Open Access
Add to BibSonomy
Abstract
A type of infrared (IR) broadband metamaterial absorbers with several typical features, such as a relatively strong IR radiation absorption, an ultrathin functioned architecture, and a layered configuration of graphene-Si3N4 films, are proposed. The principal metamaterial absorbers are constructed by alternating configuring the core film consisting of a graphene sheet and a Si3N4 dielectric layer, which are also back attached by a thin Cu substrate. The layered graphene-Si3N4/Cu architecture already experimentally achieves an average IR absorption of ∼86.71% in a wide wavelength range of 1-14 µm according to simulations and actual measurements. And their IR absorbance can also be easily adjusted by suitably matching the layer number of the graphene-Si3N4 film or applying a set of bias voltages over the functioned graphene sheets constructed. The developed metamaterial absorbers can tolerate a relatively wide beam incident angle range of 0°-80°. The improved IR absorbing architecture also presents a polarization-independent character. Finally, it is given that the addition of various superstructures to the incident surface of the structure guarantees a very high average absorption of the structure while changing the absorption peak of the structure. It is discussed that this design can be used as a substrate for different super-surface structures to ensure their very high absorption in the infrared band. The proposed IR absorbing approach will highlight the continuous development of the IR radiation absorption technique in a broad application field.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
At present, metamaterial absorbers with a very strong infrared (IR) radiation absorbing efficiency already exhibit diverse applications such as photovoltaic or photothermal energy generation [1–4], desalination [5], thermal emitters [6], modulators [7], IR detector arrays [8,9], and concealment [10]. However, most functioned metamaterials based on basic metallic materials, usually demonstrate a relatively narrow absorption bandwidth and non-adjustable and still insufficient absorption performances, so as to greatly limit their application [11]. To realize an actively adjusting absorption bandwidth and efficiency, some novel functioned materials are urgently required [12]. As indicated, current graphene materials in a single crystal or polycrystalline have presented a great potential in photoelectric applications [13–16], due to their unique electronic and optical characteristics [17,18]. In general, common graphene as a type of planar material is composed of carbon atoms, which are closely arranged in a sp2 hybridized manner according to a honeycomb-like fashion [19]. Usually, a monolayer of intrinsic graphene will reveal a broadband light absorption of only 2.3% in both the visible and near-infrared regions [20], which means that a significant absorption enhancement should be performed for achieving a strong light energy collection.
As demonstrated, a relatively strong light absorption in a single patterned sheet of doped graphene [21,22], or in a monolayer graphene sandwiched in a Fabry-Perot (FP) microcavity [23] at a single wavelength of ∼1.1 µm, were predicted, which mainly refer to a relatively intense near-infrared absorption through a graphene-based hyperbolic metamaterial [24]. However, the above approaches are normally limited into a single polarization, and then the device fabrication remains a challenge. The proposed graphene hyperbolic metamaterials, for example, consisting of alternating graphene sheet and dielectric layer, traditionally require a sophisticated and time-consuming chemical vapour deposition [25]. So far, there have been no experimental demonstration about a relatively strong radiation absorption in a wide IR wavelength range by graphene-based metamaterials, such as simultaneously covering two or three typical atmosphere windows.
In this paper, a novel graphene-based light collection strategy for achieving an ultrabroadband IR radiation absorption according to the core n-layered graphene-Si3N4 films, is proposed. Four sets of controlled simulations are performed to illustrate a remarkable IR radiation absorption effect closely related with several factors including the number of the layered graphene-Si3N4 film, an addition of a top SiO2 coating with a suitable thickness, the stacking number of single crystal graphene utilized to form a graphene sheet used in each graphene-Si3N4 film, and the graphene chemical potential obviously determining the IR absorption rate. We found that in the 1-10 µm wavelength region, the constructed graphene-Si3N4/Cu absorbers will present a similar number of the absorption peak as the functioned graphene-Si3N4 film. A type of the developed metamaterial absorber already experimentally presents an average radiation absorption of ∼87.61% in the wavelength range of 1-14 µm. Furthermore, the constructed metamaterial absorbers can tolerate a relatively wide IR beam incident angle range from ∼2° to ∼80°. And the improved IR absorbers also show an obvious polarization-independent feature. By varying the number of the single-crystal graphene layer from 0 to 50 in the simulations, it can be found that a high absorption of the graphene architecture mainly presents in the 1-8 µm band, whereas the compound of SiO2 and Si3N4 as a kind of superstructures mainly exhibits in the 8-14 µm band. In addition, it can be expected that the IR absorption rate of the functioned structures above can be easily modulated by only changing the chemical potential of graphene materials as indicated in simulations. It should be noted that the radiation absorbing architectures will still present a similar IR energy collection performance in the similar wavelength range through executing some easy configuration variance for particular applications, for instance, thermal emitters, IR imaging sensors, or photodetectors, due to their arrayed radiation sensitive architecture and absorption mechanism.
2. Layered configurating functioned films and characteristics
2.1 n-layered graphene-Si3N4/Cu absorbers
Generally, the key IR absorbing element designed by us is a kind of functioned metamaterial, which is constructed by periodically alternating a nanometer-thick graphene sheet with a 30 stacking of the single crystal graphene and a Si3N4 insulator film of 200 nm thickness. As demonstrated, graphene materials present a very high electrical conductivity and almost no magnetic property, so as to lead to a poor impedance. Through adding an insulating film between the graphene sheets, the overall impedance of the layered architecture can be increased. Considering the factor that the radiation absorption capacity can be enhanced through increasing the number of the graphene sheet, a graphene sheet sandwiched between Si3N4 films is considered in a layered film configuration. As known, a layered periodic film configuration will also increase the chance of the beam interference extinction, and thus remarkably restrain the reflection of electromagnetic radiation between the medium layers. Because of the Si3N4 insulator film’s thickness being far less than the incident radiation wavelength, an effective coupling of the surface plasmons in the dielectric nanocavity can be realized. However, if the dielectric film thickness is too small, a significant amount of reflection will still occur. So, the lossy dielectric layers including SiO2 [26] and Si3N4 [27,28] have still an imaginary part of the refractive index in the target spectral band, which means a non-neglectable incident radiation absorption.
Although the absorption efficiency is better when using SiO2 as an insulator layer, the Si3N4 materials can also be used as an oxide protection layer for graphene, because growing a SiO2 film over a graphene sheet by PECVD maybe oxidize graphene materials. Currently, the metal backplane model [29] is used to explore the electromagnetic radiation absorption properties. Lightbeams incident upon the surface of an electromagnetic absorbing material with a certain thickness, will be gradually attenuated when entering the interior of the absorbing material and then reach the surface of the metal backplate after passing through the absorbing material, where the beam reflection will occur over the metal surface mentioned above and then re-entering into the absorbing material again. Therefore, the reflected beams are continuously attenuated, so as to result in a secondary loss within the absorbing material. What’s more, the above beam reflection of the metal end plane will also increase the chance of a quarter-wavelength interference extinction.
