Surface Evaluation by X-Ray Photoelectron Spectroscopy of High Performance Polyimide Foams After Exposure to Oxygen Plasma

8 Pages·2002·0.45 MB·English

by  Melendez, Orlando

AIAA 2002-15_'_tAA 2002-1592
SURFACE EVALUATION BY X-RAY PHOTOELECTRON SPECTROSCOPY OF HIGH
PERFORMANCE POLYIMIDE FOAMS AFTER EXPOSURE TO OXYGEN PLASMA
Orlando Melendez a,Michael D. Hampton b,Martha K. Williams a,
Sylvia F. Brown a, Gordon L. Nelson c, Erik S. Weiser d
aNASA, Labs and Testbeds Division, YA-F, Kennedy Space Center, FL 32899, U.S.A.
Emaih Orlando.Melendez-1 @ksc.nasa.gov; Martha.Williams-1 @ksc.nasa.gov
bDepartment of Chemistry, University of Central Florida, Orlando, FL 32816
CFiorida Institute of Technology, 150 W. University Blvd., Melbourne, FL 32901, U.S.A.
bNASA, Mail Stop 226, Langley Research Center, Hampton, VA 23681, U.S.A.
ABSTRACT INTRODUCTION
Aromatic polyimides have been attractive in NASA LaRC has been developing a new and
the aerospace and electronics industries for diverse foam technology allowing for the next
applications such as cryogenic insulation, generation of polyimide foam materials that
flame retardant panels and structural sub- will be utilized for such things as cryogenic
insulation, flame retardant panels and
components. Newer to the arena of
structural sub-components. _-3 Depending
polyimides is the synthesis of polyimide
foams and their applications. In the present upon the application, foams can be designed
work, three different, closely related, and engineered with specific properties from
polyimide foams developed by NASA a large number of monomers and monomer
Langley Research Center (LaRC) are studied blends. These polyimide foams have a
by X-ray Photoelectron Spectroscopy (XPS) number of properties that make them highly
desirable materials for use on structures that
after exposure to radio frequency generated
Oxygen Plasma. Although polyimide films are exposed to extreme conditions. They are
exposure to atomic oxygen and plasma have strong, fire resistant, minimally outgassing,
been studied previously and reported, the data stable over a large temperature range,
relate to films and not foams. Foams have resistant to chemical attack, and can be made
much more surface area and thus present new in varying densities.
information to be explored. Understanding
In addressing extreme aerospace conditions,
degradation mechanisms and properties
versus structure, foam versus solid is of the question of polyimide foam stability and
interest and fundamental to the application degradation in Low Earth Orbit (LEO) is of
interest. Although polyimide films exposure
and protection of foams exposed to atomic
to atomic oxygen and plasma have been
oxygen in Low Earth Orbit (LEO).
studied previously and reported, 6'7'12the data
relate to films and not foams. The study of
polyimide foams versus film behavior is of
Copyright©2002 by the American Institute of scientific value. Before utilizing polyimide
Aeronautics and Astronautics, Inc. No copyright is foams in the aggressive environment of LEO,
asserted inthe United States under Title 17,U.S.Code.
it is important to understand and predict
The U.S. Government has a royalty-free license to
performance characteristics and the
exercise all rights under the copyright claimed herein
mechanisms of degradation. This information
for Governmental purposes. All other rights are
reserved bythe copyright owner. is also important to the protective measures
1
American Institute ofAeronautics and Astronautics
Copyright © 2002 by the American Institute ofAeronautics and Astronautics, Inc. No copyright isasserted in the United States under Title 17, U.S. Code,
The U.S. Government has aroyalty-free license toexercise all rights under the copyright claimed herein for Governmental purposes
All other rights are reserved by the copyright owner,
which might be required in the utilization of Table 1.Foam Materials Description
these materials. Foam, Description
Density
TEEK-HH ODPA/3,4'-ODA (4,4
The atmosphere at LEO altitudes has a
(0.082 g/cc) oxydiphthalic anhydride/3,4-
composition that is essentially the reverse of
oxydianiline)
that in the troposphere, 20% N2 and 80% 02.
