A ground-based experimental study of atomic oxygen effects on polymer-matrix composite

Abstract

Polymer-matrix composite is widely used in spacecraft for its excellent performance. However, when it is exposed to space, it is severely affected by the environmental effects of Low Earth Orbit (LEO), especially atomic oxygen (AO), which can cause composite erosion and degradation. In ground-based simulation experiments, the effects of AO on carbon fiber reinforced composites and glass fiber reinforced composites were studied. The following conclusions were drawn when comparing the appearance, mass, and surface morphology of samples before and after the experiment: Firstly, samples experienced severe AO erosion, characterized by substantial mass loss and notable changes in surface morphology. AO preferentially eroded the surface resin, causing the fibers to be exposed. The exposed carbon fibers were also eroded, resulting in surface protuberances. Secondly, the erosion of glass fiber-reinforced composites was primarily due to resin erosion. Because glass fibers are non-reactive to AO, they protected the underlying resin from further erosion. As the AO fluence increased, the material erosion yield decreased.

Keywords

Polymer-matrix composite AO erosion erosion patterns spacecraft

Introduction

Low Earth Orbit, at an altitude of 200 to 1000 km, is the main travelling orbit for several kinds of spacecraft, such as space shuttles, spaceships, and space stations. When spacecraft travel at speeds of 7–8 km/s in orbit, they are impacted by AO. AO is the primary and most reactive species in the low Earth orbit atmosphere. It has a flux of about 10¹²–10¹⁵ atoms/(cm²·s), and the energy that impinges on the spacecraft surface is about 5 eV. Owing to its high oxidising nature, AO can directly react with various materials, leading to the erosion and degradation of spacecraft surface materials. At the same time, the spacecraft’s operating state makes it easier for AO to react with the material. An AO reaction with spacecraft materials may also cause erosion products to stick to the spacecraft. These could be optical parts, solar arrays, or thermal control coatings. This is one of the main sources of the spacecraft’s molecular contamination. The erosion and contamination may affect the proper operation of spacecraft and the success of space missions. Therefore, it is of enormous significance to study the effect of AO on spacecraft.^1–6

Advanced composites, including polymer-matrix composite, metal-matrix composite, and carbon-carbon composite, are widely used in spacecraft for their excellent performance. Among them, fibre reinforced polymer-matrix composite is the most commonly used. The matrix can be epoxy, phenolics, polyimide, polysulfone resin, etc. Generally, the fiber is carbon fiber, glass fiber, boron fiber, etc. Composite materials are increasingly being considered for use in space structures, particularly those requiring high stiffness, such as bus structures and solar array frames, as well as those demanding dimensional stability, such as optical tables and antenna systems. The key material properties of concern to space designers include modulus, thermal expansion coefficient, moisture expansion coefficient, outgassing, and thermal conductivity.¹ In various spaceflight experiments, such as MISSE, LDEF, EOIM-1∼3, LDCE, and COMES/Mir, composite materials have been exposed to LEO environments to assess the cumulative and synergistic effects of orbital conditions on the physical and mechanical properties of the materials. The results of these experiments help infer the behavior of materials under long-term exposure, providing guidance for future material development and system design. Additionally, the impact of AO, as a significant factor in the orbital environment, has also been a key focus of these studies on composite materials.^1–3,7–9 Mazur et al.¹⁰ suggested that the effect of ultraviolet radiation and condensation conditioning does not significantly impact the shear strength properties of carbon fiber-reinforced polyether ketone thermoplastic composites. Kord et al.¹¹ investigated the influence of varying percentages of nanoclay on the corrosion resistance and physicomechanical properties of composites. The results indicated that the flexural strength and modulus exhibited a tendency to increase initially and then decrease. Jian et al.¹² designed and prepared a novel surface treatment agent to enhance interfacial strength through covalent bonding between carbon fibers and the reinforced polyimide matrix. This approach facilitates the uniform dispersion and impregnation of carbon fibers within the polyimide, thereby improving the mechanical properties of carbon fiber/polyimide composites to some extent. Although these studies systematically elucidated the regulatory mechanisms of surface modification strategies on key composite properties, fundamental research regarding their performance evolution under atomic oxygen erosion environments remains notably absent.

