University of Southern California Dissertations and Theses

Long term lunar radiation degradation of potential lunar habitat composite materials

(USC Thesis Other)

Long term lunar radiation degradation of potential lunar habitat composite materials

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

LONG TERM LUNAR RADIATION DEGRADATION OF POTENTIAL LUNAR
HABITAT COMPOSITE MATERIALS

by

Kristina Rojdev

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(ASTRONAUTICAL ENGINEERING)

December 2012

Copyright 2012 Kristina Rojdev
ii

Dedication
To my husband, Kent, who never asked me when I would finish, but always knew
that I would and believed in me every step of the way.

iii
Acknowledgments
First, I would like to thank the professors of the Astronautical Engineering
department at USC for giving me the opportunity to pursue my research and passion
in an unconventional manner. This work would not have been possible without
their support. I would also like to thank Professor Nutt and Bill Atwell, who both
took a chance on me to help guide me in this endeavor. They have given me
strength, advice, and knowledge throughout this process, and I am grateful for that.
I would like to thank NASA and my management for giving me the chance to
pursue this work and helping me accomplish this study. I could not have
accomplished this feat without the dedication of the people who worked on this
project for several years with me and my management for providing me with the
flexibility to pursue my goals both professionally and personally. Thank you for
helping me, teaching me, and putting up with me.
Finally, I want to acknowledge my friends and family for believing in me
during this time. I am grateful for their patience and support while I’ve pursued this
endeavor.
iv
Table of Contents
Dedication ii
Acknowledgments iii
List of Tables vi
List of Figures vii
Abstract xv
Chapter 1: Introduction 1
1.1. The Mission 2
1.1.1. The Lunar Habitat Design 3
1.1.2. The Moon 5
1.2. Previous Studies 6
1.2.1. General Radiation Effects on Materials 6
1.2.2. Polyethylene and Polypropylene 8
1.2.3. Epoxies and Composites 10
1.2.4. Materials in Space Environments 14
1.3. Literature Summary and Problem Definition 16
1.4. Dissertation Outline 19

Chapter 2: Preliminary Analyses and Assumptions 20
2.1. Habitat Radiation Exposure 20
2.1.1. Lunar Radiation Environment 20
2.1.2. HZETRN 30
2.1.3. Limitations 37
2.2. Habitat Stresses 38
2.2.1. Habitat Loads 38
2.2.2. Potential Habitat Materials 41
2.2.3. Habitat Stresses for Minimum Gauge Thickness 43

Chapter 3: Materials and Methods 44
3.1. Material Selection and Manufacture 44
3.2. Radiation Exposure 50

Chapter 4: In-situ Strain Analysis 52
4.1. Test Setup 52
4.2. Calculations of Material Strain Due to Thermal Changes 55
4.3. Results 57
4.3.1. Fast Dose Rate Samples 57
4.3.2. Slow Dose Rate Samples 58
4.4. Discussion 60
4.5. Summary and Conclusions 65
v

Chapter 5: Radiation and Dose Rate Effects 67
5.1. Radiation Exposure 67
5.2. Characterization Methods 69
5.2.1. Tensile Testing 69
5.2.2. Flexure Testing 70
5.2.3. Differential Scanning Calorimetry 72
5.2.4. Fourier Transform Infrared Spectroscopy 74
5.2.5. Scanning Electron Microscopy 75
5.3. Results 75
5.3.1. Tensile Results 75
5.3.2. Flexure Results 82
5.3.3. DSC Results 87
5.3.4. FTIR Results 91
5.3.5. SEM Results 93
5.4. Discussion 96
5.4.1. Composite Durability 97
5.4.2. Toughening 102
5.4.3. Dose Rate Effects 104
5.4.4. Aging 105
5.4.5. Fiber Debonding 105
5.5. Conclusions 107

Chapter 6: Synergistic Effects of Tension and Radiation 110
6.1. Test Setup and Methods 110
6.2. Results and Discussion 112
6.2.1. Tensile Results 112
6.2.2. Flexure Results 117
6.2.3. DSC Results 122
6.2.4. FTIR Results 124
6.3. Conclusions 127

Chapter 7: Surface Investigation of Visible Aging 128
7.1. Test Setup and Methods 128
7.2. Results and Discussion 129
7.3. Conclusions 135

Chapter 8: Conclusions 136

References 139

Appendix A: Basic Radiation Terminology 146
Appendix B: Proton Range in Epoxy - SRIM Study 152
Appendix C: Strain Correlation to Worst-Case Hoop Stress 155
vi
List of Tables

2.1. Habitat structure minimum gauge 42

2.2. Hoop stress and longitudinal stress calculations 43

3.1. Ply angles for the composite layup design 45

4.1. Description of radiation exposures conducted 55

4.2. Coefficient of thermal expansion for each material 56

5.1. Exposure details and characterization method for each panel. In
the method column, T represents tensile test, F represents flexure
test, DSC represents differential scanning calorimetry, and FTIR
represents Fourier Transform Infrared Spectroscopy 67

5.2. A comparison of the percent change in glass transition
temperature for the BF-CF samples irradiated to 5,000 Gy at fast
and slow dose rates 89

5.3. Calculated dose at the interface between the fiber and epoxy
compared with a neat epoxy at the same location for each
material 107

6.1. Exposure details and characterization method for each panel
subjected to tension. In the method column, T represents tensile
test, F represents flexure test, DSC represents differential
scanning calorimetry, and FTIR represents Fourier Transform
Infrared Spectroscopy 111

7.1. Exposure details for each panel of the BF-CF and CF materials
investigated 129

C.1. Data gathered from tensile tests and used to calculate the
average strain on these materials at ~41 MPa 157

vii
List of Figures

1.1. Conceptual representation of the cylindrical hard-shell lunar
habitat 4

1.2. Primary track, delta ray tracks, and spurs 7

2.1. Overview of primary radiation flux densities in free space 21

2.2. Solar particle events for cycles 19-21 22

2.3. Time evolution of an SPE in December 2006 23

2.4. Differential and integral spectra of the band function fit of the
combined October 1989 solar particle events 24

2.5. GCR composition as compared to the solar system composition 25

2.6. Solar cycle modulation of GCRs. The solar cycle is shown in red
and the GCR flux is shown in blue. 26

2.7. Differential energy spectrum of GCR secondary radiation
production with CREME86 solar minimum model 28

2.8. Differential energy spectrum of SPE secondary radiation
production with CREME96 worst day model 29

2.9. Differential flux at the lunar surface for GCRs at solar minimum 32

2.10. Differential flux at the lunar surface for GCRs at solar maximum 32

2.11. Historically large SPEs fitted to a double power law (Band
function) 33

2.12. The dose accumulated by a material completely exposed to the
lunar environment 35

2.13. The accumulated dose for a material shielded behind MLI and
MMSE 35

2.14. Total dose to materials over the mission 36

2.15. Free-body diagram of a cylindrical pressure vessel 39

2.16. Free-body diagram of hoop stress on a cylindrical pressure vessel 40
viii

3.1. Hand layup process for pre-preg composites 46

3.2. Temperature profile for autoclave curing of CF composite 47

3.3. Hydraulic press for curing composites 48

3.4. Temperature profile for press curing of BF-CF composite 49

3.5. C-scan comparison of two samples 50

3.6. Stacked configuration at IUCF 51

4.1. A close up view of the grips of the test stand 52

4.2. A sample tensioned in the test stand with a bi-axial strain gauge
located in the center of the sample 53

4.3. Data acquisition system used to collect data during radiation
exposures 54

4.4. Example screen shot of data collection during radiation exposure.
The top graph shows the strain data and the bottom graph shows
the temperature data 54

4.5. Strain data from Exposure 6, example of a fast dose rate in-situ
strain response 57

4.6. Measured temperature data from Exposure 6 58

4.7. Strain data from Exposure 1, example of a slow dose rate in-situ
strain response 59

4.8. Measured temperature data from Exposure 1 60

4.9. An example (BF-CF-3) of measured strain overlaid with
calculated thermal expansion of the aluminum frame and the
BF-CF-epoxy material sample. This figure shows an example
of a sample’s measured strain that did not match the calculated
thermal expansion of either the aluminum frame or the laminate 61

4.10. An example (CF-12) of measured strain matching the calculated
thermal expansion of the aluminum frame 62

4.11. The percent error between the measured strain and the calculated
thermal expansion of the aluminum frame for CF-12 63
ix

4.12. An example (BF-CF-9) of measured strain matching the calculated
thermal expansion of the laminate 63

4.13. The percent error between the measured strain and the calculated
thermal expansion of the laminate for BF-CF-9 64

5.1. A representative stress-strain curve resulting from a tensile test of
an irradiated coupon. From the data, the following quantities are
gathered: modulus, ultimate strength, fracture strength, strain-to-
failure, and first fracture point 70

5.2. Example of a sample undergoing flexure testing 71

5.3. A representative stress-strain curve resulting from a flexure test of
a control coupon. From the data, the following quantities are
gathered: modulus, ultimate strength, fracture strength, strain-to-
failure, and first fracture point 72

5.4. Example of a DSC curve for a sample before radiation exposure
(BF-CF #22) 74

5.5. Preparation of a CF coupon for SEM investigation 75

5.6. Stress-strain curve for one representative BF-CF coupon of the
control group, fast group, and slow group 76

5.7. Stress-strain curve for one representative CF coupon of the control
group, fast group, and slow group 77

5.8. Calculated modulus (■ – left side), ultimate strength (Δ – right
side), and fracture strength (○ – right side) for each BF-CF coupon
investigated and plotted against the dose rate exposure 78

5.9. Calculated fracture energy (■ – left side) and strain-to-failure (◊ -
right side) for each BF-CF coupon investigated and plotted against
the dose rate exposure 79

5.10. Averaged BF-CF first fracture point data for the dose rates
investigated 79

5.11. Calculated modulus (■ – left side), ultimate strength (Δ – right
side), fracture strength (○ – right side), and first fracture point (⟡ -
right side) for each CF coupon investigated and plotted against the
dose rate exposure 81

x
5.12. Calculated fracture energy (■ – left side) and strain-to-failure (◊ -
right side) for each CF coupon investigated and plotted against the
dose rate exposure 81

5.13. Flexural stress-strain curve for representative BF-CF coupons of
the control group, fast group, and slow group 83

5.14. Flexural stress-strain curve for representative CF coupons of the
control group, fast group, and slow group 83

5.15. Calculated flexural modulus (■ – left side), first fracture point (◊ -
right side), ultimate strength (Δ – right side), and fracture strength
(○ – right side) for each BF-CF coupon investigated and plotted
against the dose rate exposure 84

5.16. Calculated flexural strain-to-failure (◊ - left side) for each BF-CF
coupon investigated and plotted against the dose rate exposure 85

5.17. Calculated flexural modulus (■ – left side), first fracture point (⟡ -
right side), ultimate strength (Δ – right side), and fracture strength
(○ – right side) for each CF coupon investigated and plotted
against the dose rate exposure 86

5.18. Calculated flexural strain-to-failure (◊ - left side) for each CF
coupon investigated and plotted against the dose rate exposure 86

5.19. Average percent change in glass transition temperature for each
exposure group evaluated of the BF-CF material 88

5.20. Tg of the CF material as a function of dose rate 90

5.21. FTIR averaged spectra of the center locations of the BF-CF
samples for fast (___) and slow (---) exposures and the averaged
pre-irradiation scan is subtracted from the averaged post-
irradiation scan 91

5.22. FTIR averaged spectra of the center locations of the CF samples
for fast (___) and slow (---) exposures and the averaged pre-
irradiation scan is subtracted from the averaged post-irradiation
scan 92

5.23. SEM micrographs of BF-CF tensile coupons post-fracture that
were not exposed to radiation 94

xi
5.24. SEM micrographs of BF-CF tensile coupons post-fracture exposed
to 5,000 Gy. The bottom right micrograph shows the interface
between a boron fiber and the epoxy 94

5.25. SEM micrographs of CF tensile coupons post-fracture that were
not exposed to radiation 95

5.26. SEM micrographs of CF tensile coupons post-fracture exposed to
5,000 Gy 96

5.27. Averaged pre-radiation FTIR spectrum of all ten BF-CF (___) and
CF (---) samples scanned with FTIR prior to being exposed to
radiation 98

5.28. Averaged pre-radiation FTIR spectrum of all ten BF-CF (___)
and CF (---) samples scanned with FTIR prior to being exposed
to radiation in the 1800-600 range. The black arrows indicate
locations in the BF-CF spectrum and the blue arrows indicate
locations in the CF spectrum 99

5.29. IR spectrum of DGEBA 99

5.30. Overlaid averaged FTIR spectra of the BF-CF and CF materials
pre-radiation and peak comparison with PES. The black arrows
indicate locations in the BF-CF spectrum and the blue arrows
indicate locations in the CF spectrum 103

5.31. Model of BF-CF for HZETRN dose calculations at the boron
fiber and epoxy interface 106

6.1. Representative BF-CF tensile stress-strain curves for each
exposure group. Blue represents the slow dose rate, black
represents the fast dose rate, (___) represents a coupon that
underwent only radiation, and (---) represents a coupon that
underwent radiation and tension 112

6.2. Representative CF tensile stress-strain curves for each exposure
group. Blue represents the slow dose rate, black represents the
fast dose rate, (___) represents a coupon that underwent only
radiation, and (---) represents a coupon that underwent radiation
and tension 113

xii
6.3. The averaged tensile modulus, first fracture point, ultimate
strength, and fracture strength for the BF-CF material as a
function of dose rate. The blue represents samples only exposed
to radiation and the black represents samples exposed to both
radiation and tension 114

6.4. The averaged tensile strain-to-failure for the BF-CF material as
a function of dose rate. The blue represents samples only
exposed to radiation and the black represents samples exposed
to both radiation and tension 114

6.5. The averaged tensile modulus, first fracture point, ultimate
strength, and fracture strength for the CF material as a function
of dose rate. The blue represents samples only exposed to
radiation and the black represents samples exposed to both
radiation and tension 115

6.6. The averaged tensile strain-to-failure for the CF material as a
function of dose rate. The blue represent samples only exposed
to radiation and the black represents samples exposed to both
radiation and tension 115

6.7. Representative BF-CF flexure stress-strain curves for each
exposure group. Blue represents the slow dose rate, black
represents the fast dose rate, (___) represents a coupon that
underwent only radiation, and (---) represents a coupon that
underwent radiation and tension 117

6.8. Representative CF flexure stress-strain curves for each
exposure group. Blue represents the slow dose rate, black
represents the fast dose rate, (___) represents a coupon that
underwent only radiation, and (---) represents a coupon that
underwent radiation and tension 118

6.9. The averaged flexural modulus, first fracture point, ultimate
strength, and fracture strength for the BF-CF material as a
function of dose rate. The blue represents samples only
exposed to radiation and the black represents samples exposed
to both radiation and tension 119

6.10. The averaged flexure strain-to-failure for the BF-CF material as
a function of dose rate. The blue represents samples only
exposed to radiation and the black represents samples exposed
to both radiation and tension 119

xiii
6.11. The averaged flexural modulus, first fracture point, ultimate
strength, and fracture strength for the CF material as a
function of dose rate. The blue represents samples only
exposed to radiation and the black represents samples exposed
to both radiation and tension 120

6.12. The averaged flexural strain-to-failure for the CF material as a
function of dose rate. The blue represent samples only exposed
to radiation and the black represents samples exposed to both
radiation and tension 120

6.13. Percent change in Tg of the BF-CF material as a function of
dose rate. The blue represents samples that underwent
radiation only and the black represents samples that underwent
both tension and radiation 123

6.14. The Tg as a function of dose rate for the CF material. The blue
represents samples that underwent radiation only and the black
represents samples that underwent both tension and radiation 124

6.15. FTIR data of the BF-CF material with fast dose rate data on the
left and slow dose rate data on the right. The blue represents
radiation and tension exposure and the black represents
radiation only exposure 125

6.16. FTIR data of the CF material with fast dose rate data on the left
and slow dose rate data on the right. The blue represents
radiation and tension exposure and the black represents
radiation only exposure 126

7.1. Micrographs of the resin surface of BF-CF#2 where the sample
was only exposed to radiation. The top left micrograph was
taken prior to radiation exposure, the top right micrograph was
taken ~1 month after radiation exposure, and the bottom
micrographs were taken ~9 months after radiation exposure 130

7.2. Micrographs of the resin surface of BF-CF#5 where the sample
was exposed to both radiation and tension. The top left
micrograph was taken prior to radiation exposure, the top right
micrograph was taken ~1 month after radiation exposure, and
the bottom micrographs were taken ~9 months after radiation
exposure 131

xiv
7.3. Micrographs of the resin surface of CF#7 where the sample
was only exposed to radiation. The top left micrograph was
taken prior to radiation exposure, the top right micrograph was
taken ~1 month after radiation exposure, and the bottom
micrographs were taken ~9 months after radiation exposure 133

7.4. Micrographs of the resin surface of CF#27 where the sample
was exposed to both radiation and tension. The top
micrograph was taken ~1 month after radiation exposure and
the bottom micrographs were taken ~9 months after radiation
exposure 134

B.1. SRIM setup screen showing proton radiation with a maximum
of 200 MeV energy range and a generic epoxy compound
from the SRIM internal “compound dictionary” 153

B.2. Stopping power and range for the simulation of protons in an
epoxy 154

C.1. Stress-strain curve for the control BF-CF coupons examined 156

C.2. Stress-strain curve for the control CF coupons examined 156

xv
Abstract

NASA’s charter for exploration missions could take humans to deep space,
asteroids, the Moon, and eventually Mars. Each of these missions necessitates a safe and
productive place for the crew to live and work, namely deep space habitats. Long-term
habitation requires the use of large structures which must withstand the environment for
the duration of the extended-stay missions.
Recently, fiber-reinforced composite materials have gained interest as a potential
structural material for deep space and planetary habitats. These materials can provide
weight savings, potentially enhanced radiation protection for the crew, and lend
themselves to cutting-edge research when compared to existing metals. However, these
materials have not been characterized for the space environment, and particularly the
space radiation environment, which is known to cause damage to polymeric materials.
Thus, this study focused on a lunar habitation element and the integrity of potential
composite materials after exposure to a simulated long-term lunar radiation environment.
Two aerospace composites of interest to NASA were chosen for the study and
were subjected to a thirty-year simulated lunar radiation environment. The durability of
the materials was investigated during the radiation exposure, and the material properties
were characterized post-radiation exposure. Additionally, the effects of dose rate and
synergistic tendencies between radiation and tension were examined. Finally, an
exploration of the surface of the materials was completed to determine whether there
were indications of accelerated aging as a result of the radiation exposure.
1

Chapter 1

Introduction
NASA is interested in long-duration missions to various destinations, and
missions that involve long-term stays of crew require large structures to fulfill volumetric
habitability requirements for crew survival equipment, stowage, and living space. To
keep costs low and allow for the launch of such large structures, these structures need to
be extremely lightweight. Thus, there is an increased interest and focus on lightweight
polymeric composites that can provide similar strength to those metals typically used on
spacecraft.
Composites may have an advantage over metals with respect to weight, and there
may be added benefits to studying and investing in composite technologies. For instance,
composites are being investigated for implementation of new technologies, such as
embedded sensors, which could be used in structural health monitoring. There is also
new technology being developed in self-healing polymers, which may be applicable to
habitats exposed to a micrometeorite environment. A third benefit to using composites is
shown in preliminary studies of radiation shielding properties in which there are
potentially enhanced radiation mitigation properties of polymeric composites when
compared to metals for structures exposed to harsh space radiation environments [5, 66].
Hence, composites lend themselves to multi-use benefits by providing strong structural
support, potential enhanced radiation shielding capability, and the infusion of new
technologies to maintain structures over long-durations.
2
At this point, there are still several unknowns in using composite materials for
space applications. For instance, composites are not homogenous, and have very different
properties when compared to metals. In short, previous knowledge of strength
characteristics, aging properties, or failure modes of metals is not applicable to
composites. While composites are used extensively in terrestrial commercial
applications, their durability and aging characteristics in a long-term deep space
environment are unknown. Polymers are known for their susceptibility to radiation
degradation, and contemporary aerospace composites have not been characterized for
tolerance in the harsh space radiation environment on the lunar surface or in deep space.
Since the lifetime and durability of these materials is essentially unknown in the space
environment, we have no information on the impacts of radiation tolerance of the
material as it ages. Accordingly, it is extremely important to understand and test these
materials in similarly harsh environments that they will be exposed to during their service
on a space structure before there is full-confidence in these materials.
The following dissertation discusses a three-year study that examined the effects
of long-term radiation exposure to certain fiber-reinforced composite materials of interest
for deep space habitation structures.

1.1 The Mission
When this study began, NASA’s mission, Constellation, was to return to the
Moon, establish a long-term outpost, and provide a safe and productive place for the crew
to live and work. Throughout this study, NASA’s charter has evolved to encompass
potential deep space missions or even missions to an asteroid. Whether the destination
3
remains the Moon, many of the assumptions and aspects of this study are still applicable
to long-term stays on asteroids or deep space missions. So, even though NASA’s mission
has changed, the details of this study are relevant and will be presented with respect to
the original mission of a lunar outpost.

1.1.1 The Lunar Habitat Design
With long-term deep space habitation, there are many important structures that
need to be considered. One of the main functional pieces is the habitat, in which
astronauts will spend the majority of their time living and working in a “shirt-sleeve”
environment. In planning a deep space mission, NASA has considered several different
types of habitats: (1) a hard, cylindrical pressure vessel; (2) an inflatable torus; and (3) a
cylindrical hybrid, the ends of which are hard and the middle section of which is
inflatable. For the purposes of this study, the focus will be on the cylindrical hard-shelled
pressure vessel, (1).
This cylindrical, rigid, pressure vessel is a self-contained unit that can support a
crew of four on the lunar surface. This habitat has a three meter internal diameter with an
8.35 meter length, providing an approximate pressurized volume of 59 cubic meters. A
conceptual representation of this habitat is shown below in Figure 1.1.
4

Figure 1.1: Conceptual representation of the cylindrical hard-shell lunar habitat [57].

