table
15). The failure of the reaction to make high molecular
Table 15. Solution viscosity of carborane polyamides in DMAc.
POLYMER
MC-1,8 DDA MC-1,10 DDA MC-1,12 DDA MC-ODA MC-MDEA MC-MDMA
SPECIFIC VISCOSITY
(dl/g)
0.040 0.047 0.072 0.066 0.035 0.032
150
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
weight polyamides from m-carborane diacid chloride and
various diamines was not remedied by the change in reaction
solvent. Further attempts to synthesize high polymer could
not be conducted owing to a shortage of m-carborane.
THERMAL CHARACTERIZATION
DSC and TGA were used to characterize the thermal
stability of the carborane polyamides. Analysis of the
glass transition temperature and onset of degradation of the
polymers were straightforward.
DSC thermograms for all the polymers are found in
Appendix 4. No glass transition temperatures were observed
for the polymers, in agreement with the literature. 12 The
information from DSC was limited to the onset of
degradation. The polymers were thermally stable up to
250°C. The second heats of the polymers showed no
transitions, either exothermic or endothermic. Above this
temperature, degradation of the polymers was observed as
large exothermic peaks in the thermograms.
The reason for the temperature stability of the
polyamides is the carborane moiety in the polymer backbone.
H-carborane cage units act as energy sinks, absorbing
thermal energy from the environment along the polymer chain.
The degradation of carborane polymers has been found to
occur via formation of dicarbaundecaborate ions [~B~1or
151
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
which catalyze the thermo-oxidative cleavage of the polymer
molecule. 13
Thermogravimetric analyses of the polyamides were
conducted by heating the samples in air up to 650°C at
2.5°C/min. The relatively slow heating rate allowed the
boron in the polyamides to form oxides, resulting in a net
weight gain in some polymers. Figure 59 shows the graph of
the TGA curves of the three linear aliphatic carborane
polyamides, MC-1,8 DCA, MC-1,10 DCA, and MC-1,12 DCA.
Gradual weight gain up to 400°C is followed by a steep
weight loss to about 500°C, after which the curve levels
off. Over 80% of these aliphatic polyamides remained after
heating to 650°C. Generally, aliphatic polymers degrade
quickly in air, yet these carborane-polyamides showed a much
slower rate of weight loss. This stabilizing effect of mcarborane has been noted before by Korshak and coworkers14 ,
and its effects on degradation rate are examined here. The
formation of a coating of boron oxide may insulate the
polymers from further oxidation. Boron oxidation may also
explain the increase in weight of the polymer samples up to
the temperature of degradation.
The stabilizing effect of m-carborane on the aromatic
polyamides is not as pronounced. The polymers MC-MDEA and
MC-MDMA, which have aliphatic side groups on the aromatic
rings, degrade at much lower temperatures than the linear
aliphatic polymers. Figure 60 illustrates the onset of
152
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
.. ... ~ • Ul
--'tS~ OCl 1-'>' '< J»O s~ .... t1 C.< CDCD rnrn •
0 HI
1-' .... ~ ::s U1 CD w J»
t1
J» 1-' .... 'tS :r J» rt .... 0
0 J» t1 t:r 0 t1 J» ::s CD
Weight Loss of Aliphatic Polyamides in Air
TGA
%:~----------------110.00
1oo.ool ::::: :==-- "" --....,
90.00'
80.00
70.00!
200.00 400.00 Temp[C]
600.00
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
degradation of the two polymers at about 225-250°C, followed
by additional degradation above 400°C and linear weight loss
up to 650°C. The stability of MC-MDEA and MC-MDMA is
remarkable, given the fact that the molecular weight of
these two polymers is close to 3,000 gfmol. Oligomeric
species such as these usually do not display such
temperature stability, instead oligomers usually degrade in
the temperature range of degradation of small molecules
(<225°C). The m-carborane moiety preserves the structure of
the polymer until the onset of degradation.
The weight loss curve of MC-ODA shows an interesting
phenomenon (see Fiqure 61). The polymer rapidly gains
weight until it peaks at about 400°C. Subsequent weight
loss is slow initially, then faster as the final temperature
is approached. MC-ODA suffers a weight loss of only about
10% up to 650°C. This curious TGA behavior prompted testing
the polymer again in an argon atmosphere. Figure 62 is the
TGA curve for MC-ODA in argon. The polymer does not lose
any appreciable mass, but gains almost 10% of its original
mass up to 650°C. In the absence of air, the polymer gains
weight. The reason for this weight gain is not known, but
may be due to small amounts of oxygen forming oxides of
boron. Further examination of this behavior is needed.
154
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
'Ill ... 11 Weight Loss of Aromatic Polyamides G\ 0 .
TGA
t-3 %
~ 100.0
0 t ~~ MC-MDEA t:: 11 < (I) (I!
0 I 90.0
t1)
3: .... ?I I \ MC-MDMA U1 U1 :l:
cl ~
80.0
PI ::s g.
3: 7001 \ 0 I
3: \ c \ ~ .... ::s
60.0 PI .... 0.00 200.00 400.00 600.00 11 Temp[C] .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 0 0
$....( 0 10
·~
<C s= ·~
<C 0
Q 0 0
0 0,.... ~u ......
I c:l.
u e '
~
~ ~ ~ 0 0
0
00 0 0
00 M
0 ~
.:d b1) ·~
(1) 0
~ 0 0
<~ 0 00 0 0 0 0
f-. 0 0 0 0 M ... 0 0\ ... ...
Figure 61 TGA curve for MC-ODA in air.
156
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
r-------------------------------------
~ /
/ 0
0 0 0 on \ 0
< I()
~ -~
~ 0
0 0 - 0
0 0 .... vU .....
I 0.
u e u
~ !-<
C+-c 0 0
0
rn 0 rn 0
N
0 ~ ~
( ..c: on ........ <1.) 0
~ 0 0
<~ 0 0 0 0 0 0 q 0 q 0
!-< 0 0 0 0 N .... 0 01 ... ....
Piqure 62 TGA curve of MC-ODA in argon.
157
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FILM PROPERTIES
Thin films were cast of each polymer onto glass plates
with a doctor blade. The thicknesses of the films were in
the range 0.002 to 0.005 inches. All of the polymers except
one, MC-1,12 DDA, formed transparent, colored films. The
MC-1,8 DDA film was colorless. The MC-1,12 DDA film was
transparent until the solvent was removed, then the film
seemed to crystallize on the plate, forming whitened
regions. The crystallization of this polymer from solution
was unexpected since the m-carborane forms a kinked
structure which usually prevents alignment of the polymer
chains.
All of the polymer thin films were brittle and
uncreasable, as expected from the low molecular weights
determined by gel permeation chromatography. The oligomers
do not have the chain length that permits intermingling,
thus the films have almost no crease strength.
158
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.3 PMDA-ODA POLYIMIDE FILMS
Pi1a properties
The polyamic acid from the condensation reaction of
4,4'-oxydianiline (ODA) and pyromellitic dianhydride (PMDA)
was isolated as a 10-15% solid solution in DMAc. Thin films
of the neat polyimide were prepared according to the
procedure found in the Experimental section above. Five and
ten percent boron-containing PMDA-ODA formulations were
produced from a slurry of boron powder in DMAc and polyamic
acid solution. Polyimide films of boron-containing PMDA-ODA
polymer solutions were prepared in a similar procedure as
for the neat polymer. Elemental analysis of the films
(Chapter 2, Table 1) produced by the given methods indicated
close agreement between the calculated and actual
compositions.
Neat and boron-loaded PMDA-ODA films were strong and
creasable. The boron-filled polymer produced opaque films
with little surface roughness. The boron was evenly
dispersed throughout the films. Surface characterization of
the films is discussed in Section 3.1.3. The results of
boron-loaded polyimide films exposed to atomic oxygen are
described elsewhere.n Neutron exposure experiments were
conducted to examine the ability of the boron-loaded films
to absorb neutrons.
159
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Beutron results
Graphs of the loqarithm of the counts measured by a
Geiqer counter vs. time were plotted and least squares
reqression analysis utilized to calculate the initial
activity of the indium foils at time t 0 • The countinq data
and linear reqression results are found in Appendix 2.
