You are on page 1of 12

High Perform. Polym. 12 (2000) 1–12.

Printed in the UK PII: S0954-0083(00)10143-6

Polyimide foams for aerospace vehicles∗

Erik S Weiser†, Theodore F Johnson†, Terry L St Clair†, Yoshiaki Echigo‡,


Hisayasu Kaneshiro‡ and Brian W Grimsley§
† 6A West Taylor Street M/S 226, NASA Langley Research Center, Hampton, VA 23681, USA
‡ Unitika Ltd, Kyoto, Japan 611
§ Old Dominion University, Norfolk, VA 23529, USA

Received 1 November 1999, accepted for publication 7 December 1999

Abstract. Due to a demand by the aerospace industry, NASA has begun developing the next
generation of polyimide foams which could be utilized to reduce vehicle weight for the X-33 and
Reusable Launch Vehicle (RLV) programmes. The activity at NASA Langley Research Center
focuses on developing polyimide foam and foam structures which are made using monomeric
solutions or salt solutions formed from the reaction of a dianhydride and diamine dissolved in a
mixture of foaming agents and alkyl alcohols. This process can produce polyimide foams with
varying properties from a large number of monomers and monomer blends. The specific densities
of these foams can range from 0.008 g cc−1 to 0.32 g cc−1 .
Polyimide foams at densities of 0.032 g cc−1 and 0.08 g cc−1 were tested for a wide range
of physical properties. The foams demonstrated excellent thermal stability at 321 ◦ C, a good
thermal conductivity at 25 ◦ C of 0.03 W m−1 K−1 , compressive strengths as high as 0.84 MPa at
10% deflection and a limiting oxygen index of 51%. Thermomechanical cyclic testing was also
performed on these materials for 50 cycles at temperatures from −253 ◦ C to 204 ◦ C. The foams
survived the cyclic testing without debonding or cracking. Thermal forming of the 0.032 g cc−1
foam was performed and a minimum radius curvature of 0.0711 m was achieved. The foams
exhibited excellent properties overall and are shown to be viable for use as cryogenic insulation on
the next generation RLV.

1. Introduction

Foamed plastics, also known as cellular plastics or plastic foams, have been important since
primitive man began to use wood, a cellular form of the polymer cellulose [1]. Cellulose is
derived from the Latin word cellula, which means very small cell or room, and the polymer in its
natural form exists mostly in its cellular form as seen in wood and straw. The first commercial
cellular polymer was sponge rubber, introduced between 1910 and 1920 [1]. Cellular polymers,
or foams, were first introduced in the United States and Europe in the mid-to-late 1930s [2].
The term polymer foams or cellular polymers refers to a two-phase gas–solid system in which
the solid polymer is continuous and the gaseous cells are dispersed throughout the solid [3].
Polymeric materials are foamed to meet various application needs, such as weight reduction,
insulation, buoyancy, energy dissipation, conveyance and comfort. These polymeric foams
can be produced by several different methods including extrusion, compression moulding,
injection moulding, reaction injection moulding and solid-state methods. In the late 1960s, a
new type of polymeric foam was developed by Monsanto and DuPont based on a polyimide
precursor. These new polyimide foams showed improved properties when compared to
∗ This paper is declared a work of the US Government and is not subject to copyright protection in the United States.

© 2000 US Government 1

Downloaded from hip.sagepub.com at PENNSYLVANIA STATE UNIV on May 17, 2016


2 E S Weiser et al

traditional polymer foams, polyurethane or StyrofoamTM , in areas such as thermal stability,