TEEK-HL ODPA/3,4'-ODA (4,4
Without the overlying atmosphere to filter
(0.032 g/cc) oxydiphthalic anhydride/3,4-
short wavelength UV radiation (<243 nm), the
oxydianiline)
molecular oxygen present is largely TEEK-LL BTDA/4,4'-ODA (3,3,4,4-
photodissociated to atomic oxygen (AO). (0.032 g/cc) benzophenenone-tetracarboxylic
Atomic oxygen is a highly reactive substance dianhydride/4,4-oxydianiline)
and thus is prone to rapidly oxidize materials TEEK-CL BTDA/4,4'-DDSO2 (3,3,4,4-
exposed to it. Making the situation more (0.032 g/cc) benzophenenone-tetracarboxylic
extreme is the fact that structures in LEO are dianhydride/4,4-diaminodiphenyl
sulfone)
typically moving rapidly, as fast as 8 km/s, to
maintain the orbit. Moving at that speed, it is
typical for structures to collide with atomic
oxygen with an energy of as much as 5 eV
and to encounter 10]5 oxygen atoms per cm 2
of surface area per second. 4 In this study, an
oxygen plasma generator was utilized to / o o /n
produce an atmosphere of atomic oxygen that TEEK-HH and HL (ODPA/3,4"ODA)
would simulate the atmosphere of LEO.
9, o \
No ° o
Experimental
The synthesis of the precursor powders and
TEEK-LL (BTDA/4,4'-ODA)
the fabrication of the foams in this study were
reported previously. _-3 Three different
chemical formulations were used to fabricate
the polyimide foams, see Figure 1and Table 1
for abbreviations and densities. The first
letter after TEEK indicates the series and the
TEEK-CL (BTDA/4,4'-DDSO2)
second letter indicates the density.
Figure 1, Chemical structures offoams
The foams were supplied by LaRC and
Surface analyses of samples were performed
Unitika, Japan. The Kapton tape, 3M
TM
with a Kratos XSAM X-ray photoelectron
Mystic 7362, was obtained from NASA
(XPS) spectrometer. XPS is a surface
federal stock no. 5970-01-350-5283. The
analysis technique that looks at the upper
Kapton HN films, HN 100 (1 mil thick) and
TM
atomic layers of a solid surface. In XPS
HN 200 (2 mil thick) were provided by the
electrons are ejected from a sample surface
manufacturer.
with a particular binding energy
characteristic of the elements present. Shifts
in binding energy can be related to oxidation
or chemical states. Foam samples were
2
American Institute of Aeronautics and Astronautics
typically placed in the sampleinsertion The effective atomic oxygen flux distribution
chamberandallowedtopumpdownfor 1-3 over the placement of the samples (front,
days before analysis.Film samplesonly middle, and rear) were determined by using 4
required2-3hoursofpumpdowntime. samples of Kapton HN film mounted to the
TM
cleaned sample cups with a very small piece
TheoxygenplasmawasgeneratedwithaSP1 of carbon tape. The tape was situated such
PlasmaPrepII plasmaetcher. The Plasma that the sample covered and protected it. The
Prep II was operatedwith an oxygen Kapton HN film samples were placed in the
TM
atmosphereata feedpressureof 5 psi. The
4 positions within the etcher that were used
etcherwasusedat full powerfor cleaning for foam exposures and the flux determined
samplepansandfortheinitialfoamexposure. as above. When the samples were removed
This etcherutilizes radio frequency,13.56 from the etcher for weighing, they were kept
MHz, to exciteanoxygenatmosphereat a
in a nitrogen atmosphere to exclude
pressureof 2-5 torr into a plasma.This atmospheric oxygen and moisture
atmospherewill containsomeundissociated interference. Data indicated that the effective
oxygen molecules,oxygenions, electrons, atomic flux was distributed consistently
and oxygen atoms. For this work, the
throughout the chamber.
atmospherewas consideredto consistof
oxygenatomsonly and the terms,oxygen The foam samples were cut into 2x2 cm
plasma and atomic oxygen will be used squares. Foams were placed in a reduced
interchangeably. pressure environment to reduce the amount
of absorbed moisture before initial weighing.
The effective atomic oxygen flux was
After obtaining baseline XPS data, they were
determined,using the following procedure,
exposed to oxygen plasma at 25% power for
ASTM E2089-00,StandardPracticesfor
30 min increments and weighed after each
Ground Laboratory Atomic Oxygen
exposure. When significant weight loss
InteractionEvaluationof Materialsfor Space
occurred, samples were re-analyzed with
Applicationsat 100%,50%and25%power
XPS, and then oxygen plasma exposure
levels5. SamplesofKaptonTM tape, roughly 1- continued.