Fiqure 63 illustrates the shieldinq effects of various
boron-containinq polyimide films. The amount of neutrons
absorbed is related to the concentration of boron in the
films (by Beer's Law) and film thickness. Calculations of
the relative absorption of neutrons between 5% and 10%
boron-containing materials are found in Appendix 2. Thicker
boron-loaded shieldinq materials such as boron-filled epoxy
resin demonstrate the dramatic neutron shielding ability of
boron incorporated into a continuous matrix material. Films
of boron-loaded PMDA-ODA can be cast up to 0.015 inch
thickness, but in order to achieve maximum neutron shielding
per volume, thicker shielding materials are the best
candidate materials for neutron shielding applications.
Possible uses of boron powder-loaded thin films include
cloth fabrication and radiation dosimetry devices.
Processing, neutron shielding effectiveness, and cost of
boron-containing polymeric materials and their industrial
applications are discussed in Chapter 4.
160
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
"'C CD
"'C ca 0 -I c 0 tn ... 0 E m -·-'1- u.
·e .E ·e .E N 1.0 1.0 T""
1.0 <( ...-0 c c: c: 0 0 e 0 L..
I 0 L.. <( 0 0 CD CD CD 0 ~
~ ~· ~~ 0 ~ 0 • O$
1.0' 0
T"" • 1.0
0 N ~
0 0
o<C cc .2 0 ..... I C.<( oc
~
o-. co c
E o-(0 (])
E tn:E .Ca.
<C oi-'V
0 c N 0 ... ..... 0 ~ 0 CD ~
z co 1.()
t'-LO 0> lO . . (j) co
(SJUn08)Ul
Fiqure 31 Neutron attenuation effects of boron loading on shielded indium foils.
161
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER FOUR: APPLICATIONS AND FUTURE WORK
4.1 INTRODUCTION
Neutron shielding materials have applications in several
industries, including aerospace, nuclear power, and
accelerators. Polymer-based shields having a high boron
content effectively absorb neutrons via thermalization and
neutron capture. Because of the enerqy dissipated by the
neutron capture reaction, high temperature thermal stability
of polymers used as neutron shields is required. The
polymeric materials characterized in this study show promise
as neutron shielding materials for their excellent thermal
properties, ease of processing, good mechanical properties,
and their effective removal of neutrons from a neutron-rich
environment. These attractive material properties combined
with the weight and volume savings of low density matrices,
make these polymeric materials good candidates for neutron
shielding applications in industry.
162
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.2 AEROSPACE
The large weight savings of polymeric composite materials
over conventional metals drives the aerospace industry to
develop materials with lower operational costs. Incorporation
of polymeric materials not only lowers flight costs, but
provides flexibility in aircraft structure design and lower
tooling costs in manufacturing.
The need for neutron shielding materials has been present
for years in spacecraft design. Protection of electronic
circuitry and human occupants of spacecraft is vital to the
success of interplanetary travel. However, the need for
neutron shielding materials in the aircraft industry has been
ignored because of the relatively low risk. The exposure dose
of high-altitude aircraft flights is much lower than space
travel beyond the Van Allen radiation belts which trap most of
the galactic cosmic radiation, protecting Earth from the
harmful effects of GCR. Research and development work in High
Speed Research (HSR) will require some form of neutron
shielding because of the ulta-high design altitudes of HSR
aircraft. Design of polymeric materials that provide neutron
shielding without parasitic weight and volume penalties is
crucial to the HSR effort.
The boron-epoxy materials processed in this study may
fulfill the neutron shielding needs of the aerospace industry
while serving a structural role. The baseline mechanical
163
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
properties of these materials are good, and the thickness of
material needed for complete shielding from neutrons caused by
GCR cascade reactions is probably less than one inch.
Shielding from the relatively low neutron flux created by
ionizing radiation striking the primary structure materials is
easily accomplished. However, further neutron exposures along
with mechanical property testing is needed to determine the
operational life of boron-epoxy structures.
Finally, the cost of polymeric materials is always a
consideration in the design of aircraft. The cost of the
boron powder-filled epoxy composites evaluated in this study
is approximately $250/ft2 for o.s inches of shielding. This
cost can be lowered with the use of boron carbide, which is
considerably cheaper than elemental boron powder. In addition
the properties of boron carbide reinforcement and more
complete mixing are advantages over boron powder
reinforcement.
164
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.3 NUCLEAR POWER
Boron-containing polymeric materials are already used
extensively in the nuclear power industry. Reactor shielding
in nuclear-powered ships including submarines is primarily
borated polyethylene for thermalization and absorption of fast
neutrons. By using the boron-filled epoxy resin discussed in
the above sections of this study, the volume dedicated to
neutron shielding materials can be minimized aboard nuclear
ships. Although the need to minimize the volume of neutron
shielding in nuclear ships is an immediate concern of the u.s.
Navy, budget constraints on future shipbuilding budgets may
result in the examination of more effective cost/volume use of
materials.
165
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.4 ACCELERATOR FACILITIES
Small numbers of neutrons are produced in accelerators
such as the Continuous Electron Beam Accelerator Facility
(CEBAF) located in Newport News, Virginia, from collisions of
electrons with the walls of the beam path. While the flux of
neutrons is extremely small, shielding is needed for worker
safety. The neutron shielding currently used at CEBAF
consists of a water barrier only, with no neutron-capturing
materials.
A one-eighth inch thick slab of 17.43% boron-filled epoxy
resin absorbs over 92% of the incident neutrons from a
plutonium/beryllium source. Although more neutron exposures
are needed to measure neutron attenuation factors, boron-epoxy
materials clearly could be utilized as neutron shielding at
accelerators. The good compressive properties of boron-epoxy
materials should provide enough strength for the design of
wall panels. The panels can be placed at locations where the
neutron flux is greatest for reduction of exposure to
background levels.
166
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 5: CONCLUSIONS
The effects of galactic cosmic radiation on materials in
space orbit and in the Earth • s atmosphere are potentially
harmful to materials and man. Secondary particles produced
from collisions of heavy ions with matter (neutrons, protons,
helium nuclei) may cause single event upsets in electronic
devices, weakening and embrittlement of synthetic materials
and composites, and possibly cancer in human tissue. The most
effective radiation shielding can provide protection from all
types of secondaries formed from GCR interactions, both
charged and uncharged. Few materials known possess such
outstanding qualities. Contemporary shielding materials are
designed to shield against one type of radiation, i.e. gamma
rays, .charged fragments, or neutrons.
Neutrons, having no charge, can travel long distances in
even the most dense materials. Neutrons with a large range of
energies are produced in nuclear reactors, neutron point
sources, and in cascade reactions resulting from the
interaction of GCR with matter. Neutron shielding materials
must accomplish two basic tasks: slow down neutrons to
167
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
thermal energies and capture the neutrons to form stable
isotopes. Shielding materials with high hydrogen content are
the most efficient at reducing fast neutrons to thermal
energies. Polymeric materials are essentially composed of
carbon and hydrogen. Therefore, polymers are good choices for
thermalizing neutrons. Materials that contain large amounts
of boron-10 (which has a natural abundance of about 20% in
elemental boron) are very effective neutron absorbers. The
large thermal neutron capture cross section of 10s (about 4000
barns) is a factor of 106 greater than that of typical
elements. The products of the reaction between 10s and a
neutron (7Li and 4He) are energetic but stable and are easily
stopped within a few angstroms in most materials. In summary,
some of the most effective neutron shielding materials are
polymers that contain a high concentration of boron.
Polymeric materials were synthesized and characterized
for thermal stability, neutron absorption ability, and
possible structural properties. Three different types of
polymers were made: boron powder-filled epoxy resin, boron
containing polyamides, and boron-loaded polyimide films.
