Open Access
Add to BibSonomy
Abstract
During their lifetime, total solar irradiance (TSI) radiometers are exposed to different environmental conditions and processes that may degrade the coating that is used as absorber of the incoming radiation to be measured. We were able to reproduce in the laboratory the degradation observed in space with PMO6-type radiometers in instruments like PREMOS, using high-intensity UV light and show that a novel carbon nanotube coating does not degrade under the same conditions.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Starting with the SOVA2 experiment launched on the EURECA platform in 1992, the Physikalisch- Meteorologisches Observatorium Davos (PMOD) has been involved in different TSI missions with PMO6-type radiometers (Schmutz [1], Fröhlich [2], Crommelynck [3]). On all these instruments, the degradation of the coated receivers has been observable during their lifetime and has been described in the literature (Anklin [4], Schmutz [1], Romero [5], Fröhlich [6], Cessateur [7]). Herein, we will focus on three instruments in which PMOD was involved. All three experiments in our study carried two PMO6-type radiometers each. The PMO6 features an inverted conical cavity shaped receiver (Fig. 1). A general overview of the PMO6 and other instruments measuring TSI in space can be found in Solar Irradiance Variations by Fröhlich [8].
Fig. 1. – Schematic image of PMO6-type cavity by Suter [9]. Yellow represents the incoming radiation, and the red areas indicates the position of the electrical heater foil. Part of the radiation is absorbed on the cylindrical wall, where no electrical heating occurs.
1.1 SOVA
As a first example of the degradation effect suffered by the PMO6 radiometers in space, we recall the experiment SOVA2 (SOlar VAriability) as part of the EUropean REtrievable CArrier (EURECA) in 1992 (Romero [5]). On this mission, a set of two PMO6 radiometers can be found, one as an active radiometer and the second one as a backup, which can be compared to find the degradation on the active cavity. Romero [5] found that the ratios at the beginning and at the end of the mission might suggest a loss of sensitivity of about 1 ppm/day.
1.2 VIRGO
The Variability of solar IRradiance and Gravity Oscillations (VIRGO) experiment on the ESA/NASA mission SOHO (SOlar and Heliospheric Observatory) started in 1996 with two different types of TSI radiometers (described by Fröhlich 2]): two with a PMO6 cavity, and a dual-channel radiometer (DIARAD) with a different geometry and coating, i.e., diffuse instead of a specular paint.
In Fig. 2, the level-1 data (before degradation correction) of the four TSI channels on VIRGO are shown. The violet line of PMO6V-A shows the early increment of the absorptivity during the first phase of the experiment, which was explained by Anklin as an adjustment to the space environment [4]. The ratio between channels B and A is shown in Fig. 3. As well, Fig. 2 shows the non-noticeable degradation effect in PMO6V B (red points) due to the much less operative time, as a backup cavity.
Fröhlich [8] studied that the primary PMO6 cavity on the VIRGO instrument has suffered a degradation of around 0.4% over the first ten years, similar in terms of amplitude to that observed in PREMOS, as we will see in the next section, but throughout a larger time frame.
As it is studied by Fröhlich [11], in the VIRGO experiment two types of radiometers were compared. In this experiment it was found that radiometers like PMO6 do not present non-exposure dependent changes, the degradation came from sunlight.
1.3 PREMOS
As a final case for this introduction, the PREcision MOnitoring Sensor (PREMOS) was launched in 2010 as part of the satellite mission PICARD described in Schmutz [1].
This instrument contained two PMO6 radiometers (channels A and B) coated with Aeroglaze Z302. The active channel (A) was continuously operated for almost two years, while the backup channel (B) was only exposed for a few days during this period. Schmutz [1] found that channel A has shown a loss of sensitivity with respect to channel B of around 0.3% along the mission.
Figure 4 shows this 0.3% degradation as described by the ratio between the two cavities during the PREMOS mission, where cavity B is much less exposed than cavity A (after two years of instrument operation, the total exposure time of PREMOS B remains less than 2.5 days).