2cm square, were placed in cleaned aluminum
sample pans, mastic side down, and weighed.
Samples were then individually placed in the RESULTS AND DISCUSSION
etcher and exposed to oxygen plasma for 30
min periods at the desired power level. After Figure 3 indicates the effective oxygen flux
each 30 min period the samples were versus time at 25% power for the Kapton
TM
weighed. Analyses were run in triplicate at witness coupon ran with the foam samples.
each power level. The effective atomic The average effective oxygen flux is
oxygen flux, F, was then calculated according 1.14x 1016 O atom/cm2.s.
to the following equation in which AM is the
mass loss of the sample in grams, 9 is the In addressing the interactions and reactivity of
density in g/cm 3, A is the sample surface area these foams with atomic oxygen it is to be
in the cmZ, E is the in-space erosion rate of noted that although the structures are very
Kapton 3x10 -24 cm3/Oxygen atom and t is similar (see Figure 1), there are some
TM,
time in seconds: important differences. Both the C and the L
series foams have a carbonyl group linkage in
F = AM/(pA E t) the dianhydride groups, while the H series
3
American Institute of Aeronautics and Astronautics
foamshaveanetherlinkage. In theL series and their chemistries not containing an S02
foams,thediamineportionhasoxygen para- linkage, only the H and L series were chosen
bonded in both benzene rings, while in the H to be studied further by High Resolution XPS
series foams, the right most benzene ring is to investigate the relationships of density and
meta bonded. In the C series foams the chemistry to atomic oxygen effects. High
diamine has an SO2 linkage. In previous Resolution XPS is capable of
thermal degradation studies of polyimide separating/resolving different oxidation states
films, the diamine structure (Figure 2) appears or binding energies.
to have a greater influence on the stability,
with the ranking being SO>SO2>CH2>O for
X. I°
0.012]7 --I--Kapton [ •
0.01 -_ --I--HL [__/z-____
w ]] _HH I
NH2-_X-_ NH2 000,1 ../"
Figure 2, Generic Chemical Structure for 0. 4
0.00844 @LL
Diamine
0.002 _
0_
200
0 50 Time (i_n_anutes) 150
2E+16
Figure 4, Mass Loss of TEEK HL, HH, LL &
_ 1.5E+l
6 Kapton rMTape
O IE+16
<
g 5E+15 High Resolution XPS Data
O 0
2000 4000 6000 8000 10000 A high-resolution spectra baseline was
Time (seconds) obtained for each of the three polymers, HH,
HL and LL before and after plasma exposure
Figure 3, Atomic Oxygen flux versus time at
(see Figure 5, (a)---_(e) for Cls peaks). The
25% power for Kapton rMWitness Coupon
spectra were acquired in the small aperture
XPS mode for higher resolution using
magnesium X-rays at 12 KV and 12 mA.
Preliminary analyses indicate that the samples
Following the initial baseline characterization
had a small mass loss for the first hour of
by XPS the samples were exposed to oxygen
oxygen plasma exposure and then began a
plasma for 2.5 hours. A second XPS spectrum
more rapid weight loss. The CL sample with
was obtained for each sample after the plasma
the SO2 linkage in the diamine has the
exposure. The XPS results are summarized in
greatest resistance to weight loss.
Table 2. The carbon peak shows two main
features, one is a peak with binding energy
Figure 4 includes the mass loss of the HL,
(BE) around 284.8 eV (designated C-1 in
HH, LL series, plus Kapton The mass loss
TM.
Table 1), and a second peak with binding
of the HL series is the greatest, followed by
energy around 288 eV (designated C-2 in
HH, although that is of a higher density than
Table 1). The peak C-1 at 284.8 eV results
the LL series. Here the mass loss appears to
from signals due to C-C, C-N and C-O bonds.