This habitat contains all the subsystems needed for human survival on the lunar surface,
including power, thermal, life support, avionics, and environmental protection.
Part of the life support system is the atmosphere in which the crew must live and
work. The intent is to make the atmospheric pressure in the vessel as similar to Earth’s as
possible (approximately 101.4 kPa or 14.7 psi) which creates more comfort for the crew.
However, there are certain considerations that need to be contemplated in order to
balance the comfort of the crew with their safety. The main consideration is the
Extravehicular Activity (EVA) that the astronauts perform. As part of their duties on the
lunar surface, the crew will be required to leave the habitat and explore the lunar surface
in pressurized suits. Since it is difficult to work in a 101.4 kPa pressurized suit, the
pressure of the suit must be significantly reduced. This creates complications if the
habitat is at ambient pressure and the crew must perform an EVA because the
crewmembers can undergo decompression sickness, in which nitrogen bubbles from the
blood and tissue move to other areas of the body. Thus, the trade-offs are to either have
the astronauts perform a “pre-breathe” procedure or to equalize the pressures in the
5
habitat and suit so they can eliminate the “pre-breathe” procedure. The “pre-breathe”
protocol is performed by astronauts breathing 100% oxygen at ambient and then slowly
reducing the pressure over a long period of time to acclimate their bodies to the lower
pressure of the suit before EVA, similar to the current procedure used on the International
Space Station [23]. Due to the multitude of EVAs planned for the lunar surface, the lunar
habitat pressurized environment is defined as a low pressure of ~55.2 kPa (8 psi) and a
high oxygen environment of 32% (NASA, 2008).

1.1.2. The Moon
The Moon is the closest body to Earth at a distance of 356,410 kilometers (km) at
the closest portion of its orbit, and 406,697 km at the farthest. The Moon and the Earth
orbit around the sun at 1 astronomical unit (AU) or 149,597,870 km, and the Moon orbits
the Earth in a slightly elliptical fashion, such that the rate of the Moon’s orbit and its
rotation are identical. Thus, the same hemisphere of the Moon always points towards
Earth [25].
Compared to the Earth, the lunar environment is extremely harsh. Some of the
most difficult lunar environments within which a habitat will need to survive are the
almost non-existent atmosphere and vacuum environment, large thermal extremes,
micrometeorite, surface ejecta, and radiation. Every aspect of the lunar environment
poses a challenge for polymeric composites, and the focus of this study is on the radiation
environment’s effect on potential habitat composite materials.

6
1.2 Previous Studies
The field of radiation effects on materials has primarily focused on materials for
the nuclear industry (targets, dosimeters, electronic components, etc.) and the medical
industry (sterilization of components). Thus, the types of radiation that have been
investigated are typically gamma radiation or electron radiation, and at low energies.
Additionally, the materials of interest in these industries are typically aliphatic (linear
polymer chains) and aromatic polymers (polymer chains including ring structures), which
would not be used as aerospace structural materials. Few studies have investigated early
carbon fiber reinforced composites in the space environment. These studies focused on
low Earth orbit environments, and primarily on electron radiation at low energies. The
following discussion will detail some of the previous work that has been accomplished in
the field of radiation effects on materials and discuss what is lacking with regards to
understanding fiber reinforced composite durability in deep space radiation
environments.

1.2.1. General Radiation Effects on Materials
There have been several studies that have looked at radiation exposure of
polymeric materials for different applications. In Al-Sheikhly’s tutorial [4], radiation
effects on materials are discussed in general, and he shows that incident radiation
interacts with materials through ionization and excitation of the molecules in the material,
leading to secondary radiation which creates further ionization and excitation. With
proton radiation, an ion track is created in the material as a result of high ionization
density along the path of the incident ion [22, 79, 61]. The track tends to grow radially
7
outward due to secondary electron production from the incident ions. These areas of
dense ionization located close to the original track are called “spurs.” The secondary
electrons produced can also create additional tracks moving away from the incident track,
referred to as “δ-rays.” The spurs and δ-rays are shown in the diagram below (Figure
1.2).

Figure 1.2: Primary track, delta ray tracks, and spurs [22].

The ionizations in these tracks are locations of chemical changes that occur in the
material as a result of energy transfers from the initial particles to the electrons with
which they interact. These energy transfers leads to events such as chain scission,
production of free radicals, excited species, volatiles, oxidation, and the cross-linking of
chains [4]. The most basic chemical change that occurs in the material is the creation of
free radicals when electrons are stripped from the molecules [61, 15]. These free radicals
typically prefer to have the valence shell of their atom filled; it follows that, they can be
very reactive and create bonds to other free radicals in the system or bond to other
molecules such as oxygen.
δ-ray
Primary particle
track
spurs
8
Free radicals lead to three processes which cause degradation within polymers
exposed to radiation [4, 63, 79]. The first is chain scission, in which the backbone bonds
of the polymer chain are broken. This results in a weakened material. The second is
cross-linking, in which the radicals from within the main polymer chain begin bonding to
one another. Cross-linking is unique in that it initially stiffens the material and
strengthens it. But, as the cross-link density increases, the material becomes brittle. The
third is known as oxidative degradation. Oxidative degradation occurs when a material
with free radicals is exposed to an environment in which oxygen is present. The oxygen
atom is highly reactive and will bond to the free radicals. Then, because the oxygen atom
has six electrons in its valence shell, and the free radical only takes up one location in the
valence shell, the oxygen continues to bond with other radicals and other oxygen atoms
around it. Thus, oxygen increases degradation within the material, and in the appropriate
conditions, oxygen can greatly enhance the degradation process.

1.2.2. Polyethylene and Polypropylene
Parkinson and Sisman [63] reviewed several aliphatic and aromatic polymers that
have been investigated and describe their durability to radiation exposure. Polyethylene
[42, 43] and polypropylene [13, 39, 81, 2, 3, 11] are perhaps the most studied aliphatic
materials in this field; because of their simple chemistry, it is easier to isolate chemical
changes in polyethylene and polypropylene than in more complex polymers. The
literature [42, 43, 13, 39, 2, 3, 11] shows that polyethylene is typically shown to cross-
link, becoming hard and brittle as it degrades, and polypropylene is shown to
predominantly exhibit scission, becoming softer and flexible.
9
In a study by Kudoh et al. [43], polyethylene was irradiated at room temperature
under vacuum with various ions at different energies and electrons at 2 MeV. The study
found that the elongation-at-break decreased with increasing radiation dosage and the
tensile strength increased at higher doses, two indications of cross-linking occurring with
radiation exposure. This study also found the results to be independent of the type and
energy of radiation exposure, meaning there was no linear energy transfer (LET) effect
exhibited by this material. Typically the process of creating chemical changes in a
polymer (ionization tracks, free radicals, etc.) is the same regardless of the radiation type.
However, the concentration of radicals in the ionization tracks could differ with the
radiation type as a result of different rates of energy deposition. The difference in the
yield of reactive radicals is known as the LET effect.
In another study by Kudoh et al. [42], polyethylene was irradiated with electrons
and gamma radiation. After radiation exposure, the polyethylene formed an insoluble gel
fraction, further supporting the conclusion that cross-linking is occurring. If polyethylene
is irradiated in the presence of oxygen, a competition between cross-linking and scission
occurs, with scission predominating at lower dose rates [29, 17] as a result of oxidative
degradation. A typical indicator of oxidative degradation is the carbonyl group in the
infrared spectra [4, 29, 17, 65].
In contrast to polyethylene, polypropylene is considered to be a polymer that
predominantly scissions with radiation exposure, as discussed in the work by Black and
Lyons [13]. In this study, polypropylene was irradiated under vacuum with 2 MeV
electrons. The number molecular average weight decreased and the sol-gel study showed
swelling of the polymer in toluene. The samples were also investigated with infrared
10
spectroscopy. Evidence of cross-linking, chain scission, as well as unsaturation of
vinylidene was found. Several studies [39, 81, 2, 3, 11] have also investigated
polypropylene irradiated in air at room temperature (as polypropylene is a common
material used in the medical industry and is typically sterilized with gamma radiation).
When polypropylene is irradiated in air with gamma radiation, oxidative degradation
causes further chain scission, as shown by carbonyl growth in infrared spectroscopy [39,
3] and changes in mechanical properties [39, 2, 3].
In another study of polypropylene, Hegazy et al. [33] focused on the oxidative
degradation that takes place post-exposure. Polypropylene was studied both in powder
form and in film sheets made from powder. Polypropylene samples were exposed to
gamma rays with a dose rate of 2.78 Gy/s. The study showed that oxygen enhances the
degradation of polypropylene as revealed by the decrease of the tensile strength and
percent elongation at break for the polypropylene films.
When polypropylene was irradiated with 62 MeV protons in air, however,
Tripathy et al. [81] found evidence of polypropylene cross-linking rather than undergoing
scission. The differences observed between the two studies could be a result of LET
effects being apparent for polypropylene, which were not apparent for polyethylene.

1.2.3. Epoxies and Composites
Most aliphatic polymers are known to be susceptible to radiation effects. Studies
have shown that aromatic polymers are more radiation resistant because their ring
structure absorbs excitation energy without concentrating it at a particular location within
the molecule [4, 63, 79, 61, 15]. Furthermore, the aromatic rings can act as “sinks” for
11
energy absorbed by other parts of the molecule [79]. These protective features are less
apparent when irradiated in the presence of oxygen, where oxidation can occur.
Parkinson and Sisman [63] note that there are differences in effects when epoxies are
irradiated in air with electrons versus gamma radiation, which is most likely a result of
the differing dose rates. At high dose rates, oxygen may not be able to diffuse into
material as easily and, thus, depletes quickly. For this reason, the thickness of the epoxy
sample also plays an important role when observing the oxidation that results from
radiation.
Longieras et al. [48] studied the epoxy system diglycidyl ether of bisphenol-A
with hardener triethylene tetramine (DGEBA/TETA) and exposed it to electrons in
oxygen and in helium to a total dose of 5 MGy. Using NMR and IR spectroscopy,
Longieras et al. showed an increase in radiolysis products such as vinyl amine, carboxylic
acids, esters, and amides, with losses in methylene amine. The analysis concluded that
the main product of DGEBA/TETA irradiated in oxygen was carboxylic acid end-groups
as a result of chain scission which further produced more chain scission reactions. In
addition, the study showed that the oxidation was limited to the surface (a few microns).
If there are longer timescales from the irradiation to the characterization, it is
possible that these oxidation effects will proliferate deeper into the thickness of the
material as a result of long-lived radicals slowly interacting with oxygen as it diffuses
into the material. Kent et al. [37] investigated the tetraglycidyl-4,4’-diaminodiphenyl
methane (TGDDM) resin cured with the 4,4’-diamino-diphenyl sulfone (DDS) hardener
at differing ratios to understand how cross-link density affected irradiated epoxies. Kent
et al.’s samples were irradiated with gamma radiation in nitrogen and at cryogenic
12
temperatures to impede radical decay so that the samples could be examined via electron
spin resonance (ESR) spectroscopy. The results showed the same basic ESR structure
regardless of the amount of hardener and exhibited a fast decay of radicals initially with a
much slower decay of radicals following. Kent et al. concluded that radical decay rates
are related to inhomogeneities in the system and areas which are more cross-linked will
contain radicals with longer decay rates. In less cross-linked regions, there is more chain
mobility which, in turn, facilitates rapid recombination. Conversely, highly cross-linked
regions will be less mobile and restrict recombination of radicals. Furthermore, the type
of radicals produced is also a factor. Larger organic radicals formed from large
molecules tend to be less reactive since the unpaired electron can be distributed over a
greater molecular volume [79]. Since radicals from highly cross-linked epoxy systems
are available over longer time scales, it is possible that oxygen may have a greater chance
of interaction as the oxygen diffuses into the material, thereby causing additional
oxidation.
A study by Fornes [27] also considered the effects of oxygen during irradiation on
the epoxy (TGDDM/DDS) of the composite T-300/5208. For this study, cured samples
were placed in bags filled with either air, nitrogen, or were vacuum sealed. Then, the
samples were exposed to 0.5 MeV electrons. Using contact angle measurements, infrared
(IR) measurements, and electron spectroscopy for chemical analysis measurements,
Fornes found a significant increase in oxygen content observed on the irradiated epoxy
surface (37% increase), irradiated graphite fiber (159% increase), and the fracture surface
of the irradiated composite (51% increase). The increased oxygen content was due to
radicals that interacted with oxygen at the interface during and after exposure to
13
radiation. Furthermore, IR spectroscopy showed an increase in the absorbance of the
carbonyl region, indicating oxidative degradation. In addition, the study showed that the
concentration of long-lived radicals increases with the dose.
Romanov et al. [68] investigated epoxies exposed to 9 MeV protons in air and at
room temperature. They found that with proton radiation the epoxy increased in strength
characteristics up to a point (65 MPa at 0.1 MGy) and then started to decrease in strength
with further dosage (50 MPa at 1MGy). The analysis pointed to cross-linking as the
mechanism for the initial strength increase, with additional dose leading ultimately to the
degradation of the epoxy and weakening of the composite. The discussion did not
mention oxidation effects.
Hoffman and Skidmore [34] investigated carbon fiber composites with gamma
radiation and examined oxidation effects. The epoxy was based on bisphenol-A with an
aliphatic amine hardener and the composite was irradiated to 0.5, 1, and 2 MGy (50, 100,
and 200 Mrad). The results showed a decrease in modulus and fracture stress with color
changes in the resin of the irradiated tensile coupons. Using differential scanning
calorimetry (DSC), an increase in the glass transition temperature was observed.
Scanning electron microscopy confirmed degradation of the epoxy with increased dosage
and Fourier Transform Infrared Spectroscopy (FTIR) showed an increase in absorbance
of the carbonyl region, indicating oxidation.
Of the studies discussed above, the study by Romanov [68] had results different
from others studying both epoxies and fiber reinforced composites in air. In the study by
Romanov, the samples were irradiated with protons at low energies, whereas the other
studies were irradiated with electrons or gamma radiation. The overall conclusion was
14
that the type of radiation and the energy of the radiation could affect the results. That
said, the dose levels of Romanov’s study were much lower than the rest of the studies. It
is likely that, at the lower doses, the effects of oxidation were not as apparent since the
study by Fornes [27] mentioned that the concentration of radicals increased with dose,
providing more opportunities for reactions.

1.2.4. Materials in Space Environments
In general, there are few reports on the susceptibility of composites to harsh space
environments [18, 51, 46, 45, 49]. In 2004, Maji and Mahnke [49] wrote a review paper
on the degradation of composites in mid-Earth orbit. In one study, 10-100 MGy (1,000-
10,000 Mrads) of 1 MeV electrons were used to simulate a dose of thirty years in a
Geostationary orbit (GEO). The material consisted of carbon fibers in a matrix of
polyetherimide. Results in the matrix-dominated direction showed a 10% increase in
modulus, a 22% decrease in strength and a 96% decrease in strain-to-failure as a result of
additional cross-linking. There was no change in these properties in the fiber-dominated
direction, indicating that the degradation occurred in the matrix instead of within the
fibers. In another study discussed in this review, three epoxies were subjected to 0.1 and
10 MGy (10 and 1,000 Mrad) doses. One of the epoxies studied became too brittle for
the applicable tests. This result occurred mainly due to excessive cross-linking that took
place in the material.
Milkovich et al. [51] studied a common aerospace composite of the 1980s,
T300/934. Several laminate configurations were investigated and exposed to 1 MeV
electron radiation at a dose rate of 1.8E9 Gy/s to a total dose of 100 MGy, simulating a
15
composite in Earth orbit exposed to the Van Allen radiation belts. In addition to the
radiation exposure, Milkovich et al.’s samples were either subjected to a cold
temperature, room temperature, or a hot temperature to investigate the material properties
in the combined thermal and radiation environments. The authors reported that the
electron radiation degraded the in-plane strength properties as a result of radiation
interaction with the matrix chemistry and noted that this was most severe in laminate
configurations that were tested perpendicular to the fiber direction.
Leung et al. [46] studied T300/934 and exposed the material to gamma radiation
of several doses ranging from 0.44 to 3.2 MGy (comparable to three years at GEO) at a
dose rate of 0.714 Gy/s. The results of this study showed the glass transition temperature
(T
g
) decreased with radiation exposure, indicating an increased molecular mobility of the
matrix, but also showed the interlaminar shear strength to initially increase and then
decrease to its original value through the range of doses. The authors did not discuss
these results in great detail and only mentioned that longer-term radiation exposure would
become detrimental to the material, which seems consistent with the results found by
Milkovich [51] at the higher doses studied.
Kurland et al. [45] also studied T300/934 along with T300/5208 and
C6000/P1700. They exposed the composites to a simulated GEO radiation environment
of 700 keV electrons to doses between 10 MGy and 100 MGy at a dose rate ~80-100
Gy/s. This again simulated the electron environment in the trapped Van Allen belts. The
results showed no significant differences between the composites exposed to radiation as
compared with the controls, but the trends showed slightly increased stiffness and
strength with decreased ultimate elongation. These results indicate matrix cross-linking
16
which contradicts the results suggested in Milkovich’s study [51]. Furthermore, Kurland
et al. [45] recommended expanding the study to include other types of radiation in the
space environment such as high energy protons, low energy charged particles, and
ultraviolet radiation.
Coulter et al. [18] expanded upon Kurland’s [45] work by investigating Narmco
5208 epoxy exposed to 3 MeV protons under vacuum. The doses studied ranged from
0.5-100 MGy, comparable to the other studies, and the dose rates ranged from 1.5-26
kGy/s. Coulter [18] mentioned in his introduction that the mechanisms and results of
proton radiation are different than electron radiation because protons are more massive,
produce increased ionization density and excitation of atomic electrons, and can cause
more displacements in materials. The results of Coulter’s study showed that there were
no measureable changes in mechanical properties or FTIR spectra. However, UV-visible
absorption spectra showed indications of cross-linking and ESR spectra showed evidence
of radical decay species that decayed over weeks, assumed by Coulter to be a
recombination of two radicals to form new cross-links. Coulter’s [18] results for Narmco
5208 epoxy are in line with the results from Kurland [45] with T300/5208.

1.3. Literature Summary and Problem Definition
The above discussion reviews previous studies in the field of radiation effects on
materials. In general, there are several papers on the effects of radiation with respect to
polyethylene and polypropylene, which show the basic effects prominent in this field:
cross-linking or chain scission. The reasons for predominant scissions or cross-linking in
some materials are unclear, although there are indications that the chemical structure of
17
the material can provide some insight. Reviewing the exhaustive work on polyethylene
and polypropylene is useful for understanding the basic underlying effects. However, as
the chemistries of polyethylene and polypropylene are different from that of aerospace
composites, studies focused on polyethylene or polypropylene do not provide an adequate
comparison for the ways in which composites will react to radiation. In addition, the
work on polyethylene and polypropylene has primarily centered on the nuclear industry
or the medical industry, in which different forms of radiation are encountered and
different doses are experienced.
Epoxies are more complicated chemically than are polyethylene and
polypropylene, and given the aromaticity in the chemical structure, it seems as though
epoxies are affected to a lesser degree. Thus, there are significantly fewer studies on
epoxies than there are on aliphatic polymers. As was shown by the literature review
above, when irradiated in the presence of oxygen, epoxies can be susceptible to
degradation. Furthermore, when reviewing the literature on space radiation effects on
epoxies and composites, there are very few published studies. These space environment
studies were completed in the 1980s when composites became of interest to NASA for
Earth orbit missions, particularly GEO. At that time, the focus was on Earth orbit
environments, primarily the Van Allen radiation belts which contain electrons and
protons. The majority of these studies focused on the electron environment in the Van
Allen belts and fairly low energies of electrons. NASA is looking to expand explorations
to deep space, in which the radiation environment is quite different from LEO, and
includes high-energy protons from solar particle events (SPEs) and very high-energy
particles from galactic cosmic rays (GCRs). The studies published to date have not
18
investigated these very high energies, alternate radiation sources, and their impact on
composite structural materials that might be considered for deep space habitats.
Furthermore, the composites of the 1980s were typically untoughened polymer
matrices. As composite usage has expanded in the aerospace industry, toughening of the
matrix has become a standard practice to reduce the brittle nature of epoxies, which
typically include either rubber particles or thermoplastics [75, 7, 53]. Previous studies of
composites in space radiation environments have not considered the current state of the
art of toughened composites.
Finally, previous studies on composites for space environments were heavily
focused on irradiations in a vacuum since the materials were being used as exterior
components in Earth orbit. For a deep space habitat in which crews will be living and
working, however, the habitat’s interior will be pressurized with air and the habitat will
be exposed to higher energies of radiation, which penetrate more deeply than some of the
radiation experienced in Earth orbit. This means structural habitat composites will not
only be exposed to radiation in the presence of oxygen, but will also be subjected to
internal pressure stresses during the irradiation. As was shown in the literature review of
epoxies, irradiation in an oxygenated environment can significantly affect the epoxy. If
spacecraft designers consider using these commercially-available composites for the
structural components of the habitat, it is necessary to understand the ways in which these
composites will fare when exposed to high-energy deep-space radiation in an air
environment to determine possible affects arising from interactions with oxygen.
Furthermore, there have not been any studies to investigate composites exposed to
radiation while simultaneously subjected to stress. Hence, the goals of this study are to
19
expand scientific knowledge with respect to composites exposed to deep space radiation
environments, and particularly structural composites for lunar habitats. The objectives of
this dissertation are to better understand the durability of two candidate fiber-reinforced
semi-toughened habitat composites a thirty-year mission.

1.4. Dissertation Outline
This dissertation continues in Chapter 2 with background on the lunar radiation
environment and simulations of that environment to calculate dose estimates for the
study. Additionally, there is a discussion of habitat stress and calculations of worst-case
potential stresses an internal pressure environment might impart to a habitat material.
Chapter 3 reviews the materials chosen for this study, the process of manufacturing the
materials, and the radiation exposures. Chapter 4 is the first experiment that examines
the effects to these materials during radiation exposure, while Chapter 5 assesses the
mechanical durability and chemical changes after radiation exposure. Chapter 6
investigates whether there are any synergistic effects of combined radiation and tension.
Chapter 7 is an investigation of the surface of the materials over time to determine
whether there are indications of enhanced aging. Finally, Chapter 8 briefly summarizes
the main findings, conclusions, and discusses areas in need of further investigation.
20
Chapter 2

Preliminary Analyses and Assumptions
Prior to developing the experimental portion of this study, there were several
details that needed to be understood. The following sections will provide the
assumptions for this study and some analyses of the radiation exposures and stresses a
habitat might receive during a long-term lunar surface mission.

2. 1 Habitat Radiation Exposure
There are two primary sources of radiation in the lunar environment, those from
solar particle events (SPE) and those from galactic cosmic radiation (GCR). There is also
secondary radiation on the lunar surface due to the interaction of high energy GCRs with
the lunar regolith. The following description will provide background on these radiation
sources and a detailed analytical analysis of the total radiation exposure a habitat might
receive over a long-duration stay on the lunar surface.