Fig. 4. – TSI ratio between the active and backup channels of PMO6/PREMOS on PICARD showing a sensitivity loss from Schmutz [1]
Three different frames are noticed in the Fig. 4. An early increase during the first light, a 0.22% decrement during the first hundred days of the mission, and a 0.1% decrement keeping decreasing until the end between the hundredth day and the five hundredth of mission.
A similar degradation as for VIRGO, shown in Anklin [4] and Thuillier [12], was found. As shown, the degradation of the instruments depends on the time that the coating is exposed to the sunlight. An improvement in the receiver absorptance happened in the first days of measurements. This early increase is not a unique effect of the PREMOS instrument, it can be found in other instruments, and we will try to reproduce it in an experimental approach.
As noticed before with the different satellite comparisons, exposure to sunlight appears to be a key factor in the degradation of TSI radiometers. Within the electromagnetic spectrum of the sun radiation, UV and EUV is the most likely source of damage of the coatings. The UV/EUV spectrum at the top of the Earth’s atmosphere is very different from that of the ground, where the radiometers do not suffer degradation due to the exposure, as we experienced in the long-term series measurement performed by the World Standard Group (WSG) at PMOD since it was stablished in 1977 [13].
In order to further study and quantify the expected degradation due to UV/EUV radiation in space, we performed an experiment irradiating probes in the laboratory to the radiation levels expected for a full space mission.
2. UV ageing experiment
A commercially available deuterium lamp (2H) L11798 from Hamamatsu was utilized. The irradiance of this lamp at 50 cm in the wavelength range 100 to 200 nm has been determined from information provided by the manufacturer. Its integrated irradiance in this range is 6 times higher than the solar reference spectrum at the top of the atmosphere (Fig. 5) where the Sun’s irradiance integrated for the same range is 0.100775 W/m2 calculated from the extraterrestrial solar spectrum by Wehrli [14]
For this experiment, the lamp was directly connected to a vacuum tank through a MgF2 window, transparent for the UV region, and the flat samples were located at 50 cm distance, inside and at the bottom of the tank, receiving the direct radiation from the lamp (Fig. 6). The experiment was carried out according to the vacuum conditions for space qualifications of the ECSS protocol according to which the pressure during the test must be maintained below 10- 5 hPa [15].
The experiment started with a reflectance measurement at 633 nm of the samples to be exposed with an integrated sphere setup described in [9]. Then, the samples were located inside the tank and exposed to the UV/EUV radiation for one hour. After this time, the reflectance at 633 nm with the same integrated sphere setup is measured. The reflectance will be measured while observing changes in its value. The exposure time between measurements was increased throughout the experiment as the degradation saturated. The lamp didn’t present any degradation as consequence of the duration of the experiment.
A glossy paint MLS-85-SB from AZ Technologies with two different substrates was used at this first stage of the experiment. One of the samples had a 0.05mm Kapton foil under the black absorber coating, like the PMO6 cavities, whereas the second sample had no Kapton foil in order to assess a potential effect of the aluminum substrate where the coating is applied. Furthermore, we performed a control experiment to check that the degradation process is not produced by vacuum alone. Thus, a third MLS-85-SB (without Kapton) sample was analyzed under the same vacuum environment as the first two samples but without the UV/EUV radiation.
The results of these experiments are shown in Fig. 7. The two datapoints at 0 hours refer to a first measurement before any treatment has been carried out and a second one after introducing the sample in the tank under vacuum conditions but without irradiation with the UV lamp.
Fig. 7. – Degradation suffered by the glossy samples under UV irradiation and vacuum environment without radiation.
