Solar selective absorbers are the most critical part of solar water heaters that can be integrated into architecture. A high-performance absorber with a solar absorptance α higher than 95% and an infrared emissivity ɛ below 4% is fabricated by sputtering using TiNxOy based multilayers. The highest absorptance is 97.5% and the corresponding energy utilization efficiency (α/ɛ) value is as high as 26.2. The absorber has excellent thermal stability that can maintain its property after heating at 400 °C for 100 hr in air. It can even be tempered on the glass substrate, which is of great significance for lowering the cost and expanding its applications.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Perfect absorbers with high absorptance have attracted extensive attention during the last decade for harvesting solar energy [1–5], which are known to enhance the efficiency of photothermal conversion for architecture integratable solar thermal technologies, such as solar water heaters and solar thermoelectric generators. For such applications, the solar selective absorbers should have high absorptance in the wavelength range of 0.3-2.5 µm to obtain solar energy as much as possible and low infrared emissivity with wavelength longer than 2.5 µm to avoid loss of energy by radiation [6,7]. Solar energy utilization efficiency (ratio of α to ɛ) is used to evaluate the performance of solar selective absorbers. The larger value, the better performance. In addition to the excellent solar energy efficiency of solar selective absorber, good thermal stability is also important for its extensive applications [8,9], especially in the temperature range of room temperature to 400 °C, because photothermal utilization at low-temperature (below 100 °C) and industrial utilization at mid-temperature (between 100 and 400 °C) are the main applications in the market [10,11]. Recently, various concepts and materials have been explored for the aim of achieving spectral selectivity at higher operating temperatures [12,13]. Most of them are multilayered structures made from transition metal oxides and cermets with appropriate compositions of nitrides, carbides, and borides [14–17]. For example, solar selective absorbing coating of NbTiON/SiON exhibits high thermal stability at 500 °C in vacuum for 40 hr with absorptance of 95% and emissivity of 7% and energy efficiency of 13.6 . Also, the coating of SS–(Fe3O4)/Mo/TiZrN/TiZrON/SiON is reported to have excellent energy efficiency (95%/8%) even after heating treatment at 500 °C in vacuum . Although high solar absorptance has been widely reported, the suppression of infrared radiation is not sufficient, resulting the energy efficiency is still not high enough. Recent experiments in this area indicate that it is very difficult to have good thermal stability and high energy efficiency simultaneously. Moreover, most of their thermal stability experiments were carried out only in vacuum, not in real environment of air. Therefore, it is of great significance to obtain air-stable solar selective absorbers for practical applications.
Furthermore, integrating solar water heaters and solar thermoelectric generator panels into architecture is a prospective trend, since building roofs and facades are potentially usable areas . Currently, the solar selective absorber coatings mainly adopt metal substrate for infrared reflection to obtain low infrared emissivity , which is expensive and cumbersome for extensive applications. These problems can be easily solved by using glass substrate with a thin metal layer for low infrared emissivity. However, the glass used in high-rise buildings must be tempered to gain good mechanical properties and ensure safety for practical applications. Therefore, the solar selective absorber must have good temperability to withstand the tempered process of the glass substrate. But up to now, no temperable solar selective absorber has been reported as far as we know.
In this work, we presented a high performance solar selective absorber on glass with dielectric anti-reflection multilayers to dramatically suppress the reflection loss and obtain high solar absorptance, along with a thin copper layer between the glass and absorber to lower the infrared emissivity and get high solar energy utilization efficiency. The thermal stability and temperability performance are also found to be very good in such a multilayered structure.
