Open Access
Add to BibSonomy
Abstract
In this paper, a simple thermal tunable metamaterial absorber with broadband absorption in the mid-infrared regime is designed and fabricated. The TiN/VO2/Al2O3/Al four-layer structure (with top Al circular patch arrays) introduces a specific dual-dielectric method to manipulate the coupling between standing wave and magnetic resonance for the metal-insulator-metal sandwich absorber. After structural optimization, the 80% absorption bandwidth reaches 2.9 μm in both simulation and experimental results. By virtue of the insulator-to-metal transition (IMT) of VO2, the reflectance increases from 5% to 42% by heating up the absorber from room temperature to 358 K leading to a relative stable radiation temperature at the crucial IMT stage (323~348 K). This design method is quite useful for active IR absorption (or camouflage) as well as thermal tuning.
© 2017 Optical Society of America
1. Introduction
Since Jagadish Chandra Bose researched substances with chiral properties in 1898, IR metamaterials started its booming development and have attracted more and more attention due to its particular properties, such as negative refractive index [1], perfect absorption [2] and frequency selective absorption [3]. With the development of IR metamaterial absorbers, the requirement for its performance has become increasingly high, especially for the absorption bandwidth. Broadband absorption is usually realized by multilayer structure [4], geometric gradient structure [5] and special dispersive material [6]. Utilizing the properties of localized surface plasmon polariton (LSPP) mode, a dual band absorber was experimentally achieved by manipulating the filling ratios and disk array sizes [7]. Yanxia Cui has proposed an ultrabroadband infrared absorber made of sawtoothed anisotropic metamaterial of which the absorptance higher than 95% at normal incidence is supported and the full absorption width at half-maximum is about 86% [8]. By engineering the frequency dispersion of metamaterial surface to mimic an ideal absorbing sheet, a polarization-independent absorber with absorption larger than 97% is demonstrated over a larger than one octave bandwidth but only numerically [9]. Compared with our design proposed in this paper, most of the broadband absorbers require complex fabrications or treatments to realize better performance, while our design realized broadband absorption with an easy-to-fabricate sandwich structure. Moreover, with the fast development of functional materials and infrared metamaterials, tunable absorption has become a hotspot in the IR absorber and camouflage research field.
Recently, vanadium dioxide (VO2) has been extensively studied for its insulator-to-metal transition (IMT) near the room temperature at picosecond time scales. This phase transition can be induced thermally, optically or electrically. As a result of IMT, the conductivity of VO2 can increase by four orders of magnitude and the optical transmission in the near-IR will decrease significantly [10]. Utilizing the IMT property of VO2, frequency-tunable metamaterials with arrays of Ag split ring resonators (SRRs) have been reported in the near-IR range from 1.5 to 5 microns [11]. Mikhail A. Kats fabricated Y-shaped antennas on a 180-nm-thick VO2 film deposited on sapphire substrate, and achieved approximately 10% tunability of the resonance frequency at the center wavelength of 10 μm [12]. With the development of practical application of IMT, VO2 has been used in many research hotspots. For example, a sandwiched structure ZnO/VO2/ZnS forming a VO2-based smart window exhibited a high solar modulation ability (∆Tsol = 13.01%), and maintained a high Tlum at 63.24% and 57.39% both in semiconducting and metallic phases [13]. Our group has also been involved in the work of designing functional materials of which the variable dielectric constant might realize high dielectric constant and thermal tuning properties.
In this paper, an easy-to-fabricate sandwich metamaterial absorber (illustrated in Fig. 1 (a)) with VO2 was proposed to realize broadband thermal tuning absorption by utilizing the IMT properties of VO2 and the coupling between standing wave and magnetic resonance. Here, we firstly deposited VO2 by pulsed laser deposition (PLD), and conducted the dielectric analysis by a Drude-Lorentz classical oscillator model and Bruggeman equivalent medium theory. Secondly, we proposed the mechanism of broadband thermal tuning absorption design based on the dual-dielectric method and carried out experimental verification. Thirdly, we also measured the reflectance revolution of our absorber with its true temperature ranging from 303 K to 358 K.
