A Wide Incident Angle, Ultrathin, Polarization-Insensitive Metamaterial Absorber for Optical Wavelength Applications

Polarization-insensitive metamaterial absorbers (MMAs) are currently attracting a great deal of interest throughout the optical region. These kinds of MMAs can be used as energy harvesting, thermal imaging, plasmonic sensor, magnetic imaging, light trapping, optical modulator, and can help to decipher an inherence for the field. Here, a polarization-insensitive ultrathin MMA has been proposed with an extensive incident angular stability for the optical region with a sandwiched three-layer structure. Tungsten (W) and silicon dioxide (SiO2) have been chosen as materials for their various advantages in the optical spectrum with higher temperature stability. The design simulated with the finite integration technique (FIT), and validation of the simulated data is assured with the interference theory model (ITM). Underlying physics of the absorption characteristics and performance of the structure are immensely explained. The design acquired an average absorption of 96.7% from 380nm to 700nm and 96.61% for both transverse electric (TE) and transverse magnetic (TM) polarization. It has a peak absorption of 99.99% at 498.25nm, and over 99% from 460nm to 540nm, due to a very good impedance match with plasmonic resonance characteristics. For both TE and TM mode, it has wide angular independence up to 60°, which can be used for solar cell and solar thermo-photovoltaics (STPV). The various parametric sweep also analyzed for acquiring the geometric shape with a wide bandwidth to find resonance wavelength. As the MMA shifts its resonance wavelength with the change of its dielectric thickness, it can be used as an optical sensor. Changing dielectric in a Pyrex will enable the structure to be used as a light detector. Such outstanding absorption properties render the proposed MMA a successful candidate for the optical wavelength applications mentioned above.


I. INTRODUCTION
Theoretical explanation of V. Veselago devotee Smith et al. to find an artificial double negative (DNG) metamaterial with negative permeability and permittivity [1], [2]. These proper-ties of metamaterials depends on their physical structure that enables them to be used as antennas [3], waveguides [4], filters [5], invisible cloaks [6], absorbers [7], [8], The associate editor coordinating the review of this manuscript and approving it for publication was Santi C. Pavone . specific absorption rates (SAR) [9], [10], detectors [11], imaging [12], RF lenses [13], superlenses [14], polarization converter [15], various kind of sensors [16]- [19]. After the first corroboration of metamaterial absorbers (MMA) in 2008 by Landy et al., they have been studied vigorously, both theoretically and practically [20]. In the same year, Tao et al. demonstrated a single and THz MMA at 1.3 THz, which opened the door for THz applications [21]. THz metamaterials can be differentiated by single band [22], dualband [23], [24], multi-band [25], [26], tunable [27], [28], VOLUME 8, 2020 This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/ broadband [29], [30]. A broadband near-perfect MMA in the solar spectrum generally consists of a three-layer structure with metal-dielectric-metal (resonator) layers. The metal layer halts the transmission of the electromagnetic (EM) wave, and the dielectric layer helps the structure to create a coupling capacitance between the metal and the resonator layer [31]- [33]. However, these may not be the only reasons for a near-perfect MMA. Perfect impedance match of the metal with the free space of the solar spectrum [34], [35], the plasmonic resonance characteristics of the dielectric layer [36], [37], a symmetrical structure resonator with good confinement of the EM wave [38], [39], and a good E-field, H-field and distributed surface charge are also reasons for high absorption [40], [41]. MMA may also consist of twolayer [42], four-layer [43], [44], multilayer stacks [45]- [47] along with most common three-layer structure. As a result of the periodic structure of the artificial design of MMAs, they show high absorption with a wide wavelength even though they are ultrathin [48]- [50]. Broadband MMA was first discussed by Lee et al. in 2012 with FR4 and copper in the GHz frequency region [51]. After that, broadband MMA in the optical region was also demonstrated by many types of research until today. Some of them are discussed below with the results the studies found, the materials they used, and their angular and polarization stability. A multi-circular PMA with Al-GaAs-Al with three nearunit absorption peaks for TE and TM polarization in both the visible and near-infrared (NIR) regions [52]. A four-layer polarization-independent plasmonic metamaterial absorber based on nanocomposite (SiO 2 /Gold) found above 85% absorbance from the 400nm to the 750nm