2. Experimental section
2.1 Solar selective absorption structures
For solar selective absorption structures, it is ideal to have as high absorptance as possible in solar spectral range of 0.3-2.5 µm and as low infrared emissivity as possible for wavelength longer than 2.5 µm. Therefore, solar absorbance (α) and thermal emissivity (ɛ) are two main parameters and very important for the solar energy photo-thermal conversion performance of the absorbing coatings. In this work, solar absorptance is calculated in the range of 0.3-2.5 µm, covering almost all of the solar radiation energy at AM1.5. Thermal emissivity is calculated in the range of 2.5-25 µm, covering most of the thermal radiation at 373 K (100 °C). They can be calculated from the experimental spectral reflectance when without transmittance according to the following equations:
Here, the solar selective absorption structure of Sub/Cu/TiNxOy/Si3N4/SiO2 has been designed as shown in Fig. 1 to have high solar absorptance and low infrared emissivity, simultaneously. TiNxOy is the core absorption layer and will absorb most of the solar light. Si3N4/SiO2 is used as a good gradient anti-reflection stack to significantly lower the reflection loss of solar light. The Si3N4 film can effectively prevent the invasion of external oxygen and impurities , as well as maintain the stability of the composition ratio of the TiNxOy absorption layer and improve the temperability of solar selective absorber. SiO2 used in the structure is a kind of superhard wear-resistant and stable material at high temperatures, which shows good anti-reflection performance and can greatly improve the weathering resistance and stability of the solar selective absorber.
2.2 Fabrication and measurements
All the TiNxOy thin films and antireflection films were deposited by a multi-target magnetron sputtering system. Since the physicochemical properties of TiNxOy thin films are sensitive to their chemical composition and it is difficult to control the composition precisely from metallic Ti target with two reactive gases mixture of O2 and N2 simultaneously, the way of deposition process by only one reactive gas of O2 is adopted to better control the physicochemical properties of TiNxOy films . Where titanium nitride (TiN) target (99.95% in purity) and Si target (99.99% in purity) were used with the size of 322 mm × 140 mm, as well as the purity of argon (Ar) gas and reactive oxygen gas are both 99.99%. The substrates are 40 mm × 40 mm K9 glass with a thin metal Cu layer. All the depositions were carried out under the sputtering power of 1 kW with a frequency of 30 kHz at room temperature with a base pressure of 3.0 × 10−4 Pa. Here, the TiNxOy thin films were deposited by the TiN target in Ar/O2 mixture gas with flow rate kept constant at 30.0 sccm (mL/min). The thickness of the films was optimized to get better solar energy photo-thermal conversion performance, which was controlled by the sputtering time according to the deposition rate.
All thermal stability experiments were carried out in the muffle furnace in air. The samples were firstly warmed up to the experimental temperature with a heating rate of 20 °C/min, then maintained at the temperature for a period of time for annealing. It was finally cooled down to room temperature at the rate of 20 °C/min. In the temperability experiments, the samples were carried out in the muffle furnace in air at 600 °C and kept for 7 minutes. And then they were taken out for rapid cooling and tempering in air.
The Perkin Elmer Lamda 950 UV/VIS/NIR spectrometer with integrated sphere was used to measure the transmittance and reflectance spectra of the 0.3-2.5 µm band, and the Bruker IFS 125HR Fourier Transform Infrared Spectroscopy(FTIR) was used to measure the transmittance and reflectance spectra of 2.5 - 25 µm band.
3. Results and discussion
In the experiment, the samples were heated in the atmosphere of a muffle furnace with no seal and air connection. We designed three sets of thermal stability experiments in middle temperature range, as shown in Table 1. In the first two experiments, two batches of different samples with absorptance of 95.8% and 96.8%, named S1 and S2, were heated at 250 °C with Cu/TiNxOy/Si3N4/SiO2 structure.
3.1 Microstructural characterization of multilayered coatings
In order to study the microstructure of multilayered coatings, structure of Glass/Cu/TiNxOy/Si3N4/SiO2 was analyzed by SEM, as shown in Fig. 2(a) The constituent layers of the optical stack and the respective thicknesses (SiO2 ∼ 106 nm, Si3N4 ∼ 81 nm and TiNxOy ∼ 59 nm) can be seen in the cross-sectional image clearly.