Fig. 1 (a) Schematic of a unit cell of the thermal tunable infrared absorber. (b) Simulated and measured reflectance of the absorber at room temperature.
2. Structure design and experiment detail
Our easy-to-fabricate IR absorber/emitter is illustrated in Fig. 1(a). The VO2 film with thickness t2 = 560 nm and Al2O3 film with thickness t3 = 200 nm construct the double-dielectric spacing layer sandwiched between a layer of Al circular patch arrays (with diameter d = 2.2 μm, period a = 5 μm, and thickness t4 = 50 nm) and a layer of 500 nm thick TiN film. The Al circular patch arrays behave like frequency selective surface (FSS), which can support magnetic resonance and control the resonant wavelength. The optical properties of VO2 both in metallic phase and insulating phase are described by a classical Drude-Lorentz model [14] (Please check out the attachment for details). The optical properties of Al were expressed by the Drude model [15] with plasma frequency ωp = 2π × 2895 THz and collision frequency γ = 2π × 15.5 THz. The reference dielectric constant data of TiN and Al2O3 are obtained from Ref [16,17].
For experiment, the Al film, TiN film and Al2O3 film were deposited by electron beam evaporation technology (EBE), and the patterning of Al film was conducted by contact lithography followed by an Al lift-off procedure. The VO2 film was deposited by PLD (Pulsed Laser Deposition). Before sputtering, the PLD chamber was evacuated to 1 × 10−4 Pa. Then VO2 thin films were deposited on the TiN-deposited silicon substrate at room temperature in an O2 ambient of 0.9 Pa using a vanadium target (purity, 99.99%) at a distance of 5 cm from the substrate. After sputtering, VO2 was annealed at 753 K in an O2 ambient of 180 Pa for 50 minutes. Fourier transform infrared (FTIR) spectrometer associated with a temperature controlled stage was used to measure the reflectance of our metamaterial absorber from 303 K to 358 K. To gain a better insight into the mechanism of our design, full wave simulations of the structure are obtained by the commercial simulation software CST and a home-made numerical code based on RCWA (Rigorous Coupled Wave Analysis) under the normal incidence of a transverse electromagnetic plane wave (TEM) with its polarization along the x direction.
Since the thickness of the bottom TiN layer is much larger than its skin depth, there is no transmission in the mid-infrared regime, and the absorptance of this structure is equal to A(ω) = 1 – R(ω) (R(ω) is the reflectance). Figure 1(b) depicts the simulated and measured reflectance of the infrared metamaterial absorber at room temperature. Although the possible existence of V2O5 and V2O3 in the VO2 film may lead to the deviations between the measured results and the simulated results, the measured results are basically in accordance with both the full-wave simulation and RCWA results. Owing to the coupling between the 1st order standing wave (at λs1 = 7.72 μm for measurement, λs2* = 7.79 μm for CST full-wave simulation and λs2* = 7.62 μm for RCWA simulation) and the magnetic resonance (at λMP = 9.33 μm for measurement, λMP* = 9 μm for CST full-wave simulation and λMP* = 9.22 μm for RCWA simulation), which will be discussed thoroughly in the next session, the experimental bandwidth with the 80% absorption reaches 2.90 μm. Moreover, it is worth noting that there exists an absorption peak at the wavelength of about 2.85 μm, which is caused by the 2nd order standing wave.
3. Broadband absorption analysis
As we all known, broadband absorption is usually realized by the coupling of absorption peaks which are caused by different mechanism or resonant structures. Figure 2 has illustrated the reflectance simulated by RCWA with the thickness of VO2 varying from 0.1 μm to 1.5 μm. As is shown in the chromaticity diagram, there are both magnetic resonance and standing wave modes. Generally speaking, magnetic resonance is triggered within the dielectric layers to enhance the coupling between upper and lower metallic layers and therefore attenuates rapidly with the increasing dielectric layer thickness. On the other hand, wavelength of the standing wave modes is directly proportional to the thickness of dielectric layer. In this kind of dual-dielectric spacing layer case, although both the magnetic resonance and the standing wave shift towards longer wavelength when increasing the thickness of VO2, the wavelength of the 1st order standing wave resonance increases faster than that of the magnetic resonance, thus leading to the coupling between the magnetic resonance and the 1st order standing wave.