wavelength [53]. An average absorption above 95% had been achieved with a three-layer MMA consisting of TiN-SiO 2 -TiN for the full visible region at TE and TM mode. Here TiN was used as a refractory plasmonic material [54]. A unique three-layer MMA was proposed by C. Tun et al. with a phase-change material (Ge 2 Sb 2 Te 5 ) layer along with gold as a metal for NIR and the visible wavelength. They found two-peak absorption at 96.8% and 96.2% at 610nm and 870nm, respectively [55]. S. Cao et al. used a meta-nanopillar array consisting of Au/Si to achieve an average 96% absorbance for the whole visible wavelength with 60 • angular stability for both TE and TM polarization [56]. A silicon-based MMA with conical and circular holes with Au as the metal and a dielectric layer of SiO 2 had absorption above 80% from 437.9nm to 578.3 nm with peak 98.2% absorption. It had angular stability up to 60 • [57]. An MMA based on a metasurface formatted with a sandwiched metal-dielectric-metal layer found a peak of 92% absorption in the optical wavelength, with an average of an 83% absorption value [58]. Si-and Ni-based MMAs can be applicable with above 90% absorption for the optical wavelength and above 99% from 500nm to 560nm [59]. A genetic algorithm (GA)-based MMA consisting of Au-SiO 2 -Au had above 90% absorption with 40 • angular stability for TE and TM polarizations [60]. An MMA with an arbitrary cutting top layer with quartz substrate with two gold layers had above 80% absorption from 575nm to 760nm with four near-unity peaks [61].
Inspired by previous work with numerous applications in the related field, an MMA for the whole optical region has been proposed here. From the earlier literature review, it can be seen that an MMA with an average absorption up to 95% is not highly obtainable with a comprehensive incident angle like 50 • and up for TE and TM mode. Also, it is also known that the optical region is the primary source of the sun's electromagnetic emission (EM) waves. Another paramount and common problem that arises with an MMA for optical use is temperature stability, as electromagnetic emissions of the solar spectrum are very high, which creates excessive temperatures. To eradicate this type of problem, the proposed MMA uses an excellent symmetrical structure, with high-temperature solid materials, namely tungsten and silicon dioxide. The design shows 96.7% absorptivity for the optical region and 90.28% for the ultraviolet-opticalnear-infrared region. It has a near-unity absorption level of 99.99% at 498.25% and above 99% from 460nm to 540nm. The key explanation for the strong absorption is a reasonable impedance match, along with other purposes discussed later. The design is simulated, and the simulated results were proved by the interference theory model (ITM) for the whole region. Various parametric analyses and changes in dielectric and metal help to optimize the design. As it has been known that the optical region is the main source of the sun's EM waves, these forms of MMAs will open up a new age with specific implementations for the relevant sector.

II. DESIGN AND SIMULATION SETUP A. MATERIALS CHOICE
As the proposed MMA structure works in the optical wavelength, it is crucial to choose metal and dielectric layers with high-temperature stability and proper optical properties. Innumerable materials options that can be used in a three-layer structure are shown in Fig. 1, regarding wavelength with the help of various literature reviews on recent works. Only ultraviolet (UV)-optical (OPT)-near infrared (NIR) regions shown here for less complexity.
The proposed structure uses tungsten (wolfram [w]) (optical, Palik) as a metal and resonator layer and lossless silicon dioxide (SiO 2 ) (optical, Ghosh) with invariant -refractive index of 1.5 at the full visible wavelength as the dielectric layer. The optical properties of the W and SiO 2 with dispersion equation and band structure are given in [92], [93]. A perfect impedance match with free space in the optical region of tungsten is the main reason for choosing it as a metal and resonator layer with its high intrinsic losses. Tungsten itself is also an excellent absorber on its own with a low ohmic loss. High temperature stability (3422 • ), however, often renders tungsten the best choice for metal for MMAs in the optical wavelength region. Lossless resonance characteristics help SiO 2 to achieve a good impedance match with the free space of the optical region, with a large imaginary part for a wide range. As the dielectric layer helps the structure to create a suitable coupling capacitance and inductance with the resonator and metal layer, its low refractive index (1.5) helps the structure acquire higher absorbance with a larger bandwidth. SiO 2 also has a very high melting point (1710 • ), which will help the structure withstand a very high electromagnetic waveform of the solar structure.