3.2 Thermal stability experiments
In the first group of experiments, we divided the sample of S1(Glass/Cu/TiNxOy/Si3N4/SiO2 with initial solar absorptance 95.8%) into 8 pieces and put them together into the muffle furnace connection with air. They were kept at 250 °C for 0.5 hr, 1 hr, 2 hr, 4 hr, 8 hr, 16 hr, 32 hr, 64 hr, respectively. Then, the reflectance spectra of S1 heated at 250 °C for different times were measured by FTIR, as shown in Fig. 3. It can be seen that the reflectivity at 0.3-2.5 µm increased slightly with the heating time, indicating that the solar absorptance decreased slightly. Since thickness of the metal Cu layer is 200 nm and thick enough to avoid the transmission, the absorptance A can be calculated as 1 - R.
The relationship between the solar absorptance of S1 and heating time is shown in Fig. 4. It’s obvious that the solar absorptance of S1 decreased rapidly from 95.8% to 94.9% during the first 10 hr, and then remained almost unchanged at about 94.8%. Totally, the solar absorptance decreased slightly from 95.8% to 94.8% during the 64 hr of heating time for only 1%. The relationship between solar absorptance and heating time can be fitted by the following formula: A = 0.85%*exp(-T/2.8) + 94.8%, where A is solar absorptance, and T is the heating time. The results show that the high performance solar selective absorber constructed of Glass/Cu/TiNxOy/Si3N4/SiO2 can keep the absorption property after longtime heating at 250 °C in air.
In order to verify the repeatability of the heating experiment and the stability after longtime heating, another batch of sample S2 with the initial solar absorptance of 96.8% and the initial infrared emissivity of 3.7% was heated directly at 250 °C for 200 hr in the atmosphere of air. The reflectance spectra of S2 before and after heating were measured, as shown in Fig. 5. Similarly, it can be seen that the absorptance of S2 decreased from 96.8% to 95.5% by only 1.3%, which agrees well with the results of sample S1. It is worth noting that the emissivity does not get worse but contrarily decreases from 3.7% to 2.7% and apparently improves by 27%, which results in the solar energy efficiency increasing from 26.2 to 35.4. The highest one reported is only 19.2 . The results reveal that the solar selective absorber constructed of Cu/TiNxOy/Si3N4/SiO2 has excellent thermal stability in air at 250 °C, whose absorptance decreases slightly (only about 1%) and then keeps constant while the emissivity and corresponding solar energy efficiency improves remarkably. Therefore, it is a structure good for both high solar energy efficiency and thermal stability which is suitable for both low and middle temperature solar thermal utilization.
Although the absorptance of Glass/Cu/TiNxOy/Si3N4/SiO2 multilayer can reach 96%, it can be hardly further improved due to the energy loss associated with a reflectance peak in blue as can be seen in both Figs. 3 and 5. The way to further improve the absorptance of solar energy is to reduce and eliminate this reflection loss. Therefore, a thin TiO2 layer is inserted between TiNxOy and Si3N4 to further increase the absorptance, since the bandgap of TiO2 is in the near-ultraviolet band (anatase structure’s bandgap is 3.2 eV and that of rutile structure is 3.0 eV) which has a large absorption coefficient below 500 nm. The structure is changed to Glass/Cu/TiNxOy/TiO2/Si3N4/SiO2 which is named S3 and its schematic diagram is shown in the inset of Fig. 6. The cross-sectional TEM observation of Glass/Cu/TiNxOy/TiO2/Si3N4/SiO2 was shown in Fig. 2(b). Each layer and their interfaces can be clearly distinguished in the cross-sectional image, and the layer thicknesses are SiO2 ∼ 86 nm, Si3N4 ∼ 42 nm, TiO2 ∼ 20 nm and TiNxOy ∼ 83 nm, respectively.