non-flammability, radiation resistance, improved toughness and reduced smoke and toxic fume
generation [2].
Recently, the aerospace industry has had a growing need for high-performance polymer
foams for applications such as cryogenic insulation, fireproofing, energy absorbers, etc, and has
turned to polyimide foams to meet these needs. For example, the X-33, a single stage to orbit
(SSTO) reusable launch vehicle (RLV) requires a lightweight foam which retains its structural
integrity at temperatures ranging from −250 ◦ C to 250 ◦ C [4]. The High Speed Research
Programme has a need for low density core-splice material with a density of 0.50 g cc−1 and
a potting material with a density of 0.56 g cc−1 [5].
Commercially available polyimide foams are not widely produced. However, with an
increase in interest for next generation launch vehicles, they will become more popular due
to their many beneficial physical properties such as low density, low moisture adsorption and
high strength. Many foams are suitable for the space transportation industry, but they do not
meet all of the requirements for surviving extreme aerospace environments. Some foams with
densities as low as 0.066 g cc−1 are capable of co-curing at temperatures as high as 182 ◦ C.
Other foams can be resin transfer moulded at temperatures up to 180 ◦ C, with a modulus of
elasticity of 0.19 MPa, tensile strength of 3.7 MPa and compressive strength of 3.6 MPa.
However, these foams have a coefficient of thermal expansion comparable to aluminum and
can crack at liquid hydrogen (LH2 ) temperatures (−253 ◦ C). Other foams are able to withstand
temperatures from −253 ◦ C to 200 ◦ C, but they do not have the required structural integrity.
The activity at NASA Langley Research Center (LaRC) focuses on developing low-
density polyimide foam and foam structures which are made using monomeric solutions or
salt solutions formed from the reaction of a dianhydride and diamine dissolved in a mixture
of foaming agents and alkyl alcohol at room temperature. This process produces foams
from monomers such as 4,40 -oxydiphthalic anhydride (ODPA) or 3,30 ,4,40 -benzophenone-
tetracarboxylic dianhydride (BTDA) and 3,40 -oxydianiline (ODA) or 3,30 -diaminodiphenyl
sulfone (DDS), etc. Monomer blends may be used to make a variety of polyimide foams
with varying properties. Foams can be fabricated to specific densities from 0.008 g cc−1 to
0.32 g cc−1 .

Dianhydride Alcohol Dialkylester-Diacid


Salt with Agent
Agent
Alkyl Alcohol
Diamine
Polyimide Stable Salt at
Foam Ambient Conditions
Agent, Water, & (Stablized by the Agent)
Alcohol By-Products

Figure 1. The TEEK polyimide foam imidization reaction.

Downloaded from hip.sagepub.com at PENNSYLVANIA STATE UNIV on May 17, 2016


Polyimide foams for aerospace vehicles 3

Polyimide foams with a density of 0.032 g cc−1 and 0.08 g cc−1 were analysed for a
wide range of physical properties. Tests were based on the properties required to qualify a
material for use as a cryogenic insulator or structural foam on next generation RLVs. The
foams were tested at cryogenic (−253 ◦ C to −193 ◦ C) and flight temperatures (177 ◦ –260 ◦ C)
for compressive and tensile strengths, thermal conductivity, liquid oxygen (LOx) compatibility,
thermal and hydrolytic stability, and uniaxial tension. Thermal forming tests were also
conducted on the 0.032 g cc−1 polyimide foam to determine the maximum radius of curvature
the foam could sustain without any dimensional change.
The significance of structural polyimide foams can be realized with a reduction in the
overall weight of a RLV. Due to many polyimides’ high operating temperature (∼260 ◦ C),
structural polyimide foams can potentially reduce the amount of the thermal protection system
(TPS) integration structure that is required on an RLV and the total amount of TPS required.
The reduction in the TPS integration structure and TPS would reduce the total weight and
perhaps cost to build an RLV. This would allow the maximum payload weight to increase and
make the vehicle more efficient for commercial applications.