Amorphous submicron boron powder and two types of
crystalline boron-containing powders were mixed up to 20%
nominal weight percent with an aerospace qualified epoxy
resin. The epoxy was cured to manufacturer's specifications,
and the dispersion of the boron in the epoxy matrix was
characterized with optical micrographs. The mixing was not
168
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
homogeneous; rather, clusters of boron measuring up to 50
microns by 80 microns were observed. To improve the mixing,
crystalline, 250-micron boron powder was mixed at 50% weight
ratio with submicron boron. Although several batches were
processed, only a small improvement in boron dispersion was
observed. The thermal properties of the epoxy were not
adversely affected by the addition of boron powder. The glass
transition temperature increased by more than 40°C while the
coefficient of thermal expansion decreased by approximately 5
ppm/°C and the thermal degradation temperature rose slightly
to just over 500°C in air. Mechanical test specimens were
prepared for tension, compression, and flexure tests. The
boron-epoxy materials' compressive failure strengths and
compressive moduli increased by 28% and 68% respectively.
However, the materials performed worse in tension. Tensile
and flexural strength decreased with addition of boron up to
about 15% boron by weight. Above 15%, the mechanical
properties increased. one rationale for this behavior is that
as boron content is increased beyond a critical percentage,
clusters of boron are dispersed by mechanical shear forces
during mixing. Boron-epoxy materials have good mechanical
strength under compressive loads, hence they could be used as
construction applications in ceiling and wall panels. Neutron
shielding results of boron-loaded epoxy materials were
excellent. Indium foils shielded by 17.43% boron-epoxy slabs
of 0.125 inch thickness absorbed over 92% of incident neutrons
169
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
from a 5-Curie Pu/Be neutron source. The 3.95% boron-epoxy
material absorbed over 70% of incident neutrons. Thicker
plaques would absorb a larger percentage of neutrons. The
development of neutron shielding structural materials from
boron-epoxy composites would be straightforward. 10% and 20%
boron-epoxy materials were exposed to atomic oxygen in the LEO
aboard the Space Shuttle. Mass loss and reaction efficiency
of the boron-filled epoxy specimens decreased compared to the
neat epoxy owing to formation of boron oxides on the surface.
Six polyamides containing the m-carborane moiety {~B1oH12 )
were synthesized from the polycondensation of m-carborane
diacid chloride with aliphatic and aromatic diamines. The
amorphous polymers had a hydrogen/boron ratio ranging from 2. 0
to 3. 8, contained from 21-35% boron, and were soluble in
organic solvents. Thermal stability measurements by
thermogravimetric analysis indicated that all but two of the
polyamides did not degrade below 400°C in air. Transparent,
colored polymer films from 0.001 to 0.005 inches thick were
cast from solution onto glass plates. However, the molecular
weight of the polymers was lower than expected {<25,000 gfmol)
and the films were brittle. Several attempts to improve the
molecular weight were unsuccessful.
Boron powder in DMAc was mixed in a slurry with the
polyamic acid of PMDA-ODA. Thin films from 0.005 to 0.015
inches thick were cast and thermally imidized up to 300°C
under vacuum. The opaque films were strong and creasable.
170
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The boron was evenly dispersed in the polymer, yet no boron
was observed on the polyme~ surface by XPS and SSIMS. Films
containing up to 10% boron were exposed to neutrons in a s
curie PufBe neutron source. The fraction of neutrons absorbed
increased in a Beer's Law relationship with boron content and
film thickness. Boron-containinq polymer films could be used
for neutron shieldinq of areas with complex surface qeometries
such as electrical wirinq, electronic circuitry, and cloth
shielding. Aerospace applications of the boron-containing
materials produced in this study include spacecraft which
travel beyond the Van Allen radiation belts and hiqh-altitude,
hiqh speed aircraft currently in research under NASA's Hiqh
Speed Research program.
nuclear-powered ships
facilities, and high
Accelerator facilities
Nuclear power applications include
and submarines, nuclear power
level nuclear waste containers.
including hospitals and electron
accelerators such as CEBAF can also benefit from the neutron
shieldinq attributes of these boron-containing materials.
171
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LITERATURE CITED
CHAPTER ONE
1. Wilson, J.W.; Townsend, L.W.; Schimmerling, W.;
Khandelwal, G.S.; Khan, F.; Nealy, J.E.; CUcinotta,
F.A.; Simonsen, L.C.; Shinn, J.L.; and Norbury, J.W.:
Transport Methods and Interactions for Space
Radiations. NASA RP-1257, p. 12, 1991.
2. Ibid, p. 15.
3. Wilson, J.W.; Chun, S.Y.; Badavi, F.F.; Townsend, L.W.;
and Lamkin, S.L.: HZETRN: A Heavy Ion/Nucleon
Transport Code for Space Radiations. NASA TP-3146,
1991.
4. Kim, M.Y.: "Calculations of the Interactions of
Energetic Ions with Materials for Protection of
Computer Memory and Biological Systems," doctoral
thesis, College of William & Mary, 1995.
5. Ibid, p. 68.
172
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6. Friedlander, G.; Kennedy, Joseph W.; Macias, EdwardS.;
and Miller, Julian Malcolm: Nuclear and Radiochemistry,
3rd ed., p. 623, John Wiley & Sons, New York, 1981.
7. Jaeger, R.G.(ed): Engineering Compendium on Radiation
Shielding, vol. I, Springer-Verlag, New York, 1968.
8. Friedlander, p. 610.
9. Soloway, A.H.: Progress in Boron Chemistry, vol. 1, p.
203, Pergamon Press, New York, 1970.
10. Ibid, p. 204.
11. Green, J.; Mayes, N.; and Cohen, M.S.: Journal of
Polymer Science, Part A, vol. 2, p. 3113, (1964).
12. Green, J.; Mayes, N.; Kotolby, A.; and Cohen, M.S.:
Journal of Polymer Science, Part A, vol. 2, p. 3135
(1964).
13. Green, J.; Mayes, N.; Kotolby, A.; Fein, M.M.; O'Brien,
E.L.; and Cohen, M.S.: Journal of Polymer Science, Part
B, vol. 2, p. 109 (1964).
14. Haworth, D.L.: Endeavor, vol. 31, p. 17 (1972).
173
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15. Grimes, R.N.: Carboranes, pp. 181-192, Academic Press,
New York, 1970.
16. Papetti, s.; Obenland, c.; and Heying, T.L.: Ind. Eng.
Chem., Prod. Res. Develop., vol. 8, 334 (1966).
17. Green, J.; Mayes, N.; Kotolby, A.; Fein, M.M.; O'Brien,
E.L.; and Cohen, M.S.: Journal of Polymer Science, Part
B, vol. 2, p. 111 (1964).
18. Heying, T.L.: Progress in Boron Chemistry, vol. 2, p.
119, Pergamon Press, New York, 1970.
19. Williams, R.E.: Pure and Appl. Chem., vol. 29(4), p.
569 (1972).
20. Grimes, p. 182.
21. Schroeder, H.A.; Reiner, J.R.; and Knowles, T.A.:
Inorganic Chemistry, vol. 2, p. 393 (1963).
22. Korshak, V.V.; Sobolevskii, M.V.; Zhigach, A.F.;
Sarishvili, I.G.; Frolova, Z.M.; Gol'din, G.S.; and
Baturina, L.S.: Vysokomol. Soedin. Seriya 2, vol. 10,
p. 584 (1968) •
174
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23. Bekasova, N.I.: Russian Chemical Reviews, vol. 53(1),
p • 69 1 ( 1984) •
24. Greqor, V.; Plesek, J.; and Hermanek, s.: IUPAC
International Symposium on Macromolecular Chemistry,
1965.
25. Heyinq, T.L.; Reid, J.A.; and Trotz, S.I.: u.s. Patent
#3,311,593.
26. Kricheldorf, H.R.; Bruhn, c.; Russanov, A.; and
Komarova, L.G.: Journal of Polymer Science, Part A,
vol. 31, p. 279, (1993).
27. Bekasova, N.I.; Komarova, N.G.; Komarova, L.G.; and
Serqeyev, V.A.: Vysokomol. Soedin. Seriya A, vol.
33(10), p. 2028, (1991).
28. Papetti, s.; Schaeffer, B.B.; Troscianiec, H.J.; and
Heyinq, T.L.: Journal of Polymer Science, Part A, vol.
4, p. 1623, (1964).
29. Peters, E.N.: Ind. Enq. Chem. Prod. Res. Develop., vol.
23, p. 28, (1984).
175
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30. Korshak, v.v.; Prigozhina, M.P.; and Bekasova, N.I.:
Vysokomol. Soedin. Seriya A. vol. 18{4), pp. 827-830,
{1976).
