Scalable all-ceramic nanofilms as highly efficient and thermally stable selective solar absorbers

Scalable all-ceramic nanofilms as highly efficient and thermally stable selective solar absorbers

Journal Pre-proof Scalable all-ceramic nanofilms as highly efficient and thermally stable selective solar absorbers Yang Li, Chongjia Lin, Dan Zhou, Y...

1MB Sizes 0 Downloads 1 Views

Journal Pre-proof Scalable all-ceramic nanofilms as highly efficient and thermally stable selective solar absorbers Yang Li, Chongjia Lin, Dan Zhou, Yiming An, Dezhao Li, Cheng Chi, He Huang, Shihe Yang, Chi Yan Tso, Christopher Y.H. Chao, Baoling Huang PII:

S2211-2855(19)30654-8

DOI:

https://doi.org/10.1016/j.nanoen.2019.103947

Reference:

NANOEN 103947

To appear in:

Nano Energy

Received Date: 15 May 2019 Revised Date:

19 July 2019

Accepted Date: 28 July 2019

Please cite this article as: Y. Li, C. Lin, D. Zhou, Y. An, D. Li, C. Chi, H. Huang, S. Yang, C.Y. Tso, C.Y.H. Chao, B. Huang, Scalable all-ceramic nanofilms as highly efficient and thermally stable selective solar absorbers, Nano Energy (2019), doi: https://doi.org/10.1016/j.nanoen.2019.103947. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Ltd.

We first introduce all-ceramic structures into selective solar absorbers and demonstrate state-of-the-art overall performance (spectral selectivity and thermal stability).

Scalable all-ceramic nanofilms as highly efficient and thermally stable selective solar absorbers Yang Lia, Chongjia Lina, Dan Zhoub, Yiming Anb, Dezhao Lic, Cheng Chia, He Huanga, Shihe Yangb, Chi Yan Tsod, Christopher Y. H. Chaoe, Baoling Huanga,* a

Department of Mechanical and Aerospace Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong

b

Department of Chemistry, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong c College of Science, Zhejiang University of Technology, Hangzhou 310023, China d School of Energy and Environment, City University of Hong Kong, Kowloon, Hong Kong e

Department of Mechanical Engineering, The University of Hong Kong, Pokfulam, Hong Kong *Corresponding author. Email address: [email protected]

Keywords: Concentrating solar power; Solar-thermal energy conversion; All-ceramic nanofilms; Thermal stability; Spectrally selective absorber

Abstract The pressing demands for next-generation concentrating solar power drive the pursuit of high-efficiency, thermally stable, and scalable spectrally selective absorbers. Multilayer metal/ceramic nanofilms are promising candidates owing to their strong sunlight absorption provided by extremely simple configurations and facile fabrication. However, the commercial success of such absorbers is still hindered by their unsatisfactory spectral selectivity and high-temperature stability associated with metal/ceramic interfaces. Here we first propose an all-ceramic TiN/TiNO/ZrO2/SiO2 absorber with highly selective absorption, i.e., a high solar absorptance (92.2%) yet an ultralow thermal emittance (17.0% at 1000 K), producing an unprecedented solar-thermal conversion efficiency (82.6% under 100 suns). Remarkably, the absorber shows great thermal stability even after long-term (150 hrs) annealing at 1000 K, boosting the operating temperature of conventional multilayer absorbers by at least 227 K. 1

The efficient and stable all-ceramic absorber can be readily produced in quantity via low-cost processes, rendering it attractive for high-temperature solar-thermal technologies.

1. Introduction Sunlight is a wide-spread and sustainable source of energy, which can be converted into various desired energy forms using different technologies, such as photovoltaic (PV) modules and solar-thermal systems [1,2]. Concentrating solar power (CSP) plants, known as high-temperature (673-823 K) solar-thermal systems, have been widely installed for supplying power on-demand. The use of heat transfer fluids in CSP plants provides a feasible and inexpensive means of energy storage for power demand at peak time and night. Different from PV modules, CSP systems fall into the category of dispatchable power generation technologies that are much more preferred [2,3]. By 2017, the total CSP capacity even reached ~ 6 GW worldwide with a levelized cost of electricity (LCOE) of ~0.22 USD/kWh, being still higher than that of fossil fuels (~0.07 USD/kWh) [3]. To further accelerate the switching from fossil fuels to solar energy, next-generation CSP plants operating at higher temperatures (>873 K) [3-6], as widely acknowledged, are desperately demanded for enhancing the electricity generation efficiency (from ~30% to ~40%) and reaching the ambitious cost target set for 2030 (~0.05 USD/kWh) [7]. But at the same time, such high temperatures bring unprecedented challenges, such as thermal radiative heat losses and structural stability [4]. A spectrally selective and thermally stable solar absorber, as a key component in a CSP system, plays a critical role in addressing the challenges, especially for the most installed parabolic trough systems with relatively low solar concentration ratios [4,8]. Various spectrally selective absorbers (SSAs) have been developed over recent decades, including intrinsic absorbers [9], semiconductor-metal tandems [10], ceramic-metal composites (cermets) [11-13], multilayer metal/ceramic nanofilms [14,15], plasmonic metamaterials 2

[16,17], and photonic crystals (PhCs) [18,19]. Among them, cermets stand out with great spectral selectivity and lithography-free fabrication, and have achieved wide commercial success. However, the thermal stability of the commercial cermets (typically <873 K) still falls behind the requirement of next-generation CSPs [13]. In contrast, metallic PhCs with ceramic coatings exhibit better thermal stability, but their carefully designed nanostructures have to trade spectral selectivity for cost-effective fabrication

[20]. Therefore,

high-performance SSAs of alternating metal/ceramic multilayer nanofilms have attracted great attention in recent years. Without the use of complex cermet composites or nanophotonic structures, multilayer SSAs produced by facile thin-film deposition processes offer low fabrication cost and highly predictable performance [4,8]. Unfortunately, current multilayer SSAs face an unfavorable trade-off between spectral selectivity and thermal stability, precluding their commercial progress. Infrared (IR) reflectors made of highly reflective metals, such as Au, Ag, Cu, Al, and stainless steel tend to be unstable at high temperatures (>673 K) [21-23]. For higher temperatures, refractory metals like W, Mo, and Zr have been used, but inevitably the strong IR reflection and therefore the spectral selectivity are reduced [14,15,23-27]. Actually, refractory metal-based SSAs still suffer from various high-temperature issues associated with metal/ceramic interfaces such as pronounced diffusion, oxidation [15,24,26], and surface detachment [14]. It is well known that ceramics (e.g., Si3N4, Al2O3, SiO2, ZrO2, HfO2, etc.) exhibit more stable behaviors than metals at high temperatures. For this reason, they have been widely employed in SSAs as diffusion barriers [14,27], surface coatings [11,17,18,25,28], or even a few ceramics (e.g., TiAlN, TiNO, and ZrNO) as absorptive layers [21,23,25]. Nevertheless, the usage of highly reflective ceramics as IR reflectors in multilayer SSAs or even in all classes of SSAs has rarely been explored so far [16]. In light of the excellent stability of ceramics, we were motivated to develop a scalable SSA of all-ceramic multilayer nanofilms to achieve improved high-temperature stability and spectral selectivity simultaneously. In 3