In addition, TiO2 is a material with a high refractive index and can form an impedance matching layer together with Si3N4  to reduce the reflectance of the whole structure further. Then a better impedance matching with air is formed. The reflectance peak splits into two smaller peaks centered at 360 nm and 495 nm when a TiO2 layer is introduced into the structure. Since the left reflectance peak is away from the peak of solar radiation and the right one is much smaller than the former one, the solar absorptance is further improved to 97.5% as shown in Fig. 6, calculated from the formula (1).
The samples of S3 (Glass/Cu/TiNxOy/TiO2/Si3N4/SiO2 with the highest solar absorptance of 97.5%) have also been used to conduct the thermal stability test at a higher temperature of 400 °C for 100 hr in air using a muffle furnace. The reflectance spectra of S3 before and after heating were shown in Fig. 6. The two reflectance peaks in visible increased after heating and resulted that the solar absorptance decreased by 1.1%. The reflectance between 1 µm and 25 µm changed slightly and the infrared emissivity increased only by 0.7% after heating at 400 °C in air for 100 hr. The absorptance and the emissivity are found to be stable at 96.4% and 5.0%, respectively. Thus, the corresponding solar energy efficiency decreased slightly from 22.7 to 19.3. The results show that the multilayered structure of solar selective absorber based on TiNxOy can even bear high temperature of 400 °C in air for a long time.
3.3 Temperability experiment
Since the conventional Cu or Al metal substrates of solar selective absorbers are substituted by glass to lower the cost, it’s necessary to be tempered for the application in high-rise buildings to ensure its safety. It is known that a thin NiCr layer is usually coated on the tempering glass to sperate the glass and film on it . We have carried out the tempering experiments for solar selective absorber structure of TiNxOy/Si3N4/SiO2 on both “Glass/Cu(200nm)” and “Glass/NiCr(10nm)/Cu(200nm)” substrates for comparison. Figure 7 are the photos of these two samples before and after tempered. The pictures of Fig. 7(a-1) show that there are many gray spots existing on the surface of sample without adding a NiCr(10nm) layer. The film is cracked or wrinkled up and damaged after tempering. The intensity of reflection peaks below 700 nm increased obviously and the intensity of reflectance peak in visible increases by 30.5%, as shown in Fig. 7(b), indicating that the structure and property of solar selective absorber without NiCr layer was degraded apparently during the tempering process. In contrast, there was no significant change on the surface of sample with a 10nm thick NiCr layer before and after tempering, as shown in Fig. 7(a-2). Also, the reflectance spectra of the sample have little difference (the intensity of reflectance peak in visible increases only 4.6%) before and after tempering, as shown in Fig. 7(b), indicating that the structure and property of solar selective absorber with a thin NiCr(10nm) layer can bear the tempering process and keep its properties. The function of the thin NiCr seperating layer between the glass substrate and Cu is to prevent the diffusion between Cu layer and glass substrate. In a word, the results of the tempering experiments show that our solar selective absorber can maintain stable above 600 °C in air by just adding a thin NiCr(10nm) isolation layer to the glass substrate, which satisfies the tempering condition.
In this paper, we fabricated high energy efficiency solar selective absorbers constructed with Cu/TiNxOy/Si3N4/SiO2 multilayers on glass with excellent thermostability and temperability. The highest absorptance is 97.5% and corresponding energy efficiency (α/ɛ) value is 26.2, which is the reported highest one as far as we know. The thermal stability is also very good. Its absorptance can be stable at higher than 94.8% after heating at 250 °C in air. After heating at 250 °C for 200 hr, the absorptance decreased only from 96.8% to 95.5% while the emissivity decreased from 3.7% to 2.7%, corresponding to the solar energy efficiency increasing from 26.2 to 35.4. Furthermore, the absorptance. property can also be maintained at 400 °C for such a multilayered structure. When heated at 400 °C in the air for 100 hr, the absorptance decreased from 97.5% to 96.4% and the emissivity changed from 4.3% to 5.0%, respectively. In addition, the multilayered solar selective absorber can even be tempered at 600 °C in air by just adding a thin NiCr isolation layer on the glass substrate, which is of great significance to its architectural integration applications and can be widely used in the future. The solar selective absorber is a good candidate for both low and middle temperature solar utilizations.