2.1.1 Lunar Radiation Environment
The sources of primary radiation are known as ionizing radiation because they
have the ability to remove electrons (or ionize) atoms or molecules with which they
21
interact. An overview of the flux densities for these types of radiation is given in Figure
2.1.

Figure 2.1: Overview of primary radiation flux densities in free space [69].
As illustrated in the figure, solar storm and solar flare protons have a lower
energy than galactic cosmic rays, but higher flux densities. Galactic cosmic rays are the
highest energy particles but the lowest flux density. The energy levels and the flux
density are important parameters that define the level of interaction these types of
radiation will have with materials on the lunar surface.

Solar Particle Events
Solar particle events (SPE) are events that occur periodically and generate high
fluxes of energetic charged particles over a short time period. The particles are primarily
22
composed of protons but can also contain electrons and some heavy elements [31].
These events can occur during any phase of the solar cycle, but they tend to occur more
frequently during the time of solar maximum. Figure 2.2 below shows several SPEs
during solar cycles 19-21.

Figure 2.2: Solar particle events for cycles 19-21 [69].
In the figure, the curved line represents the approximate eleven-year solar cycle.
Peaked areas on the smoothed line represent the maxima of the respective cycles, and the
valleys represent the corresponding minima. The vertical lines are representative of
events and the fluence of protons that they produced. As can be seen from this figure, the
majority of particle events occur during the peak of the solar cycle. However, it can also
be seen that there are other events that occur during the minimum of the cycle. NASA’s
Human Integration Design Handbook [55] states that SPE data collected since 1956
23
shows “…that about 30 to 50 major SPE events occur per cycle, most during the 5 years
corresponding to solar maximum.”
These solar events also carry a time evolution with them. They are not
instantaneous events, but they can last anywhere from a day to weeks. An example of
this SPE time evolution is given in Figure 2.3.

Figure 2.3: Time evolution of an SPE in December 2006 [14].
This data shows the integral flux for each day of the event, and it is broken out by
those particles that are greater than 10 MeV in energy and those particles that are greater
than 60 MeV in energy. This particular event lasted seven days, and the flux shown here
is from a detector at the L1 Lagrangian point.
24
Generally, solar particle events are characterized by a shock emanating from the
low solar corona. This shock then accelerates the particles to fast speeds that initially
propagate along the interplanetary magnetic field (IMF) lines. However, as time
progresses, the directionality decreases and the particles are more isotropic in nature [31].
While there are several solar particle events in a solar cycle, there are typically
only one or two very large events in a solar cycle.” An example of one event that
contained a large fluence at very high energies was in October of 1989 (Figure 2.4),
where there was a series of solar particle events that occurred on October 19
th
, 22
nd
, and
24
th
[5].

Figure 2.4: Differential and integral spectra of the band function fit of the combined October
1989 solar particle events [5].

25
Galactic Cosmic Rays
Galactic Cosmic Rays (GCRs) are a type of very high energy radiation
penetrating the heliosphere from outside the solar system. It is believed that these rays
originate from supernovae outside the solar system. Every element in the periodic table
is included in the GCR particles, and the chemical composition is very similar to that of
the solar system (Figure 2.5).

Figure 2.5: GCR composition as compared to the solar system composition [58].

Of the elements shown here, protons are still the most abundant. However, the
concern with GCRs typically results from the prevalence of other heavy elements, such as
Iron, included in the spectrum.
26
Unlike the solar particle events, GCRs are an ever-present background radiation
and are isotropic within the solar system. However, rather than having a constant fluence
characteristic, the GCRs are modulated by the solar cycle. This modulation is inversely
correlated with the solar maximum and solar minimum of the solar cycle as shown in
Figure 2.6.

Figure 2.6: Solar cycle modulation of GCRs [1]. The solar cycle is shown in red and the GCR
flux is shown in blue.

In this graph, the solar cycle is shown in red and the GCR flux is shown in blue.
As can be seen from the plot, when the solar cycle is at a peak (maximum), the GCR flux
is in a valley (minimum) and vice versa. The cosmic ray modulation is a result of the
interaction of the solar wind with the GCR. During solar maximum, there are more
27
particles traveling within the IMF, and the magnetic field has more irregularity than
during the solar minimum. Therefore, at solar maximum there is much less GCR
intensity due to these interactions.

Secondary Radiation
For the purposes of this study, the primary radiation sources are of most interest.
However, for completeness and to better explain the importance of choosing materials
with radiation mitigating properties for lunar habitats, the following brief discussion will
consider the secondary radiation environment for the lunar surface.
Primary radiation interacting with the lunar regolith creates secondary radiation
primarily composed of neutrons and photons. At this time, little information is known
about the secondary radiation environment on the lunar surface other than information
yielded by a few simulation studies that have been performed. Current efforts are
underway in the space community to gather more in-situ data on secondary radiation to
better understand that environment.
Neutrons are of concern to both crew and electronics because they are able to pass
through material much more easily than ionizing radiation as a result of their lack of
charge. Neutrons are 1800 times larger than electrons, and only slightly larger than
protons (0.16%). When traveling through a material, if a neutron strikes an atom with
enough energy to remove an electron, it becomes charged and can react more easily with
the material through which it is passing. One study used the CREME86 and CREME96
28
models to understand the magnitude and energy levels of secondary radiation production
from GCRs and SPEs. The results are shown below (Figure 2.7 and 2.8).

Figure 2.7: Differential energy spectrum of GCR secondary radiation production with
CREME86 solar minimum model [41].

In Figure 2.7, the primary GCR radiation spectrum is shown along with the
secondary radiation (orange and red curves) that is created due to the primary GCRs. As
can be seen from the graph, the GCRs are very high energy, and the secondary radiation
is much lower energy. However, there is a higher abundance of the secondary radiation
when compared to the GCRs.

Secondary
radiation
GCR
spectrum
29

Figure 2.8: Differential energy spectrum of SPE secondary radiation production with
CREME96 worst day model [41].

In Figure 2.8, the CREME96 “worst day” SPE model was used to generate a SPE
spectrum and the secondary particle production from that spectrum. As in the previous
figure, the secondary radiation is lower in energy than the SPE spectrum. However, the
flux of the secondary radiation is more in line with the SPE flux than it was with the GCR
flux.
In comparing the two graphs above, the lower GCR flux instigated a high
secondary flux where as a high SPE flux instigated a similar secondary flux. Thus, the
GCR radiation creates much more secondary radiation than SPEs create. Given that there
is continuous GCR radiation (different fluxes depending on the cycle), this could create a
significant neutron flux over a large time period, adding to the radiation exposure of crew
and electronics.
SPE Worst Day
Secondary
radiation
30
As more information is gained on the lunar secondary radiation environment, a
better understanding of the lunar radiation environment will result. This information will
also allow for the additional study of how secondary radiation impacts material
durability. Thus, as we learn more about this environment, follow-up studies will need to
examine the effects of this secondary radiation.

2.1.2. HZETRN
To better characterize the radiation environment to which a habitat would be
subjected on the lunar surface, preliminary radiation transport calculations were
performed using a deterministic high-charge-and-energy (HZE) software package
(HZETRN). HZETRN is a one-dimensional formulation of the Boltzmann transport
equation with a straight-ahead approximation and a continuous slowing-down process
[6]. This software contains pre-programmed environments and performs dose
calculations very quickly for aid in the design of spacecraft in various radiation
environments. For this study, HZETRN was used to investigate the absorbed doses a
habitat might experience on the lunar surface when exposed to SPEs and GCRs as
described above. The environments selected in these simulations include the October
1989 series of SPEs, 1989 GCR environment during solar maximum, and 1977 GCR
environment during solar minimum. The GCR environments were arbitrarily chosen and
the October 1989 series of events was chosen for the particularly large amount of
particles at very high energies, as described earlier. One of the habitat materials was
31
simulated in the software by a single slab composed of 24% epoxy, 38% carbon, and
38% boron.

Radiation type and doses
The modeling for this preliminary analysis focused on understanding the total
radiation exposure to a habitat on the lunar surface. To do this, the differential spectra of
primary radiation on the lunar surface were acquired. The following graphs (Figure 2.9
and Figure 2.10) show the spectra at the lunar surface for GCRs during solar minimum
and solar maximum. These spectra depict the differential flux over one year. These
graphs were created from data generated through the CREME96 model, a model
developed by the Naval Research Lab to create numerical models of the space radiation
environment in low earth orbit, to evaluate radiation effects on electronics in spacecraft,
and to estimate the high LET radiation environment in manned spacecraft [83].
32

Figure 2.9: Differential flux at the lunar surface for GCRs at solar minimum.

Figure 2.10: Differential flux at the lunar surface for GCRs at solar maximum.
33
As can be seen from these graphs, the flux during solar minimum conditions is
much higher than the flux at solar maximum. This data is consistent with information
above, in which the GCR exposure is modulated by the solar cycle.
Additionally, there are large solar particle events that take place during the solar
cycle. For studying the exposure on the lunar surface, several historically large SPEs
were examined to understand which one might pose the worst case exposure. These
spectra of large SPEs are shown below (Figure 2.11).

Figure 2.11: Historically large SPEs fitted to a double power law (Band function).

The data for these events were fitted with a double power law in rigidity, known
as the Band function [6]. Compared to previous exponential rigidity fits, this one does
34
not underestimate the fluence at higher energy levels. In comparing the different events,
those that have the highest fluences throughout the spectra, and especially at high
energies, will produce the largest absorbed dose in materials. Thus, the event chosen as
the worst case for this study was the October, 1989 event.
Using HZETRN, the exposures for a ten, twenty or thirty-year mission was
investigated. The software output provides dose information as if the spacecraft were in
free space. Given our assumption that the habitat is on the lunar surface, half of the dose
is removed, since the habitat is protected on one side due to inherent shielding of the
moon. Thus, the dose values were divided by two to accurately represent the absorbed
dose on the lunar surface. The SPE results are given as a total dose for the entire event
since it is a single event, whereas the GCR results are presented as a dose-per-day, since
GCR particles are always present in the background.
The dose was calculated for two different cases. The first case was for a material
on the lunar surface completely exposed to the radiation environment (Figure 2.12). The
second case was for a material that was slightly shielded behind multi-layer insulation
(MLI), typically used for thermal control, and micrometeorite and surface ejecta
shielding (MMSE), which would be used to protect the material from impingement
(Figure 2.13). This second case incorporates more realistic elements of an environment
for materials on the lunar surface. However, the first case would be the worst-case
scenario, which was ultimately used for the study. The dose gathered for these two cases
was further broken into the dose accumulated due to GCR exposure and the dose
accumulated due to SPE exposure. The results are shown below (Figure 2.12 and Figure
2.13).
35

Figure 2.12: The dose accumulated by a material completely exposed to the lunar
environment.

Figure 2.13: The accumulated dose for a material shielded behind MLI and MMSE.
36
In our simulation, the total calculated dose over the five-day October 1989 event
to the skin of a habitat material on the lunar surface is approximately 20,000 cGy (1
centiGray [cGy] = 1 rad), at a dose rate of 4.63E-4 Gy/s (averaged over the five days). In
contrast, the skin of a habitat material on the lunar surface over a five-day period during
solar minimum (worst case) receives a GCR dose of approximately 0.07995 cGy at a
dose rate of 1.85E-9 Gy/s (averaged over the five days), which is 200% lower dose and
subjected to radiation 200% slower than the SPE exposure. Given these values, the
majority of the dose a habitat will experience will be a result of large SPEs that occur
throughout the mission, as shown in the graphs above (Figure 2.12 and Figure 2.13).
Therefore, it is necessary to simulate SPEs with proton radiation, as 98% of SPEs are
composed of protons.

Figure 2.14: Total dose to materials over the mission.
37
Combining this simulation data, the total dose for a ten, twenty, or thirty-year
mission can be calculated, as shown in Figure 2.14. The assumptions used in arriving at
a total dose are the following: the GCR exposure is an average over the solar minimum
and solar maximum conditions, there is one very large SPE per solar cycle
(approximately eleven years), and there is a factor of safety of ten applied to account for
any other SPEs that may have occurred during the solar cycle. The resulting total doses
calculated for these mission lifetimes fall between 1x10
5
and 1x10
6
cGy.
The simulation data collected from this preliminary work was an important step in
determining the type of radiation exposure that needed to be considered and the levels of
exposure for the experimental test setup. Based on the results of the simulations, the
materials in this study were exposed to proton radiation at a total dose of 5x10
5
cGy.

2.1.3 Limitations
These transport calculations provided the basis for determining the radiation
exposure during the experiment. However, simulating the space radiation environment in
its entirety is difficult and cost-prohibitive. Therefore, certain concessions were required
to allow for this type of research. As described earlier, the spectrum of SPEs or GCRs
spans a range of energies, and exposing materials over the entire range of energies is
extremely expensive. In addition, the cyclotron used for the radiation exposures, Indiana
University Cyclotron Facility (IUCF), provides proton energies up to 200 MeV. To
perform irradiation at a lower energy, a degrader would be required, and this can cause a
multitude of secondary radiation. Thus, to preserve the type of primary radiation a
38
habitat would experience in space, we chose to expose the material to only 200 MeV
protons.
A thirty-year mission will include bursts of radiation over time, but materials will
also age with time in addition to the radiation exposure. Ideally, the materials should be
exposed to radiation approximately equal to one solar cycle, then aged equivalent to
eleven years and repeat the procedure twice to arrive at the thirty-year exposure.
However, this process is both cost-prohibitive and time-prohibitive. Therefore, the
radiation exposure was accelerated for practicality, the thirty-year exposure occurred
within one day, and no accelerated aging of the material was performed.

2.2 Habitat Stresses
In addition to the environment, the habitat material is also subject to stresses due
to the internal pressure environment.

2.2.1 Habitat Loads
The internal pressure environment of the habitat set by NASA in combination
with the lunar environment has specific design implications on the habitat [36]. Due to
the lack of atmosphere on the lunar surface and the internal pressure of the habitat, the
predominant stress on the habitat while on the lunar surface is the internal pressure. This
stress is manifested as a biaxial tensile stress on the material of the pressure vessel. From
mechanics of materials, these stresses can be defined in the following manner.
39
Given a cylindrical pressure vessel, with radius R, thickness i, and internal
pressure p
i
, we have the following free-body diagram (Figure 2.15). Note that the
external pressure, P
o
, is zero as this habitat will be on the lunar surface which has no
atmosphere.

Figure 2.15: Free-body diagram of a cylindrical pressure vessel [9].

Since there is no net force exerted on an object with a uniform pressure
distribution, the sum of the forces must equal zero (Equation 2.1).

( ) ( ) 0 2
2
= − R p Rt
i
π π σ
(2.1)

Rearranging Equation 2.1 and solving for the stress gives Equation 2.2.

( )
( ) t
R p
Rt
R p
i i
2 2
2
= =
π
π
σ
(2.2)

This is known as the longitudinal stress and is along the length of the cylinder.
40
The second stress is known as the hoop stress and runs along the circumference of
the habitat. The free-body diagram for this stress is given by isolating a slice of the
cylindrical wall and is shown below (Figure 2.16).

Figure 2.16: Free-body diagram of hoop stress on a cylindrical pressure vessel [9].

Again, since there is no net force exerted on an object with a uniform pressure
distribution, the sum of the forces must equal zero (Equation 2.3).

( ) ( ) 0 2 2 = Δ − Δ x R p x t
i h
σ
(2.3)

Rearranging Equation 2.3 and solving for the stress gives Equation 2.4.

( )
( ) t
R p
x t
x R p
i i
h
=
Δ
Δ
=
2
2
σ
(2.4)

Δx
σ h
σ h
p i
41
Notice that the hoop stress is twice the longitudinal stress on the pressure vessel. Thus, it
is more important to design to the “hoop stress, worst case” than to the longitudinal
stress.

2.2.2. Potential Habitat Materials
There are several different materials being considered for the lunar habitat
structure. These materials range from metallic materials to polymer matrix composite
materials. Aluminum is the reference material which has been used since the beginning
of the space program. There is also interest in composite aluminum-lithium since it is a
metallic material and slightly lighter than aluminum. In the space industry, metallic
materials are well known and have heritage; thus, they are considered less risky.
There has recently been a renewed interest in polymer matrix composites with
reinforced fibers. These materials offer the advantage of lighter weight, potential ease of
manufacturing complex objects, and multifunctionality since they can contribute to the
overall radiation shielding of the habitat. The risk (or concern) with polymer composite
materials is that there is little known about them and they have very little space heritage.
Some of the unknowns with composites are strength characteristics due to differing
orientations of the material during layup, strength characteristics due to deviations in
manufacture, crack propagation and damage tolerance, and life time and durability of the
materials.
There are two types of configurations being considered for polymer composite
materials used on the habitat, and they are a sandwich composite or a skin-stiffened
42
composite. The sandwich composite consists of two face-sheets with honeycomb
sandwiched between the face-sheets. The honeycomb is a lightweight perforated core
material, and the face-sheets are the fiber reinforced polymer composite laminates. The
skin-stiffened composite is essentially several plies of the fiber reinforced polymer
composite with a frame and stringer structural arrangement [24].
For polymeric composite materials, there is typically a minimum gauge thickness
that is allowed to withstand the stresses to which the material will be subjected. Based on
the study by Dorsey and his coauthors, “The minimum gauge for polymeric composite
materials assumes a standard ply thickness of 0.0127 cm and standard laminate design
rules that result in durable and damage tolerant structure….” From this study, the
minimum gauge thickness of the laminate is as follows (Table 2.1).
Table 2.1: Habitat structure minimum gauge [24].
Material
Sandwich application: thickness,
mm
Skin/Laminate application:
thickness, mm
Polymeric Composite
(quasi-isotropic)
1.40 (face-sheet) 2.03
Aluminum honeycomb core 12.70 N/A

This minimum gauge thickness provides the upper limit threshold and represents the
worst case scenario when calculating the stresses imparted on the habitat material due to
the internal pressure.

43
2.2.3. Habitat Stresses for Minimum Gauge Thickness
NASA’s lunar hard-shell habitat concept has an internal radius of 1.5 m and an
internal pressure of 55.2 kPa (8 psi) [57]. Given the equations for hoop stress (Equation
2.4) and longitudinal stress (Equation 2.2), and considering a variety of thicknesses for
the material, a worst-case realistic stress can be generated. In both of these equations, the
pressure and internal radius are constants for this scenario. Therefore, when comparing
the thickness, a thinner material will generate a higher stress-state in the material, as
shown in Table 2.2.
Table 2.2: Hoop stress and longitudinal stress calculations.
Thickness
(mm)
Thickness
(in)
Longitudinal
stress (MPa)
Longitudinal
stress (psi)
Hoop
stress
(MPa)
Hoop
stress
(psi)

0.51 0.02 81.43 11811.02 162.87 23622.05
1.02 0.04 40.72 5905.51 81.43 11811.02
1.52 0.06 27.14 3937.01 54.29 7874.02
2.03 0.08 20.36 2952.76 40.72 5905.51
skin-
stiffened
2.54 0.1 16.29 2362.20 32.57 4724.41
3.05 0.12 13.57 1968.50 27.14 3937.01
3.56 0.14 11.63 1687.29 23.27 3374.58
4.06 0.16 10.18 1476.38 20.36 2952.76
4.57 0.18 9.05 1312.34 18.10 2624.67
5.08 0.2 8.14 1181.10 16.29 2362.20
10.16 0.4 4.07 590.55 8.14 1181.10
15.24 0.6 2.71 393.70 5.43 787.40 sandwich
20.32 0.8 2.04 295.28 4.07 590.55
25.40 1 1.63 236.22 3.26 472.44

This data shows that as the thickness increases, the stress in the material decreases. Thus,
to consider a realistic worst-case scenario, a skin-stiffened thickness would be the
thinnest structure when compared to a sandwich type structure. Given the minimum
gauge thickness of a skin-stiffened structure [24], the worst-case stress imparted to
NASA’s lunar habitat due to internal pressure is approximately 41 MPa.
44
Chapter 3

Materials and Methods
In the following study, two materials were investigated in each of the experiments
performed. The materials were purchased and manufactured in-house. Subsequently, the
materials were exposed to radiation at Indiana University Cyclotron Facility (IUCF) and
several characterization methods were performed to ascertain the material durability to
the radiation. The following details the materials, the manufacturing process, the
radiation exposures, and the characterization methods used in this study.

3.1 Material Selection and Manufacture
There were two commercially available aerospace composite materials chosen for
this study because NASA was most interested in materials that could be used currently
and to save manufacturing time. The first material (IM7/977-3 [19], Cytec Engineered
Materials) was chosen as a baseline carbon fiber composite as it was currently in use at
NASA for some secondary structures and interior components, such as crew quarters on
the International Space Station. The second material (Hybor-208 [78], Specialty
Materials, Inc.) was chosen for its potential as a multifunctional material in providing
primary structural capabilities while also acting as a radiation mitigator. The second
material contained both boron and carbon fibers and the boron fibers were of interest as
neutron absorbers for the secondary neutron albedo on the lunar surface. Both
45
composites contained semi-toughened epoxies and were purchased as unidirectional tapes
with the fibers pre-impregnated (pre-preg) into a partially cured resin.
For ease of use, we wanted a laminate design that would behave similar to an
isotropic material. Thus, a quasi-isotropic, balanced, and symmetric layup was chosen
for both materials. To create this type of layup, the material had to be stacked at different
angles, and the angle of each layer (or ply) is found using Equation 3.1, where k is the
layer number, N is the number of total layers, and θ
0
is an arbitrary angle [8].
(3.1)
In addition, when θ
k
> 90°, Equation 3.2 is used to describe the layer angle.
(3.2)
To keep the layup relatively thin, three was chosen for N, which is the lowest number of
plies that can be used, and an arbitrary initial angle of zero was chosen. These assumed
numbers provided the following angles for the layup design:
Table 3.1: Ply angles for the composite layup design.
Ply Number Ply Angle (θ
k
)
1 60°
2 -60°
3 0°

To make this balanced and symmetric, the layers were doubled and mirrored for a final
layup configuration of [60/-60/0/0/-60/60]. There was one deviation from this layup
design with the CF material used for tensile and flexure tests. Unfortunately, we were
unable to acquire pre-preg material from the manufacture to complete these tests and had
46
to use material available from another project. The layup of the CF material for these
tests was [0/30/-30/-30/30/0] which is not a quasi-isotropic layup, but it is balanced and
symmetric. This means that the material will have slightly anisotropic properties. Given
that we are testing the CF samples in the same way and using control coupons, the
collected data will be comparable within the material data set. But, comparing the BF-CF
results with the CF results for the tensile and flexure tests will not be possible.
The layup procedure for each material was the same. The material was received
from the vendor in a roll where all the fibers were oriented in the same direction (uni-
directional). The material was cut at the correct angles and then laid up by hand as
shown in the following images (Figure 3.1). The images do not show each step of the
layup process, but rather a few images to give a representation of the process.