From Fig. 2, (a)-(d) the progressive development of the design of the unit cell is exhibited step by step. In Fig. 2, (e) and (f) the physical dimension of the proposed structure can be perceived with the front-view and the side-view from the x-axis, respectively. From Table 1, the parameters of the structure can easily be understood. The most common three-layer sandwiched design with metal-dialectic-metal and a unique symmetrical-shaped resonator (SSR) is used in this structure. First, the back-layer and dielectric layer are implemented in the design with a total length and width of ''a = 1000nm''. Here, the back-layer (metal) thickness was ''tm = 150nm'' and the dielectric layer thickness was ''td = 60nm''. The back-layer metal thickness was greater than the skin depth, (δ) = (2ρ/2πf µ R µ 0 ) 1/2 to guarantee near-zero transmission in the full optical region. Then in step 1, four symmetrical shapes in four directions were applied with length ''rl = 750nm'' seen in Fig. 2(a). In step 2, a square box was added with the previous symmetrical shape shown in Fig. 2(b). And then, in step 3, a circular cuts in the resonator with a diameter of ''cd = 260nm'' has been made, which is visible in Fig. 2(c). For the final step, a cross shape with a length of -''pl = 160nm'' and a width of ''pw = 40nm'' was added inside the cylindrical cut, thus completing the proposed structure, as demonstrated in Fig. 2(d). The resonator layer thickness was ''tr = 15nm''. This design is a rational, symmetrical structure as will be proved below with sufficient data. Note that the total thickness of the unit cell is 225nm, which is ultrathin. These types of ultrathin structures can be easily rolled over to any solar thermo-photovoltaic (STPV) cell. Also note that tungsten is shown as yellow and silicon dioxide as cyan in Fig. 2.

C. SIMULATION SETUP
To gain the desired absorption level from scattering parameter, it is essential to have the correct simulation setup with good boundary conditions. Boundary conditions are used to restrict approaches for the electromagnetic amounts correlated with this. In this design for the transverse electromagnetic (TEM) mode, periodic boundary condition with the perfect electric conductor (PEC) and the perfect magnetic conductor (PMC) were taken from the x-axis and y-axis respectively. For the PEC, the electric field is asymmetrical and the magnetic field is symmetrical, while for the PMC the fields are vice versa. The operating wavelength source will pass through the z-axis as it was considered as open space with a perfectly matched layer (PML) in the negative direction which also abolish scattering. All three axis are perpendicular to each other in the setup. Linearly polarized planar wide-spectrum wave incidence on the planned absorber's top surface for simulation. For transverse electric (TE) and transverse magnetic (TM), the floquet port is used in the z-axis, with the master and the slave being in the x-axis and the y-axis, respectively. In order to render the simulation results more precise, a higher mesh order is used. All the data were simulated in the computer simulation technology microwave studio (CST MWS) based on the finite integration technique (FIT). Frequency-domain analysis has been used to acquire all the data. Data refinement was done with the CST-MATLAB interference.