particular, we resorted to transition metal nitrides like TiN and ZrN, which have recently aroused renewed interests by virtue of their tunable properties and exceptional stability up to 1073 K [16,29]. Their optical properties can be altered by adjusting their compositions and/or preparation processes. For instance, they can provide highly reflective properties in the infrared region similar to Au when containing excess nitrogen [30]. After the adoption of sufficient oxygen or aluminum, the resulting oxynitrides or nitrides (e.g., TiNO, ZrNO, and TiAlN) exhibit enhanced absorption at visible-NIR wavelengths, making them eligible for absorptive layers [21,23,25,31]. As shown in Fig. 1a, the all-ceramic SSA consists of a TiN IR reflector, a TiNO absorptive layer, and ZrO2 and SiO2 anti-reflection coatings. Note that here TiN also serves as a diffusion barrier, which has been widely used in the area of semiconductor technology, especially for high-temperature service [32]. Titanium oxy-nitride (TiNO) was selected as the absorptive layer because of its highly tunable and strong absorption, as well as excellent compatibility with TiN. We experimentally demonstrated that the all-ceramic structure endowed the resulting SSA with state-of-the-art and even greater overall performance, i.e., a solar-thermal conversion efficiency of 82.6% at 1000 K under 100 suns and thermal stability up to 1000 K. The superior performance, scalable simple structure, and improved thermal stability of the all-ceramic multilayer SSA render it an ideal candidate for the next-generation CSP systems.

2. Results and discussion 2.1. Design and fabrication of all-ceramic SSAs Both the TiN reflector and the TiNO absorptive layer, which are key components in the proposed absorber, are non-stoichiometric ceramic materials. Before embarking on the absorber design, it is necessary to obtain the highly reflective TiN and absorptive TiNO. The details of the fabrication processes and optimization are provided in the Materials and methods part and Supporting Information. Fig. 1b depicts the XRD patterns of the fabricated 4

TiN and TiNO nanofilms with optimized properties. Both the layers have a typical NaCl crystal structure with preferential growth along the (111) orientation, which is the most stable orientation for TiN [30,33]. For the TiNO film, both the slight shift (~0.60°) of XRD peaks to higher angles and the appearance of (200) orientation indicate it is composed of titanium oxynitride [33]. According to the Braggs law, the lattice parameters of the TiN and TiNO layers are calculated to be 0.428 and 0.421 nm, respectively. The same crystal structure and the similar lattice parameters (mismatch <2%) allow the epitaxial growth of TiNO nanofilm on the TiN crystals, forming a very peculiar “lattice-matched heterostructure”. Such heterostructures have been demonstrated to display low roughness, uniform texture, negligible stress, and enhanced thermal stability [30,34,35]. As shown in Fig. 1c, the TiN nanofilm has obvious Ti 2p and 2s, and N 1s peaks in the XPS spectrum. As a comparison, the TiNO nanofilm displays a stronger O 1s peak, yet a weaker Ti 2p and N 1s peaks, directly suggesting its higher oxygen concentration. High-resolution XPS spectra of Ti 2p were analyzed to further determine the chemical states of Ti (Fig. 1d), which can be deconvoluted into three major components including Ti-N (455.0 eV), Ti-N-O (456.5 eV), and Ti-O (458.4 eV) chemical bonds [33,36]. The TiN reflector is mainly composed of the Ti-N bond (53.1%), while the TiNO absorptive layer is mainly composed of the Ti-N-O (41.4%) and Ti-N (36.5%) bonds. Therefore, in this study, we named the TiN and TiNO nanofilms referring to their major components. Besides, to prepare for the structure optimization, the complex refractive index of each layer was measured using a spectroscopic ellipsometer (Fig. 1e; See details in Supporting Information). The performance of SSAs can be evaluated by the solar-thermal conversion efficiency ηsolar-th, defined as [37] ηsolar −th = α − ε

σ TA4 C × I solar

, where σ is the Stefan-Boltzmann constant,

TA is the SSA temperature, Isolar is the total solar radiation (AM 1.5G), and C is the concentration ratio, respectively. The spectrally averaged solar absorptance α and thermal 5

emittance ε are calculated by

∫ α=

4µm

0.3 µ m

d λ α (λ ) Esolar (λ ) I solar

and ε

∫ =

+∞

0

d λ ε (λ ) EB (λ , TA )

σ TA4

,

respectively. Here, Esolar(λ), EB(λ,TA), α(λ), and ε(λ) represent the wavelength-dependent solar radiation, the blackbody radiation at TA, the absorptance, and the emittance at a wavelength λ, respectively. Apparently, ηsolar-th depends on both TA and C. To optimize an SSA, its cut-off wavelength under certain operating conditions (i.e., TA, and C) needs to be determined, where the blackbody radiation starts to exceed the solar radiation. One-axis line-focus concentrator systems, such as the most installed CSP systems, i.e., parabolic trough collectors (PTCs), are much simpler and less expensive compared to those two-axis point-focus systems. PTCs coupled with molten salts as heat transfer fluids are promising for next-generation CSP systems because of their high operating temperatures (>823 K) and low-cost installation [4,17]. In this study, the proposed SSA was designed for a next-generation PTC system operating at 1000 K with a concentration ratio of 100, corresponding to an optimal cut-off wavelength of ~1.8 µm. By using the measured optical properties of each layer, the optimization in the optical design of TiN/TiNO/ZrO2/SiO2 four-layer structure was conducted by the finite-difference time domain (FDTD) method to achieve high overall performance. Firstly, the thickness of the TiN reflector is set to be > 100 nm, which is enough to reflect all the IR light to the top layers. As shown in Fig. 2a, the TiN reflector itself possesses highly selective absorption, which owns ~80% absorption around 0.4 µm with a golden appearance and near-perfect reflection beyond visible wavelengths. Such a highly IR reflective performance is attributed to both intrinsic material properties and great process control during fabrication, benefiting the spectral selectivity of the resulting SSA built on it. Secondly, after adding a TiNO absorptive layer with an optimized composition and thickness (~40 nm), the absorption band is broadened considerably as expected. It is worth emphasizing that the thickness of the TiNO is of utmost importance in controlling the spectral selectivity of the absorber. Here we used a 6