Figure 3.1: Hand layup process for pre-preg composites.
The curing process was slightly different for each set of materials as there were
different manufacturing procedures defined by the vendors. The carbon fiber composite
47
(CF) was assembled as a large panel and subsequently cured in an autoclave (pressure
vessel that uses compressed gas and heat) using the following cure process. In the image
above, it can be seen that the sample was encased in a vacuum bag and after the layup a
vacuum was continuously applied. This vacuum facilitates consolidation of the fibers,
and removal of excess resin and air from the matrix [8]. Additionally there was 586 kPa
(85 psi) of pressure applied to the sample inside the autoclave. When the pressure inside
the autoclave reached 138 kPa (20 psi), the vacuum in the vacuum bag was released. The
temperature followed the profile in Figure 3.2 once adequate pressure was reached in the
autoclave.

Figure 3.2: Temperature profile for autoclave curing of CF composite.
48
Once the temperature reached 60°C, the pressure was removed from the sample and the
part was subsequently removed from the autoclave to cool.
The individual boron fiber/carbon fiber composite (BF-CF) panels (152.4 mm x
152.4 mm) were cured in a press as shown in the image below (Figure 3.3).

Figure 3.3: Hydraulic press for curing composites.
With a vacuum still applied to the sample, the sample was placed in a press and a
pressure of 345 kPa (50 psi) was placed on the sample. Once the pressure from the press
plate was applied, the vacuum was removed from the sample and the tube connected to
the vacuum pump was placed in a vent hood to expel any volatiles during the cure
process. Each plate on the press had a separate heater and the heaters were manually
controlled to follow the temperature profile shown below (Figure 3.4).
49

Figure 3.4: Temperature profile for press curing of BF-CF composite.
Once the temperature reached 60°C, the pressure was removed from the sample and the
part was subsequently removed from the press to cool.
After curing was completed, all the samples were investigated with an ultrasonic
evaluation technique known as “c-scan” to determine whether the samples contained
defects, such as voids, from the manufacturing process. An example of two samples is
shown in Figure 3.5. The black dot in the corners of each sample was a sticker affixed to
the sample to use as a reference in determining the defects. The figure on the left showed
no evidence of defects and was accepted as adequate manufacturing quality. The image
on the right shows evidence of internal defects which could affect the material properties.
So, the sample on the right was rejected for use in the study.
50

Figure 3.5: C-scan comparison of two samples.
After samples with defects were removed from the sample set, the samples were
all cut to 139.7 mm x 139.7 mm to accommodate the size of the radiation beam available.
The excess material from the BF-CF samples was saved for differential scanning
calorimetry (DSC) analysis, which will be discussed later.

3.2. Radiation Exposure
All panels were irradiated at Indiana University Cyclotron Facility (IUCF) under
ambient conditions with 200 MeV protons to the following doses: 1,000 Gy (100 krad),
5,000 Gy (500 krad), 10,000 Gy (1,000 krad = 1 Mrad), and 180,000 Gy (18,000 krad).
The panels were divided into “fast” exposures (1.973 Gy/s or 197.3 rad/s) and “slow”
51
exposures (0.177 Gy/s or 17.7 rad/s), and panels were irradiated in a stacked
configuration (Figure 3.6) at their respective dose rates.

Figure 3.6: Stacked configuration at IUCF.

A maximum of 15 panels were stacked to save time at the facility, and given the high
proton energy of 200 MeV, the protons were able to penetrate through the entire
thickness (Appendix B). A laser was used to align the proton beam with the center of the
samples (bottom image of Figure 3.6). The area of the beam was limited to 2,827 mm
2

due to the limitations of the facility. After the radiation exposures were complete, the
panels remained at IUCF for approximately two weeks before safely transporting.
52
Chapter 4

In-situ Strain Analysis
The first experiment in this study investigated the strain changes of these two
materials during the radiation exposure. The materials were additionally subjected to bi-
axial tension to mimic the internal pressure stresses imparted to a habitat structural
material. The following chapter discusses the experiment, as well as the strain and
temperature response of these two materials during in-situ radiation exposure.

4.1. Test Setup
To simulate the pressure stresses on a habitat material, a test stand was designed
and manufactured specifically for this study to impart a bi-axial tensile stress. The design
for the grips were inspired by tensile test machines in which a jagged grip “bites” into the
material, and as the material begins to experience tension, the grip further tightens into
the jaws (Figure 4.1).

Figure 4.1: A close up view of the grips of the test stand.
53

Prior to tensioning, all samples for irradiation were affixed with a bi-axial strain
gauge in the center of the sample to collect strain data during the irradiation (Figure 4.2).

Figure 4.2: A sample tensioned in the test stand with a bi-axial strain gauge located in the
center of the sample.

The strain gauge was also used when tensioning the samples to place the correct amount
of strain on the sample that corresponded to the 41 MPa pressure stress calculated in
Section 2.2.3 (Appendix C). Samples were tensioned at Johnson Space Center and then
packed for transportation to Indiana University Cyclotron Facility (IUCF) for radiation
treatment.
At IUCF, the samples were additionally affixed with a thermocouple placed very
close to the strain gauge. The strain gauge leads and some of the thermocouples were
connected to a data acquisition system (National Instruments) as shown in Figure 4.3.
54

Figure 4.3: Data acquisition system used to collect data during radiation exposures.

Additional thermocouples were connected to hand-held readers placed in the radiation
beam room. Web cameras were used to acquire the readings from the hand-held readers.
The data that was collected during the radiation exposures included the change in strain
in both directions of the sample and temperature of the sample surface (Figure 4.4).

Figure 4.4: Example screen shot of data collection during radiation exposure. The top graph
shows the strain data and the bottom graph shows the temperature data.

55
During this experiment, samples were only exposed to 5,000 Gy (500 krads) of
200 MeV protons, but were subjected to both “fast” exposures (1.973 Gy/s or 197.3
rad/s) and “slow” exposures (0.177 Gy/s or 17.7 rad/s). There were a total of six
exposures completed for these experiments. All the samples for each exposure were
stacked in front of the radiation beam. The following table (Table 4.1) shows the
conditions for each exposure.
Table 4.1: Description of radiation exposures conducted [67].
Exposure # Dose Rate # of Samples Material
Exposure 1 Slow 5 BF-CF
Exposure 2 Slow 5 CF
Exposure 3 Fast 1 CF
Exposure 4 Fast 1 CF
Exposure 5 Slow 4 2 – BF-CF, 2 – CF
Exposure 6 Fast 4 2 – BF-CF, 2 – CF

4.2. Calculations of Material Strain Due to Thermal Changes
Considering both the temperature data with the strain data collected for each
sample revealed that the temperature trends for each exposure were similar to the strain
trends of the respective exposures. Thus, calculations were carried out to compare
thermal strains with the strain data collected during the radiation exposures.
First, the thermal strain change of the material sample was calculated, using
Equation 4.1. Here, α
lam
is the coefficient of thermal expansion for the material sample
and ∆T is the change in temperature.

T
lam lam Strain
Δ = Δ α
_
(4.1)
56
The thermal strain change of the test stand was calculated next (Equation 4.2). The outer
ring of the test stand was made of aluminum, and the equation for the test stand is shown
below. Here, α
Al
is the coefficient of thermal expansion for the aluminum frame and ∆T
is the change in temperature.

T
Al Al Strain
Δ = Δ α
_
(4.2)
The values used for the coefficients of thermal expansion are shown in the following
table (Table 4.2).
Table 4.2: Coefficient of thermal expansion for each material [67].
Material CTE
aluminum
2.358E-5
1 −
°C
CF-epoxy
1.925E-6
1 −
°C

BF-CF-epoxy
4.79E-6
1 −
°C

After these two calculations were completed, the change in strain between the two
equations was compared, and the greater value was assumed to be the driving force for
the strain measurements observed. These calculations were then compared with the
change in strain measured on the samples during radiation exposure.

57
4.3. Results

4.3.1. Fast Dose Rate Samples
The samples that were exposed to a fast dose rate were part of Exposures 3, 4, and
6. All of the samples in each of these exposures showed an overall decrease in strain
with respect to time, in both axes. Using Exposure 6 as an example, the decreasing strain
was apparent in both materials, as shown in Figure 4.5.

Figure 4.5: Strain data from Exposure 6, example of a fast dose rate in-situ strain response
[67].

58
In Figure 4.6, the corresponding temperature data collected for these materials is shown.

Figure 4.6: Measured temperature data from Exposure 6 [67].
Both of these graphs correspond to data collected during Exposure 6 in which two BF-
CF-epoxy samples (BF-CF-9 and BF-CF-10) were stacked with two CF-epoxy samples
(CF-1 and CF-2). In the temperature data, CF-1 did not get recorded due to a hardware
malfunction.

4.3.2. Slow Dose Rate Samples
In Exposures 1, 2, and 5, samples were exposed to a slow dose rate. These
samples showed strain increasing with time. Using Exposure 1 as an example, Figure 4.7
shows several BF-CF-epoxy samples with an increased strain with respect to time in both
axes, followed by a gradual leveling out of the data towards the latter part of the
exposure.
59

Figure 4.7: Strain data from Exposure 1, example of a slow dose rate in-situ strain response
[67].

In this plot, two dashed lines do not follow the rest of the data and correspond to data
recorded from BF-CF-6B and BF-CF-11B. These two axes did not record any strain data
for Exposure 1, most likely due to software and hardware failures from radiation
exposure to the data acquisition system. The corresponding temperature data for these
samples are shown in Figure 4.8.
60

Figure 4.8: Measured temperature data from Exposure 1 [67].

4.4. Discussion
In this experiment, samples exposed to a fast dose rate exhibited a decrease in
strain with respect to time, whereas those exposed to a slow dose rate exhibited an
increase in strain with respect to time. The temperature profiles of the respective samples
followed a similar trend of increasing slope in temperature versus time with a slow dose
rate and decreasing slope of temperature versus time with a fast dose rate. To verify that
the strain observed in the composite samples was not thermally induced, thermal
expansion calculations of the aluminum test stand frame and the composite samples were
performed and compared. If the measured strain were thermally induced, and if the
calculated thermal expansion of the aluminum frame was greater than the calculated
thermal expansion of the composite sample, we expected the measured strain to match
the calculated thermal expansion of the aluminum frame. Overall, this was not found to
be the case, as will be shown below.
61
Other potential sources of error include any possible radiation effects on the strain
gauges and/or bonding agent used to affix the gauges to the samples. To assess any
inaccuracies in the measurements induced by these possible effects, measurements could
be made on another material with a known radiation response using these same gauges
and bonding agents, and comparisons made between known and measured responses.
Of the nineteen samples studied, 68% of the samples’ measured strain did not
match the calculated thermal expansion of either the aluminum frame or the laminate.
An example of measured strain inconsistent with calculated thermal expansion is shown
in Figure 4.9 with BF-CF-3 during Exposure 1.

Figure 4.9: An example (BF-CF-3) of measured strain overlaid with calculated thermal
expansion of the aluminum frame and the BF-CF-epoxy material sample. This figure shows an
example of a sample’s measured strain that did not match the calculated thermal expansion of
either the aluminum frame or the laminate [67].

However, 21% of the measured strain of the sample corresponded with the
thermal expansion calculation of the aluminum frame and 11% of the samples’ measured
strain matched the calculated thermal expansion of the laminate. An example of the
62
measured strain matching the calculated thermal expansion of the aluminum frame is
shown in Figure 4.10 with CF-12 during Exposure 2.

Figure 4.10: An example (CF-12) of measured strain matching the calculated thermal
expansion of the aluminum frame [67].

Figure 4.11 presents the percent error between the measured strain and the calculated
thermal expansion of the aluminum frame for CF-12. As shown in the figure, initially
there is significant error between the two. However, the error quickly declines
throughout the remainder of the exposure.
63

Figure 4.11: The percent error between the measured strain and the calculated thermal
expansion of the aluminum frame for CF-12 [67].

In addition, Figure 4.12 displays an example of the measured strain of BF-CF-9 matching
the calculated thermal expansion of the laminate.

Figure 4.12: An example (BF-CF-9) of measured strain matching the calculated thermal
expansion of the laminate [67].

64
The corresponding percent error between the measured data and calculated expansion is
given in Figure 4.13. In the case of BF-CF-9, there is more error apparent between the
comparison of the measured strain data and the calculated thermal expansion of the
laminate.

Figure 4.13: The percent error between the measured strain and the calculated thermal
expansion of the laminate for BF-CF-9 [67].

Given that a majority of the samples were not equivalent to the calculated thermal
expansion for the aluminum frame or the laminate, and those samples that seemed to
agree still contained some error in the comparison, demonstrates that the strain changes in
the samples are a result of the radiation affecting the material.
Potential mechanisms responsible for the observed material behavior have been
considered through comparison with the literature. Chain scission and cross-linking
occur upon exposure of polymeric materials to radiation [4, 60, 62, 18]). Generally, these
two mechanisms occur simultaneously, with one predominating over the other. Cross-
65
linking typically results in increased strength, increased average molecular weight, and
embrittlement of the matrix, whereas chain scission results in decreased strength,
decreased molecular weight, and degradation of the matrix (i.e., ductility). Comparing
the outcome of these two mechanisms with the measured strain data during the radiation
exposure suggests that with a fast dose rate exposure, the materials are shrinking as a
result of enhanced matrix cross-linking. However, with a slow dose rate exposure, the
samples are stretching, suggesting a possible degradation of the matrix through scission.
A similar observation was reported in a study by Gillen and Clough [29], where
dose rate effects existed for four different materials. The study concluded that scission
effects became more important as the dose rate decreased, and that oxidative degradation
was a possible explanation for the scission dominance at decreased dose rates. Other
studies [28, 30, 71, 74] have also reported similar findings of scission-related effects due
to oxidation at low dose rates. Given that the current study was performed in air at two
extreme dose rates, the strain results suggest that oxidative degradation is the dominant
degradation mechanism in the slow dose rate samples, giving rise to the increased strain
behavior. Further characterization of these materials will be required to validate these
assertions.

4.5. Summary and Conclusions
Two composite materials were evaluated in a long-term radiation environment at
two different dose rates. These materials were subjected to a simulated pressure stress
while being exposed to radiation, and the strain of the materials was recorded. Exposure
66
to a fast dose rate (0.1478 krad/s) produced a decrease in strain as a result of matrix
shrinkage, and exposure to a slow dose rate (0.0139 krad/s) produced an increase in strain
as a result of matrix stretching. We also concluded that the strain changes observed in
the samples resulted from radiation exposure, and not from thermally induced strain
changes. Finally, investigating previous studies of radiation exposed polymeric materials
showed scission dominated effects as a result of oxidative degradation in polymers
exposed to a decreased radiation dose rate in air. The measured results were compared
with these studies to show potential mechanisms.
67
Chapter 5

Radiation and Dose Rate Effects
The next experiment investigated what material properties changed with only
radiation exposure and what might have caused those changes. These samples were not
subjected to bi-axial tension, only radiation. The following chapter will discuss the
characterization methods used, the results, and the analysis of the data.

5.1. Radiation Exposure
For this experiment, there were varying doses employed, as well as a “fast” (1.973
Gy/s or 197.3 rad/s) and “slow” (0.177 Gy/s or 17.7 rad/s) dose rate. The doses
investigated were 1,000 Gy (100 krad), 5,000 Gy (500 krad), 10,000 Gy (1,000 krad = 1
Mrad), and 180,000 Gy (18,000 krad), and the exposure details are shown in Table 5.1.
Table 5.1: Exposure details and characterization method for each panel. In the method
column, T represents tensile test, F represents flexure test, DSC represents differential
scanning calorimetry, and FTIR represents Fourier Transform Infrared Spectroscopy.
Panel # Speed Dose (Gy) Method Panel # Speed Dose (Gy) Method
BF-CF #56 N/A 0 T,F CF #31 N/A 0 T, F, DSC
BF-CF #58 N/A 0 T,F CF #32 N/A 0 T, F, DSC
BF-CF #59 N/A 0 T,F CF #33 N/A 0 T, F, DSC
BF-CF #60 N/A 0 T,F CF #34 N/A 0 T, F, DSC
BF-CF #62 N/A 0 T,F CF #35 N/A 0 T, F, DSC
BF-CF #63-1 N/A 0 T CF #61-1 N/A 0 T
BF-CF #63-2 N/A 0 T CF #61-2 N/A 0 T
BF-CF #63-3 N/A 0 T CF #61-3 N/A 0 T
BF-CF #9 Fast 5,000 FTIR CF #62 N/A 0 F
BF-CF #10 Fast 5,000 FTIR CF #1 Fast 5,000 FTIR

68
Table 5.1 (continued): Exposure details and characterization method for each panel.
Panel # Speed Dose (Gy) Method Panel # Speed Dose (Gy) Method
BF-CF #17 Fast 5,000 FTIR CF #2 Fast 5,000 FTIR
BF-CF #19 Fast 5,000 FTIR CF #22 Fast 5,000 FTIR
BF-CF #20 Fast 5,000 FTIR CF #24 Fast 5,000 FTIR
BF-CF #46 Fast
1,000 DSC CF #29 Fast 5,000 FTIR
BF-CF #61 Fast
1,000 DSC CF #46 Fast 5,000 T, F, DSC
BF-CF #36 Fast
5,000 T, F, DSC CF #47 Fast 5,000 T, F, DSC
BF-CF #38 Fast
5,000 T, F, DSC CF #48 Fast 5,000 T, F, DSC
BF-CF #39 Fast
5,000 T, F, DSC CF #49 Fast 5,000 T, F, DSC
BF-CF #41 Fast
5,000 T, F, DSC CF #50 Fast 5,000 T, F, DSC
BF-CF #42 Fast
5,000 T, F, DSC CF #3 Slow 5,000 FTIR
BF-CF #29 Fast
10,000 DSC CF #5 Slow 5,000 FTIR
BF-CF #47 Fast
10,000 DSC CF #7 Slow 5,000 FTIR
BF-CF #48 Fast
10,000 DSC CF #8 Slow 5,000 FTIR
BF-CF #55 Fast
10,000 DSC CF #9 Slow 5,000 FTIR
BF-CF #57 Fast
10,000 DSC CF #10 Slow 5,000 FTIR
BF-CF #64 Fast
10,000 DSC CF #12 Slow 5,000 FTIR
BF-CF #65 Fast
10,000 DSC CF #13 Slow 5,000 FTIR
BF-CF #40 Fast
180,000 DSC CF #20 Slow 5,000 FTIR
BF-CF #66 Fast
180,000 DSC CF #21 Slow 5,000 FTIR
BF-CF #67 Fast
180,000 DSC CF #25 Slow 5,000 FTIR
BF-CF #3 Slow 5,000 FTIR CF #26 Slow 5,000 FTIR
BF-CF #4 Slow 5,000 FTIR CF #27 Slow 5,000 FTIR
BF-CF #5 Slow 5,000 FTIR CF #28 Slow 5,000 FTIR
BF-CF #6 Slow 5,000 FTIR CF #36 Slow 5,000 T, F, DSC
BF-CF #7 Slow 5,000 FTIR CF #37 Slow 5,000 T, F, DSC
BF-CF #8 Slow 5,000 FTIR CF #38 Slow 5,000 T, F, DSC
BF-CF #11 Slow 5,000 FTIR CF #39 Slow 5,000 T, F, DSC
BF-CF #12 Slow 5,000 FTIR CF #40 Slow 5,000 T, F, DSC
BF-CF #13 Slow 5,000 FTIR
BF-CF #14 Slow 5,000 FTIR
BF-CF #15 Slow 5,000 FTIR
BF-CF #16 Slow 5,000 FTIR
BF-CF #43 Slow
5,000 T, F, DSC
BF-CF #50 Slow
5,000 T, F, DSC
BF-CF #51 Slow
5,000 T, F, DSC
BF-CF #52 Slow
5,000 T, F, DSC
BF-CF #53 Slow
5,000 T, F, DSC

69
5.2. Characterization Methods
In characterizing the samples, several methods were employed: tensile testing,
flexure testing, differential scanning calorimetry (DSC), Fourier transform infrared
spectroscopy (FTIR), and scanning electron microscopy (SEM). The following sections
will detail the methods and the procedures used in this study.

5.2.1. Tensile Testing
The tensile testing commenced approximately four months after the radiation
exposure and followed the procedures outlined in ASTM-D-3039, with the exception of
the coupon geometry. The parameters of the test were a speed of 1.27 mm/min to pull
the coupons and data was collected at a rate of 100 Hz. The radiation exposure was
limited to an area of 2,827 mm
2
due to the limitations of the beam at IUCF and this
exposure area limited the size of the tensile coupons to 12.7 mm (0.5”) wide and 139.7
mm (5.5”) long. The coupons were cut from the panel such that tensile forces would be
applied perpendicular to the zero-degree ply of the layup to enhance matrix-dominated
effects. An axial strain gauge was affixed to the center of the coupon to collect strain-
data during the test. Stress-strain curves were generated from the data and the following
properties were measured using the procedures in ASTM-D-3039: Modulus, ultimate
strength, fracture strength, strain-to-failure, fracture energy, and first fracture point.
Some of these quantities are shown in the representative graph below (Figure 5.1). The
fracture energy is the area under the stress-strain curve and represents the total energy
required to break the coupon. The first fracture point is the first point at which the curve
70
deviates from a straight line. Figure 5.1 is intended for illustration purposes to
demonstrate the information gathered from a stress-strain curve. Details of the stress-
strain curve will be presented in the Results and Discussion section.

Figure 5.1: A representative stress-strain curve resulting from a tensile test of an irradiated
coupon. From the data, the following quantities are gathered: modulus, ultimate strength,
fracture strength, strain-to-failure, and first fracture point.

5.2.2. Flexure Testing
The flexure testing (Figure 5.2) occurred three months after the radiation exposure
and followed the procedures outlined in ASTM-D-790-03. The parameters of the test
were a span separation of 25.4 mm, speed of 1.524 mm/min to flex the coupons, and data
71
collection at a rate of 20 Hz. The flexure coupons were cut in the same manner as the
tensile coupons, and were approximately 12.7 mm (0.5”) wide and 50.8 mm (2”) long.

