NASA Technical Reports Server (NTRS) 19920006816: Evaluation of selected thermal control coatings for long-life space structures

18 Pages·1.4 MB·English

by  NASA Technical Reports Server (NTRS)

NASA Technical 9
_..,_,:..v,,a_l.u..aEtion
I
Thermal Contro gs for
Long-Life Sp_a_°._i;_'" ..c._+..u..r.e.-s
arid William GI Wittel Jr;:
Unc I _s
0064315
!
_ram
m
NASA Technical Memorandum 4319
Evaluation of Selected
Thermal Control Coatings for
Long-Life Space Structures
Louis A. Teichman, Wayne S. Slemp,
and William G. Witte, Jr.
Langley Research Center
Hampton, Virginia
NA. A
National Aeronautics and
Space Administration
Office of Management
Scientific and Technical
Information Program
1992
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 the National Aeronautics and
Space Administration.
i
!
!
i
z:
i
!
i
Abstract One way to accomplish this objective is to apply pro-
tective coatings which have the proper ratio of solar
Graphite-reinforced resin matrix composites are
absorptance (as) to thermal emittance (e).
currently being considered for spacecraft structural
applications because of their light weight, high stiff- The thermal control coatings program at Lang-
ness, and low thermal expansion. These materi- ley Research Center has focused on the develop-
als must, however, be protected against degradation ment of stable thermal control coatings for composite
caused by the various elements of the natural space (largely graphite/epoxy) structures (tubes and pan-
environment. Thin protective coatings with stable els) for long-life space platforms in low Earth orbit
optical properties which minimize orbital thermal ex- (LEO). Research has concentrated on the develop-
tremes are attractive for this purpose. One way to ment of sputtered coatings applied directly to the
accomplish this objective is to apply protective coat- graphite/epoxy (Gr/Ep) composite surface and on
ings which have the proper ratio of solar absorptanee anodized thin aluminum foil. Both coating systems
(as) to thermal emittance (e). can be used as an atomic oxygen barrier between the
graphite-reinforced resin matrix composite and the
Research at Langley Research Center has concen- natural space environment. A small additional effort
trated on the development of sputtered coatings ap-
was also made to develop nickel-ba_sed coatings which
plied directly to the graphite/epoxy (Gr/Ep) com-
could be applied directly to the composite.
posite surface and on anodized thin aluminum foil.
Sputtered, anodized, and nickel-based coatings
Both coating systems can bc used as an atomic oxy-
were selected for study over commercial white paints
gen barrier between the graphite-reinforced resin ma-
trix composite and tile natural space environment as because their inherent tenacity made them more at-
well as for thermal control mechanisms. A small ad- tractive from a reliability standpoint for long-life mis-
ditional effort was also made to develop nickel-based sions. White paints have been used on many space-
coatings which could be applied directly to the com- craft with acceptable space environmental stability
posite. These coating systems were selected for study over 3- to 5-year missions (refs. 1 and 2), but no data
over commercial white paints because their inherent are available for extremely long-life (a0-year) mis-
tenacity made them more attractive from a reliability sions. To avoid possible chipping or discoloration of
standpoint for long-life space missions. paint coatings, several tenacious chemically bonded
coatings were chosen for more intensive study. The
Of all the protective coating techniques described, purpose of this paper is to describe results from
anodized aluminum foil coatings arc clearly the lea(l- research conducted on these tenacious coatings,
ing candidates for use on tubular and flat composite which were expected to be substantially more ad-
structures for large platforms in low Earth orbit. The herent than paints or adhesively bonded coatings,
anodized foil provides the composite substrate ma- and which met certain solar absorptance and ther-
terial with protection against many of the elements mal emittance criteria.
of the natural space environment atomic oxygen,
ultraviolet and particulate radiation and can offer a Selection Criteria for Spacecraft
broad range of tailored c_s/e. Both the aluminum foil Temperature Balance
and the anodizing process are commercially available,
Temperature cycling occurs as a spacecraft orbit-
and the foil can be produced in the large quantities
ing the Earth at low altitude proceeds from sunlight
required for large space structures.
to Earth shadow approximately every 90 minutes.
