Open Access
Add to BibSonomy
Abstract
Blocking the near-infrared region (NIR) is indispensable for saving energy consumed to maintain the interior temperature in buildings. However, simultaneously enhancing transmission in visible light and blocking in the NIR remains challenging. Here, we theoretically demonstrate a transparent all-dielectric metasurface selectively blocking the NIR by using TiO2 nanocylinders and an indium tin oxide (ITO) layer. The ITO layer is implemented as a back reflector because ITO is transparent in visible light, whereas the ITO becomes a reflective material in the long-wavelength region (λ > 1500 nm). The designed metasurface exhibits high average transmittance of 70% in visible light and high solar energy rejection (SER) of 90% in the NIR. Furthermore, the blocking capability in the NIR of the designed metasurface is maintained over a wide range of an incident angle and polarization angle of light. Therefore, the metasurface gives a guideline for designing energy-saving applications.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
With the drastic population growth and industrialization, excessive energy has been annually consumed in buildings and automobiles. A significant percentage of energy is used to maintain the interior temperature because of the poor sun-shielding of windows [1–3]. To save the energy, blocking near-infrared region (NIR) in the solar spectrum has been received great attention because NIR accounts for about 50% of the solar spectrum and it has a high heat transfer rate [3]. Thus, selectively blocking the NIR in sunlight is a key technology for the radiative cooling effect which can suppress the elevation of indoor temperature [3–7]. Recently, metasurfaces have been receiving great attention due to efficiently control of absorption and reflection of light [8,9]. The metasurfaces compose of nanostructures with a sub-wavelength scale, tailoring electromagnetic properties by controlling the geometry of nanostructures [10–29]. Therefore, the capability to block the NIR can be improved by employing metasurfaces.
The first metasurfaces were based on metal–insulator–metal (MIM) configurations [10–20] such as Ti nanodisk on SiO2/Au/Si substrate [19] and Ag/SiO2 multilayer grating on Al substrate [20]. Because metals have a high intrinsic absorption coefficient, a magnetic field can be strongly confined in the insulator layer between metals, thereby leading to strong absorption of light due to localized surface plasmon resonance. However, the MIM metasurfaces suffer from a problem of the thermal stability of metals because the absorbed energy heats them. To solve the problem, all-dielectric metasurfaces have been extensively investigated because of the high thermal stability of dielectric materials.
There were previously reported dielectric absorbers such as TiN/Si cylinder on silicon [21], Ge disk on GaF2 substrate [22], and SiNx/TiN/quartz cylinder on a quartz substrate [23]. They blocked NIR by absorption rather than reflection because those dielectric materials have a high extinction coefficient in NIR. But, the part of the absorbed energy is released into inside buildings, resulting in an inefficient radiative cooling effect. In other words, the released energy acts as a heating source and reduces the radiative cooling effect. To solve the problems, all-dielectric metasurfaces have been designed to maximize reflection in NIR because reflection can block all radiative energy, unlike absorption.
However, the previous all-dielectric metasurfaces have difficulty in reflecting overall NIR because the bandwidth of reflection is too narrow [24–26]. Recently, several attempts have been conducted to extend the bandwidth of reflection by using multilayer nanostructure [27–29]. An alternately stacked nanocylinder with GaAs/AlGaO is designed on GaAs substrate [29]. High refractive index difference between GaAs and AlGaO allows to reflect over a spectral wavelength between 1010 nm and 1420 nm, but it does not block the whole range of NIR (800 nm ≤ λ ≤ 2500 nm), resulting in a low blocking efficiency. And the other disadvantage of the stacked nanocylinder is no transparency in the visible wavelength region, so it is restricted to apply to transparent windows. Such problems could be solved by inserting an ITO layer as a back reflector because ITO is reflective at λ > 1500 nm, but transparent in visible wavelengths [30].
Here, we design an all-dielectric metasurface with ITO back reflectors that overcomes the poor transparency and improves the solar energy rejection in NIR. A TiO2 nanocylinder and ITO layer were systemically optimized to block overall NIR. The TiO2 nanocylinder was designed to reflect a spectral range of 800 nm ≤ λ ≤ 1130 nm, while transmitting visible light for application on windows. The ITO layer was designed to block the NIR in λ > 1130 nm. Consequently, the designed metasurface shows high solar energy rejection (SER) of 90% in NIR (reflection 53% and absorption 37%) as well as high transmission of 70% in visible light.