2. Material synthesis†

NASA LaRC and Unitika Ltd have jointly developed a salt-like polyimide precursor powder
which exhibits excellent foaming capabilities. Over 25 different polyimide foams have been
fabricated utilizing monomeric solutions or salt-like solutions [6]. The salt-like foam precursor
was synthesized by mixing monomer reactants of a dianhydride with a foaming agent in
methanol (MeOH) at room temperature. This material was converted into dimethyl ester after
3 h of mixing at 60 ◦ C. The resulting solution was then mixed with calculated amounts of
diamine and stirred for 2 h to yield a homogeneous polyimide precursor solution. Figure 1
shows a flow diagram of the imidization process necessary to form a polyimide precursor
solution. A fine powder was obtained by drying off excess solvent from the solution. The
polyimides LaRC-IA (ODPA/3,40 ODA), BTDA/4,40 ODA and BTDA/4,40 DDS were utilized
[7] for these experiments. The trade names TEEK-HH and -HL are used for the polymer
LaRC-IA, TEEK-LL for BTDA/4,40 ODA, and TEEK-CL for BTDA/4,40 DDS. Figures 2–4
show the chemical structures for these polyimides.

O O
O
N N O

O O n

Figure 2. Chemical structure for LaRC-IA (ODPA/3,40 ODA).

O O O
C
N N O

O O
n

Figure 3. Chemical structure for BTDA/4,40 ODA.

† The use of trademarks or names of manufacturers in this report is for accurate reporting and does not constitute an
official endorsement, either expressed or implied, of such products or manufacturers by NASA.

Downloaded from hip.sagepub.com at PENNSYLVANIA STATE UNIV on May 17, 2016


4 E S Weiser et al

O O O
C
N N SO 2

O O
n

Figure 4. Chemical structure for BTDA/4,40 DDS.

Heat Plate

Upper Graphite Plate

Mold Chamber
Expandable Powder
Lower Graphite Plate

Heat Plate

Figure 5. Schematic diagram of the mould concept for TEEK foaming.

3. Experimental details

Four different polyimide foams were fabricated for scale-up and testing. The testing criteria
were designed to meet the necessary requirements for use on a next generation launch vehicle.
These polyimide foams were tested for thermal, mechanical and flame resistant properties.
Experiments such as the thermal conductivity, maximum use temperature, glass transition
temperature and thermal stability were conducted to determine the ability of these new
polyimide foams to perform in the hostile environment of space. Mechanical testing included
compressive strength at room, elevated and cryogenic temperatures and tensile strength at room
temperature. Flammability experiments were performed to test the material’s resistance to an
oxygen-rich environment and the foam’s resistance to prolonged exposure to extreme burn
temperatures. Thermomechanical testing was performed to determine each material’s ability
to withstand flight conditions experienced during ground hold, launch and re-entry. Thermal
forming tests were performed to ensure that these novel polyimide foams could be shaped to
the radius of a cryogenic tank.

3.1. Foam fabrication


The amount of polyimide foam precursor powder necessary to completely foam a desired
volume was placed in a chamber. The mould, shown in figure 5, composed of a metal chest
with graphite plates on the top and bottom, was heated to 140 ◦ C for 60 min using heat plates
on the top and bottom. The mould was then rapidly transferred to a nitrogen convection oven
set at 300 ◦ C and held at 300 ◦ C for 60 min. The mould was then cooled to room temperature.
At this point the foam was post cured for several hours at 200 ◦ C to remove all trace volatiles.
The foam was removed from the mould and was ready for use. Figure 6 shows the cure cycle
for the TEEK foams. Open cell content tests were performed on several of the foam specimens
using a Quantachrome Incorporated Ultrafoam 1000.

3.2. Thermal properties


Thermal testing was performed on three different polyimide foams. A Shimadzu DSC-50
thermal analyser was used to measure the glass transition temperature (Tg ). A heating rate of

Downloaded from hip.sagepub.com at PENNSYLVANIA STATE UNIV on May 17, 2016


Polyimide foams for aerospace vehicles 5

Foaming Step Imidization Step


300
60 Minute Hold

Temperature (°C)
250

200

150
60 minute Hold
100

50 Air Nitrogen

25 50 75 100 125 150 175


Time (minutes)

Figure 6. Cure cycle for the TEEK foams.

