Skip to main content

Full text of "Assessment of Technologies for the Space Shuttle External Tank Thermal Protection System and Recommendations for Technology Improvement"

See other formats

NAS A/TM-2004-2 13238 

Assessment of Technologies for the Space 
Shuttle External Tank Thermal Protection 
System and Recommendations for Technology 

Part 1: Materials Characterization and Analysis 

Erik S. Weiser and Michael P. Nemeth 
Langley Research Center, Hampton, Virginia 

Terry L. St. Clair 

National Institute of Aerospace, Hampton, Virginia 

July 2004 

The NASA STI Program Office ... in Profile 

Since its founding, NASA has been dedicated to the 
advancement of aeronautics and space science. The 
NASA Scientific and Technical Information (STI) 
Program Office plays a key part in helping NASA 
maintain this important role. 

The NASA STI Program Office is operated by 
Langley Research Center, the lead center for NASA's 
scientific and technical information. The NASA STI 
Program Office provides access to the NASA STI 
Database, the largest collection of aeronautical and 
space science STI in the world. The Program Office is 
also NASA's institutional mechanism for 
disseminating the results of its research and 
development activities. These results are published by 
NASA in the NASA STI Report Series, which 
includes the following report types: 

completed research or a major significant phase 
of research that present the results of NASA 
programs and include extensive data or 
theoretical analysis. Includes compilations of 
significant scientific and technical data and 
information deemed to be of continuing 
reference value. NASA counterpart of peer- 
reviewed formal professional papers, but having 
less stringent limitations on manuscript length 
and extent of graphic presentations. 

and technical findings that are preliminary or of 
specialized interest, e.g., quick release reports, 
working papers, and bibliographies that contain 
minimal annotation. Does not contain extensive 

• CONTRACTOR REPORT. Scientific and 
technical findings by NASA-sponsored 
contractors and grantees. 

papers from scientific and technical 
conferences, symposia, seminars, or other 
meetings sponsored or co-sponsored by NASA. 

technical, or historical information from NASA 
programs, projects, and missions, often 
concerned with subjects having substantial 
public interest. 

language translations of foreign scientific and 
technical material pertinent to NASA's mission. 

Specialized services that complement the STI 
Program Office's diverse offerings include creating 
custom thesauri, building customized databases, 
organizing and publishing research results ... even 
providing videos. 

For more information about the NASA STI Program 
Office, see the following: 

• Access the NASA STI Program Home Page at 

• E-mail your question via the Internet to 

• Fax your question to the NASA STI Help Desk 
at (301) 621-0134 

• Phone the NASA STI Help Desk at 
(301) 621-0390 

• Write to: 

NASA STI Help Desk 

NASA Center for AeroSpace Information 

7121 Standard Drive 

Hanover, MD 21076-1320 

NAS A/TM-2004-2 13238 

Assessment of Technologies for the Space 
Shuttle External Tank Thermal Protection 
System and Recommendations for Technology 

Part 1: Materials Characterization and Analysis 

Erik S. Weiser and Michael P. Nemeth 
Langley Research Center, Hampton, Virginia 

Terry L. St. Clair 

National Institute of Aerospace, Hampton, Virginia 

National Aeronautics and 
Space Administration 

Langley Research Center 
Hampton, Virginia 23681-2199 

July 2004 

The use of trademarks or names of manufacturers in the report is for accurate reporting and does not 
constitute an official endorsement, either expressed or impHed, of such products or manufacturers by the 
National Aeronautics and Space Administration. 

Available from: 

NASA Center for AeroSpace Information (CASI) National Technical Information Service (NTIS) 

7121 Standard Drive 5285 Port Royal Road 

Hanover, MD 2 1 076- 1320 Springfield, VA 22 1 6 1 -2 1 7 1 

(301) 621-0390 (703) 605-6000 




Erik Weiser\ Terry St. Clair^ and Michael Nemeth^ 

^Langley Research Center, Hampton VA 23681 
^National Institute of Aerospace, Hampton VA 23666 


The use of foam insulation on the External Tank (ET) was necessitated by the 
potentially hazardous build up of ice on the vehicle prior to and during launch. This use 
of foam was initiated on the Saturn V rocket which, like the Space Shuttle, used 
cryogenic fuel. Two major types of foam have been used on the ET. The first type is an 
'acreage' material that is automatically sprayed on in a controlled environment. This 
material is designated as NCFI 24-124. It replaced CPR 488 in 1998 and has been used 
since that time. The other major foam is a 'hand- sprayed' foam that is used to 'close out' 
regions where the various sections of the ET are attached. Two types of close-out foams 
have been used on the ET, known as BX-250 or BX-265, depending on when the ET was 
built. The external tank designated as ET-1 16 was the first tank to use BX-265 in place 
of BX-250, however, ET-93 did have some rework done with BX 265 on the fairing over 
the aft ET/Solid Rocket Booster (SRB) fitting. The bipod-ramp foam that was lost during 
the ascent of Columbia was made of the BX 250 material. 