2. Materials and design of metasurface
The solar spectrum (Air Mass 1.5) should be considered to get the actual efficiency of which sunlight is blocked. The solar spectrum has ranges in which intensity decreases sharply due to absorption by the atmosphere (Fig. l(a)) [31]. These decreases are used to divide NIR into A band (800 nm ∼ 1130 nm), B band (1140 nm ∼ 1350 nm), C band (1410 nm ∼ 1810nm) and D band (1940nm ∼ 2500 nm). The percentage of energy intensity in NIR is calculated to be 54% (A band), 20% (B band), 18% (C band) and 8% (D band), respectively. The actual efficiency to block sunlight can be evaluated by solar energy rejection (SER), which defines the percentage of solar energy that is blocked when sunlight passes through a metasurface. SER is provided as follows [32–34]:
Here, we designed a metasurface composed of TiO2 nanocylinders (NCs) with an ITO back reflector (TiO2 NC/SiO2/ITO/glass) by considering SER. The TiO2 NC was designed to block A ∼ B bands and ITO layer was designed to block C ∼ D bands. The transparency of TiO2 and ITO in the visible wavelengths enables the design of metasurface that blocks most NIR whereas transmits visible light (Fig. 1(b)).
Fig. 1. (a) Solar irradiance spectrum (Air mass 1.5) (b) Schematic diagram of TiO2 nanocylinder (NC) metasurface with ITO back reflector. TiO2 NC and ITO layer block A ∼ B band and C ∼ D band in NIR, respectively, whereas transmit visible light. (c) Schematic diagrams of TiO2 nanocylinder (NC)/glass (blue), ITO/glass (red) and TiO2 NC/SiO2/ITO/glass (black).
TiO2 is suitable for use as a transparent metasurface because TiO2 has a high refractive index in NIR as well as high transparency in visible light. The TiO2 NC can obtain perfect reflection (i.e., > 99%) due to Mie resonance. The perfect reflection is generated when the complex effective permittivity (εeff) and complex effective permeability (μeff) of the dielectric nanostructure satisfy [35]:
ITO has high transmittance in the visible wavelengths as well as high reflectance in the NIR, and is, therefore, a desirable candidate material as a transparent back reflector. The ITO has plasma wavelength λplasma ∼ 1500 nm, which is defined as λ at which transmittance is the same as reflectance (Supplement 1, Fig. S1) [30]. At λ < λplasma, ITO has high transmittance due to dielectric property, and at λ > λplasma, ITO has high reflectance and absorbance due to metallic property. Thus, the ITO layer can block the C ∼ D bands in NIR.
To design a transparent metasurface that blocks NIR, we designed TiO2 NC/glass to block A ∼ B bands (Fig. 1(c), blue) and ITO layer/glass (Fig. 1(c), red) to block C ∼ D bands. Then we stacked the two optimized metasurfaces to integrate the blocking regions (TiO2 NC/SiO2/ITO/glass, Fig. 1(c), black). Reflectance and absorbance spectrums were calculated using rigorous coupled-wave analysis (RCWA) simulation. The calculations were performed under unpolarized light with an equal contribution from both polarization states (50% TE and 50% TM). Normal incident plane waves were used with the wavelength range of 200 nm to 2500 nm in intervals of 10 nm. The refractive index and extinction coefficient of TiO2 [40], ITO [41] and SiO2 were set to be references, respectively. In addition, the optical properties of metasurface are regardless of the thickness of glass, so the glass thickness was fixed to 3 µm to reduce the simulation time (Supplement 1, Fig. S3).
3. Results and discussion
3.1. Optimization of TiO2 NC/glass and ITO/glass
To optimize the TiO2 NC/glass, we systemically modified diameter D, height H and period P of TiO2 NC (Fig. 2(a)). In a variation of 0 nm ≤ D ≤ 550 nm, with fixed H = 360 nm and P = 660 nm, the reflection peak shifts toward a longer wavelength, as D increased. At D = 440 nm, the reflection region was located in the A ∼ B bands (Fig. 2(b)). To further investigate the effectiveness of the optimized metasurface, the SER and average transmittance Tavg were calculated over the wavelength ranges of 800 nm ≤ λ ≤ 2500 nm and 400 nm ≤ λ ≤ 800 nm, respectively. The highest SER = 29% occurred at D = 440 nm. Meanwhile, Tavg gradually decreased because the increased area of TiO2 NC with a high refractive index induces a large amount of the Fresnel reflection (Fig. 2(c)). At D = 440 nm, Tavg is sufficiently high at 70%, so we determined that the optimal D is 440 nm.
Fig. 2. (a) Schematic diagrams of the TiO2 NC/glass. 3D view of a unit cell (top) and vertical view of the arrayed nanostructures (bottom). Optimization of TiO2 NC/glass metasurface with various (b-c) diameter D, (d-e) height H and (f-g) period P of TiO2 NC. Calculated solar energy rejection (SER) in NIR and average transmittance Tavg in visible light as a function of TiO2 NC parameters.