relatively thin TiNO to maintain the low emission in the IR region of the whole structure. Finally, both a 35-nm-thick ZrO2 and a 100-nm-thick SiO2 were selected as antireflection (AR) coatings, since their refractive indices n (~2.0-2.2 for ZrO2 and 1.5 for SiO2) are lower than that of TiNO in the visible-IR range (Fig. 1e), forming a gradient-index TiNO/ZrO2/SiO2 structure to reduce surface reflection. In addition, the coefficient of thermal expansion (CTE) of ZrO2 (10.5×10-6/K) [38] matches to those of TiN and TiNO (9.4×10-6/K) [39], which leads to only small interfacial stresses and benefits the thermal stability of the SSA. The absorption performance of the SSA is improved significantly after adding the AR coatings, achieving a high solar absorptance (a simulated value of ~90%, Fig. 2a). The TiNO/ZrO2/SiO2 gradient-index structure enables a considerable reduction in surface reflection and thereby enhanced solar absorption. Due to the existence of the TiN reflector, the transmitted sunlight is reflected back to the gradient-index structure again so that it is trapped and re-absorbed. As a consequence, the TiNO absorptive layer contributes 80% (relative value) to the total sunlight absorption (see details in Fig. S2). The designed SSA displays low IR absorption (i.e., emission) and satisfactory wavelength selectivity because of the careful material selection and rational optical design. According to the simulated spectra of the SSA at the four stages, its RGB colors are able to be derived (Fig. 2b), which are quite useful for self-checking during the fabrication stage by stage. After a global optimization, the optimized four-layer structure was deposited on a silicon wafer in sequence using inexpensive sputtering and chemical vapor deposition methods. The economic analysis of the four-layer absorber is provided in Supporting Information in details. The photographs in Fig. 2c record the fabrication processes at the four stages, which agree well with the predicted colors shown in Fig. 2b. An all-ceramic TiN/TiNO/ZrO2/SiO2 SSA with a purple-black color was attained after the four-stage fabrication. The absorption spectra of the SSA at different stages were measured by a UV-visible-NIR spectrometer equipped with a 150 mm integrating sphere. As shown in Fig. 2d, near-perfect absorption (>90%) in a 7

wide band (~0.3-1.6 µm) was achieved in the fabricated SSA. The measured data manifest an excellent agreement with the simulation results at each stage due to the great quality control during fabrication. This agreement also suggests that the performance can be predicted accurately via quite a simple modeling. Moreover, the near-perfect broadband absorption shows independence on the incident angle from 0° to 45°. When the incident angle further increases, the absorption (i.e., emission) in the NIR range gradually declines, leading to a decrease in absorption bandwidth (Fig. S4). The cross-sectional SEM image of the fabricated four-layer SSA is shown in Fig. 2e. The TiNO layer is difficult to be distinguished from the bottom TiN reflector because of their similar columnar grains. In Fig. 2f, the TEM image of the SSA clearly illustrates the TiN/TiNO interface with a ~2.5-nm-thick epitaxial transitional interlayer, as shown in the inset HRTEM image. Enhanced thermal stability can be expected from the resulting lattice-matched TiN/TiNO heterostructure, due to the robust adhesion and negligible stress assisted by the transitional interlayer.

2.2. Optical performance of all-ceramic SSAs Fig. 3a shows the absorptance spectrum of the all-ceramic SSA over the entire UV-visible-IR range measured at room temperature. The SSA has strong absorption in the 0.3-1.8 µm wavelength range, corresponding to the highest intensity within the solar spectrum, but the absorptance rapidly declines beyond the cut-off wavelength (~1.8 µm). Its spectrally averaged solar absorptance α and thermal emittance ε at 1000 K are calculated to be 92.2% and 17.0%, respectively. As a result, the solar-thermal conversion efficiency η reaches 82.6% at 1000 K under the irradiance of 100 suns, which is among the best values for state-of-the-art SSAs, as will be compared later. Besides, this efficiency is calculated at a concentration ratio C of only 100, and higher η values can be expected when utilizing point-focus concentrators with large C values (e.g., η = 91.2% @ C = 1000), as shown in Fig. 8

3b. It should be emphasized here that such a remarkable efficiency is achieved in a simple four-layer structure, which can be easily scaled up at low manufacturing cost in the industry. As we discussed before, such a superior overall performance of our absorbers is attributed to the careful materials selection, rational optical structure design, and outstanding process control during fabrication. In addition to the all-ceramic SSA with a TiN reflector, SSAs with metal reflectors, i.e., Au, Ta, and WTi alloy, were also fabricated as references for the comparison of spectral selectivity and thermal stability. Au is a well-known IR reflective material with extremely high selectivity. Among refractory metals, tantalum (Ta) [17,40] and tungsten (W) [11,14,20] have been proven to be effective high-temperature IR reflectors in previous works. W-Ti (90%+10%) alloy instead of pure W was used in this study because of its better adhesion to different surfaces. As shown in Fig. 3c and Table 1, both the TiN monolayer and TiN-based four-layer SSA provide more sunlight absorption and stronger IR reflection than their Ta and WTi counterparts. Moreover, the TiN-based SSA even offers a great spectral selectivity comparable to that of the Au-based SSA. Consequently, the resulting η of the TiN-based all-ceramic SSA surpasses that of the Au-based (81.1%), WTi-based (73.7%), and Ta-based (72.9%) SSAs under the same conditions (TA = 1000 K and C = 100). The all-ceramic SSA should be a more efficient candidate for CSP systems. 2.3. Thermal stability of metal-based SSAs In order to verify the effectiveness of the proposed all-ceramic structure in improving the thermal stability, multilayer SSAs with different reflectors were annealed at high temperatures in a flowing argon (99.99%) atmosphere with the equivalent air partial pressure to a vacuum condition of 10-1 mbar (7.5×10-2 Torr), which is similar to or higher than that in annulus vacuum tubes in the most widely used PTC receivers (10-2-10-3 mbar). Table 1 summarizes their optical performances before and after annealing. Fig. 4a compares the absorptance 9

spectra of the Ta-based SSA before and after annealing for 1 hr. There is a 6.4% absorptance drop after annealing at 773 K, which becomes larger as the temperature increases (13.3% for 900 K, and 17.1% for 1000 K). To clarify this drop, micro-Raman spectroscopy was used to analyze the structure variations in TiNO after annealing. In Fig. 4b, the Raman spectra show three main bands centered at ~210, ~310, and ~580 cm-1, corresponding to the transverse acoustic (TA), the longitudinal acoustic (LA), and the transverse optical (TO) phonon modes of TiNO, respectively. The acoustic modes are mainly contributed by the vibrations of the heavy Ti ions, while the optical mode results from the vibrations of the lighter non-metal (i.e., N and O) ions [41]. The relative intensity ratio of Ti vibration to non-metal vibration experiences a rapid growth after annealing, suggesting a generation of non-metal vacancies in TiNO lattices, which is likely caused by N and O diffusions [42]. Actually, it was extensively documented that the major failure mechanism for most SSAs at high temperature is diffusion [27,43]. Similar absorptance drops after high-temperature annealing were also found in WTi-based SSAs (Fig. S5 and Table 1). We further studied the depth profiles of elements for the Ta-based SSA using a secondary ion mass spectrometer (SIMS). Fig. 4c illustrates obvious N, O, and Ta inter-diffusion near the Ta/TiNO interface after thermal testing at 900 K for 1 hr. When the temperature further rises to 1000 K, the reduction in sunlight absorption becomes more considerable for the Ta-based SSA and the WTi based SSA (Fig. 4a and Table 1). In consequence, their η are drastically reduced by 16.6% and 10.1%, respectively. The SEM image in Fig. 4d shows that after annealing at 1000 K for 1 hr, the Ta-based SSA suffers from large-scale surface detachment, which was also observed in W-based multilayer SSAs reported previously [14]. Obviously, this detachment is another obstacle to multilayer SSAs operating at much higher temperatures. Some literature speculated that this detachment is attributed to the mismatch in the coefficient of thermal expansion (CTE) [14,44]. To thoroughly address this problem, more details must be figured out; for example, which layers detach from the substrate. 10

