CORROSION IN SPACE

A. de Rooij
Head of Materials Technology Section
Product Assurance and Safety Department
Materials and Components Technology Division
European Space Agency
European Space Technology and Research Centre
Noordwijk, The Netherlands
Keywords:
atomic oxygen, corrosion, space, low earth orbit, LEO, ATOX, flight testing, onground testing
Abstract
The corrosion in space is described by the effect of atomic oxygen on several
materials. The metal which was most affected is silver. Silver oxidises according to a
linear-parabolic law and due to the thermal stresses the oxide layer continuously
breaks up, resulting in a linear degradation.
Atomic oxygen not only attacks materials in the line of sight of the ram flux, but also
by reflected atomic oxygen.
Many materials form an adherent oxide layer, such as Cu, Al and stainless steel and
are as such protected once this oxide layer is formed. Some materials have a volatile
oxide such as Osmium and many polymers materials. They show a linear behaviour.
1 INTRODUCTION
Corrosion in the space environment can happen during several occasions. In satellites,
manned and unmanned, the environment is present which can contribute to all kinds
of corrosion. The presence of electrolytes in batteries and/or cooling loops can cause
galvanic corrosion, general corrosion and stress corrosion. These corrosion types are
fully equivalent to ones find on ground. In the life of a spacecraft most of the dangers
of corrosion happens on ground during manufacturing, assembly, testing and or
storage. The one corrosion type which is usually not found on ground is corrosion by
atomic oxygen (ATOX). ATOX is considered as one of the most serious hazards to
spacecraft materials and is present at altitudes between 200 and 700 km.
The degrading environment for spacecraft materials includes atomic oxygen,
Ultraviolet (UV) radiation, ionizing radiation, ultrahigh vacuum (UHV), charged
particles, thermal cycles, electromagnetic radiation, micrometeoroids, and man-made
debris, micrometeoroids and orbital debris. Synergistic or separate interactions in this
environment, the spacecraft materials might suffer corrosion, erosion, structure
modification and surface roughening, which can degrade the optical, thermal,
electrical and mechanical properties.
2 WHAT IS ATOMIC OXYGEN
ATOX is produced by the photo-dissociation of molecular oxygen in the upper
atmosphere by solar radiation of wavelength less than or equal to 243 nm [1]. It is the
main constituent of the residual atmosphere in Low Earth Orbit (LEO) as illustrated in
figure 1 [2]. The oxygen atoms have a density of 107 to 108 atoms/cm3 at International
Space Station (ISS) altitude (around 400 km) with a thermal energy of approx. 0.1 eV.

The orbital atomic oxygen density can be calculated with the aid of the MSIS86/CIRA Neutral Thermophere Model of Hedin[3].
Referring to figure 1 one can see the dominant atmospheric constituent concentrations
as a function of altitude. Space vehicles are orbiting with a velocity of 7.78 to 8
km/sec at those altitudes. For satellite-gas collision interactions, this orbital velocity
corresponds to impact gas energies of 0.34, 4.40 and 5.03 eV for hydrogen, nitrogen
and oxygen atoms, respectively.
Concentrating only on the O-atoms and taking into account the solar activity the
maximum and minimum atomic oxygen density as function of altitude is shown in
figure 2. The atomic oxygen flux depends on the altitude, the solar activity, the orbital
inclination and the time of year. The collision with the external surfaces of space
vehicles, orbiting at a velocity of 8 km/s, results in a ram flux of 1014 -1015 Oatoms/(cm2.s).
3 ATOMIC OXYGEN EFFECTS
A number of processes can take place when oxygen atoms strike a spacecraft surface
at orbital velocities. These include chemical reaction with surface atoms or adsorbed
molecules, elastic scattering, scattering with partial or full thermal accommodation,
recombination, or excitation of ram species.
The chemical reaction of ATOX with a surface may cause the formation of volatile
oxides from polymers, carbon, and osmium; or oxides which do not adhere very well
to the surface and tend to spall, as in case of silver. Volatile and spalling oxides
contribute to the erosion of the surface. In several cases the oxide remains adhered to
the surface and is not porous. In these cases the surface oxide is protective and the
underlying material will not erode away. This is the case for most metals such as
aluminium, copper, steel and also for silicon. The oxidation product for most
polymers is a gas and erosion results.
The factors that may influence the erosion yield of materials in space are impact
angle, material temperature, oxide spalling, ATOX flux, ATOX fluence, synergistic
solar radiation, and ATOX impact energy. Following the sequence of Banks and
Rutledge [4]:
3.1.1 ATOX fluence
The erosion yield of Kapton due to space exposure reveals no pronounced dependence
upon ATOX flux or fluence based on information from Space Transportation System
(STS) flight 3, 4, 5, and 8. The corrosion of silver slows down with increasing ATOX
fluence due to the protective oxide layer when still on the surface. No dependence on
ATOX flux is seen based on STS-8, 41 and Long Duration Exposure Facility (LDEF)
information.
3.1.2 ATOX impact energy
The dependence of erosion yield upon impact energy has been measured both in the
ground-based laboratory and in space simulation. In the simulations, an erosion yield
dependence of Kapton to energy E can be expressed as Yield=E0.68 [5]
The energy of atomic oxygen arrival may have a significant influence on the erosion
yield of materials relative to Kapton H. At thermal energies in RF plasma ashers,
materials with high chlorine or fluorine atomic contents have anomalously high