be more dependent upon chemistry versus
density. Because of their different densities
4
American Institute of Aeronautics and Astronautics
Table 2, XPS Data
Carbon Oxygen
Sample After Plasma After Plasma
I.D. Baseline Baseline
Ex 9osure Exposure
Peak Peak
BE Atomic BE Atomic BE Atomic BE Atomic
(eV) % (eV) % (eV) % (eV) %
C ls 284.8 70.11 284.8 55.75 O ls 533 29.89 532.3 44.25
HH C-1 284.6 43 284.7 81.3
C-2 286.6 57 288.7 18.69
C ls 284.7 76.46 284.6 48.7 O ls 532 20.43 532.2 48.7
i -!iiii i'i19))i(9_i /2"8..4..-...6? :s...................................... iiiii iiiiiiiiiiii
C-2 288.1 8.81 288 32.2
CIs 284.8 74 284.9 56.15 0 Is 532 24 532.7 43.85
LL C-1 284.8 92.7 284.7 61.3
C-2 288 7.22 288 38.7
*atomic %does not include nitrogen
5"C-1and C-2 represent 100% ofCIs
The peak C-2 at 288 eV represents the al.6 reported CO and CO2 as the main volatile
carbonyl group C=O.ll Of the three samples,
species during atomic oxygen attack.
sample HH shows the highest initial amount Although there is an overall decrease in
of oxygen, almost 30% compared to 20.4% carbon (see Figure 5, charts after plasma
for HL and 24% for LL (see also Figure 5).
exposure), samples HL and LL showed a
The higher initial oxygen content for HH is significant increase in C-2 corresponding to
related to the high signal due to the HH carbonyl, C=O relative to the peak C-1. This
carbon peak C-2. This is the peak from the is consistent with results reported in several
carbonyl C=O. HH has a larger carbonyl papers. A study by Cross et. al.7 of the
signal than HL or LL even though sample LL interaction of atomic oxygen with a Kapton
TM
has one more carbonyl group. The reason for film, found H2, H20 CO, and CO2 in the gas
the larger carbonyl signal for HH is that the phase reaction products. They also found
density of HH is two and a half times the more carbonyl-bonded oxygen. The increase
density of HL and LL. This implies that a in carbonyl was also observed by Lu et. al.8
similar unit of volume will contain one basic
after using Pulsed UV laser irradiation on a
unit of LL with 5 carbonyl groups. The same
polyimide film. Kitching et. al.9 in an XPS
unit of volume will contain two and a half
study of biomedical polyurethane modified by
unit of HH with a total of 10 carbonyi groups
oxygen plasma, also observed an increase in
carbonyl following the plasma treatment.
All three samples showed significant increases
in oxygen with a comparable reduction in
Sample HH also showed a significant increase
carbon. These results indicate that the sample
in oxygen with a comparable overall
is loosing carbon while at the same time some reduction in carbon. However, contrary to
oxygen is been incorporated to the molecule.
samples HL and LL and results for Kapton in
In a series of molecular beam-surface
the literature, the carbonyl groups were
scattering experiments on Kapton TM, Minton et reduced, (C-2 less than C-1 after exposure).
Again, it seems that the high density of HH
plays an important role. A higher density
5
American Institute ofAeronautics and Astronautics
implies that there are more C=O bonds
exposedto theplasmathanin HL andLL. _'1, {'1
Theresultsshowinga reductionin carbonyl f"_
/
seemsto indicatethattheplasmais reacting
with this grouppreferentiallyoveratomsin
thering structureresultingin somevolatile u,
products. The oppositeseemsto occurin
samplesHL andLL wheretheattackby the
plasmaseemstotakeplaceatsitesonthering
resultingin the additionof oxygenat those
locations.
_2 zxs Z_4 "__
B,_llng In¢_¢5<.x
(c) TEEK-LL Before Plasma
III (!
4O j:"/ t
/
i
llJ_ / /r !L
ul.
/ rI
/
_:_. i ¸:
(a) TEEK-HH Before Plasma
(d) TEEK-LL After Plasma
_;], cl
/
/ ',
i ,
/ 3Z. / ;
/-f
"lC _/" / ...... _\'t: " ......
/-.J/ \
(b) TEEK-HH After Plasma
Figure 5, XPS Spectra, Before and After (e) TEEK-HL Before Plasma
Plasma Exposure for HH (a-b), LL (c-d), & *After Plasma exposure chart follows trend
observed inLL series
HL (e, before)
6
American Institute ofAeronautics and Astronautics
CONCLUSIONS manufacturer's names, trademarks, or other
product identification in this document does
As indicated in previous reports, 2'3 data not constitute an endorsement or approval of
presented validate that newly developed the use of such commercial products.
polyimide foams are high performance
REFERENCES
polymers in mechanical, physical and thermal
properties. In those reports, the TEEK-C 2'3
Hou, T.H.; Weiser, E.S.; Siochi, E.J.; St.
series proved to be more thermally stable, and 1
Clair, T.L. and Grimsley, B.W.,
in preliminary data discussed in this report,
the C series also appear to have the greatest Processing and Properties of Polyimide
Foam, 44th International SAMPE
resistance to atomic oxygen with the least
Symposium, 1999, 1792-1806.
amount of mass loss on exposure.