20 ◦ C min−1 was used and the Tg was taken at the inflection point of the endotherm. The weight-
loss characteristics of the cured foam specimens were measured using a Seiko TG/DTA 220
thermogravimetric analyser (TGA) at a heat-up rate of 2.5 ◦ C min−1 in flowing (40 ml min−1 )
air. Thermal conductivity was measured for all specimens using ASTM standard C-518 at
temperatures from −100 ◦ C to 350 ◦ C.

3.3. Mechanical properties

Several different polyimide foams were tested for flatwise tensile, tensile (dogbone) and
compressive strength at room temperature using ASTM standard D-1623-C, D-638-97 and
D-3574 (E), respectively. The tensile specimens were cut to dimensions of 0.051 m×0.051 m×
0.025 m and bonded to support blocks utilizing a low-temperature epoxy adhesive. Specimens
were placed in a 9000 kg Instron test stand and loaded under tension to failure. Compression
specimens were also machined to 0.051 m × 0.051 m × 0.025 m dimensions and placed in a
9000 kg Instron testing machine. Compression specimens were tested at −253 ◦ C, 25 ◦ C, and
177 ◦ C. At 50% of the initial thickness (50% deflection), the load and stress were determined.
Load and stress values at 10% deflection were determined from the graph of the deflection
against load for the 50% deflection test.

3.4. Flammability

Several tests were performed to determine the feasibility of using these materials within
an oxygen-rich, flammable environment. The first test performed was the limiting oxygen
index test which determines a materials resistance to instantaneous combustion in an oxygen-
rich environment. The test was performed using ASTM standard D-2863. The second test
performed was the vertical burn test, FAR 25.853(a), which tests a materials resistance to smoke
generation and burning. The specimens had dimensions of 0.051 m × 0.305 m × 0.025 m and
were placed in a fixture vertically above an igniter. The igniter was lit and burned for 25 ± 5 s
at a temperature of 1093 ◦ C. The burn length must be less than 0.152 m for the specimen to
pass the test.

Downloaded from hip.sagepub.com at PENNSYLVANIA STATE UNIV on May 17, 2016


6 E S Weiser et al

Figure 7. Location of the cryogenic propellant tanks for Lockheed–Martin’s X-33 and the simulated
circumferential hoop load.

Cross-Section
of fixture NRA 8-21 Panel #2
Cryogenic foam
LaRC TEEK-HHª
Structural Fiber optic sensor
wall Gr-Ep (IM7/977-2)
Heater
LN2 chimney
inlet
Convective
LHe/LN2 heater
inlet chamber
Fan
Cryogenic
chambers
LN2
Cryogenic
inlet
insulation
Supporting
GN2
framework
inlet
Load
introduction Finned strip
structure heater Strain gage lead
Thermocouple

Figure 8. Schematic diagram and an assembled view of the test fixture for the uniaxial tension
test.

3.5. Uniaxial tension tests


Combined cyclic thermal and mechanical tests of cryogenic tank wall concepts simulating
ring and stringer tanks were performed on two flat 0.30 m × 0.60 m panel specimens. A flat
specimen closely approximates a tank wall due to its large radius. These tests were developed
from earlier tests of a cryogenic insulation tile developed for the advanced launch system (ALS)
[8]. The purpose of the tests was to simulate both the thermal and mechanical loads experienced
in an RLV mission from launch to orbit and to re-entry. The cryogenic tanks in an RLV, shown
in figure 7, must endure biaxial tension loads associated with internal pressurization as well as
maximum thermal and mechanical flight loads. However, for these tests, the only mechanical
load applied was a uniaxial tension load simulating circumferential pressure loading, shown in
figure 7. This combined cyclic, thermal-mechanical test verified the durability of the cryogenic

Downloaded from hip.sagepub.com at PENNSYLVANIA STATE UNIV on May 17, 2016


Polyimide foams for aerospace vehicles 7

Heater side Cryogenic side Tension load

250
200 1.0
150

Temperature (ÞC)

Normalized uniaxial
100 0.8

tension load
50
0 0.6
-50
-100 0.4
-150
-200 0.2
-250
-300 0.0
0

200

400

600

800

1000

1200

1400

1600

1800

2000

2200
Time (Sec.)

Figure 9. Typical thermal-mechanical cycle for a LH tank specimen in the uniaxial tension test.