31. Korshak, v.v.; Bekasova, N.I.; and Komarova, L.G.:
Vysokomol. Soedin. Seriya A, vol. 12{8), p. 1866,
{1970).
32. Korshak, V.V.; Vorob'ev, P.K.; Bekasova, N.I.;
Chizhova, Ye. A.; and Komarova, L.G.: Vysokomol.
Soedin. Seriya A, vol. 18{3), p. 632, (1976).
33. Ibid, p. 634.
34. Korshak, v.v.; Bekasova, N.I.; and Komarova, L.G.:
Journal of Polymer Science, Part A, vol. a, p. 2351,
(1970).
35. Korshak, v.v.; Bekasova, N.I.; and Komarova, L.G.:
Vysokomol. Soedin. Seriya A, vol. 24{11), p. 2424,
(1982).
36. Korshak, v.v.: J. Macromol. Sci. Rev. Macromol. Chem.,
vol. C11{1), pp. 45-142, (1974).
176
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37. Savla, M.; and Skeist, I.: "Epoxy Resins," Chapter 16
in "Polymerization Processes," C.E. Schildknecht, (ed),
with I. Skeist, Wiley-Interscience, New York, 1977.
38. Bauer, R.S. (ed): "Epoxy Resin Chemistry," Amer. Chem.
Soc. Symp. Series, vol. 114, (1979).
39. Brydson, J.A.: Plastics Materials, 2nd. ed., Chapters
19-20, Van Nostrand Reinhold, New York, 1970.
40. Chemical & Engineering News, March 25, 1996.
41. Anderson, B. Jeffrey (ed); and Smith, Robert E.:
Natural Orbital Environment Guidelines for Use in
Aerospace Vehicle Development, NASA TM-4527, 1994.
42. Maiden, Donald L.: Radiation Exposure of Air Carrier
Crewmembers, FAA Advisory Circular No. 120-52, March 5,
1990.
177
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER TWO
1. Handbook of Chemistry and Physics, 69th ed., CRC Press,
Boca Raton, Florida, p. E-75, {1988).
2. Flory, P.J.: "Melting and Glass Transition in
Polymers," Proceedings of the Simposio Internazionale
di Chimica Macromoleculare, suppl. to La Ricerca
Scientifica, 3, vol. 6.9, p. 2023, (1955).
3. DMA analysis conducted by Mr. Ian I. Nieves; DuPont
Model 682 Thermal Analyzer technical manual, p. 41.
4. Tompkins, s.s.; Bowles, D.A.; and Kennedy, W.R.:
Experimental Mechanics, vol. 26{1), pp. 1-6, {1986).
5. American Standard Test Method {ASTM 0695).
6. Hewlett-Packard dc/dt technical information, p.2,
Hewlett-Packard Corporation, Rockville, MD.
7. American Standard Test Method (ASTM 0693).
8. American Standard Test Method (ASTM 0790).
178
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9. Korshak, v.v.; Bekasova, N.I.; and Komarova, L.G.:
Vysokomol. soedin. Seriya A, vol. a, p. 2351 (1970).
10. Rabilloud, G; and Sillion, B.: Journal of
Orqanometallic Chemistry, vol. 182, p. 283, (1979).
11. Korshak, v.v., p. 2360.
12. Korshak, v.v.; Bekasova, N.I.; and Komarova, L.G.:
Vysokomol. Soedin. Seriya A, vol. 24(11), p. 2426,
(1982).
13. Srooq, C.E.; Endrey, A.L.; Abramo, S.V.; Berr, C.E.;
Edwards, W.M.; and Olivier, K.L.: Journal of Polymer
Science, Part A, vol. 3, p. 1373, (1965).
14. Glasqow, M.B.; Thibeault, S.A.; Long, E.R.; Orwoll,
R.A.; and Kiefer, R.L.: Polymer Preprints, vol. 35(2),
p. 954, (1994).
15. Organic Reactions, vol. 3, p. 339, John H. Wiley &
Sons, New York, 1946.
16. Ibid, p. 382.
17. XPS and SSIMS conducted by R.L. Kiefer and B. Wood.
179
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER THREE
1. Wampler, Wesley A.; Rajeshwar, Krishnan; Pethe, R.G.;
Hyer, R.C.; and Sharma, s.c: Journal of Materials
Research, vol. 10(7), p. 1811, (1995).
2. Flory, P.J.: "Melting and Glass Transition in
Polymers," Proceedings of the Simposio Internazionale
di Chimica Macromoleculare, suppl. to La Ricerca
Scientifica, 3, vol. 6.9, p. 2023, (1955).
3. Chaplin, A.; and Hamerton, I.: "BismaleimidefCyanate
Ester Blends," TTCP Workshop on High Temperature Resins
and Composites, October, 1995.
4. T.L. st. Clair, personal communication.
5. Tompkins, s.s.; Bowles, D.A.; and Kennedy, W.R.:
Experimental Mechanics, vol. 26(1), pp. 1-6, (1986).
6. American Standard Test Method {ASTM 0790).
7. Forsythe, JohnS.; Graeme, A.G.; Hill, D.J.T.;
O'Donnell, James H.; Pomery, Peter J.; and Rasoul,
Firas A.: LDEF Third Post-Retrieval Symposium, NASA CP-
3275 Part 2, p. 645, 1993.
180
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8. Peters, Palmer N.; Zwiener, James M.; Gregory, John C.;
Raikar, Ganesh N.; Christl, Ligia c.; and Wilkes,
Donald R.: LDEF Third Post-Retrieval Symposium, NASA
CP-3275 Part 2, p. 704, 1993.
9. Linton, Roger c.; Finckenor, MiriaM.; and Kamenetzky,
Rachel R.: LDEF Third Post-Retrieval Symposium, NASA
CP-3275 Part 2, p. 727, 1993.
10. Korshak, v.v.; Bekasova, N.I.; and Komarova, L.G.:
Journal of Polymer Science, Part A, vol. 8, p. 2352,
(1970).
11. Ibid, p. 2355.
12. Korshak, V.V.; Pavlova, S-S. A.; Gribkova, P.N.;
Balykova, T.N.; Bekasova, N.I.; Prigozhina, M.P.; and
Komarova, L.G.: Vysokomol. Soedin. Seriya A, vol.
18(9), p. 2042, (1976).
13. Bekasova, N.I.; Komarova, L.G.; Kats, Georgii A.; and
Korshak, V.V.: Makromol. Chem., vol. 185, p. 2313,
(1984).
181
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14. Korshak, V.V.; Pavlova, S-S.A.; Gribkova, P.N.;
Balykova, T.N.; Polina, T.V.; Zakharkin, L.I.; Kalinin,
V.N.; Bekasova, N.I., and Komarova, L.G.: Vysokomol.
Soedin. Seriya A, vel. 26(1), p. 119, (1984).
15. Kraus, W.B.; Glasgow, M.B.; Kim, M.Y.; Olmeijer, D.L.;
Kiefer, R.L.; Orwoll, R.A.; and Thibeault, S.A.:
Polymer Preprints, vel. 33(1), p. 1152, (1992).
182
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX 1.
Infrared Spectra of Carborane-Polyamides
183
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
.... CXI ~
Figure I. FT -IR Spectrum of poly(m-carborane-1, 7 -dicarbonyl-1 ,8-diaminooctane) [MC-1 ,8 DAO]
o/o T r 60 a n 50 s m 40
I ! 30 t a 20 n c 10
I
I I
e I I I - r- l - - ,_. -.--
4000 3500 3000 2500 2000 W~v~n~mb~r~ (~'!l~ 1J
,. -- 1 -~,- ,. 1500 1000 500
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
.... (X)
Ul
Figure 2. FT -IR Spectrum of poly(m-carborane-1, 7-dicarbonyl-1, I 0-diaminodecane) [MC-1,10 DDA]
~------------------------~--- ·---- ----------- 1
o/o 45
T r 40 a n 35 s m
I! 30
25 t a n c e 15
--, -- ·-------. ---- ... - -.-- r--~--T---·~---y---r--· ,. . -· . r··- -r·----r--·.---·· -...- ~- --,-- r------~----, ·r ----...---- .. , ---..,----· ··--r--.- --. -- ' . -r ..