The mass loss data indicates that chemical . Weiser, E.S.; Johnson, T.F.; St. Clair,
T.L.; Echigo, Y.; Kaneshiro, H.;
structure and then density appear to play the
Grimsley, B., High Temperature
greatest role on atomic oxygen resistance for
the HH, HL and LL series. The XPS data Polyimide Foams for Aerospace Vehicles,
indicates an overall oxidation of the foams. Journal of High Performance Polymers,
March 2000, 12, 1, 1-12.
The data presented on the HL and LL foams
showing an increase in carbonyl after atomic
Williams, Martha K., et al., High
oxygen exposure correlate with the data .
previously reported on polyimide films, v'9 Performance Polyimide Foams, G. L.
Nelson and C. A. Wilkie, eds., Fire and
The higher density HH series showed a
decrease in the carbonyl group. This seems to Polymers: Materials and Solutions for
Hazard Prevention, ACS Symposium
indicate that the plasma is reacting with this
Series 797, American Chemical
group preferentially over atoms in the ring
structure resulting in some volatile products. Society/Oxford Press, 2001, 49-62.
Caledonia, G.E.; Krech, H.R.; Studies of
.
ACKNOWLEDGMENTS the Interaction of 8 km/s Oxygen Atoms,
Materials Degradation in Low Earth
The authors would like to thank NASA, Orbit (LEO), Edited by Srinivasan, V.;
Banks, B.A.; 1990.
Kennedy Space Center, and the National
Space Club/Hugh L. Dryden Memorial
ASTM E 2089-00, Standard Practices for
Fellowship for doctoral study support of Ms. .
Williams. Laboratory Atomic Oxygen Interaction
Evaluation of Materials for Space
This document was prepared under the Applications.
sponsorship of the National Aeronautics and
Minton, T.K.; Seale, J.W.; Garton, D.J.;
Space Administration. Neither the United .
States Government nor any person acting on Zhang, J.; Mechanisms of Polymer
behalf of the United States Government Erosion in Low Earth Orbit: Implications
For Ground-based Atomic Oxygen
assumes any liability resulting from the use of
the information contained in this document, or Testing, 44th International SAMPE
warrants that such use will be free from Symposium, 1999, 1051 -1063.
privately owned rights. The citation of
7
American Institute of Aeronautics and Astronautics
7. Cross,J.B.,S.L.,Koontz;Gregory,J.C.,
and Edgell,M.J., Hyperthermal Atomic
Oxygen Reactions with Kapton and
Polyethylene, Materials Degradation in
Low Earth Orbit (LEO), Edited by
Srinivasan, V.; Banks, B.A.; 1990, 1-13.
8. Lu, Q. H.; Li, M.; Zhu, Z.; Wang, Z.;
Polymide Surface Modification by Pulsed
Ultraviolet Laser Irradiation with Low
Fluence, Journal of Applied Polymer
Science, 2001, Vol. 82, 2739-2743.
9. Kitching, K.J., Wilson, D.J.; Doherty,
P.J.; Williams, R.L.; An X-ray
Photoelectron Spectroscopy Study of
Biomedical Polyurethane Modified Using
Low-Power Plasma, Journal of X-ray
Science and Technology, 2001, V.9, 77-
83.
10. Cella, J.A.; Degradation and Stability of
Polymides (Polyimides), Marcel Dekker,
Inc., New York., 1996, 343-365.
l l. Briggs, D.; Beamson, G.; Primary and
Secondary Oxygen-Induced Cls Binding
Engergy Shifts in X-ray Photoelectron
Spectroscopy of Polymers, Anal. Chem.
1992, 64, 1729-1736
12. Snyder, A.; Investigation of Atomic
Oxygen Erosion of Polyimide Kapton H
Exposed to a Plasma Asher Environment,
44th International SAMPE Symposium,
1999, 1321-1326.
8
American Institute ofAeronautics and Astronautics