As H was increased from 300 nm to 900 nm, the reflection peak shifted toward a longer wavelength. The bandwidth of reflection was maximized at H = 510 nm but significantly decreased at H > 510 nm due to a split of the peak (Fig. 2(d) and Supplement 1, Fig. S4). Since large bandwidth can increase reflection in the A ∼ B bands, the TiO2 NC/glass had SER = 37% at H = 510 nm (Fig. 2(e)). Also, Tavg increased from 70% to 77%, so the optimal H was determined to be 510 nm.
In a variation of 300 nm ≤ P ≤ 900 nm, with duty cycle D/P fixed at 66%, the reflectance peak shifted from the visible region to the C band, as P increased. At P = 660 nm, the region of reflection was located in the A ∼ B bands (Fig. 2(f)), thereby leading to high SER of 37%. Also, the high Tavg of 77% is obtained at P = 660 nm (Fig. 2(g)), so we determined that optimal P is 660 nm. The optimized TiO2 NC/glass showed perfect reflection at λ = 1160 nm as well as a high reflection at 970 nm ≤ λ ≤ 1260 nm, resulting in blocking the A ∼ B bands of 48%.
Consequently, as the parameters (D, H, and P) of TiO2 NC were increased, the position of the reflection peak shifted towards a longer wavelength. This change can be explained by Mie resonance. The wavelength of the Mie resonance (λresonance) and parameters of the nanostructure are related as [42,43]:
The ITO/glass was optimized by testing the consequences of varying of the thickness HITO of the ITO layer (0 nm ≤ HITO ≤ 3000 nm) (Fig. 3). At HITO > 600 nm, the reflectance spectrum is almost maintained in the NIR regardless of HITO (Fig. 3(a)). However, as HITO increased, the intensity of absorption gradually increased in visible region and NIR (Fig. 3(b)), so SER increased whereas Tavg decreased. Thus, to design a transparent metasurface, we use HITO = 2700 nm due to high Tavg = 70% as well as SER = 78% (Fig. 3(c)). The optimized ITO/glass shows reflection at λ > 1420 nm (Fig. 3(a)) and high absorption in 1010 nm ≤ λ ≤ 1710nm (Fig. 3(b)), so it blocks 100% of the C ∼ D bands.
Fig. 3. Optimization of ITO/glass with various thickness HITO of the ITO layer. Simulated (a) reflectance and (b) absorbance spectrum. Simulated (c) solar energy rejection (SER) in NIR and average transmittance Tavg in visible light as a function of HITO.
3.2. Stacking the optimized TiO2 NC/glass on ITO/glass
To integrate two optimized TiO2 NC and ITO layer by stacking, a SiO2 layer is used as a spacer between TiO2 NC and the ITO layer (Fig. 4(a)). To determine the thickness HSiO2 of the SiO2 layer, we simulated variation of 0 nm ≤ HSiO2 ≤ 500 nm (Figs. 4(b)-(d)). Without the SiO2 spacer, the intensities of reflection peaks decreased by interaction between TiO2 NC and the ITO layer. As HSiO2 increased, the interaction gradually decreased. At HSiO2 = 300 nm, the region of reflection in the A ∼ B bands was almost saturated, so the TiO2 NC/SiO2/ITO/glass can reflect both 980 nm ≤ λ ≤ 1260 nm and λ > 1610 nm (Supplement 1, Fig. S6); this result shows the independent operation of the TiO2 NC and the ITO layer. As the region of the reflection was increased, the region of absorption was decreased by the same portion (Figs. 4(b) and (c)). Thus, the SER was almost constant over the range of HSiO2 (Fig. 4(d)). To further investigate the effect of the HSiO2 on reflection and absorption, the average reflectance Ravg and absorbance Aavg were calculated over range of 800 nm ≤ λ ≤ 2500 nm. Though the SER were invariant, Tavg, Ravg, and Aavg were changed and then almost saturated by the variation of HSiO2. At HSiO2 = 300 nm, The TiO2 NC/SiO2/ITO/glass was saturated at Ravg = 55%, Aavg = 41%, Tavg = 55% and SER = 86%. Thus, we determined optimum HSiO2 is 300 nm.
