Open Access
Add to BibSonomyAbstract
We experimentally demonstrate that at terahertz frequencies perfect plasmonic absorbers made on a highly doped silicon platform can be easily realized, exhibiting near-zero dips in the reflection spectra. The unit cell of the absorber consists of a dielectric layer of SiO2 film sandwiched between a highly doped silicon wafer and the copper structures, in the form of either one-dimensional stripe array or two-dimensional cross array. The reflection spectrum of the proposed absorbers are characterized using a terahertz time-domain spectroscopy system and the experimental results are in good agreement with numerical simulations. The dependence of the absorption on the THz polarization for both the 1D and 2D absorbers are also investigated. The high performance together with the easy fabrication processes presented in this paper show that the plasmonic absorber holds high prospect in terahertz applications.
© 2016 Optical Society of America
1. Introduction
Terahertz(THz) radiation, typically defined between 0.1THz and 10THz, is widely used in various fields, ranging from biomedical imaging, homeland security, non-destructive quality to spectroscop [1–3] etc., due to the unique properties of THz radiation including higher penetrability than visible light in many dielectrics, low photon energy and spectral resolving ability to realize fingerprint detection of chemicals etc. Although the discipline of THz photonics has evolved for a few decades and a large variety of devices have been investigated for THz photonic systems such as filters, modulators, and polarizer [4–7], there are still some functionalities yet to be explored to realize novel applications, among which perfect absorbing is a typical example. Perfect terahertz absorbers, which can completely absorb THz waves thus shielding objects from THz radiation and blocking the influence of THz, have broad practical applications in imaging [8–11], detection [10] and biochemical sensoring [12] in THz technology. What’s more, the THz absorbers can also be employed to realize THz thermal emitters according to Kirchhoff’s law of thermal radiation [13], suggesting the realization of a novel THz thermal radiation source. To date most of the efficient absorbers in the THz regime rely on the concept of metamaterials, and metamaterial based perfect absorbers attract intensive research attentions since the term “perfect absorber” was first proposed by Landy et al in microwave [14]. Subsequently various schemes based on metamaterials to realize perfect absorption have been proposed across the spectrum from the microwave to terahertz [9,10] [15–17], infrared [18–21] and visible [22,23] bands. The main approach in these metamaterial absorbers is based on impedance matching. However, most metamaterial-based THz absorbers are intrinsically narrowband due to the resonant characteristics of the unit cell and the relatively low Ohmic loss of metals in the terahertz regime. To obtain a broadband absorption, multi-band and stacked multilayer absorbers were proposed and investigated numerically [15], [24]. Nevertheless, these devices with multilayer thin films or complicated structures are difficult to realize experimentally due to the challenges with alignment in fabrication. As a result, experimental results from practical devices may not be good enough in terms of absorptivity [8] or in the agreement with numerical results. Besides the use of metamaterials for absorbing applications, in the optical frequencies people mainly use plasmonic nanostructures to realize this functionality. Metallic nanostructure array fabricated on a thin dielectric layer backed by another opaque plasmonic substrate exhibits perfect absorption to the visible light due to the excitation of gap surface plasmon (GSP) mode between the bottom metallic substrate and the top nanostructure array [25]. Quite recently we report that a geometrically similar structure can be used to realize perfect absorbers in the THz regime using highly doped silicon (HDSi) instead of noble metals to circumvent the problem that metals cannot support the regular surface plasmon (SP) modes in the THz frequencies [26]. These reported plasmonic absorbers based on HDSi are of special interest for practical applications because the absorbers can be easily fabricated using the well-developed complementary metal-oxide-semiconductor (CMOS) techniques with the further possibility of being integrated with other structures on the same chip. To make the fabrication even simpler, the top layer of HDSi stripes can be even replaced with metal stripes; then the complicated process of Si deposition with high doping concentration is not required to realize the top HDSi layer. To be strictly speaking, in this case the plasmonic mode supported between the HDSi substrate and the top layer of metal structures is no longer a GSP mode because the SP mode is only present at the lower interface between the intermediate dielectric layer and the HDSi substrate. This SP mode is reflected at the upper interface between dielectric and metals which behave resembling to perfect electric conductors in the THz regime, and is like one half of the metal-insulator-metal mode in optical frequencies when a perfect conducting wall is inserted into the center of the insulator layer. As a result a tight modal confinement similar to the GSP mode is still present in the dielectric layer. We refer to this new SP mode to be conductor-insulator-plasmonic (CIP) mode, which is quite common in the THz regime when metals are present. Numerical results show that due to the excitation of the CIP mode a perfect absorption can be realized with that structure.