erosion yields relative to in the LEO environment. However, at hyperthermal energies

of ~70 eV, there does not seem to be a clear or simple chemical composition
dependence upon the erosion yields.
The corrosion of silver has been measured in ground ATOX simulators using different
energies and during space flight and no effect is seen. The energies encountered
during these tests are generally too low to have an effect on the metal bond.
3.1.3 Material temperature
The temperature range in the laboratory experiments is too small for evaluation of
Kapton, Mylar and Tedlar to identify any dependence of the erosion yield upon
material temperature.
Corrosion of metals is related to diffusion phenomena of either metal ions of oxygen
ions through the oxide layer. This process is temperature dependent. In the case of
silver the material temperature plays also a role in the phase change between AgO and
Ag2O which occurs around 100C. This was observed several times during ground
testing in oxygen plasmas.
Also transformation temperatures of the oxides such as melting points, sublimation
temperatures play an important role. The large atomic oxygen effect on osmium can
be explained by the low melting point of OsO4 (assumed it is formed) of 40C. It
starts to sublime below its melting point (temperature in LEO is +l00/-100C).
3.1.4 Thermal stresses
Thermal stresses occur in the LEO orbit on the outer surface of satellites due to
periodic in and out of the sunshade during orbiting. At LEO this occurs each 90
minutes and is roughly +100C to -100C. This temperature change causes thermal
stresses in materials and the difference in CTE will cause spalling of the oxide layer
present on metal surfaces and results in constantly exposing fresh material to the
atomic oxygen environment.
3.1.5 Synergistic solar radiation
Synergistic effects such as temperature and ultraviolet exposure may be important for
some materials and not for others. In general metals are not affected by UV exposure.
3.1.6 Impact angle
The effect of ATOX impact angle for Kapton and Mylar in space shows that the rate
of material recession depends on the impact angle with respect to the surface normal,
as (cos).
3.1.7 Indirect attack
Indirect attack of silver by reflected atomic oxygen was seen on STS-5 where the rear
side of the silver samples were oxidised to the amount of 25-50% of that of the front
side. Experiments on Kapton were performed by Banks et.al. [6] also showed the
same effect
3.2 Atomic Oxygen on metals
Experiments on metals were carried out during many STS flights, on LDEF, Eureca
and on ground. The majority of the experiments were conducted on silver, while at the
time of the Hubble Space Telescope development silver was used as solarcell
interconnectors. On-ground experiments were conducted in line with these ones and
extended with other metals.

The oxidation of silver in atomic oxygen is essentially linear-parabolic as postulated