Vacuum Bag
TEEK-LL Foam
Metal Plate
Metal Cylinder

Vacuum Bag TEEK-LL Foam

Metal
Metal Plate Cylinder

Figure 10. Vacuum-bagged specimen before and after thermal forming was performed.

insulation when subjected to cyclic mission-profile conditions, the bond line integrity between
the cryogenic insulation and the structure and the performance of cryogenic tank fabrication
technologies on a small scale.
Specialized test fixtures [9] have been developed that allow a test specimen to be thermally
cycled according to pre-defined temperature profiles between a minimum temperature of
−253 ◦ C and a maximum temperature of 372 ◦ C. Figure 8 shows a specimen mounted in
the fixture with the cryogenic chambers mounted on the surface of the inner tank wall of
the specimen and a convective heating chamber adjacent to the external surface of the foam
insulation. Tension load and temperatures for the cryogenic and high-temperature chambers
are independently controlled in a test cycle. A typical cycle lasts 30–50 min. An example of
a thermal-mechanical load profile for a LH2 tank specimen is shown in figure 9. This figure
displays the complex tension loading and temperature profiles on the high-temperature side
and cryogenic side of the panel over a period of time.

Downloaded from hip.sagepub.com at PENNSYLVANIA STATE UNIV on May 17, 2016


8 E S Weiser et al

3.6. Thermal forming


A 0.095 m × 0.229 m × 0.0254 m TEEK-LL foam specimen was vacuum bagged over a
0.0445 m radius cylinder, shown in figure 10. The vacuum-bagged specimen was placed in an
oven and heated to 188 ◦ C where 500 Hg of vacuum was applied to shape the foam specimen
around the metal cylinder. The specimen was held under vacuum for 1 h at 188 ◦ C and then
cooled.

4. Results

4.1. Foam fabrication


Visual inspection of all of the foam samples indicated excellent quality. Cross sectional views
indicated a nearly uniform distribution of cellular structure. The centreline cellular structure
appeared to be of a slightly lower density on a minority of the samples tested. Foam specimens
at a density of 0.032 g cc−1 showed the most non-uniformity of the two densities tested.
Close-cell content measurements of the polyimide foam indicated that approximately 68% of
the volume was open-celled.

4.2. Thermal properties


All of the polymers used in this study were aromatic polyimides which exhibit excellent stability
up to 320 ◦ C. The specimens’ glass transition temperatures shown in table 1 ranged from 237 ◦ C
to 321 ◦ C. The TEEK-HH and -HL foams exhibited the lowest glass transition temperatures
(237 ◦ C), while the TEEK-LL (300 ◦ C) and TEEK-CL (321 ◦ C) had higher Tg ’s. The percent
weight loss at 204 ◦ C was measured on the samples as shown in table 1. Both TEEK-HH and
-HL showed good thermal stability with weight loss values of 0.6% and 1.07%, respectively.
TEEK-LL had the best thermal stability with no appreciable weight loss at 204 ◦ C. TEEK-CL
had excellent thermal stability with a weight loss of 0.5%. However, it had a significant amount
of moisture absorption (∼1.5%), while the TEEK-HH, -HL, and -LL had little water absorption
(∼0.25%). The water absorption was subtracted from the reported TGA data. Figure 11 shows
thermal conductivity data for the polyimide foams from −100 ◦ C to 350 ◦ C.
The conductivity data for the TEEK-HH, TEEK-HL and TEEK-CL foams exhibit good
values at room temperature and show decreasing values at sub-ambient temperatures, which
would indicate that these foams will be good cryogenic insulators. The cryogenic temperatures
are −253 ◦ C for hydrogen and −193 ◦ C for LOx. The outer surface of the exterior insulated

Table 1. Thermal properties for TEEK polyimide foams.