An unpublished computer-generated heat transfer
Introduction
analysis for graphite/epoxy tubular structures in a
Graphite-reinforced resin nmtrix composites are typical spacecraft orbit resulted in figure 1. The
currently being considered for spacecraft structural figure shows temperature cycling range as a func-
applications because of their attractive features-- tion of the ratio of solar absorptance (a.s) to ther-
light weight, high stiffness, and low thermal expan- mal emittance (e). The ideal case is illustrated by
sion. These materials must, however, be protected the 0.25/0.25 ratio, where the temperature cycle is
against degradation caused by the various elements around room temperature and extremely small com-
of the natural space environment. Thin protective pared with the temperature cycle of the uncoated
coatings are attractive for this purpose, but these graphite/epoxy tube with a ratio of 0.85/0.85. The
coatings, whatever their nature, must have stable op- values of 0.25/0.25 are not always achievable by each
tical properties which minimize the thermal extremes coating process and do not provide latitude for degra-
to which the composite structure is subjected as the dation of c_s (increase in as values) due to environ-
spacecraft moves in and out of the Earth's shadow. mental exposure and spacecraft contamination. For
30-yearmissionsa, solarabsorptancoef0.30anda
in table I and figures 2 through 4. These results indi-
thermalemittanceof0.65wereselectedtobenom-
cate that sputter coating of the composite substrate
inal valuesforinitial thermalcontrol,whichwould
dramatically lowers c_s from an undesirable value of
permitextensivedegradationof thecoatingbefore about 0.70 to a much more desirable value of 0.16
temperatures(-70 to +lT0°F) similarto thoseof
on the smooth surface (fig. 2). Coating of the sub-
the uncoatedcompositewouldbc reached.
Coat- strate beyond 420 A appears to have little effect on
ing systems which could meet these thermal criteria as, at least up to the maximum thickness of 2520 A.
were then selected for further study on the basis of Although the c_s values for the rough surfaces are
tenacity and ability to act as atomic oxygen barriers. somewhat higher, i.e., about 0.24, this, too, is an ac-
These systems are discussed in tile reimfinder of the ceptable value for space flight in LEO. However, the
paper. Sputtered aluminum coating also lowers e from 0.8
for the smooth surface of the bare composite to 0.08
(fig. 3), an imacceptably low value for effective tem-
2OO
perature balance in space. This problem could have
been anticipated because the sputtered surface is a
highly reflective, conductive metal. The aluminum-
100
Coated rough surface, however, produces a thermal
Thermal
emittance of 0.2 to 0.3 (fig. 3), yielding an c_s/¢ ratio
cycling range,
oF of about 1 (fig. 4), which is acceptable under some
space flight conditions.
-100
.85/.85 .25/.25 .30/.20 .30L65 The lower as and c values obtained for the smooth
surfaces (as compared with rough surfaces) can be at-
Solar absorptance/emittance
tributed to two effects. First, sputtered coatings on
Figure 1. Thermal cycling range forspacecraft truss structure rough surfaces tend to be nonuniform, with thinner
(assuming 2-in-id Gr/Ep tubes as truss elements). coatings deposited on highly sloped surfaces than on
flatter surfaces. Second, with uniform rates of depo-
Sputter-Deposited Aluminum sition per unit area across the planetary plate, rough
specimens and smooth ones of the same diameter re-
Tile sputtering study (ref. 3) consisted of mag-
ceive equal amounts of sputtered material. Calcu-
netron sputtering of ahlminum on graphite/epoxy
lations of surface area based on the roughness data
substrates. This study was driven by the same previ-
indicate that the rough specimens present about 1.3
ously mentioned thermal analysis which showed that
times as much surface area as the smooth ones. The
an uncoated graphite/epoxy surface could experience
coating will therefore be thinner on rough substrates
temperature cycles fl'om -70 to +lT0°F, often in-
than on smooth ones, and tile effect of the coating
ducing microcracking. Details of specimen prepa-
on optical properties will therefore be reduced.
