High-speed atomic oxygen irradiation of atomically thin graphene for astronomical applications

.


Introduction
2][3] For example, in the case of X-ray astronomy, instrumental sensitivity for detector and telescope systems is sometimes limited by filter transmission, particularly in the soft energy band below 1 keV.X-rays from astronomical objects, such as black holes, stars, galaxies, and clusters of galaxies, cannot be observed directly from the ground because they are absorbed by the atmosphere of the Earth.For this reason, the observation instruments are mounted on, for example, balloons, rockets, and satellites, to perform observations at altitudes unaffected by the atmospheric absorption.]7,8 The sensitivity of the filter depends on the thickness of the coating material and polymeric films and an open-area ratio basically limited by their physical properties, such as mechanical strength and processing techniques.As for the thickness of the plastic films, it varies from few tens to hundreds of nanometers.Thus, large-area thin filters with higher mechanical strength make it possible to realize high-sensitivity thin-film devices for future astronomical missions.Recently, carbon nanotube (CNT) thin filters with low-energy X-ray transparency, higher mechanical strength, and superior thermal conductance have been proposed for X-ray detectors in space. 9The concept is to utilize superior optical and mechanical properties of CNTs and apply the CNT thin filters in large-area optical blocking filters to protect soft X-ray detectors from the intense optical light of the Sun in astronomical space missions.Bare and aluminum-coated small self-standing pellicles were fabricated successfully, and they exhibited their effectiveness in terms of X-ray transparency quantitatively as a potential innovative material.
In this study, as a new approach of replacing polymeric materials with a new material, we explored the possibility of atomically thin graphene to achieve ultimately high X-ray transparency for applications in astronomy.Graphene is also an allotrope of carbon composed of a monolayer of atoms arranged in a hexagonal lattice structure and known to possess outstanding physical properties, such as exceptionally high mechanical strength and electronic and thermal conductivity, which makes it an ideal material for filters.To verify whether graphene sheets can be applicable to thin filters in future space missions, we need to establish fabrication processes to realize possible large-area graphene-based filters that exhibit sufficient tolerance to survive severe space environments.As a first step of a space environmental tolerance evaluation test with applications in astronomy, we focused on tolerance against atomic oxygen (AO).AO is a major component of the atmosphere at altitudes of 200 to 500 km, and mission instruments on board rockets and satellites, particularly in a low Earth orbit (LEO), are sometimes supposed to be exposed to AO directly.In such a situation, AO is known to give rise to a variety of chemical and physical complex reactions with polymeric materials because the AO flow velocity is expected to be ∼8 km∕s corresponding to ∼5 eV at an altitude of 500 km. 10 Thus, we evaluated the tolerance of uncoated bare graphene for high-speed AO for space applications.
In this paper, we report the irradiation test results for single-layer graphene films under highspeed AO irradiation simulating the space environment.Section 2 describes the graphene sample information and the details of the tests, Sec. 3 describes the evaluation method and results, and Sec. 4 summarizes the results.

Experimental Method
This section describes the preparation of the graphene samples and the experimental method employed in high-speed AO irradiation tests.

Sample
The details of the sample information used in this study are described in this section.We used single-layer graphene (8-mm square) transferred onto a SiO 2 ∕Si substrate (15-mm square) with a thickness of 625 μm.The single-layer graphene was formed on a Cu/sapphire substrate through chemical vapor deposition, 11 followed by the coating and baking of a polymethyl methacrylate (PMMA) protective film at a temperature of 180°C for 1 min.Thereafter, the following processes were carried out: etching of the Cu/sapphire substrate using ammonium persulfate (0.1 M), pure water rinsing, transferring onto the SiO 2 ∕Si substrate, and finally removing the PMMA film using acetone.To reduce the PMMA residue, the PMMA removal process was performed at a temperature of 60°C for 30 min.In addition, to prevent erosion from the edges of the graphene sheets and clarify the irradiation area, the sample was irradiated by high-speed AO on an aluminum jig (50 mm square, 6 mm thick) that had a 1-mm-diameter pinhole at the center, as shown in Fig. 1.