The NCFI 24-124 foam is primarily a polyisocyanurate. The CPR 488 was also a 
polyisocyanurate, but its exact chemistry has not been disclosed to the authors at this 
time. The BX 250 and BX 265 foams are primarily polyurethane. Chemically, all of these 
foams are quite different although they are made from the same basic starting chemical 
compounds. The difference is in the ratio of reactive compounds. In both foams, there are 
many 'side products' that can possibly form if the humidity or the reaction temperature 
varies. This report concentrates on BX 250 and BX 265 polyurethane foams that are used 
for the manual close-out sprays because these materials have been identified as the source 
of most of the observed 'foam shedding.' Comparisons of the CPR 488 to the NCFI 24- 
124 are not addressed in this document because of 1) the lack of chemical information 
about the CPR 488, and 2) the lack of major problems with this foam 

The objectives of the present report are to study the chemistries of the various 
foam materials and to determine how physical and mechanical anomalies might occur 
during the spray and curing process. To accomplish these objectives, the report is 
organized as follows. First, the chemistries of the raw materials will be discussed. This 
will be followed by a discussion of how chemistry relates to void formation. Finally, a 
TGA-MS will be used to help understand the various foams and how they degrade with 
the evolution of chemical by-products. Future plans will also be discussed with an 
emphasis on developing a more comprehensive understanding of foam chemistry. 


Comparison of the BX-250 and BX-265 foams 

The BX 265 foam evolved from BX 250 polyurethane foam, which was originally 
developed for the Saturn V Program. From the excellent laboratory analyses that have 
been performed at the Michoud Assembly Facility (MAF) (as reported in MMC-ET- 
SEO5-650), the major differences in the starting materials used for the two foams are as 
follows: (1) BX-250 foam used the blowing agent CFC-11, while BX-265 uses the more 
environmentally-friendly HCFC 141b, (2) BX-250 foam used a form of Methylene Di- 
Isocyanate (MDI) that was sold commercially but is no longer available, while BX-265 
uses a newer commercial form of MDI, and (3) BX-265 foam has an additional catalyst 
that appears to allow spraying at lower than expected temperatures. Details of each of 
these differences are discussed subsequently. 

Blowing Agent Difference. The blowing agent difference is an extremely important 
factor affecting foam performance because the two blowing agents have different boiling 
points, which greatly affects their ability to expand the cells during formation of the 
foams. CFC-11 has a boiling point of 75°F, as compared to 90°F for HCFC-141b. This 
difference in boiling points necessitates that BX 265 should be sprayed at a temperature 
15°F higher than BX-250 in order to create the same amount of gas available for cell 
expansion. According to the specifications that are in place at MAF, this step in the spray 
process is not always done. The lower temperature limit for spraying BX-250 is 65°F, 
while the lower temperature for spraying BX-265 is only 70°F. This relatively low spray 
temperature is possibly a source of major processing problems for BX-265. Even though 
one would expect that the spray temperature should always be at, or above, the boiling 
point of the blowing agent, it is possible to have a marginally successful spray at 5°F 
lower. In the case of BX-250, there is a 'heat of reaction' that drives the reaction 
components to a temperature above the ambient temperature (probably above the 75 °F 
boiling temperature of the CFC-1 1). In the case of BX-265, this elevation in temperature 
would have to be 15°F higher to generate the same amount of gas for blowing that foam. 
There is the possibility that the 'heat of reaction' in BX-265 could be higher because 
when it was formulated, an additional catalyst was added that should accelerate the 
chemical reaction. However, the Arrhenius relationship dictates that a 15°F increase in 
reaction temperature will result in a doubling of the reaction rate. This holds true 
whether the added heat comes from the environment or from the heats of the chemical 
reactions. Thus the reaction for BX-265 should proceed at a much faster rate than BX- 
250 due to the increase in foaming temperature that results from the exothermic reaction. 
Finally, incomplete foaming could cause either of the foams to have additional unreacted 
components within the system which act as a decomposition by-product and an unwanted 
increase in foam density. Likewise, a large exothermic reaction could cause unwanted 
voids to form due to an acceleration of gas by-products that can become entrapped within 
the system. 

The other major factor involved in changing the blowing agent is that HCFC- 141b is 
more likely (because of its more dipolar molecular structure) to act as a solvent as well as 
a blowing agent. Because of the molecular structure of CFC-1 1, it probably acts only as 
a blowing agent. If HCFC- 14 lb does act as a solvent, it probably speeds up the reaction 
rate between the Part A and Part B components of the foam. Likewise, it will probably 

remain physically attached (or dissolved) in the final foam. This attribute can cause the 
final foam to be plasticized and it will certainly result in a slower loss of the blowing 
agent during subsequent storage. Additionally, because of the higher polarity of HCFC- 
141b, there should be a greater affinity for BX-265 foam to absorb moisture. 

MDI Difference. The MDI component is a chemical mixture of the 'pure' 
compound methylene di-isocyanate and its higher molecular weight oligomers. The ratio 
of the 'pure' compound to each of its higher molecular weight oligomers is an extremely 
important variable that must be closely monitored and controlled to assure reproducible 
foam properties from spray to spray. There is a ratio that the manufacturer of this 
material tries to maintain and this ratio is monitored at MAF on all incoming batches of 
this material. 