A second experiment with the sprayable carbon nanotube paint from Surrey NanoSystems in parallel with a new sample of glossy MLS-85-SB paint from AZ Technologies without Kapton substrate was conducted. An area on the glossy sample was masked, covered by the carbon nanotube sample with a protective paper between these two samples, to serve as a control sample in order to determine whether the degradation was produced by the radiation hitting the coating. The setup was identical to that of the prior experiment. The coatings were exposed simultaneously to the same radiation, time, and conditions in order to compare the degradation effects. The results of these experiments are shown in Fig. 8, which shows the degradation curve obtained by measuring the reflectance of the glossy sample and the masked glossy sample, and in Fig. 9 which displays the reflectance of the carbon nanotubes sample.
Fig. 9. – Degradation suffered by the carbon nanotubes sample under UV irradiation on the same scale as in Fig. 8.
Fig. 10. – Ratio of absorptance of two glossy samples. The absorptivity calculated by the reflectance measurements against virtually generated backup cavities.
Fig. 11. – Ratio of absorptance of the glossy sample during the second set of measurements. The absorptivity calculated by the reflectance measurement against a virtually generated backup cavity.
As shown in Fig. 8 and fig. 9, no degradation was noticed for the carbon nanotubes nor the masked area of the glossy sample. On the other hand, the glossy sample under UV/EUV radiation showed a similar behavior, including the early increase of the absorptivity, as observed in the first experiment (Fig. 7).
3. Discussion
As shown in Fig. 7 and fig. 8, the MLS-85-SB coating degrades when irradiated with UV/EUV. On the time scale of the experiment (100h), the vacuum conditions alone have no influence as shown in Fig. 7.
As shown in Fig. 8 and fig. 9, no degradation was noticed for the masked area of the glossy sample nor the carbon nanotubes. Control experiments employing vacuum and a masked sample support that the degradation is produced by the UV/EUV radiation. A degradation effect on the glossy coatings was observed, while the carbon nanotubes were not affected under the same conditions, as expected from previous studies [16,17]
The glossy samples featured an increase in the absorptivity during the first hours of the UV/EUV exposure, similar to the analyses of the PMO6 cavities on VIRGO or PREMOS. It takes around 3 hours to reach the maximal absorptivity, while for the PREMOS experiment it takes around 5 days of solar exposure. This mean that our running experiment of 100 hours is equivalent to 165 days of solar exposure.
A cleaning effect due to UV radiation of dust or handling impurities on the surface of the samples is unlikely, as the samples were blown with a nitrogen pistol before the reflectance measurements and the PREMOS cavities were safely stored before the launch.
In order to compare the results of our experiment with the degradation observed in PREMOS, we obtain the absorptivity of our samples by applying the absorptivity equation (a = 1 – rn) to our reflectance measurements. The parameter “n” (effective number of reflections) is 2.35 for the glossy MLS-85 experimentally evaluated by a comparison (R = rn) between the reflectance of such a paint in a flat sample (r) and the reflectance in the cavity (R).
The degradation showed in PREMOS (Fig. 4) is a ratio between an active and a backup cavity. In our experiment a backup cavity was not utilized, then a virtual backup cavity will be generated by extrapolation of the first hours (untill the cavity reaches the maximal absorptivity) of the experiment. This is because in PREMOS it was observed [1] that the backup cavity reaches the maximal absorptivity by the time that the active cavity saturates. The initial point remains the same and the maximal absorptivity will be considered for the virtual backup cavity to happen at 100 hours of the experiment. The values between these two defined points will be calculated by a linear fit, as the early increase in PREMOS or in our measurements can be assumed linear. Once the virtual backup cavity is generated, the ratio of the absorptivity of the active and backup cavity is calculated. In order to obtain the virtual backup cavity, we must solve to equation of a linear fit (y = cx + b). Where we know the initial value at t = 0h (x = 0) which is the value of absorptivity at that point, and at the end of our experiment t = 100h (x = 100) will have a value that corresponds for the moment of maximal absorptivity for that experiment. With this well know points we can obtain for each graph the parameters “c” and “b”. And, once we know the linear fit of the backup cavity, the points between can be calculated.