Figure 5.2: Example of a sample undergoing flexure testing.

Stress-strain curves were generated from the data, and the following properties
were measured using the procedures in ASTM-D-790-03: modulus, ultimate strength,
fracture strength, strain-to-failure, and first fracture point. Some of these quantities are
shown in the representative graph below (Figure 5.3). Figure 5.3 is intended for
illustration purposes to demonstrate the information gathered from a stress-strain curve.
Details of the stress-strain curve will be presented in the Results and Discussion section.
72

Figure 5.3: A representative stress-strain curve resulting from a flexure test of a control
coupon. From the data, the following quantities are gathered: modulus, ultimate strength,
fracture strength, strain-to-failure, and first fracture point.

5.2.3. Differential Scanning Calorimetry
After manufacturing of the BF- CF panels was complete, the panels were cut
down to 139.7 mm x 139.7 mm and the excess material was saved to characterize the
material by differential scanning calorimetry (DSC) prior to radiation exposure. The CF
material was manufactured as a large panel and cut into smaller panels of 139.7 mm x
139.7 mm. Thus, non-irradiated CF panels (“controls”) were used to gather DSC samples
to compare with irradiated CF samples.
73
Two 30 mg samples were acquired from both the non-irradiated panels/excess
material and irradiated panels. All samples went through a heat-cool-heat cycle using the
following method.

DSC Method
1. Equilibrate at -90°C
2. Ramp at 50°C/min to 250°C
3. Mark end of cycle
4. Ramp at 25°C/min to -90°C
5. Mark end of cycle
6. Ramp at 50°C/min to 250°C
7. Mark end of cycle

The results for each heat cycle were plotted in the figure below (Figure 5.4). The
first heat for the samples characterized before radiation exposure were used to verify that
the samples were fully cured by confirming the absence of cure peaks after the glass
transition temperature (T
g
). The second heat was used to calculate the T
g
since the first
heat could contain thermal history from the cure.
74

Figure 5.4: Example of DSC curve for a sample before radiation exposure (BF-CF #22).

5.2.4. Fourier Transform Infrared Spectroscopy (FTIR)
Fourier transform infrared spectroscopy (FTIR) scans via the Attenuated Total
Reflectance (ATR) method were acquired from the center of the sample before radiation
and immediately after returning from radiation. The ATR method allows investigations
of the surface within a range of 0.5-3 µm. Prior to characterizing any panels, a
background spectrum was obtained and subtracted from subsequent sample spectra. This
process enabled the removal of any background noise, such as water, from the signal so
that weaker signals can be more easily observed. Pressure was applied during data
collection to ensure that there was adequate contact between the sample and the ATR
crystal. Subsequently, approximately 128 scans were acquired, averaged, and the
baseline was corrected. The scans from before and after radiation were averaged
75
according to their exposure group, and the averaged scans were subtracted (post-pre) in
order to better investigate which peaks had changed with irradiation.

5.2.5. Scanning Electron Microscopy (SEM)
Post-tensile testing scanning electron micrographs were obtained of tensile
coupons. The tensile coupons were not treated prior to observation in the SEM, but they
were attached to metallic plates with conductive carbon tape to limit charging of the
epoxy (Figure 5.5).

Figure 5.5: Preparation of CF coupon for SEM investigation.

5.3. Results

5.3.1. Tensile Results
Load and strain data were collected for each material on eighteen tensile coupons,
including eight control coupons, five coupons exposed to proton radiation at a fast dose
rate and total dose of 5,000 Gy, and five coupons exposed to proton radiation at a slow
76
dose rate and a total dose of 5,000 Gy. With the exception of the control coupons, each
coupon was cut from a different panel to account for panel variation. Of the eight control
coupons, five were prepared from separate panels and three were cut from the same panel
(BF-CF #63 and CF #61 as documented in Table 5.1). These three additional coupons in
the control data were originally intended to be test coupons to ensure quality of the test
setup, but also served as control specimens. The resulting stress-strain curves of
representative coupons (for clarity) of each exposure group and material are shown in
Figure 5.6 and Figure 5.7.

Figure 5.6: Stress-strain curve for one representative BF-CF coupon of the control group, fast
group, and slow group.

77

Figure 5.7: Stress-strain curve for one representative CF coupon of the control group, fast
group, and slow group.

These curves were created by plotting the strain collected from the strain gauge
on the coupon versus the calculated stress (ASTM-D-3039) resulting from the load
during the tensile test. In the figures, the graphs seem as if they do not start at zero stress
and zero strain because of slack in the test setup. A slack correction algorithm was
employed to remove the slack from the stress-strain curves, and produced more accurate
calculations of tensile properties. In general, the representative stress-strain results for
BF-CF show a trend of lower overall strengths and increased strain-to-failure. The
representative BF-CF coupons show high agreement and consistency between the control
and irradiated samples.
78
Subsequently, the modulus (Figure 5.8), ultimate strength (Figure 5.8), fracture
strength (Figure 5.8), strain-to-failure (Figure 5.9), fracture energy (Figure 5.9), and first
fracture point (Figure 5.10) were calculated for the BF-CF material. In Figure 5.8 and
Figure 5.9 the individual data points are plotted. To show clarity and significance in the
first fracture point data (Figure5.10), the values from each coupon were averaged for
each group, and the standard deviation of the group was calculated. The results are
shown in the following graphs.

Figure 5.8: Calculated modulus (■ – left side), ultimate strength (Δ – right side), and fracture
strength (○ – right side) for each BF-CF coupon investigated and plotted against the dose rate
exposure.

79

Figure 5.9: Calculated fracture energy (■ – left side) and strain-to-failure (◊ - right side) for
each BF-CF coupon investigated and plotted against the dose rate exposure.

Figure 5.10: Averaged BF-CF first fracture point data for the dose rates investigated.

80
In Figure 5.8, the modulus, ultimate strength, and fracture strength values
generally decreased after irradiation compared to the control, but the scatter among the
data precluded any definitive conclusions. Potential sources of the error are edge effects
resulting from non-standard coupon sizes. However, values for the first fracture point
(Figure 5.10) for the irradiated samples did show significant decrease when compared
with the control. These results show that the material strength decreased with increasing
radiation exposure. Furthermore, in Figure 5.9 (strain-to-failure and fracture energy), the
irradiated samples have increased values when compared with the control, albeit with
scatter amongst the data. These data show that after radiation exposure, the strain-to-
failure increased and the total fracture energy increased, indicating the material became
tougher. In Figures 5.8-5.10, there is negligible difference between samples irradiated at
a slow dose rate and those irradiated at a fast dose rate. Therefore, no inferences can be
made regarding dose effects on tensile properties of the BF-CF material.
The modulus (Figure 5.11), ultimate strength (Figure xx.4), fracture strength
(Figure 5.11), first fracture point (Figure 5.11), strain-to-failure (Figure 5.12), and
fracture energy (Figure 5.12) were also calculated for the CF material. The results are
shown in the following graphs where individual data points are plotted.
81

Figure 5.11: Calculated modulus (■ – left side), ultimate strength (Δ – right side), fracture
strength (○ – right side), and first fracture point (⟡ - right side) for each CF coupon
investigated and plotted against the dose rate exposure.

Figure 5.12: Calculated fracture energy (■ – left side) and strain-to-failure (◊ - right side) for
each CF coupon investigated and plotted against the dose rate exposure.

82
In Figure 5.11 and Figure 5.12, there are few to no differences between the control and
the irradiated samples for the CF material in modulus, fracture energy, and strain-to-
failure. In Figure 5.11, the trends show a slight increase in ultimate strength, fracture
strength, and first fracture point for the irradiated samples, but the scatter in the data
prohibits any conclusions. There are no apparent differences between the fast and slow
dose rate samples for the CF material.

5.3.2. Flexure Results
For each material, there were four coupons tested from each control panel, and
one coupon from each irradiated panel. This provided a total of thirty coupons for each
material. The irradiated samples were exposed to 5,000 Gy and a fast and slow dose rate
was investigated for dose rate effects. The resulting stress-strain curves of representative
coupons (for clarity) of each exposure group and material are shown in Figure 5.13 and
Figure 5.14.
83

Figure 5.13: Flexural stress-strain curve for representative BF-CF coupons of the control
group, fast group, and slow group.

Figure 5.14: Flexural stress-strain curve for representative CF coupons of the control group,
fast group, and slow group.
84

These curves were created by plotting the calculated strain from the machine’s crosshead
versus the calculated stress (ASTM-D-3039) resulting from the load during the flexure
test. A slack correction algorithm was again used to remove the slack from the stress-
strain curves. In general, the representative flexural stress-strain results for BF-CF show
a trend of lower overall strengths. The CF flexural stress-strain results again show
excellent agreement and consistency between the control and irradiated samples.
Subsequently, the modulus (Figure 5.15), ultimate strength (Figure 5.15), fracture
strength (Figure 5.15), first fracture point (Figure 5.15), and strain-to-failure (Figure
5.16) were calculated for the BF-CF material. In the following figures, the individual
data points are plotted.

Figure 5.15: Calculated flexural modulus (■ – left side), first fracture point (◊ - right side),
ultimate strength (Δ – right side), and fracture strength (○ – right side) for each BF-CF coupon
investigated and plotted against the dose rate exposure.
85

Figure 5.16: Calculated flexural strain-to-failure (◊ - left side) for each BF-CF coupon
investigated and plotted against the dose rate exposure.

In Figures 5.15 and 5.16 the results show minimal to no changes with radiation exposure.
In Figure 5.15, the data shows a slight increase in first fracture point, ultimate strength,
and fracture strength with a fast radiation dose rate when compared with a slow radiation
dose rate. However, given the scatter in the data, no conclusions can be drawn from
these results regarding dose rate effects.
The modulus (Figure 5.17), ultimate strength (Figure 5.17), fracture strength
(Figure 5.17), first fracture point (Figure 5.17), and strain-to-failure (Figure 5.18) were
also calculated for the CF material. The results are shown in the following graphs where
individual data points are plotted.
86

Figure 5.17: Calculated flexural modulus (■ – left side), first fracture point (⟡ - right side),
ultimate strength (Δ – right side), and fracture strength (○ – right side) for each CF coupon
investigated and plotted against the dose rate exposure.

Figure 5.18: Calculated flexural strain-to-failure (◊ - left side) for each CF coupon investigated
and plotted against the dose rate exposure.
87

In Figure 5.17 the data show agreement between control and irradiated samples.
However, in Figure 5.18 the data shows an increase in strain-to-failure with both a fast
and a slow dose rate when compared with the control. In addition, the strain-to-failure is
increased in the slow dose rate when compared with the fast dose rate. These results
indicate that the material is toughening with radiation exposure and possibly incurring
enhanced toughening at a slower dose rate, although these results are again inconclusive
due to the scatter in the data.

5.3.3. DSC Results
The method of gathering DSC data from the BF-CF and CF materials was slightly
different as a result of the differences in manufacturing of the samples. The BF-CF
samples were individually cured and excess material from the curing process was
removed. This excess was used to gather the glass transition temperature (T
g
) for each
panel prior to irradiation and was compared with the T
g
of the irradiated panel. The
exposure groups for the BF-CF panels were 1,000 Gy, 5,000 Gy, 10,000 Gy, and 180,000
Gy. The pre-irradiation and post-irradiation values for each of these exposure groups
were averaged, and the variation was plotted against the dose (Figure 5.19).
88

Figure 5.19: Average percent change in glass transition temperature for each exposure group
evaluated of the BF-CF material.

At a very low dose, there was a small increase in the T
g
, but as the dose increased,
the T
g
decreased in a logarithmic fashion. Fitting a curve to the data, an empirical
relation was developed (Equation 5.1) which closely fit the data. In this equation, Y is the
percent change in the glass transition temperature and x is the total dose in Gy.
Y = -0.025 * ln(x) + 0.1894 (5.1)
Decreasing T
g
values with radiation exposure indicates that chain scission increases with
increased radiation dose. Similar effects were reported in [21, 84], and one report
concluded that scission predominates when the material is fully cross-linked [21]. The
lack of any cure peaks in the DSC data indicates that the samples were fully cross-linked
89
before radiation. Thus, chain scission in the epoxy is expected with increased radiation
exposure [21].
Furthermore, previous work has shown an empirical correlation (Equation 5.2)
between T
g
and the degree of cross-linking [70, 54, 38, 59].
M
c
= (3.9x10
4
) / (T
g
-T
g0
) (5.2)
In this equation M
c
is the number average molecular weight between cross-links, T
g
is the
glass transition temperature, and T
g0
is the glass transition temperature of the pre-cured
epoxy. As the glass transition temperature decreases, the molecular weight between
cross-links increases, indicating a degradation of the epoxy network structure.
For the panels irradiated at a total dose of 5,000 Gy (500 krad), the T
g
for the fast
dose rate was compared to the slow dose rate. The results are shown in Table 5.2.
Table 5.2: A comparison of the percent change in glass transition temperature for the BF-CF
samples irradiated to 5,000 Gy at fast and slow dose rates.
Average (°C) % change
Fast (pre) 126.03
-1.95%
Fast (post) 123.574
Slow (pre) 128.168
-2.43%
Slow (post) 125.056

As shown in the table, the slow dose rate caused a slightly larger depression of the T
g

than is observed in the fast dose rate. Hence, BF-CF samples exposed at a slow dose rate
experienced slightly more chain scission than samples receiving radiation at a fast dose
rate.
90
The CF material was cured as one large panel and samples were subsequently cut
from the panel. Thus, there was no excess material to use for pre-radiation analysis of
each sample. Rather, control samples were created, the pre-radiation T
g
was calculated
from the control samples, and subsequently compared with the T
g
of irradiated samples.
The CF irradiated samples were only irradiated to a 5,000 Gy dose. The T
g
values as a
function of dose rate are plotted in Figure 5.20.

Figure 5.20: Tg of the CF material as a function of dose rate.

The data do not show a significant difference between the control samples and the
irradiated samples, nor is there a difference between those samples irradiated at a fast
dose rate versus a slow dose rate.

91
5.3.4. FTIR Results
Ten panels for each material were investigated with FTIR prior to radiation
exposure and immediately after returning from irradiation. Of the two irradiation groups,
there were seven panels investigated for the slow irradiation and three panels investigated
for the fast irradiation. The scans from before and after radiation were averaged
according to their exposure group, and the averaged scans were subtracted (post-pre) to
highlight peaks that had changed with irradiation.
The averaged and subtracted scans for both the fast and slow exposure groups of
the BF-CF material are shown in Figure 5.21, and several peak regions are identified.

Figure 5.21: FTIR averaged spectra of the center locations of the BF-CF samples for fast (___)
and slow (---) exposures and the averaged pre-irradiation scan is subtracted from the
averaged post-irradiation scan.

92
In general, with radiation exposure, many of the peaks increased in absorbance when
compared with pre-radiation values. In addition, the panels that underwent fast exposures
exhibited greater absorbance values for the peaks than the slow exposure, although the
differences were relatively small (~13% for most peaks with carbonyl at 27% and
hydroxyl at 34%). Given that the carbonyl and hydroxyl peaks were elevated for the fast
dose rate as compared to the slow dose rate, indicates that more oxidation is occurring on
the surface of the fast dose rate samples.
The averaged and subtracted scans for both the fast and slow exposure groups of
the CF material are shown in Figure 5.22, and several peak regions are identified.

Figure 5.22: FTIR averaged spectra of the center locations of the CF samples for fast (___) and
slow (---) exposures and the averaged pre-irradiation scan is subtracted from the averaged
post-irradiation scan.

93
The results of the CF material show a marked difference between the fast and the
slow dose rate spectra. In the fast dose rate spectra there is very little difference between
the pre and post scans, as evidenced by the small peaks in the 600-1800 cm
-1
range,
indicating that radiation exposure at a fast dose rate imparted minimal chemical changes
in the CF material. However, with the slow dose rate there are significantly larger peaks
across the spectrum when compared with the fast dose rate, indicating that a slow dose
rate causes greater chemical changes in the CF material. The results also showed no
evidence of a carbonyl peak in either the fast or the slow dose rate samples, indicating a
lack of oxidation with radiation in this material.

5.3.5. SEM Results
Tensile coupons were investigated after fracture via scanning electron
microscopy. Control coupons of the BF-CF material without radiation exposure (Figure
5.23) were compared with BF-CF coupons exposed to 5,000 Gy (500 krads) of 200 MeV
protons (Figure 5.24). In these micrographs, only the boron fibers are shown, as the
carbon fibers were deeper in the coupon and difficult to image without pulling apart the
layers of the coupon and essentially destroying the tensile coupon.
94

Figure 5.23: SEM micrographs of BF-CF tensile coupons post-fracture that were not exposed to
radiation.

Figure 5.24: SEM micrographs of BF-CF tensile coupons post-fracture exposed to 5,000 Gy.
The bottom right micrograph shows the interface between a boron fiber and the epoxy.
95

In Figure 5.23, boron fibers in various control samples (i.e. no radiation exposure) are
shown. The fibers are covered with attached matrix indicating strong adhesion between
the boron fibers and the matrix. However, in Figure 5.24, after 5,000 Gy irradiation, the
boron fibers are relatively free of resin. The change in residual adherence of resin to the
fiber surface (bottom right image of Figure 5.24) indicates a weakened fiber-matrix
interface bond as a result of the irradiation exposure.
Control coupons of the CF material without radiation exposure (Figure 5.25) were
also compared with CF coupons exposed to 5,000 Gy (500 krads) of 200 MeV protons
(Figure 5.26). The micrographs below focus on the carbon fibers only.

Figure 5.25: SEM micrographs of CF tensile coupons post-fracture that were not exposed to
radiation.

96

Figure 5.26: SEM micrographs of CF tensile coupons post-fracture exposed to 5,000 Gy.

In Figures 5.25, carbon fibers in various control samples are shown with attached
matrix indicating strong adhesion between the carbon fibers and the matrix. In Figure
5.26, no change is discernible in the interface between the carbon fibers and the matrix.
Thus, the CF material did not experience an effect in the interface between the fiber and
the matrix as a result of radiation exposure.

5.4. Discussion
Previous studies of radiation effects on polymeric materials have generally shown
one of two effects on the mechanical properties of these types of materials. In some
cases, the material shows increased strength and reduced ductility with radiation
exposure, and this is generally attributed to increased cross-linking [45, 18, 63, 44]. In
97
other cases, decreased strength is observed and this is attributed to chain scission in the
polymer backbone [63, 64, 50, 20, 48]. These two mechanisms are considered below in
the interpretation of the current results.

5.4.1. Composite Durability
The investigation of material properties through tensile, flexure, and DSC testing
of the CF material yielded few changes when the material was exposed to radiation. The
results showed some increases with radiation exposure in ultimate strength, fracture
strength, and first fracture point of the tensile data, as well as a slight increase in strain-
to-failure in the flexure data. These differences signify potential cross-linking occurring
in the matrix. However, there was scatter in the data and the changes were not significant
enough to provide definitive conclusions regarding degradation in the material durability.
Conversely, the BF-CF material experienced a decrease in material strength,
increase in matrix ductility, and decrease in T
g
, which can be attributed to scission effects
in the matrix. In other material studies [21, 84, 63, 44, 88] similar degradation was
reported and attributed to chain scission as the primary degradation mechanism. The
epoxy chemistry typically determines whether the matrix will undergo scission or cross-
linking (or both). For instance, in two studies [63, 50] several different types of
chemistry are discussed with regards to radiation effects. If the epoxy contains tertiary
carbons, these carbons will undergo chain scission most readily, followed by secondary
bonded and primary bonded carbons. This hierarchy is due to the dissociation energy of
these bonds. The epoxy systems used in the present study are typical aerospace epoxy
98
formulations with toughening agents. One of the most common aerospace epoxy bases is
diglycidyl ether of bisphenol-A (DGEBA) [44, 50]. Comparing the FTIR spectra of the
control BF-CF and CF composite surface (Figure 5.27 and Figure 5.28) with the FTIR
spectra of DGEBA (Figure 5.29 [47]) shows several similarities, such as the aliphatic
CH-stretch peaks at ~3000 cm
-1
, the carbon double bond from the aromatic ring at ~1500
cm
-1
, the CH-bend in-plane at ~1250 cm
-1
, and the CH-bend out-of-plane at ~850 cm
-1
.
Thus, we can assume that both the BF-CF and CF epoxy is based on a DGEBA
formulation with toughening agents.

Figure 5.27: Averaged pre-radiation FTIR spectrum of all ten BF-CF (___) and CF (---) samples
scanned with FTIR prior to being exposed to radiation.
99

Figure 5.28: Averaged pre-radiation FTIR spectrum of all ten BF-CF (___) and CF (---) samples
scanned with FTIR prior to being exposed to radiation in the 1800-600 range. The black
arrows indicate locations in the BF-CF spectrum and the blue arrows indicate locations in the
CF spectrum.

Figure 5.29: IR spectrum of DGEBA [47].