by De Rooij [7] and experimentally confirmed by Chambers et.al. [8]. The silver
oxidises according to described equation until internal stresses in the oxide layer
either to mechanical or thermal excursions causes the oxide layer to crack and
eventually spall and flake off, leaving bare silver exposed to atomic oxygen again.
This process repeats itself until the silver is completely oxidised. The silver oxide is
very fragile and porous with a volume ratio of the oxide to the metal of >2. The
typical oxide thickness at which the spalling and flaking occurs is around 0.5 m.
This periodic flaking of the oxide layer results in an overall linear degradation of the
silver. During the relative short Space shuttle missions the silver reduced 3 to 4.5 m
in thickness depending surface normal to the velocity vector.
An oxide layer model was developed to predict the silver oxidation by atomic oxygen
in the ISS orbit. (figure 3). The mechanism of oxide growth at increased temperatures
is diffusion through the oxide layer. As the diffusion is a temperature-controlled
process, the observation of a substantial amount of oxidation at room temperature
indicates a different oxidation process or at least a parallel process capable of
oxidising the metal in a parabolic manner and operational at low temperatures. The
process responsible for the low temperature oxidation is known as gas flow through
micro-pores.
Several mechanisms of gas transport trough micro-pores can occur. These are free
molecular flow (Knudsen flow), molecular flow, viscous flow and surface diffusion.
At very low pressures where the mean free path of the oxygen atoms is larger than the
dimensions of the pore, the wall collisions within the pore are more frequent than
collisions between other oxygen atoms. This regime is called the free-molecular flow
regime. The density of O-atoms in low earth orbit is approx. 108 atoms/cm3 and the
mean free path is calculated ad 107 m which is much larger then the expected pore
dimension.
The number of atoms flowing through a pore is determined by the conductance of that
pore. These are two aspects:
1) The rate at which atoms enter the pore
2) The probability that these atoms are transmitted though the system.
A gas with a density of n1 at the entrance of a pore and a backflow at the other side
gives a gasflow Fk of

Fk = ( n1 n 2 )

(1)

= a dimensionless probability factor

= the mean velocity of the atoms in the pore

where

The atoms entering the pore loose their velocity due to scattering in the pores during
the first wall collisions. Their speed is than reduced to
The expression for n1 is then

n1 =
where

N
v
Nv

Nv

= the atomic oxygen density

= velocity of the spacecraft
= = the flux of atomic oxygen on the spacecraft

8 RT
M

(2)

from gas kinetic theory

(3)

R = gas constant
T = temperature
M = molecular mass
The probability factor depends on the geometry of the pore and is given by the
Clausing factor

3 Lr H

= 1 +
16 A

(4)

where Lr = the real length of the micropore

H = perimeter of the micropore
A = cross sectional area of the micropore
The pores have a certain path through the oxide layer and this can be related to the
straight direct path y, which is the oxide thickness, by the tortuosity .
(5)
Lr
y
Only a percentage of the surface consists of open pores where the Knudsen flow can
take place.
In summary we can write for the flow of oxygen atoms through porous layer by
substituting eq. 2, 3 and 5 in eq. 1:

3 Hy

Fk = 1 +
16 A

{ n2 }

(6)

Diffusion through short circuits and lattice is governed by the concentration

difference of the diffusion atoms at both sides of the porous layer. At the front side
the density is given by eq. 2. At the metal side the density is n2.
The differential flux of atoms through the oxide layer (with the exception of the
pores) is then written as:
Fd =

D (1 )

n2
y

(7)

Q
where D =diffusion coefficient = D0 exp

RT
The oxygen atoms reaching the metal surface will finally react with the base metal
and form the oxide. This type of reactions is usually linear and with an oxide flux of:

Fo = Kn 2

(8)

where K = rate of metal oxidation = K 0 exp

RT

For a steady state situation the oxygen flux through the pores Fk and from the
diffusional transport Fd must be equal to the oxide flux Fo produced by the oxidation:

Fk + Fd = Fo

(9)

The unit for the oxide flux Fo is related to the oxide growth rate by

dy
= C r Fo
dt ox

(10)

Where
C
= conversion factor from atoms/cm3 to cm3 metal-oxide
= 1.70520 10-23 for silver
r = real volume ration oxide/metal due to porosity

r =

th
(1 )

(11)

In which th = theoretical volume oxide/metal ratio = 1.6 for silver

Substituting eq. 6, 7 and 8 into eq. 9 enables us to express the generally unknown n2,
in the gasflow and diffusion expressions and using this in eq. 10 one obtains the
master equation for the oxide growth governed by the gasflow and diffusion transport
mechanism through the oxide layers.
th
CK
(1 )
dy
(12)
=
Ky
dt ox 1 +
y
+ D(1 )
3Hy
1+
16 A
Equation 12 can solved using standard methods and after some algebra this results in:

KP
thCK.t
K KP D
K KP D
= y 2 c + y 1 + c 2 V c 2
W
2W
W
(1 )
W
W
Where

(13)