ration and sputtering conditions are presented in
appendix A. For the rough surfaces tile minimum c_s occurred
at around 1000 A of alumimlm (see fig. 2); for
Optical reflectance generally decreases with sur- the smooth surfaces the minimum occurred at less |
face roughness, thereby increasing c_,. In the case of |
than 400 A. At greater coating thicknesses, solar B
composite laminates, surface roughness is controlled i
absorptances generally remained constant, although
mainly by the texture of the caul plates, separation
there is the hint of slight increases out to about |
sheets, and bleeder cloths used during fabrication. E
2520 __.Oxidation of the ahnninum during sputtering
The fabrication procedure employed in this study |
may have caused these slight increases, since electron
produced a laminate that had a "rough" side with
dispersive X-ray analysis of the coatings shows the
an average roughness of 170 p,in. and a "smooth"
presence of oxygen. The decrease of en_fittance with
side with an average roughness of about 25 pin., as E
increasing aluminum coating thickness, to the point i
measured by a profilometer. Specimens of both sur-
of aluminum opacity, was expected (fig. 3). The ratio
face finishes were sputter coated with ahnninum for
of c_s/e is approximately 1 for the rough surfaces IIi
different lengths of time, resulting in coating thick-
nesses ranging from 420 to 2520 A. and increases slightly with coating thickness (fig. 4).
The a_/c ratio for the smooth surfaces (fig. 4) rises
Solar absorptance and total normal emittanee from 2 to 4 over the coating thickness range of 420
E
were determined for six coating thicknesses and two to 2520 /_. Use of these coatings on composites in
surface textures, The results obtained are presented large space structures in LEO would have limited
=
2
iif--Smsuorofathce
application, due to lack of uniformity and complexity
of application.
- - - Rough surface
Nickel-Based Coatings
Solar .6 _ ] Standard
An investigation wa_s launched by Composite Op-
C_s .4 tics, Inc. into ways in which the surfaces of a propri-
etary nickel-based moisture barrier coating could be
.3
absorptance, .5 __ de.v_iatio-no----y--_ altered to provide desirable spectral characteristics
.2 for LEO applications. The coating offered an ex-
.1 cellent method of protecting a composite substrate
I I I I / against atomic oxygen attack but as was 0.2 and
0 500 1000 1500 2000 2500
was 0.1, an undesirable combination for the intended
Sputtered coating thickness, applications in space. Coating material composition,
mechanical abrasion, and chemical oxidation of the
Figure 2. Solar absorptance of sputtered aluminum on
surface were investigated as potentially viable tech-
T300/5209 as a flmction of sputtered coating thickness.
niques by which to raise c_s to about 0.3 and c to
about 0.6.
Alterations in surface preparation and nickel plat-
ing composition proved fruitless in raising E. Al-
terations to the specularity of the surface were the
i_ -- Smooth surface
most effective way to alter e, and values of 0.2 to
0.35 were obtained. Unfortunately, as was also al-
Total .6 tered to a fairly high (and undesirable) value of 0.5
or more. Based on these results, the decision was
emittance, ..75 ',_ - - - Rough surface
made to cease further consideration of this coating
i_ .4
system. The final report (COI-0988-5769, Sept. 21,
.3
1988) on this effort contains proprietary information
.2 and was given extremely limited distribution.
.1
0 500I 100a0 150t0 200I0 250!0 Chromic Acid Anodizing
Sputtered coating thickness, A series of contractual studies (refs. 4 and 5) was
established to develop and then optimize chromic
Figure 3. Total normal emittance of sputtered aluminum on
acid anodizing (CAA) applications for large plat-
T300/5209 as a function of sputtered coating thickness.
forms in LEO. CAA techniques were developed for
foil 24 ft long, 8 ft wide, and 3 rail thick. This foil
was then slit into 8-in-wide pieces 24 ft long, the size
required to protectively wrap the longest struts on
5 -- Smooth surface the then-proposed Space Station Freedom. Details
.... Rough surface of CAA procedures and the development of anodized
aluminum foil coatings are given in appendix B.