Property Test method TEEK-HH TEEK-HL TEEK-LL TEEK-CL

Density ASTM 0.08 g cc−1 0.032 g cc−1 0.032 g cc−1 0.032 g cc−1
D-3574 (A)
Thermal TGA 0.6% 1.07% 0.0% 0.5%
stability Percent weight loss
at 204 ◦ C
Glass DSC 237 ◦ C 237 ◦ C 300 ◦ C 321 ◦ C
transition
temperature
Thermal ASTM C-518 0.031 W m−1 K −1 0.029 W m−1 K−1 0.044 W m−1 K−1 0.03 W m−1 K −1
conductivity at 24 ◦ C

Downloaded from hip.sagepub.com at PENNSYLVANIA STATE UNIV on May 17, 2016


Polyimide foams for aerospace vehicles 9

0.00125

Thermal Conductivity (W/cm*K)


0.001

0.00075 TEEK-HH
TEEK-HL
0.0005 TEEK-CL

0.00025

0
-200 -100 0 100 200 300 400
Temperature (°C)

Figure 11. Thermal conductivity data for the TEEK foams.

cryogenic tanks must be at a temperature high enough to prevent air liquefaction, frost build-up
or the condensation of moisture. All of the foams have conductivity data which demonstrate
that the tank would be sufficiently insulated to prevent frost build-up and air liquefaction with a
relatively small thickness (<0.025 m). The elevated thermal conductivities are not as efficient
as the room temperature and sub-ambient values. This indicates that these foams are better
suited for cryogenic insulation than elevated temperature insulation. However, in the case of
next generation launch vehicles, TPS will be the main high-temperature insulator and the foam
will provide adequate protection of the cryotank wall.

4.3. Mechanical properties


Mechanical testing was performed to determine whether these new polyimide foams could be
used as a structural material as well as thermal insulation. In many new design concepts for
future RLVs, the cryogenic insulation will play a larger role if the material is a load carrying
member. Flatwise tension and compression testing was performed to determine the feasibility
of these materials as structural insulation. Table 2 shows the compression and flatwise tension
results. During the flatwise tension test, shear induced failure had to be overcome by insuring
that the specimen was loaded in pure tension. Several specimens were prepared and tested.
Only the specimens at the higher density of 0.08 g cc−1 did not exhibit shear induced failure.
These specimens had a mean tensile strength of 1.40 MPa. All of the specimens at the lower
density, 0.032 g cc−1 , failed due to shearing. This can be attributed to the formation of
elongated cells, which, when under load, tend to orient in an off-axis direction and fail in shear.
Additional testing was performed utilizing a dogbone specimen according to ASTM D-638-97
and tensile strengths could be determined for all the specimens. The TEEK-HH, -HL, -LL and
-CL had tensile strengths of 1.68 MPa, 0.28 MPa, 0.26 MPa and 0.09 MPa, respectively. The
compression specimens performed adequately for all three polyimide foams. Measurements
were taken for both 50% deflection and 10% deflection, which was measured by reading the
appropriate value of the load against deflection graph. The TEEK-HH foam had a compression
strength of 0.84 MPa at a 10% deflection and a modulus of 6.13 MPa. The TEEK-HL, -LL,
and -CL foams had values between 0.098 MPa and 0.30 MPa. Compressive strength was
also measured at 177 ◦ C and −253 ◦ C. The compressive strength of the 0.08 g cc−1 foam