In Fig. 4e, the surface EDX mapping results show that compared to the left-side surviving regime, the right-side failure regime displays a weaker O and N intensity, yet a stronger Ta intensity. This phenomenon suggests that the three top nano-films altogether peel off so that the Ta is directly exposed. In addition, the presence of N and O signals in the exposed Ta reflector consolidates their inter-diffusion before surface delamination. The cross-sectional SEM image (Fig. 4f) further verifies the delamination between the TiNO/ZrO2/SiO2 ceramic nano-films and the Ta reflector. It is worth noting that the three ceramic nano-films act as a whole during delamination. Coincidentally, this behavior was also observed in the 900 K-annealed Au-based SSA (Fig. S6), indicating that ceramic/ceramic interfaces are more robust than metal/ceramic ones at elevated temperatures. We therefore conclude that the underlying mechanisms for the high-temperature failure of metal-based multilayer SSAs are inter-diffusion, and following surface detachment at higher temperatures associated with metal/ceramic interfaces. 2.4. Thermal stability of all-ceramic SSAs Long-term annealing with heating and cooling cycles was performed to test the thermal stability of the all-ceramic SSA in practical applications. The long-term stability tests were conducted at 1000 K in argon, where the holding time was 5 hrs for each cycle. As shown in Fig. 5a and Table 1, after 50 hrs annealing at 1000 K, the absorber still sustains its high solar absorption and low IR emission with only a 0.5% drop in η. In comparison, η of the Ta-based and WTi-based SSAs degrade by 7.1% and 4.7%, respectively, after annealing at 773 K for 1 hr (Table 1). In other words, the all-ceramic structure boosts the maximum operating temperature of SSAs made of metal/ceramic multilayers by at least 227 K. Based on the above results, it is clear that the all-ceramic absorber is able to offer stable and efficient solar-thermal energy conversion at a high temperature of 1000 K where its η is calculated. Clearly, the efficiency drop of our absorber after 50 hrs annealing is due to a slight blue-shift 11

in the absorption band. Since the thin flims in the multilayer absorber were deposited at lower temperatures, annealing at such a high temperature (1000 K) would introduce intrusive stresses, resulting in minor changes in the morphology. No further blue-shift in the absorption band is found after a longer annealing time of 150 hrs, indicating that such degradation would not become more significant with the increase of operating time. As shown in Fig. 5b, the depth profiles of Ti, N, and O for the SSA annealed at 1000 K are coincided with those before annealing, confirming that the stable performance of the all-ceramic SSA results from the effective suppression of interlayer diffusion between the oxides and TiN. Although the thermal stability test time is much shorter than the expected long life time of CSP systems (>20 years), the performance is potential to be maintained owing to the suppressed inter-diffusion. Furthermore, no remarkable structural failure behaviors such as delamination and nanofilm degradation are observed within the four-layer all-ceramic SSA annealed at 1000 K, as evidenced by the cross-sectional SEM image in Fig. 5c. The absence of delamination between the top three layers and the TiN reflector is mainly attributed to the judicious selection of constituent ceramic materials. As mentioned above, the lattice-matched TiN/TiNO heterostructure leads to excellent adhesion and improved thermal stability. In addition, the CTE of TiN and TiNO (~9.0×10-6/K) are close to that of ZrO2 (10.5×10-6/K), resulting in small thermal stresses. Even though the long-term thermal annealing was mainly conducted in Ar (99.99%), our absorber is able to offer comparable stability in real annulus vacuum tubes of PTC receivers, since the air contamination (10-1 mbar) in the Ar is more than that in the vacuum tubes (10-2-10-3 mbar). We also tested the thermal stability of the all-ceramic absorber in vacuum (~7.5×10-2 torr) at 1000 K for 50 hrs and obtained almost the same absorption spectrum as that in Ar (see Fig. S8). In Fig. 5d, both the thermal stability and the solar-thermal conversion efficiency η of the proposed all-ceramic SSA are compared to those of recently reported high-performance SSAs. We classified these SSAs according to their structures [4,8], including photonic crystals (PhCs) 12

[18,20,40,45], plasmonics [16,17], cermets [11,12,46], and multilayers [14,15,27,47,48]. As the η heavily depends on the operation conditions (solar concentration C, operating temperature TA, etc.), we calculated the η of the reported SSAs under the same conditions (C = 100 and TA = 1000 K) from their absorptance spectra before thermal stability testing for comparison. The efficiency value (82.6%) of the all-ceramic SSA in this work equals or surpasses those of state-of-the-art SSAs. Its maximum operating temperature (1000 K) is higher than cermet SSAs [11,12,46], and most of the multilayer SSAs [14,15,47]. A refractory SSA made of ten-layered W/HfO2 was thermally stable at an elevated temperature of 1273 K due to the utilization of the extremely stable ceramics HfO2, but its efficiency was not very high [27]. A recently reported black chrome/ITO/SiO2 absorber that was stable at 1173 K provided an η of 50.5% because of the drop in spectral selectivity after pre-annealing [48]. The thermal stability of the all-ceramic SSA is even comparable to some of the reported PhCs [18,40], which are well-known high-temperature SSAs due to the combining effects of thicker refractory metals and stable protective ceramics (e.g., HfO2, and Al2O3). It is worth noting that such great thermal stability and high conversion efficiency of the all-ceramics SSA are enabled by a facile and scalable structure, without the need for metal-ceramic composites or nanoscale pattering.

3. Conclusion In this work, we developed a selective solar absorber (SSA) through the use of all-ceramic nanofilms, i.e., TiN/TiNO/ZrO2/SiO2 from bottom to top. Despite the simple structure, the fabricated all-ceramic four-layer structure is able to perfectly absorb most of the solar radiation (92.2%), yet strongly reflect the IR light with an ultralow thermal emission (17.0% @ 1000 K). As a result, the solar-thermal conversion efficiency reaches as high as 82.6% under the irradiation of 100 suns. This excellent spectral control, and therefore the dramatic efficiency is attributed to the inherently spectral selectivity of TiN reflector as well as the 13

rational design of nanofilm structure. More importantly, the all-ceramic structure avoids the thermal issues of current multilayered alternating metal/ceramic nanofilms, boosting their operating temperature from 773 to 1000 K. Further analysis results clarify that the great stability is achieved by effectively suppressing the diffusion and delamination at the interfaces. The all-ceramic SSA with high efficiency, high operating temperatures, yet the facile design and therefore great potential for low-cost fabrication would potentially push the deployment of next-generation CSP plants forward.