3 H
16 A
D = D (1 )

Pc =

W = + Pc D
V=

Wy
D

log 1 +
W
D

This general solution reduces to the well known linear-parabolic equation when
gasflow is absent:
y 2 2 y 2 thC .t
+
=

D K

(14)

In case diffusion is absent equation 13 reduces to

y 2 Pc +

2
2thC .t
y=
K
1

(15)

It may be noted that also the gasflow equation 15 is linear-parabolic. The temperature
dependence of eq. 15 is weak and only present in the linear part, meaning when y is
very small. With thicker oxide layers the gasflow through the pores behaves parabolic
because of the collisions of the atoms with the wall of the pore. The diffusion
equation 14 has the temperature dependence both in the linear and in the parabolic
part. It demonstrates that at low temperatures the gasflow through the pores is the
dominating process, while at high temperature the diffusion is rate controlling.
At a typical ISS orbit the silver loss due to oxidation is calculated as 11.5 m per year
with an oxide thickness of 23 m when no spallation is assumed. When complete
flaking is assumed the maximum loss is 300 m per year.
The solar cell interconnectors are typically silver foil with a thickness of 20 m. A
stress relief loop is formed in the foil to accommodate for the displacements between
the solar cells. This out-of plane-loop is mechanically formed. The atomic oxygen
effect is usually most pronounced on out-of-plane stress relief loops. In this area the
manufacturing stresses and the stresses as a result of the temperature excursions are
large. Past experience showed that in these areas the oxide layer tends to flake off and
a fresh surface is exposed to the atomic oxygen environment. On flat surfaces this
effect is less pronounced while there mainly the difference in volume between the
oxide and the parent metal is responsible for cracking of the oxide layer. The largest
effect is seen on the ram surface. However, also on surface not directly exposed to the
ram fluence, atomic oxygen oxidation is seen as evidenced by the fact that the rear
side of the silver samples were always oxidized although to a lesser extend. In general
this effect is about 25-50% of the front side of the exposed samples.
On way to protect the silver of being oxidized by atomic oxygen is by plating with
gold. Gold is supposed to not be affected by atomic oxygen and when very thin it also
does not influence the working of the stress relief loops. As with all protective
coatings they should remain intact. Deformations required for manufacturing or
during services can jeopardize the integrity of coatings and make them unreliable.
The gold plated interconnectors were finally rejected for out-of-plane stress relief
loops because it was impossible to apply a gold coating with such properties that it
remained intact after forming the stress relief loop. It was seen several times that these
gold layers were either cracking (especially in the out-of-plane loop) or contained
small pinhole type defects.
Through these defects the underlying silver was oxidised. Large holes under the gold
plating were created and silver-oxide was found at the surface.
In the low Earth orbital environment, one should not use materials which suffer from
atomic oxygen corrosion on external surfaces nor on surfaces which can be reached
by atomic oxygen neither should one protect materials such as silver by materials with
low erosion or corrosion yields.