Zheng et al.¹³ applied a silane coupling agent (SCA) to the surface of carbon fibers, which improved the AO erosion resistance of CF/EP composites. Chen et al.¹⁴ conducted research in which silane and graphene oxide (GO) were incorporated into poly (p-phenylene-benzodibenzoisoxazole) (PBO) fiber composites, effectively mitigating AO penetration. Yagnamurthy et al.¹⁵ discovered that Epon 862-based epoxy composites exposed to AO and ultraviolet (UV) radiation exhibited significant and non-uniform erosion depths, resulting in an average thickness of the surface damage layer ranging from 5 to 100 μm. Liu et al.¹⁶ concluded that the rate of fracture strength decay in the composite fibers significantly decreased with an increase in the ammonium salt content of polyhedral oligomeric silsesquioxane (POSS) under the same AO exposure conditions. He et al.¹⁷ observed through scanning electron microscopy (SEM) images that the resin on the surface of the trilaminate was almost completely eroded after exposure to AO. The conclusion was that AO will erode and degrade most composites, changing their mass, surface morphology, and other properties.

However, the mechanism of the effect of atomic oxygen on polymer matrix composites remains unclear. Meanwhile, large-scale spacecraft in low orbit faces the problems of de-orbiting fall around the end of life, and disintegrates during reentering back to the earth because they suffer tremendous aerothermodynamics environment and overloads.^18–21 The high-temperature thermo-chemical non-equilibrium gas flows produced by spacecraft re-entering atmospheric surrounding will create cumulative effect on the metal truss and the embedded composite. Under this circumstance, the tests in this paper have been conducted at the ground-based simulation facility in Beihang University to investigate the properties of AO reactions and measure how AO erosion affects various types of polymer-matrix composites.

Experimental facility and samples

Experimental facility

The configuration of the AO effects experimental system is illustrated in Figure 1. This system is a hot cathode filament discharge plasma-type atomic oxygen effects simulation facility and comprises several subsystems, including the hot cathode filament discharge, surface-restrained multiple magnetic diagnostics, and parameter diagnostic subsystems. The operating principles of the experimental system are as follows: the influx of oxygen into the vacuum chamber is regulated by a flow controller to achieve a specific working pressure, which is about 0.01 ∼ 1.0 Pa. In general, the working pressure is 0.1 ∼ 0.3 Pa in the AO - exposure experiments. Subsequently, the cathode filament is heated electrically. And the temperature is risen, the surface of the cathode filament begins to emit electrons. The discharge voltage between the filament and the vacuum chamber walls accelerates these electrons to attain a high-energy state. Oxygen plasma is generated through collision ionizations and the dissociation of oxygen molecules by the electrons. The primary components of the mixed gas include O₂, O₂⁺, O, O⁺, and e.

Figure 1.

The schematic diagram of the ground-based AO effects experimental system.

In addition, a significant number of permanent magnets are arranged alternately by polarity (north and south) along the top and side walls of the vacuum chamber. This configuration creates a surface-restrained magnetic field at the chamber’s surface, which reflects electrons traveling toward the chamber walls back into the plasma. Consequently, the density of the electrons is increased. Furthermore, the magnetic field is present only near the surface of the chamber walls. The intensity of the magnetic field decreases to 1% of the surface intensity at a distance of 4 cm from the chamber surface (the chamber diameter is 30 cm).

The key characteristics of this system are as follows: (1) the filament has a long life span, reaching up to 50–100 h; (2) the electron density can be increased by two orders of magnitude, reaching 10¹⁰/cm³, due to the confinement provided by the multiple magnetic fields; (3) the plasma exhibits excellent uniformity, as measurements indicate that it remains uniform within a diameter of 16 cm in the vacuum chamber; and (4) the AO flux is approximately 10¹⁷ atoms/(cm²·s), which is significantly higher than the ionic oxygen flux of 10¹⁴ ions/(cm²·s). A detailed introduction to the system’s configuration and the operational characteristics of this AO effects experimental system can be found in Refs. 22–24.

Samples and analyzing devices

The samples consist of two kinds of carbon fiber reinforced composites, T300/648 and M40J/S-2, and two kinds of glass fiber reinforced composites, SW280/E51 and Beta cloth (Teflon impregnated glass fabric). The sample size is approximately 20 × 20 mm.