Fig. 4. (a) Schematic diagram of the integration of the optimized TiO2 NC/glass and ITO/glass by stacking. A SiO2 layer is implemented as a spacer between TiO2 NC and ITO layer (TiO2 NC/SiO2/ITO/glass). Simulated (b) reflectance and (c) absorbance spectra with various thickness HSiO2 of SiO2 layer. (d) Average transmittance Tavg in visible light, average reflectance Ravg in NIR, average absorbance Aavg in NIR, and calculated SER.
3.3. Optimization of TiO2 NC/SiO2/ITO/glass
To optimize the TiO2 NC/SiO2/ITO/glass metasurface (Fig. 5(a)), we modified D, H, and P of TiO2 NC (Fig. 5), and also HITO, and HSiO2 (Supplement 1, Fig. S7). In the range of 0 nm ≤ D ≤ 600 nm, with fixed H = 510 nm, P = 660 nm, HITO = 2700 nm, and HSiO2 = 300 nm, the reflection peak shifted from the A band to the B band and then split it at D = 600 nm (Fig. 5(b)). At D = 320 nm, the reflection region was located in the A band, and SER was maximized at 89%. In addition, Tavg increased from 57% to 63% due to the decrease in D (Fig. 5(c)). Thus, we determined that the optimal D is 320 nm. In a variation of 0 nm ≤ H ≤ 880 nm, the reflection peak shifted toward a longer wavelength as H increased (Fig. 5(d)). The maximized reflected region was 840 nm ≤ λ ≤ 1070 nm at H = 470 nm, which was located in A band. Thus, the metasurface showed a high SER = 89% (Fig. 5(e)). Also, Tavg was maintained at 63%, so the optimal H was determined to be 470 nm.
Fig. 5. (a) Schematic diagrams of the transparent metasurface selectively blocking NIR (TiO2 NC/SiO2/ITO/glass). 3D view of a unit cell (top) and vertical view of the arrayed nanostructures (bottom). Simulation of the metasurface with various (b-c) diameter D, (d-e) height H and (f-g) period P of TiO2 NC. Reflectance and absorbance are solid and dashed lines, respectively.
As P was increased from 300 nm to 900 nm, with the duty cycle fixed at 50%, the reflection peak shifted dramatically from visible region to the B band (Fig. 5(f)). At P = 610 nm, the reflection peak was located in the A band, so the SER was maximized at 91% (Fig. 5(g)). In addition, Tavg slightly increased from 63% to 65%, so we determined that the optimal P is 610 nm and the optimal D is 305 nm, respectively.
In a variation of 0 nm ≤ HITO ≤ 3000 nm (Fig. 6 and Supplement 1, Figs. S7(a)-(b)), Ravg in the NIR was almost constant when HITO > 900 nm (Fig. 6(a)). On the other hand, as HITO increased, the intensities of absorption peak in the visible range and in the NIR gradually increased because the optical path length passing through the ITO layer increased (Supplement 1, Fig. S7(b)). Thus, Aavg gradually increased as HITO increased. As a result, as HITO increased, SER increased whereas Tavg gradually decreased. However, to improve the applicability of the metasurface, we should increase Tavg. For example, in automobiles, a strict regulation about transmission of visible light has been enacted to prevent traffic accidents. The regulation states that window tinting film for automobiles must have transmission ≥ 70% in visible light [44]. Therefore, to design a highly-transparent metasurface with Tavg ≥ 70%, we determined that the optimal HITO is 2000nm.
Fig. 6. Simulation of the metasurface with various thickness HITO of ITO layer of TiO2 NC/SiO2/ITO/glass metasurface. To investigate the effect of the HITO on the effectiveness of the metasurface, (a) average transmittance Tavg in visible light, average reflectance Ravg in NIR, average absorbance Aavg in NIR, and (b) SER were calculated.
HSiO2 was increased from 0 nm to 500 nm (Supplement 1, Figs. S7(c)-(f)). At HSiO2 = 0 nm, the reflectance was low because of the interaction between TiO2 NC and the ITO layer [45] (Supplement 1, Fig. S7(c)). As HSiO2 increased, this interaction gradually decreased. At HSiO2 ≥ 400 nm, Tavg, Ravg, Aavg and SER were almost saturated at 70%, 48%, 46% and 87%, respectively (Supplement 1, Figs. S7(e)-(f)). Thus, we determined that the optimal HSiO2 is 400 nm. These results indicate that HSiO2 and HITO are related to the intensities of absorption and reflection peaks rather than to the shifts of peaks because thicknesses of SiO2 and ITO layers do not affect the geometry of the nanostructure related to Mie resonance. The optimized metasurface can be obtained by repeating variations of parameters such as D, H, P, HITO, and HSiO2.