Other metals investigated are Cu, Au, Al, stainless steel, Ta, Al alloys and Mo. These
materials were exposed with and without coatings, such as silicones. Cu was exposed
during the LDEF mission and showed a severe darkening (dark red) of the surface,
changing the optical properties significantly. The cuprous oxide was adherent to the
surface. Other groups investigated Os, Pt, Ni, Fe-alloys and carbon.
3.3 Atomic Oxygen on non-metals
Glasses and ceramics which are based on oxides are expected not to oxidise further.
No degradation is reported.
3.4 Atomic Oxygen on organic materials
It is well documented through orbital and ground measurements that most
hydrocarbon polymers and active metals are highly reactive towards the orbital
atomic oxygen. Materials containing silicones, fluorides, oxides and noble metals are
believed to be moderately inert for short exposures to atomic oxygen; however,
samples recovered from LDEF indicate that many materials are severely degraded on
long-term exposure to atomic oxygen
The tests conducted by Miller et.al. in [9] indicate that environments which produce
synergistic effects with regard to the magnitude of erosion by atomic oxygen exposure
are not the same for each polymer. Each polymer appears to be sensitive to a different
component of the environment. Predicting the atomic oxygen durability of a material
in the space environment can be a very complex task complicated by the fact that each
material may be sensitive to a different synergistic component in the environment.
3.5 Atomic Oxygen on lubricants
Atomic oxygen can gives serious problem in wear-life of MoS2 lubricant. For MoS2
films sputter-deposited at 5070 C, friction measurements showed a high initial
friction coefficient (up to 0.25) for MoS2 surfaces exposed to atomic oxygen, which
dropped to the normal low values after several cycles of operation in air and ultrahigh
vacuum.[10]. Mo-oxide, once formed, act as a protective layer against further atomic
oxygen attack. Delamination of cracking of the oxide layer caused by operation of the
parts can reduce the wear-life of the film.
3.6 Trapping of Atomic Oxygen between Defected Protective Surfaces
Although atomic oxygen attack on internal or interior surfaces may not have direct
exposure to the LEO atomic oxygen flux, scattered impingement can have can have
serious degradation effects where sensitive interior surfaces are present.
Atomic oxygen not only attacks surfaces which are exposed directly to the flux, also
surfaces not directly exposed, but in one way or another accessible to the atoms can
be oxidised. An example is the thermal blanket covering the Ultra High Cosmic Ray
(UHCR) experiment exposed on Eureca. This blanket consists of a four layers with
Teflon as outer layer and a silver and Inconel as internal layers with black paint at the
back side. The internal silver was oxidised by ATOX through a hole created by a
micro-meteorite.
The lack of atomic oxygen protection provided by the aluminized Kapton blanket
cover for the ISS photovoltaic arrays box cushion is thought to be due to due to the
trapping of atomic oxygen between the two aluminized surfaces on the 0.0254 mm
thick Kapton blanket. Defects in the space exposed aluminized surface allow atomic
oxygen to erode undercut cavities.

Prediction of the attenuation of the oxidizing effects of the atomic oxygen flux can be
reasonably achieved through the use of Monte Carlo computational modelling. A twodimensional Monte Carlo model has been developed which has been used for such
predictions. [11]
4 TESTING
The atomic oxygen erosion measurement techniques should follow the protocols
designated in the ASTM procedure E 2089-00 for Standard Practices for Ground
Laboratory Atomic Oxygen Interaction Evaluation of Materials for Space
Applications[12].
4.1 Flight testing
Surfaces of materials used in the Space Shuttle Orbiter payload bay and exposed
during STS-1 through STS-3 were examined after flight.[13]
Films made of carbon, silver, and osmium, were exposed during the Shuttle STS-4
mission examined for the anticipated reaction with atomic oxygen[14]
Bare silver samples and partly coated silver samples were exposed to low earth orbit
conditions during Shuttle flights STS-8 and STS-17.[15] The influence of atomic
oxygen on the amount of erosion-corrosion of metal surfaces, especially silver, was
examined. Protection schemes for silver were also investigated.
The STS Mission 46 carried in orbit a large number of experiments collectively
grouped under the designation of Evaluation of Oxygen Interaction with Materials
(EOIM-III)-third phase
Some results have been given by the Solar Max (1984) returned materials, but not as a
planned experiment.
LDEF was a large cylindrical space experiment rack that exposed various material
samples to outer space for about 5.7 years. It was retrieved from space in 1990.
The retrieval of Eureca gave the opportunity to study the effects of atomic oxygen on
exposed spacecraft surfaces in Low Earth Orbit. Large areas of identical surface
materials, were exposed during the 1992-1993 period,
The first solar array of Hubble Space Telescope was returned after 3 years of
operation. All materials presents were examined including Ag, Inconel, stainless steel,
copper and kapton.
Samples of Kapton H were transported to the International Space Station attached to
the exterior of the ISS during the STS 105 Shuttle mission. The materials were
retrieved during the STS-114 Shuttle mission.
There have been five Materials International Space Station Experiment (MISSE)
passive experiment carriers (MISSE 1-5) to date that have been launched, exposed in
space on the exterior of International Space Station (ISS) and than returned to Earth
for analysis.
The Material Exposure and Degradation ExperimenT (MEDET) will actively monitor
material degradation dynamics in low Earth orbit and acquire information about the
International Space Station environment in terms of contamination, atomic oxygen,
ultraviolet and X-ray radiation, micrometeoroids and debris.
4.2 Ground based Atomic Oxygen testing
The existing AO beam sources can be classified according to the physical process
involved in dissociation of molecular oxygen: laser-induced breakdown, microwave

dissociation, ion neutralization, electron-simulated desorption, and photo-dissociation.