According to an obvious beam reflection effect, a thin metal substrate is utilized to further increase the electromagnetic wavebeam loss in the film structure constructed. A type of n-layered graphene-Si3N4/Cu absorber developed by us is presented in Fig. 1. As demonstrated, the bottom metallic layer’s thickness is considerably greater than the skin depth to prevent beam transmission [30]. Currently, we have many options for the material selection of the bottom metal substrate such as Au, Ag, Al, Cu, and so on. Based on the initial material configuration and theory constraint, the core film system designing for effectively shaping the n-layered graphene-Si3N4/Cu absorber has been conducted, where the brown Cu substrate thickness h 1= 0.1 cm, and the purple Si3N4 film thickness h 2= 200 nm, and the grey graphene sheet formed by 30 stacking number of the single crystal graphene, as shown in Fig. 1(a). The photographs of a 1-layered graphene-Si3N4/Cu absorber sample and a 2-layered graphene-Si3N4/Cu absorber sample are given in Fig. 1(b), respectively.
Fig. 1. Functioned film configuration schematic of the n-layered graphene-Si3N4/Cu absorber developed by us. (a) Layered configuration of the n-layered graphene-Si3N4/Cu architecture where the brown Cu substrate thickness h 1= 0.1 cm and the purple Si3N4 film thickness h 2= 200 nm and the stacking number of the grey single crystal graphene being 30. (b) Photographs of a 1-layered graphene-Si3N4/Cu absorber sample and a 2-layered graphene-Si3N4/Cu absorber sample fabricated.
As demonstrated, the incident radiation absorption can be commonly expressed as [29]:
Where A(ω) and R(ω) and T(ω) are absorption and reflection and transmission coefficients, and ω is angular frequency. When the effective impedance of the absorber matches free space, reflection reaches a minimum. At the same time, if the thickness of the bottom metal ground plane is greater than the target incident wave’s penetration depth, almost zero transmission will be caused. So, perfect absorption can then be achieved. The formula for the absorption rate can be expressed as:Parameters Z 0 (377Ω) and Z in are the effective impedance of free space and the n-layered graphene-Si3N4/Cu structure developed, respectively. The above relation shows that Z in = Z 0 is a critical condition for achieving a perfect radiation absorption. In addition, to efficiently match the impedance to free space, both the electric and magnetic resonances are of equal importance. The input impedance Z in and the reflection coefficient R(ω) for a double-layer absorbing material can be obtained by [31–33]:
The input impedance Z in(2) in the above equation can be obtained by relations of :
And the reflection coefficient R(ω) can also be calculated as:
The common optical properties of a monolayer graphene can be modeled using a thickness Δ according to the function in Vakil and Engheta [34]. The conductivity of graphene materials including both the interband and intraband (Drude-like) contributions [35,36] can be obtained as the relations of :
Here, ε(ω) is the dielectric constant of monolayer graphene materials, and σS the monolayer graphene conductivity, and ε 0 the vacuum permittivity, and e the electron charge, and k B the Boltzmann’s constant, and ħ the reduced Planck’s constant, and T the temperature, and µ C the chemical potential of the graphene materials used, and τ the electron relaxation time.
In this work, several related parameters are calculated using a chemical potential µ C = 0.094 eV, and the electron relaxation time τ = 10−13 s, and the temperature T = 300 K, and a single graphene thickness Δ= 0.34 nm. Strictly, the surface conductivity model specified by Eq. (8) is valid only to single crystal graphene. However, the model can also be used to roughly represent multiple layer graphene mixture by scaling the total conductivity according to the layer number. The stacking number of the single crystal graphene for shaping a single graphene sheet used in this section is 30. The thin Cu substrate is assumed to be thick enough to sufficiently ensure the incident lightbeams can not be transmitted through the CU substrate. The basic permittivity of the Cu substrate is calculated by the Drude model [37]. The dielectric property of the top SiO2 coating is directly taken from Palik [18], and that of Si3N4 film from Kischkat [39].
The simulations about the absorption spectrum of the proposed samples including a n-layered graphene-Si3N4 films indicated by brown, and a n-layered graphene-Si3N4/Cu absorber indicated by blue, and a SiO2/n-layered graphene-Si3N4/Cu absorber indicated by green, are shown in Fig. 2. During modeling and simulations, the parameter n is selected as 1, 2, 3, 4, 6, and 8, respectively. As demonstrated, the brown absorption spectrum of the n-layered graphene-Si3N4 films present a relatively gentle shifting trend as wavelength increasing in the spectral region selected. And their average value is varied from ∼0.4 in1-layered graphene-Si3N4film to ∼0.66 in 8-layered graphene-Si3N4films, as gradually increasing the layer number of the graphene-Si3N4 film. But, the blue absorption spectrum of the n-layered graphene-Si3N4 /Cu films presents a remarkable variance trend as the wavelength increasing, which is obviously different with that of the n-layered graphene-Si3N4films without any Cu film. The average absorption rate of the n-layered graphene-Si3N4/Cu films exhibits an apparent variance from ∼0.07 in1-layered graphene-Si3N4/Cu, which is far less than the corresponding value of 0.4, is rapidly increased to approach an average value 0.5 of the 3-layered graphene-Si3N4/Cu, and then exceed the average value 0.6 of the 4-layered graphene-Si3N4 /Cu, and thus almost sustain a relatively stable average value 0.8 corresponding to the graphene-Si3N4 /Cu samples with more than 4 layers of graphene-Si3N4 film. It should be noted that the green absorption spectrum of the SiO2/n-layered graphene-Si3N4/Cu samples firstly present a completely different trend as the wavelength increasing, when the layer number of the graphene-Si3N4 film is only 1 or 2, and then roughly exhibit a similar variance trend and an average absorption rate corresponding to the graphene-Si3N4 /Cu samples with more than 3 graphene-Si3N4 films. In general, the absorption spectrum of both the n-layered graphene-Si3N4/Cu sample and the SiO2/n-layered graphene-Si3N4/Cu sample are seemingly guided by the n-layered graphene-Si3N4 films, which maybe imply a core role of the n-layered graphene-Si3N4 films.
Fig. 2. Simulated absorption spectra corresponding to the n-layered graphene-Si3N4 films and the n-layered graphene-Si3N4/Cu absorber and the SiO2/n-layered graphene-Si3N4/Cu absorber, respectively. (a) to (d) Absorption spectra where parameter n is set from 1 to 4. (e) and (f) Absorption spectrum when parameter n is 6 and 8, respectively.