A. METHODOLOGY FOR CALCULATION OF ABSORBANCE
For calculation of absorbance, scattering parameters (S-parameter) are extracted by the Nicolson-Ross-Weir (NRW) method discussed in reference [101]. As absorbance is highly dependent on the impedance match, the impedance of the unit cell Z (ω) = {µ r (ω).µ 0 /[ε r (ω).ε 0 ]} 1/2 and characteristic impedance Z 0 = (µ 0 /ε r ) 1/2 = 376.73 ≈ 377 . Where, µ 0 = permeability of free space, ε 0 = permittivity of free space, µ r = relative permeability, and ε r = relative permittivity. By calibrating the physical dimensions of the design, condition (Z (ω) = Z o ), can be achieved, and the design will give almost unity absorption at a wavelength as it highly depends on wavelength. Here, Z (ω) is Z o very close to the desired value; if we can equal the value, and design may be act as a super absorber with unity absorption. As absorption is inversely proportional to the reflection and the transmission coefficients, the absorption formula will be, Here, R(ω) is the linear value of s-parameter S 11, which is also called the reflection coefficient, and T(ω) is the linear value of the S 21 parameter, which can also be called the transmission coefficient. However, the high thickness of tungsten blocks almost all the EM waves, which is why T(ω) can be rewritten as near zero. So the final absorption formula will be, Figure 3(a) manifested the absorption characteristics for transverse electromagnetic (TEM), transverse electric (TE), Transverse (TM), from wavelengths of 300nm to700nm, using the metal and resonator as tungsten and the dielectric layer as silicon dioxide. These characteristics are calculated from (1) or (2) using the s-parameter attained from the simulation. This design shows an excellent absorbance for all three modes from the ultraviolet-optical-near-infrared region. For the TEM mode, absorption consists of an outstanding level at 300nm-1000nm at 72% and 73.36%, respectively, with an average absorption of 90.28%. A near-unity peak is found at 498.25nm with 99.99% absorbance. As the design is a very good symmetrical design, the TE and TM modes find almost the same absorption level with an average absorption of 90.22% for a large bandwidth of 700nm. However, despite a high absorption in the UV-OPT-NIR region, we will focus only on the optical region for all the sections discussed below, as it is considered to be the most important region to work with. We will consider the optical region from 380nm to 700nm as it is appraised as the standard. For a better understanding of the phenomenon of absorption, the design was divided into two different layers. One is the back-layer, which is only the metal layer, and another is the front-layer, which consists of the metal resonator and the dielectric. Absorption with the front-layer and the back-layer is shown in Fig. 3(b). From the figure, it is realized that the back-layer absorbs almost half of the absorption. Because the design used tungsten as a back-layer, it is known that surface plasmons do not operate in tungsten in the optical region, so it examined outstanding impedance match with the free space of the same region. However, the front-layer also helps the structure to achieve the impedance match. As silicon dioxide shrinks the distance from the resonator to the metal with a shallow thickness of 60nm, it creates resonance characteristics. The resonator layer's intrinsic losses are also a reason for the high impedance match. For a high impedance match, real parts of the normalized impedance must be close to zero, where the imaginary parts must remain near zero. These two phenomena were achieved by the unit cell design, which can be easily seen from Fig. 3(c). Also, an excellent coupling capacitive and inductance value consists of the metal, and the resonator layer with the help of dielectric is a major reason too for the good absorption in the whole region with a near-unity peak. However, these phenomena may not be able to capture the fallen wave if the resonator is not a symmetrical resonator with an excellent design. Four symmetrical shapes with plus-shape circular cuts imprison the electromagnetic wave as it reflects the wave with the back-layer from the front end with complex dispersion and confines it in the dielectric layer.

C. CO-POLARIZATION AND CROSS-POLARIZATION WITH POLARIZATION CONVERSION RATIO (PCR)
The absorption curve is acquired from (1) and (2), as shown above. But one question still arises for an MMA that the unit cell must not work as a polarization converter and give the PCR value instead of the absorption. The structure is symmetrical, but to clinch the matter, we demonstrated both co-polarization and cross-polarization components in Fig. 4(a) with the help of (3) and (4). From Fig. 4(a), it can be easily seen that the cross-polarization component is almost zero in the linear magnitude scale, and this ensures that the design did not convert the waves in the scrutinized wavelength region.   here, |S TE,TE (ω)| 2 = |S TM ,TM (ω)| 2 = R yy = R xx = co-polarization reflectivity and |S TE,TM (ω)| 2 = |S TM ,TE (ω)| 2 = R yx = R xy = cross-polarization reflectivity. In addition, the PCR of the cell can be calculated by (5) or (6), which can be observable from Fig. 4(b). The PCR value of the proposed design is near zero for both TE and TM mode, which eliminates the matter of polarization conversion characteristics entirely.