High-Speed AO Irradiation Tests
This section summarizes the experimental setup and conditions for high-speed AO irradiation tests.
In this study, we investigated tolerance for high-speed AO in a low-orbit case with an altitude of ∼500 km corresponding to a relative velocity of ∼8 km∕s and a kinetic energy of ∼5 eV.To achieve our purposes, a laser-detonation AO beam source at Kobe University 12 was used for the high-speed AO irradiation simulations.The expected AO fluence depends strongly on a variety of factors of launch conditions and space environments, such as the location of instruments, the operational altitude of the spacecraft, and a solar activity.4][15] Considering the fact that high-speed AO tolerance for single-layer graphene has not been investigated in detail even though AO adsorption is known to occur on graphene in not so high-speed AO environments, [16][17][18] in this study, we evaluated the behaviors of single-layer graphene by irradiating high-speed AO without a metal coating.Thus, high-speed AO was irradiated with fluence values of 2 × 10 15 , 2 × 10 16 , 2 × 10 17 , 2 × 10 18 , and 2 × 10 19 atoms∕cm 2 .The fluence value of 2 × 10 15 atoms∕cm 2 corresponds to the minimum value in our experimental setup.The energy distribution during irradiation is estimated by measuring the time of flight (TOF) of ionized high-speed AO using a quadrupolar mass spectrometer located downstream of the high-speed AO beam line.The velocity was converted from the TOF and the averaged velocity was calculated to be ∼6 km∕s.The fluence was determined using the mass loss data of a reference polyimide sample, and the flux was 1.8 × 10 15 atoms∕cm 2 ∕shot.These tests were conducted under a vacuum level of ∼10 −6 Pa and almost vertical (∼83 °) AO impact was expected for all of the samples during the irradiation tests. 19The beamline and inside of the sample chamber are shown in Fig. 2.

Analysis and Results
This section describes the evaluation method of the samples before and after the tests and presents the test results.

Analysis
To evaluate the status of single-layer graphene before and after high-speed AO irradiation, we used a Raman spectrometer that is usually employed to evaluate the molecular and crystal structures of carbon materials, such as graphite, CNT, and graphene and a scanning electron microscope (SEM) to observe the surface condition.The Raman spectroscopy measurements were performed using a Renishaw instrument with an exposure time of 3 s, laser power of 5%, and wavelength of 532 nm.Because the laser light is blocked near the edge of the jig aperture, the measurements were performed at the 1 mm center aperture, and their averaged value was adopted for certain points as the initial condition before the tests.Mapping measurements were also performed to estimate an averaged value and a (1-sigma) spatial variation before and after the tests under the identical conditions as described above in the central 200 μm × 200 μm with a pitch of 6 to 10 μm.
We extracted the G and D band intensities through a fitting procedure, individually assuming a single Gaussian function in each Raman spectrum, and finally focused on a D/G ratio, which is an indicator usually used to quantify the degree of disorderliness and structural defects present in carbon materials.The G band (∼1580 cm −1 ) is the primary mode in graphene caused by a planar molecular motion, whereas the D ∼1350 cm −1 band is activated by disorderliness and defects.Thus, the ratio increases with decreasing structural quality of the graphene sample.As supplementary data, the intensity of the 2D mode, which also results from the characteristic mode of graphene, was also obtained as in previous studies on carbon materials.The 2D band intensity is also expected to decrease in a decreasing graphene structure.The SEM observations were performed using a JEOL SEM system with a magnification of ×70 at 2.00 kV.Next, we extracted each intensity of each band in each Raman spectrum and found that the D/G ratios before and after the high-speed AO irradiation changed from 0.02 ± 0.01 to 0.02 ± 0.01, 0.03 ± 0.02 to 0.02 ± 0.01, and 0.04 ± 0.03 to 0.05 ± 0.03 for 2 × 10 15 , 2 × 10 16 , and 2 × 10 17 atoms∕cm 2 , respectively.Thus, we conclude that there are no significant changes.In contrast, the D/G ratio before and after the high-speed AO irradiation changed from 0.04 ± 0.03 to 0.8 ± 0.4 for 2 × 10 18 atoms∕cm 2 drastically in both the averaged value and 1-sigma range.Furthermore, the D/G ratio could not be measured following the 2 × 10 19 atoms∕cm 2 high-speed AO irradiation because no peaks were observed in both the G and D bands, which suggests that very few graphene sheets exist following the 2 × 10 19 atoms∕cm 2 irradiation.We confirmed that the behavior of the 2D band is similar to that of the G band.The observed D/G ratios are depicted and summarized in Fig. 4 and Table 1, respectively.Consequently, to prevent the single-layer graphene sheets from erosion, a special treatment such as coating is needed to survive in the LEO for ≳ a day.Further, as in the case of the Raman spectroscopic results, there is a significant change in the contrast of the SEM images between the irradiated and unirradiated areas following the 2 × 10 18 atoms∕cm 2 irradiation.We could not see the difference in contrast up to the fluence value of 2 × 10 17 atoms∕cm 2 ; however, a clear contrast could be seen at larger fluence values of 2 × 10 18 and 2 × 10 19 atoms∕cm 2 .In particular, in the case of 2 × 10 19 atoms∕cm 2 , the contrast of the irradiated area is consistent with that of the SiO 2 ∕Si substrate area, suggesting that very few graphene sheets exist.The SEM images following the high-speed AO irradiation with fluence values of 2 × 10 17 , 2 × 10 18 , and 2 × 10 19 atoms∕cm 2 are presented in Fig. 5.