As shown in Fig. 2(a), there is a significant light absorbing enhancement at the central wavelength of ∼10 µm in the spectral range of 1-14 µm, which can also be viewed as a symmetric Fano-typed oscillation, when a thin Cu backplate and a SiO2 coating are closely attached onto the top and the bottom of the 1-layered graphene-Si3N4 film, respectively. And a FP-typed oscillation with three small peaks can also be found in the left near-infrared region ended at the wavelength of less than ∼4 µm. The maximum height of the absorption peak is already more than a value of ∼0.8. The above absorption spectral character can be attributed to a nano-cavity architecture of the graphene-Si3N4 film sandwiched between the top SiO2 coating and the bottom thin Cu backplate. As shown in Fig. 2(b), the Fano-typed oscillation is obviously weakened in the long wavelength region, and two FP-typed oscillations in the near-infrared region mentioned above is greatly enhanced with a similar peak value of ∼0.8 form less than 0.2 of the 1-layered graphene-Si3N4/Cu sample corresponding to that of Fano-typed oscillation. It can be seen that the left FP peak of the 1-layered graphene-Si3N4/Cu sample has been transformed into the first Fano peak of the 2-layered graphene-Si3N4/Cu sample, and two peak positions are slightly moved towards the long wavelength end. After continuously mounting a new film of graphene-Si3N4 onto the 2-layered graphene-Si3N4/Cu sample, three absorbing peaks with a slightly increased value of 0.97 in the near-infrared region can be viewed. Where two left peaks are obviously Fano-typed mode, and the new generated peak can be attributed to a common FP-typed mode. And the radiation absorption demonstrates a relatively slow variance trend in the long wavelength region, due to a secondary absorption by both graphene sheet and the Si3N4 film and further a quarter-wavelength interference phase extinction.
According to Figs. 2(c) to 2(d), it can be seen that the number of the Fano-typed peak with a similar and relatively stable peak value of 0.97 in the near-infrared region above is gradually increased as the increase of the graphene-Si3N4 film number in the samples, for instance, two Fano-typed peaks adding a FP-typed peak of the 2-layered graphene-Si3N4/Cu, three Fano-typed peaks adding a FP-typed peak of the 3-layered graphene-Si3N4/Cu, five Fano-typed peaks adding a FP-typed peak of the 6-layered graphene-Si3N4/Cu, and seven Fano-typed peaks adding a FP-typed peak of the 8-layered graphene-Si3N4/Cu.
Based on the simulation model above, the average radiation absorption rate of different sample without any Cu attachment in the wavelength range of 1-14 µm are displayed in table. 1. In terms of geometrical optics, lightbeams propagating in the medium with a periodically shifting dielectric constant, can be easily modulated according to the periodically configurated dielectric constants, so as to create a photonic band gap and thus result in an increase in the radiation absorption rate of the n-layered graphene-Si3N4 architecture. As shown in this table, the average absorbance of the graphene-Si3N4 film will undergo an initial rapid enhancement and then into a relatively slow increase district, for example, from initial ∼5.81% and ∼30.58% rapidly increasing to ∼30.31% and ∼56.77%, and then relatively slow increasing to ∼76.66% and ∼86.68% under the condition of without or with a top SiO2 coating of 1µm thickness. Considering the situation that the layer number of the graphene-Si3N4 film being more than 6 as shown in the table, their radiation absorptivity presents a very slight or even non-variance trend, a maximum layer number of less than value 6 is a suitable reference parameter.
Table 1. Average radiation absorbance in the wavelength range of 1-14 µm
In Simulations, when NGL (graphene-Si3N4 film layer) = 1, which means only having 1-layered graphene-Si3N4 film, the resonant wavelength of the 1-layered graphene-Si3N4/Cu sample is ∼2.244 µm, while the absorptivity is ∼24.04%. As demonstrated in [39], the absorption bandwidth can be remarkably increased by using a multilayer structure which supports several resonant modes closely positioned in the absorption spectrum. Moreover, the resonant frequency corresponding to an obvious radiation absorption caused by the magnetic polariton is primarily determined by the physical parameters of graphene. As NGL being increased to 2, two local resonances with a relatively large absorption up to ∼84.34% can be clearly observed in the near-infrared region. Owing to these resonant peaks, we can define a relatively wide absorption frequency band. It can be expected that by overlapping more graphene-Si3N4 films, more absorption peaks and a wider absorption band can be acquired. Accordingly, four resonances can be observed at λ 1= 1.151 µm, λ 2= 1.577 µm, λ 3= 2.574 µm, and λ 4= 6.306 µm, corresponding to the radiation absorption of 99.73%, 99.97%, 99.94%, and 99.14%, respectively. It should be noted that a perfect absorption already occurs at the above four absorption peaks, and thus the average absorption rate can be greatly improved. This indicates that a multilayer configuration of the graphene-Si3N4 film has presented a perfect matching with free space impedance, and thus resulting in the reflectivity approaching 0, which also a consequence of the multiple interference extinction in the layered structure constructed.
Because practically constructing a highly periodic film architecture is very difficult and then the absorption rate improvement also weakened after the parameter n being more than 3, only three samples with different n configuration from 1 to 3 are made, respectively. The optical measurements using a common Fourier Transform Infrared Spectroscopy (FTIRS, Nicolet 8700/American Thermoelectric Corporation) are given in Fig. 3. When the effective impedance of the samples is fine matched with that of free space, the beam reflection will approach a minimum value. In experiments, when NGL = 1, 2, 3, the average absorption in the wavelength range of 1-14 µm is 13.02%, 53.43%, and 78.1%, which are obviously higher than the simulation value of 5.81%, 30.31%, and 52.77%, respectively. This phenomenon maybe results from an incomplete removal of PMMA from graphene structures, which will lead to an apparent decrease of the electrical conductivity of graphene materials, and therefore increase their dielectric loss capacity and the absorption capacity. What’s more, since the graphene sheets used in the architecture are prepared through thirty transferring and stacking of the single crystal graphene utilized, so as to lead to an appearance of wrinkles in some districts of a sheet formed, and also the appearance of very small air gaps between adjacent single crystal graphene during stacking them.
Fig. 3. Measured absorption spectra of the n-layered graphene-Si3N4/Cu absorber, where n is 1, 2, and 3, respectively.
In addition, the Si3N4 film grown by PECVD will demonstrate a slight difference in refractive index with that in the simulations. To avoid being oxidized, the graphene sheet unable to be exposed in air at high temperature, and thus the Si3N4 films are grown at 100 degrees Celsius by PECVD, so as to result in less dense films [40]. The phenomenon presenting different reflectance and absorptance can also be explained by Eq. (6). So, the measured curves of the samples including the 1-layered graphene-Si3N4/Cu absorber, the 2-layered graphene-Si3N4/Cu absorber, and the 3-layered graphene-Si3N4/Cu absorber, demonstrate a relatively flat variance trend, which is obviously different with the simulations. But the increasing trend of the radiation absorption with the layer number increasing of the graphene-Si3N4 film in the wavelength range of 1-14 µm, is the same as the simulation prediction. As shown, the absorptivity is already increased from a small average value of ∼0.15 (1-layered graphene-Si3N4/Cu sample) to an intermediate average value of 0.58 (2-layered graphene-Si3N4/Cu sample) to a relatively large average value of 0.82 (3-layered graphene-Si3N4/Cu sample).