Toray Inc. manufactures T300, a PAN-carbon fiber with a strength of 2758 MPa and a modulus of 228 GPa. 648 is a phenolic-epoxy resin. The T300/648 composite is a cross-ply laminate with 16 plies.

M40J is also a PAN-carbon fiber manufactured by Toray Inc. The strength and modulus are 4400 MPa and 400 GPa, respectively. S-2 is a dicyandiamide epoxy resin. There is only one direction of carbon fibers and one ply of M40J/S-2 composite laminate, which is different from the T300/648 sample.

SW280 is a high-strength glass fiber fabric manufactured by the Nanjing Fiberglass Research and Design Institute. E51 is a bisphenol-A based epoxy resin commonly used in the preparation of composite materials.

Beta cloth, made of Teflon-impregnated glass fabric, is a commonly used thermal control material in spacecraft. It can be used as the outer layer of multilayer insulation blankets (MLI).

To study the reaction characteristics, the composite samples were directly exposed in the ground-based simulation facility for AO effects. Before and after the AO experiments, the samples were compared in appearance, mass, and surface morphology. The mass was measured by a DT-100 balance with a sensitivity of 0.00005 g. The surface morphology was measured using a Cambridge LEO 1450 scanning electron microscopy (SEM).

Experimental procedure

Estimation of the atomic oxygen flux

There is a lack of consensus about how to directly and accurately measure the AO flux in a ground-based facility. As a result, for this experiment, Kapton is chosen as the standard material for estimating AO flux. It is well known that Kapton is a commonly used polymer for spacecraft, as well as one of the main materials in space flight and ground-based simulation experiments. While AO can cause severe erosion of Kapton in LEO environments, the erosion yield typically remains unchanged. So the mass loss of Kapton can be used as a basis for calculating the AO flux and a criterion to evaluate the erosion degree of other materials. The AO flux in this paper is the equivalent AO flux calculated from the Kapton mass loss following this equation:

F = Δ M / (ρ AtEy)

(1)

In this equation, F is the AO flux (atoms/cm²·s), ΔM is the Kapton sample mass loss (g), ρ is Kapton density (1.4 g/cm³), A is the sample surface area (cm²), t is the experiment time (s), and E_y is Kapton erosion yield (in LEO, Ey = 3.0 × 10⁻²⁴ cm³/atom).

Experimental condition

In the atomic oxygen ground simulation facility, various exposure environments with different atomic oxygen flux accumulation can be achieved by adjusting parameters such as vacuum chamber pressure, discharge voltage, discharge current, and exposure time. In this facility, oxygen is used as the working gas, which undergoes dissociation or ionization due to electron collisions. Dissociation of an oxygen molecule into two oxygen atoms requires approximately 5 eV of energy, while ionization into an electron and ion requires about 13 eV of energy. The resulting mixture in the facility consists of plasma, oxygen molecules, and oxygen atoms, as shown in Figure 2. As shown in the figure, the use of this device results in a mixture of atoms, ions, and molecules.

Figure 2.

Emission spectra of the plasma.

By applying a voltage to the ion collection plate and measuring the current, the voltage-current curve is obtained, which is then used to calculate the ion oxygen flux in the facility. Additionally, using Kapton mass loss and other measurement methods, the atomic oxygen flux of 0.04 eV is evaluated. Although the speed of the atomic oxygen was not directly measured, the atomic oxygen energy in this type of plasma-based device is relatively low. Literature on similar devices from NASA JSC, GSFC, and MSFC indicates that the atomic oxygen energy is between 0.04 and 0.06 eV, and therefore, this energy range is adopted in the present study. The experimental condition of the AO-exposure experiments in this article, is listed in Table 1.

Table 1.

The experimental condition of the AO-exposure experiments.

Experimental condition	1	2	3	4	5	6	7	8
Pressure (Pa)	0.15	0.15	0.15	0.15	0.15	0.15	0.15	0.15
Discharge voltage (V)	120	120	120	120	120	120	120	120
Discharge current (mA)	120	140	160	160	160	160	140	140
Exposure time (h)	3	7	9	10	11	11	23	35
AO fluence (atoms/cm²)	7.13 × 10¹⁹	1.96 × 10²⁰	2.8 × 10²⁰	3.07 × 10²⁰	3.27 × 10²⁰	3.34 × 10²⁰	6.04 × 10²⁰	9.32 × 10²⁰
Experimental materials	Beta cloth	T300/648, M40J/S-2, SW280/E51	SW280/E51	Beta cloth	T300/648, M40J/S-2	Beta cloth	Beta cloth Teflon SW280/E51	T300/648, M40J/S-2, SW280/E51

Results and discussion

AO effects experiments on T300/648 composite

Before the AO experiment, the composite sample was black in appearance. After the experiment, the sample remained black, but some dark black stripes were visible on its surface. The stripes had the same orientation as the carbon fibers.

Table 2 lists the samples masses before and after the AO experiment. It is clear that the samples have been significantly eroded, resulting in a significant mass loss. The erosion yield of the T300/648 sample in the facility is approximately 2.29 × 10⁻²⁴ cm³/atom, which is identical to the results of the short-term space flight experiments on carbon/epoxy composites (see Table 3).

Table 2.

Mass results of T300/648 samples before and after the AO experiments (ρ = 1.50 g/cm³).

Parameters	Sample
Parameters	1 (size: 19 × 18.3 mm)	2 (size: 20 × 19.6 mm)	3 (size: 19.4 × 18.8 mm)
AO fluence (atoms/cm²)	1.96 × 10²⁰	3.27 × 10²⁰	9.32 × 10²⁰
Pre-experiment (g)	1.75790	1.93220	1.90335
Post-experiment (g)	1.75555	1.92780	1.89270
Mass loss (g)	0.00235	0.00440	0.01065
Erosion yield (Ey) (cm³/atom)	2.30 × 10⁻²⁴	2.28 × 10⁻²⁴	2.09 × 10⁻²⁴

Table 3.

Erosion yield (Ey) of carbon-fiber/epoxy composite in the space flight experiments.¹.

Flight experiment		Sample	Ply	Fluence (atoms/cm²)	Erosion yield (10⁻²⁴ cm³/atom)
Short-term	EOIM – 1	T300/5208		1.0 × 10²⁰	2.6
	EOIM – 1	T300/1034C		1.0 × 10²⁰	2.1
	EOIM – 2	T300/5208	[±45]	3.5 × 10²⁰	2.9
	EOIM – 2	T300/934	[±45]	3.5 × 10²⁰	2.6
Long-term	LDEF A0180	T300/934	[90]₄	1.33 × 10²¹	1.9
Long-term	LDEF A0180	T300/934	[±45]_2s	1.33 × 10²¹	1.25

Figure 3 is an SEM photograph of the T300/648 sample before AO exposure. It shows that the sample surface is smooth, and the fiber is invisible. However, there are some changes after the AO experiment. The samples after the AO tests, which had a fluence of 1.96 × 10²⁰ atoms/cm² and 9.32 × 10²⁰ atoms/cm², are shown in Figures 4 and 7, respectively. After the AO experiment, the surface resin is clearly eroded, and a corduroy-like morphology is formed (see Figure 4(c)). The surface roughness increases, and many carbon fibers are exposed (see Figure 4(a)). However, as AO fluence increases, the changes in surface morphology become more severe, and more carbon fibers are exposed (see Figure 5(a)). The resin and fibers are eroded more significantly. The interface between resin and fibers becomes unclear, and a lot of protuberances appear on the surface of carbon fibers (see Figure 5(c)).

Figure 3.

SEM photograph of the T300/648 sample before the AO experiment (2000×).

Figure 4.

Figure 5.

AO effects experiments on M40J/S-2 composite

M40J/S-2 is also a carbon-fiber/epoxy resin composite. In appearance, the changes in the M40J/S-2 sample are similar to those in the T300/648. Before and after the AO experiment, the sample surface is black. And after the experiment, there were some dark black stripes on the sample surface, whose orientation was the same as the carbon fiber.

Table 4 presents the mass results of the M40J/S-2 samples both before and after the AO experiment. It is obvious that the erosion is significant and the mass loss is considerable. The erosion yield is about 2.61 × 10⁻²⁴ cm³/atom, which is the same as what was found in short-term space flight tests on other types of carbon/epoxy composites (see Table 3).

Table 4.

Mass results of M40J/S-2 samples before and after the AO experiments (ρ = 1.49 g/cm³).

Parameters	Sample
Parameters	1 (size: 20 × 20.1 mm)	2 (size: 19.4 × 19.3 mm)
AO fluence (atoms/cm²)	1.96 × 10²⁰	3.27 × 10²⁰
Pre-experiment (g)	1.21554	1.19320
Post-experiment (g)	1.21235	1.18850
Mass loss (g)	0.00310	0.00470
Erosion yield (Ey) (cm³/atom)	2.64 × 10⁻²⁴	2.58 × 10⁻²⁴

Figure 6 is an SEM photograph of the M40J/S-2 sample before the AO experiment. The sample surface is smooth. However, there are some obvious changes after the AO experiment. SEM photographs of the samples after the AO experiments, with the fluence of 1.96 × 10²⁰ atoms/cm² and 9.32 × 10²⁰ atoms/cm², are shown in Figures 7 and 8, respectively. It can be seen that the changes are similar to the T300/648 sample. After the AO experiment, the surface resin is eroded and a corduroy-like morphology is formed (see Figure 7(c)). Many carbon fibers are exposed (see Figure 7(a)). The changes in surface morphology become more significant as the AO fluence increases. The resin’s erosion is more severe, and more fibers are exposed; the interface between fibers and resin is obscure. Furthermore, the erosion also affects the carbon fibers, leading to the appearance of numerous protuberances on their surface (see Figure 8(c)). This may impact the mechanical performance of the fibers, as well as the composite.

Figure 6.

SEM photograph of the M40J/S-2 sample before the AO experiment (2000×).

Figure 7.

Figure 8.

AO effects experiments on glass SW280/E51 composite

The appearance of the SW280/E51 sample is unchanged before and after the AO experiment. Table 5 lists the mass results of the SW280/E51 samples before and after the AO experiment. The data in the table indicate a significant mass loss of the sample, which suggests that considerable erosion has occurred. The erosion rate is approximately 1.88 × 10⁻²⁴ cm³/atom.

Table 5.

Mass results of SW280/E51 samples before and after the AO experiments (ρ = 1.77 g/cm³).

Parameters	Sample (size: 20 × 20 mm)
Parameters	1 (size: 20 × 20 mm)	2 (size: 20 × 20 mm)	3 (size: 20 × 20.1 mm)
AO fluence (atoms/cm²)	2.80 × 10²⁰	1.96 × 10²⁰	1.96 × 10²⁰
Pre-experiment (g)	0.34965	0.34970	0.34955
Post-experiment (g)	0.34600	0.34710	0.34685
Mass loss (g)	0.00365	0.00260	0.00270
Erosion yield (Ey) (cm³/atom)	1.84 × 10^–24	1.87 × 10⁻²⁴	1.94 × 10⁻²⁴

Figure 9 shows SEM photographs of the SW280/E51 sample prior to AO exposure. Most of the sample surface is epoxy resin, and the crosses of the fiber bundles are exposed. The surface resin is relatively smooth in the photograph with magnifications of 1000 and 2000.

Figure 9.

Figures 10 and 11 are SEM photographs of the sample after the AO experiments, with the fluence of 1.96 × 10²⁰ atoms/cm² and 9.32 × 10²⁰ atoms/cm², respectively. Compared with Figure 9, there are some obvious changes after the AO experiment. The epoxy resin is eroded, and a corduroy-like morphology is formed (see Figure 10(c)). Some fibers appear on the surface (see Figure 10(b)). And with the increase of AO fluence, the erosion of the resin is more severe (see Figure 11(c)), the surface roughness is increased, and more fibers are exposed (see Figure 11(a)). However, because the glass fiber does not react with AO, the surface of the fiber remains smooth, a feature that sets it apart from the carbon fiber/epoxy composite. Its erosion yield is lower than that of carbon fiber composites.

Figure 10.

Figure 11.

AO effects experiments on beta cloth

In appearance, the sample surface is gray and unchanged after the AO experiments. Table 6 presents the mass results of the beta cloth following the AO experiment. The mass loss is considerable, and the average erosion yield is approximately 2.22 × 10⁻²⁴ cm³/atom. This indicates that the AO erosion effect on beta cloth is also severe.

Table 6.

Mass results of beta cloth before and after the AO experiment (ρ = 1.32 g/cm³).

Parameters	Sample (size: 20 × 20 mm)
Parameters	1 (size: 20 × 20.5 mm)	2 (size: 19.9 × 19.7 mm)
AO fluence (atoms/cm²)	3.34 × 10²⁰	3.07 × 10²⁰
Pre-experiment (g)	0.04945	0.05200
Post-experiment (g)	0.04550	0.04845
Mass loss (g)	0.00395	0.00355
Erosion yield (Ey) (cm³/atom)	2.19 × 10⁻²⁴	2.24 × 10⁻²⁴

Figure 12 shows SEM photographs of beta cloth before AO exposure. As shown in Figure 12(a), the glass fiber is bundling, and the fiber bundles are weaved in an orderly manner. Additionally, a layer of Teflon resin is present on the surface. The fiber, the hole, and the resin among the fibers can be seen clearly from the photograph of high magnification. SEM photographs of the sample after the AO experiments, with the fluence of 7.13 × 10¹⁹ atoms/cm² and 3.07 × 10²⁰ atoms/cm², are shown in Figures 13 and 14, respectively. It can be seen that after the low fluence AO experiment, Teflon is significantly eroded, and only a little resin remains. The fiber surface is also smooth. Comparing Figures 12(d) and 14(b), it can be seen that after the long-term AO experiment, the Teflon is almost entirely eroded. This suggests that beta cloth erosion is primarily caused by PTFE Teflon erosion. The effects of AO and vacuum ultraviolet radiation on PTFE Teflon were investigated. The detailed results are introduced in Ref. 21.

Figure 12.

SEM photograph of the beta cloth sample before the AO experiment. (a) 20×. (b) 200×. (c) 1000×. (d) 2000×.

Figure 13.

SEM photograph of the beta cloth sample after the 7.13 × 10¹⁹ atoms/cm² AO experiment. (a) 200×. (b) 1000×.

Figure 14.

Discussion

Based on ground-based experimental studies of LEO AO effects on various types of polymer-matrix composites, we can conclude that the AO action preferentially erodes the polymer matrix of the composites. After that, the fibers are exposed. The exposed carbon fibers are also eroded, and some protuberances appear on the fiber surface. But in glass fiber/epoxy composites, the glass fiber doesn’t react with AO, so the surface of the fiber that is exposed is smooth. This may prevent AO from erosion away the underlayer resin even more.

The detailed AO experiment involving two types of glass fiber composites was conducted. During the experiment, the sample mass was weighed every 2 h. The mass loss and volume loss (mass loss/density) were calculated sequentially. Table 7 presents the results for both mass loss and volume loss, while Figure 15 illustrates the volume loss curve. To highlight the impact of glass fiber on the erosion effects of the composite, Table 7 also includes the erosion results for Kapton and Teflon, two commonly used polymers.

Table 7.

Mass loss ΔM (10⁻⁵ g) and volume loss ΔV (10⁻⁵ cm³) of kapton, teflon, SW280/E51 and beta cloth in AO experiment (10⁻⁵ g).

Time (h)	Fluence (10¹⁹)	Mass loss and volume loss of the AO experiment
		Kapton		Teflon		SW280/E51		Beta cloth
		ΔM	ΔV	ΔM	ΔV	ΔM	ΔV	ΔM	ΔV
1	1.49	25	17.86	90	32.73	60	33.90	65	49.24
3	5.36	90	64.29	340	123.64	140	79.10	180	136.36
5	11.01	185	132.14	595	216.36	235	132.77	250	189.39
7	16.67	280	200	845	307.27	320	180.79	285	215.91
9	22.32	375	267.86	1095	398.18	410	231.64	315	238.64
11	28.27	475	339.29	1345	489.09	505	285.31	350	265.15
13	33.93	570	407.14	1610	585.45	585	330.51	370	280.30
15	38.99	655	467.86	1845	670.91	655	370.06	385	291.67
17	44.05	740	528.57	2085	758.18	715	403.95	405	308.82
19	49.11	825	589.29	2340	850.91	775	437.85	420	318.18
21	55.06	925	660.71	2580	938.18	815	460.45	440	333.33
23	60.42	1015	725	2840	1032	850	480.23	455	344.70

Figure 15.

Volume loss of Teflon, SW280/E51 and beta cloth after the AO experiment.

As can be seen in Table 7, the mass loss of the polymer sample in every 2 h interval is similar. However, the SW280/E51 and beta cloth display disparities. As the AO fluence increases, the mass loss of the glass fiber composite gradually decreases every two-hour interval. In Figure 15, the volume loss of Teflon increases linearly with the AO fluence. But the gradient of the volume loss curve of SW280/E51, and beta cloth decreases, and in the end, those of the beta cloth are nearly horizontal. This indicates that the erosion of these samples slowed over experimental time and the erosion yield decreased.

As is well known, glass fiber is inert to AO. The erosion of beta cloth by AO is primarily caused by the surface Teflon resin. Generally, beta cloth should have a higher surface area exposed to AO than pure Teflon due to its 3D woven structure. Consequently, during the initial exposure period, beta cloth is expected to exhibit a higher mass loss than pure Teflon. However, experimental results revealed that the mass loss rate of beta cloth was consistently lower than that of Teflon. This discrepancy can be primarily attributed to the following mechanisms: (1) Beta cloth is a Teflon-impregnated glass fiber fabric (see Figure 12), and it features a hierarchical porous structure with inter- and intra-bundle voids created by its unique cross-woven architecture; (2) The Teflon on the fiber surfaces exhibits thickness inhomogeneity. During exposure to atomic oxygen, the preferential erosion of the surface Teflon progressively exposes the atomic oxygen-resistant glass fibers, resulting in a characteristic gradual reduction in the overall mass loss rate.

Conclusion

The following conclusions can be drawn from the experimental studies of LEO AO effects on several types of polymer-matrix composite in a ground-based simulation facility:

(1) After the AO experiments, the appearance of the samples was almost unchanged, but the erosion was significant. All the samples had a considerable mass loss and erosion yield. Furthermore, there are some obvious changes in the surface morphology. The action of AO preferentially eroded the surface resin of composites, which was the primary cause of the mass loss. The eroded sample surface formed a corduroy-like morphology, and a lot of fibers were exposed. The surface roughness was increased significantly.

(2) For carbon fiber reinforced composites, the fibers are exposed when the surface resin is eroded. AO will erode the exposed fibers, and some protrusions will appear on the fiber surface. Furthermore, as the AO fluence increased, the erosion of resin and carbon fiber became more severe, and the changes in surface morphology became more obvious.

(3) In the case of the glass fiber reinforced composite, the exposed fibers surface remained smooth after the AO experiment, due to the glass fiber non-reactivity with AO. Furthermore, the exposed glass fibers protected the underlying resin from further AO erosion. The mass of the sample lost more and more slowly, and the erosion yield decreased with the increasing AO fluence. The degradation of polymer-matrix composites in an AO environment, particularly the glass fiber reinforced composite, was primarily caused by resin erosion.

This study conducted AO erosion experiments to examine its effects on glass fiber-reinforced composites and carbon fiber-reinforced composites. The findings offer valuable insights for the design and optimization of spacecraft materials.

Footnotes

Author contributions

Xiaohu Zhao: Investigation, Data curation, Formal analysis, Writing - original draft. Zhihui Li: Writing - review & editing, Project administration, Resources, Supervision. Pingping Yang: Validation, Supervision. Yongdong Liang: Conceptualization, Investigation.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research is supported by National Natural Science Foundation of China (Grant Nos. 11930003, 12332013 and 11325212); The Project of Manned Space Engineering Technology (Grant No. ZS2020103001).

Data Availability Statement

Data will be made available on request.*

References

Silverman

. Spacecraft environmental effects on spacecraft: LEO materials selection guide. Washington, DC: NASA CR-4661, 1995.

Lillis

Youngstrom

Marx

, et al. Space flight experiments to measure polymer erosion and contamination on spacecraft. Washington, DC: NASA TM 2002- 211553, 2002.

Banks

de Groh

Miller

, et al. Lessons learned from atomic oxygen interaction with spacecraft materials in low earth orbit. Washington, DC: NASA TM 2008-215264, 2008.

Ogbonna

Popoola

, et al. A review on recent advances on improving polyimide matrix nanocomposites for mechanical, thermal, and tribological applications: challenges and recommendations for future improvement. J Thermoplast Compos Mater 2023; 36: 836–865.

Liu

, et al. Numerical analysis of deformation and failure mechanism of metaltruss structure of spacecraft under re-entry aerothermodynamic environment. Aero Sci Technol 2022; 130: 107878.

Peng

, et al. Gas-kinetic unified algorithm for computable modeling of Boltzmann equation and application to aerothermodynamics for falling disintegration of uncontrolled Tiangong-no. 1 spacecraft. Adv Aerodyn 2019; 1: 4–21.

Shalin

Minakov

Deev

. Study of polymer composite specimens surface changes after the long-term exposure in space. In: Proceedings of the 7th international symposium on materials in space environment, Toulouse, France, 16–20 June 1997, 375–383.

Tate

Bartlett

Robinson

, et al. Synergies between space research and space operations - examples from the international space station. In: 45th AIAA aerospace sciences meeting and exhibit, AIAA 2007-547, Reno, NV, 8–11 January 2007.

Evans

Robinson

Tate-Brown

, Research on the international space station an overview. In: 47th AIAA aerospace sciences meeting including the new horizons forum and aerospace exposition, AIAA 2009-186, Orlando, FL, 5–8 January 2009.

10.

Mazur

Cândido

Rezende

, et al. Accelerated aging effects on carbon fiber PEKK composites manufactured by hot compression molding. J Thermoplast Compos Mater 2016; 29: 1429–1442.

11.

Kord

Jari

Najafi

, et al. Effect of nanoclay on the decay resistance and physicomechanical properties of natural fiber-reinforced plastic composites against white-rot fungi (Trametes versicolor). J Thermoplast Compos Mater 2014; 27: 1085–1096.

12.

Jian

. Effect of surface treatment on enhancing interfacial strength of carbon fiber/polyimide composites. J Thermoplast Compos Mater 2022; 35(5): 708–719.

13.

Zheng

Zhao

, et al. Improvement of atomic oxygen erosion resistance of carbon fiber and carbon fiber/epoxy composite interface with a silane coupling agent. Mater Des 2016; 109: 171–178.

14.

Chen

Wei

Liu

, et al. Grafting of silane and graphene oxide onto PBO fibers: multifunctional interphase for fiber/polymer matrix composites with simultaneously improved interfacial and atomic oxygen resistant properties. Compos Sci Technol 2015; 106: 32–38.

15.

Yagnamurthy

Chen

, et al. Erosion yield of epoxy-silica nanocomposites at the lower earth orbit environment of the international space station. J Compos Mater 2013; 47: 107–117.

16.

Liu

Guo

Zhao

, et al. Atomic oxygen-resistant polyimide composite fibers based on wet spinning of polyamic acid-POSS ammonium salts. Polym Degrad Stabil 2019; 168: 108959.

17.

Suliga

Brinkmeyer

, et al. Atomic oxygen degradation mechanisms of epoxy composites for space applications. Polym Degrad Stabil 2019; 166: 108–120.

18.

Cui

. Multi-scale asymptotic analysis and computation of the elliptic eigenvalue problems in curvilinear coordinates. Comput Methods Appl Mech Eng 2018; 340: 340–365.

19.

Liang

, et al. Monte Carlo simulation of spacecraft reentry aerothermodynamics and analysis for ablating disintegration. Commun Comput Phys 2018; 23: 1037–1051.

20.

Cui

. Finite element algorithm for dynamic thermoelasticity coupling problems and application to transient response of structure with strong aerothermodynamic environment. Commun Comput Phys 2016; 20: 773–810.

21.

Cui

. Second-order two-scale finite element algorithm for dynamic thermo–mechanical coupling problem in symmetric structure. J Comput Phys 2016; 314: 712–748.

22.

Zhao

Shen

Xing

, et al. A study of the reaction characteristics and mechanism of Kapton in a plasma-type ground-based atomic oxygen effects simulation facility. J Phys D Appl Phys 2001; 34: 2308–2314.

23.

Shen

Zhao

Chen

, et al. Ground-based atomic oxygen effects simulation facility with the filament discharge and bound of magnetic field. ACTA Aeronautics ET Astronautics, Sinica 2000; 21: 425–430 (in chinese).

24.

Zhao

Shen

Xing

, et al. An experimental study of low earth orbit atomic oxygen and ultraviolet radiation effects on a spacecraft material- polytetrafluoroethylene. Polym Degrad Stabil 2005; 88: 275–285.