Unfortunate circumstances in the chemical industry have resulted in the 
discontinuation of the original MDI mixture that had been used for years in BX-250 
foam. A source for a nearly identical replacement resulted in the acceptance of another 
lighter colored MDI mixture for use in BX-265 foam. Although the two MDIs are 
probably very similar, chemically, they certainly have some differences. This is 
evidenced by the difference in color. The MAF laboratory has done an excellent job in 
monitoring the quality of the old and new MDI mixtures. This monitoring will certainly 
need to continue due to the probability that another change in vendors for this MDI 
component will occur because of the hazards associated with its production. Most 
chemical compounds of this type are being manufactured outside the USA because of 
more lenient restrictions regarding safety in manufacturing issues. Thus, it is all the more 
important for the Analytical Lab at MAF to develop as much comprehensive chemical 
data as possible on this material. 

Other Raw Material Differences. A third situation with the 'other raw materials' 
is the same as for the MDI; that is, there will be a continual environment of change in 
sources for these materials. As an example, the Analytical Lab at MAF noted a change in 
a catalyst within the last two years. When this change was reported to Mr. John Bzik 
(MAF), he found out that the supplier had stopped making the catalyst and had replaced it 
with a product that had a very similar numerical designation. The Analytical Lab noted a 
color change and a significant drop in the overall available catalyst level in the new 
material. Scotty Sparks of MSFC alerted the authors to this change in a technical 
interchange meeting at MAF in September 2003. The authors found that many such 
compounds are being made in other countries and are being shipped to the USA for use. 
It was determined that most of these compounds have been 'stabilized' with a 'radical 
trapping' chemical called butylated hydroxy toluene (BHT). When the authors met with 
the Analytical Lab personnel (Rando, Spiers and Smith), they found that they had 
detected this chemical in the newer form of the catalyst. The function of this BHT is to 
stabilize chemicals during shipping and storage. This function is also likely to cause this 
compound to behave differently in the mixture that is used to create the foam. In 
particular, the BHT may inhibit the catalytic effect of the catalyst compound or it may 
affect some other reaction that occurs in the complex chemistry associated with foam 

This example is not meant to convey that significant changes occurred in the final 
foam properties, it is simply being used to show that the Analytical Lab personnel will 
need to continue to update their analytical capabilities to maintain a careful watch over all 
incoming starting/raw materials. Failure to do so may result in deleterious effects on the 

At this time it is not possible to comment on the effect of the additional catalyst in 
BX-265 because the exact composition of BX-250 and BX-265 have not been disclosed. 
This situation is in the process of being corrected and this matter will be addressed in a 
subsequent document by the authors. 

Manual-Spray Process for BX-250 and BX-265. Both BX-250 and BX-265 
foams are manual-spray, closeout foams, in contrast to NCFI 24-124 foam, which is 
automatically sprayed. Manual spraying of the close out foams can lead to problems that 
are hard to isolate to one specific cause. Manual spraying requires a trained operator who 
sprays the foam at the exact same distance each time and allows each foam layer to rise 
the same amount of time between passes to insure adequate rise and cure before the next 
pass. Each pass needs to have the same gun angle and sweeping motion to ensure that the 
foam in one section of the ET is the same as in another. Consistent and repeatable 
spraying is very hard to achieve, even for the highly skilled of operators at MAF. In 
addition to the immense amount of skill required by the operator, these foams require a 
specific temperature and humidity to insure a quality product. In order to insure that the 
BX 265 foam is sprayed at an adequate temperature, a spray-parameter chart has been 
developed (see MAF document number RTF-TOX-001) that is shown in figure 1. 
However, as reported earlier, the minimum temperature allowed for spraying BX-265 
foam is 70°F and this temperature is well below the boiling point of HCFC-141B Freon 
blowing agent in the foam. Spraying at this minimum temperature could present many 
problems, as stated earlier, and could result in some of the defects associated with these 
foams that will be discussed in a later section of the present report. The best solution to 
insure that quality foam is produced is to develop a method by which the entire closeout 
foam is automatically sprayed onto the ET. While this procedure is a very difficult 
proposition, it would produce a quality foam with more uniformity on a repeatable basis. 
In the meantime, improvements to the guns that are used to spray the foam, as well as 
increasing the temperature in the chamber, will help to produce foams that are similar 
from lot to lot. 

Temperature versus Relative Humidity Requirements 


















r urriiiuuu 








70 80 90 


Figure 1. Processing parameters for manually sprayed BX 265 TPS foam. 

It is also recommended that the human environment for manually spraying the 
foam should be made reasonably comfortable for the operators. As previously stated, the 
temperature for manually spraying that has been used is too low. Increasing this 
temperature by 10-15°F will make that environment more uncomfortable, which could 
lead to operator error. For this reason, the authors strongly recommend that the personnel 
who perform this process be outfitted with "Cool Suits" that are similar to those worn by 
race-car drivers. The need to place high value on the comfort of these valuable people is 
critical to quality manually sprayed foam. 