DGEBA contains tertiary carbons, and, thus, is susceptible to scission during
irradiation. However, epoxies also contain aromatic rings that increase resistance to
radiation effects by absorbing and dissipating excitation energy [87]. Woods and Pikaev
100
[87] report that the aromatic groups dissipate much of the energy that is absorbed during
these tertiary carbon bond cleavages because the product yields in aromatic structures of
this nature tend to be lower than one would expect. The efficiency with which the
aromatic group can dissipate the energy decreases when the site of energy absorption is
farther from the aromatic group.
The FTIR spectra of BF-CF (Figure 5.21) includes peaks indicative of aromatic
features in the epoxy at 3050 cm
-1
(CH-stretch in the aromatic ring), 1300-1550 cm
-1

(carbon double bond stretch in the aromatic ring), and 700-900 cm
-1
(CH-bend out-of-
plane) [80]. With radiation exposure, these peaks increased, indicating an increase in
aromaticity. However, aromatic structures are less durable when the epoxy is irradiated
in the presence of oxygen (often causing oxidation), and many epoxies tend to undergo
chain scission under such conditions. Indication of oxidation in an epoxy [48, 65]
appears as an increase in the carbonyl region (1750 cm
-1
) and in the hydroxyl region
(3300 cm
-1
). In the BF-CF composite, there is an increase in both of these regions,
indicating oxidative degradation. However, studies have shown that the oxidation
occurring in epoxies is generally limited to near-surface regions (<20 µm) due to low
oxygen diffusion through the material [48, 65]. Thus, any oxidation that occurred in the
BF-CF composite is likely to be limited to the near-surface regions and have negligible
effects on the mechanical properties. Furthermore, when epoxies undergo chain scission
because of radiation, the molecular mobility increases, further enhancing radical
recombination. Through FTIR scans, scission can be indicated through increases in
aliphatic groups, such as the aliphatic CH-stretch region (2800-3000 cm
-1
). In Figure
5.21, increases in this peak are visible, suggesting an increase in aliphatic groups after
101
radiation as well. Given that the FTIR data shows increases in all of these regions
(aromatic, oxidation, and aliphatic), we conclude that competing processes occur in this
material, making it difficult to ascertain which process is dominant. While the aromatic
groups increased after radiation, there are also increases in areas that indicate oxidation
and enhanced scission. Coupling the FTIR results with the trends shown in the tensile
and DSC data, the results suggest that scission is the dominant process in the BF-CF
material given that oxidation would affect only near-surface regions and would not affect
the tensile properties or DSC results.
The CF material did not experience the same mechanical property changes as the
BF-CF material, even though the BF-CF and CF materials appear to share the same base
DGEBA epoxy formulation. In the CF FTIR data (Figure 5.22), the aromaticity (carbon
double bond stretch in the aromatic ring: 1500-1600 cm
-1
and CH-bend out-of-plane:
600-900 cm
-1
) slightly increased with a fast dose rate and increased further with a slow
dose rate. This would suggest enhanced cross-linking occurring in the CF material.
While inconclusive, this increased cross-linking could be the reason for the increased
ultimate strength, fracture strength, and first fracture point in the tensile data, albeit with
scatter. Furthermore, there was no indication of oxidation occurring in the fast dose rate
samples, as evidenced by the lack of carbonyl and hydroxyl peaks. However, in the slow
dose rate samples, there was evidence of a hydroxyl peak (3000-3600 cm
-1
) but no
evidence of a carbonyl peak. Moreover, the slow dose rate showed evidence of an
increasing aliphatic CH-stretch peak (~2800 cm
-1
), indicating possible scission occurring.
These results may have contributed to the slight increase in the strain-to-failure of the
flexure data. Given these results, we conclude that in the fast dose rate samples there is
102
evidence of slight cross-linking occurring, but in the slow dose rate samples, there is
competition of cross-linking and scission. However, in both of these cases, the chemical
changes are not significant enough to show any dominance within the mechanical
properties of the CF material. The lack of dominant effects may be a result of the
hardener and additional additives in the matrix acting as a protectant against the radiation,
or in the slow dose rate case, the competition between cross-linking and scission could be
cancelling out effects.

5.4.2. Toughening
The role of the toughening agent on degradation of composites is worthy of
consideration. Toughening agents, often rubber particles or thermoplastic inclusions, are
added to epoxies to increase fracture toughness and damage tolerance [75, 7, 53]. In
aerospace composites, thermoplastic additives such as polyethersulfone (PES), a high-
temperature thermoplastic, are commonly used [32, 52]. In these studies, the amount of
toughening is generally in the 10-20 wt% range. However, during radiation exposure
PES reportedly undergoes chain scission [72, 73], decreasing tensile strength and T
g
.
Both the BF-CF and CF composites are a toughened epoxy, and chain scission of the PES
toughening agent may contribute to the observed degradation in the material properties of
the BF-CF material.
In comparing the FTIR spectra of PES (Figure 1(b) in [10]) with the BF-CF and
CF materials (Figure 5.30 below), there are several similarities evident: carbon double
bond stretch in the aromatic ring structure at ~1500 cm
-1
and ~1600 cm
-1
; the CH-bend
103
in-plane at ~1300 cm
-1
, ~1230 cm
-1
, ~1140 cm
-1
(only in CF), and ~1100 cm
-1
; and the
CH-bend out-of-plane at ~810 cm
-1
and ~720 cm
-1
.

Figure 5.30: Overlaid averaged FTIR spectra of the BF-CF and CF materials pre-radiation and
peak comparison with PES from Figure 1(b) in [10]. The black arrows indicate locations in the
BF-CF spectrum and the blue arrows indicate locations in the CF spectrum.

However, in comparing Figure 5.30 above with Figure 5.28, it is also evident that there
are similarities in the peaks between DGEBA and PES. Thus, it is difficult to ascertain a
signature indicative of PES, and consequently to what degree the toughening agent
changes as a result of the radiation exposures.

104
5.4.3. Dose Rate Effects
One of the compromises made in this study was to increase the speed of the dose
rate to perform the experiment within budget and time limitations. To achieve the dose
rate experienced during an SPE (approximately 4.63E-4 Gy/s), would require ~125 days
to reach the 5,000 Gy dose being investigated in this study. Therefore, we selected a
faster dose rate of 1.973 Gy/s (fast dose rate) as well as a slower dose rate (0.177 Gy/s) to
investigate possible dose rate effects.
The tensile and flexure data showed no significant effects of dose rate for both
materials, although the DSC data revealed a slightly larger decrease in T
g
for the slow
dose rate of the BF-CF material. Dose rate effects, especially in the presence of oxygen,
are typically manifested in results as additional scission occurring as a result of oxidation
[16, 85, 65].
However, in the FTIR data of both the BF-CF and CF materials, we see
inconsistencies in the slow dose rate difference spectrum when compared with the fast
dose rate difference spectrum. In the BF-CF material the slow dose rate difference
spectrum exhibits slightly reduced peaks when compared with the fast dose rate
difference spectrum. The FTIR data indicate that less oxidative degradation, scission
effects, and aromaticity occurred at the slow dose rate compared to the fast dose rate.
Conversely, in the CF material, the slow dose rate difference spectrum exhibited
increases in all peaks when compared with the fast dose rate spectrum. This data indicate
that there are competitions occurring between scission and cross-linking, potentially
cancelling any dominant observable effects in the mechanical properties. Thus, the
105
overall observed differences in mechanical properties of both materials are minor, and
there is insufficient evidence to support dose rate effects occurring in this study.

5.4.4. Aging
Tensile tests were performed four months after radiation exposure, and the
indications of composite degradation at these low doses in the BF-CF material might
result from enhanced aging initiated by the radiation exposure. Prior studies showed that
radiation exposure of a DGEBA/TETA epoxy system incurs chain scission with radiation
exposure if the system is fully cross-linked before radiation exposure [21, 84]. Chain
scission enhances the molecular mobility, and three months after radiation exposure, the
aging kinetics of the epoxy system is accelerated. The results shown in our study do not
provide enough information to ascertain whether accelerated aging is in fact occurring in
these materials.

5.4.5. Fiber Debonding
The SEM micrographs of the CF tensile coupons after radiation showed no
change in the fiber-matrix interface regarding the bonding. However, the micrographs of
BF-CF tensile coupons after radiation exposure did provide evidence of interface
degradation. Previous studies [40, 26] have shown the effects of irradiated composites
composed of fibers with borated material, including fiber-matrix interface debonding, and
the debonding was attributed to increased radiation in the epoxy. The effect was
106
compounded by enhanced secondary radiation at interfaces induced by the primary
radiation interaction with the boron.
To investigate the effects of doses the BF-CF/semi-toughened epoxy material
might experience at the fiber-matrix interface, a simulation was conducted using the
HZETRN software [76, 77]. In this simulation, the composite was modeled after an SEM
image (Figure 5.31), and the environment used was the October 1989 SPE.

Figure 5.31: Model of BF-CF for HZETRN dose calculations at the boron fiber and epoxy
interface.

The dose at the interface in the model was compared with the dose at the same location in
unreinforced epoxy. The results (Table 5.3) show that for the October 1989 SPE
environment, the boron fiber increases the radiation dose by about 23% compared to a
neat epoxy. The same simulation was run for the CF material and the results show that
only a 3% increase in dose occurred at the interface of the carbon fiber with epoxy. The
higher overall dose of the CF material as compared to the BF-CF material is a result of
107
the CF simulation containing smaller thicknesses than the BF-CF simulation. The
magnified dose at the BF-CF interface adds to the enhanced degradation observed in this
study.
Table 5.3: Calculated dose at the interface between the fiber and epoxy compared with a neat
epoxy at the same location for each material.
Material Composite (cGy) Epoxy Only (cGy) % increase
BF-CF 20,583.59 16,760.4 22.81%
CF 79,006.01 76,641.27 3.09%

5.5. Conclusions
In the present study, we focused on doses a habitat material might experience
during a lunar mission, (approximately 5,000 Gy). The selected dose is well below that
of other studies [45, 18, 70, 63, 86], which typically involved doses on the order of
10
6
Gy (10
8
rads) before significant effects were observed. Given the low doses
evaluated in this study, it is not surprising that the effects are minimal and are not as large
as reported in other studies. However, most previous studies were performed with low
energy gamma radiation or electron radiation, and few if any studies involved proton
radiation at high energies. In addition, previous studies generally involved pure epoxy
composites, whereas the present study focused on toughened epoxy composites. Both
factors, high-energy proton radiation and toughened epoxy composites, distinguish the
present study from earlier work and could cause the effects reported here.
We concluded that radiation exposure of a BF-CF/semi-toughened epoxy system
degrades due to chain scission within the polymer network and subsequent debonding of
108
the resin from the boron fibers when exposed to 200 MeV protons in an oxygenated
environment. The factors that lead to and enhance the scission effects are the chemical
composition of the epoxy, the presence of toughening agents that are susceptible to
radiation damage and potentially enhance aging, and expected increased radiation dose at
the interface between the boron fiber and the epoxy. There was some evidence of
oxidative degradation at the sample surface, but due to low oxygen diffusion through the
sample, the oxidation is not expected to affect bulk properties such as T
g
or tensile
strength. While most of the tensile results were inconclusive, the first fracture point in
the tensile data is significant evidence of the scission effects and may be used as a
sensitive initial indicator of composite degradation due to this type of radiation damage.
Furthermore, the DSC results showed evidence of decreased T
g
with increasing radiation
exposure, providing further evidence of scission effects in the composite. There was no
evidence of dose rate effects.
Conversely, the CF material showed no evidence of degradation of the epoxy.
The mechanical properties showed some indication of enhanced cross-linking in the
matrix, but the effects were not significant enough to conclude cross-linking definitively.
The FTIR results also allude to enhanced cross-linking as a result of increased
aromaticity with a fast dose rate, but with a slow dose rate, evidence of competition
between scission and cross-linking is apparent. This competition between the two
mechanisms could negate any dominant effect visible in the mechanical properties.
Furthermore, there was no evidence of oxidative degradation in the CF material. Finally,
there was no evidence of changes in the interface properties of the matrix with the carbon
109
fibers and the subsequent dose calculations at the interface showed only a slight increase
of 3%, which did not enhance any effects.
110
Chapter 6

Synergistic Effects of Tension and Radiation
This experiment investigated whether material properties changed when a
composite was subjected bi-axial tension during irradiation. The results were compared
with the results from Chapter 5 to highlight whether there was enhanced degradation
from synergistic effects. The discussion examines the analysis of these comparisons.

6.1. Test Setup and Methods
Prior to irradiation, the materials were placed into the test stand detailed in
Section 4.1 to simulate the internal pressure stresses imparted to a habitat material.
Subsequently, the samples were exposed to a 5,000 Gy dose of 200 MeV protons at a
“fast” (1.973 Gy/s) and “slow” (0.177 Gy/s) dose rate. In addition, there were five
samples from each material that were subjected to only tension (no radiation) for ~30
days to replicate the length of time the irradiated samples were held under tension. These
“tension only” samples were compared with control samples (no tension and no
radiation) to investigate how the tension may affect the samples without the radiation.
The exposure details of the samples subjected to both radiation and tension are shown in
Table 6.1. Note that BF-CF#21 and BF-CF#22 are only used for DSC analysis because
the panels had some porosity as a result of manufacturing that negatively impacted the
tensile and flexure properties.

111
Table 6.1: Exposure details and characterization method for each panel subjected to tension.
In the method column, T represents tensile test, F represents flexure test, DSC represents
differential scanning calorimetry, and FTIR represents Fourier Transform Infrared
Spectroscopy.
Panel # Speed Dose (Gy) Method Panel # Speed Dose (Gy) Method
BF-CF#33 N/A 0 T, F, DSC CF#56 N/A 0 T, F, DSC
BF-CF#34 N/A 0 T, F, DSC CF#57 N/A 0 T, F, DSC
BF-CF#35 N/A 0 T, F, DSC CF#58 N/A 0 T, F, DSC
BF-CF#49 N/A 0 T, F, DSC CF#59 N/A 0 T, F, DSC
BF-CF#54 N/A 0 T, F, DSC CF#60 N/A 0 T, F, DSC
BF-CF#9 Fast 5,000 FTIR CF#1 Fast 5,000 FTIR
BF-CF#10 Fast 5,000 FTIR CF#2 Fast 5,000 FTIR
BF-CF#21 Fast 5,000 DSC CF#51 Fast 5,000 T, F, DSC
BF-CF#22 Fast 5,000 DSC CF#52 Fast 5,000 T, F, DSC
BF-CF#23 Fast 5,000 T, F, DSC CF#53 Fast 5,000 T, F, DSC
BF-CF#24 Fast 5,000 T, F, DSC CF#54 Fast 5,000 T, F, DSC
BF-CF#26 Fast 5,000 T, F, DSC CF#55 Fast 5,000 T, F, DSC
BF-CF#1 Slow 5,000 FTIR CF#3 Slow 5,000 FTIR
BF-CF#3 Slow 5,000 FTIR CF#5 Slow 5,000 FTIR
BF-CF#4 Slow 5,000 FTIR CF#7 Slow 5,000 FTIR
BF-CF#5 Slow 5,000 FTIR CF#8 Slow 5,000 FTIR
BF-CF#6 Slow 5,000 FTIR CF#9 Slow 5,000 FTIR
BF-CF#8 Slow 5,000 FTIR CF#10 Slow 5,000 FTIR
BF-CF#11 Slow 5,000 FTIR CF#12 Slow 5,000 FTIR
BF-CF#27 Slow 5,000 T, F, DSC CF#41 Slow 5,000 T, F, DSC
BF-CF#28 Slow 5,000 T, F, DSC CF#42 Slow 5,000 T, F, DSC
BF-CF#30 Slow 5,000 T, F, DSC CF#43 Slow 5,000 T, F, DSC
BF-CF#31 Slow 5,000 T, F, DSC CF#44 Slow 5,000 T, F, DSC
BF-CF#32 Slow 5,000 T, F, DSC CF#45 Slow 5,000 T, F, DSC

The materials were characterized in the same manner as described in Section 5.2, with the
exception of SEM (not performed), and the characterization details will not be repeated
here.

112
6.2. Results and Discussion

6.2.1. Tensile Results
For the samples under tension there were a total of fifteen tensile coupons per
material, including five coupons exposed to tension only, five coupons exposed to tension
and a fast dose rate, and five coupons exposed to tension and a slow dose rate. Each
coupon was cut from a different panel to account for panel variations. The stress-strain
curves of representative coupons for the fast and slow dose rates of both material are
shown in Figures 6.1-6.2 and compared with the fast and slow representative coupons
from Chapter 5.

Figure 6.1: Representative BF-CF tensile stress-strain curves for each exposure group. Blue
represents the slow dose rate, black represents the fast dose rate, (___) represents a coupon
that underwent only radiation, and (---) represents a coupon that underwent radiation and
tension.
113

Figure 6.2: Representative CF tensile stress-strain curves for each exposure group. Blue
represents the slow dose rate, black represents the fast dose rate, (___) represents a coupon
that underwent only radiation, and (---) represents a coupon that underwent radiation and
tension.

In Figure 6.1, the combined exposure of radiation and tension generally appears to
increase the modulus and strength, as well as decrease the strain-to-failure. In Figure 6.2,
the slow dose rate samples have similar results to that of BF-CF, but with the fast dose
rate, the strength generally decreases with tension and radiation.
Subsequently, the modulus, first fracture point, ultimate strength, fracture
strength, and strain-to-failure were calculated for each material from these stress-strain
curves. The values from each coupon were averaged and the standard deviation was
calculated and represented by the error bars. Figures 6.3 and 6.4 below show the results
for the BF-CF material and Figures 6.5 and 6.6 show the results for the CF material.
114

Figure 6.3: The averaged tensile modulus, first fracture point, ultimate strength, and fracture
strength for the BF-CF material as a function of dose rate. The blue represents samples only
exposed to radiation and the black represents samples exposed to both radiation and tension.

Figure 6.4: The averaged tensile strain-to-failure for the BF-CF material as a function of dose
rate. The blue represent samples only exposed to radiation and the black represents samples
exposed to both radiation and tension.

115

Figure 6.5: The averaged tensile modulus, first fracture point, ultimate strength, and fracture
strength for the CF material as a function of dose rate. The blue represents samples only
exposed to radiation and the black represents samples exposed to both radiation and tension.

Figure 6.6: The averaged tensile strain-to-failure for the CF material as a function of dose rate.
The blue represent samples only exposed to radiation and the black represents samples
exposed to both radiation and tension.

116
When comparing the control to the tension only data for the BF-CF material, the
data shows an overall trend of a decrease in strength and modulus and an increase in
strain-to-failure, which would be indicative of the material weakening with prolonged
tensile stress. This result could be indicative of creep effects from the prolonged stress.
The CF material showed little difference between the control and tension only data, with
the exception of the fracture strength and strain-to-failure, where an increase was
observed in the fracture strength and a decreased was observed in the strain-to-failure
when the material was exposed to tension.
However, under the condition of radiation with tension, the BF-CF data shows
trends of increased modulus and strength, with slight decrease in strain-to-failure when
the tensioned data is compared with the non-tensioned data of the corresponding dose
rate. These trends indicate strengthening of the material with radiation and tension.
The CF material in general showed decreased strength and increased strain-to-
failure with radiation and tension when compared with the coupons that only experienced
radiation. The CF trends indicate enhanced degradation occurring with radiation and
tension. Nevertheless, for both materials and in all calculations, the error in the data
precludes any definitive conclusions.
Furthermore, in comparing the fast dose rate with the slow dose rate of samples
that underwent both tension and radiation, in general there are no differences. The BF-
CF material shows a slight increase in modulus with fast dose rate when compared with
slow dose rate, and the CF material shows decreases in first fracture point and ultimate
strength with the fast dose rate when compared with the slow dose rate. However, again,
the error in the data prohibits any conclusions regarding dose rate effects in the samples.
117

6.2.2. Flexure Results
There were a total of fifteen flexure coupons per material, including five coupons
exposed to tension only, five coupons exposed to tension and a fast dose rate, and five
coupons exposed to tension and a slow dose rate. The coupons were each cut from a
different panel to account for panel variations. The stress-strain curves of representative
coupons for the fast and slow dose rates of both materials are shown in Figures 6.7-6.8
and compared with the fast and slow representative coupons from Chapter 5.

Figure 6.7: Representative BF-CF flexure stress-strain curves for each exposure group. Blue
represents the slow dose rate, black represents the fast dose rate, (___) represents a coupon
that underwent only radiation, and (---) represents a coupon that underwent radiation and
tension.
118

Figure 6.8: Representative CF flexure stress-strain curves for each exposure group. Blue
represents the slow dose rate, black represents the fast dose rate, (___) represents a coupon
that underwent only radiation, and (---) represents a coupon that underwent radiation and
tension.

In Figure 6.7, the general trends show an increase in modulus and strength with
tension in the BF-CF material when considering the fast dose rate. However, with a slow
dose rate, the tension decreases the strength and the modulus. In Figure 6.8, the CF
material appears to have no change in modulus and possibly a decrease in strength with
the fast dose rate and tension.
The modulus, first fracture point, ultimate strength, fracture strength, and strain-
to-failure were calculated for each material, averaged, and the standard deviation was
calculated and represented by the error bars. Figures 6.9 and 6.10 below show the results
for the BF-CF material and Figures 6.11 and 6.12 show the results for the CF material.
119

Figure 6.9: The averaged flexural modulus, first fracture point, ultimate strength, and fracture
strength for the BF-CF material as a function of dose rate. The blue represents samples only
exposed to radiation and the black represents samples exposed to both radiation and tension.

Figure 6.10: The averaged flexure strain-to-failure for the BF-CF material as a function of dose
rate. The blue represent samples only exposed to radiation and the black represents samples
exposed to both radiation and tension.
120

Figure 6.11: The averaged flexural modulus, first fracture point, ultimate strength, and
fracture strength for the CF material as a function of dose rate. The blue represents samples
only exposed to radiation and the black represents samples exposed to both radiation and
tension.

Figure 6.12: The averaged flexural strain-to-failure for the CF material as a function of dose
rate. The blue represent samples only exposed to radiation and the black represents samples
exposed to both radiation and tension.

121
In the flexural data of the BF-CF material, the tension increased the modulus,
strengths, and strain-to-failure when compared with the control. With radiation, the
modulus, first fracture point, and ultimate strength either decreased or remained the same
with tension when compared to the samples without tension. In the fracture strength and
strain-to-failure, the radiation and tension either increased the quantities or the values
remained unchanged. With the exception of the increased fracture strength with radiation
and tension, these results indicate scission effects are dominantly expressed when tension
is applied to the BF-CF material in conjunction with radiation exposure.
The CF results also show increases in ultimate strength, fracture strength, and
strain-to-failure in samples exposed to tension only when compared with the control
samples. The modulus decreases and the first fracture point remains unchanged of
samples exposed to tension only when compared with the control samples. Overall, there
is little change in the modulus with radiation and tension when compared with radiation
only. However, similarly to the BF-CF material, the CF material experiences an overall
decrease in ultimate strength, fracture strength, and strain-to-failure with combined
radiation and tension when compared with radiation only. Additionally, the first fracture
point at a fast dose rate decreases, but the slow dose rate sees an increase with radiation
and tension when compared with radiation only. These results indicate evidence of
scission-dominated effects with the decreases in strength, but also indicates potential
cross-linking with the decreased strain-to-failure and increase in first fracture point of the
slow dose rate samples. However, for both materials, the errors overlap and prevent any
definitive conclusions regarding the results.
122
Furthermore, there are no conclusive results concerning a dose rate effect in these
data. There are either slight increases or no change in the slow dose rate of the quantities
for both the BF-CF and CF materials when compared with the fast dose rate. However,
the error in the data prohibits any conclusions from these results.