In this paper, we report the experimental realization of perfect THz plasmonic absorbers composed of metallic stripe array on an HDSi substrate using commercially available HDSi wafers and CMOS-compatible techniques. Both stripe arrays composed of one-dimensional (1D) and two-dimensional (2D) structures made of copper are investigated and fabricated. These structures are then characterized using a terahertz time domain spectroscopy (THz-TDS) system, and the results from the reflection spectra demonstrate that the proposed plasmonic absorbers have a near-zero reflection at resonances, which imply a perfect absorption (>96%) in the terahertz region. The absorption resonance can be adjusted by changing the metal stripe width and the absorption with 1D or 2D structures also demonstrates different polarization-dependence, which implies distinct potential applications in different occasions. All these features make our absorber quite promising in practical applications.
2. Model
We now give the design and explain the working principal for the investigated structures. Figure 1(a) depicts the cross section of one unit cell of the 1D plasmonic absorber considered first, which consists of a continuous SiO2 film of thickness t sandwiched between a highly doped n-type silicon substrate and a top layer of copper stripe with height h and width W. The permittivity of HDSi can be well described by the Drude model:
in which is the permittivity of intrinsic Si, ω is the angular frequency. The plasma frequency ωp and the damping rate γ are related with the free carrier concentration N by the following equations: where e is the electron charge, m* is the effective mass of free carries, μ is the free carrier mobility and calculated by the following empirical formula
Fig. 1 (a) The schematic diagram of cross section for one unit-cell of the THz plasmonic absorber. (b) The distribution of magnetic field for resonance frequency 1.01THz, with W = 60μm. (c) One typical optical microscopy image of the fabricated sample with W = 80μm.
The empirical parameters of μmin, μmax, Nref and α in Eq. (4) are chosen with values of 65cm2/V⋅s, 1330cm2/V⋅s, 8.5 × 1016cm−3 and 0.72 for n-type Si working at room temperatures [27] in our calculations. In practice, the free carrier concentration can be reflected by the conductivity σ of bulk material and these two quantities are related with each other by the equation:
The value of N is chosen so that the resulted conductivity of HDSi agrees with that of the Si wafer that we will use in our experiment. In our simulations the conductivity of copper is set to be 5.998 × 107S/m and the refractive index of SiO2 deposited with plasmon enhanced chemical vapor deposition (PECVD) is chosen as 1.857, which will be explained in this paper later. The cladding above the structure is assumed to be air, and the period of the stripe array is P. The THz beam is incident downward normal to the plane of the structure. The finite element method (FEM) together with the periodic boundary conditions implemented in the commercial software of Comsol Multiphysics is used for the numerical simulations. By scanning the frequency of the incident THz beam, the reflection spectrum can be obtained.
3. Experimental results and dissicussion
We start with THz plasmonic absorbers composed of 1D stripe array, whose cross section is shown in Fig. 1(a). Figure 2 presents the numerically simulated reflection spectra for different width value W of the copper stripe as shown in Fig. 1(a). In all these simulations the other structural parameters are set constantly as follows: P = 200 μm, t = 11.6 μm, h = 500 nm. It is quite clear that each reflection spectrum has two resonance dips, one at the lower frequency and the other higher. The first resonance shifts as a function of the copper stripe width, as shown in Fig. 2 and this resonance is attributed to the excitation of the first order Fabry-Perot (FP) resonance. The excited CIP mode will propagate back and forth inside the SiO2 layer and gets reflected at the stripe terminations to form the FP resonance. Figure 1(b) illustrates the distribution of the magnetic field amplitude at the frequency of 1.01THz for the cooper stripe width of 60μm and one can clearly see the feature of the first order FP resonance. It is also visible that the CIP mode, which is tightly confined in the dielectric layer, extends to some extent into the HDSi layer while the penetration depth into copper is almost zero. This agrees quite well with the modal properties of the CIP mode that we mentioned above. As for the second resonance around 1.5THz, it doesn’t move as a function of the copper stripe width and has a typical profile of Fano resonance. This resonance is attributed to the excitation of the surface plasmon polariton mode propagating along the HDSi-SiO2-air interface due to the periodic metal stripe array working as the excitation grating.