NCFI 24-124 Acreage Foam 

The NCFI 24-124 foam covers most of the ET and is automatically sprayed in a 
cell where the temperature and humidity are very tightly controlled. This is opposed to 
BX-265 that is manually sprayed in an environment that is not as rigidly controlled. It is 
very important to note that the allowed low-end spray temperature for NCFI 24-124 is 
85°F. The BX-265 uses the exact same blowing agent, HCFC-141b, and has an allowed 
low-end spray temperature of 70°F. It was pointed out earlier in this report that HCFC- 
141b has a boiling point of 90°F. In order for the blowing agent to generate the proper 
amount of gas required to expand the foam cells, it must be used at or near its boiling 
point. Since there are fewer problems with the NCFI 24-124, the importance of the lower 
temperature limit for spraying needs to be thoroughly investigated. 

The NCFI 24-124 has a tightly controlled environmental chamber in which the 
foam is sprayed. In addition, NCFI 24-124 foam is automatically sprayed and other than 
small-scale cohesive failures, known as 'popcorning,' in the intertank region, this foam 
does not pose a significant threat to the Space Shuttle in the form of large debris. The 
'popcorning' issue was identified after post-flight analysis of STS-87, which showed 
higher than normal tile damage on the Orbiter. It was determined that temperature (slow 
heating) and the differential between the foam internal pressure and the external vacuum 
was the cause of 'popcorning.' To fix this problem, small holes of 0.1 inch diameter 
were placed in the foam every 0.3 inches to serve as a venting mechanism for alleviating 
'popcorning.' This simple fix has significantly reduced the 'popcorning' in the intertank 
region and, for the most part, has eliminated the debris threat of NCFI 24-124. However, 
the punching of holes raises other issues such as allowing water to penetrate the foam in 
this area. This penetration could lead to water freezing and thawing in these holes which 
would cause further damage to the foam. Such damage could create loci for foam failure 
and 'popcorning' in cases where the Space Shuttle is left on the launch pad for long 
periods of time. 

The NCFI 24-124 has the same problems as all of the other foams previously 
discussed with regard to having continuity and consistency of raw materials. In fact, the 
catalyst that was found to contain a radical-trapping agent is a component in NCFI 24- 
124 foam. Also, since this foam is primarily composed of the lighter colored MDI, the 
constant constituency of this raw material is extremely important. 

Void Formation Issues 

A possible reason for formation of large voids in the bipod-ramp area of the ET 
may be related to the fact that the reaction of the raw materials can be greatly affected by 
the temperature at the time of spraying. As previously stated herein, the Arrhenius 
equation indicates that a doubling of reaction rates occurs with a 15°F increase in 
temperature. Therefore, the reactivity can be greatly diminished when a manual-spray 
operation occurs at lower temperatures. 

In foam formation, there are two critical factors that can often be problematic in 
creating a high quality cellular structure. The first factor is cell growth, which is a direct 
effect of how the blowing agent creates 'bubbles' in the liquid mixture of reactants. If the 
reaction occurs too slowly, these 'bubbles' will form and collapse prematurely because 
the molecular weight buildup (polymerization) has not reached a critical level when the 
reaction mixture has reached a gel point. Secondly, if the reaction proceeds too rapidly, 
the 'bubbles' do not grow to their optimum size or shape prior to gellation. Both factors 
degrade the performance of the foam significantly. 

A large void such as the one that was discovered in the bipod ramp of a test panel 
sprayed at MAF may have resulted from spraying the foam at too low a temperature. In 
such a case, the various layers of spray may have created foam structures that collapsed 
prematurely. This weak structure was most likely over sprayed in subsequent spray 
passes with foam that formed better cell structures because the heat of reaction of the 
underlying layers raised the temperature of the part. Since the bipod region requires 
numerous overpasses of material to build-up the thickness desired for that region, the 
exothermic reaction would be significantly higher in this area than in sections of the 

vehicle where less foam is required. The result would be a significant amount of heat 
being generated through the thickness of the foam before the material has become fully 
cured. This heat could then result in large amounts of volatiles being produced that 
cannot escape due to the outer rind layer. The result would be formation of a void similar 
to the void found on the test panel sprayed at MAF (figure 2 left). In addition, any 
material that failed to attain proper molecular weight could continue to react with 
ambient moisture at a later time and enlarge the void because such a chemical reaction 
evolves carbon-dioxide gas, with accompanying build up of internal pressure. 

Figure 2. Cross section of test-panel bipod ramp. 

Assessment of Foam Defects 

In a memo from A. Clouatre and L. Shows of MAF, dated 31 January 2004, 
definitions and causes were listed for a variety of defects that have been found in foam 
sprayed on flight hardware surfaces at MAF. In particular, this memorandum outlines 
twelve TPS defect categories that appear to have been discovered during the Dissection 
Test Plan (809-9341) activities, and it includes a short description of their formation. 
These defects and descriptions are itemized subsequently, along with comments of the 
authors of the present report that are given parenthetically in italic type. 

1. Delamination from the primer (figure 3-1) - The foam pulls away from primer. 
The causes include inadequate surface preparation, improper substrate 
temperature (too cold), overly thick foam application, or an inadequate overlap 
time. {The given causes appear to cover the primary scenarios.) 

Delamination from the adhesive (figure 3-2) - Formation of an inadequate bond 
between the foam and adhesive. The causes include stress build-up in foam that 
results from thick passes or an excessive total foam thickness. {Other causes 
could be a cold surface or contaminants on the adhesive.) 