The phenomenon with a relatively high absorption rate of the measured samples can be explained clearly. As a beam of incident electromagnetic wave directly hits graphene, the carriers in graphene are very prone to an energy leap to produce photogenerated carriers. So, a partial incident light energy is used to stimulate the carrier leap in graphene, which thus leads to an obvious absorption of the incident radiation. And then a relatively weak local inductive electron oscillation can be formed over the illuminated structural surface, and continuously diffused outside, so as to shape a surface diffusion vibration or electron density wave or a kind of outward micro-current field, which means that the incident electromagnetic energy is already dissipated by converting it into internal energy according to a conductive loss mechanism. In addition, the basic lossy dielectric layer composed of a Si3N4 film [27,28] and a graphene sheet has an imaginary part of the refractive index in the target spectral band, which is also capable of absorbing a portion of the incident radiation. The reflective lightwaves from the interfaces between adjacent Si3N4 film and graphene sheet are able to interfere with incident lightwaves and then generates phase cancel if the phase difference being 180°, so as to increase the chance of interference phase extinction, due to the multilayer membrane configuration.
2.2 SiO2/n-layered graphene-Si3N4/Cu absorber
In this section, a new design for realizing a near-perfect broadband radiation absorption in the wavelength range of 1-14 µm is outlined. The key operation is to adjust the component of the dielectric layers currently focusing on the top coating. Here, a proposed construction of the top dielectric layer is outfrom two common materials. As shown, the lossy dielectric materials such as SiO2 have an obvious imaginary part of the refractive index for fully absorbing incident lightwaves, and therefore exhibit a higher radiation absorption than that of Si3N4 materials in the 1-14 µm wavelength region. So, a certain thickness of SiO2 coating is continuously mounted on the top of the n-layered graphene-Si3N4/Cu architecture, where the SiO2 coating is already optimized according to the impedance matching model given above. Based on the initial insulator-graphene-metal configuration, a type of the SiO2/n-layered graphene-Si3N4/Cu absorber is further constructed. The main architecture schematic of the absorber is illustrated in Fig. 4, where the brown Cu substrate thickness h 1= 0.1 cm and the purple Si3N4 film thickness h 2= 200 nm and the top SiO2 coating thickness h 3= 1000 nm, and the stacking number of the grey single crystal graphene being still 30. The chemical potential of graphene is taken as µ C = 0.094 eV. The experimentally measured and simulated absorption spectra of the SiO2/3-layered graphene-Si3N4 /Cu absorber sample are shown in Fig. 4.
Fig. 4. Experimental measurement and simulation absorption spectra of the SiO2/3-layered graphene-Si3N4/Cu absorber sample indicated by brown and blue, respectively.
From the simulations and experimental measurements, it can be found that the simulation spectrum of the sample also demonstrates two Fano-typed and a FP-typed peaks in the near-infrared wavelength region and a similar variance trend in the long wavelength range compared with that of the sample without any top SiO2 coating. And the measuremental spectrum also exhibits a similar varying character with the sample indicated by green shown in Fig. 4. It can be expected that an intrinsic radiation absorption characteristic of SiO2 materials plays a vital role. In general, the sample already achieves an average absorption rate of ∼83% in the 1-14 µm wavelength range. To such absorbers, the imaginary-part refractive index of SiO2 coating usually governs the excitation of the intrinsic absorption of incident lightwaves in the long wavelength region, and thus the lightwave absorption action can also be coupled with typical plasmon resonances between the lossy dielectric films with a typical nanometer-scale thickness being far less than the incident IR wavelength, so as to jointly enhance a broadband light energy absorption. So, the increase of the absorption peak can be attributed to an amount increasing of the Graphene-Si3N4 film, which directly lead to an enhancement of plasma resonances among the Graphene-Si3N4 films. According to experimental measurements, the tested absorption rates are generally higher than that of simulations, and then a slightly red-shifted movement of the absorption peaks can be viewed apparently, which maybe due to impurities distributed over the interfaces during sample fabrication process and experimental measurements. Considering the manufacturing cost and complexity and the needed IR radiation absorption level, for instance, an average absorption of ∼87.61% being acquired by a 3-layered graphene-Si3N4/Cu absorber as tested by FTIR spectroscopy, it is suggested that an optimal layer number of the graphene-Si3N4 film is determined as 4.
As known, the common optical characteristics closely related with an obvious plasmonic action based on the typical FP-typed resonator with a special patterned layout or architecture, are highly sensitive to the geometry dimensions of the functioned micro-nano-structure [38]. So, the polarization-insensitive absorption property is mainly attributed to the axisymmetric geometry of structure, whose edge lengths in both x- and y-axis directions are finite and equal. Therefore, for both TM and TE incidence, the electric dipoles formed by the accumulation of charges with opposite sign oscillate in the same way. Besides, they are greatly coupled with their own images, which oscillate in antiphase on the metallic film. Consequently, magnetic polaritons [41,42] are formed, which will induce a strong magnetic response and then cause a resonant dip in the same reflection spectrum. Thus, the proposed 3D structure has a polarizations-insensitive property, finally.
In order to further investigate the absorption effect of the SiO2/4-layered graphene-Si3N4 /Cu absorber under the condition of oblique incidence, both the broadband and narrowband absorption spectra are simulated in a wide incidence angle range of 0°-80°. The chemical potential of graphene is taken as µ C = 0.094 eV. The IR radiation absorption spectra according to IR radiation incident angle from 0° to 80° in steps 2° are shown in Fig. 5. Where a type of TE-mode (x-polarized) absorption spectra is given in Fig. 5(a), and the TM-mode (y-polarized) absorption spectra in Fig. 5(b). Two typical absorption spectra corresponding to lightbeam oblique incidence for TE-polarized lightwaves at 0° and TM-polarized lightwaves at 90°, can be seen in Fig. 5. Corresponding to small lightbeam incident angle, the proposed absorber almost exhibits an angle-independent absorption character for both TE- and TM-modes, and the absorption spectrum remains almost completely unchanged character for both the polarization directions with incident angle up to 60°.
Fig. 5. Absorption spectra according to IR radiation incident angle from 0° to 80° in steps 2°. (a) TE-mode (x-polarized) absorption spectra. (b) TM-mode (y-polarized) absorption spectra.