4. Materials and methods 4.1. Absorber fabrication At first, a TiN reflector (>100 nm) was deposited on a cleaned silicon wafer using the reactive mid-frequency high-temperature magnetron sputtering (EG 1000 system, Techwinner Co., Ltd., Shenzhen, China). The sputtering chamber was evacuated down to a high vacuum (6.0×10-6 Torr) before the deposition. The TiN reflector was deposited at 10 kW DC power, and 150 sccm Ar and 100 sccm N2 flow rates using high-purity (99.9%) Ti targets. Afterward, a 40-nm-thick TiNO absorptive layer was deposited using reactive DC sputtering (Plasma Science, ARC-12M) with a high-purity (99.9%) Ti target. A gas mixture of Ar (20 sccm) and N2 (5 sccm) was employed as the deposition atmosphere and the DC power was 130 W. The base vacuum for TiNO deposition (~1.0×10-5 Torr) was lower than that for TiN to introduce a little oxygen. After that, a high-quality ZrO2 anti-reflection layer of around 35 nm was deposited onto the TiNO layer by the atomic layer deposition (ALD) technique. The Zr and the O precursor, i.e., tetrakis[ethylmethylamino]zirconium (TEMAZ) and water vapor, were introduced into the vacuum (1.0×10-2 Torr) chamber and reacted at 200 °C. Finally, a SiO2 anti-reflection layer of around 100 nm was deposited at 300 °C with the plasma-enhanced chemical vapor deposition (PECVD, STS). The ZrO2 and SiO2 can also be deposited by magnetron sputtering, which are mature techniques already in the industry. Unfortunately, 14

limited by the equipment conditions, we were not able to fabricate high-quality ZrO2 and SiO2 using our sputtering machine. Besides, although in this work, we fabricated our all-ceramic absorber on Si wafers, the absorber that consists of four layer thin films is compatible with diverse substrates used in CSP technology, such as stainless steel, copper, and aluminum. 4.2. Absorber characterizations The scanning electron microscopy with energy dispersive X-ray spectroscopy (SEM/EDX, Jeol, JSM-7100F) was used to observe the morphologies and detect the element distribution. The transmission electron microscope (TEM, Jeol, JEM-2010F) was employed to analyze crystal structures and interfaces. The phase identification of the deposited nanofilms was performed by an X-ray diffractometer (XRD, PANalytical) using Cu Kα radiation at 45 kV and 40 mA. Raman scattering spectra were obtained using a Micro-Raman system (Renishaw, RM 3000) equipped with a 514 nm laser at 25 mW. The surface chemistry of the TiN and TiNO nanofilms were characterized by X-ray photoelectron spectroscopy (XPS, Physical Electronics, PHI 5600 multi-technique system). The depth profiles of elements were analyzed by a secondary ion mass spectrometer (SIMS, Physical Electronics, PHI 7200). The reflectance spectra of the SSAs in the wavelength range of 0.3-2.5 µm were collected using a UV-visible-NIR spectrometer (Perkin Elmer, Lambda 950) equipped with a 150 mm integrating sphere. The angle of incidence is 8°. Regarding the specular reflectance spectra in the mid-IR (MIR) range (>2.5 µm), the SSAs were measured by a Fourier transform infrared spectrometer (FTIR, Bruker, Vertex 70) at normal incidence. The mid-IR reflectance spectra were also measured by an FTIR spectrometer (Thermo Fisher Scientific, Nicoletis 50) equipped with a gold-coated integrating sphere. The diffuse reflectance of planar multilayer thin films without nanostructured surfaces is negligible especially for longer wavelengths [14]. All the optical measurements mentioned above were performed at room temperature. Here the absorptance (A) was calculated according to A = 1 − R , because the bottom IR reflectors were 15

thick enough (>100 nm) to allow no transmission. To characterize the properties of each layer, the four nanofilms with a thickness >100 nm each, including TiN, TiNO, ZrO2 and SiO2 layers, were deposited on Si wafers individually. The fabrication processes were the same as those introduced above. It should be noted here that the real IR emittance at 1000 K would be slightly larger (~2.3%) than the result measured at room temperature due to the changed optical properties of components, as discussed in the Supporting Information. 4.3. Thermal stability tests The thermal stability of the four-layered SSAs was tested in an encapsulated quartz tube. The tube was purged with high purity argon (99.99%) for 1 hr before annealing, which also kept flowing to serve as the atmosphere during annealing. Theoretically, the air contamination level in the argon used here (10-1 mbar, 7.5×10-2 Torr) is equal to or higher than that in a typical commercial vacuum tube (10-2-10-3 mbar). The samples were heated up to a target annealing temperature Tt at a heating rate of 20 K/min. After being annealed at Tt for a predetermined period, the samples cooled down to the room temperature naturally inside the tube. The annealing temperatures Tt were set as 773, 900, and 1000 K. To test the long-term thermal stability, the TiN-based SSAs were annealed for 150 hours in total with 30 heating and cooling cycles (tt was 5 hrs for each cycle). For the Au-based, Ta-based, and WTi-based SSAs, the total annealing time tt was 1 hr.

Acknowledgments The authors are thankful for the financial support from the Hong Kong General Research Fund (Grant Nos. 16213015, 16245516, and 16214217) and the Hong Kong Collaborative Research Fund (C6022-16G) and technical support of the Nanosystem Fabrication Facility (NFF) of HKUST for the device fabrication.

Conflict of interest 16

The authors declare no conflict of interests.

Appendix A. Supporting Information Supplementary data associated with this article can be founded in the online version at

References [1] M.A. Green, S.P. Bremner, Energy conversion approaches and materials for high-efficiency photovoltaics, Nat. Mater., 16 (2017) 23-34. [2] M. Romero, A. Steinfeld, Concentrating solar thermal power and thermochemical fuels, Energy Environ. Sci., 5 (2012) 9234-9245. [3] H. Anuta, P. Ralon, M. Talyor, Renewable power generation costs in 2018, in, International Renewable Energy Agency, Abu Dhabi, 2019. [4] L.A. Weinstein, J. Loomis, B. Bhatia, D.M. Bierman, E.N. Wang, G. Chen, Concentrating solar power, Chem. Rev., 115 (2015) 12797-12838. [5] M. Sarvghad, S. Delkasar Maher, D. Collard, M. Tassan, G. Will, T.A. Steinberg, Materials compatibility for the next generation of concentrated solar power plants, Energy Storage Materials, 14 (2018) 179-198. [6] M. Caccia, M. Tabandeh-Khorshid, G. Itskos, A.R. Strayer, A.S. Caldwell, S. Pidaparti, S. Singnisai, A.D. Rohskopf, A.M. Schroeder, D. Jarrahbashi, T. Kang, S. Sahoo, N.R. Kadasala, A. Marquez-Rossy, M.H. Anderson, E. Lara-Curzio, D. Ranjan, A. Henry, K.H. Sandhage, Ceramic–metal composites for heat exchangers in concentrated solar power plants, Nature, 562 (2018) 406-409. [7] C. Murphy, Y. Sun, W. Cole, G. Maclaurin, C. Turchi, M. Mehos, The Potential Role of Concentrating Solar Power within the Context of DOE's 2030 Solar Cost Targets, in, CO: National Renewable Energy Laboratory, Golden, 2019, pp. NREL/TP-6A20-71912. [8] P. Bermel, J. Lee, J.D. Joannopoulos, I. Celanovic, S. M., Selective solar absorbers, in: Annual review of heat transfer, Begell House, Inc., 2010, pp. 231-254. [9] E. Randich, D.D. Allred, Chemically vapor-deposited ZrB2 as a selective solar-absorber, Thin Solid Films, 83 (1981) 393-398. [10] J. Moon, D. Lu, B. VanSaders, T.K. Kim, S.D. Kong, S.H. Jin, R.K. Chen, Z.W. Liu, High performance multi-scaled nanostructured spectrally selective coating for concentrating solar power, Nano Energy, 8 (2014) 238-246. [11] F. Cao, D. Kraemer, T.Y. Sun, Y.C. Lan, G. Chen, Z.F. Ren, Enhanced thermal stability of W-Ni-Al2O3 cermet-based spectrally selective solar absorbers with tungsten infrared reflectors, Adv. Energy. Mater., 5 (2015) 1401042. [12] X.Y. Wang, J.H. Gao, H.B. Hu, H.L. Zhang, L.Y. Liang, K. Javaid, Z.G. Fei, H.T. Cao, L. 17