Downloaded from hip.sagepub.com at PENNSYLVANIA STATE UNIV on May 17, 2016


10 E S Weiser et al
Table 2. Mechanical properties of the TEEK polyimide foams.

Property Test method TEEK-HH TEEK-HL TEEK-LL TEEK-CL

Density ASTM 0.08 g cc−1 0.032 g cc−1 0.032 g cc−1 0.032 g cc−1
D-3574 (A)
Tensile ASTM 1.68 MPa 0.28 MPa 0.26 MPa 0.09 MPa
strength D-638-97
Compressive ASTM 0.84 MPa at 0.19 MPa at 0.30 MPa at 0.098 MPa at
strength D-3574 (C) 10% deflection 10% deflection 10% deflection 10% deflection
Compressive ASTM 6.13 MPa 3.89 MPa 11.03 MPa In test
modulus D-3574 (C)
Compressive ASTM 0.31 MPa 0.06–0.10 MPa 0.06–0.09 MPa In test
strength at D-3574 (C) at 10% deflection at 10% deflection
177 ◦ C
Compressive ASTM 0.72 MPa 0.14–0.46 MPa 0.13–0.40 MPa In test
strength at D-3574 (C) at 10% deflection 10% deflection
−253 ◦ C

Table 3. Flammability properties of the TEEK foams.

Property Test method TEEK-HH TEEK-HL TEEK-LL TEEK-CL

Density ASTM 0.08 g cc−1 0.032 g cc−1 0.032 g cc−1 0.032 g cc−1
D-3574 (A)
Oxygen ASTM 51% 42% 49% 46%
index D-2863
Vertical burn FAR 25.853(a)a Not available
Burn length 0 cm 1 cm 0 cm
Dripping None None None
Smoke Visual None None None None
generation inspection
a The flame time applied was 25 s.

was 0.31 MPa and 0.72 MPa, respectively. The TEEK-HL and -LL had values ranging from
0.06 MPa to 0.10 MPa and 0.13 MPa to 0.46 MPa, respectively.

4.4. Flammability
Resistance to an oxygen atmosphere, smoke generation and flammability are critical properties
that polyimide foams will be required to have if they are to be used on the next generation RLV.
Table 3 shows the results from the limiting oxygen index test the vertical burn test, and the
visual inspection of gas evolution. The three polyimide foams tested for the limiting oxygen
index show excellent resistance to combustion in an oxygen-rich environment. The foams
of lower density, 0.032 g cc−1 , exhibit lower values of about 42–49% as compared to the
polyimide foam at a higher density of 0.08 g cc−1 which had a limiting oxygen index of 51%.
Figure 12 shows the visual effects of the vertical burn test on the three samples tested. Two
of the samples were the same polyimide foam at two densities. The largest burn length was
0.013 m, while the largest value permissible for a material to pass the test was 0.152 m. All
three foams showed excellent resistance to burning.
The test also showed that the materials were non-ignitable and would not self-combust.
The TEEK-HL was the only specimen to exhibit any shrinkage due to the extreme heat.
This shrinkage was determined to be less than 10% of the original thickness. Finally, during
the visual inspection of the tests, none of these materials exhibited any smoke generation.

Downloaded from hip.sagepub.com at PENNSYLVANIA STATE UNIV on May 17, 2016


Polyimide foams for aerospace vehicles 11

Figure 12. Vertical burn specimens of TEEK-LL, TEEK-HL and TEEK-HH (from left to right).

4.5. Uniaxial tension tests


Uniaxial tension panels were tested to simulate the hoop stress in a cryogenic tank. The panels
utilized the IM7/977-2 Gr–Ep (±45◦2 , 902◦ , 01.5

)S substrate with TEEK-HH and fibre optic
temperature and strain sensors. The foam was bonded to the Gr–Ep substrate with the room-
temperature cure epoxy shown in figure 8. The panel was subjected to 50 thermomechanical
cycles. The maximum temperature of the high-temperature side (outer or upper surface of
foam) was 177 ◦ C for panel number one and increased to 207 ◦ C for panel number two. The
temperature on the cryogenic side (tank inner wall) was below −253 ◦ C. After the 50 cycles,
panel number one displayed no signs of degradation except for a slight darkening of the upper
surface on the foam. Panel number two was darkened and had a slight blistering on the upper
surface after 50 cycles, indicating that the foam outgassed volatiles at high temperatures.