To a relatively large lightbeam incident angle, the absorber behaves a completely different manner to TE- and TM-polarized lightwaves, respectively. As the lightbeam incident angle increasing, both the average radiation absorptivity and bandwidth are gradually decreased with an obvious fluctuation. In addition, the incident magnetic field’s horizontal component will be decreased as the incident angle increasing, and also the coupling strength between adjacent graphene-Si3N4 films and then radiation absorption weakened. So, it can be expected that the proposed SiO2/4-layered graphene-Si3N4 /Cu absorber can tolerate a wide lightbeam incident angle range for both the TE- and TM-polarized lightwaves, so as to highlight the future development of the IR broadband absorber with polarization-insensitive and angle-independent feature.
2.3 Graphene chemical potential and number of single crystal graphene
As demonstrated, the light energy absorption of a single graphene sheet in common infrared band is only 2.3%, which can be remarkably increased by continuously stacking graphene [43,44]. Strictly speaking, the surface conductivity model specified by Eq. (5) is still valid only for a single graphene. However, the model can also be used to roughly represent the typical characters of the multiplayer stacking graphene materials by scaling the total conductivity simply according to the layer number selected. For this purpose, a conductivity scaling factor has been included in the parameter choice about graphene material in the FDTD. In this section, an effect about the stacking number of the single crystal graphene for shaping a single graphene sheet and their chemical potential determining the absorbance of the SiO2/4-layered graphene-Si3N4/Cu sample, is outlined.
The simulated absorption spectra of the SiO2/4-layered graphene-Si3N4 /Cu absorber using different stacking number of the single crystal graphene from 0 to 50 indicated by different color, is illustrated in Fig. 6. And the chemical potential of graphene is taken as µ C = 0.094 eV. As shown, single crystal graphene presents a weak light absorption due to its typical zero-bandgap energy band structure. Continuously stacking single crystal graphene together can effectively improve their light absorption efficiency. And then the absorption of the structure without any graphene is very low in the wavelength range of 1-7 µm. In addition, it can be seen that increasing the layer number of the stacked single crystal graphene can apparently vary their electrical conductivity and absorption efficiency. The obvious absorption of the SiO2/4-layered graphene-Si3N4/Cu structure in the 1-7 µm band depends mainly on the stacking number of the single crystal graphene, while in the 7-14 µm band depends on the intrinsic absorption of the top SiO2 coating and the Si3N4 film.
Fig. 6. Simulated absorption spectra of the SiO2/4-layered graphene-Si3N4/Cu absorber using different stacking number of single crystal graphene from 0 to 50 indicated by different color.
With the increase of the stacking number of the single crystal graphene, the number of absorption peak and their central wavelength position remains almost unchanged, but the height of absorption peaks increases continuously. The height rising of the absorption peak leads to a corresponding increase of the average absorption rate of the functioned structure. Almost coincidence of the absorption spectra as that without any graphene materials can be seen when the stacking number of the single crystal graphene indicated by blue is only 1. If the stacking number is relatively large, the conductivity Eq. (5) will not be applicable corresponding to the graphene sheet formed. Considering the difficulty of making graphene sheet with a relatively large stacking number of the single crystal and then maintenance their basic physical properties, the optimized stacking number of the single crystal graphene is less than 30.
Because the Fermi energy of the graphene materials can be modulated by electrostatic bias [45,46], the absorbance of the functioned structure based on the layered graphene materials insulated by dielectric medium can also be adjusted through applying a set of bias voltage over an objective SiO2/4-layered graphene-Si3N4 /Cu architecture. Firstly, the influence of shifting the chemical potential of graphene sheet on the absorption rate is investigated, as shown in Fig. 7. It can be seen that by applying a bias voltage over the functioned structure, the tunable range of the absorption rate in the 1-8 µm band is more than 50%. The phenomenon can be attributed to a presence of a 4-sheet graphene structure formed according to a 30 stacking of the single crystal graphene. Due to the presence of the Si3N4 insulating film between adjacent graphene sheets, the conductive tape should be applied over the graphene sheet prior to PECVD growth of the Si3N4 film for effectively applying bias voltage onto each graphene sheet. At present, the experimental approaches are still optimized because a suitable power-up method has not been found practically. However, the possibility that the structural absorptivity can be modulated in a large scale is successfully simulated.
Fig. 7. Infrared radiation absorption curves of the absorber through shifting the Fermi energy of graphene sheet in a range from 0.05 eV to 0.6 eV indicated by different color.
2.4 New designing
According to the above designing about periodically configurating functioned nano-films or nano-structures, two new patterned nano-structural layouts based on the optical antenna strategy for highly efficient absorbing incident radiation, are proposed as shown in Fig. 9. In recent years, numerous new methods have been attempted to achieve broadband absorption. Initially, the designs of the metallic layers of MPAs (field-metamaterial perfect absorbers) are limited mostly to high-Q-factor noble metals, including Au [47] and Ag [48], which already cause excellent resonant selectivity. It has been discovered that some refractory metals can excite broadband plasmon resonances, including Mo [49], W [50], and Ti [51]. Considering the wavelength range such as from the visible to mid- or even long-wave infrared, these metals present a large imaginary-part in their refractive indices, and thus exhibit a highly lossy property in a broadband absorption. According to the situation of already presenting a relatively strong broadband IR absorption performance through layered configurating graphene-Si3N4 metamaterials also attached by top and bottom insulating layers, the radiation absorption efficiency will be further improved by continuously coupling or even directly growing an array of patterned metallic micro-nano-structures leading to an arrayed optical antenna onto the top SiO2 coating, which should be suitable to the 1-14 µm infrared waveband. Compared to other metals, Ti has great potential and also exhibits better performances when used as a metallic layer of infrared broadband MPAs.
Fig. 8. Schematic diagram of the top-patterned absorber based on the basic graphene-Si3N4 film system configuration and the absorption spectrum features. The thickness of the SiO2 coating is 1 µm and the patterned Ti layer 20 nm and the Si3N4 film 0.8 µm. The 4-layered graphene-Si3N4 films are evenly constructed into a Ti/SiO2/4-layered graphene-Si3N4/Cu absorber. (a) Schematic of the structure-1 with an arrayed nano-convex-crosses-shaped resonator. In this layout designing, p = 1 µm, l 2= 0.8 µm, and l 1= 0.2 µm. (b) Schematic of the structure-2 with an arrayed nano-double-convex-rectangle-shaped resonator. In this designing, p = 1 µm, l 3= 0.8 µm, l 2= 0.6 µm, and l 1= 0.4 µm. (c) Simulated absorption spectra of the SiO2/n-layered graphene-Si3N4/Cu and both the structure-1 and -2 indicated by green and brown and blue, respectively. Their detailed absorption spectra in the 1-3 µm waveband are also inserted in the figure.