National Natural Science Foundation of China (11874376); Shanghai Science and Technology Foundation (18590712600, 18DZ2282200, 19DZ2293400, 19ZR1465900); Development Program of China (973 Program) (2017YFC0111400); Youth Innovation Promotion Association of the Chinese Academy of Sciences..
The authors declare no conflicts of interest.
1. C. G. Granqvist, “Solar Energy Materials,” ChemInform 35(2), 2278 (2004). [CrossRef]
2. T. K. Ghosh and M. A. Prelas, “Solar Energy,” Energy Resources & Systems 81(2), 79–156 (2011). [CrossRef]
3. S.-A. Cho, R. Fookes, and C. A. Garris, “Efficiency of ceramic absorber coatings for solar-thermal conversion,” Ceram. Int. 7(1), 8–12 (1981). [CrossRef]
4. A. Ghobadi, H. Hajian, M. Gokbayrak, B. Butun, and E. Ozbay, “Bismuth-based metamaterials: from narrowband reflective color filter to extremely broadband near perfect absorber,” Nanophotonics 8(5), 823–832 (2019). [CrossRef]
5. S. Wang, F. Chen, R. Ji, M. Hou, F. Yi, W. Zheng, T. Zhang, and W. Lu, “Large-Area Low-Cost Dielectric Perfect Absorber by One-Step Sputtering,” Adv. Opt. Mater. 7(9), 1801596 (2019). [CrossRef]
6. T. K. Vien, C. Sella, J. Lafait, and S. Berthier, “Pt-Al2O3 selective cermet coatings on superalloy substrates for photothermal conversion up to 600°C,” Vacuum 126(1-2), 17–22 (1985). [CrossRef]
7. H. C. Barshilia, N. Selvakumar, K. S. Rajam, D. V. S. Rao, K. Muraleedharan, and A. Biswas, “TiAlN/TiAlON/Si3N4 tandem absorber for high temperature solar selective applications,” Appl. Phys. Lett. 89(19), 191909 (2006). [CrossRef]
8. S. Yue, S. Yueyan, and W. Fengchun, “High-temperature optical properties and stability of AlxOy–AlNx–Al solar selective absorbing surface prepared by DC magnetron reactive sputtering,” Sol. Energy Mater. Sol. Cells 77(4), 393–403 (2003). [CrossRef]
9. L. Wu, J. Gao, Z. Liu, L. Liang, F. Xia, and H. Cao, “Thermal aging characteristics of CrNxOy solar selective absorber coating for flat plate solar thermal collector applications,” Sol. Energy Mater. Sol. Cells 114, 186–191 (2013). [CrossRef]
10. G. Gong, X. Huang, J. Wang, and M. Hao, “An optimized model and test of the China's first high temperature parabolic trough solar receiver,” Sol. Energy 84(12), 2230–2245 (2010). [CrossRef]
11. D. Barlev, R. Vidu, and P. Stroeve, “Innovation in concentrated solar power,” Sol. Energy Mater. Sol. Cells 95(10), 2703–2725 (2011). [CrossRef]
12. A. Dan, K. Chattopadhyay, H. C. Barshilia, and B. Basu, “Angular solar absorptance and thermal stability of W/WAlN/WAlON/Al2O3-based solar selective absorber coating,” Appl. Therm. Eng. 109, 997–1002 (2016). [CrossRef]
13. L. Zheng, F. Gao, S. Zhao, F. Zhou, J. P. Nshimiyimana, and X. Diao, “Optical design and co-sputtering preparation of high performance Mo–SiO2 cermet solar selective absorbing coating,” Appl. Surf. Sci. 280, 240–246 (2013). [CrossRef]