3 The 1145-H19 A1 alloy was the only alloy evalu-
125 ated that achieved the desired optical goals of a solar
a 2 absorptance of 0.35 or less and a thermal emittance
of 0.55 to 0.70. The 1145 foil also was the only high-
1('-- purity foil readily available in a variety of thicknesses
and tempers. The fully hardened temper (H19) min-
I I I I l
0 500 1000 1500 2000 2500 imizes chances of wrinkling and creasing of the foil
during processing, while the half-hard temper (H24)
Sputtered coating thickness, _,
was the easiest to work with when wrapping Gr/Ep
Figure 4. The ratio o_s/Eofsputtered aluminum as a function tubes. Optical properties achievable by this process
of coating thickness on T300/5209. are given in table II.
3
Exposure Facility confirm minimal changes in physi-
cal and optical properties of thin anodized aluminum
absorptance, 3 o o o o o o o
Solar .4[ after ahnost 6 years in LEO.
°(s 2
d 10;020 0 30 0 40'0050 0
Summary of Results
Exposure time, equivalent sun hours
Several tenacious thermal control coating systems
Figure 5. Effccts of ultraviolet radiation (2 times equivalent which met certain thermal, adherence, and atomic
sun hours; zero air mass) on the solar absorptanee of oxygen resistance criteria were chosen from intensive
water-scaled anodized 3-rail, 1145 A1foil. study as potential coverings for composite tubes on
long-life space platforms in low Earth orbit (LEO).
The results of these studies indicated the following:
I
Chromic-acid-anodized 3-mil-thick 1145-H19 A1
.
absorptance, 31
S°lar i:_3I o F7 ! adhesively bonded to Cr/Ep tubular structures
was shown to provide excellent protection and
Before 5000 hr t x 107rad thermal control in the LEO environment. The
exposure UV 600 keV electrons
anodized foil protected the Gr/Ep from degrada-
(3yr inLEO) (30 yr in LEO)
tion caused by atomic oxygen (see appendix B),
Figure 6. Effects of ultraviolet and electron radiation on minimized the temperature gradients in the com-
anodized aluminum foil.
posite struts, and provided passive thermal con-
trol. Techniques were successfully developed for
Sealing the surface of the anodized foil by sub- anodizing foil large enough to wrap, as a sin-
merging it in hot water through which an electrical gle piece, around diagonal struts of large truss
current was passed was performed to increase resis- structures.
tance to soiling and staining during handling. The
, Sputtering directly onto composites proved to be
sealing process is easily performed and possesses a
only a marginally succcssflfl method of providing
side benefit of increasing emittance while the absorp-
a surface within the desired range of the ratio
tanee remains constant.
of solar absorptanec to thermal emittance (a._/e)
CAA of foil 25 ft long by 44 in. wide was accom- and, at best, will be of limited use.
plished with up to three pieces of foil being processed
. Nickel-based coatings, like all metallic materials,
at the same time. Uniformity of optical properties
offer excellent protection against various elements
throughout the 25-ft lengths was excellent. A pro-
of the space environment, and can readily be mass
cess specification was developed and included as an
produced, but have inherently low values of c.
appendix in reference 5.
Preliminary efforts to find methods to alter the
The anodizing techniques developed were used spectral characteristics were unsuccessfifl.