Firstly, by the traditional surface equipartition excitation theory, a scheme of directly mounting a certain thickness of patterned Ti metal mask consisting of an arrayed surface wave resonator onto the SiO2 coating, such as outfrom typical nano-convex-cross or nano-double-convex-rectangle with several key concave-tips and convex-tips, is given in Fig. 8(a). After conducting FDTD simulations with initial structural parameters selected empirically, the most suitable geometry parameters of each graphical unit are obtained. In this layout designing, the key parameters of p = 1 µm and l 1 = 0.2 µm and l 2 = 0.8 µm. The thickness of the SiO2 coating is 1 µm, and the top Ti cap or a patterned mask 20 nm, and the Si3N4 film 0.8 µm. The core 4-layered graphene-Si3N4 films are evenly constructed into a type of Ti/SiO2/4-layered graphene-Si3N4/Cu absorber.
Another design is also achieved by etching the top SiO2 coating in the SiO2/4-layered graphene-Si3N4/Cu architecture. The absorptivity of initial patterns with different size configuration has been simulated, and thus the best is consequently selected. A schematic of the functioned structure-2 with an arrayed nano-double-convex-rectangle-shaped resonator is shown in Fig. 8(b). In this design, the key parameters of p = 1 µm and l 3 = 0.8 µm and l 2 = 0.6 µm and l 1 = 0.4 µm. The rest of the thickness configuration is the same as in the structure-1. The chemical potential of the graphene materials is taken as µ C = 0.094 eV. The simulated absorption spectra of the SiO2/4-layered graphene-Si3N4/Cu and both the structure-1 and -2 indicated by green and brown and blue, respectively, are shown in Fig. 9(c).
From the simulations, it can be seen that the position and the number of the absorption peaks already present an obviously shifting trend corresponding to the variance of the surface plasma mode, which practically response the spectral oscillation behaviors in the wavelength range of 1-14 µm. This phenomenon means that the absorption spectrum of the functioned structure can be adjusted somewhat by changing the appearance of the SiO2 coating continuously attached by a patterned cap or mask, which can also be viewed as a typical optical antenna array for highly efficient absorbing incident radiation, as shown in Fig. 8(c). This structure utilizes an optical metasurface to generate a highly localized enhancement of incident lightwave collected mainly at the tip of the surface Ti structures. Generally, the excited surface plasmon polaritons (SPPs), as a highly localized electromagnetic mode, exist near the insulator’s top metallic structure and then gradually diffuse into the entire graphene-Si3N4. So, this arrayed optical antenna allows the incident lightwaves to be more fully absorbed by the crystal lattices and valence electrons in graphene-Si3N4 layers. Due to the SPPs, we can find that the structure-1 shows two absorption peaks at ∼1.89 µm and ∼4.58 µm with a strong absorption of more than 99%, as shown in Fig. 9(c).
On the other hand, the partial etching of the SiO2 also allows some of the lightbeams to go directly into the graphene-Si3N4 layers, and then absorbed by the structure-2. The graphene-Si3N4 absorbing layers do not exhibit an obvious reflection peak at ∼9 µm, which means that the structure-2 already eliminate the reflection peaks of SiO2 at ∼9 µm, and its absorptivity is greater than 70% in the wavelength range of 2.5-14µm. The average absorption of both functioned structures is already more than 80%. So, through efficiently configurating the micro-nano-appearance of the SiO2 coating with an ultrathin metallic cap patterned for stimulating local surface resonance, it is possible to obtain a desired absorption spectrum with a stronger IR radiation absorption according to the requirements. The fabrication processing based on the new layout designing and optical parameter configuration will be carried out.
3. Discussion and conclusion
In this paper, a tunable broadband metamaterial absorber with several features including a very strong IR radiation absorption, an ultrathin architecture in several micron scale, and an easy film system configuration, is proposed. Firstly, a type of n-layered graphene-Si3N4 /Cu absorber sample is demonstrated, which can be used to achieve an average IR radiation absorption of more than 75% in simulations and ∼80% in actual measurements. We find that the number and the height of the absorption peak according to the spectral absorption simulations will be obviously increased as gradually increasing the layer number of the core graphene-Si3N4 films. And the average radiation absorption in the IR wavelength range of 1-14 µm can be effectively enhanced by at least 10% after mounting a certain thickness of SiO2 coating onto the top or IR lightwave incident surface of the graphene-Si3N4 films. When the layer number of graphene-Si3N4 film is more than 7, their average IR radiation absorption in the 1-14 µm wavelength range already reaches a relatively large value of ∼86.71% in simulations. The proposed SiO2/n-layered graphene-Si3N4/Cu absorber demonstrates an obvious polarization-independent character, and then fine radiation absorption performance over a relatively wide incidence angle range. The intrinsic absorption of SiO2 coating remarkably enhances the radiation absorption efficiency of the n-layered graphene-Si3N4/Cu structure in the 8-14 µm IR waveband, while the graphene-Si3N4 films act mainly in the 1-8 µm IR band.
Since the Fermi energy level of the graphene materials can be easily varied by applying a set of bias voltage onto the functioned micro-nano-structure constructed, their IR radiation absorbance performance can be adjusted in a relatively wide bias voltage range corresponding to those with multiple sheets of graphene configuration. Based on the structural model proposed in this paper, a patterned thin metal cap or mask can be coated or mounted onto the top of the n-layered graphene-Si3N4/Cu metamaterials structure for further enhancing the IR radiation absorption efficiency, so as to highlight a continuous development of the SiO2/n-layered graphene-Si3N4/Cu absorber architecture approach. According to modern micro-electronic techniques, a variety of micro-nano-patterned mask can be designed and coupled with the absorber architecture proposed for electrically adjust or tune the optical properties. This work has exhibited a great potential in a variety of applications, such as thermal emitters, infrared sensors and imaging detector arrays.
4. Materials and method
4.1 Simulations
The numerical radiation absorption spectra of the metamaterials absorbers constructed and also the electromagnetic field distributions corresponding to the IR radiation absorption peaks are theoretically simulated using common finite difference time domain (FDTD) method. A unit cell of the investigated functioned micro-nano-structure is simulated using traditional periodic boundary conditions along x- and y-axis, respectively, and then perfectly matched with a layered film materials configuration along the propagation direction of incident IR lightwaves (z-axis). A mesh accuracy of 6 is chosen and the calculation time set as 5000 fs with an automatic shutoff minimum of 1 × 10−5. The refractive indices of Cu and SiO2 are derived from Palik’s handbook [52], and the refractive index of Si3N4 derived from Kischkat [39]. The actual refractive indices of SiO2 and Si3N4 are 1.51 and 2.1, respectively. The stacking number of the single crystal graphene for shaping a graphene sheet is 30, and its chemical potential 0.094 eV. In FDTD simulations, a conductivity scaling represents the layer number of the graphene sheet. A planar lightwaves are launched from the top of the functioned structure constructed, and then the electromagnetic wavefield distributions are effectively stimulated using a TM-polarized planar lightwave. To oblique incidence, the lightbeam incident angle can be varied from 0° to 80° with a step of 2°.