D. THEORETICAL VERIFICATION
Verification of simulated result with the interference theory model discussed in this section for normal and oblique angles from 0 • to 45 • . For the verification, decoupled simulated data were compared with the calculated data acquired from the interference theory model. The comparison graph is given below in Fig. 5, with the schematic model in Fig. 5(b) and the decoupled model in Fig. 5(c). According to interference theory from the references [102]- [104], with the ground plane, total reflection S 11total for layer 1 can be calculated from (7).
here, S 11 = |S 11 |e jθ 11 , reflection coefficient for layer 1 from area 1 to 2  And, β = kd Where, β = complex propagation phase, k = wave number in are 2, d = propagation distance The above discussion proved that the design is symmetrical. So, S 12 = S 21 . Then we can rewrite the equation (7), in (8).
Finally, the absorption can be theoretically calculated from the simple equation A(ω) = 1−|S 11total | 2 , using 3 and 4, here the back layer obstructs the transmission, so T(ω) = 0.
From Fig. 5(a), we can easily realize that the simulated value of absorbance is very close to the calculated absorbance from the interference theory we discussed above. Both of the results demonstrate an excellent absorption level with peak values above 99.9% and with average absorption of 96.72% and 96.90% for simulated and calculated results, respectively. These results will help to verify the proposed design under normal conditions.

E. ABSORPTION OF HALF OF THE UNIT CELL STRUCTURE
A unique feature of the proposed design is discussed here. When the scale of the suggested unit cell configuration y-x axis parameters is half as large as the real structure, the model also exhibits an outstanding average absorption level of 96.25% with a peak of 98.24% in the full optical region, which can be easily seen from Fig. 6a-c. As it is understood, a metamaterial unit cell may be rendered considerably smaller than the free space of the wavelength and considerable phase variance at practical operating frequencies [105]. This is the main reason why such absorption has occurred for the structure. Here, the thicknesses of the resonator, dielectric, and metal layers are unchanged. When the small structure size is considered, the design may be used with half of the size.

F. POLARIZATION AND INCIDENT ANGLE STABILITY
As mentioned earlier, the structure is wide-angle stable for both the TE and the TM modes. In this study, the performance of the MMA with the various oblique incidences of the TE and TM waves are discussed. From Fig. 7(a) and (b), it can be seen that the design is entirely polarization angle sensitive from 0 • to 90 • because of its symmetrical structure, which was proved above. However, an ideal MMA must also be stable for the incident angle to use it as a solar cell, or for energy harvesting, as a solar sensor or a detector, or for many other important applications. Figure 7(c) demonstrate the design from 0 • to 60 • with the increments of 10 • for the various incident angles in TE mode.
Here, the unit cell shows minimal alteration, and we find average absorption of above 82.17% for all incident angles, with peak absorption of 70%. As the path length of the wave is directly proportional to the angle of incidence, the higher the angle, the higher the path length. Because of the long path, the coupling effect of the resonator and metal decreases for the weaker magnetic dipolar resonance. Confinement of the wave in the dielectric layer is affected by the phenomenon and creates lower absorptivity.
A different scenario occurs for the TM mode exhibited in Fig. 7(d) from the TE mode we discussed above. We have better stability with average absorptivity, which changes from 96.61% to 94.15% for the incident angle of 0 • to 90 • this time.  The MMA absorptivity level does not change as much as with the incident angle in the TM mode. The larger electric field concerning the perpendicular incident angle rather than the parallel component is the reason for better stability in the TM mode. Excellent coupling with good confinement of the wave makes this design a polarization-and incident-angle stable MMA, which enriches its application criteria for many fields. Table 1 showed that the structure used nine parameters when it was designed. Geometric shape highly depends on the parameters, and absorption depends on the geometric structure. The most important parameter sweep has been discussed here, which showed in Fig. 8.