The optimized metasurface with TiO2 NC (D = 330 nm, P = 550 nm and H = 430 nm)/SiO2 (HSiO2 = 440 nm)/ITO (HITO = 1800nm)/glass showed high transmission over all visible wavelengths. Also, the designed metasurface blocked NIR by reflection (810 nm ≤ λ ≤ 1030 nm and λ > 1660 nm) and absorption (1090 nm ≤ λ ≤ 1650 nm) (Fig. 7(a)). The solar irradiance spectrum passing through the optimized metasurface was shown in Fig. 7(b). The spectrum exhibited a high solar irradiance in the visible region, which meant the designed metasurface had high transparency, whereas, in the NIR, the SER of each band was 90% (A band), 78% (B band), 100% (C band) and 100% (D band), respectively. Thus, the metasurface exhibited Tavg of 70% as well as SER in NIR of 90% (reflection of 53% and absorption of 37%). To compare the performances, we summarized previously reported dielectric metasurfaces in Table S1. The blocking capability in NIR was dramatically improved, which is almost twice, compared to the references.
Fig. 7. (a) Optical spectra of the designed metasurface. (b) Solar irradiance spectrum passing through the metasurface (red); the reference spectrum is Air Mass 1.5 (black).
To examine the contributions of the TiO2 NC and the ITO layer in the designed metasurface, we simulated it using a finite-difference time-domain (FDTD) method. We determined seven wavelengths (λ1 ∼ λ7), such as peaks and crossing points in the optical spectrum (Fig. 7(a)). The top view of electromagnetic field distribution (xy plane, Figs. 8(a) and (b)) and the cross-section view of absorption distribution (zx plane, Fig. 8(c)) were obtained at steady state by using a discrete Fourier transform (DFT) monitors. At λ3 = 950 nm, strong electric and magnetic field distribution are observed in the TiO2 NC. These field distributions indicate a magnetic dipole Mie resonance, so the metasurface has a perfect reflection at the wavelength [35–37]. In contrast, at the other wavelengths (λ ≠ λ3), electric and magnetic fields in the TiO2 NC become weak, so the reflection is lower than at λ = λ3. At λ ≥ λ4, the intensity of the magnetic dipole resonance gradually declined (λ = λ4), and then disappeared (λ > λ5). Therefore, we demonstrated that the perfect reflection at 910 nm ≤ λ ≤ 960 nm is induced by a magnetic dipole Mie resonance of TiO2 NC.
Fig. 8. (a) Electric, (b) magnetic field (on xy plane, top view) and (c) absorption distribution (on zx plane, cross-section view) of the designed metasurface at wavelengths of λ1 ∼ λ7. These wavelengths were peaks and crossing points in the optical spectrum in Fig. 7(a).
The intensity of absorption in the ITO layer gradually increased as the wavelength of incident light increased (λ < λ5). However, at λ = λ3, the absorption is near zero because of the perfect reflection of TiO2 NC. At λ = λ5, absorption is maximized and the intensity of absorption in the ITO layer continuously increased toward the top of the ITO layer. At λ > λ6, the absorption decreased due to reflection by the ITO layer (λ6 > λplasma). Thus, the absorption of the designed metasurface in the range of 1090 nm ≤ λ ≤ 1630 nm is mainly due to the ITO layer.
For a metasurface to have ideal blocking capability, it should be independent of the incident angle of light. Calculated transmittance spectrum was almost constant for an incident angle of 0° to 50° (Fig. 9(a)). To precisely investigate the effect of the incident angle, the SER and Tavg were calculated using the transmittance spectrum (Fig. 9(b)). As the incident angle of the light increased from 0° (normal direction) to 50°, the SER in NIR almost maintained from 90% to 86%. The average transmittance in visible region decreased slightly from 70% to 52% due to a reflection induced by changing in an optical path length passing through the metasurface. Consequently, the designed metasurface effectively transmits and blocks sunlight over a wide range of incident angles.
Fig. 9. (a) Transmittance contour map as a function of the wavelength and incident angle of the light source in simulation. (b) Solar energy rejection (SER) in NIR and average transmittance Tavg in visible light as functions of the incident angle of the light source.
In addition, the performance of the designed metasurface should be independent of the polarization angle of the incident light because sunlight is unpolarized light. Simulated transmittance spectrum was almost constant regardless of the polarization angle of the normal incident light (Supplement 1, Fig. S8). As the polarization angle of the incident light increased from 0° (TM mode) to 90° (TE mode), the average transmittance in visible light and the SER in NIR are almost constant at 70% and 90%, respectively. These results are because cylinder nanostructure constituting the metasurface has a symmetrical shape. Thus the designed metasurface can be applied to application for blocking sunlight.