Fig. 2 The chromaticity diagram of reflectance with the thickness of VO2 (t2) varying from 0.1 μm to 1.5 μm (the dash lines are computed by Eq. (5).
To gain a better insight into the underlying physics, the electric/magnetic field amplitude distribution is obtained by CST full-wave simulation (consistent with the RCWA simulation) to illustrate the magnetic resonance at 9 μm (see Fig. 3). The electric charges gathered on the Al circular patch edges and the magnetic field localized in the dielectric layers under the Al circular patch characterize the field distribution. Seeing from Fig. 3(c), it can also be understood by an equivalent electric current loop formed by the antiparallel currents excited in the top and bottom metallic layer and passing through the equivalent media with dielectric constant εr and thickness h = t2 + t3. In this case, each unit cell of the absorber behaves like a magnetic moment strongly interacting with the incident magnetic field and can be mimicked by a classical LC equivalent circuit model [18] shown in Fig. 3(d). Consequently, the magnetic resonant peak wavelength is given by [19]
Here, the modifying factor c varies from 0.2 to 0.4 according to the geometrical parameters, and εr is the equivalent relative permittivity of the VO2/Al2O3 dual-dielectric spacing layer. According to the equivalent medium theory, εr is in proportion to the volume fraction of each components and increasing with the content of high dielectric constant component, VO2 in this case. Therefore, Eq. (1) is related to the VO2 thickness, leading to the redshift of the magnetic resonance in Fig. 2.Fig. 3 The electric/magnetic field amplitude distribution at λMP* = 9 μm with t2 = 560 nm. (a) The amplitude distribution of electric field in zox plane at y = 0 and (b) The amplitude distribution of magnetic field in zox plane at y = 0. (c) Schematic of a unit cell of the typical infrared sandwich metamaterial absorber/emitter. (d)The equivalent circuit model of magnetic resonance.
The measured absorption peaks at the wavelength of 7.8 μm and 2.46 μm are attributed to two standing wave modes. Considering that the surface area of Al circular patch is far less than one unit cell, the influence of standing wave caused by the Al circular patch can be ignored, which is quite different from our former work [20]. Since the extinction coefficients of Al2O3 and VO2 are much less than their refractive index in the mid-infrared regime, the phase difference of the reflection on the TiN/VO2 interface can be simplified to π. According to geometrical optics [21], excitation of standing wave modes should satisfy the following formula:
Here, the mode order n = 1, 2, 3…, t2 and t3 are the thickness of VO2 and Al2O3 layers, λVO2 = λ/nVO2 and λAl2O3 = λ/nAl2O3 are the effective wavelength in Al2O3 and VO2. Similarly, setting the thickness of VO2 as variable, we have the following relationship:
And the formula above can also be simplified asHere, λsn is the wavelength of the nth order standing wave mode. The calculation results of standing waves based on the Eq. (3) are illustrated in Fig. 2 (the dotted lines), which tallies well with the simulated results and experimental results. Therefore, the absorption at 7.62 μm for t2 = 560 nm is attributed to the 1st order standing wave as the total thickness of the dielectric layers is 1/4 the effective wavelength. Likewise, the absorption at 2.63 μm corresponds to the 2nd order standing wave because the total thickness of the dielectric layers equals to 3/4 the effective wavelength.Comparing Eq. (1) with Eq. (4), it is not difficult to find that
According to the derivative relationships shown in Eq. (5), when the thickness of VO2 film increases, the wavelength of the 1st order standing wave increases faster than that of the magnetic resonance. Therefore, the coupling between this two kinds of absorption is realized by the double dielectric method. Moreover, a thin Al2O3 layer on VO2 film can improve the impedance matching of dielectric spacing layer to air with a more moderate dielectric constant and protect VO2 layer from denaturation. All these facts are necessary for our broadband absorption design.