Wang, High-temperature tolerance in WTi-Al2O3 cermet-based solar selective absorbing coatings with low thermal emissivity, Nano Energy, 37 (2017) 232-241. [13] F. Cao, K. McEnaney, G. Chen, Z.F. Ren, A review of cermet-based spectrally selective solar absorbers, Energy Environ. Sci., 7 (2014) 1615-1627. [14] H. Wang, H. Alshehri, H. Su, L. Wang, Design, fabrication and optical characterizations of large-area lithography-free ultrathin multilayer selective solar coatings with excellent thermal stability in air, Sol. Energy Mater. Sol. Cells, 174 (2018) 445-452. [15] Z.Y. Nuru, M. Msimanga, T.F.G. Muller, C.J. Arendse, C. Mtshali, M. Maaza, Microstructural, optical properties and thermal stability of MgO/Zr/MgO multilayered selective solar absorber coatings, Sol. Energy, 111 (2015) 357-363. [16] W. Li, U. Guler, N. Kinsey, G.V. Naik, A. Boltasseva, J.G. Guan, V.M. Shalaev, A.V. Kildishev, Refractory plasmonics with titanium nitride: broadband metamaterial absorber, Adv. Mater., 26 (2014) 7959-7965. [17] Y. Li, D. Li, D. Zhou, C. Chi, S. Yang, B. Huang, Efficient, scalable, and high-temperature selective solar absorbers based on hybrid-strategy plasmonic metamaterials, Solar RRL, 2 (2018) 1800057. [18] P. Li, B. Liu, Y. Ni, K.K. Liew, J. Sze, S. Chen, S. Shen, Large-Scale Nanophotonic Solar Selective Absorbers for High-Efficiency Solar Thermal Energy Conversion, Adv. Mater., 27 (2015) 4585-4591. [19] V. Rinnerbauer, S. Ndao, Y.X. Yeng, W.R. Chan, J.J. Senkevich, J.D. Joannopoulos, M. Soljacic, I. Celanovic, Recent developments in high-temperature photonic crystals for energy conversion, Energy Environ. Sci., 5 (2012) 8815-8823. [20] K. Cui, P. Lemaire, H. Zhao, T. Savas, G. Parsons, A.J. Hart, Tungsten–carbon nanotube composite photonic crystals as thermally stable spectral-selective absorbers and emitters for thermophotovoltaics, Adv. Energy. Mater., 8 (2018) 1801471. [21] F.L. Chen, S.W. Wang, X.X. Liu, R.N. Ji, L.M. Yu, X.S. Chen, W. Lu, High performance colored selective absorbers for architecturally integrated solar applications, J. Mater. Chem. A, 3 (2015) 7353-7360. [22] C.B. Wang, W. Cheng, P.J. Ma, R.B. Xia, X.M. Ling, High performance Al-AlN solar spectrally selective coatings with a self-assembled nanostructure AlN anti-reflective layer, J. Mater. Chem. A, 5 (2017) 2852-2860. [23] H.C. Barshilia, N. Sevakumar, K.S. Rajam, D.V.S. Rao, K. Muraleedharan, Deposition and characterization of TiAlN/TiAlON/Si3N4 tandem absorbers prepared using reactive direct current magnetron sputtering, Thin Solid Films, 516 (2008) 6071-6078. [24] J. Zhang, T.P. Chen, Y.C. Liu, Z. Liu, H.Y. Yang, W/Cu thin film infrared reflector for TiNxOy based selective solar absorber with high thermal stability, J. Appl. Phys., 121 (2017) 203101. [25]

B.

Usmani,

A.

Dixit,

Spectrally

selective

response

of

ZrOx/ZrC-ZrN/Zr

absorber-reflector tandem structures on stainless steel and copper substrates for high 18

temperature solar thermal applications, Sol. Energy, 134 (2016) 353-365. [26] M. Chirumamilla, A.S. Roberts, F. Ding, D.Y. Wang, P.K. Kristensen, S.I. Bozhevolnyi, K.