3. Delamination from the super light ablator (SLA, see figure 3-3) - The foam pulls 
away from SLA. The causes are similar to the causes of the previous defect and 
also result from inadequate cleaning of SLA prior to foam application. {Since all 
the SLA is coated with the adhesive, Conithane, prior to spraying, the 
delaminations may be associated with poor adhesion of the adhesive to the SLA 
or to a cold surface. Another factor may be excessive moisture pickup on the SLA 
or adhesive.) 

Delamination along a knit line (figure 3-4) - Separation along foam-to-foam 
interface. The causes include inadequate overlap time, excessive thickness per 
foam pass, or contamination. (The given causes appear to cover the primary 

Delamination from the substrate (figure 3-5) - Separation between foam and the 
substrate that results in a layer of broken cells remaining on the primed aluminum 
surface (appears as a thin fuzzy film). The causes are normally related to thick 
spray passes, a substrate temperature that is too cold, or to surface contamination. 
(Another cause may be excessive moisture on the substrate, which leads to the 
MDI component reacting with the moisture to form carbon dioxide that could lead 
to weak cell walls in the layer adjacent to the substrate.) 

6. Delamination between the PDL repair material and the adhesive (figure 3-6) - A 
condition where the PDL foam that is used in a repair area tears away from the 
adhesive tie coat. The causes include excessive charge size of the foam being 
applied to the repair and premature release of pressure in the mold during the first 
five minutes of foam rise and cure. (Other possibilities include excessive moisture 
being present and contamination of the adhesive surface.) 

7. Presence of concentrated material ejected by a gun spit (figure 3-7) - This effect 
occurs when gun closes or opens in an irregular manner, ejecting material that 
rich in either the A or B component. (The explanations given are the most likely 

Formation of elongated cells (figure 3-8) - This defect is characterized by the 
presence of cells with excessive expansion and is typically found in thick foam 
applications where blowing agent escapes during foam rise. (This situation could 
occur when the spray temperature is on the high side, causing excessive gas 
formation by the blowing agent. Excessive temperature could also be an artifact 
of the 'heat of reaction' of the reactants. This situation should result in a lower 
viscosity in the forming' foam that could result in overly large cell formation.) 

9. Porosity (figure 3-9) - This defect is characterized by concentrations of elongated 
cells. (A high spray temperature or excessive 'heat of reaction' may cause this 
phenomenon. If one performs the spraying near the upper temperature limit, the 
additional heat from the reacting raw materials can cause an 'out -of -limit' 
temperature condition within the foam itself.) 

10. Crack (figure 3-10) - A break in the TPS that does not exhibit material loss. The 
causes are stress build up or physical inducement. The cracks can be 
perpendicular or parallel to the substrate. Cracks in foams at cryogenic 
temperatures normally run perpendicular to the substrate and are caused by 
differences in thermal contraction between the TPS and substrate, or by substrate 
bending. {In the absence of weak cellular structure, this explanation covers the 
most likely scenario.) 

11. Rollover (figure 3-11) - During foam application, air pockets can be covered by 
foam that rebounds off of adjacent surfaces. Rollovers are normally associated 
with gun over spray, regions with closely packed geometric features, and 90° 
corners. {This cause is the best explanation. There is no chemistry problem to 
consider in this case.) 

12. Voids (figure 3-12) - These defects are characterized by an absence of foam. 
Voids can be caused by air pockets in bolt heads when applying foam or any 
condition where air can become trapped during the foam application process. 
{Another explanation is that moisture could be present that could cause the MDI 
component to react with it to form carbon dioxide which is an excess volatile. An 
elevated temperature in the forming foam could exacerbate this problem by 
causing a lower viscosity. This condition makes the foam more susceptible to 
creating large voids. Problems with the surfactant could also cause this 


Figure 3. The twelve known TPS-foam defects. 


Results from TGA-Mass Spectrometry 

Use of a Thermogravimetric Analyzer, coupled with a Mass Spectrometer (TGA- 
MS), was instituted by the authors to measure the total weight loss and weight loss of 
individual chemical by-products for the BX-265, NCFI-24-124, and NCFI-27-68 foams. 
A TGA-MS allows the chemical identification of component materials that come from a 
system during heating from room temperature to the decomposition temperature of the 
material. This analytical tool is critical to understanding what additional material is being 
released from the foams during prelaunch and ascent, and more importantly what 
chemical reactions, if any, are taking place during ascent. All three foams were heated 
from room temperature to 1000°C (capacity of instrument) at 10°C/min, but the major 
weight losses were in the range of 200 to 400°C. 

The results of the testing are significant to understanding the potential cause for 
the ET foam shedding. The data shown in figure 4 indicates that BX-265 has the lowest 
thermal stability of the three materials. It began showing appreciable weight loss below 
200°C and 50% weight loss at approximately 365°C. The fact that BX-265 has the 
lowest use temperature of the three materials was expected because BX-265 is a 
polyurethane foam and, in general, this type of polymer has a lower thermal stability 
compared to polyisocyanurate foams, like the NCFI foams. This point is significant 
because some of the data generated have shown that the ET foams see temperatures well 
in excess of 200°C at around 80 seconds into flight, which is approximately the time of 
foam loss on STS-107. 