4000 3500 3000 2500 2000 1500 1000 500 ____ Wavenumbers ( c!'"-1 ~ . _ ...
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
Figure 3. FT -IR Spectrum of poly(m-carborane-1, 7 -dicarbonyl-1,12-diaminododecane) [MC-1,12 DDA]
------ - ---------------
% 65 T r a 60 n s m ... ,. OJ . I
0'1 t 55
t a n c e 45
····--- ------.
---,---, - -,.-- ~~-T--.,.- -~-~-T---,----.,.-- -,- --,--,.- - r·--r-r----~-~--r~--.L.._,.-,----.. ---.-~---,---.---r---,-- --,
4000 3500 3000 2500 2000 1500 1000 500
. ~~~~~-U_f!l_b!:~~ _(~~~ 1)
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
.... 00 -..J
Figure 4. FT-IR Spectrum ofpoly(m-carborane-1,7-dicarbonyl-4.4'-oxydianiline) [MC-ODA]
------------------- --------------------~---------------- -------------,
% 60-T r 50 a n s m i
I: a 20 n c 10-e
-~-~-~-------r- --..-- - ..... -~--.- r·--r-----.,.----......--- r-~--..- ..... ---..---r--r·-----y--~---.----r----,--- .. ----.---.,.--r---.. -- -.. -- --. " - t-
4000 3500 3000 2500 2000 1500 1000 500 _. _ ~~ven~~_ers ~m~l_ _________________ . ___ . __ ·----. __
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
.... 0) 0)
Figure 5. FT -IR Spectrum of poly(m-carborane-1, 7 -dicarbonyl-4, 4'-methylenebis( dimethylaniline) [MC-MDMA]
-----------------% 70 T r a 60 n s m I
I: a n c 30 e
4000 3500 3000 2500 2000 1500 1000 500 __ W_avenumbers (cm-1_) _________________________ -·-·---
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
.... 0)
\0
Figure 6. FT -IR Spectrum of poly(m-carborane-1, 7 -dicarbonyl-4, 4'methylenebis( diethylaniline) [MC-MDEA]
··- - --% T
40
m i t t a n c e
-------
--· ·- -...--------- -.----,----- .. -·-. --,---.--, '- ·- -,.-----· ..,. ··- . --. ····--r---·r--- ... -,---.- .---. --.--,.---.-
4000 3500 3000 2500 2000 1500 1000 yvay~~~mb~S. (<?~:~J __ ·-···-·· __
., I
500
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
..... \0 0
Figure 7 (a). FT-IR spectrum of m-carborane. [MC]
220 ~/""-'"-~" 200
180 I % I
I T 160 !
I
140 I I ,
120 I I 100 I
II
4000 3500 3000 2500 2000 1500 Wavenumbers (cm-1)
Figure 7 (b). FT-IR spectrum of m-carborane-dicarboxylic acid. [MC-DA]
/~ Jo.
. ~. /' 100
\
~\,
"~
1000 500
% T SOJ I \ I
j ·~,\/~\ t\ I // I 1 ~ I I'
\ \
\} \1 " .. . '}
4000 3500 3000 2500 2000 1500 1000 500 Wavenumbers ( cm-1 )
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX2.
Counting Data for Boron-Containing Materials
191
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
0% B Epoxy 5% B Epoxy 10% Epoxy 15% Epoxy 20% Epoxy In foil 3/17/ 94 3/22/94 3/23/94 3/29/94 3/30/94 3/10/94
~ Cts Ln Cts Ln Cts Ln Cts Ln Cts Ln Cts Ln Tim BG=16 BG=16 BG=14 BG=16 BG=14 BG=l4 = (m) ~ 11 4792 8.475 1373 7.225 759 6.632 485 6.184 303 5.714 4134 8.327 ~ 16 4471 8.405 1298 7.169 723 6.583 453 6.116 319 5.765 4138 8.328
& 21 4153 8.332 1268 7.145 702 6.554 424 6.050 274 5.613 3844 8.254 26 3925 8.275 1188 7.080 633 6.450 425 6.052 315 5.753 3616 8.193 31 3580 8.183 1103 7.006 610 6.413 335 5.814 340 5.829 3391 8.129 =· Q I 36 3400 8.132 993 6.901 558 6.324 323 5.778 291 5.673 3295 8.100 az .... 41 3198 8.070 957 6.864 519 6.252 371 5.916 253 5.533 3092 8.037
\0 46 2955 7.991 861 6.758 547 6.304 290 5.670 246 5.505 2891 7.969 cr ~ N
51 2802 7.938 864 6.762 474 6.161 306 5.724 216 5.375 2669 7.889 ~~ 56 2716 7.907 752 6.623 483 6.180 261 5.565 213 5.361 2474 7.814 , ... 61 2465 7.810 723 6.583 379 5.938 244 5.497 200 5.298 2304 7.742 Q Q 66 2391 7.779 718 6.576 366 5.903 230 5.438 166 5.112 2170 7.682 ~~ 71 2110 7.654 663 6.497 366 5.903 225 5.416 165 5.106 2106 7.653 76 2085 7.643 619 6.428 346 5.846 204 5.318 162 5.088 1840 7.518 cr 81 1933 7.567 572 6.349 330 5.799 200 5.298 153 5.030 1829 7.512 rl.l 86 1828 7.511 524 6.261 281 5.638 180 5.193 157 5.056 1656 7.412
Q ... 91 1712 7.445 521 6.256 249 5.517 188 5.236 132 4.883 1509 7.319 ., 96 1594 7.374 487 6.188 245 5.501 158 5.063 121 4.796 1446 7.277 =· 101 1448 7.278 450 6.109 243 5.493 139 4.934 129 4.860 1408 7.250 Q
106 1430 7.265 449 6.107 251 5.525 123 4.812 115 4.745 1297 7.168 =-111 1370 7.223 403 5.999 184 5.215 159 5.069 112 4.718 1174 7.068
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Indium Foil Analysis Boron-Epoxy
Linear Best-Fit Equation: Ln (Counts) - -M x time + B0
KATEBIAL Slope. M Intercept. Be
no shielding 0.0129 ± 0.0002 8.53 ± 0.03
O%B-Epoxy 0.0127 ± 0.0001 8.60 ± 0.02
5%B-Epoxy 0.0124 ± 0.0002 7.37 ± 0.03
10%B-Epoxy 0.0134 ± 0.0005 6.82 ± 0.06
15%B-Epoxy 0.0128 ± 0.0006 6.31 ± 0.08
20%B-Epoxy 0.0116 ± 0.0006 5.99 ± 0.08
Slope (M) and intercept (B0 ) data from regression analysis.
193
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
~ 5%8,5 mil 5%8, 15 mil PMDA-ODA, 2 mil 10%8, 15 mil Indium Foil t;1
0 9.547 0 9.500 0 9.590 0 9.305 0 9.604 ~ 11 9.423 13 9.339 12 9.427 11 9.174 12 9.457
~7 16 9.360 18 9.281 17 9.377 16 9.108 17 9.414 21 9.285 23 9.214 22 9.314 21 9.050 22 9.348 26 9.229 28 9.158 27 9.261 26 8.990 27 9.261
~~ 31 9.151 33 9.085 32 9.193 31 8.919 32 9.202 36 9.067 38 9.043 37 9.151 36 8.864 37 9.1~8 oa .... 41 9.050 43 8.951 42 9.057 41 8.786 42 9.104 \0 t:;JIIIl ~ 46 8.966 48 8.933 47 9.006 46 8.723 47 9.056 51 8.934 54 8.836 52 8.973 51 8.679 52 8.972 >8 56 8.863 58 8.768 57 8.888 56 8.638 57 8.915
f~ 61 8.786 63 8.711 62 8.814 61 8.548 62 8.840 66 8.741 68 8.664 67 8.776 66 8.474 67 8.797 71 8.664 73 8.597 72 8.709 71 8.449 72 8.7~8 76 8.615 78 8.538 77 8.629 76 8.376 77 8.668 rn 0 .., 81 8.561 83 8.491 82 8.588 81 8.297 82 8.649 "C 86 8.477 88 8.419 87 8.503 86 8.266 87 8.553 fll"t-...... 91 8.420 93 8.344 92 8.436 91 8.168 92 8.485 0
=
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Indium Foil Analysis PMDA-ODA
Linear Best-Fit Equation: Ln (Counts) - -M x time + B0
MATERIAL Slope. M Intercept. Be
no shieldinq 0.0121 ± 0.0001 9.60 ± 0.01
PMDA-ODA, 2 mil 0.0124 ± 0.0001 9.59 ± 0.01
5% Boron, 5 mil 0.0123 ± 0.0001 9.55 ± 0.01
5% Boron, 15 mil 0.0123 ± 0.0001 9.50 ± 0.01
lOt Boron, 15 mil 0.0123 ± 0.0001 9.31 ± 0.01
Slope (M) and intercept (B0 ) data from reqression analysis.