4. Conclusions
In summary, we designed the transparent metasurface selectively blocking NIR by implementing the ITO layer as a back reflector because the ITO is transparent in visible light whereas the ITO becomes reflective materials in the long-wavelength region (λ > 1500 nm). On the basis of SER and Tavg, the optimal metasurface was determined by modifying parameters of TiO2 NC (D, P and H), SiO2 (HSiO2) and ITO (HITO) layer. Therefore, a high transparent metasurface with the Tavg = 70% in visible light was achieved, exhibiting SER = 90% in a spectral range of 800 nm ∼ 2500 nm (reflection of 53% and absorption of 37%). From FDTD simulation, the contributions of the TiO2 NC and ITO layer in the designed metasurface were demonstrated. Furthermore, the blocking capability in NIR of the designed metasurface was maintained over a wide range of an incident angle and polarization angle of light. Therefore, it can lead to practical applications in the field related to sunlight such as heat blocking film, window tinting film, and energy-efficient buildings. We believe this proposed metasurface gives an advanced guideline to design next-generation energy-saving applications.
Funding
Brain Korea 21 PLUS project for Center for Creative Industrial Materials (F18SN25D1706); Global Frontier Hybrid Interface Materials (2013M3A6B1078874).
Acknowledgments
This research was supported by Global Frontier Program through the Global Frontier Hybrid Interface Materials (GFHIM) of the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT (2013M3A6B1078874). This research also was financially supported by Brain Korea 21 PLUS project for Center for Creative Industrial Materials (F18SN25D1706).
Disclosures
There are no financial conflicts of interest to disclose.
See Supplement 1 for supporting content.
References
1. J. Liu, Z. Zhou, J. Zhang, W. Feng, and J. Zuo, “Advances and challenges in commercializing radiative cooling,” Mater. Today Phys. 11100161 (2019). [CrossRef]
2. Z. Xia, Z. Fang, Z. Zhang, K. Shi, and Z. Meng, “An easy way to achieve the passive radiative materials’ self-adaptive cooling,” ACS Appl. Mater. Interfaces 12(24), 27241–27248 (2020). [CrossRef]
3. B. Ko, D. Lee, T. Badloe, and J. Rho, “Metamaterial-based radiative cooling: Towards energy-free all-day cooling,” Energies 12(1), 89 (2018). [CrossRef]
4. A. P. Raman, M. Abou Anoma, L. Zhu, E. Rephaeli, and S. Fan, “Passive radiative cooling below ambient air temperature under direct sunlight,” Nature 515(7528), 540–544 (2014). [CrossRef]
5. R. Yang and X. Yin, “Passive cooling in an urban setting,” Nat. Sustain. 2(8), 663–664 (2019). [CrossRef]
6. T. Li, Y. Zhai, S. He, W. Gan, Z. Wei, M. Heidarinejad, D. Dalgo, R. Mi, X. Zhao, and J. Song, “A radiative cooling structural material,” Science 364(6442), 760–763 (2019). [CrossRef]
7. Y. Zhai, Y. Ma, S. N. David, D. Zhao, R. Lou, G. Tan, R. Yang, and X. Yin, “Scalable-manufactured randomized glass-polymer hybrid metamaterial for daytime radiative cooling,” Science 355(6329), 1062–1066 (2017). [CrossRef]
8. S. S. Bukhari, J. Y. Vardaxoglou, and W. Whittow, “A metasurfaces review: Definitions and applications,” Appl. Sci. 9(13), 2727 (2019). [CrossRef]
9. S. Sun, Q. He, J. Hao, S. Xiao, and L. Zhou, “Electromagnetic metasurfaces: Physics and applications,” Adv. Opt. Photonics 11(2), 380–479 (2019). [CrossRef]
10. A. S. Rana, M. Q. Mehmood, H. Jeong, I. Kim, and J. Rho, “Tungsten-based ultrathin absorber for visible regime,” Sci. Rep. 8, 2443 (2018). [CrossRef]