PY OA*^ 





-20 - 

-40 - 

\ ^^ 

-60 - 

-80 - 



100 - 

' 1 1 1 1 

1 1 







Temperature (°C) 

Figure 4. TGA weight loss for ET foams at 10°C/min in helium. 

The mass spectrometer portion of the TGA-MS identified 12 different component 
materials released from the three foams tested during the heating from room temperature 
to 1000°C. Figures 5-10 show the rate at which each component gas was evolved from 


the system, with the area under the curve being proportional to the total approximate 
weight loss for each material identified. The data show that the most significant weight 
loss is attributed to two components. The first component is Freon, which begins coming 
out of the foam at 100°C and continues until a maximum loss has been achieved at just 
below 200°C (figure 5). The Freon loss can be attributed to two causes. The first cause is 
due to entrapped Freon in the foam cells and the second is due to Freon that was 
physically attached to the polymer itself. 

PY ^C 

.^ 6e-10 - 






/ ■ \ 



■§ 4e-10 - 




CD 2e-10 - 


400 600 800 

Temperature (°C) 



Figure 5. Freon evolution from ET foams at 10°C/min in helium. 

The second most significant component that came off the foams tested is carbon 
dioxide (see figure 6). CO2 is the by-product of the reaction between water vapor and 
unreacted isocyanate. It's presence at the lower temperature regime of the experiments 
suggests that the foam reaction is incomplete and can be triggered by the elevated 
temperature experienced during launch. For all cases, the carbon dioxide begins 
releasing from the foams at around 200°C, with a maximum in the case of BX-265 at 
365°C. In addition, the data shows that at approximately 600°C for the BX-265 and 
slightly higher for the NCFI foams, a second peak which can be associated with the 
decomposition of the foams begins to form. Figure 7 confirms that decomposition occurs 
above 600°C due to the evolution of hydrogen gas, which is indicative of decomposition 
in these polymers. 



PY 'iRc; 

4e-9 - 



3e-9 - 


2e-9 - 

/ \\ 

/ M„ 

1e-9 - 


p' ^^ 


/ ^ 


' 1 

— 1 1 

1 1 

200 400 600 800 1000 1200 

Temperature (°C) 

Figure 6. CO2 evolution from ET foams at 10°C/min in helium. 

1 .8e-9 

1 .6e-9 

-^ 1 .4e-9 


3 1 .2e-9 

■■§ 1.0e-9 


l3 8.0e-10- 



^ 6.0e-10- 




200 400 600 800 1000 

Temperature (°C) 


Figure 7. Hydrogen-gas evolution from ET foams at 10°C/min in helium. 

In addition to the Freon, carbon dioxide and hydrogen gas, nine other components 
were released from the foam during heating from room temperature to 1000°C. Figures 8- 
10 show the most significant materials that were released by the BX-265 foam. These 
components are important because they all began coming out of the foam at 
approximately 180°C. This finding is significant because these additional losses of 
gases could increase the volume of entrapped gases within the foam cells or defects, and 
as the external pressure drops during ascent, the internal pressure of the cell or defect 
could exacerbate the shedding of popcorn-size foam. 





^ 6e-1 1 

I 4e-11 







DY Ofic; 



1 1 1 1 1 1 







Temperature (°C) 

Figure 8. Phenol evolution from ET foams at 10°C/min in helium. 

Figure 8, shows that phenol begins evolving from the foam at approximately 
190°C and that the rate rapidly increases up to a maximum peak of 9.5 x 10'^^ g/min and 
then drops off to essentially zero at approximately 350°C. Figures 9 and 10 show results 
for two components that could not be identified with complete certainty because the exact 
chemistry of the BX-265 foam was never divulged. For both gases, significant amounts 
of both components, m/z 31 and m/z 152 (unidentified mass numbers), began evolving 
from the foam at approximately 180°C and return to zero evolution at approximately 
400°C and 250°C for the m/z 31 and m/z 152, respectively. 

5e-10 - 


^ 4e-10 - 


■■§ 3e-10 - 

CO 2e-10 - 



i \ 

1e-10 - 


200 400 600 800 

Temperature (°C) 

1000 1200 

Figure 9. Evolution of m/z 31 from ET foams at 10°C/min in helium. 




-g 1.2e-11 

5? 1.0e-11 


■■^ 8.0e-12 


LJJ 6.0e-12 


^ 4.0e-12 










Temperature (°C) 

Figure 10. Evolution of m/z 152 from ET foams at 10°C/min in helium. 
Summary and Recommendations 

The foams that are used on the thermal protection system (TPS) for the Space 
Shuttle external tank (ET) has evolved over the last 20 plus years. There have been 
major changes in foam formulation and foam component suppliers. Also, during the term 
of each of these contracts, the foam raw materials have been supplied to the contractors 
from different suppliers. In some cases, the supplier changed the raw materials, making 
it difficult to identify a change in TPS performance. 