4. Thermal tuning properties
Thermal tuning properties of our absorber is also studied by measuring its reflectance and infrared radiation properties in the heating cycle and cooling cycle. With the temperature varying from 303 K to 358 K, we have measured the reflectance and infrared images by FTIR and FLIR (see Fig. 4). According to Kirchhoff's law of thermal radiation [22], the relationship between emissivity and absorptance of arbitrary object under a specific temperature T0 is
Here, eλ and Aλ are emissivity and absorptance at different wavelength. Although the absorptance of our absorber is dispersive, it can still be taken as a gray body with average emissivity eavg. According to infrared physics and Stefan-Boltzmann law [23], radiation temperature can be obtained from the following formula:Here, To and Tr are the true temperature and radiation temperature of the absorber. Since the FLIR works at the wavelength range of 8-14 μm, reflectance shown in Fig. 4(a) at the corresponding wavelengths determines the average emissivity. Basically, the average emissivity decreases with increasing true temperature, consistent with the results shown in Fig. 4(b). Detailed experimental data obtained in the heating cycle and cooling cycle is shown in Fig. 5. It is interesting that the radiation temperature in cooling cycle varied only 2.4 K (while the variation in heating cycle is 6.3 K) when its true temperature changed from 323K to 348 K. According to the measurement results shown in Fig. 5, the phase transition temperature can be defined as Tc ≈336 K and the temperature span of IMT, ∆T, is around 25 K (323~348 K) in the heating cycle, which can be applied to camouflage heating objects from infrared detection. In our former work [24], we have studied the V4+/V5+ mixed valence states which might lead such wide temperature span of IMT by investigating the intensity distribution and variation of the characteristic Raman modes in over oxidized VO2. In the process of IMT, the volume fraction of metallic phase increases with increasing temperature and the dielectric constant of VO2 also changes rapidly. Due to the changes of absorption properties, the average emissivity of our absorber also decreases with increasing temperature, thus the radiation temperatures keep relatively stable during the phase transition process.Fig. 4 (a) The reflectance measured by FTIR at different temperatures (heating process). (b) The infrared images measured by Forward Looking Infrared (FLIR working at 8-14μm) camera at different temperatures (To, Tr and eavg are the true temperature, radiation temperature and average emissivity).
Fig. 5 The radiation temperatures and average emissivity at different temperatures both in heating cycle and cooling cycle.
Combining the FTIR measurement (see Fig. 4(a)) and the infrared radiation temperature (see Fig. 5), the IMT process can be divided into three periods according to the volume fraction of metallic phase. For the first period, insulating phase remains dominant component of VO2 and metallic phase grows relatively slowly with increasing temperature. For the second period, the volume fraction of metallic is equal to that of insulating phase near Tc and the metallic phase grows faster than that in the first period. For the last period, metallic phase becomes dominant component of VO2 and VO2 shows metallic properties. Thermal tuning mainly embodies in the continuous tuning of the position of absorption peaks and absorptance. Seeing from Fig. 6(a), the magnetic resonant peak shifted form 9.49 μm to 7.972 μm, while its absorptance decreased from 95% to 58%. As for the absorptance tuning, taking the measures reflectance at λ = 4.5 μm in Fig. 6(b) for example, it showed first decrease then increase properties in the heating cycle and cooling cycle and can be tuned from 23% to 76%. Based on this thermal tuning properties, our broadband absorber is of much value in sensing and thermal imaging.
Fig. 6 (a)Measuring magnetic resonant peak and (b) the reflectance at the wavelength of 4.5 μm with the temperature ranging from 303 K to 358 K.
5. Conclusion
In summary, we have analyzed the dielectric properties of VO2 by a Drude-Lorentz classical oscillator model and experimentally demonstrated a broadband IR absorber based on the coupling between magnetic resonance and standing wave. The sandwich metamaterial absorber absorbs above 80% light energy over a 2.9 μm spectral bandwith from 6.8 μm to 9.7 μm at room temperature. Due to the IMT property of VO2, this absorber can also be tuned by temperature but keep a relative stable radiation temperature during the process. Our experimental results agree well with numerical simulation results. Based on this method, broadband absorption can be realized with proper materials with large dielectric constant. The demonstrated thermal tuning wideband absorption design can be applied for future applications in thermal management of optical devices, infrared detection, camouflage and other fields.