Fig. 2 Numerical simulation results of reflection spectra at normal incidence for three different widths of the metal stripe W.
The fabrication of the investigated THz plasmonic absorber starts from a phosphorus-doped n-type silicon wafer which has a resistivity of 1.845mΩ⋅cm. Details of the discussion about the permittivity of the HDSi as a function of doping concentration can be found in ref [26]. A SiO2 layer was deposited on the silicon substrate using plasma enhanced chemical vapor deposition (PECVD) and the thickness was measured using prism coupler to be 11.6μm. After that a 6μm-thick layer of negative resist (Futurrex NR9-8000) was spin onto the wafer. After patterning using photolithography and developing in RD6 developer, the sample was coated with a 500nm-thick of Copper layer using electron beam evaporation for the 1D sample and magnetic sputtering for the 2D sample due to the availability of different metal deposition equipment. Finally a lift-off process was performed in acetone to get the copper array on the SiO2 layer deposited on the HDSi wafer.
With an electron concentration of 4.3 × 1019 cm−3 and Eqs. (4) and (5), one can calculate that the resulted conductivity corresponds to this value. Using Eqs. (2) and (3), a plasmon frequency of 725THz and a collision frequency of 85.5THz can be established in the numerical simulations for this HDSi. Figure 1(c) presents a typical microscopy image of the fabricated structure with the copper stripe width W equaling to 80μm. One can see clearly the 1D copper stripe array on the sample surface.
Since the substrate of HDSi is opaque in the THz regime, the absorption of the fabricated structures can be conveniently characterized by using the reflection spectrum. A commercial THz-TDs system (Z2 from Zomega Terahertz Corp, USA) was used to measure the reflection from the sample. The reflection spectrum was obtained by normalizing the reflected THz radiation from the sample surface to that from an indium tin oxide (ITO) coated glass slide. The sample and its holder was accommodated into a home-made plastic box and during the measurement, nitrogen gas was introduced into the box to minimize the influence of humidity to the measured results. The red lines in Fig. 3 present the experimentally measured reflection spectra for different copper stripe width of 60μm, 70μm and 80μm respectively. We also provide some numerical results of reflection spectra using the FEM technique for comparison. Because that there is no reported value of the refractive index in the THz regime for the SiO2 film obtained using PECVD technique, we first used the SiO2 refractive index as a fitting parameter to fit the numerically calculated reflection spectrum to the experimentally obtained one solely for the copper stripe width of 60μm. A refractive index value of 1.857 was found to provide the perfect fitting between the experimental result (red solid line) and the numerical result (black dashed line). This value was further used in the numerical calculations for the copper stripe widths of 70μm and 80μm respectively and the results are shown in Fig. 3(b) and 3(c) as the black lines. One can see that the experimental results are in excellent agreement with the numerical results quite well, both in the resonant frequency and bandwidth. For the three copper stripe width of 60μm, 70μm and 80μm, the measured reflectivity at the three resonant frequencies of 1.01THz, 0.885THz and 0.785THz is 1.8%, 2.2% and 3.1% respectively, which show that the overall absorptivity of these investigated THz plasmonic absorbers is above 96%. The bandwidth of resonances at the three reflection dips is 0.169THz, 0.171THz and 0.166THz respectively. Besides the near-zero reflection at resonances, it is also quite visible in the results that some fringes are evidently present in all the experimentally obtained reflection spectra. These fluctuations are attributed to the interference between THz beams resulting from some reflections at the surfaces of THz photonic component in our THz-TDS system, e.g. the Si plate that is used as the THz beam splitter. Due to these fringes and also possibly the imperfect reflection from the ITO surface used as the reference, at some frequencies shown in Fig. 3 the reflectivity is slightly higher than 1.0. The second reflection dip shown in Fig. 2 due to the excitation of the SP mode propagating along the HDSi surface, can also be seen in the experimental results in Fig. 3 at around 1.5 THz.