4.6. Thermal forming


Several specimens were successfully thermo-formed to the desired 0.0445 m radius; however,
upon removal from the vacuum bag, spring back occurred. The initial spring back was an
increase in the radius from 0.0445 m to 0.0635 m. A final radial measurement was taken after
24 h of the unrestrained relaxation. The resultant radius was 0.0711 m. Upon visual inspection,
slight buckling of the foam was noticed along the centreline due to the non-uniformity of the
cellular structure. As a result of the spring back and the slight centreline buckling, more process
optimization must occur to ensure that the foam will not relax away from a tight radius. To best
utilize a thermo-formed TEEK foam one should thermal form the structure at a temperature
above the use temperature to ensure relaxation does not occur at elevated temperatures. The
TEEK foam, however, can be thermo-formed to the required radius for an RLV cryogenic tank.

Downloaded from hip.sagepub.com at PENNSYLVANIA STATE UNIV on May 17, 2016


12 E S Weiser et al

5. Conclusions

A new class of polyimide foam has been developed using monomeric solutions from the
reaction of a dianhydride and diamine dissolved in a mixture of foaming agents and alkyl
alcohol. These new polyimide foams exhibit excellent stability at cryogenic, ambient and
elevated temperatures while maintaining high mechanical strength. The results from the
material characterization tests performed on these polyimide foams indicate that these foams
will be excellent for use on next generation launch vehicles. The higher operating temperature
and strength of the foam can potentially reduce the weight of the entire vehicle by reducing
TPS and TPS integration structures. Thermal, mechanical, combined thermal/mechanical
and flammability testing has indicated that these materials can endure the rigours of space
and space-like environments. Flammability results indicate that these materials are well
suited for extremely combustible environments and will maintain dimensional stability during
exposure to high temperature and flames. The uniaxial tension test exhibits the ability of
these materials to survive simulated flight conditions for 50 cycles. Initial thermal forming
tests indicate that these materials can be formed to a cryogenic tank radius and possibly to
tighter-radius parts such as fuel lines. Based on the results contained in this report, this new
high-temperature, hydrophobic, low-density, structural polyimide foam will support the needs
of the next generation of launch vehicles.

References

[1] Cellular materials Encyclopedia of Polymer Science and Engineering 1985 vol 3 (New York: Wiley) pp 1–59
[2] NASA Contractor’s Report 1994 NASA Contract No. NASW-4367, Research Triangle Institute, North Carolina
[3] Cellular materials Encyclopedia of Polymer Science and Technology 1965 vol 3 (New York: Wiley) pp 80–130
[4] McConnell V P 1997 NASA gets hands-on with X-33 design High-Performance Composites 5 56–8
[5] Fisher K 1997 Resin flow control is the key to RTM success High-Performance Composites 5 34–8
[6] Weiser E S, Baillif F F, Grimsley B W and Marchello J M 1998 High temperature structural foam Proc. 43rd Int.
SAMPE Symp. (May 31, 1998) pp 730–44
[7] Hou T H, Johnston N J and St Clair T L 1995 IM7/LARCTM –IA polyimide composites High Performance
Polymers 7 105–24
[8] McAuliffe P S, Davis R C and Taylor A H 1986 Development of reusable, flight-weight cryogenic foam insulation
system American Institute of Aeronautics and Astronautics (AIAA) Space Systems Technology Conf. (June,
1986) AIAA-1189-CP
[9] Johnson T F, Natividad R, Rivers H K and Smith R 1998 Thermal structures technology development for reusable
launch vehicle cryogenic propellant tanks Space Technologies and Application Int. Forum (Albuquerque, NM,
January 1998)

Downloaded from hip.sagepub.com at PENNSYLVANIA STATE UNIV on May 17, 2016

You might also like