195
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Boron-containing PMDA-ODA Film
Neutron Absorption Calculations
ln C(S1S) = 9.546644 - 0.0~234 (t)
ln C(Kap1:on) 9.590223 - 0.01237 (t)
ln C(10 115) = 9.305481 - 0.0~232 ( t)
ln C(foil) = 9.604140 - 0.01206 (1:)
Evaluating each equation a-c tille - 30 minu1:es and taking the na-cural log yields
C( foil) { 30 I = 101325 counts
C(Kapton){30} a 10 1088 counts
C{5 1 5)(301 9 1667 counts
C(l0 115){30} 7 1600 coun1:s
Su.bt.rac-.:inq the counts of each boron-loaded film from the counts of pure Kap1:on film yields
C(Kap1:on)(JO} - C(S~5)(30! = 10088 - 9667 = 421 counts
C(Kap1:on){JO} - C(10 1l5)(30} = 10088- 7600 2488 coun1:s
Theore1:ically I the lOt B-loaded film at 15 mils thick should provide six times as much pro1:ection from neu1:rons as the 5% film a1: 5 mils thick, 2 times for the percent boron loading, and 3 times for the thickness difference.
Comparing the above numbers shows
5% film coun1:s 1 10% film counts • 2488 1 421 • 5.9
196
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX3.
XPS, SSIMS Spectra
197
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
.... \0 0)
XPS Spectra of Pure Polypyromellitimide Films
12·~--------------------------------~
en 1o .... ·-c: ;:)
~ 8 I! .... :e 6 ~ ~4~ ·u; c: Cl)
'E 2 -
C1s
01s
N 1s
(+3)
0~--------~--------~------~~~~~~ -600 -500 -400 -300 -200
-Binding Energy (eV)
r=------control - Exposed in LEO I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
.... \0 \0
XPS Spectra of Pure Polypyromellitimide Films, C 1s Peak
4~----------------------------------------------~
3.5 -.!! ·- 3 c ~
~2.5 l! .. :e 2 ~ ~1.5 ·-en 1: s 1 c
0.5 -~~~=--.. ______ _
0 +-----~----~------~----~----~----~~----~----~ -295 -293 -291 -289 -287 -285 -283 -281 -279
-Binding Energy (eV)
r.::..:.-LEoE;~sed (x3.4)- Not Exposed I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
8
7.5 -~ 7 c ::J
~6.5 b :a 6 ...
1\J ~ 0 0
b5.5 ·-en c s 5 c -
4.5
4 -410
XPS Spectra of Pure Polypyromellitimide Film, N 1s Peak
-408 -406 -404 -402 -400 -398 -Binding Energy
j- Exposed in LEO -Not Exposed (x1.4) I
-396 -394
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
N 0 ....
XPS Spectra of Pure Polypyromellitimide Films, 0 1 s Peak
4 .
-3.5 lJ ·c: ::J 3 ~ l! .. :e 2.5 ~ ~ ·- 2 en c s c -1.5
1~----~----.-----.------.-----.-----.----~ ~40 -538 -536 -534 -532 -530 -528 -526
-Binding Energy (eV)
[-=. E~posed in LEO - Not Exposed I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
l\l 0 l\l
XPS Spectra of Cured 934 Epoxy C1s Peak
6 -
-5 J!J "2 :::J4 ~ f! ... €3 ~ ~ ·u; 2 c s c - 1
OT-----.----.-----.----~----~--~----~----~--~
~00 -294 -292 ~90 ~88 ~86 ~84 -Binding Energy (eV)
-282 -280
I _..:_ LEO Exposed (x1. 7) - Not Exposed I
-278
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
N 0 w
XPS Spectra of Cured 934 Epoxy with 10°/o Boron Powder, C1s Peak
7 ~----------------------------------------~-------.
6 -~ :§5 ~ ~4 :0 ... ~3 ~ en C2 s c
1
0 +-----~--~----~----~----~----~----~--~----~ -296 -294 -292 -290 -288 -286 -284 -282 -280 -278
-Binding Energy (eV)
r==LEO E~p~s;d(x3) - Not Expos_;._-]
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
8
7 -J!) ·- 6 c ::J
~5 e ... :e4
1\) <( -0
~ ~3 en c .!2 c -
1
0
-292 -290
XPS Spectra of Cured 934 Epoxy with 20°/o Boron Powder, C1s Peak
-288 -286 -284 -282 -Binding Energy (eV)
~------LEO E;o;;d (x1. 7) - Not Exposed I
-280 -278
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
IV 0 U1
XPS Spectra of Cured 934 Epoxy N1s Peak
1.6 .
-lJ "2 1.4 ::J ~ l! .... :e 1.2 ~ ~ ·u; c s 1 c -
0.8 +----.----r----.------,-------r------r------,-----i
-408 -406 -404 -402 -400 -398 -396 -394 -392 -Binding Energy (eV)
1- LEO Exposed -··_ Not Expo~d-(x1~) J
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
"' 0 CJ'I
XPS Spectra of Cured 934 Epoxy with 10% Boron Powder, N1s Peak
1.8 .
-J!J ·- 1.6 c ::J ~ l! ... :e1.4 ~ ~ ·-en c s 1.2 c
1 +-----~--~----~----~----~--~----~----~--~
-410 -408 -406 -404 -402 -400 -398 -396 -394 -392 -Binding Energy (eV)
f=-t-EOExposed - Not Exposed (x1.6) I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
N 0 -...I
XPS Spectra of Cured 934 Epoxy with 20°/o Boron Powder, N1s Peak
1.5~-------------------------------------------~
s1.4 ·-c :J ~ e 1.3 ., :a :c -~ 1.2 ·-f/J c .s ..5 1.1
1 +-----------~----------~----------~--------~ -410 -405 -400 -395 -390
-Binding Energy (eV)
,...._ LEOE;posed- Not Exposed j
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
1\J 0 (X)
XPS Spectra of Cured 934 Epoxy with 20°/o Boron Powder, 01s Peak
4 -
3.5 -.!! ·-:5 3 ~ l! ~ 2.5 .Q ... c( - 2 b ·-In ~ 1.5 .... c
1
0.5 +-----~------~----~~----~------~----~----~
~40 -538 -536 -534 -532 -530 -528 -526 -Binding Energy (eV)
j- LEO E~~;d-- - Not Exposed (x 1. 7) j
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
N 0 \0
XPS Spectra of Cured 934 Epoxy with 10% Boron Powder, 01s Peak
3.5 ~--------------------------------------~------~
- 3 J!l ·-c ::J ~2.5
e == -e 2 c( -~ ·u; 1.5 c s c
1
0.5 +-----,.----..-------.------.------~--~-----.-------r-------1
-542 -540 -538 -536 -534 -532 -530 -528 -526 -524 -Binding Energy (eV)
1-LEO Exposed - Not Exposed (x1.5) I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
N ~ 0
XPS Spectra of Cured 934 Epoxy 01s Peak
5~----------------------------------~------------.
~4 ·-c ::>
~3 .... :c .. ~ ~2 ·-en c s .!:1
0+---~~--~----~----~----~--~~--~----~--~
-542 -540 -538 -536 -534 -532 -530 -528 -526 -524 -Binding Energy (eV)
1-LEO Exposed - -·-· Not Exposed (x1.9) I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
N .... ....