Appendix
The complex refractive index of VO2 was measured by spectroscopic ellipsometry. However, the measurement of spectroscopic ellipsometry is usually restricted to the ultraviolet, near-infrared and visible light ranges. To obtain the mid-far-infrared optical parameters, the refractive index of VO2 is measured in the range of 1.7 μm to 4 μm, and substituted into the classical Drude-Lorentz oscillator model to derive the dielectric constants with an extended spectrum range by
On the right side of Eq. (8), the first term represents a constant contribution to the real part of the dielectric constant from high-frequency electronic transition. The second term is the Drude model which represents the free-electron contribution to the dielectric constant. Here is the plasma frequency of VO2, where N is the carrier density and m* is the effective mass of electrons, and γp is the scattering rate of free-electrons. The third term is the classical Lorentz model formula of k orders. Here the weight factor Sk, resonant frequency ωk and scattering rate γk of the kth oscillator are only related to the inherent properties of materials. The measured result and the fitted result (using the fitting parameters in Table 1) shown in Fig. 7 are quite in accordance with that in Ref. [14]. For the insulating state of VO2 at room temperature (303 K), both the real part and imaginary part of dielectric constant of VO2 show first decrease then tending to be constants with increasing the wavelength. When heating VO2 up to 358 K, it completely translated into metallic state, and the real part of dielectric constant of VO2 becomes negative while the imaginary part is proportional to wavelength.
Table 1. Fitting Parameters of the Drude-Lorentz model expressed by Eq. (8)
Funding
National Natural Science Foundation of China (NSFC) (61471097,61475031, 51302027, 51522204); Program for Changjiang Scholars and Innovative Research Team in University.
References and links
1. D. R. Smith, J. B. Pendry, and M. C. K. Wiltshire, “Metamaterials and negative refractive index,” Science 305(5685), 788–792 (2004). [CrossRef] [PubMed]
2. N. I. Landy, S. Sajuyigbe, J. J. Mock, D. R. Smith, and W. J. Padilla, “Perfect metamaterial absorber,” Phys. Rev. Lett. 100(20), 207402 (2008). [CrossRef] [PubMed]
3. G. Nie, Q. Shi, Z. Zhu, and J. Shi, “Selective coherent perfect absorption in metamaterials,” Appl. Phys. Lett. 105(20), 201909 (2014). [CrossRef]
4. F. Ding, Y. Cui, X. Ge, Y. Jin, and S. He, “Ultra-broadband microwave metamaterial absorber,” Appl. Phys. Lett. 100(10), 103506 (2012). [CrossRef]
5. H. B. Zhang, P. H. Zhou, L. W. Deng, J. L. Xie, D. F. Liang, and L. J. Deng, “Frequency-dispersive resistance of high impedance surface absorber with trapezoid-coupling pattern,” J. Appl. Phys. 112(1), 014106 (2012). [CrossRef]
6. H. Zhang, S. B. Lu, J. Zheng, J. Du, S. C. Wen, D. Y. Tang, and K. P. Loh, “Molybdenum disulfide (MoS2) as a broadband saturable absorber for ultra-fast photonics,” Opt. Express 22(6), 7249–7260 (2014). [CrossRef] [PubMed]
7. C. W. Cheng, M. N. Abbas, C. W. Chiu, K. T. Lai, M. H. Shih, and Y. C. Chang, “Wide-angle polarization independent infrared broadband absorbers based on metallic multi-sized disk arrays,” Opt. Express 20(9), 10376–10381 (2012). [CrossRef] [PubMed]
8. Y. Cui, K. H. Fung, J. Xu, H. Ma, Y. Jin, S. He, and N. X. Fang, “Ultrabroadband light absorption by a sawtooth anisotropic metamaterial slab,” Nano Lett. 12(3), 1443–1447 (2012). [CrossRef] [PubMed]
9. Q. Feng, M. Pu, C. Hu, and X. Luo, “Engineering the dispersion of metamaterial surface for broadband infrared absorption,” Opt. Lett. 37(11), 2133–2135 (2012). [CrossRef] [PubMed]
10. D. Ruzmetov and S. Ramanathan, “Metal-Insulator Transition in Thin Film Vanadium Dioxide” in Thin Film Metal-Oxides (Springer US, 2010).