in the fabrication of A1 foil-covered 2-in-diameter
graphite/epoxy (T-300/934 and P75/934) tubes. Of all the protective coating techniques described,
Both eoeuring and adhesive bonding of the A1 were anodized ahnninum foil coatings are clearly the lead-
used. The chromic-acid-anodized AI foil graphite/ ing candidates for use on tubular and flat composite
epoxy system was evaluated for durability to the structures for large platforms in tow Earth orbit. The
LEO space environment. For example, ultraviolet ra- anodized foil provides the composite substrate mate-
diation exposure in a vacuum, using xenon short-arc rim with protection against many of the elements of
lamps with quartz envelopes producing wavelengths the natural space environment atomic oxygen, ul-
of 200 400 nm for 5000 equivalent sun hours (equiva- traviolet an(t particulate radiation and can offer a
lent to 3 years in LEO), resulted in an increase of less broad range of tailored e_s/e. Both the aluminum foil
than 0.01 in solar absorptance (sec fig. 5). Also, no and the anodizing process are commercially available,
disbonding or change in optical properties occurred and the foil can be produced in the large quantities
after 25 000 thermal cycles of +i50°F in dry nitrogen required for space platforms.
(sinmlating 3 years in LEO). In addition, radiation
exposure of 107 rads with 600-keV electrons (equiv-
alent to 30 years in LEO) resulted in a negligible NASA f,anglcy Research Center
change in solar absorptance (see fig. 6). Preliminary, Hampton, VA 23665-5225
as yet unpublished, results from the Long Duration December 6, 1991
4
Appendix A
smooth side up and five with the rough side up.
Sputtering Study for Aluminum on During sputtering, the plate was stationary, about
3 in. under the alumimm_ target.
Graphite/Epoxy
TestSpecimenDescription All coatings were sputtered at 1 kW power.
The sputtering chamber was evacuated for at least
Thespecimenwsere1-in-diametedriscsof8-ply
30 minutes before sputtering, resulting in initial
[0,0,0,90]Ts300/5209graphite/epoxcyompositme a-
chamber pressures in the range of 4 x 10-6 to
terialcutfromlaminates.Thelaminatewasfabri-
1 x 10-5 torr. The chamber was backfitlcd with ar-
catedfromunidirectionaclommerciaplrcprcg.The
gon to a pressure of 8 pm and an arc was struck to
laminatewaslaid-uponasmoothsurfaceTefloncaul
form a plasma. The system was programmed to ramp
platewithastandardtexturedbleedecrlothontop,
up power to reach 1kW in 1minute after the plasma
sothatthe
cured composite had a smooth side and formed. At that moment, the large fan valve shield-
a rough side. Peak-to-trough measurements of the ing the specimens from the plasma was opened to
surface profile on thc smooth surface were in the expose the specimens. When the desired sputtering
range of 20 to 80 pin. The surface profile of the time was achieved, the valve was closed and power
rough surface yielded peak-to-trough measurements was turned off. The sputtering times were based
of up to 1600/*in. in variation, with the average being on previous experience and were chosen to provide
170 #in. a range of coating thicknesses. The thicknesses of
aluminum deposited on the sapphire thickness moni-
Specimen Preparation tors were assumed to be the same as the coatings on
the specimens.
After the specimens were cut from the sheet,
they were lightly sanded around the edges with SiC
paper to remove projecting fibers and were then Optical Properties
wiped with trichloroethane and rinsed with dcionized
Solar reflectance was measured in the wavelength
water. They were stored for several weeks in a
range of 0.3 to 2.5 #m with a Gicr Dunkle MS-251
desiccator. The specimens were weighed before and
solar reflectometer. The source, optics, and sphere
after sputter coating in an attempt to determine
characteristics of this instrument as it was used ap-
coating weights, but the results wcrc inconclusive
proximate the solar spectrum. For an opaque sur-
because of the extremely low masses of the coatings.
face, the solar absorptance can be computed by sub-
tracting the reflectance from unity. Total normal
Sputter Conditions
cmittancc of the specimens was determined from in-
Six sputter coating runs were made. For each frared reflcctivity measurements made with a Gier
run, ten 1-in-diameter specimens were placed on a Dunkle DB-100 infrared rcflectometer in the wave-
plate in the vacuum chamber. A sapphire thickness length range of 5 to 25 pro. For each surface texture
monitor was placed in the center of the plate, and and coating thickness, five specimens were coated
arranged around it were five specimens with the and measured.