Fig. 3 (a) ~(c) Experimentally measured (red) and numerically simulated (black) reflection spectra for plasmonic absorber with the width of the metal stripe W equaling to 60μm, 70μm and 80μm. (d) Simulated (black) and measured (red) reflection spectrum for plasmonic absorber with W = 60μm when the polarized direction is changed along the stripes.
Note that these THz plasmonic absorbers only work for the x polarization (TM), i.e. the electric field perpendicular to the stripe array. With this polarization, the CIP mode can be excited by the stripe array and form the FP resonance. For the y polarization (TE) along the stripe axis, however, the CIP mode is not excited and the structure exhibits high reflection. Figure 3(d) presents the reflection spectra for both the calculated (black dashed line) and measured results (red solid line) for the y polarization for the copper width of 60μm. It is evident that in the frequency range from 0.4 to 1.5THz the sample shows a high reflection for TE-polarized THz waves, while for TM-polarized light there is a broadband absorption peak at 1.01THz, as is shown in Fig. 3(a).
Although the polarization dependence of absorbers made from 1D stripe array may find special applications such as efficient polarizers with high extinction ratio working in the reflection mode, under some circumstances perfect absorbers working independently of the polarization may also be required. Then one needs to design the absorbers with the unit cell composed of 2D symmetrical structures such like cross shaped metal stripe or metal disk. In this paper, we will use the cross shaped copper stripe array as an example. The thickness of the SiO2 film layer remains the same as t = 11.6μm and the period of the array has been shrunk to 150μm to ensure a near-zero reflection at resonance. The inset in Fig. 4(a) presents a schematic figure of the cross unit cell. The width of the copper stripe in the cross structure is set as 30μm while the length of the cross can be changed to tune the absorption resonance frequency. Figure 4(a) gives a microscopy image of the fabricated 2D structures in which the copper stripe length L is 100μm. Besides the absorbing performance, we are also interested in the polarization dependence of the 2D absorbers. So in the measurement of the fabricated 2D sample using THz-TDS to get the reflection spectra, the sample was rotated so that the polarization of the incident THz pulse would have an angle α with respect to one axis of the 2D periods, as shown in the inset of Fig. 4(a). Numerical results calculated using FEM for the 2D array demonstrate as the dashed lines shown in Fig. 4(b) that the reflection spectrum remains the same for different polarization angels. Experimental results from our measurement show the same resonance frequency fixed around 0.79THz for different polarizations. The reflection level at those frequencies off the resonance differ slightly for different polarizations, especially in the high-frequency regime. This discrepancy was mainly attributed to different humidity in our plastic box used to accommodate our sample for measurement when it was opened for the sample rotation during the measurement. Further results both from the numerical simulations and the experimental measurement show that the absorption resonance can be moved with high flexibility by changing the copper stripe length L in the cross structure, in a similar way as what has been demonstrated for the 1D structure in Fig. 3.
Fig. 4 (a) A typical microscopy image of the 2D fabricated structure. The inset illustrates the schematic of cross shaped structure within one unit cell. (b) Numerical (black dashed lines) and experimental (red solid lines) results for the 2D sample with the copper stripe width of 100μm at different polarization angle.
For the 2D sample, a similarly high level of absorptivity (>96%) can also experimentally obtained at resonance as for the 1D sample. In principal, the absorptivity for these absorbers is determined by the period and the filling ratio of area of the top metallic part to that of one unit cell while the central frequency of the absorption resonance is determined by the dimension (width W for the 1D stripe array in Fig. 1(a) and length L for the 2D cross array shown in Fig. 4(a)) of the metal stripe which should fulfill the requirement for the formation of the FP resonance. As a result, using more complicated shapes of the metal structures, a broadband THz absorber can be envisioned.