Pedersen,

Multilayer

tungsten-alumina-based

broadband

light

absorbers

for

high-temperature applications, Opt. Mater. Express, 6 (2016) 2704-2714. [27] P.N. Dyachenko, S. Molesky, A.Y. Petrov, M. Stormer, T. Krekeler, S. Lang, M. Ritter, Z. Jacob, M. Eich, Controlling thermal emission with refractory epsilon-near-zero metamaterials via topological transitions, Nat. Commun., 7 (2016) 11809. [28] K.A. Arpin, M.D. Losego, A.N. Cloud, H.L. Ning, J. Mallek, N.P. Sergeant, L.X. Zhu, Z.F. Yu, B. Kalanyan, G.N. Parsons, G.S. Girolami, J.R. Abelson, S.H. Fan, P.V. Braun, Three-dimensional self-assembled photonic crystals with high temperature stability for thermal emission modification, Nat. Commun., 4 (2013) 2630. [29] U. Guler, A. Boltasseva, V.M. Shalaev, Refractory plasmonics, Science, 344 (2014) 263-264. [30] G.V. Naik, J.L. Schroeder, X.J. Ni, A.V. Kildishev, T.D. Sands, A. Boltasseva, Titanium nitride as a plasmonic material for visible and near-infrared wavelengths, Opt. Mater. Express, 2 (2012) 478-489. [31] P. Carvalho, F. Vaz, L. Rebouta, L. Cunha, C.J. Tavares, C. Moura, E. Alves, A. Cavaleiro, P. Goudeau, E. Le Bourhis, J.P. Riviere, J.F. Pierson, O. Banakh, Structural, electrical, optical, and mechanical characterizations of decorative ZrOxNy thin films, J. Appl. Phys., 98 (2005) 023715. [32] C.Y. Ting, TiN as a high-temperature diffusion barrier for arsenic and boron, Thin Solid Films, 119 (1984) 11-21. [33] A. Achour, R.L. Porto, M.A. Soussou, M. Islam, M. Boujtita, K.A. Aissa, L. Le Brizoual, A. Djouadi, T. Brousse, Titanium nitride films for micro-supercapacitors: Effect of surface chemistry and film morphology on the capacitance, J. Power Sources, 300 (2015) 525-532. [34] A. Gadanecz, J. Blasing, A. Dadgar, C. Hums, A. Krost, Thermal stability of metal organic vapor phase epitaxy grown AlInN, Appl. Phys. Lett., 90 (2007) 221906. [35] Z.C. Wen, T. Kubota, K. Takanashi, Epitaxial CuN films with highly tunable lattice constant for lattice-matched magnetic heterostructures with enhanced thermal stability, Adv. Electron. Mater., 4 (2018) 1700376. [36] A. Glaser, S. Surnev, F.P. Netzer, N. Fateh, G.A. Fontalvo, C. Mitterer, Oxidation of vanadium nitride and titanium nitride coatings, Surf. Sci., 601 (2007) 1153-1159. [37] J.A. Duffie, W.A. Beckman, Solar Engineering of Thermal Processes, Wiley, Hoboken, New Jersey, 2013. [38] R.P. Haggerty, P. Sarin, Z.D. Apostolov, P.E. Driemeyer, W.M. Kriven, Thermal Expansion of HfO2 and ZrO2, J. Am. Ceram. Soc., 97 (2014) 2213-2222. [39] C.S. Shin, D. Gall, N. Hellgren, J. Patscheider, I. Petrov, J.E. Greene, Vacancy hardening in single-crystal TiNx(001) layers, J. Appl. Phys., 93 (2003) 6025-6028. [40] V. Rinnerbauer, A. Lenert, D.M. Bierman, Y.X. Yeng, W.R. Chan, R.D. Geil, J.J. 19

Senkevich, J.D. Joannopoulos, E.N. Wang, M. Soljacic, I. Celanovic, Metallic photonic crystal

absorber-emitter

for

efficient

spectral

control

in

high-temperature

solar

thermophotovoltaics, Adv. Energy. Mater., 4 (2014) 1400334. [41] M. Franck, J.P. Celis, J.R. Roos, Microprobe Raman-spectroscopy of TiN coatings oxidized by solar beam heat-treatment, J Mater. Res., 10 (1995) 119-125. [42] N.K. Ponon, D.J.R. Appleby, E. Arac, P.J. King, S. Ganti, K.S.K. Kwa, A. O'Neill, Effect of deposition conditions and post deposition anneal on reactively sputtered titanium nitride thin films, Thin Solid Films, 578 (2015) 31-37. [43] D. Peykov, Y.X. Yeng, I. Celanovic, J.D. Joannopoulos, C.A. Schuh, Effects of surface diffusion on high temperature selective emitters, Opt. Express, 23 (2015) 9979-9993. [44] T.K. Kim, B. VanSaders, E. Caldwell, S. Shin, Z.W. Liu, S.H. Jin, R.K. Chen, Copper-alloyed spinel black oxides and tandem-structured solar absorbing layers for high-temperature concentrating solar power systems, Sol. Energy, 132 (2016) 257-266. [45] J.B. Chou, Y.X. Yeng, Y.E. Lee, A. Lenert, V. Rinnerbauer, I. Celanovic, M. Soljacic, N.X. Fang, E.N. Wang, S.G. Kim, Enabling ideal selective solar absorption with 2D metallic dielectric photonic crystals, Adv. Mater., 26 (2014) 8041-8045. [46] F. Cao, L. Tang, Y. Li, A.P. Litvinchuk, J.M. Bao, Z.F. Ren, A high-temperature stable spectrally-selective solar absorber based on cermet of titanium nitride in SiO2 deposited on lanthanum aluminate, Sol. Energy Mater. Sol. Cells, 160 (2017) 12-17. [47] M.H. Liu, E.T. Hu, Y. Yao, K.Y. Zang, N. He, J. Li, Y.X. Zheng, S.Y. Wang, O. Yoshie, Y. Lee, C.Z. Wang, D.W. Lynch, L.Y. Chen, High efficiency of photon-to-heat conversion with a 6-layered metal/dielectric film structure in the 250-1200 nm wavelength region, Opt. Express, 22 (2014) A1843-A1852. [48] H. Wang, I. Haechler, S. Kaur, J. Freedman, R. Prasher, Spectrally selective solar absorber stable up to 900  °C for 120 h under ambient conditions, Sol. Energy, 174 (2018) 305-311.

20

Figures

Fig. 1. Design of all-ceramic SSAs and characterizations of each layer. (a) The 3D schematic of the all-ceramic TiN/TiNO/ZrO2/SiO2 four-layer structure for selective sunlight absorption. (b) XRD patterns of both the TiN reflector and TiNO absorptive layer. (c-d) The XPS spectra and high-resolution Ti 2p XPS spectra of the TiN and TiNO nanofilms. The curves are artificially offset along the y-axis for comparison. (e) Complex refractive index of each layer obtained from ellipsometer measurement.

21

Fig. 2. Fabrication of all-ceramic SSAs. (a) Simulated absorption spectra of the SSA made of all-ceramic four-layered nanofilms at different stages. Stage 1: TiN; Stage 2: TiN/TiNO; Stage 3: TiN/TiNO/ZrO2; and Stage 4: TiN/TiNO/ZrO2/SiO2. (b) RGB colors of the all-ceramic SSA at the four stages derived from the simulated spectra. (c-d) Photographs and the corresponding absorption spectra of the fabricated SSA at the four stages. (e-f) The cross-sectional SEM image and TEM image of the SSA at Stage 4. Inset: the high-resolution TEM image of the marked regime in (f).

22

Fig.3. Optical performance of all-ceramic SSAs. (a) The measured absorptance spectrum of the all-ceramic SSA. AM 1.5G solar spectral irradiance (0.3-4.0 µm) with a concentration ratio C of 100, the utilized solar radiation by the SSA, and the thermal emission spectra for a blackbody and the SSA at 1000 K. (b) The solar-thermal energy conversion efficiency η of the all-ceramic SSA versus concentration ratios C and operating temperatures TA. (c) The measured absorptance spectra of the Au, Ta, and TiN nanofilms, and those of the SSAs with Au, Ta, or TiN reflectors.

23

Fig. 4. Thermal stability of metal-based SSAs. (a) The absorptance spectra of the Ta-based SSA before and after annealing at 773, 900, and 1000 K for 1 hr. (b) The Raman spectra of the Ta-based SSA before and after annealing at 773 and 900 K. The spectra are artificially offset along the y-axis for comparison. (c) The depth profile of elements before and after annealing at 900 K. (d-f) The surface SEM image, the surface EDX mapping, and the cross-sectional SEM image of the Ta-based SSA after annealing at 1000 K. The arrows in (d) indicate the failure areas. The right side in (e) corresponds to the failure regime. See also Fig. S5-S6.