A major change that occurred in the late 1990's was the use of a new blowing 
agent known as HCFC 141b. The previous CFC 11 blowing agent had been a standard 
that dates back to the Saturn V program. CFC 1 1 was known to be an ozone-depleting 
substance and the entire foam insulation market was transitioning over to HCFC 141b. 
NASA was forced to follow this trend because CFC 11 was disappearing from the 
commercial market. This change in blowing agents was a major change since the HCFC 
141b had a higher boiling point, requiring a higher application temperature for the ET 
foam. This higher application temperature inadvertently increased the reaction rate for 
the foam formation which could, in some cases, cause the foams to be sprayed with 
voids. Also, the chemical structure of HCFC 141b has an appreciably larger dipole that 
most certainly results in more solubility of the blowing agent in the foam cell walls and a 
greatly increased tendency to absorb water. This change in blowing agents affected both 
the hand sprayed material, BX 250, resulting in BX 265, and the acreage foam, CPR 488, 
resulting in NCFI 24-124. 

The foams that were studied have been shown by some initial Thermogravimetric 
Analysis/Mass Spectrometry (TGA-MS) tests (performed by Phillip Morris for NASA 
Langley) to generate volatile materials as the temperature increases in a manner that 
simulates that encountered during ascent. The authors recommend that an independent, 
comprehensive TGA-MS study of the various ET foams be undertaken in the near future 


in order to verify preliminary data related to the evolution of blowing agent; to study the 
volatiles associated with a continuation of the chemical reactions of the residual raw 
materials; and to study the decomposition by-products as the foams are exposed to 
temperatures that approach their stability limits. It is strongly recommended that a 
similar program be put in place at MAF. At the present time there is no chemical quality 
assurance performed on the final foam materials. This activity will greatly enhance the 
reliability of these materials and will afford a chemical database that will be easy to use 
in tracking future problems with foams. 

Since the acreage foam, NCFI 24-124, is automatically sprayed on in an 
atmospherically controlled cell, the higher temperature needed for the HCFC 141b was 
accommodated. A serious concern is that the lower limit for spraying the BX 265 (with 
the higher boiling point HCFC 141b) is 70°F; a temperature well below the boiling point 
of the blowing agent. The boiling point of HCFC 141b is 90°F. This lower- limit spray 
temperature could result in inadequate blowing of the foam and in having an excessive 
amount of entrapped blowing agent well below the surface of the foam. For the BX 250 
foam, the boiling point of CFC 11 is 75°F and the lower spray temperature is 65°F. 
Thus, it is strongly suggested that this lower-limit temperature for BX 265 blowing agent 
be re-examined. Failure to do so could result in significant void formations, similar to 
those pictured in figures 2 and 3. 

Another major concern is that the exact chemical structure of each of the 
components used in the manufacture of the various foams is not always known. This lack 
of information makes it difficult for the Analytical Lab at MAF to detect changes in the 
raw materials that are supplied. This problem was evident recently when a supplier made 
a change in a catalyst that is used in one of the foams. MAF was not alerted to the exact 
nature of the change. They were only alerted to a change in its numerical designation. In 
actuality, the supplier had made a major change in this product. The Analytical Lab noted 
a decrease in the active catalyst content and made adjustments to add more catalyst. The 
undocumented change that the vendor had made was to add a 'free radical trapping 
agent', designated as BHT, to the catalyst. This change allowed them to have a product 
with a longer shelf life, but no one can predict what effect this stabilizing agent will have 
on the many reactions that occur in the foaming process. Thus, there exists potential for 
degradation of the foam performance. 

Another very important difference between CPR 488 and NCFI 24-124 and BX 
250 and BX 265 is the change in the major MDI mixture component. The CPR 488 and 
BX 250 utilized an MDI mixture that is no longer made and the NCFI 24-124 and BX 
265 utilizes a new MDI mixture source. These materials are mixtures of methylene 
diisocyanate (MDI) and oligomers (higher molecular weight forms) of this chemical 
compound. The ratio of the pure MDI to its oligomers is extremely important to the 
physical and mechanical properties of the foams. A thorough chemical knowledge of this 
component is critical since it is the major component in all of the ET foams. It is highly 
recommended that this issue be studied in much more detail in the very near future. 

The need for more chemical- structure knowledge (with the concomitant 
acquisition of new analytical equipment) is important for analyzing all of the components 
in the foams. An expansion of this knowledge is recommended for the Analytical Lab at 
MAF. This task will involve learning about, documenting, analyzing and tracking all 
components on a routine basis. This expanded knowledge of the raw-material chemistry, 


along with TGA-MS data studies on the final foams, will allow the MAF Analytical Lab 
to operate in the same manner that their counterparts in the commercial manufacturing 
sector operate today. Analytical Chemistry is a rapidly changing science that is a very 
powerful tool in the quality manufacture of any material. 