G. GEOMETRIC PARAMETER SWEEP
First of all, the parameter ''a'' has been changed from 1000nm to 1200nm, which will increase the length and width of the back-layer and dielectric layer proportionally. This parameter sweep will make sure the matter that, when the resonator fit with the dielectric and metal layer, the design exhibited the highest average absorption value shown in Fig. 8(a). However, average absorption is almost near 96.70% for all value of ''a'', but the resonant absorption finds with the value ''a = 1000nm''. Another parameter sweep demonstrated in Fig. 8(b) where the width of the outer side of the resonator has been changed from 50nm to 110nm. Also, here, not an immense variation occurred with the average absorption, which maintained an excellent level around 96.6%. But, the resonant absorption is found with ''rw = 80nm''.These two phenomena happened as the self and mutual inductance changes with the sweep of the parameter. Successive increment of ''a'' and ''rw'' increase the mutual inductance but decreased the self-inductance which produces with the metal and the resonator layer. That's why the resonance wavelength decreased with both of them. These two parameter sweeps also help to find the best geometric design with high absorption.
A major and important sweep with parameter ''td'' takes place here. Here the dielectric layer thicknesses increased by a step of 10nm from 40nm-80nm, which initiate very imp important exhibited in Fig. 8(c). Here, resonant wavelength increased linearly at 405.23nm, 442.17nm, 498.25nm, 558.51nm, and 620.64nm for 40nm-80nm width of the dielectric layer respectively. Here peak absorption is changes with 98.07%, 99.77%, 99.99%, 99.94%, and 99.93%. As the structure irradiated by an EM wave, it creates capacitance with the metal layer and resonator with the help of the dielectric layer. However, the capacitance of a structure highly depends on the dielectric thickness and inversely proportional to it. Successive increment of the thickness has decreased the capacitance; hence the resonant wavelength shifted from left to right linearly. With this important phenomenon, this structure may be used as an optical sensor, dielectric layer thickness sensor. Also, the dielectric layer confined most of the wave within it and made a structure highly procurable for absorption. So, a maintained thickness of dialectic is a must for good absorption. Here we select the thickness of the dielectric at 60nm after completing the parameter sweep with different thicknesses.

H. ABSORPTION COMPARISON WITH DIFFERENT DIELECTRIC LAYERS AND METALS
Here, we simulated our design with some dielectric layer and metal (resonator and back-layer) choices. Fig 9 and 10 ensure that the choice of tungsten with silicon dioxide is perfect. However, as it is known that a high absorption level is not always a must for application, some other applications have been discussed here with different dielectric or metal layers.
First, the dielectric substance was changed with four different materials along with silicon dioxide and exhibited in Fig. 9. The new materials are gallium arsenide (GaAs), silicon nitride, graphene, and amorphous silicon. Here we can see that silicon dioxide gives far better average absorption than the other materials. The reason for this kind of variation is the different refractive indices of those materials. As we know, the lower the refractive index, the higher the absorption and the more extensive the bandwidth. The refractive indices of SiO 2 , GaAs, graphene, Si 3 N 4 , and A-Si are 1.5, 3.9, 2.6, 2.0, and 4.4, respectively. However, as the absorption level was linearly increased with Si 3 N 4 , it can be used as a light detector for the visible regime. And GaAs and A-Si can be used as half-power absorbers as they exhibited half-absorption level for almost the full region.
Finally, the unit cell was simulated with different metals, namely aluminum, copper, platinum, and silver along with tungsten, as shown in Fig. 10. A good impedance VOLUME 8, 2020  match of tungsten with a full optical region makes it perfect for the proposed cell. Here platinum and copper indicate a good absorption level from 380nm to 500nm and 980 to 575nm, respectively, with above 90% absorption. Thus, they can be used as a wideband absorber for various applications. Aluminum and silver didn't show proper absorption as they didn't match the desired impedance with this particular structure.