6.2.3. DSC Results
As was mentioned in Section 5.3.3., the method of collecting DSC data from each
material was different as a result of the manufacturing process of each material. The BF-
CF material had excess removed from each panel during manufacturing, and the pre-
radiation DSC data was gathered from this excess material. The CF samples did not have
excess material from each panel tested and the DSC was gathered only from those panels
that underwent radiation and tension. Thus, the analysis of the data is slightly different
for each material.
The pre-radiation and post-radiation data for the BF-CF material of each panel
was averaged according to exposure group, and the percent change in the glass transition
temperature (T
g
) was calculated. The percent change was plotted against the dose rate, as
shown in the figure below (Figure 6.13).
123

Figure 6.13: Percent change in Tg of the BF-CF material as a function of dose rate. The blue
represents the samples that underwent radiation only and the black represents the samples
that underwent both tension and radiation.

The results show that for the fast dose rate, the T
g
was decreased by ~2% with radiation
and tension when compared to the samples only exposed to radiation, indicating
enhanced scission occurring. However, with the slow dose rate, the T
g
was only slightly
increased with radiation and tension when compared to the radiation only.
The CF material had five fast dose rate and tension panels, and five slow dose rate
and tension panels. The data gathered from these panels were averaged and the standard
deviations were calculated. The data for the tensioned and irradiated panels was
compared with the irradiated only panels from Section 5.3.3., as shown below (Figure
6.14).
124

Figure 6.14: The Tg as a function of dose rate for the CF material. The blue represents the
samples that underwent radiation only and the black represents the samples that underwent
both tension and radiation.

In the CF material, the combined exposure of radiation and tension decreases the T
g
,
indicating enhanced scission. However, this data has overlapping error bars, and no
conclusions can be made.

6.2.4. FTIR Results
Nine panels for each material were investigated via FTIR. The center of the
sample was tested prior to irradiation and immediately after returning from radiation
exposure. The scans from before radiation and after radiation were averaged, and
subsequently subtracted to reveal the differences from the radiation exposure. Then the
data from radiation and tension were plotted with the data from radiation only (presented
in Section 5.3.4.) to highlight any changes from the combined effect of radiation and
tension.
The BF-CF material FTIR data is shown below (Figure 6.15), where the data from
the slow dose rate is shown side by side with the data from the fast dose rate.
125

Figure 6.15: FTIR data of the BF-CF material with fast dose rate data on the left and slow dose
rate data on the right. The blue represents radiation and tension exposure and the black
represents radiation only exposure.

The results show similar chemistry effects in the BF-CF material when exposed to
combined radiation and tension, in which the peaks increase with radiation exposure.
However, in comparison to those samples exposed to only radiation, the radiation and
tension exposed samples have slightly decreased responses. With a fast dose rate, the
tensioned samples show a decrease in the hydroxyl peak (3300 cm
-1
) and a slight
decrease in the carbonyl peak (1750 cm
-1
), indicating less oxidation occurring. There are
also decreases in the aliphatic CH-stretch (2800-3000 cm
-1
), the aromatic carbon double
bond stretch (1300-1550 cm
-1
), and the CH-bending (600-1400 cm
-1
), indicating
competition of scission and cross-linking occurring but to a lesser degree than in the
samples that underwent only radiation exposure.

126
The CF material results for both the fast and slow dose rates are shown in Figure
6.16.

Figure 6.16: FTIR data of the CF material with fast dose rate data on the left and slow dose rate
data on the right. The blue represents radiation and tension exposure and the black
represents radiation only exposure.

In Figure 6.16, both the fast and slow dose rate with tension experiences decreases in the
peaks with radiation and tension when compared with the pre-radiation values of the
samples. In both dose rates, there are decreases in the hydroxyl peak (~3400 cm
-1
) and
the CH-stretch peaks (~3000 cm
-1
). In the fast dose rate, there are also large decreases in
the carbon double bond stretch in the aromatic ring (1400-1700 cm
-1
), the CH-bend in-
plane (900-1400 cm
-1
), and the CH-bend out of plane (600-900 cm
-1
). These results
indicate that with a fast dose rate there is enhanced scission taking place when the
samples are irradiated under stress. However, in the slow dose rate, there appears to be a
slight decrease in the carbon double bond stretch of the aromatic rings (1400-1700 cm
-1
)
and no differences in the CH-bending at the lower wavenumbers. The slow dose rate also
indicates the effects of enhanced scission occurring with radiation and tension, but is not
as severe as in the fast dose rate. Furthermore, in both dose cases, there are no
appearances of a carbonyl group and the hydroxyl group decreases with radiation and
tension, indicating there are no effects of oxidation taking place in the CF material.
127

6.3. Conclusions
The experiment discussed in this chapter was designed to investigate whether
additional degradation in the materials occur as a result of synergistic effects when the
samples are exposed to both radiation and tension simultaneously. The mechanical
properties of both materials were inconsistent in the outcome when samples exposed to
radiation and tension were compared with samples only exposed to radiation. These
inconsistencies could be a result of competing scission and cross-linking occurring in the
materials. Furthermore, there was scatter in the data of both materials, leading to
inconclusive results. Therefore, we conclude there are no significant changes to the
mechanical properties as a result of combined radiation and tension.
It should also be noted that the FTIR data of both materials did show consistency
with respect to a decrease in all the peaks examined when radiation and tension samples
were compared with radiation only samples. While these results show evidence of
competition with scission and cross-linking effects, they are occurring to a lesser extent
in the samples exposed to radiation and tension, perhaps leading to the lack of significant
changes in the mechanical properties. Overall, these results indicate that the tension
improves the material properties when it is simultaneously exposed to radiation, but it is
not clear why this is the case. Further investigation into effects of radiation and tension
should be undertaken to ascertain whether there is indeed synergisms, perhaps at larger
doses to magnify any effects.
128
Chapter 7

Surface Investigation of Visible Aging
The final experiment explored the surface of the materials to determine whether
there were any visible effects of aging occurring as a result of radiation exposure or
combined radiation and tension. Given that there was evidence of surface chemistry
modifications through FTIR, and particularly oxidation of the surface in the BF-CF
material, necessitates understanding whether surface chemistry deviations are significant
enough that there are visible changes to the material.

7.1. Test Setup and Methods
Four panels of the BF-CF material and four panels of the CF material were
studied with scanning electron microscopy prior to radiation exposure, immediately after
returning from radiation exposure (about one month after radiation exposure), and
approximately nine months after the radiation exposure. Two panels, CF#27 and CF#28,
were not investigated prior to radiation exposure due to a lack of time and equipment
availability. However, there were little differences in the CF surface visual properties
amongst the panels prior to radiation and we determined that the lack of these images
would not be detrimental to the study.
All the materials were subjected to a slow radiation dose rate (0.177 Gy/s), a total
dose of 5,000 Gy, and two panels of each material were exposed to simultaneous
129
radiation and tension. The slow dose rate was chosen because we wanted to investigate
surface oxidation in particular and whether it was creating an observable effect on the
sample surface. The exposure details are listed in the table below (Table 7.1).
Table 7.1: Exposure details for each panel of the BF-CF and CF materials investigated.
Panel # Speed Dose (Gy) Tension Panel # Speed Dose (Gy) Tension
BF-CF#2 Slow 5,000 No CF#7 Slow 5,000 Yes
BF-CF#5 Slow 5,000 Yes CF#8 Slow 5,000 Yes
BF-CF#7 Slow 5,000 No CF#27 Slow 5,000 No
BF-CF#11 Slow 5,000 Yes CF#28 Slow 5,000 No

7.2. Results and Discussion
Four panels of the BF-CF material were investigated, two exposed only to
radiation and two exposed to both radiation and tension. Using a scanning electron
microscope, the resin surface was studied to determine whether any visible changes could
be ascertained to support enhanced degradation and aging. Micrographs representing
each type of exposure, BF-CF#2 (Figure 7.1) and BF-CF#5 (Figure 7.2), are shown
below. In these micrographs, the images captured approximately nine months after the
radiation exposure appear slightly different than the other images as a result of laboratory
equipment upgrades to a newer SEM.
130

Figure 7.1: Micrographs of the resin surface of BF-CF#2 where the sample was only
exposed to radiation. The top left micrograph was taken prior to radiation exposure, the top
right micrograph was taken ~1 month after radiation exposure, and the bottom micrographs
were taken ~9 months after radiation exposure.

131

Figure 7.2: Micrographs of the resin surface of BF-CF#5 where the sample was
exposed to both radiation and tension. The top left micrograph was taken prior to radiation
exposure, the top right micrograph was taken ~1 month after radiation exposure, and the
bottom micrographs were taken ~9 months after radiation exposure.

In these images, a cross-hatch pattern is observed on the surface. This pattern is
an imprint in the resin of the layered materials used in the manufacturing process. There
are also raised areas (Figure 7.2, pre and post micrographs) visible, which are resin rich
areas as a result of the imprints from the manufacturing process. In addition, in Figure
7.2, there are dark lines that are diagonal across the image. These diagonal lines are the
boron fibers underneath the resin in places where the resin is thin. In observing the resin-
filled and resin-rich areas of the surface, there is no apparent difference in the
132
micrographs before radiation, about one month after radiation, and approximately nine
months after radiation.
Four panels of the CF material were also imaged via SEM before radiation, about
one month after radiation, and roughly nine months after radiation. Micrographs
representing each type of exposure, CF#7 (Figure 7.3) and CF#27 (Figure 7.4), are shown
below. In these micrographs, the images captured approximately months after the
radiation exposure appear slightly different than the other images as a result of laboratory
equipment upgrades to a newer SEM. In addition, we were unable to collect data on
CF#27 prior to radiation as a result of equipment availability.

133

Figure 7.3: Micrographs of the resin surface of CF#7 where the sample was only
exposed to radiation. The top left micrograph was taken prior to radiation exposure, the top
right micrograph was taken ~1 month after radiation exposure, and the bottom micrographs
were taken ~9 months after radiation exposure.

134

Figure 7.4: Micrographs of the resin surface of CF#27 where the sample was exposed
to both radiation and tension. The top micrograph was taken ~1 month after radiation
exposure and the bottom micrographs were taken ~9 months after radiation exposure.

In the CF micrographs, the carbon fibers are clearly visible underneath the resin
as diagonal lines. In addition, there are fewer instances of the cross-hatching and resin
rich areas in these micrographs as a result of smoother material used during the
manufacturing process. Comparing the pre-radiation micrographs with the post-radiation
micrographs show no differences in the surface resin properties.

135
7.3. Conclusions
The experiment detailed herein was an investigation of the visible surface
characteristics of these materials and whether they were impacted as a result of radiation
or radiation and tension. The results discussed above for both materials indicate that
while there may be surface chemistry changes occurring, as evidenced by the FTIR
results in Chapters 5 and 6, the chemistry has not changed significantly such that there
are visible features evident on the surface. Thus, we conclude that radiation or radiation
and tension do not produce appreciable changes to the material surface chemistry to yield
visible differences.
136
Chapter 8

Conclusions
The purpose of this study was to investigate composite materials for a lunar
habitat and their durability over a protracted exposure to the lunar radiation environment.
Previous studies lacked investigation of an environment similar to the lunar radiation
environment, consisting of high-energy protons. Additionally, little research has been
completed on epoxies and fiber reinforced composites, and the completed studies thus far
have not investigated toughened composites or epoxies. Furthermore, studies involving
the combined exposures of radiation and stress have been minimal. Thus, to gain insight
into potential contemporary aerospace composites for lunar habitats, we studied two
aerospace composites of interest that contained semi-toughened epoxies, and exposed
them to high-energy proton radiation, simulating a thirty-year mission on the lunar
surface. Additionally, we investigated in-situ effects, synergistic effects of radiation and
tension, as well as dose rate effects.
From the results, we concluded that generally there are chemical changes on the
surface of these composites as a result of high-energy proton radiation exposure in air,
but the overall mechanical changes are minimal. During in-situ radiation of the samples
of both materials subjected to bi-axial tension, we found that those exposed to a fast dose
rate experienced shrinkage, whereas those exposed to a slow dose rate experienced
stretching. These results would indicate that the fast dose rate caused in-situ cross-
linking in the material, whereas the slow dose rate caused in-situ scission of the material.
However, these effects were not necessarily confirmed by the mechanical results
137
following the radiation exposures. Rather, the CF material had no conclusive evidence of
degradation in the mechanical results, whereas the BF-CF material validated evidence of
degradation as a result of chain scission and debonding of resin from the boron fibers.
The changes observed in the BF-CF material were a result of the epoxy chemistry,
toughening agents, oxidative degradation, and enhanced dose in the matrix as a result of
secondary radiation. Furthermore, we concluded that there were no significant
synergistic effects from combined radiation and tension exposure, and there was no
conclusive evidence of dose rate effects observed. Additionally, there was no
confirmation of degradation of the material surfaces as a result of radiation or radiation
and tension over a nine-month time period post-radiation exposure.
While this study provided a comprehensive first-step at investigating
contemporary aerospace composite durability for a deep space environment, it has also
provided questions for further exploration. This work primarily focused on the radiation
exposure, and did not take into account the aging of the material over the thirty-year
period. Given that there is evidence of enhanced scission occurring in the BF-CF
material and evidence of competing effects in the CF material, it is important to also
understand whether these chemistry changes will cause enhanced aging of the material
with time. If there is enhanced aging occurring as well, the lifetime expectancies of the
material will be affected and this information is vital to engineers and designers using
these materials for long-term space applications. Furthermore, the lack of comprehensive
chemical information on the epoxies in the composites prevented deeper understanding of
the chemical changes, especially in understanding the effects of toughening agents in the
composites. Thus, it will be important to further investigate toughening agents in epoxies
138
in a systematic way to better understand the implications of their radiation susceptibility
in the overall matrix system. Currently, the radiation chemistry community has a basic
understanding of how radiation affects polymeric systems. However, this understanding
is not deep enough to the point where simulation tools can be used by designers for
understanding a material’s capabilities. Therefore, additional systematic studies need to
occur on epoxies to provide better comprehension of the radiation chemistry effects and
their implications in the mechanical properties. This knowledge will aid in the
development of simulation tools that designers and engineers can use in the future to
understand safe lifetime limitations of these materials in harsh space environments.

139
References

1. K. Addison, 2008. Sources of Space Weather. Solar Dynamics Observatory.
Retrieved 22 March 2009, http://sdo.gsfc.nasa.gov/mission/spaceweather.php

2. S. Alariqi, A. Kumar, B. Rao, and R. Singh. Effect of γ-dose rate on crystallinity and
morphological changes of γ-sterilized biomedical polypropylene. Polymer
Degradation and Stability, 94(2):272–277, 2009.

3. S. Alariqi, R. Singh, and B. Rao. γ-Initiated Oxidation of Isotactic Polypropylene:
Effect of γ-Dose Rate on Durability. Arabian Journal for Science and Engineering,
36(1):29–38, 2011.

4. M. Al-Sheikhly and A. Christou. Tutorial: How Radiation Affects Polymeric
Materials. IEEE Transactions on Reliability, 43(4), 551-556, 1994.

5. W. Atwell, S. Koontz, B. Reddell, K. Rojdev, and J. Franklin. Meeting Radiation
Protection Requirements and Reducing Spacecraft Mass - A Multifunctional
Materials Approach. AIAA Space 2010 Conference and Exposition, Anaheim, CA,
Aug 30 – Sept 2, 2010.

6. D. Band, J. Matteson, L. Ford, B. Schaefer, D. Palmer, B. Teegarden, T. Cline, M.
Briggs, W. Paciesas, G. Pendleton, G. Fishman, C. Kouveliotuou, C. Meegan, R.
Wilson, and P. Lestrade. BATSE Observations of gamma-ray burst spectra I. –
Spectral diversity. Astrophysical Journal, 413: 281-292, 1993.

7. R. Bagheri, B. Marouf, and R. Pearson. Rubber-Toughened Epoxies: A Critical
Review, Polymer Reviews, 49(3): 201-225, 2009.

8. E. Barbero. Introduction to Composite Material Design. Taylor & Francis Group,
LLC, 1999.

9. A. Bedford, K. Liechti. Mechanics of Materials. Prentice Hall, 2000.

10. S. Belfer, R. Fainchtain, Y. Purinson, and O. Kedem. Surface characterization by
FTIR-ATR spectroscopy of polyethersulfone membranes-unmodified , modified and
protein fouled. Journal of Membrane Science, 172: 113–124, 2000.

11. R. Bernstein, S. Thornberg, A. Irwin, J. Hochrein, D. Derzon, S. Klamo, and R.
Clough. Radiation-oxidation mechanisms: Volatile organic degradation products
from polypropylene having selective C-13 labeling studied by GC/MS. Polymer
Degradation and Stability, 93(4):854–70, 2008.

12. H. Bethe and J. Ashkin. Experimental Nuclear Physics. (E. Segre, Ed.) John Wiley
& Sons, Inc., pages 166–251, 1953.
140

13. R. Black and B. Lyons. Effect of High-Energy Radiation on Polypropylene. Nature,
180:1346–1347, 1957.

14. Center for Integrated Space Weather Modeling (CISM). Developing Solar Energetic
Particle Predictions with the CISM Model Suite. Retrieved 22 March 2009,
www.bu.edu/cism/News/highlights.html

15. D. Clegg, and A. Collyer. Irradiation Effects on Polymers. Elsevier Science
Publishers LTD, 1991.

16. R. Clough and K. Gillen. Combined Environment Aging Effects: Radiation-Thermal
Degradation of Polyvinylchloride and Polyethylene. Journal of Polymer Science
Part A: Polymer Chemistry, 19(8): 2041-2051, 1981.

17. L. Costa, I. Carpentieri, and P. Bracco. Post electron-beam irradiation oxidation of
orthopedic UHMWPE. Polymer Degradation and Stability, 93(9):1695–703, 2008.

18. D. Coulter, A. Gupta, M. Smith, and R. Fornes. The Effects of Energetic Proton
Bombardment on Polymeric Materials: Experimental Studies and Degradation
Models. 1986. NASA-CR-177161.

19. Cytec. Composite Product Detail: CYCOM 977-3. Retrieved 20 February 2009,
http://www.cemselectorguide.com/comp_product_detail.php?prodname=977-3

20. F. Dasilva, K. Dasaquino, and E. Araujo. Effects of Gamma Irradiation on
Poly(vinyl chloride)/Polystyrene Blends: Investigation of Radiolytic Stabilization
and Miscibility of the Mixture, Polymer Degradation and Stability, 93(12): 2199-
2203, 2008.

21. J. Davenas, I. Stevenson, N. Celette, S. Cambon, J. Gardette, A. Rivaton, and L.
Vignoud. Stability of Polymers Under Ionizing Radiation: The Many Faces of
Radiation Interactions with Polymers, Nuclear Instruments and Methods in Physics
Research, Section B, 191: 653-661, 2002.

22. A. Denaro. Fundamentals of Radiation Chemistry. Ann Arbor Science Publishers,
Inc., 1972.

23. K. Dismukes, 2004. Space Station Extravehicular Activity: Working Outside the
International Space Station. Retrieved 8 March 2009,
http://spaceflight.nasa.gov/station/eva/outside.html

24. J. Dorsey, K. Wu, and R. Smith. Structural Definition and Mass Estimation of Lunar
Surface Habitats for the Lunar Architecture Team Phase 2 (LAT-2) Study.
Proceedings of Earth and Space 2008, Long Beach, CA, Mar 3 – Mar 8, 2008.

141
25. P. Eckart. The Lunar Base Handbook. The McGraw-Hill Companies, Inc., 1999.

26. S. Egusa, M. Kirk, and R. Birtcher. Neutron Irradiation Effects on the Mechanical
Properties of Organic Composite Materials, Journal of Nuclear Materials, 126(2):
152-159, 1984.

27. R. Fornes, R. Gilbert, and J. Memory. Effects of High Energy Radiation on the
Mechanical Properties of Epoxy-Graphite Fiber Reinforced Composites. Jul 1, 1987.
NASA-CR-181110.

28. K. Gillen and R. Clough. Radiation-Thermal Degradation of PE and PVC:
Mechanism of Synergism and Dose Rate Effects. Radiation Physics and Chemistry,
18(3-4), 661-669, 1981a.

29. K. Gillen and R. Clough. Occurrence and Implications of Radiation Dose-Rate
Effects for Material Aging Studies. Radiation Physics and Chemistry, 18(3-4):679–
687, 1981b.

30. K. Gillen and R. Clough. Time-Temperature-Dose Rate Superposition: A
Methodology for Extrapolating Accelerated Radiation Aging Data to Low Dose Rate
Conditions. Polymer Degradation and Stability, 24:137-168, 1989.

31. T. Gombosi. Physics of the Space Environment. Cambridge University Press, 1998.

32. J. Hedrick, I. Yilgor, M. Jurek, J. Hedrick, G. Wilkes, and J. McGrath. Chemical
modification of matrix resin networks with engineering thermoplastics: 1. Synthesis,
morphology, physical behavior and toughening mechanisms of poly(arylene ether
sulphone) modified epoxy networks, Polymer, 32(11): 2020–2032, 1991.

33. E. Hegazy, T. Seguchi, K. Arakawa, and S. Machi. Radiation-Induced Oxidative
Degradation of Isotactic Polypropylene. Journal of Applied Polymer Science, 26(4):
1361-1372, 1981.

34. E. Hoffman and T. Skidmore. Radiation Effects on Epoxy/Carbon-Fiber Composite.
2008. SRNL-STI-2008-00527.

35. ICRU. Fundamental Quantities and Units for Ionizing Radiation (Revised). Journal
of the ICRU, 11(1): 1–35, 2011.

36. A. Jablonski and K. Ogden. Technical Requirements for Lunar Structures. Journal of
Aerospace Engineering, 21(2): 72-90, 2008.

37. G. Kent, J. Memory, R. Gilbert, and R. Fornes. Variation in radical decay rates in
epoxy as a function of crosslink density. Journal of Applied Polymer Science,
28(10): 3301–3307, 1983.

142
38. A. Kenyon, and L. Nielsen. Characterization of Network Structure of Epoxy Resins
by Dynamic Mechanical and Liquid Swelling Tests, Journal of Macromolecular
Science, Part A, 3(2): 275-295, 1969.

39. F. Khoylou and A. Katbab. Radiation Degradation of Polypropylene. Radiation
Physics and Chemistry, 42(1-3):219–222, 1993.

40. C. Klabunde, and R. Coltman. Debonding of Epoxy from Glass in Irradiated
Laminates. Journal of Nuclear Materials, 117: 345-350, 1983.

41. T. Komiyama and H. Matsumoto. Calculation of Radiation Exposure on the Moon
Surface by using PHITS. 5
th
GEANT4 Space Users Workshop, Tokyo, Japan, Feb 13
– Feb 15, 2008.