11. M. J. Dicken, K. Aydin, I. M. Pryce, L. A. Sweatlock, E. M. Boyd, S. Walavalkar, J. Ma, and H. A. Atwater, “Frequency tunable near-infrared metamaterials based on VO2 phase transition,” Opt. Express 17(20), 18330–18339 (2009). [CrossRef] [PubMed]
12. M. A. Kats, R. Blanchard, P. Genevet, Z. Yang, M. M. Qazilbash, D. N. Basov, S. Ramanathan, and F. Capasso, “Thermal tuning of mid-infrared plasmonic antenna arrays using a phase change material,” Opt. Lett. 38(3), 368–370 (2013). [CrossRef] [PubMed]
13. Y. Zhao, R. Xu, X. R. Zhang, X. Hu, R. J. Knize, and Y. L. Lu, “Simulation of smart windows in the ZnO/VO2 /ZnS sandwiched structure with improved thermochromic properties,” Energy Build. 66, 545–552 (2013). [CrossRef]
14. W. V. Hans, A. S. Barker, and C. N. Berglund, “Optical Properties of VO2, between 0.25 and 5 eV,” Phys. Rev. 40(4), 737 (1968).
15. S. Adachi, “Aluminum (Al),” in The Handbook on Optical Constants of Metals (World Scientific, 2012).
16. S. Bagheri, C. M. Zgrabik, T. Gissibl, A. Tittl, F. Sterl, R. Walter, S. D. Zuani, A. Berrier, T. Stauden, G. Richter, E. L. Hu, and H. Giessen, “Large-area fabrication of TiN nanoantenna arrays for refractory plasmonics in the mid- infrared by femtosecond direct laser writing and interference lithography,” Opt. Mater. Express 5(11), 2625–2633 (2015). [CrossRef]
17. E. D. Palik, “Aluminum Oxide (Al2O3)” in Handbook of Optical Constants of Solids II (Academic Press, 1991).
18. J. Zhou, E. N. Economon, T. Koschny, and C. M. Soukoulis, “Unifying approach to left-handed material design,” Opt. Lett. 31(24), 3620–3622 (2006). [CrossRef] [PubMed]
19. D. M. Cheng, J. L. Xie, H. B. Zhang, C. D. Wang, N. Zhang, and L. J. Deng, “Pantoscopic and polarization-insensitive perfect absorbers in the middle infrared spectrum,” Josa B 29(6), 1503–1510 (2012). [CrossRef]
20. G. Zhen, P. Zhou, X. Luo, J. Xie, and L. Deng, “Modes Coupling Analysis of Surface Plasmon Polaritons Based Resonance Manipulation in Infrared Metamaterial Absorber,” Sci. Rep. 7, 46093 (2017). [CrossRef] [PubMed]
21. M. Born and E. Wolf, Principles of Optics (Cambridge University Press, 2003), Chap. 3.
22. G. Kirchhoff, “Ueber das Verhältniss zwischen dem Emissionsvermögen und dem Absorptionsvermögen der Körper für Wärme und Licht,” Ann. Phys. 185(2), 275–301 (2010). [CrossRef]
23. L. Boltzmann, “Ableitung des Stefan’schen Gesetzes, betreffend die Abhängigkeit der Wärmestrahlung von der Temperatur aus der electromagnetischen Lichttheorie,” Ann. Phys. 258(6), 291–294 (1884). [CrossRef]
24. T. Huang, L. Yang, J. Qin, F. Huang, X. P. Zhu, P. H. Zhou, B. Peng, H. G. Duan, L. J. Deng, and L. Bi, “Study of the phase evolution, metal-insulator transition, and optical properties of vanadium oxide thin films,” Opt. Mater. Express 6(11), 3609–3621 (2016). [CrossRef]