4. Conclusion
In summary, we have experimentally demonstrated perfect plasmonic absorbers both in the 1D and 2D form with copper structure array on a SiO2 layer deposited on a doped silicon substrate in the terahertz regime. The samples were fabricated by CMOS-compatible technology and characterized via a THz-TDS system. Compared with the case when the copper layer is also made from HDSi [26], the current configuration is preferable due to its ease of fabrication. The experimental results are in good agreement with numerical simulation results and both of the 1D and 2D absorbers show perfect absorption at resonance. Difference polarization dependence has also been experimentally demonstrated for the 1D and 2D absorbers. The easy fabrication process and high performances of the absorbers make us believe that the investigated THz plasmonic absorbers hold high promise for a variety of important applications in the terahertz regime.
Acknowledgments
This work was supported by Zhejiang Provincial Natural Science Foundation of China (LY15F050008). The authors would like to thank Dr. Yaocheng Shi at Zhejiang University and Dr. Miaogen Chen at China Jiliang University for helps with SiO2 deposition and metal sputtering respectively. The authors acknowledge financial support from Zhejiang Key Discipline of Instrument Science and technology and T. Wang also acknowledges Science and Technology innovation activity plans for Zhejiang college students (2015R409028).
References and links
1. M. Tonouchi, “Cutting-edge terahertz technology,” Nat. Photonics 1(2), 97–105 (2007). [CrossRef]
2. R. M. Woodward, B. E. Cole, V. P. Wallace, R. J. Pye, D. D. Arnone, E. H. Linfield, and M. Pepper, “Terahertz pulse imaging in reflection geometry of human skin cancer and skin tissue,” Phys. Med. Biol. 47(21), 3853–3863 (2002). [CrossRef] [PubMed]
3. M. C. Beard, G. M. Turner, C. A. Schmuttenmaer, Y. U. V. P. Street, and P. O. Box, “Terahertz spectroscopy,” J. Phys. Chem. B 106(29), 7146–7159 (2002).
4. L. Liang, B. Jin, J. Wu, Y. Huang, Z. Ye, X. Huang, D. Zhou, G. Wang, X. Jia, H. Lu, L. Kang, W. Xu, J. Chen, and P. Wu, “A flexible wideband bandpass terahertz filter using multi-layer metamaterials,” Appl. Phys. B 113(2), 285–290 (2013). [CrossRef]
5. H. T. Chen, W. J. Padilla, J. M. Zide, S. R. Bank, A. C. Gossard, A. J. Taylor, and R. D. Averitt, “Ultrafast optical switching of terahertz metamaterials fabricated on ErAs/GaAs nanoisland superlattices,” Opt. Lett. 32(12), 1620–1622 (2007). [CrossRef] [PubMed]
6. N. K. Grady, J. E. Heyes, D. R. Chowdhury, Y. Zeng, M. T. Reiten, A. K. Azad, A. J. Taylor, D. A. R. Dalvit, and H. T. Chen, “Terahertz Metamaterials for Linear Polarization Conversion and Anomalous Refraction,” Science 340(6138), 1304–1307 (2013). [CrossRef] [PubMed]
7. Z. H. Zhu, C. C. Guo, K. Liu, J. F. Zhang, W. M. Ye, X. D. Yuan, and S. Q. Qin, “Electrically tunable polarizer based on anisotropic absorption of graphene ribbons,” Appl. Phys., A Mater. Sci. Process. 114(4), 1017–1021 (2014). [CrossRef]
8. H. Tao, N. I. Landy, C. M. Bingham, X. Zhang, R. D. Averitt, and W. J. Padilla, “A metamaterial absorber for the terahertz regime: Design, fabrication and characterization,” Opt. Express 16(10), 7181–7188 (2008). [CrossRef] [PubMed]
9. N. Landy, C. Bingham, T. Tyler, N. Jokerst, D. Smith, and W. Padilla, “Design, theory, and measurement of a polarization-insensitive absorber for terahertz imaging,” Phys. Rev. B 79(12), 125104 (2009). [CrossRef]
10. D. Y. Shchegolkov, K. Azad, J. F. O’Hara, and E. I. Simakov, “Perfect subwavelength fishnetlike metamaterial-based film terahertz absorbers,” Phys. Rev. B 82(20), 205117 (2010). [CrossRef]
11. W. Withayachumnankul, C. M. Shah, C. Fumeaux, B. S.-Y. Ung, W. J. Padilla, M. Bhaskaran, D. Abbott, and S. Sriram, “Plasmonic Resonance toward Terahertz Perfect Absorbers,” ACS Photonics 1(7), 625–630 (2014). [CrossRef]
12. P. Kung and S. M. Kim, “Terahertz Metamaterial Absorbers for Sensing and Imaging,” PIERS Proc. 1, 232–235 (2013).
13. X. Liu, T. Tyler, T. Starr, A. F. Starr, N. M. Jokerst, and W. J. Padilla, “Taming the Blackbody with Infrared Metamaterials as Selective Thermal Emitters,” Phys. Rev. Lett. 107(4), 045901 (2011). [CrossRef] [PubMed]
14. N. I. Landy, S. Sajuyigbe, J. J. Mock, D. R. Smith, and W. J. Padilla, “A Perfect Metamaterial Absorber,” Phys. Rev. Lett. 100(20), 207402 (2008). [CrossRef] [PubMed]
15. M. Amin, M. Farhat, and H. Bağcı, “An ultra-broadband multilayered graphene absorber,” Opt. Express 21(24), 29938–29948 (2013). [CrossRef] [PubMed]
16. C. Shi, X. Zang, Y. Wang, L. Chen, B. Cai, and Y. Zhu, “A polarization-independent broadband terahertz absorber,” Appl. Phys. Lett. 105(3), 031104 (2014). [CrossRef]
17. J. Zhu, Z. Ma, W. Sun, F. Ding, Q. He, L. Zhou, and Y. Ma, “Ultra-broadband terahertz metamaterial absorber,” Appl. Phys. Lett. 105(2), 021102 (2014). [CrossRef]
18. M. Pu, C. Hu, M. Wang, C. Huang, Z. Zhao, C. Wang, Q. Feng, and X. Luo, “Design principles for infrared wide-angle perfect absorber based on plasmonic structure,” Opt. Express 19(18), 17413–17420 (2011). [CrossRef] [PubMed]
19. J. Hao, L. Zhou, and M. Qiu, “Nearly total absorption of light and heat generation by plasmonic metamaterials,” Phys. Rev. B 83(16), 165107 (2011). [CrossRef]
20. C. Wu and G. Shvets, “Design of Metamaterial Surfaces with Broad-band Absorbance,” Opt. Lett. 37(3), 308–310 (2012). [CrossRef] [PubMed]
21. L. Meng, D. Zhao, Q. Li, and M. Qiu, “Polarization-sensitive perfect absorbers at near-infrared wavelengths,” Opt. Express 21(S1Suppl 1), A111–A122 (2013). [CrossRef] [PubMed]
22. J. Hao, J. Wang, X. Liu, W. J. Padilla, L. Zhou, and M. Qiu, “High performance optical absorber based on a plasmonic metamaterial,” Appl. Phys. Lett. 96(25), 251104 (2010). [CrossRef]
23. A. Tittl, P. Mai, R. Taubert, D. Dregely, N. Liu, and H. Giessen, “Palladium-Based Plasmonic Perfect Absorber in the Visible Wavelength Range and Its Application to Hydrogen Sensing,” Nano Lett. 11(10), 4366–4369 (2011). [CrossRef] [PubMed]
24. B. Zhang, J. Hendrickson, and J. Guo, “Multispectral near-perfect metamaterial absorbers using spatially multiplexed plasmon resonance metal square structures,” J. Opt. Soc. Am. B 30(3), 656 (2013). [CrossRef]
25. M. G. Nielsen, A. Pors, O. Albrektsen, and S. I. Bozhevolnyi, “Efficient absorption of visible radiation by gap plasmon resonators,” Opt. Express 20(12), 13311–13319 (2012). [CrossRef] [PubMed]
26. Y. Zhang and Z. Han, “Efficient and broadband Terahertz plasmonic absorbers using highly doped Si as the plasmonic material,” AIP Adv. 5(1), 017113 (2015). [CrossRef]
27. D. M. Caughey and R. E. Thomas, “Carrier mobilities in silicon empirically related to doping and field,” Proc. IEEE 55(12), 2192–2193 (1967). [CrossRef]