42. H. Kudoh, I. Celina, G. Malone, R. Kaye, K. Gillen, and R. Clough. Pulsed E-Beam
Irradiation of Polymers - A Comparison of Dose Rate Effects and LET Effects.
Radiation Physics and Chemistry, 48(5): 555–562, 1996.

43. H. Kudoh, T. Sasuga, T. Seguchi, and Y. Katsumura. High Energy Ion Irradiation
Effects on Polymer Material: 4. Heavier Ion Irradiation Effects on Mechanical
Properties of PE and PTFE. Polymer, 37(16): 3737–3739, 1996.

44. B. Kumar, R. Singh, and T. Nakamura. Degradation of Carbon Fiber-reinforced
Epoxy Composites by Ultraviolet Radiation and Condensation. Journal of
Composite Materials, 36(24): 2713-2733, 2002.

45. R. Kurland, J. Thomasson, and W. Beggs. Radiation Testing of Composite
Materials, In situ versus ex situ effects. 1981. NASA CR-3475.

46. C. Leung. Space Environmental Effects on Graphite Epoxy Composites. Composites
for Extreme Environments. American Society for Testing and Materials, 110-117,
1982.

47. P. Linstron, and W. Mallard, Eds. NIST Chemistry WebBook. NIST Standard
Reference Database Number 69, National Institute of Standards and Technology. ,
Retrieved 15 July 2012, http://webbook.nist.gov

48. N. Longiéras, M. Sebban, P. Palmas, A. Rivaton, and J. Gardette. Degradation of
epoxy resins under high energy electron beam irradiation: Radio-oxidation. Polymer
Degradation and Stability, 92(12): 2190–2197, 2007.

49. A. Maji and S. Mahnke. A Review of the Degradation of Composites in Mid-Earth
Orbit. Earth and Space 2004, League City, TX, Mar 7 – Mar 10, 2004.

143
50. J. Memory, R. Fornes, and R. Gilbert. Radiation Effects on Graphite Fiber
Reinforced Composites. Journal of Reinforced Plastics and Composites, 7(1): 33-65,
1988.

51. S. Milkovich, C. Herakovich, and G. Sykes. Space Radiation Effects on the Thermo-
Mechanical Behavior of Graphite-Epoxy Composites. Journal of Composite
Materials, 20(6): 579-593, 1986.

52. K. Mimura, H. Ito, and H. Fujioka. Improvement of thermal and mechanical
properties by control of morphologies in PES-modified epoxy resins. Polymer,
41(12): 4451-4459, 2000.

53. K. Mimura, H. Ito, and H. Fujioka. Toughening of Epoxy Resin Modified with In-
situ Polymerized Thermoplastic Polymers. Polymer, 42(22): 9223-9233, 2001.

54. T. Murayama, and J. Bell. Relation Between the Network Structure and Dynamic
Mechanical Properties of a Typical Amine-Cured Epoxy Polymer. Journal of
Polymer Science, Part A-2, 8: 437-445, 1970.

55. NASA. Human Integration Design Handbook. 2010. NASA SP-2010-3407.

56. NASA. NASA-STD-3000. Man-Systems Integration Standards, Rev B, 1(5): 227-
275, 1995.

57. NASA. Surface Architecture Reference Document. 2008.

58. National Research Council. Managing Space Radiation Risk in the New Era of
Space Exploration. The National Academies, 2008.

59. L. Nielsen. Cross-linking: Effect on Physical Properties of Polymers, Journal of
Macromolecular Science, Reviews in Macromolecular Chemistry and Physics,
C3(1): 69-103, 1969.

60. J. O’Donnell, N. Rahman, and D. Winzor. Evidence of Crosslinking and Chain
Scission in the Degradation of Poly(tert-butyl Crotonate) by γ-Irradiation. Journal of
Polymer Science and Polymer Chemistry, 15: 131-139, 1977.

61. J. O’Donnell and D. Sangster. Principle of Radiation Chemistry. American Elsevier
Publishing Company, Inc., 1970.

62. H. Otaguro, L. de Lima, D. Parra, A. Lugao, M. Chinelatto, and S. Canevarolo.
High-energy radiation forming chain scission and branching in polypropylene.
Radiation Physics and Chemistry, 79: 318-324, 2010.

63. W. Parkinson and O. Sisman. The Use of Plastics and Elastomers in Nuclear
Radiation. Nuclear Engineering and Design, 17: 247-280, 1971.
144

64. D. Queiroz, F. Fraïsse, B. Fayolle, M. Kuntz, and J. Verdu. Radiochemical Ageing
of Epoxy Coating for Nuclear Plants, Radiation Physics and Chemistry, 79(3): 362-
364, 2010.

65. A. Rivaton, and J. Arnold. Structural modifications of polymers under the impact of
fast neutrons. Polymer Degradation and Stability, 93(10):1864–1868, 2008.

66. K. Rojdev, W. Atwell, R. Wilkins, B. Gersey, F. Badavi. Evaluation of Multi-
Functional Materials for Deep Space Radiation Shielding. National Space and
Missile Materials Symposium, Henderson, NV, June 22 – June 25, 2009.

67. K. Rojdev, M. O’Rourke, C. Hill, S. Nutt, and W. Atwell. In-Situ Strain Analysis of
Potential Habitat Composites Exposed to a Simulated Long-Term Radiation
Exposure. Radiation Physics and Chemistry, in press, 2012.

68. V. Romanov, G. Khorasanov, I. Konstantinov, A. Smolyanskii, E. Klinshpont, V.
Tupikov, and V. Milinchuk. Durability changes of epoxy resins under action of
protons and gamma rays. Radiation Physics and Chemistry, 46(4-6): 863–866, 1995.

69. R. Rose, 2008. Natural and Induced Environments. Man-Systems Integration
Standards. Retrieved 15 March 2009, http://msis.jsc.nasa.gov/sections/section05.htm

70. T. Sasuga and M. Hagiwara. Disintegration of Network Structure of Bismaleimide-
triazine Resin by Electron Beam Irradiation, Polymer, 27: 681-685, 1986.

71. T. Sasuga, N. Hayakawa, K. Yoshida, and M. Hagiwara. Degradation in Tensile
Properties of Aromatic Polymers by Electron Beam Irradiation. Polymer, 26: 1039-
1045, 1985.

72. T. Sasuga, S. Kawanishi, M. Nishii, T. Seguchi, and I. Kohno. Effects of ion
irradiation on the mechanical properties of several polymers. International Journal
of Radiation Applications and Instrumentation, Part C, 37(1): 135-140, 1991.

73. T. Sasuga, H. Kudoh, and T. Seguchi. High energy ion irradiation effects on polymer
materials – Changes in mechanical properties of PE , PSF and PES. Polymer, 40:
5095-5102, 1999.

74. T. Seguchi, K. Arakawa, N. Hayakawa, and S. Machi. Radiation Induced Oxidative
Degradation of Polymers – IV. Dose Rate Effects on Chemical and Mechanical
Properties. Radiation Physics and Chemistry, 18(3-4): 671-678, 1981.

75. N. Sela and O. Ishai. Interlaminar Fracture Toughness and Toughening of Laminated
Composite Materials: A Review. Composites, 20(5): 423-435, 1989.

145
76. T. Slaba, S. Blattnig, S. Aghara, L. Townsend, T. Handler, T. Gabriel, L. Pinsky, and
B. Reddell. Coupled Neutron Transport for HZETRN. Radiation Measurements,
45(2): 173-182, 2010.

77. T. Slaba, S. Blattnig, and F. Badavi. Faster and More Accurate Transport Procedures
for HZETRN. 2010. NASA/TP-2010-216213.

78. Specialty Materials, Inc. Hy-Bor Prepreg Tape. Retrieved 20 February 2009,
http://specmaterials.com/hyborprepregtape.htm

79. J. Spinks, and R. Woods. An Introduction to Radiation Chemistry. John Wiley &
Sons, Inc., 3
rd
edition, 1990.

80. B. Stuart. Infrared Spectroscopy: Fundamentals and Applications. John Wiley &
Sons, Ltd., 71-93, 2004.

81. S. Tripathy, R. Mishra, K. Dwivedi, D. Khathing, S. Ghosh, and D. Fink. Dose-
dependent modification of polypropylene by 62 MeV protons. Radiation
Measurements, 35:95–98, 2002.

82. J. Turner. Atoms, Radiation, and Radiation Protection. Wiley-VCH Verlag GmbH &
Co., 2007.

83. A. Tylka. Cosmic Ray Effects on Micro Electronics. Naval Research Lab. Retrieved
24 April 2007, https://creme96.nrl.navy.mil/

84. L. Vignoud, L. David, B. Sixou, and G. Vigier. Influence of Electron Irradiation on
the Mobility and on the Mechanical Properties of DGEBA / TETA Epoxy Resins,
Polymer, 42: 4657-4665, 2001.

85. H. Wilski. The Radiation Induced Degradation of Polymers. Radiation Physics and
Chemistry, 29(1): 1-14, 1987.

86. K. Wolf, J. Memory, R. Gilbert, and R. Fornes. Effects of 0.5-MeV Electrons on the
Interlaminar Shear and Flexural Strength Properties of Graphite Fiber Composites.
Journal of Applied Physics, 54(10): 5558-5582, 1983.

87. R. Woods and A. Pikaev. Applied Radiation Chemistry: Radiation Processing. John
Wiley & Sons, Inc., 197-199, 1994.

88. G. Yu, Y. De-zhuang, X. Jing-dong, H. Shi-Yu, and L. Zhi-jun. Effect of Proton
Irradiation on Mechanical Properties of Carbon/Epoxy Composites. Journal of
Spacecraft and Rockets, 43(3): 505-508, 2006.

89. J. Ziegler, 1984. The Stopping and Range of Ions in Matter. Retrieved 27 October
2008, www.srim.org
146

Appendix A

Radiation Terminology
In radiation physics, there are fundamental definitions and equations that should
be mentioned for clarity. In this study, the focus was on deep space radiation, composed
of particles, and the defining features of particle radiation are its atomic number and
energy. Furthermore, the radiant energy, R, is defined as the energy of the particles
emitted, transferred, or received [35]. It is given by the following equation, and does not
include rest energy in this definition.
R = NE (A.1)
In Equation A.1, N is the number of particles that are emitted and E is the energy of the
particles. The flux ( N
&
) of the particles is the increment of particle number (dN) in a time
interval (dt) as given by the following equation [35].

dt
dN
N =
&
(A.2)
The fluence (Φ) is the number of particles (dN) on a sphere of cross-sectional area (da),
where da is perpendicular to the direction of each particle [35].

da
dN
= Φ (A.3)

147
Following is the fluence rate (Φ
&
) [35], which is the incremental fluence (dΦ) over a
particular time interval (dt).

dt
dΦ
= Φ
&
(A.4)
The above equations provide a method to describe the source radiation, or radiation field.
The following will describe primary radiation interaction with a target material.
When a proton travels through a material, it loses energy by interacting with
electrons of the atoms bound in the material. The probability of this interaction (P) when
an atom or molecule of the target material is subjected to a particle fluence (Φ) is called
the cross section (σ) [79].

Φ
=
P
σ (A.5)
The energy losses occur via collisions that create ionization and excitation of atoms.
These collisions can be classified as elastic when they involve only an exchange of
translational kinetic energy, or inelastic when the collision involves excitation or
ionization [22].
The energy loss experienced by the incident particle is due to both secondary
radiation that is produced and to the inelastic collisions that take place. The secondary
radiation emitted from the material does not occur unless the incident proton has
relatively high energy and is dependent on the atomic number of the material being
traversed. This secondary radiation can escape the material or be absorbed through
inelastic collisions. These inelastic collisions, for both the incident particle and the
148
secondary particles, are interactions of the particle with the electrons of the material
being traversed. As the particle approaches a molecule, it electrostatically affects the
electrons bound in the outer shell. Depending on the energy of the incident particle, the
outer shell electron of the molecule can transfer the incident energy to its internal energy,
and raise its energy level to a higher state, or leave the molecule completely [22].
As the energy of the incident particle is lost to the absorbing target material, the
incident particle’s speed is also reduced. At some point before all the energy of the
incident particle is depleted, the incident particle loses large amounts of energy to the
absorbing target material. This effect is known as the Bragg peak. After this peak, the
incident particle does not have enough energy to overcome the binding energy of the
target electrons to cause ionization, thereby losing its final energy to the molecules.
The target material chemistry determines the degree to which the incident
particle’s energy is decreased. This effect is known as the stopping power of the material
and is defined as the average linear rate of energy loss of a proton in the material. In
effect, this describes the energy loss of the particle to the material. The stopping power is
closely linked with the dose the particle imparts on the material, which is the energy
deposited within the material [82]. The stopping power (S) [35] is defined as the amount
of energy the incident particle loses to the medium (dE) while traversing a certain
distance (dl).

dl
dE
S = (A.6)
149
Using relativistic quantum mechanics, Bethe theoretically derived a detailed equation for
stopping power based on the Born approximation of heavy particle collisions with
electrons. The Born approximation makes two assumptions [12]. The first is that the
charge (ze) to velocity (v) ratio of the primary particle must be very small, as given by the
following equation. In this equation, h is the reduced Planck constant (1.05E-34 J-s).
1
2
<<
v
ze
h
(A.7)
The second is that the velocity of the primary particle is large compared with the
velocities of the electrons within the atoms. This is given by the following equation,
where E is the energy of the incident particle, M is the mass of the incident particle, m is
the mass of the electron, and E
el
is the ionization potential of the electrons.

el
E
m
M
E >> (A.8)
Using these assumptions, Bethe derived the following equation for relativistic velocities
[82, 12].

( )






−
−
= −
2
2
2 2
2 2
4 2 2
0
1
2
ln
4
β
β
β
β
π
I
mc
mc
n e z k
dx
dE

(A.9)
In this equation, the following parameters are used:
k
0
= 8.99 x 10
9
N-m
2
-C
-2

z = atomic number of the heavy particle
e = magnitude of the electron charge
n = number of electrons per unit volume in the medium
m = electron rest mass
c = speed of light in vacuum
β = V/c = speed of particle relative to c
I = mean excitation energy of the medium
150
Another term closely linked with stopping power is the linear energy transfer
(LET). The LET (L
Δ
) is the mean energy lost (dE
Δ
) by the incident charged particle due
to electronic interactions in traversing a distance (dl), minus the mean sum of the kinetic
energies in excess of a cutoff energy Δ of all the secondary electrons [35].

dl
dE
L
Δ
Δ
= (A.10)
This definition is focused on the secondary electron production in the close vicinity along
the track of the incident particle, which is the reason for an imposed threshold on the
secondary electrons energies. If the secondary electron energy is very high, the electron
may traverse very large distances away from the incident particle track. Therefore, this
definition excludes those high energy electrons.
The range of an incident particle (R(T)) is also closely related to the stopping
power of the material and is defined as the total distance a particle travels before
exhausting its energy and coming to rest. In this equation, T is the kinetic energy of the
particle and -dE/dx is the stopping power of the material.

( )
∫
−






− =
T
dE
dx
dE
T R
0
1

(A.11)
Similar to the stopping power, the absorbed dose (D) is defined as the mean
energy (dε ) of the radiation field transmitted to the target mass (dm) [35].

dm
d
D
ε
= (A.12)
151
The mean energy (ε ) is the radiant energy of all the particles entering the material (R
in
),
minus the radiant energy of the all the particles exiting the material (R
out
), plus the mean
sum of all changes in the rest energies (Q) of the nuclei and elementary particles involved
with the interaction [35].

∑
+ − = Q R R
out in
ε (A.13)
The dose rate ( D
&
) is the amount of dose (D) imparted over a specific time period (t) [35].

dt
dD
D =
&
(A.14)
152

Appendix B

Proton Range in Epoxy – SRIM Study
During the radiation exposure, panels were in a stacked configuration to save time
and money at Indiana University Cyclotron Facility (IUCF). However, it was necessary
to ensure that the radiation penetrated through all the samples in a similar way so that we
maintained consistency with the radiation exposure amongst all samples. Thus, we
performed a simulation using SRIM (The Stopping and Range of Ions in Matter), a
simulation software package that transports ions through matter and performs various
calculations [88].
To perform these calculations the composites were approximately modeled by a
generic epoxy built into the software (Figure B.1). The software did not have the
capability to model the composites in detail. The epoxy was impacted by hydrogen ions
(protons) with a maximum energy of 200 MeV.
153

Figure B.1: SRIM setup screen showing proton radiation with a maximum of 200 MeV energy
range and a generic epoxy compound from the SRIM internal “compound dictionary”.

The output from the software provides protons from energies 10 keV to 20 MeV,
the stopping power, and the range of the protons at each of these energy levels. The
stopping power and range from this simulation are shown below (Figure B.2).
154

Figure B.2: Stopping power and range for the simulation of protons in an epoxy.

From these graphs, you can see that as the proton energy increases, the stopping
power of the material decreases and the range of the proton increases. At 200 MeV, the
range of the protons in epoxy would be ~225.45 mm based on these calculations.
The panels investigated in this study had an average thickness of ~0.889 mm and
given that the stacked configuration had a maximum of fifteen panels, the total thickness
a proton would need to traverse is 13.34 mm. This value is far below the 225.45 mm
calculated as the range of 200 MeV protons with the SRIM software, and we concluded
that a fifteen-panel stacked configuration would receive consistent radiation through the
panels.
155
Appendix C

Strain Correlation to Worst-Case Hoop Stress
In Chapter 2, calculations were performed to determine the worst-case stress
imparted to a skin-stiffened habitat pressure vessel. These calculations showed that the
hoop stress is ~41 MPa in this configuration. Thus, in tensioning the samples for the case
of radiation and tension, the bi-axial stress on the material needed to be at least 41 MPa.
The tool used to determine the stress on the material was a bi-axial strain gauge placed in
the center of the sample. However, the strain gauge measures strain, not stress, and a
correlation between stress and strain was needed.
Using control panels from both the CF and BF-CF material, tensile tests were
performed on three coupons for each material. The coupon dimensions and the test setup
followed those described in Chapter 5. Stress-strain curves were generated from each
coupon and the strain corresponding to ~41 MPa was gathered from the data. The stress-
strain curves from the BF-CF material (Figure C.1) and the CF material (Figure C.2) are
shown below.
156

Figure C.1: Stress-strain curve for the control BF-CF coupons examined.

Figure C.2: Stress-strain curve for the control CF coupons examined.

157
The strain data corresponding to the stress is given in Table C.1. The strains for the
coupons were averaged and these strain values were used for the bi-axial tension
imparted to the samples in the test stands.
Table C.1: Data gathered from tensile tests and used to calculate the average strain on these
materials at ~41 MPa.
Stress Microstrain Stress Microstrain
BF-CF-T1 40.66 551 CF-T1 41.01 659
BF-CF-T2 41.37 554 CF-T2 41.72 686
BF-CF-T3 41.01 515 CF-T3 41.01 589
Averages 41.01 540 Averages 41.25 645

Abstract (if available)

Abstract NASA’s charter for exploration missions could take humans to deep space, asteroids, the Moon, and eventually Mars. Each of these missions necessitates a safe and productive place for the crew to live and work, namely deep space habitats. Long-term habitation requires the use of large structures which must withstand the environment for the duration of the extended-stay missions. ❧ Recently, fiber-reinforced composite materials have gained interest as a potential structural material for deep space and planetary habitats. These materials can provide weight savings, potentially enhanced radiation protection for the crew, and lend themselves to cutting-edge research when compared to existing metals. However, these materials have not been characterized for the space environment, and particularly the space radiation environment, which is known to cause damage to polymeric materials. Thus, this study focused on a lunar habitation element and the integrity of potential composite materials after exposure to a simulated long-term lunar radiation environment. ❧ Two aerospace composites of interest to NASA were chosen for the study and were subjected to a thirty-year simulated lunar radiation environment. The durability of the materials was investigated during the radiation exposure, and the material properties were characterized post-radiation exposure. Additionally, the effects of dose rate and synergistic tendencies between radiation and tension were examined. Finally, an exploration of the surface of the materials was completed to determine whether there were indications of accelerated aging as a result of the radiation exposure.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Laboratory investigations of the near surface plasma field and charging at the lunar terminator

PDF

Systems engineering and mission design of a lunar South Pole rover mission: a novel approach to the multidisciplinary design problem within a spacecraft systems engineering paradigm

PDF

Unreal engine testbed for computer vision of tall lunar tower assembly

PDF

Thermal and deformation analysis of multiphase sulfur concrete extrusion for planetary construction

PDF

Reduction of large set data transmission using algorithmically corrected model-based techniques for bandwidth efficiency

PDF

Extravehicular activity (EVA) emergency aid for extended planetary surface missions: through-the-spacesuit intravenous (IV) administration

PDF

Experimental and numerical investigations of charging interactions of a dusty surface in space plasma

PDF

An experimental investigation of cathode erosion in high current magnetoplasmadynamic arc discharges

PDF

Autonomous interplanetary constellation design

PDF

Incorporation of mission scenarios in deep space spacecraft design trades

PDF

Increased fidelity space weather data collection using a non-linear CubeSat network

PDF

A declarative design approach to modeling traditional and non-traditional space systems

PDF

Development of composite oriented strand board and structures

PDF

Three-dimensional exospheric hydrogen atom distributions obtained from observations of the geocorona in Lyman-alpha

PDF

Stochastic peridynamics and upscaling

PDF

Defect control in vacuum bag only processing of composite prepregs

PDF

Processing and properties of phenylethynyl-terminated PMDA-type asymmetric polyimide and composites

PDF

Material and process development and optimization for efficient manufacturing of polymer composites

PDF

Numerical and experimental investigations of dust-plasma-asteroid interactions

PDF

Developing efficient methods for the manufacture and analysis of composites structures

Asset Metadata

Creator Rojdev, Kristina (author)

Core Title Long term lunar radiation degradation of potential lunar habitat composite materials

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Astronautical Engineering

Publication Date 11/12/2012

Defense Date 10/11/2012

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag aging,composite,deep space,lunar habitat,OAI-PMH Harvest,proton radiation,scission

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Erwin, Daniel A. (committee chair), Atwell, William (committee member), Kunc, Joseph (committee member), Nutt, Steven R. (committee member)

Creator Email kristina.rojdev@gmail.com,rojdev@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-110150

Unique identifier UC11289217

Identifier usctheses-c3-110150 (legacy record id)

Legacy Identifier etd-RojdevKris-1283.pdf

Dmrecord 110150

Document Type Dissertation

Rights Rojdev, Kristina

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA