Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2022 Mar 11;38(11):3339–3349. doi: 10.1021/acs.langmuir.1c02605

Formation of Nanoscale Protrusions on Polymer Films after Atomic Oxygen Irradiation: Changes in Morphologies, Masses, and FT-IR Spectra

Aki Goto †,‡,*, Shinichi Yamashita §,‡,*, Masahito Tagawa ∥
PMCID: PMC8945384  PMID: 35276044

Abstract

graphic file with name la1c02605_0009.jpg

Atomic oxygen (AO) is the main component of the residual atmosphere in a low Earth orbit. AO with a translational energy of 5 eV colliding with artificial satellites forms nano- and microscale protrusions on polymeric materials. This study investigated the influences of AO (fluence and velocity distribution) and a polymer’s chemical structure on such surface morphologies. The correlations between samples’ mass losses and positions in the irradiation field of an AO beam were analyzed with polyimide (Kapton) films, a standard reference material for AO fluence measurements. The characterizations of polyethylene (PE), polypropylene (PP), and polystyrene (PS) films were studied using gel permeation chromatography and X-ray diffraction. The sample surfaces were observed using a field emission scanning electron microscope. Nanoscale protrusions were formed on all the samples and were larger but fewer with increasing AO fluence. The numerical density of protrusions formed on PE and PP was lower than that on PS. However, the erosion yields and functional groups of PE, PP, and PS were similar per FT-IR spectra.

1. Introduction

Spacecraft missions are becoming indispensable. In particular, a low Earth orbit (LEO) provides an environment ripe for commercial utilization, especially Earth observations and high-speed communications. The performances of spacecraft and their components are vulnerable to environmental factors in space, such as highly energetic charged particles (protons and electrons)1 and ultraviolet (UV) rays from the Sun.2,3 One of the principal environmental factors in an LEO is the residual atmosphere, whose dominant constituent at an altitude of 200–700 km is atomic oxygen (AO).4 AO collides with the polymers used in thermal control blankets at a relative velocity of 8 km/s, the orbital velocity of a satellite which corresponds to a translational energy of 5 eV.5,6 Such collisions with hyperthermal energy oxidize and erode polymer surfaces, degrading their thermo-optical and mechanical properties.7−9 Before we can ensure successful spacecraft missions in an LEO, evaluating polymers’ durability against AO in that environment will be necessary.

An artificial AO-irradiation apparatus was developed and used to investigate the reaction processes between AO and polymer surfaces.5,7,10−12 Minton and co-workers found that incident AO atoms directly abstract hydrogen atoms from a polymer via a hyperthermal or thermal mechanism, forming OH radicals and volatile reaction products, such as OH and H2O.5,10 They reported that the resultant radical sites on the polymer are susceptible to further reaction with AO; volatile carbon-containing products such as CO and CO2 are formed and desorbed from the surface.5,11 Tagawa and co-workers measured the mass of the AO-irradiated polymer and showed that it increases due to oxidation at low AO fluences (typically 1016 atoms/cm2) and then decreases linearly with increasing fluence.12 The decrease in linear mass would be due to the removal of carbon atoms such as CO and CO2. Furthermore, polymer erosion yields (the eroded volume per AO irradiation) have been investigated using the AO-irradiation apparatus5,6,13−16 and via flight experiments.7,17,18 Erosion yields depend on the polymers’ chemical structures (carbon content, degree of aromaticity, whether hydrocarbon or fluorocarbon, for instance);13−16 however, the differences among different polymers are not significant,5 implying that the reaction processes of hydrocarbon polymers with AO are roughly similar.

Studies with AO-irradiation equipment8,19−22 and exposures in an LEO6,7 have revealed that needle-like nano- and microscale protrusions can form on polymers and carbon-containing materials (e.g., diamond and highly oriented pyrolytic graphite, or HOPG). Other techniques using ion beams,23 plasma,24,25 and templates26 can also produce nano- and microprotrusions on polymer surfaces, adding desirable physical properties such as super-hydrophobicity or hydrophilicity24−27 and high photo-absorption.23 These physical properties depend strongly on morphological characteristics such as aspect ratios and numerical density of the protrusions. If the surface morphology can be determined by AO irradiation, AO might be useful in modifying the surface of a polymer as desired. Banks and co-workers evaluated the roughness of an irradiated surface using the root mean square (RMS) or height of each protrusion and found that it increases with increasing AO fluence.28 Later, Buczala et al. investigated the influence of temperature on the roughness of AO-irradiated polyimide films, showing that surfaces irradiated at low temperatures were rough and those at high temperatures were relatively smooth.19 They deduced that a hyperthermal temperature-independent mechanism dominates at low temperatures and a thermal temperature-dependent mechanism dominates at high temperatures. Moreover, the morphological characteristics depended on the chemical structure of the polymer or carbon-containing material.6,7 In past flight experiments, samples of polyimide, fluoropolymer, diamond, and HOPG were exposed in an LEO. Their surfaces were observed with a scanning electron microscope.6,7 The aspect ratios and numerical densities of the formed protrusions were not quantitatively estimated but differed among the four materials. However, the reason for such differences has not been clarified.

This study aims to clarify the influences of AO fluence and velocity distribution and the chemical structures of polymers on the surface morphologies, mass losses, and Fourier transform infrared spectroscopy (FT-IR) spectra. Three hydrocarbon polymers with a common backbone were selected: polyethylene (PE), polypropylene (PP), and polystyrene (PS), and the molecular weights and crystallinities of their films were evaluated by gel permeation chromatography (GPC) and X-ray diffraction (XRD). AO-irradiated films were observed using a field emission scanning electron microscope, and their mass losses were measured gravimetrically. FT-IR analyzed chemical structural changes. This article discusses factors that determine the morphologies of these AO-irradiated polymer surfaces.

2. Experimental Method

2.1. Sample Preparations

PE, PP, and PS films were cut into 2.5 cm squares as samples for AO irradiation. The polymers consist of only hydrogen and carbon atoms and share a common backbone (Figure S1, Supporting Information). Commercially available films were used, although it was known that the films included additives, such as plasticizers, antioxidants, UV absorbers, and lubricants.29,30 In determining the impurities’ influences indirectly, films from three manufactures were used for each polymer. Low-density PE films were purchased from Sankyo Polyethylene Co., Ltd. (PE-A), Towakako Co., Ltd. (PE-B), Futamura Chemical Co., Ltd. (PE-C), with thicknesses of 23, 25, and 30 μm, respectively. PP films were purchased from Kyowa Sangyo Co., Ltd. (PP-A), Futamura Chemical Co., Ltd. (PP-B), Mitsui Chemicals Tohcello Co., Ltd. (PP-C), whose thicknesses were 35, 30, and 30 μm, respectively. PS films were purchased from Towakako Co., Ltd. (PS-A), Ohishi Sangyo Co., Ltd. (PS-B), and FPCo Alright Co., Ltd. (PS-C), with thicknesses of 25, 30, and 30 μm, respectively. In addition, polyimide (Kapton) films with a thickness of 25 μm were obtained from Du Pont-Toray Co., Ltd. and used to evaluate AO fluence quantitatively from its mass loss using the following equation

2.1. 1

where F is the Kapton-equivalent fluence (atoms/cm2), Δm is the mass loss per unit area (g/cm2), ρ is the density (1.42 g/cm3), and Ey is the erosion yield for AO irradiation (3.0 × 10–24 cm3/atom).5,31

GPC evaluated the molecular weights of the PE, PP, and PS samples with a refractive index detector. The eluent profiles for PE and PP were obtained at 145 °C with o-dichlorobenzene and the flow rate was 1.0 mL/min (HLC-8321GPC/HT, Tosoh Corporation), as shown in Figure S2 (Supporting Information). Those for PS were obtained at 40 °C with tetrahydrofuran (THF), and the flow rate was 0.8 mL/min (Shodex GPC-101, Showa Denko K.K.). The weight and number-averaged molecular weights (Mw and Mn) were evaluated for their PS-equivalent values by calibrating with the eluent profiles of PS standard samples.

The crystallinities were also evaluated for PE and PP, semicrystalline polymers, using an X-ray diffractometer (MiniFlex600, Rigaku Corporation). The XRD profiles were obtained at room temperature in 2θ of 3–60° using Cu Kα radiation (40 kV, 15 mA). The crystallinities were calculated by dividing the area of crystalline peaks (assigned to PE or PP) by that of the amorphous and the PE- or PP-assigned peaks. The detailed results of the peak deconvolutions are shown in Figure S3 (Supporting Information). The average crystallite sizes were also evaluated by the Halder–Wagner method.32 The Bragg angle and widths of the PE- or PP-assigned peaks were substituted for the following equation

2.1. 2

where β is the width of a PE- or PP-assigned peak, θ is the Bragg angle of a PE- or PP-assigned peak, K is the Scherrer constant (0.9), λ is the wavelength of X-ray (0.15418 nm, Cu Kα), L is the crystallite size, and e is the lattice strain. The average crystallite sizes were calculated from the slopes of the regression lines, as shown in Figure S4 (Supporting Information).

2.2. AO Irradiations

Sample films were irradiated with a laser detonation AO beam source. The irradiation apparatus was developed at Kobe University12 based on the original design developed by Physical Sciences Inc.33,34 A schematic view of the apparatus is shown in Figure 1a. A pulsed supersonic valve (PSV) injected pure molecular oxygen (O2) gas through a nozzle. A concave Au mirror located 50 cm from the nozzle focused a beam from a pulsed carbon-dioxide laser (10.6 μm, 5–7 J/pulse) into the nozzle throat. The laser beam and PSV were synchronized. The laser pulse induced the inverse Bremsstrahlung process, leading to partial dissociation of O2 molecules into AO and accelerating the AO/O2 mixture. The AO velocity was controlled by the delay of the laser pulse to O2 injection by opening the PSV. The longer the delay, the more O2 molecules are injected before the arrival of the laser pulse. The velocity of AO is related to the energy that each O2 molecule obtains from the laser pulse. The shot number was counted to control the AO fluence. Note that the AO fluence was carefully estimated with the procedure described below. Figure 1b shows a schematic view of the sample holder, which is 22 cm in diameter and was placed 42 cm from the nozzle, as shown in Figure 1a. A 6 cm hole in the holder allowed the laser light to pass through it. The samples were placed on the holder and held individually by square aluminum frames that make exposed areas constant (1.5 × 1.5 cm2). Five Kapton samples (K-1–K-5) were always mounted on the holder to estimate the AO fluence and the spatial distribution.

Figure 1.

Figure 1

Open in a new tab

Schematics of the (a) AO irradiation apparatus12 at Kobe University and the (b) 22 cm sample holder showing the mounted polymers (PE, PP, and PS) and Kapton samples (K-1–K-5). TMP in (a) is a turbomolecular pump.

The generated beam was characterized using a time-of-flight mass spectrometer12 with a flight length of 211 cm. TOF distributions of the mass-to-charge ratios (m/z) corresponding to AO and O2 were used to calculate their velocity distributions. The timing of the PSV and the laser pulse can be set for different beam characteristics.33 Two beams with different mean velocities were used; the slower and faster ones are called beam 1 and beam 2, respectively. The flux-weighted TOF spectra of AO (m/z = 16) and O2 (m/z = 32) are shown in Figure 2. The distribution of AO was broad for beam 1 but narrow for beam 2. The mean velocities of AO and O2 in beam 1 were 5.3 and 4.1 km/s, respectively, and those in beam 2 were 7.6 and 4.6 km/s, respectively. The ratio of the numbers of AO and O2 strikes was calculated from the peak areas of the TOF distributions, assuming the relative electron ionization cross-sectional ratio of AO/O2 = 27:79.35 The ratios of those for beams 1 and 2 were 70:30 and 93:7, respectively.

Figure 2.

Figure 2

Open in a new tab

TOF spectra of (a) beam 1 and (b) beam 2 [black: AO (m/z = 16), gray: O2 (m/z = 32)].

2.3. Evaluations of Masses, Morphologies, and FT-IR Spectra

The samples were dehydrated in a desiccator at 23 ± 2 °C and a 50 ± 5% humidity for more than 6 h in advance of measuring their masses with an electronic microbalance (XP6, Mettler Toledo Inc., minimum mass: 1 μg, repeatability: 0.8 μg). After the dehydrated samples were removed from the desiccator, their weights increased gradually as the materials absorbed water from the air. To remove the effect of absorbed water, the dry mass was determined with the following procedure. (1) Each sample was removed from the desiccator and placed on the balance within 30 s; (2) immediately after that, the mass was monitored for 152 s; (3) then, the monitored time profile was fitted to a quadratic curve; and (4) the curve was extrapolated to time zero and the intercept was taken as the dry mass.

After the gravimetric measurements, each irradiated sample was cut into pieces ca. 5 × 5 mm2 for the SEM observations and FT-IR measurements. Some pieces were monitored using a field emission scanning electron microscope (S4800, Hitachi High-Tech Corporation) with an accelerating voltage of 5 kV. Before the observations, they were coated thinly with Au by a magnetron sputter coater (MSP-10, Vacuum Device Co., Ltd.) to prevent charging. FT-IR also analyzed some other pieces using an attenuated total reflectance (ATR) method (670-IR, Varian), giving IR spectra only near the surfaces. The penetration depth of IR light (dp) depends on its wavelength (λ), the incident angle (θ), and the refractive indices of an ATR crystal and the sample (n1, n2) as shown below36

2.3. 3

In this study, a Ge crystal (n1 = 4.0) was used and θ was 60°. Assuming n2 is 1.5 (typical of general organic substances), dp was estimated at less than 500 nm in the range of 1000–4000 cm–1.

2.4. Spatial Distributions of AO Fluence

Our ability to accurately measure the AO fluence of each sample for each irradiation was critical. Since AO is electrically neutral in its ground state (3P), it does not produce an electric current. AO with a low translational energy cannot induce ionization, which is utilized in the fluence measurements of high-energy particles. AO fluence is generally estimated using a standard reference material, Kapton. Its erosion yield has been reported to be 3.0 × 10–24 cm3/atom, as mentioned above.5,31 The Kapton-equivalent AO fluence can be calculated from the volume loss and the erosion yield. AO beams generated by laser detonation, one of the most common methods to produce AO on the ground, usually have an inhomogeneous spatial distribution.33,34 A past study reported that the AO fluence, in this case, depended on the radial distance from the beam center.33 In this study, we attempted to estimate the spatial distribution of Kapton-equivalent AO fluence.

There were eight batches of AO irradiations in this study, divided into two groups (Table 1). Beam 1 (the slower beam, Figure 2a) was applied to the four batches (nos. 1–4), and beam 2 (the faster beam, Figure 2b) was applied to the other four batches (nos. 5–8). To obtain samples irradiated with stepwise fluences, some samples in batch 1 were removed from the holder, and the others were irradiated again as batch 2. A similar procedure was repeated for the later batches. As a result, half of the PE, PP, and PS films were irradiated as batches 1 (B1–F1), 3 (B1–F3), 5 (B2–F1), and 7 (B2–F3), while the other half and all the Kapton films were continuously irradiated as batches 1 and 2 (B1–F2), 3 and 4 (B1–F4), 5 and 6 (B2–F2), and 7 and 8 (B2–F4). To aid in the spatial distribution of AO strikes for each condition (B1–F2, B1–F4, B2–F2, and B2–F4), five Kapton films were simultaneously placed on the holder in positions K-1 to K-5, as shown in Figure 1b. The fluences were estimated from the mass losses using eq 1. The estimated fluence at the center position (K-2) is shown in Figure 3a and Table 1. Figure 3b shows the ratios of the mass losses at the five positions (K-1 to K-5) to that at the center position (K-2). The mass loss ratios ranged from 0.5 to 1, implying that the spatial distributions have peaks around the center. The mass loss ratios for beam 1 were higher than for beam 2, showing a dependence on AO velocities. Similar spatial distributions were reported in a past study.33 There was apparent inhomogeneity in the spatial distributions, so it was necessary to evaluate the AO fluence accurately at any position on the holder, and hence, the following was assumed. First, there is a point on the holder where the AO fluence is at a maximum, the AO fluence center. Second, the AO fluence decreases as a function of d, the distance from the AO fluence center. Considering the proportional relationship between the Kapton-equivalent fluence F and the mass loss per unit area Δm (eq 1), Δm was also assumed to decrease as a function of d. Since there is no established theory for such spatial distributions, some of the simplest functions were selected as regression curves: the linear (Δm = αd + β, α ≤ 0, β ≥ 0), quadratic (Δm = αd2 + βd + γ, α ≤ 0, β ≤ 0, γ ≥ 0), and Gaussian (Δm = α exp(−d2/β), α ≥ 0, β > 0) functions. Note that α, β, and γ parameters are adjusted for the fit.

Table 1. AO Irradiation Batches.

no. beam number of shots index for samplesa fluence for K-2/atoms/cm2
1 beam 1 (slower) 8020 B1–F1 B1–F2 1.1 × 1019
2 8014 -
3 32,134 B1–F3 B1–F4 8.4 × 1019
4 48,198 -
5 beam 2 (faster) 4218 B2–F1 B2–F2 1.1 × 1019
6 4219 -
7 17,318 B2–F3 B2–F4 6.9 × 1019
8 25,918 -
Open in a new tab
a

B1 and B2 mean that irradiations were done with beams 1 and 2, respectively. F1, F2, F3, and F4 correspond to the lowest, second lowest, third lowest, and highest AO fluences, respectively.

Figure 3.

Figure 3

Open in a new tab

Mass losses per unit area of Kapton films; (a) absolute values at the K-2 position (right axis: Kapton-equivalent AO fluences calculated from eq 1) and the (b) relative values at the five positions K-1–K-5 normalized with that at K-2.

Based on these assumptions, the spatial distributions were analyzed as follows. An X–Y coordinate system was defined with its origin at the center of the holder (the 22 cm circle in Figure 4a). Grid points with 0.5 cm spacing (the dots in Figure 4a) were considered candidate coordinates as the AO fluence center. Regression analyses were conducted to investigate the correlation between Δm and d. A Python library, SciPy, was used to optimize the parameters to attain the highest determination coefficient (R2), which is defined as follows:

2.4. 4

where Δmi is the measured mass loss per unit area of K-i (i = 1, 2, 3, 4, 5), Δm̂i is the corresponding estimate using the function with optimized parameters, and Δm̅ is the mean of Δmi. This optimization was repeated for all candidate coordinates, and then the coordinate giving the maximum R2 value was regarded as an AO fluence center. Table 2 summarizes the R2 maximum and the coordinates of the AO fluence center. Regardless of the regression curves, the R2 maximum was larger than 0.8, and the AO fluence center was within 1.4 cm. Although each regression curve can reasonably be used to evaluate the AO fluence (shown as Supporting Information as Figure S5), the quadratic curve was chosen because it gave the highest R2 maximum. Note that the AO fluence was evaluated with the three regression curves, and their differences were 3.2% or lower for any position of the PE, PP, and PS samples.

Figure 4.

Figure 4

Open in a new tab

(a) X–Y coordinate system used to analyze the spatial distribution of the AO fluence. The circle is the perimeter of the sample holder; the dots are candidate coordinates for the AO fluence center. Distances d1–d5 are from a candidate coordinate (the black star) to K-1–K-5. (b–e) Distributions of the coefficient of determination (R2) obtained by quadratic regressions.

Table 2. R2 Maxima and the Corresponding Coordinates.

sample B1–F2 B1–F4 B2–F2 B2–F4
beam 1 1 2 2
linear 0.97 (−4.5, 2.0) 0.86 (−4.0, −0.5) 0.90 (−1.0, 0.5) 1.0 (0.5, 0.5)
quadratic 1.0 (−5.0, −1.5) 0.86 (−4.0, −0.5) 0.90 (−1.0, 0.5) 1.0 (0.5, 0.5)
Gaussian 0.99 (−5.0, −1.5) 0.81 (−3.0, −1.5) 0.90 (−1.0, 0.5) 1.0 (0.5, 0.5)
Open in a new tab

Figure 4b–e shows the color maps of the R2 values after optimizing each candidate coordinate. The color maps for the three regression curves are similar, so only the results of the quadratic fit are shown here (the others are found in the Supporting Information as Figures S6 and S7). For beam 1 (Figure 4b,c), the R2 distributions had maxima of 3–5 cm from the origin in the −X direction. For beam 2 (Figure 4d,e), the R2 distributions had maxima close to the origin. The R2 distributions and the AO fluence centers depended on the AO beam conditions. As mentioned above, the laser was directed into the O2 cloud using the Au mirror, but it was not aligned with the beam axis (Figure 1a), so the laser interacts with O2 molecules but not always on the axis. The AO velocity was controlled by the delay of the laser pulse after the opening of the PSV to inject O2. For the faster AO (beam 2), the O2 cloud was compact, so its interaction with the laser would occur close to the beam axis. Thus, the AO fluence center was also close to the beam axis. On the other hand, for the slower AO (beam 1), the O2 cloud was large due to the long delay, and the interaction of the laser and O2 would occur in the larger region. As a result, the AO fluence centers for beam 1 shifted a few centimeters from the beam axis toward the Au mirror.

3. Results and Discussion

3.1. Characterizations of Polymers

Table 3 summarizes the average molecular weights and polydispersity indices (PDIs, Mw/Mn) obtained using GPC. The molecular weight distributions of all polymers were rather broad (PDI of 4 ± 2); however, the samples from three different manufactures were mostly composed of similar molecular weights as seen in the GPC eluent profiles (Figure S2, Supporting Information).

Table 3. Average Molecular Weights and PDI Characterized by GPC (PS-Equivalent Values).

samples Mw Mn PDI (Mw/Mn) eluent temperature and solvent
PE-A 2.5 × 105 6.1 × 104 4.1 145 °C, o-dichlorobenzene
PE-B 1.9 × 105 4.1 × 104 4.7
PE-C 1.6 × 105 6.1 × 104 2.5
PP-A 3.9 × 105 9.3 × 104 4.2
PP-B 3.9 × 105 9.0 × 104 4.3
PP-C 4.7 × 105 7.9 × 104 6.0
PS-A 3.5 × 105 7.8 × 104 4.5 40 °C, THF
PS-B 2.7 × 105 7.0 × 104 3.9
PS-C 3.2 × 105 7.9 × 104 4.1
Open in a new tab

Table 4 lists the crystallinities and crystallite sizes obtained using XRD for the PE and PP samples (semicrystalline polymers). Figure 5a shows the XRD profiles for PE-A to C, which commonly had three diffraction peaks at 21.4–21.7°, 23.6–24.1°, and 36.2–36.5° of 2θ (110, 200, and 020 planes of orthorhombic PE, respectively).37 In addition, only PE-B had other peaks at 9.8 and 29.0°, which cannot be attributed to PE, showing the existence of crystalline impurities. Thus, the peaks of the impurities were omitted in the calculations of the crystallinities. The crystallinities of PE-A to -C were comparable (43, 45, and 48%, respectively). The crystallite sizes of PE-A to -C were comparable as well (10, 11, and 8.1 nm, respectively). As a conclusion, purity was the main difference between the three manufacturers.

Table 4. Crystallinities and Crystallite Sizes Characterized by XRD.

  % crystallinity crystallite size/nm (evaluated by a Halder–Wagner method)
PE-A 43 10
PE-B 45 11
PE-C 48 8.1
PP-A -a -
PP-B 48 7.9
PP-C - -
Open in a new tab
a

The crystallinity and crystallite size for PP-A and PP-C could not be calculated because the peaks were too broad.

Figure 5.

Figure 5

Open in a new tab

XRD profiles of (a) PE and (b) PP. The Miller indices and blue lines show the peaks attributed to PE (orthorhombic) and PP (monoclinic iPP). The peaks shown by triangles could not be assigned to PE or PP.

Figure 5b shows the XRD profiles for PP samples. A broad structure was commonly observed for all samples. Only PP-B had six sharp peaks at 14.1, 16.9, 18.6, 21.6, 25.5, and 28.5° (110, 040, 130, 140, 150, and 012 planes of monoclinic isotactic PP (iPP), respectively).38,39 The crystallinity and crystallite size for PP-B were 48% and 7.9 nm, respectively. However, PP-A and -C had no sharp peak but two broad peaks at 15 and 21°, indicating that they had a mesophase instead of semicrystalline regions.39−41 The mesophase has a hierarchical structure: an intermediate structure between amorphous and crystalline on the crystallographic scale40 and a dense packing of polygonal or spherical shape with a diameter of ∼10 nm at a mesoscopic scale.39,40 PP-C had another peak at 11.5°, which cannot be attributed to PP, showing the existence of crystalline impurities. As a conclusion, the three manufacturers differed in the conformation at the crystallographic scale and purity.

3.2. Surface Morphologies

The AO-irradiated PE, PP, and PS were observed using a field emission scanning electron microscope. Nanoscale protrusions were densely formed on the all-polymer samples. The protrusions seemed to become larger (in height and width) but fewer (smaller numerical densities) with increasing AO fluence (Figure 6). This tendency is consistent with previous studies.20,28 A comparison of beams 1 and 2 indicates that the AO velocity had little effect on the surface morphologies. However, sinkhole structures that were circular and larger than the protrusions were found only for PS irradiated with beam 1 (Figure 6c,f). Sinkholes were observed on the other PS samples (PS-B and -C) irradiated with beam 1 (Figure S8g–i, Supporting Information) but not beam 2 (Figure S9g–i, Supporting Information). These sinkholes were similar to those formed on HOPG by AO42,43 or ion beam (Cs+ or Ar+)44,45 irradiations. In these studies, it was observed that tiny sinkholes (etch pits) appeared first and then grew wider and deeper gradually. Another possibility was the sudden desorption of long-chain volatile products that might locally occur. The reason for the sinkhole production in this study is not clear, but the combination of beam 1 and PS seemed to be necessary.

Figure 6.

Figure 6

Open in a new tab

FE-SEM images of the AO-irradiated (a,d,g) PE-A, (b,e,h) PP-A, and (c,f,i) PS-A; (a–f) irradiated with beam 1 and (g–i) irradiated with beam 2. The AO fluences after the correction of the spatial distribution (see details in the text) were (a) 3.2 × 1019, (b) 2.6 × 1019, (c) 2.5 × 1019, (d) 8.2 × 1019, (e) 7.0 × 1019, (f) 7.1 × 1019, (g) 5.2 × 1019, (h) 5.5 × 1019, and (i) 5.2 × 1019 atoms/cm2. White squares in (c,f) outline examples of the sinkhole structures. The tilt angle was 30° and the scale bars show 1 μm.

It looks that the protrusions’ morphologies such as sharpness, height, width, and numerical density were specific to the type of polymer (Figure 6d–i). The numerical densities of PE and PP were lower than that of PS if they were irradiated with comparable AO fluences. Comparing the same polymers, the numerical densities were similar regardless of their manufacture (Figures S8 and S9, Supporting Information), whose XRD profiles differed as mentioned above. Thus, the polymer’s chemical structure would be a dominant factor in morphological change; in other words, the impurities and the conformation at the crystallographic scale would not be much important.

3.3. Mass Losses of Polymers

The mass losses depended on the AO fluence but not on the manufacturer or mean velocity of AO (Figure S10 in the Supporting Information). For each type of polymer, the correlation between the mass loss and the AO fluence was investigated by linear regression, resulting in R2 values above 0.98. The erosion yield of each polymer was calculated from the slope of the regression line. The unit of cm3/atom is usually used for the erosion yields; however, the unit of g/atom was used here to exclude uncertainty and fluctuation in the density estimation for the small and thin films. The erosion yields for PE, PP, and PS were almost the same (2.3 × 10–24, 2.5 × 10–24, and 2.4 × 10–24 g/atom, respectively). That for Kapton was reported to be much higher (4.3 × 10–24 g/atom).5,31 Based on the difference, it can be said that the polymer’s chemical structure is a principal factor affecting erosion yield. The erosion yields of the PE-based polymers were reported in exposure experiments in an LEO (3.4–3.8 × 10–24, 2.4–2.8 × 10–24, and 3.9 × 10–24 g/atom for PE, PP, and PS, respectively)7,18,46,47 and AO-irradiation experiments with laser detonation (3.5 × 10–24 and 3.8 × 10–24 g/atom for PE and PP, respectively).14 Note that the latter erosion yields were reported in cm3/atom and converted using the densities (PE: 0.92 g/cm3 and PP: 0.91 g/cm3)47 reported in the literature. Similarly, in the past studies, the erosion yields of Kapton were slightly higher than those of the PE-based polymers. A difference in chemical structures could cause this difference. For example, Kapton has oxygen-containing functional groups such as carbonyl (C=O) and ether groups (C–O–C), which PE, PP, and PS do not have. Some reports showed that polymers containing more oxygen atoms tend to result in higher erosion yields with some exceptions.13,14

As mentioned above, the erosion yields in the present experiment did not depend on the mean velocity of AO; however, the breadth of peaks in the AO velocity distribution might have some influence. In an LEO, the mean and the full width at half-maximum (FWHM) of AO translational energy distribution were measured at 5.6 and 1.7 eV, respectively.48 The breadth is considered almost constant because of the AO’s Brownian motion, which has a Maxwell–Boltzmann distribution.6 However, this breadth is not constant for AO beams from the laser detonation mechanism. In a study measuring the erosion yields of PE and PP, the mean and FWHM were 5.4 and 2.0 eV,14 respectively. In this study, the mean and FWHM were 2.3 and 6.5 eV (beam 1) and 4.8 and 6.9 eV (beam 2). Thus, the AO beams used in this study were broader than in others. This could be why the erosion yields of the PE-based polymers in this study were slightly lower than those in the past studies. We obtained erosion yields using the standard value of Kapton as a reference; these values are proportional to mass loss ratios of the samples to Kapton. Tagawa and co-workers reported that the erosion yield of polyimide increases with the increasing translational energy of AO.16 Such a dependence can be different for PE-based polymers and Kapton due to the difference in chemical bonds. There are two possibilities: AO with a lower energy might have decreased the mass losses of the PE-based polymers, or AO with a higher energy might have increased those of Kapton. No difference in the erosion yields was seen in beams 1 and 2, suggesting only that the mass loss ratios of the PE-based polymers to Kapton were maintained despite differences in the AO energy distributions. In addition, the variations in the erosion yields in the exposure experiments7,18,46,47 are large, probably because there are other space-environment factors besides AO.

3.4. Surface Chemical Structures

It is known that the intensities (absorbances) in FT-IR spectra depend on the contacting states of the samples with an ATR crystal, so surface morphologies of AO-irradiated samples affect the peak intensities. To consider this factor, the intensities of the spectra were normalized with the peak intensity at 1463 cm–1 for PE (C–H), 1377 cm–1 for PP (C–H), or 1600 cm–1 for PS (C=C aromatic rings). Figure 7a–c shows the normalized FT-IR spectra of the PE, PP, and PS samples before and after irradiation with beam 1. All the polymers have C–H groups, whose peaks were found in the ranges of 1300–1500 and 2800–3000 cm–1. To extract newly appearing and disappearing peaks, the difference spectra were obtained by subtracting the spectra before irradiation from those after irradiation; these are shown in Figure 7d,e. For all the polymers, the absorbance of C=O (1710–1711 cm–1, ketone, and carboxylic acid), C–O (1079–1101 cm–1, alcohol), and O–H (3370–3456 cm–1, alcohol, and adsorbed water) increased. For PP and PS, the absorbance of C=C (1632–1654 cm–1) increased. However, the peaks of N–H (1645, 3394, and 3191 cm–1) and C=O (1645 cm–1), seen only in PE samples, decreased and almost disappeared. It is known that higher fatty acid amides are often used as lubricants on polymer films,29 so the decrease of the two peaks shows that the lubricants on the PE surface were eroded by AO irradiation. The oxidized functional groups such as C=O and O–H were probably produced through two pathways. First, some AOs react directly with a polymer to break chemical bonds, resulting in broken sites (dangling bonds). The resultant radical sites on the polymer are susceptible to further reactions with AO, forming oxidized functional groups. Second, such oxidized functional groups might be produced after irradiation. Once the irradiated samples are removed from the irradiation chamber, O2 in the air can react quickly with the radical sites on the polymers. Polymeric material oxidation is typically investigated by thermal oxidation (heating in the air). Such changes in chemical structures have been consistently reported in thermal oxidation studies of PE, PP,49−52 and PS.53 In addition, there was no notable difference between the irradiations with beams 1 and 2.

Figure 7.

Figure 7

Open in a new tab

FT-IR spectra of PE-A, PP-A, and PS-A samples. (a–c) Spectra before and after irradiation with beam 1 (black and red lines, respectively). (d–f) Subtraction spectra for beam 1 (solid line) and beam 2 (dashed line). ν, δ, and ρ show stretching, scissoring, and rocking vibrations of a corresponding chemical bond, respectively. The peaks shown by triangles in (a,d) are attributed to higher fatty acid amides. The AO fluences after the correction of spatial distribution were 8.2 × 1019 (PE), 7.0 × 1019 (PP), and 7.1 × 1019 atoms/cm2 (PS) for beam 1 and 5.2 × 1019 (PE), 5.5 × 1019 (PP), and 5.2 × 1019 atoms/cm2 (PS) for beam 2, respectively.

Based on the results obtained with PE-based polymers, the nanoscale morphologies of the AO-irradiated surfaces were affected by the irradiation conditions and the polymer’s chemical structure. However, the erosion yields and chemical structural changes were not. In particular, the numerical densities of the protrusions formed on PE and PP were lower than those on PS irradiated with a comparable AO fluence. The mass loss and FT-IR measurements showed that the protrusions’ numerical density would not be determined only by the total amount of desorbed volatile gases and the types of surface oxidation reactions. Focusing on the effects of the higher-order structures of polymers, there are three possible explanations for the difference in the numerical densities, and each may contribute to some extent.

First, the heterogeneity of the polymer’s higher-order structure might limit the defect sites triggered to form protrusions. PE and PP were semicrystalline polymers with amorphous and semicrystalline (lamellar) regions. The PE-A to -C and PP-B crystallinities were 40–50%, and their average crystallite sizes were 10 nm, evaluated by XRD. Although the crystallinities and crystallite sizes for PP-A and -C could not be quantitatively evaluated, their XRD profiles showed that both samples were mainly composed of mesophase iPP, which has a dense polymer packing with a ∼10 nm diameter.39,40 These results suggest that all PE and PP samples were heterogeneous at several tens of the nanometer scale. Such a heterogeneity might limit the nanoscale erosion sites, leading to lower protrusions’ numerical densities. However, PS-A to -C were composed of general-purpose polystyrene, which has no semicrystalline regions. The heterogeneity of the PS samples would be much lower, so the defects might occur randomly without any limits, leading to a higher numerical density. Second, the polymer chains’ motion might affect the scattering behaviors of AO on the surface. PE and PP have lower glass-transition temperatures (Tg) than PS. The Tg values of PE, PP, and PS are reported as 140–230,54,55 260–266,54,56 and 360–370 K,54,55 respectively, indicating that PE and PP were rubber-like and PS was glassy during the AO irradiations conducted at room temperature. Thus, the micro-Brownian motion of PE or PP is greater than that of PS. Due to the micro-Brownian motion, PE and PP surfaces are rougher than PS surfaces for injected AO. It was reported that the angular distributions of scattering components could be broader, probably due to multiple bounces of AO at the “rough” surface.57−60 Such scattered AOs would induce erosion in directions parallel to the surface in the case of PE and PP, forming protrusions in a lower numerical density. Third, the polymer’s free-volume holes might affect the penetration depth of AO on the surface. Note that the free-volume holes are the volumes of the regions not occupied by polymer chains. Studies using positron annihilation lifetime spectroscopy (PALS) reported that the free volume holes of PE and PP are larger than those of PS at room temperature.54,61 The average radii (R) of the free volume holes were reported as 0.32, 0.31, and 0.29 nm for PE, PP, and PS, respectively, with the Tao–Eldrup model.61 AO is expected to penetrate more deeply into PE and PP than into PS. If so, the defect sites would be fewer on the outermost surface, forming protrusions with a lower numerical density.

4. Conclusions

As a basis for application to surface modifications of polymers, this study aimed to clarify the influences of AO-irradiation conditions (fluence and velocity distribution) and the chemical structure of polymers on surface morphologies.

First, the spatial distribution of Kapton-equivalent AO fluence was determined. Kapton films were irradiated with beam 1 (slow beam) or beam 2 (fast one), and their mass losses were measured. The mass loss ratios at the five positions (K-1 to K-5) to that at the center position (K-2) of the sample holder were in the range of 0.5–1, implying that the spatial distributions have peaks around the center. The following assumptions were made to evaluate the AO fluence at any position on the holder: (1) there exists a point where the AO fluence takes a maximum, the AO fluence center, and (2) the AO fluence decreases as a function of the distance from the AO fluence center. Using these assumptions, the correlations between the mass losses and positions of the Kapton samples were analyzed by linear, quadratic, and Gaussian regressions. Consequently, each regression curve gave a reasonable evaluation of the AO fluence. This study used the quadratic curve because it gave the highest R2.

For each PE, PP, and PS, three films from different manufactures were selected as samples and characterized with GPC and XRD. The three samples were different in the impurities (for PE and PP) and the conformations at the crystallographic scale (for PP) but similar in the average molecular weights and PDI. All samples were irradiated with beams 1 and 2 and observed using a field emission scanning electron microscope. Nanoscale protrusions were formed on all samples and became larger but fewer with increasing AO fluence. Regardless of the mean velocities of AO, the numerical density of protrusions formed on PE or PP was lower than that on PS irradiated with a comparable AO fluence. The small fraction of impurities and conformation did not affect the morphologies. In addition, the mass and FT-IR measurements showed that the erosion yields and formed functional groups (C=O and O–H, e.g.) were similar among PE, PP, and PS. The protrusions’ numerical densities would not be determined only by the total amount of desorbed volatile gases or the types of surface oxidation reactions. The higher-order structures of the polymer might affect the defect (erosion) sites on the surface, scattering behaviors, or penetration depth of AO, leading to different morphologies. Further investigation is expected to see crystallinity and temperature effects and utilize PALS measurement. The same type of polymers with different crystallinities would be prepared by controlling the manufacture conditions, which might show various protrusions’ numerical density. Scattering of AO would be affected by micro-Brownian motions of polymer chains, which are related to the temperature. PALS can quantify the free volume holes of a polymer, which might detect the penetration depth of AO.

Acknowledgments

We thank the engineers from Advanced Engineering Services Co., Ltd. for supporting the sample preparation and mass measurements. We also thank Matsushita, Y. PhD from NIMS for supporting the measurements and analyses of XRD. In addition, we thank Minowa, T. PhD and Li, J. PhD from NIMS for supporting the measurements and analyses of GPC. This work was supported by Inamori Research Grants and JSPS KAKENHI grant number JP20K14955. A part of this work was conducted at the AIST Nano-Processing Facility supported by the Nanotechnology Platform Program of the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) of Japan. Grant Number JPMXP09F-20-AT-0089.

Glossary

Abbreviations

AO

atomic oxygen

LEO

low Earth orbit

FE-SEM

field emission scanning electron microscopy

FT-IR

Fourier transform infrared spectroscopy

GPC

gel permeation chromatography

XRD

X-ray diffraction

ATR

attenuated total reflectance

PE

polyethylene

PP

polypropylene

PS

polystyrene

PALS

positron annihilation lifetime spectroscopy

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.langmuir.1c02605.

  • Chemical structures of PE, PP, and PS; GPC eluent profiles of PE, PP, and PS; XRD profiles of PE-A to -C and PP-B; Halder–Wagner plots of PE-A to -C and PP-B; mass losses of Kapton films as a function of their distance from the R2-maximum coordinate; distributions of the coefficient of determination (R2) obtained by linear and Gaussian regressions; FE-SEM images of PE, PP, and PS irradiated with beam 1; FE-SEM images of PE, PP, and PS irradiated with beam 2; and mass losses of PE, PP, and PS as a function of AO fluence (PDF)

Author Contributions

The manuscript was written through the contributions of all authors. All authors have approved the final version of the manuscript. All authors contributed equally.

The authors declare no competing financial interest.

Supplementary Material

la1c02605_si_001.pdf (1.4MB, pdf)

References

  1. Srour J. R.; McGarrity J. M.. Radiation Effects on Microelectronics in Space. Proceedings of the IEEE; IEEE: New York, U.S., 1988; Vol. 76, pp 1443–1469. [Google Scholar]
  2. ASTM International . Solar Constant and Zero Air Mass Solar Spectral Irradiance Tables; ASTM E490-00a; ASTM: West Conshohocken, PA, 2019. [Google Scholar]
  3. Joyce A. D.Low Earth Orbital Atomic Oxygen and Ultraviolet Radiation Effects on Polymers. NASA/TM-103711; National Aeronautics and Space Administration, Lewis Research Center: Cleveland, OH, 1991. https://ntrs.nasa.gov/api/citations/19910009981/downloads/19910009981.pdf?attachment=true. [Google Scholar]
  4. Picone J. M.; Hedin A. E.; Drob D. P.; Aikin A. C. NRLMSISE-00 empirical model of the atmosphere: Statistical comparisons and scientific issues. J. Geophys. Res.: Space Phys. 2002, 107, SIA 15-1–SIA 15-16. 10.1029/2002JA009430|. [DOI] [Google Scholar]
  5. Minton T. K.; Garton D. J.. Dynamics of atomic-oxygen-induced polymer degradation in low earth orbit. Chemical Dynamics in Extreme Environments: Advanced series in Physical Chemistry; World Scientific, 2001; Vol. 11; pp 420–489. [Google Scholar]
  6. Banks B. A.; de Groh K. K.; Miller S. K.. Low Earth Orbital Atomic Oxygen Interactions with Spacecraft Materials. NASA/TM-2014-213400; National Aeronautics and Space Administration, Glenn Research Center: Cleveland, OH, 2014. https://ntrs.nasa.gov/api/citations/20040191331/downloads/20040191331.pdf. [Google Scholar]
  7. de Groh K. K.; Banks B. A.. Atomic Oxygen Erosion Data From the MISSE 2-8 Missions. NASA/TM-2019-219982; National Aeronautics and Space Administration, Glenn Research Center: Cleveland, OH, 2019. https://ntrs.nasa.gov/api/citations/20190025445/downloads/20190025445.pdf. [Google Scholar]
  8. Shimamura H.; Nakamura T. Mechanical properties degradation of polyimide films irradiated by atomic oxygen. Polym. Degrad. Stab. 2009, 94, 1389–1396. 10.1016/j.polymdegradstab.2009.05.013. [DOI] [Google Scholar]
  9. Shimamura H.; Nakamura T.. Mechanical Properties Degradation and its Prediction Methods of Polyimide Films Exposed to Space Environment. JAXA/RR-10-009; Japan Aerospace Exploration Agency: Ibaraki, Japan, 2011. https://jaxa.repo.nii.ac.jp/?action=pages_view_main&active_action=repository_view_main_item_detail&item_id=2062&item_no=1&page_id=13&block_id=21. [Google Scholar]
  10. Zhang J.; Garton D. J.; Minton T. K. Reactive and inelastic scattering dynamics of hyperthermal oxygen atoms on a saturated hydrocarbon surface. J. Chem. Phys. 2002, 117, 6239–6251. 10.1063/1.1460858. [DOI] [Google Scholar]
  11. Garton D.; Minton T.; Alagia M.; Balucani N.; Casavecchia P.; Volpi G.. Atomic oxygen interactions with saturated hydrocarbon surfaces. In Defense and Space Programs Conference and Exhibit - Critical Defense and Space Programs for the Future, Huntsville, AL, U.S., September 23-25, 1997; AIAA: Reston, VA, U.S., 1997. [Google Scholar]
  12. Kinoshita H.; Tagawa M.; Yokota K.; Ohmae N. Nonlinear Phenomena in the Mass Loss of Polyimide Films under Hyperthermal Atomic Oxygen Beam Exposure. High Perform. Polym. 2001, 13, 225–234. 10.1088/0954-0083/13/4/301. [DOI] [Google Scholar]
  13. Iskanderova Z. A.; Kleiman J. I.; Gudimenko Y.; Tennyson R. C. Influence of content and structure of hydrocarbon polymers on erosion by atomic oxygen. J. Spacecr. Rockets 1995, 32, 878–884. 10.2514/3.26699. [DOI] [Google Scholar]
  14. Minton D. J.; Stockdale D. P.; Lee D. C.; Yu L.; Minton T. K., Probing the effects of molecular structure on the erosion of hydrocarbon-based polymers by atomic oxygen. Proceedings of the 10th International Symposium on Materials in the Space Environment and the 8th International Conference on Protection of Materials from Space, Collioure, France, June 19-23, 2006; ESA Publications Div.: Noordwijk, The Netherlands, 2006. [Google Scholar]
  15. Zhang J.; Lindholm N. F.; Brunsvold A. L.; Upadhyaya H. P.; Minton T. K.; Tagawa M. Erosion of FEP Teflon and PMMA by VUV radiation and hyperthermal O or Ar atoms. ACS Appl. Mater. Interfaces 2009, 1, 653–660. 10.1021/am800186m. [DOI] [PubMed] [Google Scholar]
  16. Tagawa M.; Yokota K.; Kishida K.; Okamoto A.; Minton T. K. Energy Dependence of Hyperthermal Oxygen Atom Erosion of a Fluorocarbon Polymer: Relevance to Space Environmental Effects. ACS Appl. Mater. Interfaces 2010, 2, 1866–1871. 10.1021/am100201k. [DOI] [Google Scholar]
  17. Visentine J. T.Atomic Oxygen Effects Measurements for Shuttle Missions STS-8 and 41-G; NASA/TM-100459; National Aeronautics and Space Administration, Jonson Space Center: Houston, TX, 1988. https://ntrs.nasa.gov/api/citations/19890004960/downloads/19890004960.pdf. [Google Scholar]
  18. Yi G. T.; de Groh K. K.; Banks B. A.; Haloua A.; Imka E. C.; Mitchell G. G.. Overview of the MISSE 7 Polymers and Zenith Polymers Experiment After 1.5 Years of Space Exposure. NASA/TM-2013-217848; National Aeronautics and Space Administration, Hathaway Brown School: Shaker Heights, OH, 2013. https://ntrs.nasa.gov/api/citations/20130011784/downloads/20130011784.pdf. [Google Scholar]
  19. Buczala D. M.; Brunsvold A. L.; Minton T. K. Erosion of Kapton H by Hyperthermal Atomic Oxygen. J. Spacecr. Rockets 2006, 43, 421–425. 10.2514/1.16402. [DOI] [Google Scholar]
  20. Banks B.; Miller S.; de Groh K.; Chan A.; Sahota M.. The Development of Surface Roughness and Implications for Cellular Attachment in Biomedical Applications. NASA/TM-2001-211288; National Aeronautics and Space Administration; Glenn Research Center: Cleveland, OH, 2001. https://ntrs.nasa.gov/api/citations/20020011682/downloads/20020011682.pdf. [Google Scholar]
  21. Shimamura H.; Nakamura T. Investigation of degradation mechanisms in mechanical properties of polyimide films exposed to a low earth orbit environment. Polym. Degrad. Stab. 2010, 95, 21–33. 10.1016/j.polymdegradstab.2009.10.017. [DOI] [Google Scholar]
  22. Shpilman Z.; Gouzman I.; Grossman E.; Shen L.; Minton T. K.; Paci J. T.; Schatz G. C.; Akhvlediani R.; Hoffman A. Oxidation and Etching of CVD Diamond by Thermal and Hyperthermal Atomic Oxygen. J. Phys. Chem. C 2010, 114, 18996–19003. 10.1021/jp1073208. [DOI] [Google Scholar]
  23. Amemiya K.; Koshikawa H.; Imbe M.; Yamaki T.; Shitomi H. Perfect blackbody sheets from nano-precision microtextured elastomers for light and thermal radiation management. J. Mater. Chem. C 2019, 7, 5418–5425. 10.1039/C8TC06593D. [DOI] [Google Scholar]
  24. Barshilia H. C.; Gupta N. Superhydrophobic polytetrafluoroethylene surfaces with leaf-like micro-protrusions through Ar + O2 plasma etching process. Vacuum 2014, 99, 42–48. 10.1016/j.vacuum.2013.04.020. [DOI] [Google Scholar]
  25. Palumbo F.; Di Mundo R.; Cappelluti D.; d’Agostino R. SuperHydrophobic and SuperHydrophilic Polycarbonate by Tailoring Chemistry and Nano-texture with Plasma Processing. Plasma Processes Polym. 2011, 8, 118–126. 10.1002/ppap.201000098. [DOI] [Google Scholar]
  26. Victor J. J.; Facchini D.; Erb U. A low-cost method to produce superhydrophobic polymer surfaces. J. Mater. Sci. 2012, 47, 3690–3697. 10.1007/s10853-011-6217-x. [DOI] [Google Scholar]
  27. Drelich J.; Chibowski E.; Meng D. D.; Terpilowski K. Hydrophilic and superhydrophilic surfaces and materials. Soft Matter 2011, 7, 9804–9828. 10.1039/C1SM05849E. [DOI] [Google Scholar]
  28. Banks B.; Rutledge S.; Stueber T.; Snyder S.; Norris M.; Banks B.; Rutledge S.; Stueber T.; Snyder S.; Norris M.. Atomic oxygen erosion phenomena. Defense and Space Programs Conference and Exhibit - Critical Defense and Space Programs for the Future, Huntsville, AL, U.S., September 23-25, 1997; AIAA: Reston, VA, U.S., 1997. [Google Scholar]
  29. Owens D. K. Friction of polymer films. I. Lubrication. J. Appl. Polym. Sci. 1964, 8, 1465–1475. 10.1002/app.1964.070080335. [DOI] [Google Scholar]
  30. Vandenburg H. J.; Clifford A. A.; Bartle K. D.; Carroll J.; Newton I.; Garden L. M.; Dean J. R.; Costley C. T. Critical ReviewAnalytical Extraction of Additives From Polymers. Analyst 1997, 122, 101R–116R. 10.1039/A704052K. [DOI] [Google Scholar]
  31. ASTM International . Standard Practices for Ground Laboratory Atomic Oxygen Interaction Evaluation of Materials for Space Application; ASTM E2089-00; ASTM: West Conshohocken, PA, 2000. [Google Scholar]
  32. Halder N. C.; Wagner C. N. J. Separation of particle size and lattice strain in integral breadth measurements. Acta Cryst. 1966, 20, 312–313. 10.1107/S0365110X66000628. [DOI] [Google Scholar]
  33. Caledonia G.; Krech R.; Upschulte B.; Sonnenfroh D.; Oakes D.; Holtzclaw K.. Fast oxygen atom facility for studies related to low earth orbit activities. 17th Aerospace Ground Testing Conference; ARC: Nashville, TN, U.S., July 6–8, 1992. [Google Scholar]
  34. Caledonia G. E.; Krech R. H.; Green B. D. A high flux source of energetic oxygen atoms for material degradation studies. AIAA J. 1987, 25, 59–63. 10.2514/3.9580. [DOI] [Google Scholar]
  35. Yokota K.; Yasuda S.; Mizutani A.; Tagawa M. Relative Electron Impact Ionization Probabilities of O, O2, and Ar Components in Laser-Detonation Hyperthermal Beams. Jpn. J. Appl. Phys. 2013, 52, 038002. 10.7567/JJAP.52.038002. [DOI] [Google Scholar]
  36. Okada T.; Ishige R.; Ando S. Analysis of Thermal Radiation Properties of Polyimide and Polymeric Materials Based on ATR-IR Spectroscopy. J. Photopolym. Sci. Technol. 2016, 29, 251–254. 10.2494/photopolymer.29.251. [DOI] [Google Scholar]
  37. Poly(ethylene) HD. Powder Diffraction File. International Centre for Diffraction Data. (PDF Card: 00-060-0984).
  38. Polypropylene. Powder Diffraction File. International Centre for Diffraction Data. (PDF Card: 00-054-1936).
  39. Konishi T.; Nishida K.; Kanaya T.; Kaji K. Effect of Isotacticity on Formation of Mesomorphic Phase of Isotactic Polypropylene. Macromolecules 2005, 38, 8749–8754. 10.1021/ma050908f. [DOI] [Google Scholar]
  40. Nishida K.; Okada K.; Asakawa H.; Matsuba G.; Ito K.; Kanaya T.; Kaji K. In situ observations of the mesophase formation of isotactic polypropylene-A fast time-resolved X-ray diffraction study. Polym. J. 2012, 44, 95–101. 10.1038/pj.2011.118. [DOI] [Google Scholar]
  41. Mollova A.; Androsch R.; Mileva D.; Gahleitner M.; Funari S. S. Crystallization of isotactic polypropylene containing beta-phase nucleating agent at rapid cooling. Eur. Polym. J. 2013, 49, 1057–1065. 10.1016/j.eurpolymj.2013.01.015. [DOI] [Google Scholar]
  42. Nicholson K. T.; Minton T. K.; Sibener S. J. Spatially Anisotropic Etching of Graphite by Hyperthermal Atomic Oxygen. J. Phys. Chem. B 2005, 109, 8476–8480. 10.1021/jp045960c. [DOI] [PubMed] [Google Scholar]
  43. Nicholson K. T.; Minton T. K.; Sibener S. J. Temperature-dependent morphological evolution of HOPG graphite upon exposure to hyperthermal atoms. Prog. Org. Coat. 2003, 47, 443–447. 10.1016/S0300-9440(03)00134-6. [DOI] [Google Scholar]
  44. Zhu Y.-J.; Hansen T. A.; Ammermann S.; McBride J. D.; Beebe T. P. Nanometer-Size Monolayer and Multilayer Molecule Corrals on HOPG: A Depth-Resolved Mechanistic Study by STM. J. Phys. Chem. B 2001, 105, 7632–7638. 10.1021/jp011377+. [DOI] [Google Scholar]
  45. Zhu Y.; McBride J. D.; Hansen T. A.; Beebe T. P. Controlled Production of Molecule Corrals Using Cesium Ion Bombardment: A TOF–SIMS, XPS, and STM Study. J. Phys. Chem. B 2001, 105, 2010–2018. 10.1021/jp0040300. [DOI] [Google Scholar]
  46. de Groh K. K.; Banks B. A.; Mitchell G. G.; Yi G. T.; Guo A.; Ashmead C. C.; Roberts L. M.; Sechkar E. A.. MISSE 6 Stressed Polymers Experiment Atomic Oxygen Erosion Data; NASA/TM-2013-217847; National Aeronautics and Space Administration, Glenn Research Center: Cleveland, OH, 2013. https://ntrs.nasa.gov/api/citations/20130011518/downloads/20130011518.pdf. [Google Scholar]
  47. de Groh K. K.; Banks B. A.; McCarthy C. E.; Rucker R. N.; Roberts L. M.; Berger L. A. MISSE 2 PEACE Polymers Atomic Oxygen Erosion Experiment on the International Space Station. High Perform. Polym. 2008, 20, 388–409. 10.1177/0954008308089705. [DOI] [Google Scholar]
  48. Cross J. B.; Koontz S. L.; Hunton D. E. Flight mass-spectrometer calibration in a high-velocity atomic-oxygen beam. J. Spacecr. Rockets 1995, 32, 496–501. 10.2514/3.26642. [DOI] [Google Scholar]
  49. George G. A.; Celina M.; Vassallo A. M.; Cole-Clarke P. A. Real-time analysis of the thermal oxidation of polyolefins by FT-IR emission. Polym. Degrad. Stab. 1995, 48, 199–210. 10.1016/0141-3910(95)00035-K. [DOI] [Google Scholar]
  50. Chiellini E.; Corti A.; D’Antone S.; Baciu R. Oxo-biodegradable carbon backbone polymers - Oxidative degradation of polyethylene under accelerated test conditions. Polym. Degrad. Stab. 2006, 91, 2739–2747. 10.1016/j.polymdegradstab.2006.03.022. [DOI] [Google Scholar]
  51. Weon J.-I. Effects of thermal ageing on mechanical and thermal behaviors of linear low density polyethylene pipe. Polym. Degrad. Stab. 2010, 95, 14–20. 10.1016/j.polymdegradstab.2009.10.016. [DOI] [Google Scholar]
  52. Gugumus F. Thermooxidative degradation of polyolefins in the solid state: Part 1. Experimental kinetics of functional group formation. Polym. Degrad. Stab. 1996, 52, 131–144. 10.1016/0141-3910(95)00229-4. [DOI] [Google Scholar]
  53. Israeli Y.; Lacoste J.; Lemaire J.; Singh R. P.; Sivaram S. Photo- and Thermoinitiated Oxidation of High-Impact Polystyrene. I. Characterization by FT-IR Spectroscopy. J. Polym. Sci. 1994, 32, 485–493. 10.1002/pola.1994.080320310. [DOI] [Google Scholar]
  54. Uedono A.; Kawano T.; Tanigawa S.; Ban M.; Kyoto M.; Uozumi T. Transition and Relaxation Processes of Polyethylene, Polypropylene, and Polystyrene Studied by Positron Annihilation. J. Polym. Sci., Part B: Polym. Phys. 1997, 35, 1601–1609. . [DOI] [Google Scholar]
  55. Han J.; Gee R. H.; Boyd R. H. Glass Transition Temperatures of Polymers from Molecular Dynamics Simulations. Macromolecules 1994, 27, 7781–7784. 10.1021/ma00104a036. [DOI] [Google Scholar]
  56. Jarrigeon M.; Chabert B.; Chatain D.; Lacabanne C.; Nemoz G. Multiple transitions in isotactic polypropylene around and above the glass transition. J. Macromol. Sci., Part B: Phys. 2006, 17, 1–24. 10.1080/00222348008212795. [DOI] [Google Scholar]
  57. King M. E.; Saecker M. E.; Nathanson G. M. The thermal roughening of liquid surfaces and its effect on gas-liquid collisions. J. Chem. Phys. 1994, 101, 2539–2547. 10.1063/1.467693. [DOI] [Google Scholar]
  58. Alexander W. A.; Zhang J.; Murray V. J.; Nathanson G. M.; Minton T. K. Kinematics and dynamics of atomic-beam scattering on liquid and self-assembled monolayer surfaces. Faraday Discuss. 2012, 157, 355–374. 10.1039/C2FD20034A. [DOI] [PubMed] [Google Scholar]
  59. Gibson K. D.; Sibener S. J.; Upadhyaya H. P.; Brunsvold A. L.; Zhang J.; Minton T. K.; Troya D. Hyperthermal Ar atom scattering from a C(0001) surface. J. Chem. Phys. 2008, 128, 224708. 10.1063/1.2924126. [DOI] [PubMed] [Google Scholar]
  60. Murray V. J.; Marshall B. C.; Woodburn P. J.; Minton T. K. Inelastic and Reactive Scattering Dynamics of Hyperthermal O and O2 on Hot Vitreous Carbon Surfaces. J. Phys. Chem. C 2015, 119, 14780–14796. 10.1021/acs.jpcc.5b00924. [DOI] [Google Scholar]
  61. Hu Y.; Qi C.; Liu W.; Wang B.; Zheng H.; Sun X.; Zheng X. Characterization of the Free Volume in High-Impact Polystyrene/Polypropylene and High-Impact Polystyrene/High-Density Polyethylene Blends Probed by Positron Annihilation Spectroscopy. J. Appl. Polym. Sci. 2003, 90, 1507–1514. 10.1002/app.2379. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

la1c02605_si_001.pdf (1.4MB, pdf)

Articles from Langmuir are provided here courtesy of American Chemical Society

RESOURCES