-J!! ·-c
XPS Spectra of Cured 934 Epoxy with 20°/o Boron Powder, S2p Peak
1000~----------------------------------~------~
900
:J 800 ~ l! ..., ·-.c 700 ... <C -~ ·- 600 en c G) ..., c -
500
400+---~----~----~----~--~----~----~--~----~
-176 -174 -172 -170 -168 -166 -164 -162 -160 -158 -Binding Energy (eV)
1- LEO Exposed --u---==N~ExPo;-d-C~~S)I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
N .... N
XPS Spectra of Cured 934 Epoxy with 10°k Boron Powder, S2p Peak
2.6 -
2.4
-.!J ·c: 2.2 =» ~ 2 l! ..., ·--e 1.8 ~ ~1.6 Cl) c s 1.4 c
1.2
1 .
-176 -174 -172 -170 -168 -166 -Binding Energy (eV)
[- LEO Exposed - Not Exposed (x2.2) I
-164 -162
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
N .... w
XPS Spectra of Cured 934 Epoxy S2p Peak
2.2 • -
2 -J!i ·;:: 1.8 =» ~1.6 ... ... :e 1.4 ~ ~1.2 UJ c s 1 c
0.8
0.6 -4----.-----.-------.-----r-----r-----r----..----i
-176 -174 -172 -170 -168 -166 -164 -162 -160 -Binding Energy (eV)
ju:.::_ LEO E;pose-d - Not Exposed (x3.8) j
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
"' ~ ~
XPS Spectra of Cured 934 Epoxy with 20°/o Boron Powder, B1s Peak
3500~----------------------------------------------~
3000 -J!! ·-:5 2500 ~ .s 2000 ·-.Q ... 5.1500 ~ ·-U)
~ 1000 .... c
500
0~------~------~r-------.--------.------~ -194 -192 -190 -188 -186 -184
-Binding Energy (eV)
[- LEO Exposed -- Not Exposed (x1 0.6) j
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
XPS Spectra of Cured 934 Epoxy · with 10°/o Boron Powder, B1s Peak
2500 ..,------------------., ........... U) ... ·-c
:::> 2000 + ...................................................................................... ..
~ ~ ...
~ :e 1500 + ............ . Ul <(
...........
~ ·~ 1000 ~ ............................................................................. . Q) ... c - 500 I I ,.____ ___ I I I I I I I I I ~:;:~<>li
-196 -191 -186 -181 -Binding Energy (eV)
r=- Not Exposed - LEO exposed I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
-J!J ·-r::: ::l
~ I! ... ·--e ~ 1:\) ....
0'1 a-·-en c Q) ... c
XPS Spectra of Cured 934 Epoxy F1s Peak
1.2 ,-----------------=--------'-------.
1
0.8 +----.----r---~--..----r----..------r-------,r------i
-696 -694 -692 -690 -688 -686 -684 -682 -680 -678 -Binding Energy (eV)
[-== LEO-E-;,;,;ed (x1.3) ---- Not Exposed ---~
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
l\J .... -..J
-J!) ·-c
XPS Spectra of Cured 934 Epoxy with 10o/o Boron Powder, F1s Peak
1.4 !""-------------------=-------
~ 1.2 I! ... ·-.a ~
~ ~ ·- 1 en c Q) ... c
0.8 +---r----.-------.------.----,......--~---.----.-----i
-696 -694 -692 -690 -688 -686 -684 -682 -680 -678 -Binding Energy (eV)
F LE-O E~po~ed (;6)=-~N~Ex;;o~i--J
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
N .... Q)
XPS Spectra of Cured 934 Epoxy with 20°k Boron Powder, F1s Peak
1.6 ,------------------......-------~
-J!J "i: :l 1 4 ~-
e .., :e ~ b ·u; 1.2 c: .s c:
1 +-----~--~----~----~----~--~----~----~----4
-696 -694 -692 -690 -688 -686 -684 -682 -680 -678 -Binding Energy (eV)
1..:....: LEO-Exp~~~d-(xi9) - Not Expos;d- I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
l\J .... \0
XPS Spectra of Cured 934 Epoxy Si2p Peak
1.8 -.-----------------_.._-----.--.......,...-------,
i1.6 ·-c ::J
~ f! 1.4 ..., :s ~ ~ 1.2 ·-U) c .s .5 1
0.8 +----r----.----.-----.------.----,---.,.---,-----i
-112 -110 -108 -106 -104 -102 -100 -98 -96 -94 -Binding Energy (eV)
~--- LEO Exposed (x1.3) - Not Exposed I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
IV IV 0
-
XPS Spectra of Cured 934 Epoxy with 10% Boron Powder, Si2p Peak
1.6 ,----------------------'"---~
~1.4 c :::> ~ ! .., ·--e 1.2 c( -~ ·-Cl) c s 1 c
0.8 -t---.-------.---~---r---..---~----,-----r-----1
-112 -110 -108 -106 -104 -102 -100 -98 -96 -94 -Binding Energy (eV)
r=- LEO Exposed - Not Exposed (x1.3) I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
N N ....
XPS Spectra of Cured 934 Epoxy with 20°/o Boron Powder, Si2p Peak
1.6 ,--------------------.----~
-J!! ·- 1 4 c . =» ~ f! .... ·--e 1.2 ~ a-·-en c J!! 1 c
0.8 +-----~--~----~----~----~--~----~----~--~
-112 -110 -108 -106 -104 -102 -100 -98 -96 -94 -Binding Energy (eV)
1-=:LEO -Exp-os~d - Not E;~sed (x2.6) j
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
N N N
20
16 -
.!112 -
c :I
8 8 -
4 -
0 120
~ '1
u l
SSIMS Spectra of Cured 934 Epoxy with 20% Boron Powder
I
} fl n
125
I u ... ~
~ I '
130 mass Units
~ (
1 u
135
[=-E;p~dlnlEOU=- Not Exposed I
n r-1
l L 1A 140
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
Static SIMS Spectra Pure Polypyromellitimide Films
450
400
350
300
J!J 250 c :I
l\J 8 200 l\J w
150
100
50
0 0 5 10 15 20 25 30 35 40 45 50
Mass Units
[- Exposed In LEO - Not Exposed j
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
"' E -·-1.1. l! CD .... "C u ·-CD E a.;; tn= -tn CD :i E U5 e u ~
;::3 ~ .ern~
e :::J
D..
0 0 ct)
0 LO N
0 0 N
0 LO ......
!QUnO:)
224
0 0 ......
0 LO
0
0 0 ......
0 CJ)
LO CX)
0 ........
LO co
0 co
0 LO
"0 Q) (I)
0 ~ )(
w 0 z
0 w ...J
.E "0 Q) (I) 0 ~ )(
w
I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
N N Ul
Static SIMS Spectra Pure Polypyromellitimide Films
40 .
30
' J!J c ::::s 20 0 0
10
0
100 105 110 115 120 125 130 135 140 145 150 Mass Units
j_:_ Exposed In LEO - Not Exposed I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
160
140
120
100 .!1
lV § 80 lV 0 0'1 0
60
40
20
0 0 5 10
Static SIMS Spectra of Cured 934 Epoxy
15 20 25 30 Mass Units
35
1-Exposed in LEO - Not.Exposed I
40 45 50
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
N N -...!