Proposed Research Activities 

As a result of the studies conducted to date, there are several areas that need more 
research. In particular, a planned effort to develop a full chemical database and 
understanding of the raw materials that are used in all of the foams is needed. It is 
proposed that Erik Weiser of the Langley Research Center will lead and conduct such an 
effort, in conjunction with contractors through the National Institute of Aerospace (Drs. 
St.Clair and Nelson). If empowered, this team will begin a literature search and will meet 
with experts in the area of polyurethane chemistry to develop an in-depth understanding 
of the complex chemistry of the foam systems that are used on the External Tank (ET). 
The expected outcome is a report that details this chemistry and discusses, in scientific 
terms, how the various reactive monomers and raw materials interact with the catalysts 
and additives to create the different foams. Also, they will chemically detail some of the 
possible side reactions that can occur as the environmental conditions change. Two 
primary foci will be on the effect of excess moisture or water and on the effect of 
temperature on the reactions. It is anticipated that this study will broaden the chemical 
understanding of the foams and will allow for better predictions of the effect of future 
changes in raw materials. The team will work closely with the experts in this area at 
MAF and at MSEC in developing this database. 

At present there is no appreciable understanding of the exact state of chemistry in 
the as-formed foams that are used on the ET. This fact means that although the foams are 
seemingly in their final physical states when the ET is completed; it is quite obvious that 
there are still chemical reactions that can occur if the ET gets heated or is exposed to 
moisture. Likewise, the chemistry continues to change as the Space Shuttle is launched. 
Work done this past year at the Langley Research Center shows that the foam samples of 
BX-265, NCFI-24-124 and NCFI-27-68 contains much residual blowing agent and 
unreacted materials that continue to react and evolve gas as the temperature is raised. 
Thermogravimetric Analysis-Mass Spectrometry (TGA-MS) work conducted shows that 
significant amounts of the blowing agent, carbon dioxide and water are liberated as the 
temperature is increased. Of course, similar temperature increases occur during the 
ascent of the Space Shuttle. The TGA-MS data also shows that chemical reactions 
continue to occur as the temperature rises. Ultimately, there is an onset of decomposition 
that occurs and the effects of this decomposition on the foam structural and thermal 
performance needs to be quantified. 

The proposed Langley/NIA team will form an alliance with engineers and 
scientists at KSC to develop a comprehensive understanding of this "Foam Chemistry." 
The TGA-MS equipment that Langley has purchased will be utilized and the close 
alliance with Florida Institute of Technology (FIT) that KSC has developed will be 
employed to study this chemistry. Professor Gordon Nelson of FIT is a world-class expert 
in polymer degradation and has the equipment at FIT to carry out this study. His former 
Ph.D. student. Dr. Martha Williams of KSC, has volunteered to work closely with him on 


this activity. The combined studies at Langley and at FIT/KSC should provide much new 
information about this "Foam Chemistry." 

As these activities progress, this team will try to correlate their findings with 
known ET-foam problems such as the formation of the twelve defect types discussed 
previously herein. Experiments that can be done at MAF or at MSEC to verify any 
postulated chemical interactions that could lead to foam defects will be identified and 
planned. It is envisioned that members of the Langley /NIA team will continue to offer 
their expertise and services to MAE and MSEC personnel in creating an in-depth 
understanding of TPS-foam parameters that affect thermal and structural performance of 
the ET. 



Form Approved 
0MB No. 0704-0188 

The public reporting burden for tliis collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, 
gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this 
collection of information, including suggestions for reducing this burden, to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and 
Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person 
shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid 0MB control number. 



Technical Memorandum 

3. DATES COVERED fFrorn - 7o; 


Assessment of Technologies for the Space Shuttle External Tank Thermal 
Protection System and Recommendations for Technology Improvement 

Part 1: Materials Characterization and Analysis 





Weiser, Erik S.; St. Clair, Terry L.; and Nemeth, Michael P. 






NASA Langley Research Center 
Hampton, VA 23681-2199 




National Aeronautics and Space Administration 
Washington, DC 20546-0001 






Unclassified - Unlimited 

Subject Category 27 

Availability: NASA CASI (301) 621-0390 

Distribution: Standard 


Weiser and Nemeth: Langley Reseafch Center; St. Clair: National Institute of Aerospace. 

An electronic version can be found at or 


The use of foam insulation on the External Tank (ET) was necessitated by the potentially hazardous build up of ice on the vehicle prior to 
and during launch. This use of foam was initiated on the Saturn V rocket, which, like the Space Shuttle, used cryogenic fuel. Two major 
types of foam have been used on the ET. The first type is NCFI 24-124, an 'acreage' material that is automatically sprayed on in a controlled 
environment. It replaced CPR 488 in 1998 and has been used since that time. The other major foams, BX-250 or BX-265, are 'hand- 
sprayed' foams that are used to 'close out' regions where the various sections of the ET are attached. The objectives of the present report are 
to study the chemistries of the various foam materials and to determine how physical and mechanical anomalies might occur during the 
spray and curing process. To accomplish these objectives, the report is organized as follows. First, the chemistries of the raw materials will 
be discussed. This will be followed by a discussion of how chemistry relates to void formation. Finally, a TGA-MS will be used to help 
understand the various foams and how they degrade with the evolution of chemical by-products. 


External tank foam; Space Shuttle; Cryogenic insulation; Chemistry; Chemical analysis 













STI Help Desk (email: 

19b. TELEPHONE NUMBER (Include area code) 


Standard Form 298 (Rev. 8-98) 

Prescribed by ANSI Std. Z39.18