[16]
AO sources have been created and are being used in many research centres around the
world: NASA MSFC, NASA JPL, Montana State University (USA), ESTEC
(Netherlands), ONERA/DESP (France), ITL/UTIAS (Canada), Kobe University
(Japan),
4.2.1 RF Atomic Oxygen Plasma (plasma etcher)
For ground-based AO this study used a 13.56-MHz Plasma asher operated with
oxygen or air. This type of asher produces ATOX by the use of a low pressure, radiofrequency induced, gaseous discharge. The plasma contains multiple potential
aggressor such as atoms, atomic ions, molecular ions, exited species, ozone etc. It
combines the oxidation reactions with possible influences of UV photons, target
heating, electrical charging effects etc
4.2.2 End Hall ion source
The End-Hall Ion Source is a permanent magnet ion source. It produces a beam of
almost exclusively O2+ with negligible O+ or O++ ions. The energy distribution
contained two distinct energy peaks (one at ~40 eV and one at ~85 eV) which resulted
in an average energy of ~70 eV. A water-cooled thermal radiation shield can be used
to shield samples located downstream and prevent sample heating from the heat
radiated by the end Hall source cathode
4.2.3 Plasma neutralization
It utilizes a microwave power source to excite O2 to produce oxygen plasma. The
plasmas become a beam under the effect of an electromagnetic field. The oxygen
plasma beam is accelerated by an electric field which is produced by negatively
biased metallic plate. The accelerated oxygen plasma beam collides with the plate,
and oxygen plasmas are neutralized by the negative charges on the plate and
rebounded to form neutral ATOX beam with impingement kinetic energy,
4.2.4 Pulsed laser breakdown
The source concept selected for ATOX is based on the Laser Pulse Induced
Breakdown (LPIB) principle, studied and experimented in the USA by Physical
Sciences Inc. (PSI, Andover, MA). First, molecular oxygen is forced through the
throat of a nozzle, in the form of gas puffs generated by a fast-switching molecular
beam valve. After partial fill of the nozzle, a high power beam from a pulsed CO2
laser
(wavelength 10.6m) synchronised with the valve is focused onto the injected gas.
This produces breakdown and dissociation of the gas into a very hot (>20,000K)
plasma. The detonation creates a blast wave that propagates through the nozzle with,
as well known from the theory of gas dynamics, conversion of the plasma thermal
energy into directed velocity. The cooling of the expansion allows the plasma to
charge neutralise into oxygen atoms, but the expansion rate is kept sufficiently high,
and the density sufficiently low, to prevent recombination of these atoms into
molecules. A thermally cold (low spread in random velocity - meaning a couple of
thousand Kelvin) ATOX beam with high directed energy finally exhausts the nozzle
and propagates towards the samples.
A regulator system is used to perform thermal cycling between -150C to +150C
with a slope of 10C.min-1 by using liquid nitrogen as a cold source and a heating

resistor which is incorporated into the sample holder. Furthermore the sample holder
can be rotated up to a beam axis angle of attack of 70.
5 CONCLUSION
The effect of atomic oxygen on surfaces in laboratory experiments might be different
from the effect encountered during low earth orbit exposure. During orbit the exposed
samples undergo a thermal cycling sequence of +100/-100C. This might have a
detrimental effect on some oxides. The simultaneous action of atomic oxygen attack
and thermal cycling might be compared with such effects as static stress and corrosion
(stress corrosion) and fatigue and corrosion (corrosion fatigue), where the result of the
combined action is more than the sum of the separate effects. The pure metals are
virtually unaffected by thermal cycling itself, also the effect of atomic oxygen on
metals such as Au, Pd and Al is in the line of expectation compared to the sputtering
yields under 20 KeV O2+ bombardment. Following these yields, the value for silver
(and osmium) has to be in the same order of magnitude. However, a dramatic effect
on silver and osmium is observed in LEO, which is due to the combined action of
atomic oxygen attack and thermal excursions. Therefore, on ground testing of
materials of materials for LEO behaviour should always incorporate both atomic
oxygen attack as well as thermal cycling. To predict materials performance in low
earth orbit, one must accurately simulate the conditions of the low earth orbit
environment or at least understand how the performance of materials under simulated
conditions relates to that in space.

1.
2.

3.
4.