I. E-FIELD, H-FIELD AND SURFACE CURRENT DISTRIBUTION
Absorption characteristics depend on the electromagnetic field and surface charge distribution for both TE and TM FIGURE 11. Representation of electric-field distribution of the unit cell (a)-(c) for TE polarization in 415nm, 498.25nm, and 690nm with linear color bar (Vm −1 ) in y-x axis, (d) cross-sectional view of the middle of the structure at 498.25 (peak) resonance wavelength in z-y axis, (e)-(g) TM polarization in 415nm, 498.25nm, and 690nm with the same linear color bar (Vm −1 ) y-x axis, (h) cross-sectional view of the middle of the structure at 498.25 (peak) resonance wavelength in z-y axis for normal incidence angle. mode. As described above, the magnetic dipolar moment created for the metal resonator and the back-layer created the resonance in the dielectric layer. Thus, the absorbance is mainly confined in the SiO 2 , and the wavelength of high absorbance broadens for this reason. In this section, all these phenomena are discussed with underlying physics by appropriate Fig. 11-13 separately. Three different wavelengths with three different absorption levels are discussed here for better understanding. They are λ = 415nm, which has midlevel absorption of 95.6%, λ = 498.25nm, which has the highest absorption of 99.99%, and λ = 690nm which has comparatively low absorption compared to the previous two wavelengths, with 91.8% for TE and TM mode for normal incidence. Figure 11 shows a strong E-field for both TE and TM mode with three different wavelengths mentioned above. This is one of the most important reason for high absorption discussed previously. It is seen that the E-field is highly confined by the dielectric layer SiO 2 , which enlivens the electric resonance dipolar moment that can be seen in Fig.  -11(d) and (h) for both TE and TM. The electric-field mainly distributed in the center of the structure. The high spots of the E-field are mainly detected on the interfaces of the metal resonator/dielectric layer, which described the increasing surface plasmon effect shown by Fig. 11(a)-(c) and (e)-(f). These surface plasmons created a perfect resonant dipole to magnify the localized E-field [40], [106]. Therefore, increased surface plasmons at the interfaces with the strong optical properties of tungsten guided the MMA for high absorption for those wavelengths.
Magnetic-field exhibited in Fig. 12, for three different wavelengths in TE and TM mode as like as E-field. Like reliable and distributed E-field, the design has a very good distributed H-field, which ushered the structure to a very good polarization-independent MMA for the whole optical region. Fig. 12(a)-(c) and (e)-(g) show that the magnetic field is highly localized on the metal resonator and extensively confined by the dielectric layer, which can be seen from Fig. 12(d) and (h). For the peak absorption level at λ = 498.25nm, the H-field is very strong for both TE and TM modes as can be seen from Fig. 12(b) and (f), which operates different directions. The cross-sectional view at VOLUME 8, 2020 λ = 498.25nm also confirms that the magnetic-field intensity is vigorous in the center of the structure. The H-field only changes its direction as the polarization mode varies from TE to TM. Anti-parallel current density from the metal layer to the metal resonator creates a loop. Artificial magnetic dipolar moment rises by the current density which is interrelated highly with the H-field [41], [107]. Therefore, the structure creates an excitation for the H-field and insinuates a very strong magnetic resonant dipole and creates a very good absorbance for the whole optical region. Another important phenomenon has been confirmed by Figs. 12(d) and (h), that the thickness of the back-layer metal completely blocks the transmission for TE, TM, or TEM mode.
With high E-field and H-field, a good surface charge distribution is another main reason for good absorbance over a wide bandwidth. Fig. 13(a)-(c) and (d)-(f), demonstrate surface charge distribution for the both TE and TM modes respectively. It can be seen from the figure that at λ = 498.25nm, the surface charge is very widely distributed throughout the whole unit cell. For the TE and TM modes, only the direction of propagation changes. However, for λ = 415nm and λ = 690nm, distribution is less than at λ = 498.25nm. This uniform anti-parallel circulating surface charge distribution is a reason for the high electromagnetic field created in the dielectric layer by the structure discussed above [22], [53], [108].