For the glossy samples the backup cavities are generated by the first hours of the glossy samples themselves. Thus, the curves begin at 1 while the PREMOS ratio as it is obtained by two different cavities with two different initial reflectance value starts at 0.996.
If these two charts (Fig. 10 and Fig. 11) are compared with the one obtained by PREMOS (Fig. 4), a qualitatively similar behavior of the degradation as a function of exposure time is found. It must be considered that the 100h of experiment correspond to 165 days of solar exposure in the PREMOS mission. The three phases of the degradation process (early increase, fast degradation, and saturation) occur similarly. While our experiment shows, for the three different samples (Kapton, no Kapton and duplicate), a decrement of circa 0.1%, PREMOS was affected by a 0.25% decrement on its absorptivity.
The difference of a factor of 2.5 may likely be produced by the different paint (Aeroglaze Z302 vs the new MLS-85-SB). Although both were glossy, the original Aeroglaze Z302 paint applied on the PMO6 likely has a smaller effective number of reflections. MLS-85-SB should be similar, but the exact composition of both paints are trade secrets and may be different. If we run a simulation with a potential effective number of reflections for the PREMOS n = 1.75, the decrement of our glossy samples is 0.3% (Fig. 12). In Fig. 13 it is shown the result obtained for the simulation with n = 1.75 overlapped to the PREMOS chart. It is a reasonable value for Aeroglaze Z302 (higher than diffuse coatings but smaller than the glossy MLS-85). A higher the diffuse component the smaller the effective number of reflections. On the other hand, the glossy paints become more diffuse with time which alters the effective number of reflections. Based on the loss of glossiness, we have to assume that indeed the parameter n decreases with time.
Fig. 12. – Ratio of absorptance of the glossy sample during the second set of measurements with n (effective number of reflections) = 1.75
Fig. 13. – PREMOS degradation curve showed in Fig. 4 overlapped with a degradation curve of a glossy sample with a n = 1.75 courtesy of Dr. Finsterle.
4. Conclusions
Degradation processes of the coating of receivers have been studied, focusing on UV/EUV radiation as the most likely source of damage.
We have been able to reproduce for the first time in a laboratory under controlled conditions the curve’s shape of the degradation effects reported in the literature for PMO6 TSI radiometers.
An experiment trying to simulate the space-based instrument conditions was conducted to quantify the impact of the UV radiation with a continuous UV lamp as light source.
A qualitatively and quantitative similar degradation curve, early increase of the absorptivity included, as the obtained in PREMOS was found in the experiment but with a factor of 2.5 smaller for an n = 2.35. This difference is likely produced for the different black paint applied in PREMOS. Although the experiment was performed with a glossy paint (like the one applied in PREMOS), the characteristics of the coatings are different, and the paint applied in PREMOS (Aeroglaze) had likely a smaller coefficient “n” (effective number of reflections) than our new paint (MLS-85). Thus, at least for this instrument, EUV is the largest contributor to the degradation effect.
The combination of two effects on the paint (i.e., change of glossiness and loss of absorptivity) enhance the degradation when applied in a cavity, because the effective number of reflections in the cavity decreases as the glossiness of the coating decreases, thus the absorptivity decreases. In addition, of course, the losses of absorptivity itself enhance the degradation effect. However, when the paint is applied in a flat sample the losses of absorptivity coefficient is a dominant effect over the change of glossiness, as it was observed in the degradation process in flat samples.
Although we cannot definitively state the process leading to the early absorptance improvement, it is likely attributable to a cleaning process, as it has been shown in both cavities and flat samples. It is not a dust-cleaning process as it was discarded before, but a removal of the superficial layer of the coating itself where structural impurities are found (e.g., solvent leftovers or the matrix that holds the pigments). While these impurities are being removed, the absorptance increases. Then, when all the impurities have been removed, the UV starts to damage the paint and the absorptance decreases.