5.
6.
7.
8.
9.

REFERENCES

Bruce A. Banks, et al., Atomic Oxygen Effects on Spacecraft Materials.

NASA/TM2003-212484, 2003.
Daniel R. Peplinski and Graham S. Arnold. INTRODUCTION TO:
SIMULATION OF UPPER ATMOSPHERE OXYGEN SATELLITE
EXPOSURE TO ATOMIC OXYGEN IN LOW EARTH ORBIT. in 13th Space
Simulation Conference. 1984. Orlando, Florida.
Hedin, A.E., MSIS-86 Thermospheric Model,. J. Geophys. Res. , 1987.
A5(92): p. 4649.
Banks, B.A. and S. K.Rutledge, Low earth orbital atomic oxygen simulation
for materials durability evaluation. Proceedings of the Fourth European
Symposium on Spacecraft Materials in the Space Environment,, 1988: p. 371
392.
Ferguson D. C. The energy dependence and surface morphology of Kapton
degradation under atomic oxygen bombardment. in Proc. 13th Space
Simulation Conf.: The Payload-Testing for Success. 1984. Orlando, FL.
Bruce A. Banks, Kim K. de Groh, and Sharon K. Miller, MISSE Scattered
Atomic Oxygen Characterization Experiment, in MISSE Post-Retreival
Conference. 2006: Orlando, Florida.
de Rooij, A., The Oxidation of Silver by Atomic Oxygen. ESA Journal, 1989.
13.
A. R. Chambers, I. L. Harris, and G. T. Roberts, Reactions of spacecraft
materials with fast atomic oxygen. Materials Letters, 1996. 26: p. 121-131.
Sharon K.R. Miller, Bruce A. Banks, and Deborah L. Waters.
INVESTIGATION INTO THE DIFFERENCES IN ATOMIC OXYGEN

10.
11.

12.
13.
14.
15.
16.

EROSION YIELDS OF MATERIALS IN GROUND BASED FACILITIES

COMPARED TO THOSE IN LEO. in Proc. of the 10th ISMSE & the 8th
ICPMSE. 2006. Collioure, France.
J. B. Cross, et al., Atomic oxygen-MoS2 chemical interactions. Surf. Coat.
Technol. 42, 1990.
B. Banks, T. Stueber, and M. Norris, Monte Carlo Computational Modeling of
the Energy Dependence of Atomic Oxygen Undercutting of Protected
Polymers, in Fourth International Space Conference, ICPMSE-4. 1998:
Toronto, Canada.
Bruce A. Banks, et al. Comparison of Atomic Oxygen Erosion Yields of
Materials at Various Energies and Impact Angles. in Proc. of the 10th ISMSE
& the 8th ICPMSE. 2006. Collioure, France.
Leger, L.J., Oxygen Atom Reaction with Shuttle Materials at Orbital Altitudes.
NASA TM 58246, 1982.
Peterson, P.N., R.C. Linton, and E.R. Miller, Results of Apparent Atomic
Oxygen Reactions on Ag, C, and Os Exposed During the Shuttle STS-4 Orbits.
Geophysical Research Letters, 1983. 10: p. 569-571.
de Rooij, A. THE DEGRADATION OF METAL SURFACES BY ATOMIC
OXYGEN. in Proceedings of the Third European Symposium on Spacecraft
Materials in Space Environment. 1985. Noordwijk, The Netherlands,.
J. Kleiman, et al. CO2-LASER ASSISTED ATOMIC OXYGEN BEAM
SOURCES: RESEARCH, DEVELOPMENT AND OPTIMIZATION OF
OPERATIONAL PARAMETERS. in Proc. of the 10th ISMSE & the 8th
ICPMSE. 2006. Collioure, France.

Figure 1: Atmospheric composition in Low Earth Orbit

Figure 2: Atmospheric Atomic Oxygen density in Low Earth Orbit

Figure 3: Model of oxide layer with pore

Figure 4: Silver interconnector (25 m) retrieved from Eureca. It shows oxidised

silver which is flaked off and exposing the underlying fresh material which is
oxidised again.

Figure 5: Gold coated silver is subjected to bending where after it is exposed to

atomic oxygen. The horizontal artefacts are oxidised silver through defects in the
gold plating.

Figure 6: A microsection through a defect in the gold plating. It shows that silver
is also attacked under the gold plating.