11. D. Lee, S. Y. Han, Y. Jeong, D. M. Nguyen, G. Yoon, J. Mun, J. Chae, J. H. Lee, J. G. Ok, and G. Y. Jung, “Polarization-sensitive tunable absorber in visible and near-infrared regimes,” Sci. Rep. 8, 12393 (2018). [CrossRef]
12. I. Kim, S. So, A. S. Rana, M. Q. Mehmood, and J. Rho, “Thermally robust ring-shaped chromium perfect absorber of visible light,” Nanophotonics 7(11), 1827–1833 (2018). [CrossRef]
13. A. K. Azad, W. J. Kort-Kamp, M. Sykora, N. R. Weisse-Bernstein, T. S. Luk, A. J. Taylor, D. A. Dalvit, and H. T. Chen, “Metasurface broadband solar absorber,” Sci. Rep. 6(1), 20347 (2016). [CrossRef]
14. A. V. Pogrebnyakov, J. A. Bossard, J. P. Turpin, J. D. Musgraves, H. J. Shin, C. Rivero-Baleine, N. Podraza, K. A. Richardson, D. H. Werner, and T. S. Mayer, “Reconfigurable near-ir metasurface based on ge 2 sb 2 te 5 phase-change material,” Opt. Mater. Express. 8(8), 2264–2275 (2018). [CrossRef]
15. L. Ouyang, W. Wang, D. Rosenmann, D. A. Czaplewski, J. Gao, and X. Yang, “Near-infrared chiral plasmonic metasurface absorbers,” Opt. Express. 26(24), 31484–31489 (2018). [CrossRef]
16. P. Zhou, G. G. Zheng, F. L. Xian, and L. H. Xu, “Dynamically switchable polarization-independent and broadband metasurface perfect absorber in the visible and near-infrared spectra regime,” Results Phys. 11, 278–282 (2018). [CrossRef]
17. J. K. Li, X. F. Chen, Z. Yi, H. Yang, Y. J. Tang, Y. Yi, W. T. Yao, J. Q. Wang, and Y. G. Yi, “Broadband solar energy absorber based on monolayer molybdenum disulfide using tungsten elliptical arrays,” Mater. Today Energy. 16, 100390 (2020). [CrossRef]
18. K. Sun, C. A. Riedel, A. Urbani, M. Simeoni, S. Mengali, M. Zalkovskij, B. Bilenberg, C. H. de Groot, and O. L. Muskens, “Vo2 thermochromic metamaterial-based smart optical solar reflector,” Acs Photonics 5(6), 2280–2286 (2018). [CrossRef]
19. F. Ding, J. Dai, Y. Chen, J. Zhu, Y. Jin, and S. I. Bozhevolnyi, “Broadband near-infrared metamaterial absorbers utilizing highly lossy metals,” Sci. Rep. 6(1), 39445 (2016). [CrossRef]
20. D. Ji, H. Song, X. Zeng, H. Hu, K. Liu, N. Zhang, and Q. Gan, “Broadband absorption engineering of hyperbolic metafilm patterns,” Sci. Rep. 4(1), 4498 (2015). [CrossRef]
21. M. Chirumamilla, A. Chirumamilla, Y. Yang, A. S. Roberts, P. K. Kristensen, K. Chaudhuri, A. Boltasseva, D. S. Sutherland, S. I. Bozhevolnyi, and K. Pedersen, “Large-area ultrabroadband absorber for solar thermophotovoltaics based on 3d titanium nitride nanopillars,” Adv. Opt. Mater. 5(22), 1700552 (2017). [CrossRef]
22. J. Tian, H. Luo, Q. Li, X. Pei, K. Du, and M. Qiu, “Near-infrared super-absorbing all-dielectric metasurface based on single-layer germanium nanostructures,” Laser Photonics Rev. 12(9), 1800076 (2018). [CrossRef]
23. S. Wu, Y. Ye, Z. Jiang, T. Yang, and L. Chen, “Large-area, ultrathin metasurface exhibiting strong unpolarized ultrabroadband absorption,” Adv. Opt. Mater. 7(24), 1901162 (2019). [CrossRef]
24. A. Karvounis, B. Gholipour, K. F. MacDonald, and N. I. Zheludev, “All-dielectric phase-change reconfigurable metasurface,” Appl. Phys. Lett. 109(5), 051103 (2016). [CrossRef]
25. J. Sautter, I. Staude, M. Decker, E. Rusak, D. N. Neshev, I. Brener, and Y. S. Kivshar, “Active tuning of all-dielectric metasurfaces,” ACS Nano 9(4), 4308–4315 (2015). [CrossRef]
26. P. Moitra, B. A. Slovick, Z. G. Yu, S. Krishnamurthy, and J. Valentine, “Experimental demonstration of a broadband all-dielectric metamaterial perfect reflector,” Appl. Phys. Lett. 104(17), 171102 (2014). [CrossRef]
27. Y. Yao and W. Wu, “All-dielectric heterogeneous metasurface as an efficient ultra-broadband reflector,” Adv. Opt. Mater. 5(14), 1700090 (2017). [CrossRef]