4.2 Fabrication
Two types of adjusted broadband IR radiation absorbers including a type of 4-layered graphene-Si3N4/Cu micro-nano-structure and a type of SiO2/4-layered graphene-Si3N4/Cu micro-nano-structures, are successfully fabricated. The functioned micro-nano-structures with an effective area of 1 cm × 1 cm and 5 cm × 5 cm are made, respectively. The single crystal graphene is prepared by CVD (KJ-CVD-1700-60*300-F3) method, and Cu materials is used as catalyst and a basic growth substrate, and methane is used as gas to prepare monolayer graphene over Cu substrate. A PMMA layer is spin-coated on the graphene using a homogenizer to protect the graphene. The Cu substrate is put into etching solution to fully etch off Cu material from graphene. The graphene is then picked up with PET and then washed in deionized water, and a new Cu substrate graphene is picked up from the deionized water to obtain a bilayer Cu substrate graphene. Repeating the above steps several times to finally obtain the target layer number of a graphene sheet. In experiments, PMMA is completely removed from the functioned nano-structure by high-temperature annealing or acetone immersion. A typical 200 nm thickness of Si3N4 film is grown using PECVD (Oxford PlasmaPro 800 Stratum PECVD) at 100°C. The multilayer configurating graphene sheet is then transferred onto the substrate using a wet transfer method. A 1000 nm thickness of SiO2 is continuously grown onto the top of a 4-layered graphene-Si3N4 films using PECVD at 300°C. The refractive indices of SiO2 and Si3N4 are 1.46 and 1.81, respectively.
4.3 Optical characterizations
An infrared Fourier transformation spectroscope measuring system (FTIRS, Nicolet 8700/American Thermoelectric Corporation) is used for demonstrating the IR reflection spectra of samples including a 4-layered graphene-Si3N4/Cu absorber and a SiO2/4-layered graphene-Si3N4/Cu absorber. We need to thank Chengming Wang of USTC's Physical and Chemical Laboratory Center for the infrared spectroscopy testing. The measured IR radiation reflection is referenced to an Ag mirror in order to determine absolute reflectivity (R). As the transmission channel is completely cancelled by the opaque metal mirror, the radiation absorption according to A = 1-R-T is determined directly to be A = 1-R. The measured wavelength range is 1.3-14 µm actually.
Funding
National Natural Science Foundation of China (61176025).
Disclosures
The authors declare that there are no conflicts of interest related to this article.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. A. Cao, X. Zhang, C. Xu, B. Wei, and D. Wu, “Tandem structure of aligned carbon nanotubes on Au and its solar thermal absorption,” Sol. Energy Mater. Sol. Cells 70(4), 481–486 (2002). [CrossRef]
2. H. Ghasemi, George Ni, Amy Marie Marconnet, et al., “Solar steam generation by heat localization,” Nat. Commun. 5(1), 4449 (2014). [CrossRef]
3. H. Ren, Miao Tang, Baolu Guan, et al., “Hierarchical graphene foam for efficient omnidirectional solar–thermal energy conversion,” Adv. Mater. 29(38), 1702590 (2017). [CrossRef]
4. M. Zhu, Yiju Li, Guang Chen, et al., “Tree-inspired design for high-efficiency water extraction,” Adv. Mater. 29(44), 1704107 (2017). [CrossRef]
5. X. Li, et al. “Graphene oxide-based efficient and scalable solar desalination under one sun with a confined 2D water path,” Proc. Natl Acad. Sci. USA 113, 13953–13958 (2016). [CrossRef]
6. J. Huang, Changxu Liu, Yihan Zhu, et al., “Harnessing structural darkness in the visible and infrared wavelengths for a new source of light,” Nat. Nanotechnol. 11(1), 60–66 (2016). [CrossRef]
7. Y. Yao, Raji Shankar, Mikhail A. Kats, et al., “Electrically tunable metasurface perfect absorbers for ultrathin mid-infrared optical modulators,” Nat. Nanotechnol. 14(11), 6526–6532 (2014). [CrossRef]
8. W. Li and J. Valentine, “Metamaterial perfect absorber based hot electron photodetection,” Nano Lett. 14(6), 3510–3514 (2014). [CrossRef]
9. I. Mellouki, N. Bennaji, and Y. acoubi, “IR characterization of graphite black-coating for cryogenic detectors,” Infrared Phys. Technol. 50(1), 58–62 (2007). [CrossRef]
10. H. Shi, J. G. Ok, H. Won Baac, and L. Jay Guo, “Low density carbon nanotube forest as an index-matched and near perfect absorption coating,” Appl. Phys. Lett. 99(21), 211103 (2011). [CrossRef]
11. Z. Xu, D. Wu, Y. Liu, C. Liu, Z. Yu, L. Yu, and H. Ye, “Design of a tunable ultra-broadband terahertz absorber based on multiple layers of graphene ribbons,” Nanoscale Res. Lett. 13(1), 143 (2018). [CrossRef]
12. H. Xiong and F. Y. ang, “Ultra-broadband and tunable saline water-based absorber in microwave regime,” Opt. Express 28(4), 5306–5316 (2020). [CrossRef]
13. P. J. Harshman, T. K. Gustafson, and P. Kelley, “Title of paper,” J. Chem. Phys. 3, (to be published).
14. K. Gallo and G. Assanto, “All-optical diode based on second-harmonic generation in an asymmetric waveguide,” J. Opt. Soc. Am. B 16(2), 267–269 (1999). [CrossRef]
15. B. R. Masters, “Three-dimensional microscopic tomographic imagings of the cataract in a human lens in vivo,” Opt. Express 3(9), 332–338 (1998). [CrossRef]
16. D. Yelin, D. Oron, S. Thiberge, E. Moses, and Y. Silberberg, “Multiphoton plasmon-resonance microscopy,” Opt. Express 11(12), 1385–1391 (2003). [CrossRef]