14. P. Patsalas, N. Kalfagiannis, S. Kassavetis, G. Abadias, D. V. Bellas, C. Lekka, and E. Lidorikis, “Conductive nitrides: Growth principles, optical and electronic properties, and their perspectives in photonics and plasmonics,” Mater. Sci. Eng., R 123, 1–55 (2018). [CrossRef]
15. F. Cao, K. McEnaney, G. Chen, and Z. Ren, “A review of cermet-based spectrally selective solar absorbers,” Energy Environ. Sci. 7(5), 1615 (2014). [CrossRef]
16. M. Joly, O. Bouvard, T. Gascou, Y. Antonetti, M. Python, M. A. González Lazo, P. Loesch, A. Hessler-Wyser, and A. Schüler, “Optical and structural analysis of sol–gel derived Cu–Co–Mn–Si oxides for black selective solar nanocomposite multilayered coatings,” Sol. Energy Mater. Sol. Cells 143, 573–580 (2015). [CrossRef]
17. I. E. Khodasevych, L. P. Wang, A. Mitchell, and G. Rosengarten, “Micro- and Nanostructured Surfaces for Selective Solar Absorption,” Adv. Opt. Mater. 3(7), 852–881 (2015). [CrossRef]
18. Y. Liu, C. Wang, and Y. Xue, “The spectral properties and thermal stability of NbTiON solar selective absorbing coating,” Sol. Energy Mater. Sol. Cells 96, 131–136 (2012). [CrossRef]
19. Y. Liu, Z. Wang, D. Lei, and C. Wang, “A new solar spectral selective absorbing coating of SS–(Fe3O4)/Mo/TiZrN/TiZrON/SiON for high temperature application,” Sol. Energy Mater. Sol. Cells 127, 143–146 (2014). [CrossRef]
20. F. Chen, S.-W. Wang, X. Liu, R. Ji, L. Yu, X. Chen, and W. Lu, “High performance colored selective absorbers for architecturally integrated solar applications,” J. Mater. Chem. A 3(14), 7353–7360 (2015). [CrossRef]
21. M. A. Estrella-Gutiérrez, F. I. Lizama-Tzec, O. Arés-Muzio, and G. Oskam, “Influence of a metallic nickel interlayer on the performance of solar absorber coatings based on black nickel electrodeposited onto copper,” Electrochim. Acta 213, 460–468 (2016). [CrossRef]
22. H. C. Barshilia, N. Selvakumar, K. S. Rajam, D. V. Sridhara Rao, and K. Muraleedharan, “Deposition and characterization of TiAlN/TiAlON/Si3N4 tandem absorbers prepared using reactive direct current magnetron sputtering,” Thin Solid Films 516(18), 6071–6078 (2008). [CrossRef]
23. F. Chen, S.-W. Wang, X. Liu, R. Ji, Z. Li, X. Chen, Y. Chen, and W. Lu, “Control of optical properties of TiNxOy films and application for high performance solar selective absorbing coatings,” Opt. Mater. Express 4(9), 1833–1847 (2014). [CrossRef]
24. L. Rebouta, P. Capela, M. Andritschky, A. Matilainen, P. Santilli, K. Pischow, and E. Alves, “Characterization of TiAlSiN/TiAlSiON/SiO2 optical stack designed by modelling calculations for solar selective applications,” Sol. Energy Mater. Sol. Cells 105, 202–207 (2012). [CrossRef]
25. F. Chen, S.-W. Wang, X. Liu, R. Ji, Z. Li, X. Chen, Y. Chen, and W. Lu, “Colorful solar selective absorber integrated with different colored units,” Opt. Express 24(2), A92–A103 (2016). [CrossRef]
26. H. J. Gläser, “History of the development and industrial production of low thermal emissivity coatings for high heat insulating glass units,” Appl. Opt. 47(13), C193–C199 (2008). [CrossRef]