24

Fig. 5. Thermal stability of all-ceramic SSAs. (a) The absorptance spectra of the all-ceramic SSA before and after annealing at 1000 K. (b) The depth profiles of Ti, N, Zr, and O for the SSA before and after annealing at 1000 K. (c) The cross-sectional SEM image of the all-ceramic SSA after annealing at 1000 K. (d) Comparison of η (under 100-sun irradiation and at 1000 K) and thermal stability of the proposed all-ceramic four-layer SSA with recently reported high-performance SSAs. Photonic crystals (PhCs): Ref. [40] Ta/HfO2, Ref. [18] Ni/Al2O3, Ref. [45] Ru/HfO2, and Ref. [20] W-C nanotube/Al2O3. Plasmonic SSAs: Ref. [17] Ti/Al2O3/Ta, and Ref. [16] TiN/SiO2/TiN. Cermets: Ref. [11] [email protected], Ref. [12] [email protected], and Ref. [46] [email protected] Multilayer SSAs: Ref. [47] Cu/SiO2/Ti/SiO2/Ti/SiO2, Ref. [15] Zr/MgO/Zr/MgO, Ref. [14] W/SiO2/W/Si3N4/SiO2, Ref. [27] W/HfO2 ten-layer, Ref. [48] black chrome/ITO/SiO2, and this work. Solid symbols indicate the maximum operating temperatures were obtained in vacuum or in inert gas conditions, while the half symbols correspond to those obtained in air.

25

Tables Table 1. The solar absorptance α , thermal emittance ε at 1000 K, and solar-thermal energy conversion efficiency η under 100 suns. Samples

Ta-based SSAs

WTi-based SSAs

All-ceramic SSAs

α

ε

η

α

ε

η

α

ε

η

Before annealing

90.8%

31.5%

72.9%

87.5%

24.3%

73.7%

92.2%

17.0%

82.6%

773 K (1 hr)

84.4%

32.8%

65.8%

85.3%

28.8%

69.0%

-

-

-

900 K (1 hr)

77.5%

29.7%

60.7%

-

-

-

-

-

-

1000 K (1 hr)

73.7%

30.6%

56.3%

80.2%

29.3%

63.6%

-

-

-

1000 K (50 hrs)

-

-

-

-

-

-

91.1%

15.9%

82.1%

1000 K (150 hrs)

-

-

-

-

-

-

90.1%

15.2%

81.5%

26

Yang Li received his B.S. degree from Huazhong University of Science and Technology in 2012 and M.S. degree from Zhejiang University in 2015. After that, He received his Ph.D. degree in the Department of Mechanical and Aerospace Engineering from The Hong Kong University of Science and Technology (HKUST) in 2019. He is currently a postdoctoral research fellow of Prof. Baoling Huang’s Group in HKUST. His research interests spectrally

include

solar-thermal

selective

energy

metasurfaces,

and

conversion, plasmonic

metamaterials.

Chongjia Lin received his B.S. degree from South China Agricultural University in China in 2015 and M.S. degree in Intelligent Building Technology and Management from The Hong Kong University of Science and Technology (HKUST) in 2017. He is currently a Ph.D. candidate at the

Department

of

Mechanical

and

Aerospace

Engineering in HKUST. His current research is mainly on the passive radiative cooling.

Dan Zhou received her B.S. degree in Material Physics from University of Science and Technology of China in 2015, and M.S. degree in Chemistry from The Hong Kong University of Science and Technology in 2018. Her research focuses on transition metal based catalysts for electrochemical water splitting.

27

Yiming An received her B.S. degree in College of Chemistry from Beijing Normal University. She is currently a Ph.D. candidate in Prof. Shihe Yang’s group in Department of Chemistry of The Hong Kong University of Science and Technology. Her current research focuses on hydrogen and oxygen generation by electrochemical (PEC) water splitting.

Dezhao Li received his B.S. degree from Tsinghua University in 2007 and Ph.D degree in Mechanical and Aerospace Engineering from The Hong Kong University of Science and Technology in 2018. He then joined the Zhejiang University of Technology as an associate professor.

His

research

mainly

focuses

on

the

MEMS/NEMS devices’ design, fabrication, and their applications in energy fields.

Cheng Chi received his B.S. degree from South China University of Technology in China in 2012 and M.S. degree in Advanced Materials and Manufacturing from The Hong Kong University of Science and Technology (HKUST) in 2015. He is currently a Ph.D. candidate at the

Department

of

Mechanical

and

Aerospace

Engineering in HKUST. His current research is mainly on the

solid

polymer

supercapacitors.

28

electrolyte

based

batteries

&

He Huang received his B.S. degree from Xi'an Jiaotong University, China in 2014 and Ph.D. degree from the Hong Kong University of Science and Technology, Hong Kong in 2019. He is currently a postdoctoral research fellow in Professor Baoling Huang’s Group. His current research interests include molecular dynamics simulations and density functional theory calculations applied to energy storage materials for lithium and sodium batteries.

Shihe Yang is Professor at The Hong Kong University of Science and Technology. Now he is also Professor at the Shenzhen Graduate School, Peking University. He has contributed

to

metalloful-lerenes,

cluster

science,

fullerenes

and

soft

molecular

interfaces

and

nanomaterials. His current research interests include the understanding,

ma-nipulation

and

applications

of

low-dimensional nanosys-tems and multiscale materials for

energy

conversion,

particularly

solar

energy

harvesting.

Chi-Yan Tso is an Assistant Professor in City University of Hong Kong. Dr. Tso received his B.S. degree in Mechanical Engineering with 1st class honors, M.S. degree in Environmental Engineering and Ph.D. degree in Mechanical Engineering from The Hong Kong University of Science and Technology

in 2010, 2012 and 2015,

respectively. His research interest covers thermofluid and energy conversion in a built environment.

29

Christopher Y.H. Chao is Dean of Engineering and Chair Professor of Mechanical Engineering at The University of Hong Kong (HKU). He received his BSc(Eng) degree in Mechanical Engineering (First Class) from HKU in 1988 and

worked

for

Swire

Industrial

Division

from

1988-1990. He obtained his M.S. and Ph.D. degrees in Mechanical California,

Engineering Berkeley,

from

USA,

The in

University

1992

and

of

1994,

respectively. He has a wide range of research interests in the areas of built environment, energy and environmental engineering. Baoling Huang is currently an Associate Professor in The Hong Kong University of Science and Technology. He received his B.S. and M.S. degrees from Tsinghua University, China in 1999 and 2001. He worked in industry from 2001 to 2004. In 2008, he received his Ph.D. degree in Mechanical Engineering from the University of Michigan, Ann Arbor, USA. Then, he worked as a postdoctoral research fellow at the University of California, Berkeley and Lawrence Berkeley National Laboratory. His research interests are in the broad area of energy transport, conversion and storage.

30

Highlights 

We first introduce all-ceramic structures into selective solar absorbers.



The absorber offers state-of-the-art spectral selectivity and thermal stability.



The maximum operating temperature of multilayer solar absorbers is boosted by 227 K.



The all-ceramic absorber can be readily produced in quantity via low-cost process.