17. M. Partridge, “Spectra evolution during coating,” figshare, 2014, https://doi.org/10.6084/m9.figshare.1004612.
18. E. D. Palik, Handbook of Optical Constants of Solids (Academic Press, 1998).
19. P. Avouris, Z. Chen, and V. Perebeinos, “Carbon-based electronics,” Nat. Nanotechnol. 2(10), 605–615 (2007). [CrossRef]
20. R. R. Nair, A. N. Grigorenko, K. S. Novoselov, et al., “Fine structure constant defines visual transparency of graphene,” Science 320(5881), 1308 (2008). [CrossRef]
21. A. Ferreira and N. Peres, “Complete light absorption in graphene-metamaterial corrugated structures,” Phys. Rev. B 86(20), 205401 (2012). [CrossRef]
22. S. Thongrattanasiri, F . H. Koppens, and de Abajo, “Complete optical absorption in periodically patterned graphene.,” Phys. Rev. Lett. 108(4), 047401 (2012). [CrossRef]
23. A. Ferreira, N. Peres, R. Ribeiro, and T. Stauber, “Graphene-based photodetector with two cavities,” Phys. Rev. B 85(11), 115438 (2012). [CrossRef]
24. I. S. Nefedov, C. A. Valaginnopoulos, and L. A. Melnikov, “Perfect absorption in graphene multilayers,” J. Opt. 15(11), 114003 (2013). [CrossRef]
25. Y.-C. Chang, Che-Hung Liu, Chang-Hua Liu, et al., “Realization of mid-infrared graphene hyperbolic metamaterials,” Nat. Commun. 7(1), 10568 (2016). [CrossRef]
26. C. H. Fann, Jihua Zhang, Mohamed ElKabbash, et al., “Broadband infrared plasmonic metamaterial absorber with multipronged absorption mechanisms,” Opt. Express 27(20), 27917–27926 (2019). [CrossRef]
27. I. Massiot, Nicolas Vandamme, Nathalie Bardou, et al., “Metal nanogrid for broadband multiresonant light-harvesting in ultrathin GaAs layers,” ACS Photon. 1(9), 878–884 (2014). [CrossRef]
28. S. L. Wu, Yan Ye, Zhouying Jiang, et al., “Large-area, ultrathin metasurface exhibiting strong unpolarized ultrabroadband absorption,” Adv. Opt. Mater. 7(24), 1901162 (2019). [CrossRef]
29. L. Wang, Y. Huang, C. Li, et al., “Hierarchical composites of polyaniline nanorod arrays covalently grafted on the surfaces of graphene@Fe3O4@C with high microwave absorption performance[J],” Compos. Sci. Technol. 108, 1–8 (2015). [CrossRef]
30. Yu Zhou, Zheng Qin, Zhongzhu Liang, Dejia Meng, Haiyang Xu, David R. Smith, and Yichun. Liu, “Ultrabroadband metamaterial absorbers from long to very long infrared regime[J].,” Light: Sci. Appl. 10(1), 138 (2021). [CrossRef]
31. N. I. Landy, S. Sajuyigbe, J. J. Mock, et al., “Perfect metamaterial absorber,” Phys. Rev. Lett. 100(20), 207402 (2008). [CrossRef]
32. Y. Li, D. Li, J. Yang, et al., “Enhanced microwave absorption and surface wave attenuation properties of Co0.5Ni0.5Fe2O4 fibers/reduced graphene oxide composites,” Materials 11(4), 566 (2018). [CrossRef]
33. H. Luo, R. Gong, X. Wang, et al., “Synthesis and excellent microwave absorption properties of reduced graphene oxide/FeNi3/Fe3O4 composite,” New J. Chem. 40(7), 6238–6243 (2016). [CrossRef]
34. B. Gao, L. Qiao, B. Wang J, et al., “Microwave absorption properties of the Ni nanowires composite,” J. Phys. D: Appl. Phys. 41(23), 235005 (2008). [CrossRef]
35. A. Vakil and N. Engheta, “Transformation optics using graphene,” Science 332(6035), 1291–1294 (2011). [CrossRef]
36. L. A. Falkovsky and S. S. Pershoguba, “Optical far-infrared properties of a graphene monolayer and multilayer,” Phys. Rev. B 76(15), 153410 (2007). [CrossRef]
37. L. A. Falkovsky, “Optical properties of graphene,” J. Phys.: Conf. Ser. 129, 012004 (2008). [CrossRef]
38. Yijun Cai and Kai-Da Xu, “Tunable broadband terahertz absorber based on multilayer graphene-sandwiched plasmonic structure,” Opt. Express 26(24), 31693 (2018). [CrossRef]
39. J. Kischkat and S. Peters, “Mid-infrared optical properties of thin films of aluminum oxide, titanium dioxide, silicon dioxide, aluminum nitride, and silicon nitride,” Appl Opt. 51(28), 6789 (2012). [CrossRef]
40. Z. Chongyou and C. Xianwu, “Study on SiNx thin film prepared by PECVD,” Semiconductor Optoelectronics 239, 233–235 (2011).
41. N. Liu, L. Fu, S. Kaiser, H. Schweizer, and H. Giessen, “Plasmonic building blocks for magnetic molecules in three- dimensional optical metamaterials,” Adv. Mater. 20(20), 3859–3865 (2008). [CrossRef]
42. J. Zhou, L. Zhang, G. Tuttle, T. Koschny, and C. M. Soukoulis, “Negative index materials using simple short wire pairs,” Phys. Rev. B 73(4), 041101 (2006). [CrossRef]
43. Y. Ye, Y. Jin, and S. He, “Omnidirectional, polarization-insensitive and broadband thin absorber in the terahertz regime,” J. Opt. Soc. Am. B 27(3), 498–504 (2010). [CrossRef]
44. A K. Geim, “Graphene: Status and Prospects,” Science 324(5934), 1530–1534 (2009). [CrossRef]
45. F. Bonaccorso, Z. Sun, T. Hasan, et al., “Graphene photonics and optoelectronics,” Nat. Photonics 4(9), 611–622 (2010). [CrossRef]
46. B. Vasić and R. Gajić, “Graphene induced spectral tuning of metamaterial absorbers at mid-infrared frequencies,” Appl. Phys. Lett. 103(26), 261111 (2013). [CrossRef]
47. E. Z. Hou, Dejia Meng, Zhongzhu Liang, et al., “Mid-wave and long-wave infrared dual-band stacked metamaterial absorber for broadband with high refractive index sensitivity.,” Appl. Opt. 59(9), 2695–2700 (2020). [CrossRef]
48. R. Feng, Jun Qiu, Linhua Liu, et al., “Parallel LC circuit model for multi-band absorption and preliminary design of radiative cooling.,” Opt. Express 22, A1713–A1724 (2014). [CrossRef]
49. D. Hasan, Prakash Pitchappa, Jiahui Wang, et al., “Novel CMOS-compatible Mo-AlN-Mo Platform for metamaterial-based mid-IR absorber.,” ACS Photon. 4(2), 302–315 (2017). [CrossRef]
50. Z. G. Li, Liliana Stan, David A. Czaplewski, et al., “Wavelength-selective mid-infrared metamaterial absorbers with multiple tungsten cross resonators,” Opt. Express 26(5), 5616–5631 (2018). [CrossRef]
51. P. Q. Yu, Hua Yang, Xifang Chen, et al., “Ultra-wideband solar absorber based on refractory titanium metal,” Renew. Energy 158(2 2), 227–235 (2020). [CrossRef]
52. L. P. Wang and Z. M. Zhang, “Measurement of coherent thermal emission due to magnetic polaritons in subwavelength microstructures,” J. Heat Transfer 135(9), 091505 (2013). [CrossRef]