5
Appendix B
of the 125-ft-long foils. The A1 foil alloys and tem-
Development of CAA Aluminum Foil pers which were evaluated are 3-rail 1145-H19 and
1145-H24, 3-mil 5024-H19, 3-rail 3003-H19, and 5-mil
Coatings
6061-0. They possessed similar solar absorptanccs of
Anodizing Procedures 0.08 to 0.17 and thermal emittances of 0.02 prior to
anodizing. The variation in absorptance values was
The anodizing of the Al foil was performed using
caused by sample orientation (because of the stria-
various contractor-developed specifications and pro-
tions in the unanodized foil) and was not attributable
duction facilities. The specifications also include the
to alloying elements. Alloy 1145 was the most read-
cleaning of the foil, which is required to ensure a sat-
ily available of all the A1 foil alloys. It wa_s avail-
isfactory anodizing. The specifications required that
able as fiflly soft (1145-0), half-hard (1145-H24), or
the foils be anodized in the following sequence:
fully hardened temper (1145-H19) and in a variety
of thicknesses. Alloy 6061 is fairly common but is
1. Vapor degreased
rarely produced in thicknesses less than 5 mils. Tile
2. Placed in racking other alloys wcrc not as readily available.
3. Alkaline cleaned
Initial Screening
4. Hot water rinsed
The CAA parameters varied were (1) immersion
5. Deoxidized time in chromic acid solution, (2) anodizing voltage
6. Cold water rinsed (22 or 40 V), (3) ramp time to desired voltage,
and (4) hot deionized water scaling. It was not
7. Anodized possible to vary the chromic acid solution percentage
8. Cold water rinsed of 7 percent (by weight), because the CAA was
performed in tanks being used for thc production of
9. Dryed (warm air) aircraft parts. Prcvious work (ref. 3) showed minimal
10. Sealed with hot water changes in _ and e of CAA aluminum as a result of
changing the chromic acid solution from 7.5 percent
to 5 percent as other parameters remained constant.
After the foil was vapor degreased, a metal rack
was clamped to tile perimeter of the foil to provide The two alloys that underwent extensive experi-
a secure electrical contact. Tile racking was kept to mentation wcrc 1145 and 6061. The 5024 and 3003
a minimum because the foil under the racking does were available in limited quantities only and there-
not anodize. This unanodized portion is trimmed off fore underwent limited characterization. The solar
after the anodizing process is completed. The rack- absorptance and emittance values as a function of
ing also provides a means for handling the foil during CAA parameters, for all foils evaluated, are shown
thc various cleaning processes performed prior to the in tables II, III, IV, and V. Examination of the re-
anodizing. Sections of A1 foil (1 ft2) were anodized sults from reference 4 shows that
and, after the anodizing was complete, 1'in 2 sam-
plcs were cut from the 1-ft 2 sections to determine . Immersion time and anodizing voltage had the
the optical values. This established the control op- greatest impact on the optical values. The
tical values that could be achieved by following the 1145 alloy anodized at 22 V, 5-minute ramp, and
anodizing parameters of the specifications. Follow-up 50-minute immersion at full voltage achieved the
samples were then fabricated using modified anodiz- targeted optical values (see table II). Increasing
ing parameters in an attempt to achieve the target the voltage to 40 V and decreasing the immersion
optical values. time to 35 minutes also achieved similar optical
values.
Aluminum Foil Selection
. CAA of 6061 did not achieve the targeted optical
The foil selection study was limited to evaluating values (see table III). The solar absorptance was
At foils that could be procured "off the shelf," be- too high (approximately 0.50) after foils were im-
cause extremely large orders are required to procure mersed long enough to achieve the minimum tar-
nonstandard foils. Four A1 foil alloys with various geted emittance of 0.55. Limited testing showed
tempers were available for evaluation as described 5024 and 3003 alloys (tables IV and V) to possess
below. The desired thickness was 3 mils, which was similar traits. Absorptancc values for 6061 were
the lightest weight A1 foil that could be handled con- approximately 40 percent higher than 1145 when
sistently without damage during the CAA processing anodized at the same parameters.