Summary
Thin-film devices play a key role in high-energy astrophysics, for example, X-ray astronomy, as passive temperature-control devices and optical blocking filters on board mission payload systems.Polymeric materials such as polyimide have been used, and sometimes thin-film devices limit sensitivity in terms of transmissivity, particularly at the soft X-ray energy range below 1 keV.Thus, we explored the possibility of replacing such polymeric materials with atomically thin graphene to achieve ultimately high X-ray transparency.As a first step, we investigated the tolerance of single-layer uncoated graphene sheets for high-speed AO, expected to be exposed in the LEO.We prepared the single-layer graphene sample and conducted high-speed AO irradiation tests using the laser-detonation AO beam source.We examined the Raman spectral features before and after the tests with fluence values of 2 × 10 15 , 2 × 10 16 , 2 × 10 17 , 2 × 10 18 , and 2 × 10 19 atoms∕cm 2 .It was found that the observed D/G ratios before and after the high-speed AO irradiation were from 0.02 ± 0.01 to 0.02 ± 0.01, 0.03 ± 0.02 to 0.02 ± 0.01, and 0.04 ± 0.03 to 0.05 ± 0.03 for 2 × 10 15 , 2 × 10 16 , and 2 × 10 17 atoms∕cm 2 , respectively, and we conclude that there are no significant changes.However, the D/G ratios changed from 0.04 ± 0.03 to 0.8 ± 0.4 for 2 × 10 18 atoms∕cm 2 drastically in both the averaged value and 1-sigma range.Furthermore, the D/G ratio could not be measured following the 2 × 10 19 atoms∕cm 2 high-speed AO irradiation because no peaks were observed in both the G and D bands, which suggests that the degradation occurs between 2 × 10 17 and 2 × 10 18 atoms∕cm 2 and very few graphene sheets exist following the 2 × 10 19 atoms∕cm 2 irradiation.The SEM images also support this conclusion because there is no significant difference in the image contrast between the irradiated and unirradiated areas up to the 2 × 10 17 atoms∕cm 2 case, and no evidence of the existence of the graphene sheets is seen in the 2 × 10 19 atoms∕cm 2 case.Consequently, to prevent the single-layer graphene sheets from erosion, a special treatment such as coating is needed to survive in the LEO for ≳ a day.The details of the impact of erosion on specific functions such as mechanical strength will be examined in future studies.

Fig. 1
Fig. 1 (a) Overview of the aluminum jigs.The top cover has a 1-mm diameter pinhole at the center to avoid the irradiation for edge parts of the graphene sheets and clarify the irradiation area.Graphene sample on (b) a SiO 2 ∕Si substrate on the sample stage and (c) its schematic diagram.

Fig. 2
Fig. 2 (a) Overview of the high-speed AO irradiation test facility 12 and (b) the single-layer graphene sample and a polyimide reference in the sample chamber.
of the typical observed Raman spectra for each high-speed AO irradiation fluence are presented shown in Fig. 3.The observed spectra with fluence values of 2 × 10 15 , 2 × 10 16 , and 2 × 10 17 atoms∕cm 2 are very similar to each other, whereas the behavior of the 2 × 10 18 atoms∕cm 2 case is found to be different from the others, and no line features are seen for the 2 × 10 19 atoms∕cm 2 case.As for the cases with fluence values of 2 × 10 15 , 2 × 10 16 , and 2 × 10 17 atoms∕cm 2 , very weak D band intensity was detected, whereas relatively very strong G band intensity was confirmed.In the 2 × 10 18 atoms∕cm 2 case, the contrast between the D and G bands becomes much smaller, which suggests that a significant change in terms of the D/G ratio occurs in the fluence values from 2 × 10 17 to 2 × 10 18 atoms∕cm 2 .As for the 2D band, emissions were detected except for the 2 × 10 19 atoms∕cm 2 case, although the relative intensities of the D and G bands changed drastically.The observed Raman spectra with the best fit models are shown in Fig. 6 of Appendix.

Fig. 4 D
Fig. 4 D/G ratios before and after the high-speed AO irradiation tests as a function of the highspeed AO fluence.The ratio after the 2 × 10 19 atoms∕cm 2 irradiation was not plotted because the G and D bands were not observed.

Table 1
Summary of the D/G ratios for each fluence obtained by Raman spectroscopy before and after the high-speed AO irradiation tests.The ratio is not listed for that fluence because the G and D peaks were not observed after the 2 × 10 19 atoms∕cm 2 irradiation.The 1-sigma ranges were estimated based on Raman mapping measurements.