Carbon nanotubes have been tested, under the same conditions as the glossy coating, presenting no measurable change in their absorptance nor visually (by a simple eye inspection) damage. This characteristic, together with the good optical properties described in detail in [18], make them a very good candidate for a next generation of space radiometers [19]. Currently a radiometer has been developed in collaboration with Davos Instruments AG [20].
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. W. Schmutz, A. Fehlmann, W. Finsterle, G. Kopp, and G. Thuillier, 2013. Conference: AIP Conf. Proc. Volume: 1531, p. 624–627. Total Solar Irradiance Measurements with PREMOS/PICARD
2. C. Fröhlich, J. Romero, H. Roth, et al., “VIRGO: experiment for helioseismology and solar irradiance monitoring,” Sol. Phys. 162(1-2), 101–128 (1995). [CrossRef]
3. D. A. H. Crommelynck, V. Domingo, and C. Fröhlich, “The SOlar VAriations (SOVA) experiment in the EURECA space platform.,” Adv. Space Res. 11(4), 83–87 (1991). [CrossRef]
4. M. Anklin, C. Fröhlich, W. Finsterle, D.A.H. Crommelynck, and S. Dewitte, “Assessment of degradation of VIRGO radiometers on board SOHO,” Metrologia. 35(4), 685–688 (1998). [CrossRef]
5. J. Romero, C. Wehrli, and C. Fröhlich, “Solar Total Irradiance Variability from SOVA 2 on board EURECA,” Sol. Phys. 152(1), 23–29 (1994). [CrossRef]
6. C. Fröhlich, B. N. Andersen, T. Appourchaux, et al., “First results from VIRGO,” Sol. Phys. 170(1), 1–25 (1997). [CrossRef]
7. G. Cessateur, W. Schmutz, Ch. Wehrli, et al., 2016. A&A. Volume 588, Article 126. Solar irradiance observations with PREMOS filter radiometers on the PICARD mission: In-flight performance and data release
8. C. Fröhlich, “Solar Irradiance Variations,” Proc. ISCS 2003 Symposium. ESA SP-535183–193 (2003).
9. M. Suter. 2015. PhD Thesis UZH. Advances in Solar Radiometry
10. C. Fröhlich2010. ISSI Scientific Report Series. Article 32, pp. 525–540. Solar radiometry
11. C. Fröhlich, 2006. Proc. SOHO 17–10 years of SOHO and Beyond. ESA SP-617. pp. 601-604. TSI Variability: what have we learned from SOHO/VIRGO about solar cycle 23
12. G. Thuillier, S. Dewitter, and W. Schmutz, “Simultaneous measurement of TSI and solar diameter by the PICARD,” Adv. Space Res. 38(8), 1792–1806 (2006). [CrossRef]
13. WMO, 1977, No. 490, Annex IV, pp. 97-101, World Radiometric Reference
14. C. Wehrli. Extraterrestrial Solar Spectrum, Publication no. 615, PMOD internal report
15. European Cooperation for Space Standarization. ECSS-E-ST-10-03C
16. D. Harber, Z. Castleman, G. Drake, et al., “Compact Total Irradiance Monitor flight demonstration.,” Proc. SPIE 11131, 12 (2019). [CrossRef]
17. J. Lehman, C. Yung, N. Tomlin, D. Conklin, and M. Stephens, “Carbon Nanotube-Based Black Coatings,” Appl. Phys. Rev. 5(1), 011103 (2018). [CrossRef]
18. A. Remesal Oliva, W. Finsterle, B. Walter, and W. Schmutz., “Characterisation of a new carbon nanotube detector coating for solar absolute radiometers.,” Journal of Physics: Conference Series, Volume 972, 012007 (2018).
19. A. Remesal Oliva and W. Finsterle, “New detector for next generation solar radiometers,” Opt. Mater. Express 12(10), 3882–3893 (2022). [CrossRef]