SSIMS Spectra of Cured 934 Epoxy with 10% Boron Powder
250~------------------------------------~----~
200
II 150 c ::s 8 100
50
0 I ... o M ssP, of'! ....... W In PI !en .. UfV I I
0 10 20 30 Mass Units
1-Exposed in LEO - Not Expo~ed --~
40 50
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
N N 03
SSIMS Spectra of Cured 934 Epoxy with 20o/o Boron Powder
·300~--------------------------------------------~
250
200
!l § 150 0 0
100
50
0 1 112 a a a a t~d rwuwt~ rnA+ , "·.,.... . w, a A'W - ...... • ._. • 4 :r::Q'I4 I I I
0 10 20 30 40 50 mass Units
1-Exposed in LEO - Not -E~p~s-ed I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
IV IV \0
SSIMS Spectra of Cured 934 Epoxy with 20% Boron Powder
100~------------------------------------------~
80
!I 60 c :I 0 0 40
20
0 I .... 1 *II' 1 WIW • .,.,, I eriJ niYf •·, It'= eeF 1 ftl.a& •fP"l
0 10 20 30 40 50 Mass Units
I =-scraped with Razor - Original I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
80
70
60
50 IJ
"' §40 w 0 0
u 30
20
10
0 50 55 60
Static SIMS Spectra of Cured 934 Epoxy
65 70 75 80 Mass Units
85
~-=- Exposed in LEO----=- N~t Exposed I
90 95 100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
~ 0 Q, w ... ... ~i -aO .a. ... c ~ 0 u ... .... o om ":::e l.o ..,0 u.,.. CD o..c tn~
tn i: ::i -tn tn
0 It)
0 N
0 en 0 co
squnoo
231
0 (f)
0
0 0 ~
0 en
0 co
0 I'
0 co
0 It)
.!! -c ::» • • ca :&
"0 CD fl) 0 Q. )( w -0 z
0 w ...J c: ·-"0 CD fl) 0 Q. )( w
I
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
>a >< 0 a. w ... llllltCD f')~ eft ~ ~0 CDQ. ... c ::::s 0
CJ ... .._ 0 om ~~ .. 0 .. 0 UN CD a..c tn~ tn ~ ::E -UJ UJ
0 It)
0 C\1
0 co
squnoo
232
0
0 0 ~
0 0)
0 co
0 .......
0 co
0 It)
.!J -c ~
• • ftl E
"C Q) (I) 0 Q.. )( w -0 z
0 w ..... .5 "C Q) (I) 0 Q.. )( w
I
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
~ )( 0 a. w .. .. CD
~i ,o CDC. ~ c o e .... 0 om Cl~ '-o ... 0 UN CD c..c tn~ tn ~ :E -tn 0
0 U)
0 0 ('I) N
squno:>
233
0
0 0 ~
0 en
0 CX)
0 t--
0 co
0 U)
.! -c :::» • ., CCI =e
'iii c:: ·a, ·;:: 0
I '-0 N aJ a:: ~ -"i "'D G) ~
l! (J
en
I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
N w
""'
.!
Static SIMS Spectra of Cured 934 Epoxy
20~--------------------------------------------~
15
§ 10 0 u
5
0 100 105 110 115 120 125 130 135 140 145 150
Mass Units
l..::. E;posed i~ LEO - Not Exposed- I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
N w Ul
SSIMS Spectra of Cured 934 Epoxy with 10% Boron Powder
20~-----------------------------------------------.
16
.a 12 c ~
8 8
4
0 100 110 120 130 140 150
Mass Units
~--- E;osed in LEO - Not Exposed I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
N w 0\
SSIMS Spectra of Cured 934 Epoxy with 20% Boron Powder
20~--------------------------------------------~
16
.rJ 12 c :I
8 8
4
0
100 110 120 130 140 150 mass Units
1- Exposed in LEO - Not Exposed I
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
"" w -.J
SSIMS Spectra of Cured 934 Epoxy with 20% Boron Powder
10~--------------------------------------------~
8
lJ 6 c ::::s 0 u
4
2
0 100 110 120 130 140 150
Mass Units
1-Scraped with-Razor -=--- Or~ina-1 - u I
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX 4.
DSC Thermograms of Carborane Polyamides
238
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
-c --cY'I Cl .r -li ;
"tt
8 >-CD - i 0
11..
< """ ::It :a 0 .! ::It I Cl • 6 0 -l.. 1&.. 0
.Q .., a 0
Cl u ::t:
...
IC 0) 0) -0) c ui ltl rr; N --a ::It
c en 0 IE
::It
c c c c c c:i lri c:i lri c:i ltl N !ij r- ltl N N -
(All) aat.:l ~atiH
239
c 0 c lri c:i .,; N 0 r-- -
0
c:i U'l
I
c d c cY'I
c d ltl N
~g jN I
c d ltl
l~ c c -
c d ltl c
.,; N
G .., Cl l.. ::r .., 0 l.. Cl a.. E Ill ....
r:: i ~
r:: i 0
a
i .. uu
oa
ii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I lc
r~ -c r--CI'J Cl ~ -li
I lc r.... •
I~
i , -8 >.CD - i 0
Q.
~ ; 5! .! I Cl .. a 0 -L LL. 0
.Q ~
I! a Cl
u :::z:
,.
CD 01 01 -r--.... • c r:: r--
I! X
:I In
~ • 1-
240
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
"' .a:. .....
Curva Ia DSC Fila lnfoa •lka3702 Thu Mar 7 19a31a47 1996 Sanpl• Walghta 2.300 •g Corborona-OOA (I> Po I ya~~l da
54.0 ...--·---------
53.5
53.0
52.5
~ 52.0
.. ~ 51.5 u. ., a
::t: 51.0
50.5
50.0
49.5
49.0
48.5
I I Carborona-ODA<I> Polya•ldaa•lka3~7_02 ____ _ Heat Flo• <•W> ••2
I 100.0
r·--------, r---z5o. o -~-.--·-· 150.0 200.0 300.0
T••paratura C"C) fUPio SO. a C T IM.Io no.z. IISO. a c o.o •'" AATKio aa.a C/•'"
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
1\J ,.. 1\J
Curve Ia DSC File lnfoa •lka31105 Tua Mer 12 0.4a 52a .42 1996 Sa.p Ia Wei ghta 2. 600 119 Carlborana-1, 12CTHF> P~lyaftlde
275.0
250.0
225.0
s; 2DO.O .!
• a 175.0 ... Ll.
ISO. 0 J +J D IJ :z::
125.0
100.0
75.0
50.0 -----1---··----- --I
so. 0 100. 0 150. 0
liR: sa.. a .so. a fi TI..-Jo a. a •In "-''TIUo aa.a C,l'•&n
I I Carborana-1. 12CTHF> Polya.ldaa•lka311DS Heat Flow <•W> ••2
-r 2DO.O
T ••perature C • C)
r--250.0
I 300.0
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited w
ithout permission.
1\J .... w
Cw-ve I• DSC File Info• •lke31104 Tue Mar 12 03• 32• 38 1996 Sa.ple lfelghta 5. 300 119 Ccrborana-1. 10 .Palyo.lde
325.0
300.0
275.0
250.0 ~ .! 225.0 .. 0 .... 200.0 1 u. ., D 175.0 Cl ~
150.0
125.0
100.0
75.0
50.0 I
50. D
Jl:'= d8:8 S ruca.
I 100. D
I ISO. 0
0. a aln ItA Rio Ia. a C/aln
I I Ccrborane-1.10 Polya~~ld•••lke311_0_4 __ _
Ueat Flow C.lf> •• 2
I 200.0
Te11perature ('C>
.-250.0
I
I
., ___ j 300.0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11"1 N
N Q at rrJ Cl ~
ii .g -8 ~ -a
Q.
Cl c: -8 .; I
Ill
j' Cl e L. a
~ L.J ..
N
i
.... • .! • a u. ., a Cl
::z:
I
I
I I
L----r--.-Q Q Q
g d N s - - -
244
Q
r~ I I r i I I
l ~d i~ L I I I I
I
r~ ..... u .
I ..... I Cl
L L. ::) .,
I a L.
I !. lo &
Cl
L· ,_
,~ c i
I ;,
L a
I ci
I
i I Ia ~ L· <
~~ II:
c ,- i
I a ci
r-I l i i Q ..
d uu Q Q Q Q Q -d d a a
r-- CD au d Q
d Cl
.:w I -
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
VITA
Michael B. Glasgow
Born in Akron, Ohio on April 25, 1966. Earned a
Bachelor of Science degree in Chemistry from Virginia
Polytechnic Institute and State University on May 5, 1989.
Married Dr. Bonnie Louise Brown, DVM, the next day. Was
employed by James R. Reed & Associates as a chemistry
technician for one year. Entered the Master's program in
Chemistry at the College of William & Mary in 1990. Entered
the Doctorate program in Applied Science in January, 1991.
Was a part-time chemistry instructor at Thomas Nelson
Community College, 1994. After completion of the Ph.D.
degree, the author will seek gainful employment as a polymer
scientist in the chemical industry.
245
top related