J. COMPARATIVE STUDY
As mentioned earlier, an MMA that covers the full optical region is highly desirable. In this section, a comparison of the structure is made with ones in several recent studies with similar features and similar wavelength ranges. From Table 2, it can be seen that in many details this design is superior to other designs published to date. First of all, in this project, the design consisted of an ultrathin with materials of a much higher temperature table than others. As this proposed MMA is supposed to operate in the solar spectrum, high temperature stability is a must. Secondly, the materials used in this structure are more cost-effective than other materials like gold, silicon, and others. Thirdly, a significant feature is that the structure is highly operative at an incident angle of up to 60 • , which is very important for energy harvesting or sensor applications. Another essential feature is that, as it uses tungsten as the base metal, it does not have to use quartz or a glass layer like other structures have to use. Thus, this feature will reduce the cost of producing this design. With high bandwidth, a high absorption peak of 99.99%, and the features mentioned above, the proposed MMA is a substantial candidate for various optical region applications.

IV. CONCLUSION
In conclusion, a polarization-insensitive MMA has been designed, simulated, theoretically proved, and optimized for the optical region with an average absorption of 96.7% and a peak of 99.99%. The proposed structure is ultrathin as its thickness is 225nm, using a low refractive index dielectric layer, with a very good impedance match metal tungsten for the optical region with a sandwiched three-layer structure. Also, lossless resonance characteristics of SiO 2 , incorporated with intrinsic losses of tungsten, enable the design to have excellent absorption. A high electromagnetic field with a distributed surface distribution also perceived by the structure. Also, excellent plasmonic coupling capacitance and inductance between the metal layer and metal resonator leads the design with excellent confinement of the EM wave. Furthermore, the proposed design can be used in UV-OPT-NIR as it has above 90.22% absorption from 300nm to 1000nm. The high melting point of tungsten and silicon dioxide with a wide incident angle makes this design a perfect candidate for solar energy harvesting, STPV, and solar cell applications. A unique feature of linear change of resonant wavelength, makes this design an excellent optical sensor, and a dielectric detector. Changing the dielectric layer to Si 3 N 4 allows the model used as a light detection sensor or THz light detector. With the large demand of the optical field, the proposed design may be used as a real invisible cloak, thermal detector, light trapper, imaging, magnetic resonance imaging, plasmonic sensor, optical modulator, and other high temperature stable optical wavelength applications.
SULTAN MAHMUD received the B.Sc. degree from the Department of Electrical and Electronic Engineering, International Islamic University Chittagong (IIUC), Bangladesh, in 2020. He is a metamaterial enthusiast. He completed his undergraduate thesis on metamaterial based absorber for optical wavelength applications. His main research interest includes metamaterials and metamaterial absorber for optical applications. He is currently serving as an Associate Professor and the Department Head with the Department of Electrical and Electronic Engineering, International Islamic University Chittagong, Bangladesh. He has one published book and many publications in international proceedings and journals. His main research interests include metamaterial for communications, electromagnetic radiation, and mobile computing. He has been serving as a reviewer for several international conferences and journals. Dr. Almutairi is a member of other professional societies. In December 1993, he has been awarded a full scholarship from Kuwait University to pursue his graduate studies. He served/serving as associate editor and a reviewer for many technical publications.
MOHAMMAD TARIQUL ISLAM (Senior Member, IEEE) is currently a Professor with the Department of Electrical, Electronic and Systems Engineering, Universiti Kebangsaan Malaysia (UKM), and a Visiting Professor with the Kyushu Institute of Technology, Japan. He is the author or coauthor of about 500 research journal articles, nearly 175 conference articles, and a few book chapters on various topics related to antennas, microwaves, and electromagnetic radiation analysis with 20 inventory patents filed. Thus far, his publications have been cited 6000 times and his H-index is 38 (Source: Scopus). His Google scholar citation is 8200 and H-index is 42. His research interests include communication antenna design, satellite antennas, and electromagnetic radiation analysis.
Dr Sc. theses, and has mentored more than ten postdocs and Visiting scholars. He is a Chartered Professional Engineer-C.Eng., a member of IET, U.K., and a Senior Member of IEICE, Japan. He currently serves as the Guest Editor of Sensors Journal, an Associate Editor of IEEE ACCESS, and he was an Associate Editor of IET Electronics Letter. VOLUME 8, 2020