28. S. Liu, M. B. Sinclair, S. Saravi, G. A. Keeler, Y. Yang, J. Reno, G. M. Peake, F. Setzpfandt, I. Staude, and T. Pertsch, “Resonantly enhanced second-harmonic generation using iii–v semiconductor all-dielectric metasurfaces,” Nano Lett. 16(9), 5426–5432 (2016). [CrossRef]
29. S. Liu, G. A. Keeler, J. L. Reno, M. B. Sinclair, and I. Brener, “Iii–v semiconductor nanoresonators—a new strategy for passive, active, and nonlinear all-dielectric metamaterials,” Adv. Opt. Mater. 4(10), 1457–1462 (2016). [CrossRef]
30. H. Kim, a. C. Gilmore, A. Pique, J. Horwitz, H. Mattoussi, H. Murata, Z. Kafafi, and D. Chrisey, “Electrical, optical, and structural properties of indium–tin–oxide thin films for organic light-emitting devices,” J. Appl. Phys. 86(11), 6451–6461 (1999). [CrossRef]
31. X. Y. Huang, S. Y. Han, W. Huang, and X. G. Liu, “Enhancing solar cell efficiency: The search for luminescent materials as spectral converters,” Chem. Soc. Rev. 42(1), 173–201 (2013). [CrossRef]
32. Y. Chen, “Preparation of tat tri-layer insulating film on automotive glass using magnetron sputtering method,” IOP Conf. Ser. Mater. Sci. Eng. 733(1), 012040 (2020). [CrossRef]
33. T. Ariosto, A. M. Memari, and R. L. Solnosky, “Development of designer aids for energy efficient residential window retrofit solutions,” Sustain. Energy Techn. 33, 1–13 (2019). [CrossRef]
34. Q. Xue, R. Xia, C. J. Brabec, and H.-L. Yip, “Recent advances in semi-transparent polymer and perovskite solar cells for power generating window applications,” Energy Environ. Sci. 11(7), 1688–1709 (2018). [CrossRef]
35. B. Slovick, Z. G. Yu, M. Berding, and S. Krishnamurthy, “Perfect dielectric-metamaterial reflector,” Phys. Rev. B. 88(16), 165116 (2013). [CrossRef]
36. Q. Zhang, C. P. Liu, G. W. Gan, and X. D. Cui, “Visible perfect reflectors realized with all-dielectric metasurface,” Opt. Commun. 402, 226–230 (2017). [CrossRef]
37. P. Moitra, B. A. Slovick, W. Li, I. I. Kraychencko, D. P. Briggs, S. Krishnamurthy, and J. Valentine, “Large-scale all-dielectric metamaterial perfect reflectors,” Acs Photonics 2(6), 692–698 (2015). [CrossRef]
38. Y. Kivshar and A. Miroshnichenko, “Meta-optics with mie resonances,” Opt Photonics News. 28(1), 24–31 (2017). [CrossRef]
39. A. E. Miroshnichenko, A. B. Evlyukhin, Y. S. Kivshar, and B. N. Chichkov, “Substrate-induced resonant magnetoelectric effects for dielectric nanoparticles,” Acs Photonics 2(10), 1423–1428 (2015). [CrossRef]
40. L. Miao, P. Jin, K. Kaneko, A. Terai, N. Nabatova-Gabain, and S. Tanemura, “Preparation and characterization of polycrystalline anatase and rutile tio2 thin films by rf magnetron sputtering,” Appl. Surf. Sci. 212-213, 255–263 (2003). [CrossRef]
41. A. Sytchkova, D. Zola, L. R. Bailey, B. Mackenzie, G. Proudfoot, M. Tian, and A. Ulyashin, “Depth dependent properties of ito thin films grown by pulsed dc sputtering,” Mater. Sci. Eng. B. 178(9), 586–592 (2013). [CrossRef]
42. T. Lepetit, E. Akmansoy, and J. P. Ganne, “Experimental measurement of negative index in an all-dielectric metamaterial,” Appl. Phys. Lett. 95(12), 121101 (2009). [CrossRef]
43. J. Kim and A. Gopinath, “Simulation of a metamaterial containing cubic high dielectric resonators,” Phys. Rev. B. 76(11), 115126 (2007). [CrossRef]
44. S. Q. Wang and H. W. Lim, Principles and practice of photoprotection (Springer, 2016).
45. M. Sakamoto, T. Kawawaki, M. Kimura, T. Yoshinaga, J. J. M. Vequizo, H. Matsunaga, C. S. K. Ranasinghe, A. Yamakata, H. Matsuzaki, and A. Furube, “Clear and transparent nanocrystals for infrared-responsive carrier transfer,” Nat. Commun. 10(1), 1–7 (2019). [CrossRef]
